2024-05-17 08:38:00 +02:00
|
|
|
/* $OpenBSD: sshconnect2.c,v 1.373 2024/05/17 06:38:00 jsg Exp $ */
|
2000-04-29 15:57:08 +02:00
|
|
|
/*
|
|
|
|
* Copyright (c) 2000 Markus Friedl. All rights reserved.
|
2008-11-05 06:20:46 +01:00
|
|
|
* Copyright (c) 2008 Damien Miller. All rights reserved.
|
2000-04-29 15:57:08 +02:00
|
|
|
*
|
|
|
|
* Redistribution and use in source and binary forms, with or without
|
|
|
|
* modification, are permitted provided that the following conditions
|
|
|
|
* are met:
|
|
|
|
* 1. Redistributions of source code must retain the above copyright
|
|
|
|
* notice, this list of conditions and the following disclaimer.
|
|
|
|
* 2. Redistributions in binary form must reproduce the above copyright
|
|
|
|
* notice, this list of conditions and the following disclaimer in the
|
|
|
|
* documentation and/or other materials provided with the distribution.
|
|
|
|
*
|
|
|
|
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
|
|
|
|
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
|
|
|
|
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
|
|
|
|
* IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
|
|
|
|
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
|
|
|
|
* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
|
|
|
|
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
|
|
|
|
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
|
|
|
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
|
|
|
|
* THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include "includes.h"
|
2006-03-15 01:09:42 +01:00
|
|
|
|
2006-03-15 01:29:24 +01:00
|
|
|
#include <sys/types.h>
|
2006-08-05 04:39:39 +02:00
|
|
|
#include <sys/socket.h>
|
2006-03-15 01:29:24 +01:00
|
|
|
#include <sys/wait.h>
|
2006-03-15 01:45:54 +01:00
|
|
|
#include <sys/stat.h>
|
2000-04-29 15:57:08 +02:00
|
|
|
|
2006-07-12 14:22:46 +02:00
|
|
|
#include <errno.h>
|
2010-01-08 07:07:22 +01:00
|
|
|
#include <fcntl.h>
|
2020-11-13 08:30:44 +01:00
|
|
|
#include <limits.h>
|
2007-05-20 07:11:33 +02:00
|
|
|
#include <netdb.h>
|
2006-08-05 04:39:39 +02:00
|
|
|
#include <pwd.h>
|
|
|
|
#include <signal.h>
|
2006-08-05 03:37:59 +02:00
|
|
|
#include <stdio.h>
|
2006-07-24 06:13:33 +02:00
|
|
|
#include <string.h>
|
2019-11-13 05:47:52 +01:00
|
|
|
#include <stdarg.h>
|
2006-07-24 06:07:35 +02:00
|
|
|
#include <unistd.h>
|
2013-03-20 02:55:14 +01:00
|
|
|
#if defined(HAVE_STRNVIS) && defined(HAVE_VIS_H) && !defined(BROKEN_STRNVIS)
|
2008-07-17 10:57:06 +02:00
|
|
|
#include <vis.h>
|
2008-07-17 11:15:43 +02:00
|
|
|
#endif
|
2006-07-12 14:22:46 +02:00
|
|
|
|
2003-05-14 06:31:11 +02:00
|
|
|
#include "openbsd-compat/sys-queue.h"
|
|
|
|
|
2006-08-05 04:39:39 +02:00
|
|
|
#include "xmalloc.h"
|
2000-04-29 15:57:08 +02:00
|
|
|
#include "ssh.h"
|
2001-01-22 06:34:40 +01:00
|
|
|
#include "ssh2.h"
|
2018-07-09 23:03:30 +02:00
|
|
|
#include "sshbuf.h"
|
2000-04-29 15:57:08 +02:00
|
|
|
#include "packet.h"
|
|
|
|
#include "compat.h"
|
2001-01-22 06:34:40 +01:00
|
|
|
#include "cipher.h"
|
2018-07-09 23:03:30 +02:00
|
|
|
#include "sshkey.h"
|
2000-04-29 15:57:08 +02:00
|
|
|
#include "kex.h"
|
|
|
|
#include "sshconnect.h"
|
|
|
|
#include "authfile.h"
|
2001-03-29 02:36:16 +02:00
|
|
|
#include "dh.h"
|
2000-08-23 02:46:23 +02:00
|
|
|
#include "authfd.h"
|
2001-01-22 06:34:40 +01:00
|
|
|
#include "log.h"
|
2004-05-13 08:15:47 +02:00
|
|
|
#include "misc.h"
|
2014-07-18 06:11:24 +02:00
|
|
|
#include "readconf.h"
|
2001-03-11 02:49:19 +01:00
|
|
|
#include "match.h"
|
2001-04-04 03:56:17 +02:00
|
|
|
#include "dispatch.h"
|
2001-04-13 01:34:34 +02:00
|
|
|
#include "canohost.h"
|
2002-06-06 21:57:33 +02:00
|
|
|
#include "msg.h"
|
|
|
|
#include "pathnames.h"
|
2006-06-13 05:05:15 +02:00
|
|
|
#include "uidswap.h"
|
2010-12-01 02:21:51 +01:00
|
|
|
#include "hostfile.h"
|
2015-01-14 21:05:27 +01:00
|
|
|
#include "ssherr.h"
|
2016-07-17 06:20:16 +02:00
|
|
|
#include "utf8.h"
|
2019-10-31 22:18:28 +01:00
|
|
|
#include "ssh-sk.h"
|
2019-11-12 23:36:44 +01:00
|
|
|
#include "sk-api.h"
|
2000-10-14 07:23:11 +02:00
|
|
|
|
2021-10-11 23:46:32 +02:00
|
|
|
#ifdef WINDOWS
|
|
|
|
#include "sshTelemetry.h"
|
|
|
|
#endif
|
|
|
|
|
2003-08-26 03:49:55 +02:00
|
|
|
#ifdef GSSAPI
|
|
|
|
#include "ssh-gss.h"
|
|
|
|
#endif
|
|
|
|
|
2000-04-29 15:57:08 +02:00
|
|
|
/* import */
|
|
|
|
extern Options options;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* SSH2 key exchange
|
|
|
|
*/
|
|
|
|
|
2020-12-21 00:40:19 +01:00
|
|
|
static char *xxx_host;
|
|
|
|
static struct sockaddr *xxx_hostaddr;
|
|
|
|
static const struct ssh_conn_info *xxx_conn_info;
|
2001-04-04 03:56:17 +02:00
|
|
|
|
2001-06-25 07:01:22 +02:00
|
|
|
static int
|
2017-05-30 10:52:19 +02:00
|
|
|
verify_host_key_callback(struct sshkey *hostkey, struct ssh *ssh)
|
2001-04-04 03:56:17 +02:00
|
|
|
{
|
Merge 9.1 (#626)
* upstream: fix poll() spin when a channel's output fd closes without
data in the channel buffer. Introduce more exact packing of channel fds into
the pollfd array. fixes bz3405 and bz3411; ok deraadt@ markus@
OpenBSD-Commit-ID: 06740737849c9047785622ad5d472cb6a3907d10
* upstream: select post-quantum KEX
sntrup761x25519-sha512@openssh.com as the default; ok markus@
OpenBSD-Commit-ID: f02d99cbfce22dffec2e2ab1b60905fbddf48fb9
* upstream: add support for the "corp-data" protocol extension to
allow server-side copies to be performed without having to go via the client.
Patch by Mike Frysinger, ok dtucker@
OpenBSD-Commit-ID: 00aa510940fedd66dab1843b58682de4eb7156d5
* upstream: add a sftp client "cp" command that supports server-side
copying of files. Useful for this task and for testing the copy-data
extension. Patch from Mike Frysinger; ok dtucker@
OpenBSD-Commit-ID: 1bb1b950af0d49f0d5425b1f267e197aa1b57444
* depend
* Skip slow tests on (very) slow test targets.
* Set Makefile SHELL as determined by configure.
This should improve compatibility for users with non-POSIX shells. If
using Makefile.in directly (eg make -f Makefile.in distprep) then SHELL
will need to be specified on the command line (along with MANFMT in that
particular case). ok djm@
* Use bash or ksh if available for SH in Makefile.
* Increase test timeout to allow slow VMs to finish
* Only run regression tests on slow VMs.
* Only return events from ppoll that were requested.
If the underlying system's select() returns bits that were not in the
request set, our ppoll() implementation can return revents for events
not requested, which can apparently cause a hang. Only return revents
for activity in the requested event set. bz#3416, analysis and fix by
yaroslav.kuzmin at vmssoftware com, ok djm@
* Specify TEST_SHELL=bash on AIX.
The system shells cause the agent-restrict test to fail due to some
quoting so explicitly specify bash until we can get configure to
autmatically work around that.
* Disable security key on fbsd6 test host.
* upstream: man pages: add missing commas between subordinate and
main clauses
jmc@ dislikes a comma before "then" in a conditional, so leave those
untouched.
ok jmc@
OpenBSD-Commit-ID: 9520801729bebcb3c9fe43ad7f9776ab4dd05ea3
* upstream: ssh: document sntrup761x25519-sha512@openssh.com as
default KEX
OpenBSD-Commit-ID: 12545bfa10bcbf552d04d9d9520d0f4e98b0e171
* upstream: openssh-9.0
OpenBSD-Commit-ID: 0dfb461188f4513ec024c1534da8c1ce14c20b64
* update version numbers for release
* update build-aux files to match autoconf-2.71
i.e. config.guess, config.sub and install-sh
* Revert "update build-aux files to match autoconf-2.71"
This reverts commit 0a8ca39fac6ad19096b6c263436f8b2dd51606f2.
It turns out that the checked-in copies of these files are actually newer
than autoconf-2.71's copies, so this was effectively a downgrade.
Spotted by Bo Anderson via github
* upstream: two defensive changes from Tobias Stoeckmann via GHPR287
enforce stricter invarient for sshbuf_set_parent() - never allow
a buffer to have a previously-set parent changed.
In sshbuf_reset(), if the reallocation fails, then zero the entire
buffer and not the (potentially smaller) default initial alloc size.
OpenBSD-Commit-ID: 14583203aa5d50ad38d2e209ae10abaf8955e6a9
* upstream: Note that curve25519-sha256 was later published in
RFC8731. ok djm@
OpenBSD-Commit-ID: 2ac2b5d642d4cf5918eaec8653cad9a4460b2743
* upstream: clear io_want/io_ready flags at start of poll() cycle;
avoids plausible spin during rekeying if channel io_want flags are reused
across cycles. ok markus@ deraadt@
OpenBSD-Commit-ID: 91034f855b7c73cd2591657c49ac30f10322b967
* Retire fbsd6 test VM.
It's long since out of support, relatively slow (it's i686) and the
compiler has trouble with PIE.
* Resync moduli.5 with upstream.
1.18: remove duplicate publication year; carsten dot kunze at arcor dot de
1.19: ssh-keygen's -G/-T have been replaced with -M generate/screen.
* upstream: Correct path for system known hosts file in description
of IgnoreUserKnownHosts. Patch from Martin Vahlensieck via tech@
OpenBSD-Commit-ID: 9b7784f054fa5aa4d63cb36bd563889477127215
* upstream: list the correct version number
for when usage of the sftp protocol became default and fix a typo
from ed maste
OpenBSD-Commit-ID: 24e1795ed2283fdeacf16413c2f07503bcdebb31
* upstream: Import regenerated moduli
OpenBSD-Commit-ID: f9a0726d957cf10692a231996a1f34e7f9cdfeb0
* upstream: Try to continue running local I/O for channels in state
OPEN during SSH transport rekeying. The most visible benefit is that it
should make ~-escapes work in the client (e.g. to exit) if the connection
happened to have stalled during a rekey event. Based work by and ok dtucker@
OpenBSD-Commit-ID: a66e8f254e92edd4ce09c9f750883ec8f1ea5f45
* upstream: Import regenerated moduli
OpenBSD-Commit-ID: f9a0726d957cf10692a231996a1f34e7f9cdfeb0
* upstream: regression test for sftp cp command
OpenBSD-Regress-ID: c96bea9edde3a384b254785e7f9b2b24a81cdf82
* upstream: Simplify forward-control test.
Since we no longer need to support SSH1 we don't need to run shell
commands on the other end of the connection and can use ssh -N instead.
This also makes the test less racy.
OpenBSD-Regress-ID: 32e94ce272820cc398f30b848b2b0f080d10302c
* upstream: Use ssh -f and ControlPersist ..
to start up test forwards and ssh -O stop to shut them down intead of
sleep loops. This speeds up the test by an order of magnitude.
OpenBSD-Regress-ID: eb3db5f805100919b092a3b2579c611fba3e83e7
* upstream: It looks like we can't completely avoid
waiting for processes to exit so retrieve the pid via controlmaster and
use that.
OpenBSD-Regress-ID: 8246f00f22b14e49d2ff1744c94897ead33d457b
* Cache timezone data in capsicum sandbox.
From emaste at freebsd.org, originally part of FreeBSD commit r339216
/ fc3c19a9 with autoconf bits added by me.
* Include stdlib.h for free() prototype.
... which is used inside the CUSTOM_SYS_AUTH_GET_LASTLOGIN_MSG block.
* Update OpenSSL and LibreSSL versions in tests.
* Add debian-riscv64 test target.
* upstream: Avoid an unnecessary xstrdup in rm_env() when matching
patterns. Since match_pattern() doesn't modify its arguments (they are
const), there is no need to make an extra copy of the strings in
options->send_env. From Martin Vahlensieck
OpenBSD-Commit-ID: 2c9db31e3f4d3403b49642c64ee048b2a0a39351
* upstream: Add missing includes of stdlib.h and stdint.h. We need
stdlib.h for malloc(3) and stdint.h for SIZE_MAX. Unlike the other xmss
files, ssh-xmss.c does not include xmss_commons.h so ssh-xmss.c must include
those headers itself. From Martin Vahlensieck
OpenBSD-Commit-ID: 70e28a9818cee3da1be2ef6503d4b396dd421e6b
* upstream: Remove unnecessary includes: openssl/hmac.h and
openssl/evp.h. From Martin Vahlensieck.
OpenBSD-Commit-ID: a6debb5fb0c8a44e43e8d5ca7cc70ad2f3ea31c3
* upstream: Check sshauthopt_new() for NULL. bz#3425, from
tessgauthier at microsoft.com. ok djm@
OpenBSD-Commit-ID: af0315bc3e44aa406daa7e0ae7c2d719a974483f
* upstream: Add authfd path to debug output. ok markus@
OpenBSD-Commit-ID: f735a17d1a6f2bee63bfc609d76ef8db8c090890
* upstream: avoid printing hash algorithm twice; from lucas AT sexy.is
OpenBSD-Commit-ID: 9d24671e10a84141b7c504396cabad600e47a941
* upstream: fix memleak on session-bind path; from Pedro Martelletto, ok
dtucker@
OpenBSD-Commit-ID: e85899a26ba402b4c0717b531317e8fc258f0a7e
* upstream: Don't leak SK device. Patch from Pedro Martelletto via
github PR#316. ok djm@
OpenBSD-Commit-ID: 17d11327545022e727d95fd08b213171c5a4585d
* upstream: mention that the helpers are used by ssh(1), ssh-agent(1)
and ssh-keygen(1). Previously only ssh(1) was mentioned. From Pedro
Martelletto
OpenBSD-Commit-ID: 30f880f989d4b329589c1c404315685960a5f153
* Remove now-empty int32_minmax.inc.
* Only run tests when source files change.
Also run tests on changes to V_9_0 branch.
* Add Mac OS X 12 test target.
* upstream: be stricter in which characters will be accepted in
specifying a mask length; allow only 0-9. From khaleesicodes via GHPR#278; ok
dtucker@
OpenBSD-Commit-ID: e267746c047ea86665cdeccef795a8a56082eeb2
* upstream: fix some integer overflows in sieve_large() that show up when
trying to generate modp groups > 16k bits. Reported via GHPR#306 by Bertram
Felgenhauer, but fixed in a different way. feedback/ok tb@
OpenBSD-Commit-ID: 81cbc6dd3a21c57bd6fadea10e44afe37bca558e
* upstream: remove an obsolete rsa1 format example from an example;
from megan batty
ok djm
OpenBSD-Commit-ID: db2c89879c29bf083df996bd830abfb1e70d62bf
* upstream: Add FIDO AUTHENTICATOR section and explain a bit how FIDO
works. The wording came mostly from the 8.2 OpenSSH release notes, addapted
to fit the man page. Then move the -O bits into the new section as is already
done for CERTIFICATES and MODULI GENERATION. Finally we can explain the
trade-offs of resident keys. While here, consistently refer to the FIDO
thingies as "FIDO authenticators", not "FIDO tokens".
input & OK jmc, naddy
OpenBSD-Commit-ID: dd98748d7644df048f78dcf793b3b63db9ab1d25
* upstream: make sure stdout is non-blocking; ok djm@
OpenBSD-Commit-ID: 64940fffbd1b882eda2d7c8c7a43c79368309c0d
* upstream: mux.c: mark argument as const; from Martin Vahlensieck
OpenBSD-Commit-ID: 69a1a93a55986c7c2ad9f733c093b46a47184341
* upstream: channel_new no longer frees remote_name. So update the
comment accordingly. As remote_name is not modified, it can be const as
well. From Martin Vahlensieck
OpenBSD-Commit-ID: e4e10dc8dc9f40c166ea5a8e991942bedc75a76a
* upstream: sshkey_unshield_private() contains a exact duplicate of
the code in private2_check_padding(). Pull private2_check_padding() up so the
code can be reused. From Martin Vahlensieck, ok deraadt@
OpenBSD-Commit-ID: 876884c3f0e62e8fd8d1594bab06900f971c9c85
* Add ubsan minimal testcase on OpenBSD.
As suggested by djm@.
* Note that, for now, we need variadic macros.
* Also retest OpenBSD upstream on .yml changes.
* upstream: When performing operations that glob(3) a remote path, ensure
that the implicit working directory used to construct that path escapes
glob(3) characters.
This prevents glob characters from being processed in places they
shouldn't, e.g. "cd /tmp/a*/", "get *.txt" should have the get operation
treat the path "/tmp/a*" literally and not attempt to expand it.
Reported by Lusia Kundel; ok markus@
OpenBSD-Commit-ID: 4f647f58482cbad3d58b1eab7f6a1691433deeef
* Remove duplicate bcrypt_pbkdf.o from Makefile
bcrypt_pbkdf.o is duplicated in the openbsd-compat Makefile's object
file list.
* upstream: improve error message when 'ssh-keygen -Y sign' is unable to
load a private key; bz3429, reported by Adam Szkoda ok dtucker@
OpenBSD-Commit-ID: bb57b285e67bea536ef81b1055467be2fc380e74
* upstream: Allow existing -U (use agent) flag to work with "-Y sign"
operations, where it will be interpreted to require that the private keys is
hosted in an agent; bz3429, suggested by Adam Szkoda; ok dtucker@
OpenBSD-Commit-ID: a7bc69873b99c32c42c7628ed9ea91565ba08c2f
* upstream: Remove errant apostrophe. From haruyama at queen-ml org.
OpenBSD-Commit-ID: dc6b294567cb84b384ad6ced9ca469f2bbf0bd10
* upstream: arrange for scp, when in sftp mode, to not ftruncate(3) files
early
previous behavious of unconditionally truncating the destination file
would cause "scp ~/foo localhost:" and "scp localhost:foo ~/" to
delete all the contents of their destination.
spotted by solene@ sthen@, also bz3431; ok dtucker@
OpenBSD-Commit-ID: ca39fdd39e0ec1466b9666f15cbcfddea6aaa179
* upstream: fix in-place copies; r1.163 incorrectly skipped truncation in
all cases, not just at the start of a transfer. This could cause overwrites
of larger files to leave junk at the end. Spotted by tb@
OpenBSD-Commit-ID: b189f19cd68119548c8e24e39c79f61e115bf92c
* upstream: Only run agent-ptrace.sh if gdb is available as all
architectures do not ship with gdb.
OpenBSD-Regress-ID: ec53e928803e6b87f9ac142d38888ca79a45348d
* upstream: regress test for in-place transfers and clobbering larger
files with smaller ones; would have caught last regression in scp(1)
OpenBSD-Regress-ID: 19de4e88dd3a4f7e5c1618c9be3c32415bd93bc2
* configure.ac: Add missing AC_DEFINE for caph_cache_tzdata test causing
HAVE_CAPH_CACHE_TZDATA to be missing from config.h.in.
Spotted by Bryan Drewery
* upstream: make SSHBUF_DBG/SSHBUF_TELL (off by default and only enabled
via #define) dump to stderr rather than stdout
OpenBSD-Commit-ID: 10298513ee32db8390aecb0397d782d68cb14318
* upstream: revert previous; it was broken (spotted by Theo)
OpenBSD-Commit-ID: 457c79afaca2f89ec2606405c1059b98b30d8b0d
* upstream: Note that ProxyJump also accepts the same tokens as
ProxyCommand. From pallxk via github PR#305.
OpenBSD-Commit-ID: 7115ac351b129205f1f1ffa6bbfd62abd76be7c5
* upstream: Avoid kill with -1 argument. The out_ctx label can be
reached before fork has been called. If this happens, then kill -1 would be
called, sending SIGTERM to all processes reachable by the current process.
From tobias@ and c3h2_ctf via github PR#286, ok djm@
OpenBSD-Commit-ID: 6277af1207d81202f5daffdccfeeaed4c763b1a8
* upstream: f sshpkt functions fail, then password is not cleared
with freezero. Unconditionally call freezero to guarantee that password is
removed from RAM.
From tobias@ and c3h2_ctf via github PR#286, ok djm@
OpenBSD-Commit-ID: 6b093619c9515328e25b0f8093779c52402c89cd
* upstream: refactor authorized_keys/principals handling
remove "struct ssh *" from arguments - this was only used to pass the
remote host/address. These can be passed in instead and the resulting
code is less tightly coupled to ssh_api.[ch]
ok dtucker@
OpenBSD-Commit-ID: 9d4373d013edc4cc4b5c21a599e1837ac31dda0d
* upstream: split the low-level file handling functions out from
auth2-pubkey.c
Put them in a new auth2-pubkeyfile.c to make it easier to refer to them
(e.g. in unit/fuzz tests) without having to refer to everything else
pubkey auth brings in.
ok dtucker@
OpenBSD-Commit-ID: 3fdca2c61ad97dc1b8d4a7346816f83dc4ce2217
* fuzzer for authorized_keys parsing
mostly redundant to authopt_fuzz, but it's sensitive code so IMO it
makes sense to test this layer too
* Test against LibreSSL 3.5.3.
* Test against OpenSSL 1.1.1o and 3.0.3.
* fix some bugs in the fuzzer
* upstream: keywords ref ssh_config.5;
from caspar schutijser
OpenBSD-Commit-ID: f146a19d7d5c9374c3b9c520da43b2732d7d1a4e
* upstream: ssh-keygen: implement "verify-required" certificate option.
This was already documented when support for user-verified FIDO
keys was added, but the ssh-keygen(1) code was missing.
ok djm@
OpenBSD-Commit-ID: f660f973391b593fea4b7b25913c9a15c3eb8a06
* upstream: ssh-keygen -A: do not generate DSA keys by default.
Based on github PR#303 from jsegitz with man page text from jmc@, ok markus@
djm@
OpenBSD-Commit-ID: 5c4c57bdd7063ff03381cfb6696659dd3f9f5b9f
* upstream: Add period at end of "not known by any other names"
message. github PR#320 from jschauma, ok djm@
OpenBSD-Commit-ID: bd60809803c4bfd3ebb7c5c4d918b10e275266f2
* upstream: Add missing *-sk types to ssh-keyscan manpage. From
skazi0 via github PR#294.
OpenBSD-Commit-ID: fda2c869cdb871f3c90a89fb3f985370bb5d25c0
* upstream: Make SetEnv directives first-match-wins in both
sshd_config and sshd_config; previously if the same name was reused then the
last would win (which is the opposite to how the config is supposed to work).
While there, make the ssh_config parsing more like sshd_config.
bz3438, ok dtucker
OpenBSD-Commit-ID: 797909c1e0262c0d00e09280459d7ab00f18273b
* upstream: test setenv in both client and server, test first-match-wins
too
OpenBSD-Regress-ID: 4c8804f9db38a02db480b9923317457b377fe34b
* upstream: move auth_openprincipals() and auth_openkeyfile() over to
auth2-pubkeyfile.c too; they make more sense there.
OpenBSD-Commit-ID: 9970d99f900e1117fdaab13e9e910a621b7c60ee
* upstream: make sure that UseDNS hostname lookup happens in the monitor
and not in the pledge(2)'d unprivileged process; fixes regression caused by
recent refactoring spotted by henning@
OpenBSD-Commit-ID: a089870b95101cd8881a2dff65b2f1627d13e88d
* fix possible NULL deref when built without FIDO
Analysis/fix from kircher in bz3443; ok dtucker@
* automatically enable built-in FIDO support
If libfido2 is found and usable, then enable the built-in
security key support unless --without-security-key-builtin
was requested.
ok dtucker@
* upstream: Log an error if pipe() fails while accepting a
connection. bz#3447, from vincent-openssh at vinc17 net, ok djm@
OpenBSD-Commit-ID: 9d59f19872b94900a5c79da2d57850241ac5df94
* upstream: Don't attempt to fprintf a null identity comment. From
Martin Vahlensieck via tech@.
OpenBSD-Commit-ID: 4c54d20a8e8e4e9912c38a7b4ef5bfc5ca2e05c2
* upstream: Make sure not to fclose() the same fd twice in case of an
error.
ok dtucker@
OpenBSD-Commit-ID: e384c4e05d5521e7866b3d53ca59acd2a86eef99
* upstream: make it clear that RekeyLimit applies to both transmitted
and received data. GHPR#328 from Jan Pazdziora
OpenBSD-Commit-ID: d180a905fec9ff418a75c07bb96ea41c9308c3f9
* request 1.1x API compatibility for OpenSSL >=3.x
idea/patch from Pedro Martelletto via GHPR#322; ok dtucker@
* fix broken case statement in previous
* Disable SK support if FIDO libs not found.
* Zero out LIBFIDO2 when SK support not usable.
Prevents us from trying to link them into ssh-sk-helper and failing to
build.
* upstream: Don't leak the strings allocated by order_hostkeyalgs()
and list_hostkey_types() that are passed to compat_pkalg_proposal(). Part of
github PR#324 from ZoltanFridrich, ok djm@
OpenBSD-Commit-ID: b2f6e5f60f2bba293b831654328a8a0035ef4a1b
* upstream: Roll back previous KEX changes as they aren't safe until
compat_pkalg_proposal and friends always allocate their returned strings.
Reported by Qualys.
OpenBSD-Commit-ID: 1c7a88a0d5033f42f88ab9bec58ef1cf72c81ad0
* upstream: allow arguments to sftp -D option, e.g. sftp -D
"/usr/libexec/sftp-server -el debug3"
ok markus@
OpenBSD-Commit-ID: 5a002b9f3a7aef2731fc0ffa9c921cf15f38ecce
* Update OpenSSL tests to the most recent releases.
* upstream: reflect the update to -D arg name in usage();
OpenBSD-Commit-ID: abdcde4f92b1ef094ae44210ee99d3b0155aad9c
* upstream: ignore SIGPIPE earlier in main(), specifically before
muxclient() which performs operations that could cause one; Reported by Noam
Lewis via bz3454, ok dtucker@
OpenBSD-Commit-ID: 63d8e13276869eebac6d7a05d5a96307f9026e47
* upstream: Always return allocated strings from the kex filtering so
that we can free them later. Fix one leak in compat_kex_proposal. Based on
github PR#324 from ZoltanFridrich with some simplications by me. ok djm@
OpenBSD-Commit-ID: 9171616da3307612d0ede086fd511142f91246e4
* upstream: Don't leak the strings allocated by order_hostkeyalgs()
and list_hostkey_types() that are passed to compat_pkalg_proposal(). Part of
github PR#324 from ZoltanFridrich, ok djm@
This is a roll-forward of the previous rollback now that the required
changes in compat.c have been done.
OpenBSD-Commit-ID: c7cd93730b3b9f53cdad3ae32462922834ef73eb
* upstream: bump up loglevel from debug to info when unable to open
authorized keys/principals file for errno != ENOENT; bz2042 ok dtucker
OpenBSD-Commit-ID: e79aa550d91ade6a80f081bda689da24c086d66b
* Skip select+rlimit check if sandboxing is disabled
It's not needed in that case, and the test can fail when being built
with some compiler memory sanitizer flags. bz#3441
* upstream: use consistent field names (s/char/byte)
in format description
OpenBSD-Commit-ID: 3de33572733ee7fcfd7db33d37db23d2280254f0
* upstream: Remove leftover line.
Remove extra line leftover from merge conflict. ok djm@
OpenBSD-Commit-ID: 460e2290875d7ae64971a7e669c244b1d1c0ae2e
* Move checks for pollfd.fd and nfds_t.
Move the checks for struct pollfd.fd and nfds_t to before the sandboxing
checks. This groups all the sandbox checks together so we can skip them
all when sandboxing is disabled.
* Skip all rlimit tests when sandboxing disabled.
The rlimit tests can hang when being run with some compiler sanitizers
so skip all of them if sandbox=no.
* Add clang sanitizer tests.
* upstream: Add TEST_REGRESS_CACHE_DIR.
If set, it is used to cache regress test names that have succeeded and
skip those on a re-run.
OpenBSD-Regress-ID: a7570dd29a58df59f2cca647c3c2ec989b49f247
* Move sanitizer logs into regress for collection.
* Add GCC address sanitizer build/test.
* Update sanitizer test targets:
- remove clang-sanitize-memory for now. It takes so long that the test
times out.
- add gcc sanitize-address and sanitize-undefined test targets.
* Test against openssl-3.0.5.
* Move unset to before we set anything.
* Refuse to use OpenSSL 3.0.4 due to potential RCE.
OpenSSL has a potential RCE in its RSA implementation (CVE-2022-2274)
so refuse to use that specific version.
* Capture stderr output from configure.
* Only refuse to use OpenSSL 3.0.4 on x86_64.
The potential RCE only impacts x86_64, so only refuse to use it if we're
targetting a potentially impacted architecture. ok djm@
* Remove special casing of crypt().
Configure goes to some lengths to pick crypt() from either libcrypt
or OpenSSL's libcrypto because they can more or less featureful (eg
supporting md5-style passwords).
OpenSSL removed its crypt() interface in 2002:
https://github.com/openssl/openssl/commit/69deec58 so these hijinks
should no longer be necessary. This also only links sshd with libcrypt
which is the only thing that needs it. ok djm@
* Clarify README.md text.
Clarify the text about the implications of building without OpenSSL, and
prefix the "configure --help" example command with a "./" so it's likely
to work as-is in more shells. From bz#3461.
* Split README.platform into its own line.
README.platform has general platform-specific information, having it
following text about FIDO2 on the same line could imply that it only
has information about FIDO2.
* Return ERANGE from getcwd() if buffer size is 1.
If getcwd() is supplied a buffer size of exactly 1 and a path of "/", it
could result in a nul byte being written out of array bounds. POSIX says
it should return ERANGE if the path will not fit in the available buffer
(with terminating nul). 1 byte cannot fit any possible path with its nul,
so immediately return ERANGE in that case.
OpenSSH never uses getcwd() with this buffer size, and all current
(and even quite old) platforms that we are currently known to work
on have a native getcwd() so this code is not used on those anyway.
Reported by Qualys, ok djm@
* Remove unintended changes.
I inadvertently included a couple of local changes with the OpenSSL
3.0.4 change. Revert, anything that should be there will be committed
separately.
* Add AUDIT_ARCH_PPC to supported seccomp arches.
Patch from dries.deschout at dodeco.eu.
* Rename bbone test target to ARM.
* Move vmshutdown to first step.
If a previous run on a physical runner has failed to clean up, the next
run will fail because it'll try to check out the code to a broken
directory mount. Make cleanup the first step.
* upstream: pull passphrase reading and confirmation into a separate
function so it can be used for FIDO2 PINs; no functional change
OpenBSD-Commit-ID: bf34f76b8283cc1d3f54633e0d4f13613d87bb2f
* upstream: when enrolling a resident key on a security token, check
if a credential with matching application and user ID strings already exists.
if so, prompt the user for confirmation before overwriting the credential.
patch from Pedro Martelletto via GHPR329
NB. cranks SSH_SK_VERSION_MAJOR, so any third-party FIDO middleware
implementations will need to adjust
OpenBSD-Commit-ID: e45e9f1bf2b2f32d9850669e7a8dbd64acc5fca4
* upstream: sk-usbhid: preserve error code returned by key_lookup()
it conveys useful information, such as the supplied pin being wrong.
Part of GHPR329 from Pedro Martelletto
OpenBSD-Commit-ID: c0647eb9290f793add363d81378439b273756c1b
* upstream: ssh-keygen: fix touch prompt, pin retries;
part of GHPR329 from Pedro Martelletto
OpenBSD-Commit-ID: 75d1005bd2ef8f29fa834c90d2684e73556fffe8
* crank SSH_SK_VERSION_MAJOR in sk-dummy.so
* Skip scp3 test if there's no scp on remote path.
scp -3 ends up using the scp that's in the remote path and will fail if
one is not available. Based on a patch from rapier at psc.edu.
* Convert "have_prog" function into "which".
"which" and its behaviour is not standardized, so convert the existing
have_prog function into "which" so we can rely on it being available
and what its semantics are. Add a have_prog wrapper that maintains the
existing behaviour.
* upstream: Test TEST_SSH_ELAPSED_TIMES for empty string not
executable. No-op on most platforms but should prevent warnings in -portable
on systems that don't have 'date %s'.
OpenBSD-Regress-ID: e39d79867b8065e33d0c5926fa1a31f85659d2a4
* upstream: Restore missing "!" in TEST_SSH_ELAPSED_TIMES test.
OpenBSD-Regress-ID: 38783f9676ec348c5a792caecee9a16e354b37b0
* Remove workarounds for OpenSSL missing AES-GCM.
We have some compatibility hacks that were added to support OpenSSL
versions that do not support AES GCM mode. Since that time, however,
the minimum OpenSSL version that we support has moved to 1.0.1 which
*does* have GCM, so this is no longer needed. ok djm@
* Remove workarounds for OpenSSL missing AES-CTR.
We have some compatibility hacks that were added to support OpenSSL
versions that do not support AES CTR mode. Since that time, however,
the minimum OpenSSL version that we support has moved to 1.0.1 which
*does* have CTR, so this is no longer needed. ok djm@
* Do not link scp, sftp and sftp-server w/ zlib.
Some of our binaries (eg sftp, sftp-server, scp) do not interact with
the channels code and thus do use libraries such as zlib and libcrypto
although they are linked with them. This adds a CHANNELLIBS and starts
by moving zlib into it, which means the aformentioned binaries are no
longer linked against zlib. ok djm@
* Group libcrypto and PRNGD checks together.
They're related more than the libcrypt or libiaf checks which are
currently between them. ok djm@
* Remove seed_rng calls from scp, sftp, sftp-server.
These binaries don't use OpenSSL's random functions. The next step
will be to stop linking them against libcrypto. ok djm@
* Move libcrypto into CHANNELLIBS.
This will result in sftp, sftp-server and scp no longer being linked
against libcrypto. ok djm@
* Move stale-configure check as early as possible.
We added a check in Makefile to catch the case where configure needs to
be rebuilt, however this did not happen until a build was attempted in
which case all of the work done by configure was wasted. Move this check
to the start of configure to catch it as early as possible. ok djm@
* Remove deprecated MacOS 10.15 runners.
* upstream: avoid double-free in error path introduced in r1.70; report
and fix based on GHPR#332 by v-rzh ok dtucker@
OpenBSD-Commit-ID: 3d21aa127b1f37cfc5bdc21461db369a663a951f
* Include CHANNEL and FIDO2 libs in configure output
* Factor out getrnd() and rename to getentropy().
Factor out the arc4random seeding into its own file and change the
interface to match getentropy. Use native getentropy if available.
This will make it easier to resync OpenBSD changes to arc4random.
Prompted by bz#3467, ok djm@.
* compat code for fido_dev_is_winhello()
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* check_sk_options: add temporary WinHello workaround
Up to libfido 1.10.0, WinHello advertises "clientPin" rather
than "uv" capability. This is fixed in 1.11.0. For the time
being, workaround it here.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* sk_sign: set FIDO2 uv attribute explicitely for WinHello
WinHello via libfido2 performs user verification by default.
However, if we stick to that, there's no way to differentiate
between keys created with or without "-O verify-required".
Set FIDO2 uv attribute explicitely to FIDO_OPT_FALSE, then check
if user verification has been requested.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* upstream: don't prompt for FIDO passphrase before attempting to enroll
the credential, just let the enroll operating fail and we'll attempt to get a
PIN anyway. Might avoid some unneccessary PIN prompts.
Part of GHPR#302 from Corinna Vinschen; ok dtucker@
OpenBSD-Commit-ID: bd5342ffc353ee37d39617906867c305564d1ce2
* Give unused param a name.
Fixes builds on platforms that do have fido2 but don't have
fido_dev_is_winhello.
* Actually put HAVE_STDINT_H around the stdint.h.
* Rename our getentropy to prevent possible loops.
Since arc4random seeds from getentropy, and we use OpenSSL for that
if enabled, there's the possibility that if we build on a system that
does not have getentropy then run on a system that does have it, then
OpenSSL could end up calling our getentropy and getting stuck in a loop.
Pointed out by deraadt@, ok djm@
* Test hostbased auth on github runners.
* fix SANDBOX_SECCOMP_FILTER_DEBUG
* Fix conditional for running hostbased tests.
* upstream: allow certificate validity intervals, sshsig verification
times and authorized_keys expiry-time options to accept dates in the UTC time
zone in addition to the default of interpreting them in the system time zone.
YYYYMMDD and YYMMDDHHMM[SS] dates/times will be interpreted as UTC if
suffixed with a 'Z' character.
Also allow certificate validity intervals to be specified in raw
seconds-since-epoch as hex value, e.g. -V 0x1234:0x4567890. This
is intended for use by regress tests and other tools that call
ssh-keygen as part of a CA workflow.
bz3468 ok dtucker
OpenBSD-Commit-ID: 454db1cdffa9fa346aea5211223a2ce0588dfe13
* upstream: add some tests for parse_absolute_time(), including cases
where it is forced to the UTC timezone. bz3468 ok dtucker
OpenBSD-Regress-ID: ea07ca31c2f3847a38df028ca632763ae44e8759
* Skip hostbased during Valgrind tests.
Valgrind doesn't let ssh exec ssh-keysign (because it's setuid) so skip
it during the Valgrind based tests.
See https://bugs.kde.org/show_bug.cgi?id=119404 for a discussion of this
(ironically there the problematic binary was ssh(1) back when it could
still be setuid).
* Rerun tests if any .github config file changes.
* Add a timegm implementation from Heimdal via Samba.
Fixes build on (at least Solaris 10).
* Replace deprecated ubuntu-18.04 runners with 22.04
* upstream: sftp-server: support home-directory request
Add support to the sftp-server for the home-directory extension defined
in draft-ietf-secsh-filexfer-extensions-00. This overlaps a bit with the
existing expand-path@openssh.com, but uses a more official protocol name,
and so is a bit more likely to be implemented by non-OpenSSH clients.
From Mike Frysinger, ok dtucker@
OpenBSD-Commit-ID: bfc580d05cc0c817831ae7ecbac4a481c23566ab
* fido_dev_is_winhello: return 0, not "false"
"false" is not used anywhere in OpenSSH, so return 0 like
everywhere else.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* Revert "check_sk_options: add temporary WinHello workaround"
Cygwin now comes with libfido2 1.11.0, so this workaround
isn't required anymore.
This reverts commit 242c044ab111a37aad3b0775727c36a4c5f0102c.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* upstream: use .Cm for "sign"; from josiah frentsos
OpenBSD-Commit-ID: 7f80a53d54857ac6ae49ea6ad93c5bd12231d1e4
* upstream: add an extra flag to sk_probe() to indicate whether we're
probing for a FIDO resident key or not. Unused here, but will make like
easier for portable
OpenBSD-Commit-ID: 432c8ff70e270378df9dbceb9bdeaa5b43b5a832
* on Cygwin, prefer WinHello FIDO device
If no FIDO device was explictly specified, then prefer the
windows://hello FIDO device. An exception to this is when
probing resident FIDO keys, in which case hardware FIDO
devices are preferred.
* Check for perms to run agent-getpeereid test.
Ubuntu 22.04 defaults to private home dirs which prevents "nobody"
running ssh-add during the agent-getpeereid test. Check for this and
add the necessary permissions.
* upstream: double free() in error path; from Eusgor via GHPR333
OpenBSD-Commit-ID: 39f35e16ba878c8d02b4d01d8826d9b321be26d4
* Add Cygwin (on windows-2019) test target.
In addition to installing the requisite Cygwin packages, we also need to
explicitly invoke "sh" for steps that run other scripts since the runner
environment doesn't understand #! paths.
* Add a bit more debug output.
* Fix cygwin conditional steps.
* upstream: Strictly enforce the maximum allowed SSH2 banner size in
ssh-keyscan and prevent a one-byte buffer overflow. Patch from Qualys, ok
djm@
OpenBSD-Commit-ID: 6ae664f9f4db6e8a0589425f74cd0bbf3aeef4e4
* upstream: remove incorrect check that can break enrolling a
resident key (introduced in r1.40)
OpenBSD-Commit-ID: 4cab364d518470e29e624af3d3f9ffa9c92b6f01
* upstream: attemp FIDO key signing without PIN and use the error
code returned to fall back only if necessary. Avoids PIN prompts for FIDO
tokens that don't require them; part of GHPR#302
OpenBSD-Commit-ID: 4f752aaf9f2e7c28bcaaf3d4f8fc290131bd038e
* Install Cygwin packages based on OS not config.
* initial list of allowed signers
* upstream: whitespace
OpenBSD-Commit-ID: d297e4387935d4aef091c5e9432578c2e513f538
* upstream: whitespace
OpenBSD-Commit-ID: a5d015efbfd228dc598ffdef612d2da3a579e5d8
* Add cygwin-release test target.
This also moves the cygwin package install from the workflow file to
setup_ci.sh so that we can install different sets of Cygwin packages
for different test configs.
* Add Windows 2022 test targets.
* Add libcrypt-devel to cygwin-release deps.
Based on feedback from vinschen at redhat.com.
* cross-sign allowed_signers with PGP key
Provides continuity of trust from legacy PGP release key to
the SSHSIG signing keys that we will use henceforth for git
signing.
* additional keys
* upstream: whitespace
OpenBSD-Commit-ID: c2bcbf93610d3d62ed206cdf9bf9ff98c6aaf232
* Move sftp from valgrind-2 to 3 to rebalance.
* upstream: sk-usbhid: fix key_lookup() on tokens with built-in UV
explicitly test whether the token performs built-in UV (e.g. biometric
tokens) and enable UV in that case. From Pedro Martelletto via GHPR#388
OpenBSD-Commit-ID: 007eb7e387d27cf3029ab06b88224e03eca62ccd
* Remove arc4random_uniform from arc4random.c
This was previously moved into its own file (matching OpenBSD) which
prematurely committed in commit 73541f2.
* Move OPENBSD ORIGINAL marker.
Putting this after the copyright statement (which doesn't change)
instead of before the version identifier (which does) prevents merge
conflicts when resyncing changes.
* Resync arc4random with OpenBSD.
This brings us up to current, including djm's random-reseeding change,
as prompted by logan at cyberstorm.mu in bz#3467. It brings the
platform-specific hooks from LibreSSL Portable, simplified to match our
use case. ok djm@.
* Remove DEF_WEAK, it's already in defines.h.
* openbsd-compat/bsd-asprintf: add <stdio.h> include for vsnprintf
Fixes the following build failure with Clang 15 on musl:
```
bsd-asprintf.c:51:8: error: call to undeclared library function 'vsnprintf' with type 'int (char *, unsigned long, const char *, struct __va_list_tag *)'; ISO C99 and laterclang -O2 -pipe -fdiagnostics-color=always -frecord-gcc-switches -pipe -Wunknown-warning-option -Qunused-arguments -Wall -Wpointer-arith -Wuninitialized -Wsign-compare -Wformat-security -Wsizeof-pointer-memaccess -Wno-pointer-sign -Wno-unused-result -Wmisleading-indentation -Wbitwise-instead-of-logical -fno-strict-aliasing -mretpoline -ftrapv -fzero-call-used-regs=all -fno-builtin-memset -fstack-protector-strong -fPIE -I. -I. -D_XOPEN_SOURCE=600 -D_BSD_SOURCE -D_DEFAULT_SOURCE -DSSHDIR=\"/etc/ssh\" -D_PATH_SSH_PROGRAM=\"/usr/bin/ssh\" -D_PATH_SSH_ASKPASS_DEFAULT=\"/usr/lib/misc/ssh-askpass\" -D_PATH_SFTP_SERVER=\"/usr/lib/misc/sftp-server\" -D_PATH_SSH_KEY_SIGN=\"/usr/lib/misc/ssh-keysign\" -D_PATH_SSH_PKCS11_HELPER=\"/usr/lib/misc/ssh-pkcs11-helper\" -D_PATH_SSH_SK_HELPER=\"/usr/lib/misc/ssh-sk-helper\" -D_PATH_SSH_PIDDIR=\"/run\" -D_PATH_PRIVSEP_CHROOT_DIR=\"/var/empty\" -DHAVE_CONFIG_H -c cipher-aes.c -o cipher-aes.o
do not support
implicit function declarations [-Wimplicit-function-declaration]
ret = vsnprintf(string, INIT_SZ, fmt, ap2);
^
bsd-asprintf.c:51:8: note: include the header <stdio.h> or explicitly provide a declaration for 'vsnprintf'
1 error generated.
```
* upstream: notifier_complete(NULL, ...) is a noop, so no need to test
that ctx!=NULL; from Corinna Vinschen
OpenBSD-Commit-ID: ade2f2e9cc519d01a586800c25621d910bce384a
* fix pester test failures
* upstream: fix repeated words ok miod@ jmc@
OpenBSD-Commit-ID: 6765daefe26a6b648cc15cadbbe337596af709b7
* upstream: .Li -> .Vt where appropriate; from josiah frentsos,
tweaked by schwarze
ok schwarze
OpenBSD-Commit-ID: 565046e3ce68b46c2f440a93d67c2a92726de8ed
* upstream: ssh-agent: attempt FIDO key signing without PIN and use
the error to determine whether a PIN is required and prompt only if
necessary. from Corinna Vinschen
OpenBSD-Commit-ID: dd6be6a0b7148608e834ee737c3479b3270b00dd
* upstream: a little extra debugging
OpenBSD-Commit-ID: edf1601c1d0905f6da4c713f4d9cecc7d1c0295a
* upstream: sk_enroll: never drop SSH_SK_USER_VERIFICATION_REQD flag
from response
Now that all FIDO signing calls attempt first without PIN and then
fall back to trying PIN only if that attempt fails, we can remove the
hack^wtrick that removed the UV flag from the keys returned during
enroll.
By Corinna Vinschen
OpenBSD-Commit-ID: 684517608c8491503bf80cd175425f0178d91d7f
* define HAVE_KILLPG
* upstream: sftp: Don't attempt to complete arguments for
non-existent commands
If user entered a non-existent command (e.g. because they made a
typo) there is no point in trying to complete its arguments. Skip
calling complete_match() if that's the case.
From Michal Privoznik
OpenBSD-Commit-ID: cf39c811a68cde2aeb98fc85addea4000ef6b07a
* upstream: sftp: Be a bit more clever about completions
There are commands (e.g. "get" or "put") that accept two
arguments, a local path and a remote path. However, the way
current completion is written doesn't take this distinction into
account and always completes remote or local paths.
By expanding CMD struct and "cmds" array this distinction can be
reflected and with small adjustment to completer code the correct
path can be completed.
By Michal Privoznik, ok dtucker@
OpenBSD-Commit-ID: 1396d921c4eb1befd531f5c4a8ab47e7a74b610b
* upstream: correct error value
OpenBSD-Commit-ID: 780efcbad76281f11f14b2a5ff04eb6db3dfdad4
* upstream: actually hook up restrict_websafe; the command-line flag
was never actually used. Spotted by Matthew Garrett
OpenBSD-Commit-ID: 0b363518ac4c2819dbaa3dfad4028633ab9cdff1
* upstream: Add a sshkey_check_rsa_length() call for checking the
length of an RSA key; ok markus@
OpenBSD-Commit-ID: de77cd5b11594297eda82edc594b0d32b8535134
* upstream: add a RequiredRSASize for checking RSA key length in
ssh(1). User authentication keys that fall beneath this limit will be
ignored. If a host presents a host key beneath this limit then the connection
will be terminated (unfortunately there are no fallbacks in the protocol for
host authentication).
feedback deraadt, Dmitry Belyavskiy; ok markus@
OpenBSD-Commit-ID: 430e339b2a79fa9ecc63f2837b06fdd88a7da13a
* upstream: Add RequiredRSASize for sshd(8); RSA keys that fall
beneath this limit will be ignored for user and host-based authentication.
Feedback deraadt@ ok markus@
OpenBSD-Commit-ID: 187931dfc19d51873df5930a04f2d972adf1f7f1
* upstream: better debugging for connect_next()
OpenBSD-Commit-ID: d16a307a0711499c971807f324484ed3a6036640
* upstream: sftp-server(8): add a "users-groups-by-id@openssh.com"
extension request that allows the client to obtain user/group names that
correspond to a set of uids/gids.
Will be used to make directory listings more useful and consistent
in sftp(1).
ok markus@
OpenBSD-Commit-ID: 7ebabde0bcb95ef949c4840fe89e697e30df47d3
* upstream: extend sftp-common.c:extend ls_file() to support supplied
user/group names; ok markus@
OpenBSD-Commit-ID: c70c70498b1fdcf158531117e405b6245863bfb0
* upstream: sftp client library support for
users-groups-by-id@openssh.com; ok markus@
OpenBSD-Commit-ID: ddb2f33a2da6349a9a89a8b5bcb9ca7c999394de
* upstream: use users-groups-by-id@openssh.com sftp-server extension
(when available) to fill in user/group names for directory listings.
Implement a client-side cache of see uid/gid=>user/group names. ok markus@
OpenBSD-Commit-ID: f239aeeadfa925a37ceee36ee8b256b8ccf4466e
* avoid Wuninitialized false positive in gcc-12ish
* no need for glob.h here
it also causes portability problems
* add debug on appveyor
* add sleep to pester test
* upstream: add RequiredRSASize to the list of keywords accepted by
-o; spotted by jmc@
OpenBSD-Commit-ID: fe871408cf6f9d3699afeda876f8adbac86a035e
* upstream: Fix typo. From AlexanderStohr via github PR#343.
OpenBSD-Commit-ID: a134c9b4039e48803fc6a87f955b0f4a03181497
* upstream: openssh-9.1
OpenBSD-Commit-ID: 5a467b2ee81da01a86adf1ad93b62b1728494e56
* crank versions in RPM spec files
* update release notes URL
* update .depend
* fix 9.1 compilation errors
* disable -p pester tests due to unreliability on older Windows versions
* remove extra sleep time from debugging scp pester tests
* modify -p tests to only run for Windows OS version 10 and above
* add windows specific code back into method moved from auth.c to auth2-pubkeyfile.c
* add preprocessor for WinHello
* revert preprocessor definition for winhello
* add windows preprocessor definition in key_lookup
* remove rdp block from appveyor since we are no longer debugging
* add ifdef to sftp-server.c
* make key_lookup compatible with winhello
* appveyor.yml
* increase debug of failing pester test
* add #ifdef SUPPORT_CRLF back into auth_check_principals_line method that was moved/renamed
* modify new scp.sh tests for windows
* remove in place tests from scp.sh
* remove rdp debug from appveyor
* retrigger appveyor
* change check of OS version in scp test
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
Co-authored-by: djm@openbsd.org <djm@openbsd.org>
Co-authored-by: Damien Miller <djm@mindrot.org>
Co-authored-by: Darren Tucker <dtucker@dtucker.net>
Co-authored-by: naddy@openbsd.org <naddy@openbsd.org>
Co-authored-by: dtucker@openbsd.org <dtucker@openbsd.org>
Co-authored-by: tj@openbsd.org <tj@openbsd.org>
Co-authored-by: millert@openbsd.org <millert@openbsd.org>
Co-authored-by: jmc@openbsd.org <jmc@openbsd.org>
Co-authored-by: florian@openbsd.org <florian@openbsd.org>
Co-authored-by: markus@openbsd.org <markus@openbsd.org>
Co-authored-by: Tobias Heider <me@tobhe.de>
Co-authored-by: anton@openbsd.org <anton@openbsd.org>
Co-authored-by: Tim Rice <tim@multitalents.net>
Co-authored-by: tobhe@openbsd.org <tobhe@openbsd.org>
Co-authored-by: Corinna Vinschen <vinschen@redhat.com>
Co-authored-by: Sam James <sam@gentoo.org>
Co-authored-by: jsg@openbsd.org <jsg@openbsd.org>
2022-11-02 17:06:45 +01:00
|
|
|
int r;
|
|
|
|
|
|
|
|
if ((r = sshkey_check_rsa_length(hostkey,
|
|
|
|
options.required_rsa_size)) != 0)
|
|
|
|
fatal_r(r, "Bad server host key");
|
2020-12-21 00:40:19 +01:00
|
|
|
if (verify_host_key(xxx_host, xxx_hostaddr, hostkey,
|
|
|
|
xxx_conn_info) == -1)
|
2001-10-10 07:03:11 +02:00
|
|
|
fatal("Host key verification failed.");
|
2001-04-04 03:56:17 +02:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2020-09-18 07:23:03 +02:00
|
|
|
/* Returns the first item from a comma-separated algorithm list */
|
|
|
|
static char *
|
|
|
|
first_alg(const char *algs)
|
|
|
|
{
|
|
|
|
char *ret, *cp;
|
|
|
|
|
|
|
|
ret = xstrdup(algs);
|
|
|
|
if ((cp = strchr(ret, ',')) != NULL)
|
|
|
|
*cp = '\0';
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2010-12-01 02:21:51 +01:00
|
|
|
static char *
|
2020-12-21 00:40:19 +01:00
|
|
|
order_hostkeyalgs(char *host, struct sockaddr *hostaddr, u_short port,
|
|
|
|
const struct ssh_conn_info *cinfo)
|
2010-12-01 02:21:51 +01:00
|
|
|
{
|
2020-09-18 07:23:03 +02:00
|
|
|
char *oavail = NULL, *avail = NULL, *first = NULL, *last = NULL;
|
|
|
|
char *alg = NULL, *hostname = NULL, *ret = NULL, *best = NULL;
|
2010-12-01 02:21:51 +01:00
|
|
|
size_t maxlen;
|
2020-09-18 07:23:03 +02:00
|
|
|
struct hostkeys *hostkeys = NULL;
|
2010-12-01 02:21:51 +01:00
|
|
|
int ktype;
|
2011-05-29 13:42:31 +02:00
|
|
|
u_int i;
|
2010-12-01 02:21:51 +01:00
|
|
|
|
|
|
|
/* Find all hostkeys for this hostname */
|
|
|
|
get_hostfile_hostname_ipaddr(host, hostaddr, port, &hostname, NULL);
|
|
|
|
hostkeys = init_hostkeys();
|
2011-05-29 13:42:31 +02:00
|
|
|
for (i = 0; i < options.num_user_hostfiles; i++)
|
2020-12-21 00:36:51 +01:00
|
|
|
load_hostkeys(hostkeys, hostname, options.user_hostfiles[i], 0);
|
|
|
|
for (i = 0; i < options.num_system_hostfiles; i++) {
|
|
|
|
load_hostkeys(hostkeys, hostname,
|
|
|
|
options.system_hostfiles[i], 0);
|
|
|
|
}
|
2020-12-22 01:15:22 +01:00
|
|
|
if (options.known_hosts_command != NULL) {
|
|
|
|
load_hostkeys_command(hostkeys, options.known_hosts_command,
|
2023-12-13 04:28:19 +01:00
|
|
|
"ORDER", cinfo, NULL, hostname);
|
2020-12-22 01:15:22 +01:00
|
|
|
}
|
2020-09-18 07:23:03 +02:00
|
|
|
/*
|
|
|
|
* If a plain public key exists that matches the type of the best
|
|
|
|
* preference HostkeyAlgorithms, then use the whole list as is.
|
|
|
|
* Note that we ignore whether the best preference algorithm is a
|
|
|
|
* certificate type, as sshconnect.c will downgrade certs to
|
|
|
|
* plain keys if necessary.
|
|
|
|
*/
|
|
|
|
best = first_alg(options.hostkeyalgorithms);
|
|
|
|
if (lookup_key_in_hostkeys_by_type(hostkeys,
|
2020-10-04 11:45:01 +02:00
|
|
|
sshkey_type_plain(sshkey_type_from_name(best)),
|
|
|
|
sshkey_ecdsa_nid_from_name(best), NULL)) {
|
2020-10-18 13:32:01 +02:00
|
|
|
debug3_f("have matching best-preference key type %s, "
|
|
|
|
"using HostkeyAlgorithms verbatim", best);
|
2020-09-18 07:23:03 +02:00
|
|
|
ret = xstrdup(options.hostkeyalgorithms);
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Otherwise, prefer the host key algorithms that match known keys
|
|
|
|
* while keeping the ordering of HostkeyAlgorithms as much as possible.
|
|
|
|
*/
|
2020-02-06 23:48:23 +01:00
|
|
|
oavail = avail = xstrdup(options.hostkeyalgorithms);
|
2010-12-01 02:21:51 +01:00
|
|
|
maxlen = strlen(avail) + 1;
|
|
|
|
first = xmalloc(maxlen);
|
|
|
|
last = xmalloc(maxlen);
|
|
|
|
*first = *last = '\0';
|
|
|
|
|
|
|
|
#define ALG_APPEND(to, from) \
|
|
|
|
do { \
|
|
|
|
if (*to != '\0') \
|
|
|
|
strlcat(to, ",", maxlen); \
|
|
|
|
strlcat(to, from, maxlen); \
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
while ((alg = strsep(&avail, ",")) && *alg != '\0') {
|
2015-01-14 21:05:27 +01:00
|
|
|
if ((ktype = sshkey_type_from_name(alg)) == KEY_UNSPEC)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_f("unknown alg %s", alg);
|
2020-05-13 11:52:41 +02:00
|
|
|
/*
|
|
|
|
* If we have a @cert-authority marker in known_hosts then
|
|
|
|
* prefer all certificate algorithms.
|
|
|
|
*/
|
|
|
|
if (sshkey_type_is_cert(ktype) &&
|
|
|
|
lookup_marker_in_hostkeys(hostkeys, MRK_CA)) {
|
|
|
|
ALG_APPEND(first, alg);
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
/* If the key appears in known_hosts then prefer it */
|
2010-12-01 02:21:51 +01:00
|
|
|
if (lookup_key_in_hostkeys_by_type(hostkeys,
|
2020-10-04 11:45:01 +02:00
|
|
|
sshkey_type_plain(ktype),
|
|
|
|
sshkey_ecdsa_nid_from_name(alg), NULL)) {
|
2010-12-01 02:21:51 +01:00
|
|
|
ALG_APPEND(first, alg);
|
2020-05-13 11:52:41 +02:00
|
|
|
continue;
|
|
|
|
}
|
|
|
|
/* Otherwise, put it last */
|
|
|
|
ALG_APPEND(last, alg);
|
2010-12-01 02:21:51 +01:00
|
|
|
}
|
|
|
|
#undef ALG_APPEND
|
2015-01-18 14:33:34 +01:00
|
|
|
xasprintf(&ret, "%s%s%s", first,
|
|
|
|
(*first == '\0' || *last == '\0') ? "" : ",", last);
|
2010-12-01 02:21:51 +01:00
|
|
|
if (*first != '\0')
|
2020-10-18 13:32:01 +02:00
|
|
|
debug3_f("prefer hostkeyalgs: %s", first);
|
2020-12-22 01:15:22 +01:00
|
|
|
else
|
|
|
|
debug3_f("no algorithms matched; accept original");
|
2020-09-18 07:23:03 +02:00
|
|
|
out:
|
|
|
|
free(best);
|
2013-06-01 23:31:17 +02:00
|
|
|
free(first);
|
|
|
|
free(last);
|
|
|
|
free(hostname);
|
|
|
|
free(oavail);
|
2010-12-01 02:21:51 +01:00
|
|
|
free_hostkeys(hostkeys);
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2000-04-29 15:57:08 +02:00
|
|
|
void
|
2020-12-21 00:40:19 +01:00
|
|
|
ssh_kex2(struct ssh *ssh, char *host, struct sockaddr *hostaddr, u_short port,
|
|
|
|
const struct ssh_conn_info *cinfo)
|
2000-10-14 07:23:11 +02:00
|
|
|
{
|
2023-03-06 13:14:48 +01:00
|
|
|
char *myproposal[PROPOSAL_MAX];
|
2024-01-08 01:34:33 +01:00
|
|
|
char *all_key, *hkalgs = NULL;
|
2020-02-06 23:48:23 +01:00
|
|
|
int r, use_known_hosts_order = 0;
|
2001-04-04 03:56:17 +02:00
|
|
|
|
|
|
|
xxx_host = host;
|
|
|
|
xxx_hostaddr = hostaddr;
|
2020-12-21 00:40:19 +01:00
|
|
|
xxx_conn_info = cinfo;
|
2000-10-14 07:23:11 +02:00
|
|
|
|
2023-03-06 13:14:48 +01:00
|
|
|
if (options.rekey_limit || options.rekey_interval)
|
|
|
|
ssh_packet_set_rekey_limits(ssh, options.rekey_limit,
|
|
|
|
options.rekey_interval);
|
2020-02-06 23:48:23 +01:00
|
|
|
|
|
|
|
/*
|
|
|
|
* If the user has not specified HostkeyAlgorithms, or has only
|
|
|
|
* appended or removed algorithms from that list then prefer algorithms
|
|
|
|
* that are in the list that are supported by known_hosts keys.
|
|
|
|
*/
|
|
|
|
if (options.hostkeyalgorithms == NULL ||
|
|
|
|
options.hostkeyalgorithms[0] == '-' ||
|
|
|
|
options.hostkeyalgorithms[0] == '+')
|
|
|
|
use_known_hosts_order = 1;
|
|
|
|
|
|
|
|
/* Expand or fill in HostkeyAlgorithms */
|
|
|
|
all_key = sshkey_alg_list(0, 0, 1, ',');
|
2020-10-30 02:50:07 +01:00
|
|
|
if ((r = kex_assemble_names(&options.hostkeyalgorithms,
|
2022-12-02 19:59:24 +01:00
|
|
|
kex_default_pk_alg(), all_key)) != 0) // CodeQL [SM02311] false positive: kex_assemble_names handle null all_key.
|
2020-10-30 02:50:07 +01:00
|
|
|
fatal_fr(r, "kex_assemble_namelist");
|
2020-02-06 23:48:23 +01:00
|
|
|
free(all_key);
|
|
|
|
|
2023-03-09 08:11:05 +01:00
|
|
|
if (use_known_hosts_order)
|
2023-03-06 13:14:48 +01:00
|
|
|
hkalgs = order_hostkeyalgs(host, hostaddr, port, cinfo);
|
|
|
|
|
2024-01-08 01:34:33 +01:00
|
|
|
kex_proposal_populate_entries(ssh, myproposal,
|
|
|
|
options.kex_algorithms, options.ciphers, options.macs,
|
|
|
|
compression_alg_list(options.compression),
|
2023-03-06 13:14:48 +01:00
|
|
|
hkalgs ? hkalgs : options.hostkeyalgorithms);
|
|
|
|
|
|
|
|
free(hkalgs);
|
2003-04-09 12:50:06 +02:00
|
|
|
|
2001-04-04 19:57:54 +02:00
|
|
|
/* start key exchange */
|
2018-12-27 04:25:24 +01:00
|
|
|
if ((r = kex_setup(ssh, myproposal)) != 0)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_r(r, "kex_setup");
|
2014-05-15 06:24:09 +02:00
|
|
|
#ifdef WITH_OPENSSL
|
2019-01-21 11:40:11 +01:00
|
|
|
ssh->kex->kex[KEX_DH_GRP1_SHA1] = kex_gen_client;
|
|
|
|
ssh->kex->kex[KEX_DH_GRP14_SHA1] = kex_gen_client;
|
|
|
|
ssh->kex->kex[KEX_DH_GRP14_SHA256] = kex_gen_client;
|
|
|
|
ssh->kex->kex[KEX_DH_GRP16_SHA512] = kex_gen_client;
|
|
|
|
ssh->kex->kex[KEX_DH_GRP18_SHA512] = kex_gen_client;
|
2018-12-27 04:25:24 +01:00
|
|
|
ssh->kex->kex[KEX_DH_GEX_SHA1] = kexgex_client;
|
|
|
|
ssh->kex->kex[KEX_DH_GEX_SHA256] = kexgex_client;
|
2015-02-22 19:04:21 +01:00
|
|
|
# ifdef OPENSSL_HAS_ECC
|
2019-01-21 11:40:11 +01:00
|
|
|
ssh->kex->kex[KEX_ECDH_SHA2] = kex_gen_client;
|
2015-02-22 19:04:21 +01:00
|
|
|
# endif
|
2014-05-15 06:24:09 +02:00
|
|
|
#endif
|
2019-01-21 11:40:11 +01:00
|
|
|
ssh->kex->kex[KEX_C25519_SHA256] = kex_gen_client;
|
2020-12-29 01:59:15 +01:00
|
|
|
ssh->kex->kex[KEX_KEM_SNTRUP761X25519_SHA512] = kex_gen_client;
|
2018-12-27 04:25:24 +01:00
|
|
|
ssh->kex->verify_host_key=&verify_host_key_callback;
|
2000-10-14 07:23:11 +02:00
|
|
|
|
2018-12-27 04:25:24 +01:00
|
|
|
ssh_dispatch_run_fatal(ssh, DISPATCH_BLOCK, &ssh->kex->done);
|
2024-01-08 01:34:33 +01:00
|
|
|
kex_proposal_free_entries(myproposal);
|
2010-01-08 06:50:41 +01:00
|
|
|
|
2000-10-14 07:23:11 +02:00
|
|
|
#ifdef DEBUG_KEXDH
|
|
|
|
/* send 1st encrypted/maced/compressed message */
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshpkt_start(ssh, SSH2_MSG_IGNORE)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, "markus")) != 0 ||
|
2018-07-11 20:55:11 +02:00
|
|
|
(r = sshpkt_send(ssh)) != 0 ||
|
|
|
|
(r = ssh_packet_write_wait(ssh)) != 0)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_fr(r, "send packet");
|
2000-10-14 07:23:11 +02:00
|
|
|
#endif
|
2000-04-29 15:57:08 +02:00
|
|
|
}
|
2000-05-30 05:44:51 +02:00
|
|
|
|
2000-04-29 15:57:08 +02:00
|
|
|
/*
|
|
|
|
* Authenticate user
|
|
|
|
*/
|
2000-09-23 08:15:56 +02:00
|
|
|
|
2015-01-14 21:05:27 +01:00
|
|
|
typedef struct cauthctxt Authctxt;
|
|
|
|
typedef struct cauthmethod Authmethod;
|
2003-05-14 05:46:00 +02:00
|
|
|
typedef struct identity Identity;
|
|
|
|
typedef struct idlist Idlist;
|
|
|
|
|
|
|
|
struct identity {
|
|
|
|
TAILQ_ENTRY(identity) next;
|
2015-01-14 21:05:27 +01:00
|
|
|
int agent_fd; /* >=0 if agent supports key */
|
|
|
|
struct sshkey *key; /* public/private key */
|
2003-05-14 05:46:00 +02:00
|
|
|
char *filename; /* comment for agent-only keys */
|
|
|
|
int tried;
|
|
|
|
int isprivate; /* key points to the private key */
|
2013-02-15 02:18:32 +01:00
|
|
|
int userprovided;
|
2003-05-14 05:46:00 +02:00
|
|
|
};
|
|
|
|
TAILQ_HEAD(idlist, identity);
|
2000-09-23 08:15:56 +02:00
|
|
|
|
2015-01-14 21:05:27 +01:00
|
|
|
struct cauthctxt {
|
2000-09-23 08:15:56 +02:00
|
|
|
const char *server_user;
|
2001-04-13 01:34:34 +02:00
|
|
|
const char *local_user;
|
2000-09-23 08:15:56 +02:00
|
|
|
const char *host;
|
|
|
|
const char *service;
|
2015-01-14 21:05:27 +01:00
|
|
|
struct cauthmethod *method;
|
2010-05-10 03:55:38 +02:00
|
|
|
sig_atomic_t success;
|
2001-03-09 01:12:22 +01:00
|
|
|
char *authlist;
|
2019-02-11 10:44:42 +01:00
|
|
|
#ifdef GSSAPI
|
|
|
|
/* gssapi */
|
|
|
|
gss_OID_set gss_supported_mechs;
|
2022-11-30 17:57:01 +01:00
|
|
|
size_t mech_tried; // fix CodeQL SM01735
|
2019-02-11 10:44:42 +01:00
|
|
|
#endif
|
2001-04-13 01:34:34 +02:00
|
|
|
/* pubkey */
|
2015-01-14 21:05:27 +01:00
|
|
|
struct idlist keys;
|
|
|
|
int agent_fd;
|
2001-04-13 01:34:34 +02:00
|
|
|
/* hostbased */
|
2002-06-06 21:57:33 +02:00
|
|
|
Sensitive *sensitive;
|
2015-01-30 12:43:14 +01:00
|
|
|
char *oktypes, *ktypes;
|
|
|
|
const char *active_ktype;
|
2001-09-12 20:29:00 +02:00
|
|
|
/* kbd-interactive */
|
|
|
|
int info_req_seen;
|
2019-01-04 04:27:50 +01:00
|
|
|
int attempt_kbdint;
|
|
|
|
/* password */
|
|
|
|
int attempt_passwd;
|
2003-08-26 03:49:55 +02:00
|
|
|
/* generic */
|
|
|
|
void *methoddata;
|
2000-09-23 08:15:56 +02:00
|
|
|
};
|
2015-01-14 21:05:27 +01:00
|
|
|
|
|
|
|
struct cauthmethod {
|
2000-09-23 08:15:56 +02:00
|
|
|
char *name; /* string to compare against server's list */
|
2019-01-19 22:40:21 +01:00
|
|
|
int (*userauth)(struct ssh *ssh);
|
|
|
|
void (*cleanup)(struct ssh *ssh);
|
2000-09-23 08:15:56 +02:00
|
|
|
int *enabled; /* flag in option struct that enables method */
|
|
|
|
int *batch_flag; /* flag in option struct that disables method */
|
|
|
|
};
|
|
|
|
|
2019-02-11 10:44:42 +01:00
|
|
|
static int input_userauth_service_accept(int, u_int32_t, struct ssh *);
|
|
|
|
static int input_userauth_success(int, u_int32_t, struct ssh *);
|
|
|
|
static int input_userauth_failure(int, u_int32_t, struct ssh *);
|
|
|
|
static int input_userauth_banner(int, u_int32_t, struct ssh *);
|
|
|
|
static int input_userauth_error(int, u_int32_t, struct ssh *);
|
|
|
|
static int input_userauth_info_req(int, u_int32_t, struct ssh *);
|
|
|
|
static int input_userauth_pk_ok(int, u_int32_t, struct ssh *);
|
|
|
|
static int input_userauth_passwd_changereq(int, u_int32_t, struct ssh *);
|
|
|
|
|
|
|
|
static int userauth_none(struct ssh *);
|
|
|
|
static int userauth_pubkey(struct ssh *);
|
|
|
|
static int userauth_passwd(struct ssh *);
|
|
|
|
static int userauth_kbdint(struct ssh *);
|
|
|
|
static int userauth_hostbased(struct ssh *);
|
2000-09-23 08:15:56 +02:00
|
|
|
|
2003-08-26 03:49:55 +02:00
|
|
|
#ifdef GSSAPI
|
2019-02-11 10:44:42 +01:00
|
|
|
static int userauth_gssapi(struct ssh *);
|
|
|
|
static void userauth_gssapi_cleanup(struct ssh *);
|
|
|
|
static int input_gssapi_response(int type, u_int32_t, struct ssh *);
|
|
|
|
static int input_gssapi_token(int type, u_int32_t, struct ssh *);
|
|
|
|
static int input_gssapi_error(int, u_int32_t, struct ssh *);
|
|
|
|
static int input_gssapi_errtok(int, u_int32_t, struct ssh *);
|
2003-08-26 03:49:55 +02:00
|
|
|
#endif
|
|
|
|
|
2019-01-19 22:40:21 +01:00
|
|
|
void userauth(struct ssh *, char *);
|
2001-03-09 01:12:22 +01:00
|
|
|
|
2019-02-13 00:53:10 +01:00
|
|
|
static void pubkey_cleanup(struct ssh *);
|
2019-01-19 22:40:21 +01:00
|
|
|
static int sign_and_send_pubkey(struct ssh *ssh, Identity *);
|
2021-12-19 23:08:48 +01:00
|
|
|
static void pubkey_prepare(struct ssh *, Authctxt *);
|
2016-12-05 00:54:02 +01:00
|
|
|
static void pubkey_reset(Authctxt *);
|
2017-05-30 10:52:19 +02:00
|
|
|
static struct sshkey *load_identity_file(Identity *);
|
2001-03-09 01:12:22 +01:00
|
|
|
|
2001-06-25 07:01:22 +02:00
|
|
|
static Authmethod *authmethod_get(char *authlist);
|
|
|
|
static Authmethod *authmethod_lookup(const char *name);
|
|
|
|
static char *authmethods_get(void);
|
2000-09-23 08:15:56 +02:00
|
|
|
|
|
|
|
Authmethod authmethods[] = {
|
2003-08-26 03:49:55 +02:00
|
|
|
#ifdef GSSAPI
|
2003-11-17 12:18:21 +01:00
|
|
|
{"gssapi-with-mic",
|
2003-08-26 03:49:55 +02:00
|
|
|
userauth_gssapi,
|
2019-02-11 10:44:42 +01:00
|
|
|
userauth_gssapi_cleanup,
|
2003-08-26 03:49:55 +02:00
|
|
|
&options.gss_authentication,
|
|
|
|
NULL},
|
|
|
|
#endif
|
2001-06-05 21:37:25 +02:00
|
|
|
{"hostbased",
|
|
|
|
userauth_hostbased,
|
2008-11-05 06:20:46 +01:00
|
|
|
NULL,
|
2001-06-05 21:37:25 +02:00
|
|
|
&options.hostbased_authentication,
|
|
|
|
NULL},
|
2001-08-06 22:57:11 +02:00
|
|
|
{"publickey",
|
|
|
|
userauth_pubkey,
|
2019-02-13 00:53:10 +01:00
|
|
|
NULL,
|
2001-08-06 22:57:11 +02:00
|
|
|
&options.pubkey_authentication,
|
|
|
|
NULL},
|
2000-10-14 07:23:11 +02:00
|
|
|
{"keyboard-interactive",
|
|
|
|
userauth_kbdint,
|
2008-11-05 06:20:46 +01:00
|
|
|
NULL,
|
2000-10-14 07:23:11 +02:00
|
|
|
&options.kbd_interactive_authentication,
|
|
|
|
&options.batch_mode},
|
2001-08-06 22:57:11 +02:00
|
|
|
{"password",
|
|
|
|
userauth_passwd,
|
2008-11-05 06:20:46 +01:00
|
|
|
NULL,
|
2001-08-06 22:57:11 +02:00
|
|
|
&options.password_authentication,
|
|
|
|
&options.batch_mode},
|
2000-10-14 07:23:11 +02:00
|
|
|
{"none",
|
|
|
|
userauth_none,
|
|
|
|
NULL,
|
2008-11-05 06:20:46 +01:00
|
|
|
NULL,
|
2000-10-14 07:23:11 +02:00
|
|
|
NULL},
|
2008-11-05 06:20:46 +01:00
|
|
|
{NULL, NULL, NULL, NULL, NULL}
|
2000-09-23 08:15:56 +02:00
|
|
|
};
|
|
|
|
|
|
|
|
void
|
2018-12-27 04:25:24 +01:00
|
|
|
ssh_userauth2(struct ssh *ssh, const char *local_user,
|
|
|
|
const char *server_user, char *host, Sensitive *sensitive)
|
2000-09-23 08:15:56 +02:00
|
|
|
{
|
|
|
|
Authctxt authctxt;
|
2015-12-04 17:41:28 +01:00
|
|
|
int r;
|
2000-09-23 08:15:56 +02:00
|
|
|
|
2001-03-11 02:49:19 +01:00
|
|
|
if (options.preferred_authentications == NULL)
|
|
|
|
options.preferred_authentications = authmethods_get();
|
|
|
|
|
2000-09-23 08:15:56 +02:00
|
|
|
/* setup authentication context */
|
2001-09-12 20:29:00 +02:00
|
|
|
memset(&authctxt, 0, sizeof(authctxt));
|
2000-09-23 08:15:56 +02:00
|
|
|
authctxt.server_user = server_user;
|
2001-04-13 01:34:34 +02:00
|
|
|
authctxt.local_user = local_user;
|
2000-09-23 08:15:56 +02:00
|
|
|
authctxt.host = host;
|
|
|
|
authctxt.service = "ssh-connection"; /* service name */
|
|
|
|
authctxt.success = 0;
|
2000-10-14 07:23:11 +02:00
|
|
|
authctxt.method = authmethod_lookup("none");
|
2001-03-09 01:12:22 +01:00
|
|
|
authctxt.authlist = NULL;
|
2003-08-26 03:49:55 +02:00
|
|
|
authctxt.methoddata = NULL;
|
2002-06-06 21:57:33 +02:00
|
|
|
authctxt.sensitive = sensitive;
|
2015-01-30 12:43:14 +01:00
|
|
|
authctxt.active_ktype = authctxt.oktypes = authctxt.ktypes = NULL;
|
2001-09-12 20:29:00 +02:00
|
|
|
authctxt.info_req_seen = 0;
|
2019-01-04 04:27:50 +01:00
|
|
|
authctxt.attempt_kbdint = 0;
|
|
|
|
authctxt.attempt_passwd = 0;
|
2019-02-11 10:44:42 +01:00
|
|
|
#if GSSAPI
|
|
|
|
authctxt.gss_supported_mechs = NULL;
|
|
|
|
authctxt.mech_tried = 0;
|
|
|
|
#endif
|
2015-01-15 12:04:36 +01:00
|
|
|
authctxt.agent_fd = -1;
|
2023-12-18 15:45:49 +01:00
|
|
|
if (authctxt.method == NULL)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_f("internal error: cannot send userauth none request");
|
2000-09-23 08:15:56 +02:00
|
|
|
|
2015-12-04 17:41:28 +01:00
|
|
|
if ((r = sshpkt_start(ssh, SSH2_MSG_SERVICE_REQUEST)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, "ssh-userauth")) != 0 ||
|
|
|
|
(r = sshpkt_send(ssh)) != 0)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_fr(r, "send packet");
|
2000-09-23 08:15:56 +02:00
|
|
|
|
2017-05-30 16:19:15 +02:00
|
|
|
ssh->authctxt = &authctxt;
|
2015-12-04 17:41:28 +01:00
|
|
|
ssh_dispatch_init(ssh, &input_userauth_error);
|
2023-12-18 18:04:18 +01:00
|
|
|
ssh_dispatch_set(ssh, SSH2_MSG_EXT_INFO, kex_input_ext_info);
|
2015-12-04 17:41:28 +01:00
|
|
|
ssh_dispatch_set(ssh, SSH2_MSG_SERVICE_ACCEPT, &input_userauth_service_accept);
|
2017-05-31 09:00:13 +02:00
|
|
|
ssh_dispatch_run_fatal(ssh, DISPATCH_BLOCK, &authctxt.success); /* loop until success */
|
2019-02-13 00:53:10 +01:00
|
|
|
pubkey_cleanup(ssh);
|
2023-10-12 04:15:53 +02:00
|
|
|
#ifdef GSSAPI
|
|
|
|
if (authctxt.gss_supported_mechs != NULL) {
|
|
|
|
u_int ms;
|
|
|
|
|
|
|
|
gss_release_oid_set(&ms, &authctxt.gss_supported_mechs);
|
|
|
|
authctxt.gss_supported_mechs = NULL;
|
|
|
|
}
|
|
|
|
#endif
|
2017-05-30 16:19:15 +02:00
|
|
|
ssh->authctxt = NULL;
|
2000-09-23 08:15:56 +02:00
|
|
|
|
2015-12-04 17:41:28 +01:00
|
|
|
ssh_dispatch_range(ssh, SSH2_MSG_USERAUTH_MIN, SSH2_MSG_USERAUTH_MAX, NULL);
|
2003-05-15 04:01:28 +02:00
|
|
|
|
2021-10-11 23:46:32 +02:00
|
|
|
#ifdef WINDOWS
|
|
|
|
send_auth_telemetry(authctxt.success, authctxt.success ? authctxt.method->name : "NULL");
|
|
|
|
#endif
|
2016-09-22 04:29:57 +02:00
|
|
|
if (!authctxt.success)
|
|
|
|
fatal("Authentication failed.");
|
2021-07-23 07:24:02 +02:00
|
|
|
if (ssh_packet_connection_is_on_socket(ssh)) {
|
|
|
|
verbose("Authenticated to %s ([%s]:%d) using \"%s\".", host,
|
|
|
|
ssh_remote_ipaddr(ssh), ssh_remote_port(ssh),
|
|
|
|
authctxt.method->name);
|
|
|
|
} else {
|
|
|
|
verbose("Authenticated to %s (via proxy) using \"%s\".", host,
|
|
|
|
authctxt.method->name);
|
|
|
|
}
|
2000-09-23 08:15:56 +02:00
|
|
|
}
|
2003-05-15 04:01:28 +02:00
|
|
|
|
2019-02-11 10:44:42 +01:00
|
|
|
static int
|
2017-05-30 16:23:52 +02:00
|
|
|
input_userauth_service_accept(int type, u_int32_t seq, struct ssh *ssh)
|
2015-12-04 17:41:28 +01:00
|
|
|
{
|
|
|
|
int r;
|
|
|
|
|
|
|
|
if (ssh_packet_remaining(ssh) > 0) {
|
|
|
|
char *reply;
|
|
|
|
|
|
|
|
if ((r = sshpkt_get_cstring(ssh, &reply, NULL)) != 0)
|
|
|
|
goto out;
|
|
|
|
debug2("service_accept: %s", reply);
|
|
|
|
free(reply);
|
|
|
|
} else {
|
|
|
|
debug2("buggy server: service_accept w/o service");
|
|
|
|
}
|
|
|
|
if ((r = sshpkt_get_end(ssh)) != 0)
|
|
|
|
goto out;
|
|
|
|
debug("SSH2_MSG_SERVICE_ACCEPT received");
|
|
|
|
|
|
|
|
/* initial userauth request */
|
2019-01-19 22:40:21 +01:00
|
|
|
userauth_none(ssh);
|
2015-12-04 17:41:28 +01:00
|
|
|
|
2023-12-18 15:45:49 +01:00
|
|
|
/* accept EXT_INFO at any time during userauth */
|
|
|
|
ssh_dispatch_set(ssh, SSH2_MSG_EXT_INFO, ssh->kex->ext_info_s ?
|
|
|
|
&kex_input_ext_info : &input_userauth_error);
|
2015-12-04 17:41:28 +01:00
|
|
|
ssh_dispatch_set(ssh, SSH2_MSG_USERAUTH_SUCCESS, &input_userauth_success);
|
|
|
|
ssh_dispatch_set(ssh, SSH2_MSG_USERAUTH_FAILURE, &input_userauth_failure);
|
|
|
|
ssh_dispatch_set(ssh, SSH2_MSG_USERAUTH_BANNER, &input_userauth_banner);
|
|
|
|
r = 0;
|
|
|
|
out:
|
|
|
|
return r;
|
|
|
|
}
|
|
|
|
|
2000-09-23 08:15:56 +02:00
|
|
|
void
|
2019-01-19 22:40:21 +01:00
|
|
|
userauth(struct ssh *ssh, char *authlist)
|
2001-03-09 01:12:22 +01:00
|
|
|
{
|
2019-01-19 22:40:21 +01:00
|
|
|
Authctxt *authctxt = (Authctxt *)ssh->authctxt;
|
2018-07-09 23:03:30 +02:00
|
|
|
|
2008-11-05 06:20:46 +01:00
|
|
|
if (authctxt->method != NULL && authctxt->method->cleanup != NULL)
|
2019-01-19 22:40:21 +01:00
|
|
|
authctxt->method->cleanup(ssh);
|
2008-11-05 06:20:46 +01:00
|
|
|
|
2013-06-01 23:31:17 +02:00
|
|
|
free(authctxt->methoddata);
|
|
|
|
authctxt->methoddata = NULL;
|
2001-03-09 01:12:22 +01:00
|
|
|
if (authlist == NULL) {
|
|
|
|
authlist = authctxt->authlist;
|
|
|
|
} else {
|
2013-06-01 23:31:17 +02:00
|
|
|
free(authctxt->authlist);
|
2001-03-09 01:12:22 +01:00
|
|
|
authctxt->authlist = authlist;
|
|
|
|
}
|
|
|
|
for (;;) {
|
|
|
|
Authmethod *method = authmethod_get(authlist);
|
|
|
|
if (method == NULL)
|
2017-06-14 02:31:38 +02:00
|
|
|
fatal("%s@%s: Permission denied (%s).",
|
|
|
|
authctxt->server_user, authctxt->host, authlist);
|
2001-03-09 01:12:22 +01:00
|
|
|
authctxt->method = method;
|
2003-05-15 04:01:28 +02:00
|
|
|
|
|
|
|
/* reset the per method handler */
|
2018-07-09 23:03:30 +02:00
|
|
|
ssh_dispatch_range(ssh, SSH2_MSG_USERAUTH_PER_METHOD_MIN,
|
2003-05-15 04:01:28 +02:00
|
|
|
SSH2_MSG_USERAUTH_PER_METHOD_MAX, NULL);
|
|
|
|
|
|
|
|
/* and try new method */
|
2019-01-19 22:40:21 +01:00
|
|
|
if (method->userauth(ssh) != 0) {
|
2001-03-09 01:12:22 +01:00
|
|
|
debug2("we sent a %s packet, wait for reply", method->name);
|
|
|
|
break;
|
|
|
|
} else {
|
|
|
|
debug2("we did not send a packet, disable method");
|
|
|
|
method->enabled = NULL;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
2002-06-23 23:23:20 +02:00
|
|
|
|
2019-02-11 10:44:42 +01:00
|
|
|
static int
|
2017-05-30 16:23:52 +02:00
|
|
|
input_userauth_error(int type, u_int32_t seq, struct ssh *ssh)
|
2000-09-23 08:15:56 +02:00
|
|
|
{
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_f("bad message during authentication: type %d", type);
|
2015-01-19 21:07:45 +01:00
|
|
|
return 0;
|
2001-01-06 16:18:16 +01:00
|
|
|
}
|
2002-06-23 23:23:20 +02:00
|
|
|
|
2019-02-11 10:44:42 +01:00
|
|
|
static int
|
2017-05-30 16:23:52 +02:00
|
|
|
input_userauth_banner(int type, u_int32_t seq, struct ssh *ssh)
|
2001-01-06 16:18:16 +01:00
|
|
|
{
|
2019-01-19 22:40:21 +01:00
|
|
|
char *msg = NULL;
|
2019-01-19 22:34:45 +01:00
|
|
|
size_t len;
|
|
|
|
int r;
|
2003-10-08 09:37:58 +02:00
|
|
|
|
2020-10-18 13:32:01 +02:00
|
|
|
debug3_f("entering");
|
2019-01-19 22:34:45 +01:00
|
|
|
if ((r = sshpkt_get_cstring(ssh, &msg, &len)) != 0 ||
|
2019-01-19 22:40:21 +01:00
|
|
|
(r = sshpkt_get_cstring(ssh, NULL, NULL)) != 0)
|
2019-01-19 22:34:45 +01:00
|
|
|
goto out;
|
2016-07-17 06:20:16 +02:00
|
|
|
if (len > 0 && options.log_level >= SYSLOG_LEVEL_INFO)
|
|
|
|
fmprintf(stderr, "%s", msg);
|
2019-01-19 22:34:45 +01:00
|
|
|
r = 0;
|
|
|
|
out:
|
2016-07-17 06:20:16 +02:00
|
|
|
free(msg);
|
2019-01-19 22:34:45 +01:00
|
|
|
return r;
|
2000-09-23 08:15:56 +02:00
|
|
|
}
|
2002-06-23 23:23:20 +02:00
|
|
|
|
2019-02-11 10:44:42 +01:00
|
|
|
static int
|
2017-05-30 16:23:52 +02:00
|
|
|
input_userauth_success(int type, u_int32_t seq, struct ssh *ssh)
|
2000-09-23 08:15:56 +02:00
|
|
|
{
|
2017-05-30 16:19:15 +02:00
|
|
|
Authctxt *authctxt = ssh->authctxt;
|
2009-10-24 02:47:58 +02:00
|
|
|
|
2000-09-23 08:15:56 +02:00
|
|
|
if (authctxt == NULL)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_f("no authentication context");
|
2013-06-01 23:31:17 +02:00
|
|
|
free(authctxt->authlist);
|
|
|
|
authctxt->authlist = NULL;
|
2009-10-24 02:47:58 +02:00
|
|
|
if (authctxt->method != NULL && authctxt->method->cleanup != NULL)
|
2019-01-19 22:40:21 +01:00
|
|
|
authctxt->method->cleanup(ssh);
|
2013-06-01 23:31:17 +02:00
|
|
|
free(authctxt->methoddata);
|
|
|
|
authctxt->methoddata = NULL;
|
2000-09-23 08:15:56 +02:00
|
|
|
authctxt->success = 1; /* break out */
|
2023-12-18 18:04:18 +01:00
|
|
|
ssh_dispatch_set(ssh, SSH2_MSG_EXT_INFO, dispatch_protocol_error);
|
2015-01-19 21:07:45 +01:00
|
|
|
return 0;
|
2000-09-23 08:15:56 +02:00
|
|
|
}
|
2002-06-23 23:23:20 +02:00
|
|
|
|
2019-02-11 10:44:42 +01:00
|
|
|
#if 0
|
|
|
|
static int
|
2017-05-30 16:23:52 +02:00
|
|
|
input_userauth_success_unexpected(int type, u_int32_t seq, struct ssh *ssh)
|
2009-10-24 02:47:58 +02:00
|
|
|
{
|
2017-05-30 16:19:15 +02:00
|
|
|
Authctxt *authctxt = ssh->authctxt;
|
2009-10-24 02:47:58 +02:00
|
|
|
|
|
|
|
if (authctxt == NULL)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_f("no authentication context");
|
2009-10-24 02:47:58 +02:00
|
|
|
|
|
|
|
fatal("Unexpected authentication success during %s.",
|
|
|
|
authctxt->method->name);
|
2015-01-19 21:07:45 +01:00
|
|
|
return 0;
|
2009-10-24 02:47:58 +02:00
|
|
|
}
|
2019-02-11 10:44:42 +01:00
|
|
|
#endif
|
2009-10-24 02:47:58 +02:00
|
|
|
|
2019-02-11 10:44:42 +01:00
|
|
|
static int
|
2017-05-30 16:23:52 +02:00
|
|
|
input_userauth_failure(int type, u_int32_t seq, struct ssh *ssh)
|
2000-09-23 08:15:56 +02:00
|
|
|
{
|
2017-05-30 16:19:15 +02:00
|
|
|
Authctxt *authctxt = ssh->authctxt;
|
2000-09-23 08:15:56 +02:00
|
|
|
char *authlist = NULL;
|
2018-07-09 23:03:30 +02:00
|
|
|
u_char partial;
|
2000-09-23 08:15:56 +02:00
|
|
|
|
|
|
|
if (authctxt == NULL)
|
|
|
|
fatal("input_userauth_failure: no authentication context");
|
|
|
|
|
2019-07-07 03:05:00 +02:00
|
|
|
if (sshpkt_get_cstring(ssh, &authlist, NULL) != 0 ||
|
|
|
|
sshpkt_get_u8(ssh, &partial) != 0 ||
|
|
|
|
sshpkt_get_end(ssh) != 0)
|
2018-07-09 23:03:30 +02:00
|
|
|
goto out;
|
2000-09-23 08:15:56 +02:00
|
|
|
|
2013-04-23 07:15:49 +02:00
|
|
|
if (partial != 0) {
|
2021-07-23 07:24:02 +02:00
|
|
|
verbose("Authenticated using \"%s\" with partial success.",
|
|
|
|
authctxt->method->name);
|
2013-04-23 07:15:49 +02:00
|
|
|
/* reset state */
|
2016-12-05 00:54:02 +01:00
|
|
|
pubkey_reset(authctxt);
|
2013-04-23 07:15:49 +02:00
|
|
|
}
|
2002-12-23 03:44:36 +01:00
|
|
|
debug("Authentications that can continue: %s", authlist);
|
2000-09-23 08:15:56 +02:00
|
|
|
|
2019-01-19 22:40:21 +01:00
|
|
|
userauth(ssh, authlist);
|
2018-07-09 23:03:30 +02:00
|
|
|
authlist = NULL;
|
|
|
|
out:
|
|
|
|
free(authlist);
|
2015-01-19 21:07:45 +01:00
|
|
|
return 0;
|
2001-03-09 01:12:22 +01:00
|
|
|
}
|
2008-11-03 09:26:18 +01:00
|
|
|
|
2018-09-14 07:26:27 +02:00
|
|
|
/*
|
|
|
|
* Format an identity for logging including filename, key type, fingerprint
|
|
|
|
* and location (agent, etc.). Caller must free.
|
|
|
|
*/
|
|
|
|
static char *
|
|
|
|
format_identity(Identity *id)
|
|
|
|
{
|
|
|
|
char *fp = NULL, *ret = NULL;
|
2019-10-31 22:18:28 +01:00
|
|
|
const char *note = "";
|
2018-09-14 07:26:27 +02:00
|
|
|
|
|
|
|
if (id->key != NULL) {
|
2021-04-03 08:18:40 +02:00
|
|
|
fp = sshkey_fingerprint(id->key, options.fingerprint_hash,
|
2018-09-14 07:26:27 +02:00
|
|
|
SSH_FP_DEFAULT);
|
|
|
|
}
|
2019-10-31 22:18:28 +01:00
|
|
|
if (id->key) {
|
|
|
|
if ((id->key->flags & SSHKEY_FLAG_EXT) != 0)
|
|
|
|
note = " token";
|
2019-11-12 20:33:08 +01:00
|
|
|
else if (sshkey_is_sk(id->key))
|
2020-02-06 23:30:54 +01:00
|
|
|
note = " authenticator";
|
2019-10-31 22:18:28 +01:00
|
|
|
}
|
2018-09-14 07:26:27 +02:00
|
|
|
xasprintf(&ret, "%s %s%s%s%s%s%s",
|
|
|
|
id->filename,
|
|
|
|
id->key ? sshkey_type(id->key) : "", id->key ? " " : "",
|
|
|
|
fp ? fp : "",
|
2019-10-31 22:18:28 +01:00
|
|
|
id->userprovided ? " explicit" : "", note,
|
2018-09-14 07:26:27 +02:00
|
|
|
id->agent_fd != -1 ? " agent" : "");
|
|
|
|
free(fp);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2019-02-11 10:44:42 +01:00
|
|
|
static int
|
2017-05-30 16:23:52 +02:00
|
|
|
input_userauth_pk_ok(int type, u_int32_t seq, struct ssh *ssh)
|
2001-03-09 01:12:22 +01:00
|
|
|
{
|
2017-05-30 16:19:15 +02:00
|
|
|
Authctxt *authctxt = ssh->authctxt;
|
2017-05-30 10:52:19 +02:00
|
|
|
struct sshkey *key = NULL;
|
2003-05-14 05:46:00 +02:00
|
|
|
Identity *id = NULL;
|
2018-09-14 07:26:27 +02:00
|
|
|
int pktype, found = 0, sent = 0;
|
2018-07-09 23:03:30 +02:00
|
|
|
size_t blen;
|
2018-09-14 07:26:27 +02:00
|
|
|
char *pkalg = NULL, *fp = NULL, *ident = NULL;
|
2018-07-09 23:03:30 +02:00
|
|
|
u_char *pkblob = NULL;
|
|
|
|
int r;
|
2001-03-09 01:12:22 +01:00
|
|
|
|
|
|
|
if (authctxt == NULL)
|
2021-10-11 23:46:32 +02:00
|
|
|
#ifdef WINDOWS
|
|
|
|
{
|
|
|
|
send_pubkey_telemetry("input_userauth_pk_ok: no authentication context");
|
2001-03-09 01:12:22 +01:00
|
|
|
fatal("input_userauth_pk_ok: no authentication context");
|
2021-10-11 23:46:32 +02:00
|
|
|
}
|
|
|
|
#else
|
|
|
|
fatal("input_userauth_pk_ok: no authentication context");
|
|
|
|
#endif
|
2018-01-23 06:27:21 +01:00
|
|
|
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshpkt_get_cstring(ssh, &pkalg, NULL)) != 0 ||
|
|
|
|
(r = sshpkt_get_string(ssh, &pkblob, &blen)) != 0 ||
|
|
|
|
(r = sshpkt_get_end(ssh)) != 0)
|
2021-10-11 23:46:32 +02:00
|
|
|
#ifdef WINDOWS
|
|
|
|
{
|
|
|
|
send_pubkey_telemetry("failure");
|
|
|
|
goto done;
|
|
|
|
}
|
|
|
|
#else
|
2018-07-09 23:03:30 +02:00
|
|
|
goto done;
|
2021-10-11 23:46:32 +02:00
|
|
|
#endif
|
2001-03-09 01:12:22 +01:00
|
|
|
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((pktype = sshkey_type_from_name(pkalg)) == KEY_UNSPEC) {
|
2021-10-11 23:46:32 +02:00
|
|
|
#ifdef WINDOWS
|
|
|
|
send_pubkey_telemetry("server sent unknown pkalg");
|
|
|
|
#endif
|
2020-10-18 13:32:01 +02:00
|
|
|
debug_f("server sent unknown pkalg %s", pkalg);
|
2003-05-14 05:46:00 +02:00
|
|
|
goto done;
|
|
|
|
}
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshkey_from_blob(pkblob, blen, &key)) != 0) {
|
2021-10-11 23:46:32 +02:00
|
|
|
#ifdef WINDOWS
|
|
|
|
send_pubkey_telemetry("no key from blob");
|
|
|
|
#endif
|
2020-10-18 13:32:01 +02:00
|
|
|
debug_r(r, "no key from blob. pkalg %s", pkalg);
|
2003-05-14 05:46:00 +02:00
|
|
|
goto done;
|
|
|
|
}
|
|
|
|
if (key->type != pktype) {
|
2021-10-11 23:46:32 +02:00
|
|
|
#ifdef WINDOWS
|
|
|
|
send_pubkey_telemetry("type mistmatch for decoded key");
|
|
|
|
#endif
|
2003-05-14 05:46:00 +02:00
|
|
|
error("input_userauth_pk_ok: type mismatch "
|
|
|
|
"for decoded key (received %d, expected %d)",
|
|
|
|
key->type, pktype);
|
|
|
|
goto done;
|
|
|
|
}
|
|
|
|
|
2003-10-15 07:55:59 +02:00
|
|
|
/*
|
|
|
|
* search keys in the reverse order, because last candidate has been
|
|
|
|
* moved to the end of the queue. this also avoids confusion by
|
|
|
|
* duplicate keys
|
|
|
|
*/
|
2004-04-20 12:07:19 +02:00
|
|
|
TAILQ_FOREACH_REVERSE(id, &authctxt->keys, idlist, next) {
|
2018-07-09 23:03:30 +02:00
|
|
|
if (sshkey_equal(key, id->key)) {
|
2018-09-14 07:26:27 +02:00
|
|
|
found = 1;
|
2001-03-09 01:12:22 +01:00
|
|
|
break;
|
|
|
|
}
|
2003-05-14 05:46:00 +02:00
|
|
|
}
|
2018-09-14 07:26:27 +02:00
|
|
|
if (!found || id == NULL) {
|
|
|
|
fp = sshkey_fingerprint(key, options.fingerprint_hash,
|
|
|
|
SSH_FP_DEFAULT);
|
2020-10-18 13:32:01 +02:00
|
|
|
error_f("server replied with unknown key: %s %s",
|
2018-09-14 07:26:27 +02:00
|
|
|
sshkey_type(key), fp == NULL ? "<ERROR>" : fp);
|
|
|
|
goto done;
|
|
|
|
}
|
|
|
|
ident = format_identity(id);
|
|
|
|
debug("Server accepts key: %s", ident);
|
2021-10-11 23:46:32 +02:00
|
|
|
#ifdef WINDOWS
|
|
|
|
send_pubkey_telemetry("success");
|
|
|
|
#endif
|
2019-01-19 22:40:21 +01:00
|
|
|
sent = sign_and_send_pubkey(ssh, id);
|
2018-07-09 23:03:30 +02:00
|
|
|
r = 0;
|
|
|
|
done:
|
|
|
|
sshkey_free(key);
|
2018-09-14 07:26:27 +02:00
|
|
|
free(ident);
|
|
|
|
free(fp);
|
2013-06-01 23:31:17 +02:00
|
|
|
free(pkalg);
|
|
|
|
free(pkblob);
|
2001-03-09 01:12:22 +01:00
|
|
|
|
2002-07-04 02:14:17 +02:00
|
|
|
/* try another method if we did not send a packet */
|
2018-07-09 23:03:30 +02:00
|
|
|
if (r == 0 && sent == 0)
|
2019-01-19 22:40:21 +01:00
|
|
|
userauth(ssh, NULL);
|
2018-07-09 23:03:30 +02:00
|
|
|
return r;
|
2000-09-23 08:15:56 +02:00
|
|
|
}
|
|
|
|
|
2003-08-26 03:49:55 +02:00
|
|
|
#ifdef GSSAPI
|
2019-02-11 10:44:42 +01:00
|
|
|
static int
|
2019-01-19 22:40:21 +01:00
|
|
|
userauth_gssapi(struct ssh *ssh)
|
2003-08-26 03:49:55 +02:00
|
|
|
{
|
2019-01-19 22:40:21 +01:00
|
|
|
Authctxt *authctxt = (Authctxt *)ssh->authctxt;
|
2003-08-26 03:49:55 +02:00
|
|
|
Gssctxt *gssctxt = NULL;
|
|
|
|
OM_uint32 min;
|
2018-07-09 23:03:30 +02:00
|
|
|
int r, ok = 0;
|
2019-02-11 10:44:42 +01:00
|
|
|
gss_OID mech = NULL;
|
2003-08-26 03:49:55 +02:00
|
|
|
|
|
|
|
/* Try one GSSAPI method at a time, rather than sending them all at
|
|
|
|
* once. */
|
|
|
|
|
2019-02-11 10:44:42 +01:00
|
|
|
if (authctxt->gss_supported_mechs == NULL)
|
|
|
|
gss_indicate_mechs(&min, &authctxt->gss_supported_mechs);
|
2003-08-26 03:49:55 +02:00
|
|
|
|
2019-02-11 10:44:42 +01:00
|
|
|
/* Check to see whether the mechanism is usable before we offer it */
|
|
|
|
while (authctxt->mech_tried < authctxt->gss_supported_mechs->count &&
|
|
|
|
!ok) {
|
|
|
|
mech = &authctxt->gss_supported_mechs->
|
|
|
|
elements[authctxt->mech_tried];
|
2003-08-26 03:49:55 +02:00
|
|
|
/* My DER encoding requires length<128 */
|
2019-02-11 10:44:42 +01:00
|
|
|
if (mech->length < 128 && ssh_gssapi_check_mechanism(&gssctxt,
|
|
|
|
mech, authctxt->host)) {
|
2003-08-26 03:49:55 +02:00
|
|
|
ok = 1; /* Mechanism works */
|
|
|
|
} else {
|
2019-02-11 10:44:42 +01:00
|
|
|
authctxt->mech_tried++;
|
2003-08-26 03:49:55 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2019-02-11 10:44:42 +01:00
|
|
|
if (!ok || mech == NULL)
|
2005-06-17 04:59:34 +02:00
|
|
|
return 0;
|
2003-08-26 03:49:55 +02:00
|
|
|
|
|
|
|
authctxt->methoddata=(void *)gssctxt;
|
|
|
|
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshpkt_start(ssh, SSH2_MSG_USERAUTH_REQUEST)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->server_user)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->service)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->method->name)) != 0 ||
|
|
|
|
(r = sshpkt_put_u32(ssh, 1)) != 0 ||
|
2019-02-11 10:44:42 +01:00
|
|
|
(r = sshpkt_put_u32(ssh, (mech->length) + 2)) != 0 ||
|
2018-07-09 23:03:30 +02:00
|
|
|
(r = sshpkt_put_u8(ssh, SSH_GSS_OIDTYPE)) != 0 ||
|
2019-02-11 10:44:42 +01:00
|
|
|
(r = sshpkt_put_u8(ssh, mech->length)) != 0 ||
|
|
|
|
(r = sshpkt_put(ssh, mech->elements, mech->length)) != 0 ||
|
2018-07-09 23:03:30 +02:00
|
|
|
(r = sshpkt_send(ssh)) != 0)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_fr(r, "send packet");
|
2003-08-26 03:49:55 +02:00
|
|
|
|
2018-07-09 23:03:30 +02:00
|
|
|
ssh_dispatch_set(ssh, SSH2_MSG_USERAUTH_GSSAPI_RESPONSE, &input_gssapi_response);
|
|
|
|
ssh_dispatch_set(ssh, SSH2_MSG_USERAUTH_GSSAPI_TOKEN, &input_gssapi_token);
|
|
|
|
ssh_dispatch_set(ssh, SSH2_MSG_USERAUTH_GSSAPI_ERROR, &input_gssapi_error);
|
|
|
|
ssh_dispatch_set(ssh, SSH2_MSG_USERAUTH_GSSAPI_ERRTOK, &input_gssapi_errtok);
|
2003-08-26 03:49:55 +02:00
|
|
|
|
2019-02-11 10:44:42 +01:00
|
|
|
authctxt->mech_tried++; /* Move along to next candidate */
|
2003-08-26 03:49:55 +02:00
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2019-02-11 10:44:42 +01:00
|
|
|
static void
|
|
|
|
userauth_gssapi_cleanup(struct ssh *ssh)
|
|
|
|
{
|
|
|
|
Authctxt *authctxt = (Authctxt *)ssh->authctxt;
|
|
|
|
Gssctxt *gssctxt = (Gssctxt *)authctxt->methoddata;
|
|
|
|
|
|
|
|
ssh_gssapi_delete_ctx(&gssctxt);
|
|
|
|
authctxt->methoddata = NULL;
|
|
|
|
}
|
|
|
|
|
2003-11-17 11:20:18 +01:00
|
|
|
static OM_uint32
|
2017-05-31 07:08:46 +02:00
|
|
|
process_gssapi_token(struct ssh *ssh, gss_buffer_t recv_tok)
|
2003-11-17 11:20:18 +01:00
|
|
|
{
|
2017-05-30 16:19:15 +02:00
|
|
|
Authctxt *authctxt = ssh->authctxt;
|
2003-11-17 11:20:18 +01:00
|
|
|
Gssctxt *gssctxt = authctxt->methoddata;
|
|
|
|
gss_buffer_desc send_tok = GSS_C_EMPTY_BUFFER;
|
2005-08-31 11:46:26 +02:00
|
|
|
gss_buffer_desc mic = GSS_C_EMPTY_BUFFER;
|
|
|
|
gss_buffer_desc gssbuf;
|
2003-11-17 12:18:21 +01:00
|
|
|
OM_uint32 status, ms, flags;
|
2018-07-09 23:03:30 +02:00
|
|
|
int r;
|
2003-11-21 13:56:47 +01:00
|
|
|
|
2003-11-17 11:20:18 +01:00
|
|
|
status = ssh_gssapi_init_ctx(gssctxt, options.gss_deleg_creds,
|
2003-11-17 12:18:21 +01:00
|
|
|
recv_tok, &send_tok, &flags);
|
2003-11-17 11:20:18 +01:00
|
|
|
|
|
|
|
if (send_tok.length > 0) {
|
2018-07-09 23:03:30 +02:00
|
|
|
u_char type = GSS_ERROR(status) ?
|
|
|
|
SSH2_MSG_USERAUTH_GSSAPI_ERRTOK :
|
|
|
|
SSH2_MSG_USERAUTH_GSSAPI_TOKEN;
|
|
|
|
|
|
|
|
if ((r = sshpkt_start(ssh, type)) != 0 ||
|
|
|
|
(r = sshpkt_put_string(ssh, send_tok.value,
|
|
|
|
send_tok.length)) != 0 ||
|
|
|
|
(r = sshpkt_send(ssh)) != 0)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_fr(r, "send %u packet", type);
|
2003-11-21 13:56:47 +01:00
|
|
|
|
2003-11-17 11:20:18 +01:00
|
|
|
gss_release_buffer(&ms, &send_tok);
|
|
|
|
}
|
2003-11-21 13:56:47 +01:00
|
|
|
|
2003-11-17 11:20:18 +01:00
|
|
|
if (status == GSS_S_COMPLETE) {
|
2003-11-17 12:18:21 +01:00
|
|
|
/* send either complete or MIC, depending on mechanism */
|
|
|
|
if (!(flags & GSS_C_INTEG_FLAG)) {
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshpkt_start(ssh,
|
|
|
|
SSH2_MSG_USERAUTH_GSSAPI_EXCHANGE_COMPLETE)) != 0 ||
|
|
|
|
(r = sshpkt_send(ssh)) != 0)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_fr(r, "send completion");
|
2003-11-17 12:18:21 +01:00
|
|
|
} else {
|
2018-07-09 23:03:30 +02:00
|
|
|
struct sshbuf *b;
|
|
|
|
|
|
|
|
if ((b = sshbuf_new()) == NULL)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_f("sshbuf_new failed");
|
2018-07-09 23:03:30 +02:00
|
|
|
ssh_gssapi_buildmic(b, authctxt->server_user,
|
2021-01-27 11:05:28 +01:00
|
|
|
authctxt->service, "gssapi-with-mic",
|
|
|
|
ssh->kex->session_id);
|
2003-11-17 12:18:21 +01:00
|
|
|
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((gssbuf.value = sshbuf_mutable_ptr(b)) == NULL)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_f("sshbuf_mutable_ptr failed");
|
2018-07-09 23:03:30 +02:00
|
|
|
gssbuf.length = sshbuf_len(b);
|
2003-11-21 13:56:47 +01:00
|
|
|
|
2003-11-17 12:18:21 +01:00
|
|
|
status = ssh_gssapi_sign(gssctxt, &gssbuf, &mic);
|
2003-11-21 13:56:47 +01:00
|
|
|
|
2003-11-17 12:18:21 +01:00
|
|
|
if (!GSS_ERROR(status)) {
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshpkt_start(ssh,
|
|
|
|
SSH2_MSG_USERAUTH_GSSAPI_MIC)) != 0 ||
|
|
|
|
(r = sshpkt_put_string(ssh, mic.value,
|
|
|
|
mic.length)) != 0 ||
|
|
|
|
(r = sshpkt_send(ssh)) != 0)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_fr(r, "send MIC");
|
2003-11-17 12:18:21 +01:00
|
|
|
}
|
2003-11-21 13:56:47 +01:00
|
|
|
|
2018-07-09 23:03:30 +02:00
|
|
|
sshbuf_free(b);
|
2003-11-17 12:18:21 +01:00
|
|
|
gss_release_buffer(&ms, &mic);
|
2003-11-21 13:56:47 +01:00
|
|
|
}
|
2003-11-17 11:20:18 +01:00
|
|
|
}
|
2003-11-21 13:56:47 +01:00
|
|
|
|
2003-11-17 11:20:18 +01:00
|
|
|
return status;
|
|
|
|
}
|
|
|
|
|
2019-02-11 10:44:42 +01:00
|
|
|
static int
|
2017-05-30 16:23:52 +02:00
|
|
|
input_gssapi_response(int type, u_int32_t plen, struct ssh *ssh)
|
2003-08-26 03:49:55 +02:00
|
|
|
{
|
2017-05-30 16:19:15 +02:00
|
|
|
Authctxt *authctxt = ssh->authctxt;
|
2003-08-26 03:49:55 +02:00
|
|
|
Gssctxt *gssctxt;
|
2018-07-09 23:03:30 +02:00
|
|
|
size_t oidlen;
|
|
|
|
u_char *oidv = NULL;
|
|
|
|
int r;
|
2003-08-26 03:49:55 +02:00
|
|
|
|
|
|
|
if (authctxt == NULL)
|
|
|
|
fatal("input_gssapi_response: no authentication context");
|
|
|
|
gssctxt = authctxt->methoddata;
|
|
|
|
|
|
|
|
/* Setup our OID */
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshpkt_get_string(ssh, &oidv, &oidlen)) != 0)
|
|
|
|
goto done;
|
2003-08-26 03:49:55 +02:00
|
|
|
|
2003-11-03 10:09:03 +01:00
|
|
|
if (oidlen <= 2 ||
|
|
|
|
oidv[0] != SSH_GSS_OIDTYPE ||
|
|
|
|
oidv[1] != oidlen - 2) {
|
|
|
|
debug("Badly encoded mechanism OID received");
|
2019-01-21 02:05:00 +01:00
|
|
|
userauth(ssh, NULL);
|
2018-07-09 23:03:30 +02:00
|
|
|
goto ok;
|
2003-08-26 03:49:55 +02:00
|
|
|
}
|
|
|
|
|
2003-11-03 10:09:03 +01:00
|
|
|
if (!ssh_gssapi_check_oid(gssctxt, oidv + 2, oidlen - 2))
|
|
|
|
fatal("Server returned different OID than expected");
|
|
|
|
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshpkt_get_end(ssh)) != 0)
|
|
|
|
goto done;
|
2003-08-26 03:49:55 +02:00
|
|
|
|
2017-05-31 07:08:46 +02:00
|
|
|
if (GSS_ERROR(process_gssapi_token(ssh, GSS_C_NO_BUFFER))) {
|
2003-08-26 03:49:55 +02:00
|
|
|
/* Start again with next method on list */
|
|
|
|
debug("Trying to start again");
|
2019-01-21 02:05:00 +01:00
|
|
|
userauth(ssh, NULL);
|
2018-07-09 23:03:30 +02:00
|
|
|
goto ok;
|
2003-08-26 03:49:55 +02:00
|
|
|
}
|
2018-07-09 23:03:30 +02:00
|
|
|
ok:
|
|
|
|
r = 0;
|
|
|
|
done:
|
|
|
|
free(oidv);
|
|
|
|
return r;
|
2003-08-26 03:49:55 +02:00
|
|
|
}
|
|
|
|
|
2019-02-11 10:44:42 +01:00
|
|
|
static int
|
2017-05-30 16:23:52 +02:00
|
|
|
input_gssapi_token(int type, u_int32_t plen, struct ssh *ssh)
|
2003-08-26 03:49:55 +02:00
|
|
|
{
|
2017-05-30 16:19:15 +02:00
|
|
|
Authctxt *authctxt = ssh->authctxt;
|
2003-08-26 03:49:55 +02:00
|
|
|
gss_buffer_desc recv_tok;
|
2018-07-09 23:03:30 +02:00
|
|
|
u_char *p = NULL;
|
|
|
|
size_t len;
|
2003-11-17 11:20:18 +01:00
|
|
|
OM_uint32 status;
|
2018-07-09 23:03:30 +02:00
|
|
|
int r;
|
2003-08-26 03:49:55 +02:00
|
|
|
|
|
|
|
if (authctxt == NULL)
|
|
|
|
fatal("input_gssapi_response: no authentication context");
|
|
|
|
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshpkt_get_string(ssh, &p, &len)) != 0 ||
|
|
|
|
(r = sshpkt_get_end(ssh)) != 0)
|
|
|
|
goto out;
|
2003-08-26 03:49:55 +02:00
|
|
|
|
2018-07-09 23:03:30 +02:00
|
|
|
recv_tok.value = p;
|
|
|
|
recv_tok.length = len;
|
2017-05-31 07:08:46 +02:00
|
|
|
status = process_gssapi_token(ssh, &recv_tok);
|
2003-08-26 03:49:55 +02:00
|
|
|
|
2018-07-09 23:03:30 +02:00
|
|
|
/* Start again with the next method in the list */
|
2003-08-26 03:49:55 +02:00
|
|
|
if (GSS_ERROR(status)) {
|
2019-01-21 02:05:00 +01:00
|
|
|
userauth(ssh, NULL);
|
2018-07-09 23:03:30 +02:00
|
|
|
/* ok */
|
2003-08-26 03:49:55 +02:00
|
|
|
}
|
2018-07-09 23:03:30 +02:00
|
|
|
r = 0;
|
|
|
|
out:
|
|
|
|
free(p);
|
|
|
|
return r;
|
2003-08-26 03:49:55 +02:00
|
|
|
}
|
|
|
|
|
2019-02-11 10:44:42 +01:00
|
|
|
static int
|
2017-05-30 16:23:52 +02:00
|
|
|
input_gssapi_errtok(int type, u_int32_t plen, struct ssh *ssh)
|
2003-08-26 03:49:55 +02:00
|
|
|
{
|
2017-05-30 16:19:15 +02:00
|
|
|
Authctxt *authctxt = ssh->authctxt;
|
2003-08-26 03:49:55 +02:00
|
|
|
Gssctxt *gssctxt;
|
|
|
|
gss_buffer_desc send_tok = GSS_C_EMPTY_BUFFER;
|
|
|
|
gss_buffer_desc recv_tok;
|
2013-04-23 07:18:51 +02:00
|
|
|
OM_uint32 ms;
|
2018-07-09 23:03:30 +02:00
|
|
|
u_char *p = NULL;
|
|
|
|
size_t len;
|
|
|
|
int r;
|
2003-08-26 03:49:55 +02:00
|
|
|
|
|
|
|
if (authctxt == NULL)
|
|
|
|
fatal("input_gssapi_response: no authentication context");
|
|
|
|
gssctxt = authctxt->methoddata;
|
|
|
|
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshpkt_get_string(ssh, &p, &len)) != 0 ||
|
|
|
|
(r = sshpkt_get_end(ssh)) != 0) {
|
|
|
|
free(p);
|
|
|
|
return r;
|
|
|
|
}
|
2003-08-26 03:49:55 +02:00
|
|
|
|
|
|
|
/* Stick it into GSSAPI and see what it says */
|
2018-07-09 23:03:30 +02:00
|
|
|
recv_tok.value = p;
|
|
|
|
recv_tok.length = len;
|
2013-04-23 07:18:51 +02:00
|
|
|
(void)ssh_gssapi_init_ctx(gssctxt, options.gss_deleg_creds,
|
2005-07-17 09:22:45 +02:00
|
|
|
&recv_tok, &send_tok, NULL);
|
2018-07-09 23:03:30 +02:00
|
|
|
free(p);
|
2003-08-26 03:49:55 +02:00
|
|
|
gss_release_buffer(&ms, &send_tok);
|
|
|
|
|
|
|
|
/* Server will be returning a failed packet after this one */
|
2015-01-20 08:56:44 +01:00
|
|
|
return 0;
|
2003-08-26 03:49:55 +02:00
|
|
|
}
|
|
|
|
|
2019-02-11 10:44:42 +01:00
|
|
|
static int
|
2017-05-30 16:23:52 +02:00
|
|
|
input_gssapi_error(int type, u_int32_t plen, struct ssh *ssh)
|
2003-08-26 03:49:55 +02:00
|
|
|
{
|
2018-07-09 23:03:30 +02:00
|
|
|
char *msg = NULL;
|
|
|
|
char *lang = NULL;
|
|
|
|
int r;
|
2003-08-26 03:49:55 +02:00
|
|
|
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshpkt_get_u32(ssh, NULL)) != 0 || /* maj */
|
|
|
|
(r = sshpkt_get_u32(ssh, NULL)) != 0 || /* min */
|
|
|
|
(r = sshpkt_get_cstring(ssh, &msg, NULL)) != 0 ||
|
|
|
|
(r = sshpkt_get_cstring(ssh, &lang, NULL)) != 0)
|
|
|
|
goto out;
|
|
|
|
r = sshpkt_get_end(ssh);
|
2005-11-05 05:07:33 +01:00
|
|
|
debug("Server GSSAPI Error:\n%s", msg);
|
2018-07-09 23:03:30 +02:00
|
|
|
out:
|
2013-06-01 23:31:17 +02:00
|
|
|
free(msg);
|
|
|
|
free(lang);
|
2018-07-09 23:03:30 +02:00
|
|
|
return r;
|
2003-08-26 03:49:55 +02:00
|
|
|
}
|
|
|
|
#endif /* GSSAPI */
|
|
|
|
|
2019-02-11 10:44:42 +01:00
|
|
|
static int
|
2019-01-19 22:40:21 +01:00
|
|
|
userauth_none(struct ssh *ssh)
|
2000-10-14 07:23:11 +02:00
|
|
|
{
|
2019-01-19 22:40:21 +01:00
|
|
|
Authctxt *authctxt = (Authctxt *)ssh->authctxt;
|
2018-07-09 23:03:30 +02:00
|
|
|
int r;
|
|
|
|
|
2000-10-14 07:23:11 +02:00
|
|
|
/* initial userauth request */
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshpkt_start(ssh, SSH2_MSG_USERAUTH_REQUEST)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->server_user)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->service)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->method->name)) != 0 ||
|
|
|
|
(r = sshpkt_send(ssh)) != 0)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_fr(r, "send packet");
|
2000-10-14 07:23:11 +02:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2019-02-11 10:44:42 +01:00
|
|
|
static int
|
2019-01-19 22:40:21 +01:00
|
|
|
userauth_passwd(struct ssh *ssh)
|
2000-04-29 15:57:08 +02:00
|
|
|
{
|
2019-01-19 22:40:21 +01:00
|
|
|
Authctxt *authctxt = (Authctxt *)ssh->authctxt;
|
2018-11-28 07:00:38 +01:00
|
|
|
char *password, *prompt = NULL;
|
2010-01-08 08:48:02 +01:00
|
|
|
const char *host = options.host_key_alias ? options.host_key_alias :
|
|
|
|
authctxt->host;
|
2018-07-09 23:03:30 +02:00
|
|
|
int r;
|
2000-04-29 15:57:08 +02:00
|
|
|
|
2019-01-04 04:27:50 +01:00
|
|
|
if (authctxt->attempt_passwd++ >= options.number_of_password_prompts)
|
- Remove references to SSLeay.
- Big OpenBSD CVS update
- markus@cvs.openbsd.org
[clientloop.c]
- typo
[session.c]
- update proctitle on pty alloc/dealloc, e.g. w/ windows client
[session.c]
- update proctitle for proto 1, too
[channels.h nchan.c serverloop.c session.c sshd.c]
- use c-style comments
- deraadt@cvs.openbsd.org
[scp.c]
- more atomicio
- markus@cvs.openbsd.org
[channels.c]
- set O_NONBLOCK
[ssh.1]
- update AUTHOR
[readconf.c ssh-keygen.c ssh.h]
- default DSA key file ~/.ssh/id_dsa
[clientloop.c]
- typo, rm verbose debug
- deraadt@cvs.openbsd.org
[ssh-keygen.1]
- document DSA use of ssh-keygen
[sshd.8]
- a start at describing what i understand of the DSA side
[ssh-keygen.1]
- document -X and -x
[ssh-keygen.c]
- simplify usage
- markus@cvs.openbsd.org
[sshd.8]
- there is no rhosts_dsa
[ssh-keygen.1]
- document -y, update -X,-x
[nchan.c]
- fix close for non-open ssh1 channels
[servconf.c servconf.h ssh.h sshd.8 sshd.c ]
- s/DsaKey/HostDSAKey/, document option
[sshconnect2.c]
- respect number_of_password_prompts
[channels.c channels.h servconf.c servconf.h session.c sshd.8]
- GatewayPorts for sshd, ok deraadt@
[ssh-add.1 ssh-agent.1 ssh.1]
- more doc on: DSA, id_dsa, known_hosts2, authorized_keys2
[ssh.1]
- more info on proto 2
[sshd.8]
- sync AUTHOR w/ ssh.1
[key.c key.h sshconnect.c]
- print key type when talking about host keys
[packet.c]
- clear padding in ssh2
[dsa.c key.c radix.c ssh.h sshconnect1.c uuencode.c uuencode.h]
- replace broken uuencode w/ libc b64_ntop
[auth2.c]
- log failure before sending the reply
[key.c radix.c uuencode.c]
- remote trailing comments before calling __b64_pton
[auth2.c readconf.c readconf.h servconf.c servconf.h ssh.1]
[sshconnect2.c sshd.8]
- add DSAAuthetication option to ssh/sshd, document SSH2 in sshd.8
- Bring in b64_ntop and b64_pton from OpenBSD libc (bsd-base64.[ch])
2000-05-07 04:03:14 +02:00
|
|
|
return 0;
|
|
|
|
|
2019-01-04 04:27:50 +01:00
|
|
|
if (authctxt->attempt_passwd != 1)
|
2000-06-07 11:55:44 +02:00
|
|
|
error("Permission denied, please try again.");
|
|
|
|
|
2018-11-28 07:00:38 +01:00
|
|
|
xasprintf(&prompt, "%s@%s's password: ", authctxt->server_user, host);
|
2000-04-29 15:57:08 +02:00
|
|
|
password = read_passphrase(prompt, 0);
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshpkt_start(ssh, SSH2_MSG_USERAUTH_REQUEST)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->server_user)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->service)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->method->name)) != 0 ||
|
|
|
|
(r = sshpkt_put_u8(ssh, 0)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, password)) != 0 ||
|
|
|
|
(r = sshpkt_add_padding(ssh, 64)) != 0 ||
|
|
|
|
(r = sshpkt_send(ssh)) != 0)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_fr(r, "send packet");
|
2018-07-09 23:03:30 +02:00
|
|
|
|
2018-11-28 07:00:38 +01:00
|
|
|
free(prompt);
|
|
|
|
if (password != NULL)
|
2018-07-09 23:03:30 +02:00
|
|
|
freezero(password, strlen(password));
|
|
|
|
|
|
|
|
ssh_dispatch_set(ssh, SSH2_MSG_USERAUTH_PASSWD_CHANGEREQ,
|
2002-03-27 18:28:46 +01:00
|
|
|
&input_userauth_passwd_changereq);
|
|
|
|
|
2000-04-29 15:57:08 +02:00
|
|
|
return 1;
|
|
|
|
}
|
2008-11-03 09:26:18 +01:00
|
|
|
|
2002-03-27 18:28:46 +01:00
|
|
|
/*
|
|
|
|
* parse PASSWD_CHANGEREQ, prompt user and send SSH2_MSG_USERAUTH_REQUEST
|
|
|
|
*/
|
2019-02-11 10:44:42 +01:00
|
|
|
static int
|
2017-05-30 16:23:52 +02:00
|
|
|
input_userauth_passwd_changereq(int type, u_int32_t seqnr, struct ssh *ssh)
|
2002-03-27 18:28:46 +01:00
|
|
|
{
|
2017-05-30 16:19:15 +02:00
|
|
|
Authctxt *authctxt = ssh->authctxt;
|
2018-07-09 23:03:30 +02:00
|
|
|
char *info = NULL, *lang = NULL, *password = NULL, *retype = NULL;
|
2017-08-27 02:38:41 +02:00
|
|
|
char prompt[256];
|
2017-01-30 01:32:03 +01:00
|
|
|
const char *host;
|
2018-07-09 23:03:30 +02:00
|
|
|
int r;
|
2002-03-27 18:28:46 +01:00
|
|
|
|
|
|
|
debug2("input_userauth_passwd_changereq");
|
|
|
|
|
|
|
|
if (authctxt == NULL)
|
|
|
|
fatal("input_userauth_passwd_changereq: "
|
|
|
|
"no authentication context");
|
2017-01-30 01:32:03 +01:00
|
|
|
host = options.host_key_alias ? options.host_key_alias : authctxt->host;
|
2002-03-27 18:28:46 +01:00
|
|
|
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshpkt_get_cstring(ssh, &info, NULL)) != 0 ||
|
|
|
|
(r = sshpkt_get_cstring(ssh, &lang, NULL)) != 0)
|
|
|
|
goto out;
|
2002-03-27 18:28:46 +01:00
|
|
|
if (strlen(info) > 0)
|
2003-04-09 12:59:48 +02:00
|
|
|
logit("%s", info);
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshpkt_start(ssh, SSH2_MSG_USERAUTH_REQUEST)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->server_user)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->service)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->method->name)) != 0 ||
|
|
|
|
(r = sshpkt_put_u8(ssh, 1)) != 0) /* additional info */
|
|
|
|
goto out;
|
|
|
|
|
2002-06-21 02:41:51 +02:00
|
|
|
snprintf(prompt, sizeof(prompt),
|
2002-03-27 18:28:46 +01:00
|
|
|
"Enter %.30s@%.128s's old password: ",
|
2010-01-08 08:48:02 +01:00
|
|
|
authctxt->server_user, host);
|
2002-03-27 18:28:46 +01:00
|
|
|
password = read_passphrase(prompt, 0);
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshpkt_put_cstring(ssh, password)) != 0)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
freezero(password, strlen(password));
|
2002-03-27 18:28:46 +01:00
|
|
|
password = NULL;
|
|
|
|
while (password == NULL) {
|
2002-06-21 02:41:51 +02:00
|
|
|
snprintf(prompt, sizeof(prompt),
|
2002-03-27 18:28:46 +01:00
|
|
|
"Enter %.30s@%.128s's new password: ",
|
2010-01-08 08:48:02 +01:00
|
|
|
authctxt->server_user, host);
|
2002-03-27 18:28:46 +01:00
|
|
|
password = read_passphrase(prompt, RP_ALLOW_EOF);
|
|
|
|
if (password == NULL) {
|
|
|
|
/* bail out */
|
2018-07-09 23:03:30 +02:00
|
|
|
r = 0;
|
|
|
|
goto out;
|
2002-03-27 18:28:46 +01:00
|
|
|
}
|
2002-06-21 02:41:51 +02:00
|
|
|
snprintf(prompt, sizeof(prompt),
|
2002-03-27 18:28:46 +01:00
|
|
|
"Retype %.30s@%.128s's new password: ",
|
2010-01-08 08:48:02 +01:00
|
|
|
authctxt->server_user, host);
|
2002-03-27 18:28:46 +01:00
|
|
|
retype = read_passphrase(prompt, 0);
|
|
|
|
if (strcmp(password, retype) != 0) {
|
2018-07-09 23:03:30 +02:00
|
|
|
freezero(password, strlen(password));
|
2003-04-09 12:59:48 +02:00
|
|
|
logit("Mismatch; try again, EOF to quit.");
|
2002-03-27 18:28:46 +01:00
|
|
|
password = NULL;
|
|
|
|
}
|
2018-07-09 23:03:30 +02:00
|
|
|
freezero(retype, strlen(retype));
|
2002-03-27 18:28:46 +01:00
|
|
|
}
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshpkt_put_cstring(ssh, password)) != 0 ||
|
|
|
|
(r = sshpkt_add_padding(ssh, 64)) != 0 ||
|
|
|
|
(r = sshpkt_send(ssh)) != 0)
|
|
|
|
goto out;
|
2002-06-21 02:41:51 +02:00
|
|
|
|
2018-07-09 23:03:30 +02:00
|
|
|
ssh_dispatch_set(ssh, SSH2_MSG_USERAUTH_PASSWD_CHANGEREQ,
|
2002-03-27 18:28:46 +01:00
|
|
|
&input_userauth_passwd_changereq);
|
2018-07-09 23:03:30 +02:00
|
|
|
r = 0;
|
|
|
|
out:
|
|
|
|
if (password)
|
|
|
|
freezero(password, strlen(password));
|
|
|
|
free(info);
|
|
|
|
free(lang);
|
|
|
|
return r;
|
2002-03-27 18:28:46 +01:00
|
|
|
}
|
2000-04-29 15:57:08 +02:00
|
|
|
|
2018-03-03 04:01:50 +01:00
|
|
|
/*
|
2018-07-03 13:39:54 +02:00
|
|
|
* Select an algorithm for publickey signatures.
|
|
|
|
* Returns algorithm (caller must free) or NULL if no mutual algorithm found.
|
|
|
|
*
|
|
|
|
* Call with ssh==NULL to ignore server-sig-algs extension list and
|
|
|
|
* only attempt with the key's base signature type.
|
2018-03-03 04:01:50 +01:00
|
|
|
*/
|
2018-07-03 13:39:54 +02:00
|
|
|
static char *
|
|
|
|
key_sig_algorithm(struct ssh *ssh, const struct sshkey *key)
|
2018-03-03 04:01:50 +01:00
|
|
|
{
|
2018-07-03 15:20:25 +02:00
|
|
|
char *allowed, *oallowed, *cp, *tmp, *alg = NULL;
|
2021-06-06 05:40:39 +02:00
|
|
|
const char *server_sig_algs;
|
2018-03-03 04:01:50 +01:00
|
|
|
|
2018-07-03 13:39:54 +02:00
|
|
|
/*
|
|
|
|
* The signature algorithm will only differ from the key algorithm
|
|
|
|
* for RSA keys/certs and when the server advertises support for
|
|
|
|
* newer (SHA2) algorithms.
|
|
|
|
*/
|
|
|
|
if (ssh == NULL || ssh->kex->server_sig_algs == NULL ||
|
2018-10-11 05:48:04 +02:00
|
|
|
(key->type != KEY_RSA && key->type != KEY_RSA_CERT) ||
|
2021-01-27 10:26:53 +01:00
|
|
|
(key->type == KEY_RSA_CERT && (ssh->compat & SSH_BUG_SIGTYPE))) {
|
2018-07-03 13:39:54 +02:00
|
|
|
/* Filter base key signature alg against our configuration */
|
2018-07-11 20:53:29 +02:00
|
|
|
return match_list(sshkey_ssh_name(key),
|
2021-01-22 03:44:58 +01:00
|
|
|
options.pubkey_accepted_algos, NULL);
|
2018-03-03 04:01:50 +01:00
|
|
|
}
|
2018-07-03 13:39:54 +02:00
|
|
|
|
2021-06-06 05:40:39 +02:00
|
|
|
/*
|
|
|
|
* Workaround OpenSSH 7.4 bug: this version supports RSA/SHA-2 but
|
|
|
|
* fails to advertise it via SSH2_MSG_EXT_INFO.
|
|
|
|
*/
|
|
|
|
server_sig_algs = ssh->kex->server_sig_algs;
|
|
|
|
if (key->type == KEY_RSA && (ssh->compat & SSH_BUG_SIGTYPE74))
|
|
|
|
server_sig_algs = "rsa-sha2-256,rsa-sha2-512";
|
|
|
|
|
2018-07-03 13:39:54 +02:00
|
|
|
/*
|
|
|
|
* For RSA keys/certs, since these might have a different sig type:
|
2021-01-22 03:44:58 +01:00
|
|
|
* find the first entry in PubkeyAcceptedAlgorithms of the right type
|
2018-07-03 13:39:54 +02:00
|
|
|
* that also appears in the supported signature algorithms list from
|
|
|
|
* the server.
|
|
|
|
*/
|
2021-01-22 03:44:58 +01:00
|
|
|
oallowed = allowed = xstrdup(options.pubkey_accepted_algos);
|
2018-07-03 13:39:54 +02:00
|
|
|
while ((cp = strsep(&allowed, ",")) != NULL) {
|
|
|
|
if (sshkey_type_from_name(cp) != key->type)
|
|
|
|
continue;
|
2020-06-05 05:24:16 +02:00
|
|
|
tmp = match_list(sshkey_sigalg_by_name(cp),
|
2021-06-06 05:40:39 +02:00
|
|
|
server_sig_algs, NULL);
|
2018-07-03 15:20:25 +02:00
|
|
|
if (tmp != NULL)
|
|
|
|
alg = xstrdup(cp);
|
|
|
|
free(tmp);
|
2018-07-03 13:39:54 +02:00
|
|
|
if (alg != NULL)
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
free(oallowed);
|
|
|
|
return alg;
|
2018-03-03 04:01:50 +01:00
|
|
|
}
|
|
|
|
|
2003-05-14 05:46:00 +02:00
|
|
|
static int
|
2015-01-14 21:05:27 +01:00
|
|
|
identity_sign(struct identity *id, u_char **sigp, size_t *lenp,
|
2018-07-03 13:39:54 +02:00
|
|
|
const u_char *data, size_t datalen, u_int compat, const char *alg)
|
2003-05-14 05:46:00 +02:00
|
|
|
{
|
2019-10-31 22:18:28 +01:00
|
|
|
struct sshkey *sign_key = NULL, *prv = NULL;
|
Merge 9.1 (#626)
* upstream: fix poll() spin when a channel's output fd closes without
data in the channel buffer. Introduce more exact packing of channel fds into
the pollfd array. fixes bz3405 and bz3411; ok deraadt@ markus@
OpenBSD-Commit-ID: 06740737849c9047785622ad5d472cb6a3907d10
* upstream: select post-quantum KEX
sntrup761x25519-sha512@openssh.com as the default; ok markus@
OpenBSD-Commit-ID: f02d99cbfce22dffec2e2ab1b60905fbddf48fb9
* upstream: add support for the "corp-data" protocol extension to
allow server-side copies to be performed without having to go via the client.
Patch by Mike Frysinger, ok dtucker@
OpenBSD-Commit-ID: 00aa510940fedd66dab1843b58682de4eb7156d5
* upstream: add a sftp client "cp" command that supports server-side
copying of files. Useful for this task and for testing the copy-data
extension. Patch from Mike Frysinger; ok dtucker@
OpenBSD-Commit-ID: 1bb1b950af0d49f0d5425b1f267e197aa1b57444
* depend
* Skip slow tests on (very) slow test targets.
* Set Makefile SHELL as determined by configure.
This should improve compatibility for users with non-POSIX shells. If
using Makefile.in directly (eg make -f Makefile.in distprep) then SHELL
will need to be specified on the command line (along with MANFMT in that
particular case). ok djm@
* Use bash or ksh if available for SH in Makefile.
* Increase test timeout to allow slow VMs to finish
* Only run regression tests on slow VMs.
* Only return events from ppoll that were requested.
If the underlying system's select() returns bits that were not in the
request set, our ppoll() implementation can return revents for events
not requested, which can apparently cause a hang. Only return revents
for activity in the requested event set. bz#3416, analysis and fix by
yaroslav.kuzmin at vmssoftware com, ok djm@
* Specify TEST_SHELL=bash on AIX.
The system shells cause the agent-restrict test to fail due to some
quoting so explicitly specify bash until we can get configure to
autmatically work around that.
* Disable security key on fbsd6 test host.
* upstream: man pages: add missing commas between subordinate and
main clauses
jmc@ dislikes a comma before "then" in a conditional, so leave those
untouched.
ok jmc@
OpenBSD-Commit-ID: 9520801729bebcb3c9fe43ad7f9776ab4dd05ea3
* upstream: ssh: document sntrup761x25519-sha512@openssh.com as
default KEX
OpenBSD-Commit-ID: 12545bfa10bcbf552d04d9d9520d0f4e98b0e171
* upstream: openssh-9.0
OpenBSD-Commit-ID: 0dfb461188f4513ec024c1534da8c1ce14c20b64
* update version numbers for release
* update build-aux files to match autoconf-2.71
i.e. config.guess, config.sub and install-sh
* Revert "update build-aux files to match autoconf-2.71"
This reverts commit 0a8ca39fac6ad19096b6c263436f8b2dd51606f2.
It turns out that the checked-in copies of these files are actually newer
than autoconf-2.71's copies, so this was effectively a downgrade.
Spotted by Bo Anderson via github
* upstream: two defensive changes from Tobias Stoeckmann via GHPR287
enforce stricter invarient for sshbuf_set_parent() - never allow
a buffer to have a previously-set parent changed.
In sshbuf_reset(), if the reallocation fails, then zero the entire
buffer and not the (potentially smaller) default initial alloc size.
OpenBSD-Commit-ID: 14583203aa5d50ad38d2e209ae10abaf8955e6a9
* upstream: Note that curve25519-sha256 was later published in
RFC8731. ok djm@
OpenBSD-Commit-ID: 2ac2b5d642d4cf5918eaec8653cad9a4460b2743
* upstream: clear io_want/io_ready flags at start of poll() cycle;
avoids plausible spin during rekeying if channel io_want flags are reused
across cycles. ok markus@ deraadt@
OpenBSD-Commit-ID: 91034f855b7c73cd2591657c49ac30f10322b967
* Retire fbsd6 test VM.
It's long since out of support, relatively slow (it's i686) and the
compiler has trouble with PIE.
* Resync moduli.5 with upstream.
1.18: remove duplicate publication year; carsten dot kunze at arcor dot de
1.19: ssh-keygen's -G/-T have been replaced with -M generate/screen.
* upstream: Correct path for system known hosts file in description
of IgnoreUserKnownHosts. Patch from Martin Vahlensieck via tech@
OpenBSD-Commit-ID: 9b7784f054fa5aa4d63cb36bd563889477127215
* upstream: list the correct version number
for when usage of the sftp protocol became default and fix a typo
from ed maste
OpenBSD-Commit-ID: 24e1795ed2283fdeacf16413c2f07503bcdebb31
* upstream: Import regenerated moduli
OpenBSD-Commit-ID: f9a0726d957cf10692a231996a1f34e7f9cdfeb0
* upstream: Try to continue running local I/O for channels in state
OPEN during SSH transport rekeying. The most visible benefit is that it
should make ~-escapes work in the client (e.g. to exit) if the connection
happened to have stalled during a rekey event. Based work by and ok dtucker@
OpenBSD-Commit-ID: a66e8f254e92edd4ce09c9f750883ec8f1ea5f45
* upstream: Import regenerated moduli
OpenBSD-Commit-ID: f9a0726d957cf10692a231996a1f34e7f9cdfeb0
* upstream: regression test for sftp cp command
OpenBSD-Regress-ID: c96bea9edde3a384b254785e7f9b2b24a81cdf82
* upstream: Simplify forward-control test.
Since we no longer need to support SSH1 we don't need to run shell
commands on the other end of the connection and can use ssh -N instead.
This also makes the test less racy.
OpenBSD-Regress-ID: 32e94ce272820cc398f30b848b2b0f080d10302c
* upstream: Use ssh -f and ControlPersist ..
to start up test forwards and ssh -O stop to shut them down intead of
sleep loops. This speeds up the test by an order of magnitude.
OpenBSD-Regress-ID: eb3db5f805100919b092a3b2579c611fba3e83e7
* upstream: It looks like we can't completely avoid
waiting for processes to exit so retrieve the pid via controlmaster and
use that.
OpenBSD-Regress-ID: 8246f00f22b14e49d2ff1744c94897ead33d457b
* Cache timezone data in capsicum sandbox.
From emaste at freebsd.org, originally part of FreeBSD commit r339216
/ fc3c19a9 with autoconf bits added by me.
* Include stdlib.h for free() prototype.
... which is used inside the CUSTOM_SYS_AUTH_GET_LASTLOGIN_MSG block.
* Update OpenSSL and LibreSSL versions in tests.
* Add debian-riscv64 test target.
* upstream: Avoid an unnecessary xstrdup in rm_env() when matching
patterns. Since match_pattern() doesn't modify its arguments (they are
const), there is no need to make an extra copy of the strings in
options->send_env. From Martin Vahlensieck
OpenBSD-Commit-ID: 2c9db31e3f4d3403b49642c64ee048b2a0a39351
* upstream: Add missing includes of stdlib.h and stdint.h. We need
stdlib.h for malloc(3) and stdint.h for SIZE_MAX. Unlike the other xmss
files, ssh-xmss.c does not include xmss_commons.h so ssh-xmss.c must include
those headers itself. From Martin Vahlensieck
OpenBSD-Commit-ID: 70e28a9818cee3da1be2ef6503d4b396dd421e6b
* upstream: Remove unnecessary includes: openssl/hmac.h and
openssl/evp.h. From Martin Vahlensieck.
OpenBSD-Commit-ID: a6debb5fb0c8a44e43e8d5ca7cc70ad2f3ea31c3
* upstream: Check sshauthopt_new() for NULL. bz#3425, from
tessgauthier at microsoft.com. ok djm@
OpenBSD-Commit-ID: af0315bc3e44aa406daa7e0ae7c2d719a974483f
* upstream: Add authfd path to debug output. ok markus@
OpenBSD-Commit-ID: f735a17d1a6f2bee63bfc609d76ef8db8c090890
* upstream: avoid printing hash algorithm twice; from lucas AT sexy.is
OpenBSD-Commit-ID: 9d24671e10a84141b7c504396cabad600e47a941
* upstream: fix memleak on session-bind path; from Pedro Martelletto, ok
dtucker@
OpenBSD-Commit-ID: e85899a26ba402b4c0717b531317e8fc258f0a7e
* upstream: Don't leak SK device. Patch from Pedro Martelletto via
github PR#316. ok djm@
OpenBSD-Commit-ID: 17d11327545022e727d95fd08b213171c5a4585d
* upstream: mention that the helpers are used by ssh(1), ssh-agent(1)
and ssh-keygen(1). Previously only ssh(1) was mentioned. From Pedro
Martelletto
OpenBSD-Commit-ID: 30f880f989d4b329589c1c404315685960a5f153
* Remove now-empty int32_minmax.inc.
* Only run tests when source files change.
Also run tests on changes to V_9_0 branch.
* Add Mac OS X 12 test target.
* upstream: be stricter in which characters will be accepted in
specifying a mask length; allow only 0-9. From khaleesicodes via GHPR#278; ok
dtucker@
OpenBSD-Commit-ID: e267746c047ea86665cdeccef795a8a56082eeb2
* upstream: fix some integer overflows in sieve_large() that show up when
trying to generate modp groups > 16k bits. Reported via GHPR#306 by Bertram
Felgenhauer, but fixed in a different way. feedback/ok tb@
OpenBSD-Commit-ID: 81cbc6dd3a21c57bd6fadea10e44afe37bca558e
* upstream: remove an obsolete rsa1 format example from an example;
from megan batty
ok djm
OpenBSD-Commit-ID: db2c89879c29bf083df996bd830abfb1e70d62bf
* upstream: Add FIDO AUTHENTICATOR section and explain a bit how FIDO
works. The wording came mostly from the 8.2 OpenSSH release notes, addapted
to fit the man page. Then move the -O bits into the new section as is already
done for CERTIFICATES and MODULI GENERATION. Finally we can explain the
trade-offs of resident keys. While here, consistently refer to the FIDO
thingies as "FIDO authenticators", not "FIDO tokens".
input & OK jmc, naddy
OpenBSD-Commit-ID: dd98748d7644df048f78dcf793b3b63db9ab1d25
* upstream: make sure stdout is non-blocking; ok djm@
OpenBSD-Commit-ID: 64940fffbd1b882eda2d7c8c7a43c79368309c0d
* upstream: mux.c: mark argument as const; from Martin Vahlensieck
OpenBSD-Commit-ID: 69a1a93a55986c7c2ad9f733c093b46a47184341
* upstream: channel_new no longer frees remote_name. So update the
comment accordingly. As remote_name is not modified, it can be const as
well. From Martin Vahlensieck
OpenBSD-Commit-ID: e4e10dc8dc9f40c166ea5a8e991942bedc75a76a
* upstream: sshkey_unshield_private() contains a exact duplicate of
the code in private2_check_padding(). Pull private2_check_padding() up so the
code can be reused. From Martin Vahlensieck, ok deraadt@
OpenBSD-Commit-ID: 876884c3f0e62e8fd8d1594bab06900f971c9c85
* Add ubsan minimal testcase on OpenBSD.
As suggested by djm@.
* Note that, for now, we need variadic macros.
* Also retest OpenBSD upstream on .yml changes.
* upstream: When performing operations that glob(3) a remote path, ensure
that the implicit working directory used to construct that path escapes
glob(3) characters.
This prevents glob characters from being processed in places they
shouldn't, e.g. "cd /tmp/a*/", "get *.txt" should have the get operation
treat the path "/tmp/a*" literally and not attempt to expand it.
Reported by Lusia Kundel; ok markus@
OpenBSD-Commit-ID: 4f647f58482cbad3d58b1eab7f6a1691433deeef
* Remove duplicate bcrypt_pbkdf.o from Makefile
bcrypt_pbkdf.o is duplicated in the openbsd-compat Makefile's object
file list.
* upstream: improve error message when 'ssh-keygen -Y sign' is unable to
load a private key; bz3429, reported by Adam Szkoda ok dtucker@
OpenBSD-Commit-ID: bb57b285e67bea536ef81b1055467be2fc380e74
* upstream: Allow existing -U (use agent) flag to work with "-Y sign"
operations, where it will be interpreted to require that the private keys is
hosted in an agent; bz3429, suggested by Adam Szkoda; ok dtucker@
OpenBSD-Commit-ID: a7bc69873b99c32c42c7628ed9ea91565ba08c2f
* upstream: Remove errant apostrophe. From haruyama at queen-ml org.
OpenBSD-Commit-ID: dc6b294567cb84b384ad6ced9ca469f2bbf0bd10
* upstream: arrange for scp, when in sftp mode, to not ftruncate(3) files
early
previous behavious of unconditionally truncating the destination file
would cause "scp ~/foo localhost:" and "scp localhost:foo ~/" to
delete all the contents of their destination.
spotted by solene@ sthen@, also bz3431; ok dtucker@
OpenBSD-Commit-ID: ca39fdd39e0ec1466b9666f15cbcfddea6aaa179
* upstream: fix in-place copies; r1.163 incorrectly skipped truncation in
all cases, not just at the start of a transfer. This could cause overwrites
of larger files to leave junk at the end. Spotted by tb@
OpenBSD-Commit-ID: b189f19cd68119548c8e24e39c79f61e115bf92c
* upstream: Only run agent-ptrace.sh if gdb is available as all
architectures do not ship with gdb.
OpenBSD-Regress-ID: ec53e928803e6b87f9ac142d38888ca79a45348d
* upstream: regress test for in-place transfers and clobbering larger
files with smaller ones; would have caught last regression in scp(1)
OpenBSD-Regress-ID: 19de4e88dd3a4f7e5c1618c9be3c32415bd93bc2
* configure.ac: Add missing AC_DEFINE for caph_cache_tzdata test causing
HAVE_CAPH_CACHE_TZDATA to be missing from config.h.in.
Spotted by Bryan Drewery
* upstream: make SSHBUF_DBG/SSHBUF_TELL (off by default and only enabled
via #define) dump to stderr rather than stdout
OpenBSD-Commit-ID: 10298513ee32db8390aecb0397d782d68cb14318
* upstream: revert previous; it was broken (spotted by Theo)
OpenBSD-Commit-ID: 457c79afaca2f89ec2606405c1059b98b30d8b0d
* upstream: Note that ProxyJump also accepts the same tokens as
ProxyCommand. From pallxk via github PR#305.
OpenBSD-Commit-ID: 7115ac351b129205f1f1ffa6bbfd62abd76be7c5
* upstream: Avoid kill with -1 argument. The out_ctx label can be
reached before fork has been called. If this happens, then kill -1 would be
called, sending SIGTERM to all processes reachable by the current process.
From tobias@ and c3h2_ctf via github PR#286, ok djm@
OpenBSD-Commit-ID: 6277af1207d81202f5daffdccfeeaed4c763b1a8
* upstream: f sshpkt functions fail, then password is not cleared
with freezero. Unconditionally call freezero to guarantee that password is
removed from RAM.
From tobias@ and c3h2_ctf via github PR#286, ok djm@
OpenBSD-Commit-ID: 6b093619c9515328e25b0f8093779c52402c89cd
* upstream: refactor authorized_keys/principals handling
remove "struct ssh *" from arguments - this was only used to pass the
remote host/address. These can be passed in instead and the resulting
code is less tightly coupled to ssh_api.[ch]
ok dtucker@
OpenBSD-Commit-ID: 9d4373d013edc4cc4b5c21a599e1837ac31dda0d
* upstream: split the low-level file handling functions out from
auth2-pubkey.c
Put them in a new auth2-pubkeyfile.c to make it easier to refer to them
(e.g. in unit/fuzz tests) without having to refer to everything else
pubkey auth brings in.
ok dtucker@
OpenBSD-Commit-ID: 3fdca2c61ad97dc1b8d4a7346816f83dc4ce2217
* fuzzer for authorized_keys parsing
mostly redundant to authopt_fuzz, but it's sensitive code so IMO it
makes sense to test this layer too
* Test against LibreSSL 3.5.3.
* Test against OpenSSL 1.1.1o and 3.0.3.
* fix some bugs in the fuzzer
* upstream: keywords ref ssh_config.5;
from caspar schutijser
OpenBSD-Commit-ID: f146a19d7d5c9374c3b9c520da43b2732d7d1a4e
* upstream: ssh-keygen: implement "verify-required" certificate option.
This was already documented when support for user-verified FIDO
keys was added, but the ssh-keygen(1) code was missing.
ok djm@
OpenBSD-Commit-ID: f660f973391b593fea4b7b25913c9a15c3eb8a06
* upstream: ssh-keygen -A: do not generate DSA keys by default.
Based on github PR#303 from jsegitz with man page text from jmc@, ok markus@
djm@
OpenBSD-Commit-ID: 5c4c57bdd7063ff03381cfb6696659dd3f9f5b9f
* upstream: Add period at end of "not known by any other names"
message. github PR#320 from jschauma, ok djm@
OpenBSD-Commit-ID: bd60809803c4bfd3ebb7c5c4d918b10e275266f2
* upstream: Add missing *-sk types to ssh-keyscan manpage. From
skazi0 via github PR#294.
OpenBSD-Commit-ID: fda2c869cdb871f3c90a89fb3f985370bb5d25c0
* upstream: Make SetEnv directives first-match-wins in both
sshd_config and sshd_config; previously if the same name was reused then the
last would win (which is the opposite to how the config is supposed to work).
While there, make the ssh_config parsing more like sshd_config.
bz3438, ok dtucker
OpenBSD-Commit-ID: 797909c1e0262c0d00e09280459d7ab00f18273b
* upstream: test setenv in both client and server, test first-match-wins
too
OpenBSD-Regress-ID: 4c8804f9db38a02db480b9923317457b377fe34b
* upstream: move auth_openprincipals() and auth_openkeyfile() over to
auth2-pubkeyfile.c too; they make more sense there.
OpenBSD-Commit-ID: 9970d99f900e1117fdaab13e9e910a621b7c60ee
* upstream: make sure that UseDNS hostname lookup happens in the monitor
and not in the pledge(2)'d unprivileged process; fixes regression caused by
recent refactoring spotted by henning@
OpenBSD-Commit-ID: a089870b95101cd8881a2dff65b2f1627d13e88d
* fix possible NULL deref when built without FIDO
Analysis/fix from kircher in bz3443; ok dtucker@
* automatically enable built-in FIDO support
If libfido2 is found and usable, then enable the built-in
security key support unless --without-security-key-builtin
was requested.
ok dtucker@
* upstream: Log an error if pipe() fails while accepting a
connection. bz#3447, from vincent-openssh at vinc17 net, ok djm@
OpenBSD-Commit-ID: 9d59f19872b94900a5c79da2d57850241ac5df94
* upstream: Don't attempt to fprintf a null identity comment. From
Martin Vahlensieck via tech@.
OpenBSD-Commit-ID: 4c54d20a8e8e4e9912c38a7b4ef5bfc5ca2e05c2
* upstream: Make sure not to fclose() the same fd twice in case of an
error.
ok dtucker@
OpenBSD-Commit-ID: e384c4e05d5521e7866b3d53ca59acd2a86eef99
* upstream: make it clear that RekeyLimit applies to both transmitted
and received data. GHPR#328 from Jan Pazdziora
OpenBSD-Commit-ID: d180a905fec9ff418a75c07bb96ea41c9308c3f9
* request 1.1x API compatibility for OpenSSL >=3.x
idea/patch from Pedro Martelletto via GHPR#322; ok dtucker@
* fix broken case statement in previous
* Disable SK support if FIDO libs not found.
* Zero out LIBFIDO2 when SK support not usable.
Prevents us from trying to link them into ssh-sk-helper and failing to
build.
* upstream: Don't leak the strings allocated by order_hostkeyalgs()
and list_hostkey_types() that are passed to compat_pkalg_proposal(). Part of
github PR#324 from ZoltanFridrich, ok djm@
OpenBSD-Commit-ID: b2f6e5f60f2bba293b831654328a8a0035ef4a1b
* upstream: Roll back previous KEX changes as they aren't safe until
compat_pkalg_proposal and friends always allocate their returned strings.
Reported by Qualys.
OpenBSD-Commit-ID: 1c7a88a0d5033f42f88ab9bec58ef1cf72c81ad0
* upstream: allow arguments to sftp -D option, e.g. sftp -D
"/usr/libexec/sftp-server -el debug3"
ok markus@
OpenBSD-Commit-ID: 5a002b9f3a7aef2731fc0ffa9c921cf15f38ecce
* Update OpenSSL tests to the most recent releases.
* upstream: reflect the update to -D arg name in usage();
OpenBSD-Commit-ID: abdcde4f92b1ef094ae44210ee99d3b0155aad9c
* upstream: ignore SIGPIPE earlier in main(), specifically before
muxclient() which performs operations that could cause one; Reported by Noam
Lewis via bz3454, ok dtucker@
OpenBSD-Commit-ID: 63d8e13276869eebac6d7a05d5a96307f9026e47
* upstream: Always return allocated strings from the kex filtering so
that we can free them later. Fix one leak in compat_kex_proposal. Based on
github PR#324 from ZoltanFridrich with some simplications by me. ok djm@
OpenBSD-Commit-ID: 9171616da3307612d0ede086fd511142f91246e4
* upstream: Don't leak the strings allocated by order_hostkeyalgs()
and list_hostkey_types() that are passed to compat_pkalg_proposal(). Part of
github PR#324 from ZoltanFridrich, ok djm@
This is a roll-forward of the previous rollback now that the required
changes in compat.c have been done.
OpenBSD-Commit-ID: c7cd93730b3b9f53cdad3ae32462922834ef73eb
* upstream: bump up loglevel from debug to info when unable to open
authorized keys/principals file for errno != ENOENT; bz2042 ok dtucker
OpenBSD-Commit-ID: e79aa550d91ade6a80f081bda689da24c086d66b
* Skip select+rlimit check if sandboxing is disabled
It's not needed in that case, and the test can fail when being built
with some compiler memory sanitizer flags. bz#3441
* upstream: use consistent field names (s/char/byte)
in format description
OpenBSD-Commit-ID: 3de33572733ee7fcfd7db33d37db23d2280254f0
* upstream: Remove leftover line.
Remove extra line leftover from merge conflict. ok djm@
OpenBSD-Commit-ID: 460e2290875d7ae64971a7e669c244b1d1c0ae2e
* Move checks for pollfd.fd and nfds_t.
Move the checks for struct pollfd.fd and nfds_t to before the sandboxing
checks. This groups all the sandbox checks together so we can skip them
all when sandboxing is disabled.
* Skip all rlimit tests when sandboxing disabled.
The rlimit tests can hang when being run with some compiler sanitizers
so skip all of them if sandbox=no.
* Add clang sanitizer tests.
* upstream: Add TEST_REGRESS_CACHE_DIR.
If set, it is used to cache regress test names that have succeeded and
skip those on a re-run.
OpenBSD-Regress-ID: a7570dd29a58df59f2cca647c3c2ec989b49f247
* Move sanitizer logs into regress for collection.
* Add GCC address sanitizer build/test.
* Update sanitizer test targets:
- remove clang-sanitize-memory for now. It takes so long that the test
times out.
- add gcc sanitize-address and sanitize-undefined test targets.
* Test against openssl-3.0.5.
* Move unset to before we set anything.
* Refuse to use OpenSSL 3.0.4 due to potential RCE.
OpenSSL has a potential RCE in its RSA implementation (CVE-2022-2274)
so refuse to use that specific version.
* Capture stderr output from configure.
* Only refuse to use OpenSSL 3.0.4 on x86_64.
The potential RCE only impacts x86_64, so only refuse to use it if we're
targetting a potentially impacted architecture. ok djm@
* Remove special casing of crypt().
Configure goes to some lengths to pick crypt() from either libcrypt
or OpenSSL's libcrypto because they can more or less featureful (eg
supporting md5-style passwords).
OpenSSL removed its crypt() interface in 2002:
https://github.com/openssl/openssl/commit/69deec58 so these hijinks
should no longer be necessary. This also only links sshd with libcrypt
which is the only thing that needs it. ok djm@
* Clarify README.md text.
Clarify the text about the implications of building without OpenSSL, and
prefix the "configure --help" example command with a "./" so it's likely
to work as-is in more shells. From bz#3461.
* Split README.platform into its own line.
README.platform has general platform-specific information, having it
following text about FIDO2 on the same line could imply that it only
has information about FIDO2.
* Return ERANGE from getcwd() if buffer size is 1.
If getcwd() is supplied a buffer size of exactly 1 and a path of "/", it
could result in a nul byte being written out of array bounds. POSIX says
it should return ERANGE if the path will not fit in the available buffer
(with terminating nul). 1 byte cannot fit any possible path with its nul,
so immediately return ERANGE in that case.
OpenSSH never uses getcwd() with this buffer size, and all current
(and even quite old) platforms that we are currently known to work
on have a native getcwd() so this code is not used on those anyway.
Reported by Qualys, ok djm@
* Remove unintended changes.
I inadvertently included a couple of local changes with the OpenSSL
3.0.4 change. Revert, anything that should be there will be committed
separately.
* Add AUDIT_ARCH_PPC to supported seccomp arches.
Patch from dries.deschout at dodeco.eu.
* Rename bbone test target to ARM.
* Move vmshutdown to first step.
If a previous run on a physical runner has failed to clean up, the next
run will fail because it'll try to check out the code to a broken
directory mount. Make cleanup the first step.
* upstream: pull passphrase reading and confirmation into a separate
function so it can be used for FIDO2 PINs; no functional change
OpenBSD-Commit-ID: bf34f76b8283cc1d3f54633e0d4f13613d87bb2f
* upstream: when enrolling a resident key on a security token, check
if a credential with matching application and user ID strings already exists.
if so, prompt the user for confirmation before overwriting the credential.
patch from Pedro Martelletto via GHPR329
NB. cranks SSH_SK_VERSION_MAJOR, so any third-party FIDO middleware
implementations will need to adjust
OpenBSD-Commit-ID: e45e9f1bf2b2f32d9850669e7a8dbd64acc5fca4
* upstream: sk-usbhid: preserve error code returned by key_lookup()
it conveys useful information, such as the supplied pin being wrong.
Part of GHPR329 from Pedro Martelletto
OpenBSD-Commit-ID: c0647eb9290f793add363d81378439b273756c1b
* upstream: ssh-keygen: fix touch prompt, pin retries;
part of GHPR329 from Pedro Martelletto
OpenBSD-Commit-ID: 75d1005bd2ef8f29fa834c90d2684e73556fffe8
* crank SSH_SK_VERSION_MAJOR in sk-dummy.so
* Skip scp3 test if there's no scp on remote path.
scp -3 ends up using the scp that's in the remote path and will fail if
one is not available. Based on a patch from rapier at psc.edu.
* Convert "have_prog" function into "which".
"which" and its behaviour is not standardized, so convert the existing
have_prog function into "which" so we can rely on it being available
and what its semantics are. Add a have_prog wrapper that maintains the
existing behaviour.
* upstream: Test TEST_SSH_ELAPSED_TIMES for empty string not
executable. No-op on most platforms but should prevent warnings in -portable
on systems that don't have 'date %s'.
OpenBSD-Regress-ID: e39d79867b8065e33d0c5926fa1a31f85659d2a4
* upstream: Restore missing "!" in TEST_SSH_ELAPSED_TIMES test.
OpenBSD-Regress-ID: 38783f9676ec348c5a792caecee9a16e354b37b0
* Remove workarounds for OpenSSL missing AES-GCM.
We have some compatibility hacks that were added to support OpenSSL
versions that do not support AES GCM mode. Since that time, however,
the minimum OpenSSL version that we support has moved to 1.0.1 which
*does* have GCM, so this is no longer needed. ok djm@
* Remove workarounds for OpenSSL missing AES-CTR.
We have some compatibility hacks that were added to support OpenSSL
versions that do not support AES CTR mode. Since that time, however,
the minimum OpenSSL version that we support has moved to 1.0.1 which
*does* have CTR, so this is no longer needed. ok djm@
* Do not link scp, sftp and sftp-server w/ zlib.
Some of our binaries (eg sftp, sftp-server, scp) do not interact with
the channels code and thus do use libraries such as zlib and libcrypto
although they are linked with them. This adds a CHANNELLIBS and starts
by moving zlib into it, which means the aformentioned binaries are no
longer linked against zlib. ok djm@
* Group libcrypto and PRNGD checks together.
They're related more than the libcrypt or libiaf checks which are
currently between them. ok djm@
* Remove seed_rng calls from scp, sftp, sftp-server.
These binaries don't use OpenSSL's random functions. The next step
will be to stop linking them against libcrypto. ok djm@
* Move libcrypto into CHANNELLIBS.
This will result in sftp, sftp-server and scp no longer being linked
against libcrypto. ok djm@
* Move stale-configure check as early as possible.
We added a check in Makefile to catch the case where configure needs to
be rebuilt, however this did not happen until a build was attempted in
which case all of the work done by configure was wasted. Move this check
to the start of configure to catch it as early as possible. ok djm@
* Remove deprecated MacOS 10.15 runners.
* upstream: avoid double-free in error path introduced in r1.70; report
and fix based on GHPR#332 by v-rzh ok dtucker@
OpenBSD-Commit-ID: 3d21aa127b1f37cfc5bdc21461db369a663a951f
* Include CHANNEL and FIDO2 libs in configure output
* Factor out getrnd() and rename to getentropy().
Factor out the arc4random seeding into its own file and change the
interface to match getentropy. Use native getentropy if available.
This will make it easier to resync OpenBSD changes to arc4random.
Prompted by bz#3467, ok djm@.
* compat code for fido_dev_is_winhello()
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* check_sk_options: add temporary WinHello workaround
Up to libfido 1.10.0, WinHello advertises "clientPin" rather
than "uv" capability. This is fixed in 1.11.0. For the time
being, workaround it here.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* sk_sign: set FIDO2 uv attribute explicitely for WinHello
WinHello via libfido2 performs user verification by default.
However, if we stick to that, there's no way to differentiate
between keys created with or without "-O verify-required".
Set FIDO2 uv attribute explicitely to FIDO_OPT_FALSE, then check
if user verification has been requested.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* upstream: don't prompt for FIDO passphrase before attempting to enroll
the credential, just let the enroll operating fail and we'll attempt to get a
PIN anyway. Might avoid some unneccessary PIN prompts.
Part of GHPR#302 from Corinna Vinschen; ok dtucker@
OpenBSD-Commit-ID: bd5342ffc353ee37d39617906867c305564d1ce2
* Give unused param a name.
Fixes builds on platforms that do have fido2 but don't have
fido_dev_is_winhello.
* Actually put HAVE_STDINT_H around the stdint.h.
* Rename our getentropy to prevent possible loops.
Since arc4random seeds from getentropy, and we use OpenSSL for that
if enabled, there's the possibility that if we build on a system that
does not have getentropy then run on a system that does have it, then
OpenSSL could end up calling our getentropy and getting stuck in a loop.
Pointed out by deraadt@, ok djm@
* Test hostbased auth on github runners.
* fix SANDBOX_SECCOMP_FILTER_DEBUG
* Fix conditional for running hostbased tests.
* upstream: allow certificate validity intervals, sshsig verification
times and authorized_keys expiry-time options to accept dates in the UTC time
zone in addition to the default of interpreting them in the system time zone.
YYYYMMDD and YYMMDDHHMM[SS] dates/times will be interpreted as UTC if
suffixed with a 'Z' character.
Also allow certificate validity intervals to be specified in raw
seconds-since-epoch as hex value, e.g. -V 0x1234:0x4567890. This
is intended for use by regress tests and other tools that call
ssh-keygen as part of a CA workflow.
bz3468 ok dtucker
OpenBSD-Commit-ID: 454db1cdffa9fa346aea5211223a2ce0588dfe13
* upstream: add some tests for parse_absolute_time(), including cases
where it is forced to the UTC timezone. bz3468 ok dtucker
OpenBSD-Regress-ID: ea07ca31c2f3847a38df028ca632763ae44e8759
* Skip hostbased during Valgrind tests.
Valgrind doesn't let ssh exec ssh-keysign (because it's setuid) so skip
it during the Valgrind based tests.
See https://bugs.kde.org/show_bug.cgi?id=119404 for a discussion of this
(ironically there the problematic binary was ssh(1) back when it could
still be setuid).
* Rerun tests if any .github config file changes.
* Add a timegm implementation from Heimdal via Samba.
Fixes build on (at least Solaris 10).
* Replace deprecated ubuntu-18.04 runners with 22.04
* upstream: sftp-server: support home-directory request
Add support to the sftp-server for the home-directory extension defined
in draft-ietf-secsh-filexfer-extensions-00. This overlaps a bit with the
existing expand-path@openssh.com, but uses a more official protocol name,
and so is a bit more likely to be implemented by non-OpenSSH clients.
From Mike Frysinger, ok dtucker@
OpenBSD-Commit-ID: bfc580d05cc0c817831ae7ecbac4a481c23566ab
* fido_dev_is_winhello: return 0, not "false"
"false" is not used anywhere in OpenSSH, so return 0 like
everywhere else.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* Revert "check_sk_options: add temporary WinHello workaround"
Cygwin now comes with libfido2 1.11.0, so this workaround
isn't required anymore.
This reverts commit 242c044ab111a37aad3b0775727c36a4c5f0102c.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* upstream: use .Cm for "sign"; from josiah frentsos
OpenBSD-Commit-ID: 7f80a53d54857ac6ae49ea6ad93c5bd12231d1e4
* upstream: add an extra flag to sk_probe() to indicate whether we're
probing for a FIDO resident key or not. Unused here, but will make like
easier for portable
OpenBSD-Commit-ID: 432c8ff70e270378df9dbceb9bdeaa5b43b5a832
* on Cygwin, prefer WinHello FIDO device
If no FIDO device was explictly specified, then prefer the
windows://hello FIDO device. An exception to this is when
probing resident FIDO keys, in which case hardware FIDO
devices are preferred.
* Check for perms to run agent-getpeereid test.
Ubuntu 22.04 defaults to private home dirs which prevents "nobody"
running ssh-add during the agent-getpeereid test. Check for this and
add the necessary permissions.
* upstream: double free() in error path; from Eusgor via GHPR333
OpenBSD-Commit-ID: 39f35e16ba878c8d02b4d01d8826d9b321be26d4
* Add Cygwin (on windows-2019) test target.
In addition to installing the requisite Cygwin packages, we also need to
explicitly invoke "sh" for steps that run other scripts since the runner
environment doesn't understand #! paths.
* Add a bit more debug output.
* Fix cygwin conditional steps.
* upstream: Strictly enforce the maximum allowed SSH2 banner size in
ssh-keyscan and prevent a one-byte buffer overflow. Patch from Qualys, ok
djm@
OpenBSD-Commit-ID: 6ae664f9f4db6e8a0589425f74cd0bbf3aeef4e4
* upstream: remove incorrect check that can break enrolling a
resident key (introduced in r1.40)
OpenBSD-Commit-ID: 4cab364d518470e29e624af3d3f9ffa9c92b6f01
* upstream: attemp FIDO key signing without PIN and use the error
code returned to fall back only if necessary. Avoids PIN prompts for FIDO
tokens that don't require them; part of GHPR#302
OpenBSD-Commit-ID: 4f752aaf9f2e7c28bcaaf3d4f8fc290131bd038e
* Install Cygwin packages based on OS not config.
* initial list of allowed signers
* upstream: whitespace
OpenBSD-Commit-ID: d297e4387935d4aef091c5e9432578c2e513f538
* upstream: whitespace
OpenBSD-Commit-ID: a5d015efbfd228dc598ffdef612d2da3a579e5d8
* Add cygwin-release test target.
This also moves the cygwin package install from the workflow file to
setup_ci.sh so that we can install different sets of Cygwin packages
for different test configs.
* Add Windows 2022 test targets.
* Add libcrypt-devel to cygwin-release deps.
Based on feedback from vinschen at redhat.com.
* cross-sign allowed_signers with PGP key
Provides continuity of trust from legacy PGP release key to
the SSHSIG signing keys that we will use henceforth for git
signing.
* additional keys
* upstream: whitespace
OpenBSD-Commit-ID: c2bcbf93610d3d62ed206cdf9bf9ff98c6aaf232
* Move sftp from valgrind-2 to 3 to rebalance.
* upstream: sk-usbhid: fix key_lookup() on tokens with built-in UV
explicitly test whether the token performs built-in UV (e.g. biometric
tokens) and enable UV in that case. From Pedro Martelletto via GHPR#388
OpenBSD-Commit-ID: 007eb7e387d27cf3029ab06b88224e03eca62ccd
* Remove arc4random_uniform from arc4random.c
This was previously moved into its own file (matching OpenBSD) which
prematurely committed in commit 73541f2.
* Move OPENBSD ORIGINAL marker.
Putting this after the copyright statement (which doesn't change)
instead of before the version identifier (which does) prevents merge
conflicts when resyncing changes.
* Resync arc4random with OpenBSD.
This brings us up to current, including djm's random-reseeding change,
as prompted by logan at cyberstorm.mu in bz#3467. It brings the
platform-specific hooks from LibreSSL Portable, simplified to match our
use case. ok djm@.
* Remove DEF_WEAK, it's already in defines.h.
* openbsd-compat/bsd-asprintf: add <stdio.h> include for vsnprintf
Fixes the following build failure with Clang 15 on musl:
```
bsd-asprintf.c:51:8: error: call to undeclared library function 'vsnprintf' with type 'int (char *, unsigned long, const char *, struct __va_list_tag *)'; ISO C99 and laterclang -O2 -pipe -fdiagnostics-color=always -frecord-gcc-switches -pipe -Wunknown-warning-option -Qunused-arguments -Wall -Wpointer-arith -Wuninitialized -Wsign-compare -Wformat-security -Wsizeof-pointer-memaccess -Wno-pointer-sign -Wno-unused-result -Wmisleading-indentation -Wbitwise-instead-of-logical -fno-strict-aliasing -mretpoline -ftrapv -fzero-call-used-regs=all -fno-builtin-memset -fstack-protector-strong -fPIE -I. -I. -D_XOPEN_SOURCE=600 -D_BSD_SOURCE -D_DEFAULT_SOURCE -DSSHDIR=\"/etc/ssh\" -D_PATH_SSH_PROGRAM=\"/usr/bin/ssh\" -D_PATH_SSH_ASKPASS_DEFAULT=\"/usr/lib/misc/ssh-askpass\" -D_PATH_SFTP_SERVER=\"/usr/lib/misc/sftp-server\" -D_PATH_SSH_KEY_SIGN=\"/usr/lib/misc/ssh-keysign\" -D_PATH_SSH_PKCS11_HELPER=\"/usr/lib/misc/ssh-pkcs11-helper\" -D_PATH_SSH_SK_HELPER=\"/usr/lib/misc/ssh-sk-helper\" -D_PATH_SSH_PIDDIR=\"/run\" -D_PATH_PRIVSEP_CHROOT_DIR=\"/var/empty\" -DHAVE_CONFIG_H -c cipher-aes.c -o cipher-aes.o
do not support
implicit function declarations [-Wimplicit-function-declaration]
ret = vsnprintf(string, INIT_SZ, fmt, ap2);
^
bsd-asprintf.c:51:8: note: include the header <stdio.h> or explicitly provide a declaration for 'vsnprintf'
1 error generated.
```
* upstream: notifier_complete(NULL, ...) is a noop, so no need to test
that ctx!=NULL; from Corinna Vinschen
OpenBSD-Commit-ID: ade2f2e9cc519d01a586800c25621d910bce384a
* fix pester test failures
* upstream: fix repeated words ok miod@ jmc@
OpenBSD-Commit-ID: 6765daefe26a6b648cc15cadbbe337596af709b7
* upstream: .Li -> .Vt where appropriate; from josiah frentsos,
tweaked by schwarze
ok schwarze
OpenBSD-Commit-ID: 565046e3ce68b46c2f440a93d67c2a92726de8ed
* upstream: ssh-agent: attempt FIDO key signing without PIN and use
the error to determine whether a PIN is required and prompt only if
necessary. from Corinna Vinschen
OpenBSD-Commit-ID: dd6be6a0b7148608e834ee737c3479b3270b00dd
* upstream: a little extra debugging
OpenBSD-Commit-ID: edf1601c1d0905f6da4c713f4d9cecc7d1c0295a
* upstream: sk_enroll: never drop SSH_SK_USER_VERIFICATION_REQD flag
from response
Now that all FIDO signing calls attempt first without PIN and then
fall back to trying PIN only if that attempt fails, we can remove the
hack^wtrick that removed the UV flag from the keys returned during
enroll.
By Corinna Vinschen
OpenBSD-Commit-ID: 684517608c8491503bf80cd175425f0178d91d7f
* define HAVE_KILLPG
* upstream: sftp: Don't attempt to complete arguments for
non-existent commands
If user entered a non-existent command (e.g. because they made a
typo) there is no point in trying to complete its arguments. Skip
calling complete_match() if that's the case.
From Michal Privoznik
OpenBSD-Commit-ID: cf39c811a68cde2aeb98fc85addea4000ef6b07a
* upstream: sftp: Be a bit more clever about completions
There are commands (e.g. "get" or "put") that accept two
arguments, a local path and a remote path. However, the way
current completion is written doesn't take this distinction into
account and always completes remote or local paths.
By expanding CMD struct and "cmds" array this distinction can be
reflected and with small adjustment to completer code the correct
path can be completed.
By Michal Privoznik, ok dtucker@
OpenBSD-Commit-ID: 1396d921c4eb1befd531f5c4a8ab47e7a74b610b
* upstream: correct error value
OpenBSD-Commit-ID: 780efcbad76281f11f14b2a5ff04eb6db3dfdad4
* upstream: actually hook up restrict_websafe; the command-line flag
was never actually used. Spotted by Matthew Garrett
OpenBSD-Commit-ID: 0b363518ac4c2819dbaa3dfad4028633ab9cdff1
* upstream: Add a sshkey_check_rsa_length() call for checking the
length of an RSA key; ok markus@
OpenBSD-Commit-ID: de77cd5b11594297eda82edc594b0d32b8535134
* upstream: add a RequiredRSASize for checking RSA key length in
ssh(1). User authentication keys that fall beneath this limit will be
ignored. If a host presents a host key beneath this limit then the connection
will be terminated (unfortunately there are no fallbacks in the protocol for
host authentication).
feedback deraadt, Dmitry Belyavskiy; ok markus@
OpenBSD-Commit-ID: 430e339b2a79fa9ecc63f2837b06fdd88a7da13a
* upstream: Add RequiredRSASize for sshd(8); RSA keys that fall
beneath this limit will be ignored for user and host-based authentication.
Feedback deraadt@ ok markus@
OpenBSD-Commit-ID: 187931dfc19d51873df5930a04f2d972adf1f7f1
* upstream: better debugging for connect_next()
OpenBSD-Commit-ID: d16a307a0711499c971807f324484ed3a6036640
* upstream: sftp-server(8): add a "users-groups-by-id@openssh.com"
extension request that allows the client to obtain user/group names that
correspond to a set of uids/gids.
Will be used to make directory listings more useful and consistent
in sftp(1).
ok markus@
OpenBSD-Commit-ID: 7ebabde0bcb95ef949c4840fe89e697e30df47d3
* upstream: extend sftp-common.c:extend ls_file() to support supplied
user/group names; ok markus@
OpenBSD-Commit-ID: c70c70498b1fdcf158531117e405b6245863bfb0
* upstream: sftp client library support for
users-groups-by-id@openssh.com; ok markus@
OpenBSD-Commit-ID: ddb2f33a2da6349a9a89a8b5bcb9ca7c999394de
* upstream: use users-groups-by-id@openssh.com sftp-server extension
(when available) to fill in user/group names for directory listings.
Implement a client-side cache of see uid/gid=>user/group names. ok markus@
OpenBSD-Commit-ID: f239aeeadfa925a37ceee36ee8b256b8ccf4466e
* avoid Wuninitialized false positive in gcc-12ish
* no need for glob.h here
it also causes portability problems
* add debug on appveyor
* add sleep to pester test
* upstream: add RequiredRSASize to the list of keywords accepted by
-o; spotted by jmc@
OpenBSD-Commit-ID: fe871408cf6f9d3699afeda876f8adbac86a035e
* upstream: Fix typo. From AlexanderStohr via github PR#343.
OpenBSD-Commit-ID: a134c9b4039e48803fc6a87f955b0f4a03181497
* upstream: openssh-9.1
OpenBSD-Commit-ID: 5a467b2ee81da01a86adf1ad93b62b1728494e56
* crank versions in RPM spec files
* update release notes URL
* update .depend
* fix 9.1 compilation errors
* disable -p pester tests due to unreliability on older Windows versions
* remove extra sleep time from debugging scp pester tests
* modify -p tests to only run for Windows OS version 10 and above
* add windows specific code back into method moved from auth.c to auth2-pubkeyfile.c
* add preprocessor for WinHello
* revert preprocessor definition for winhello
* add windows preprocessor definition in key_lookup
* remove rdp block from appveyor since we are no longer debugging
* add ifdef to sftp-server.c
* make key_lookup compatible with winhello
* appveyor.yml
* increase debug of failing pester test
* add #ifdef SUPPORT_CRLF back into auth_check_principals_line method that was moved/renamed
* modify new scp.sh tests for windows
* remove in place tests from scp.sh
* remove rdp debug from appveyor
* retrigger appveyor
* change check of OS version in scp test
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
Co-authored-by: djm@openbsd.org <djm@openbsd.org>
Co-authored-by: Damien Miller <djm@mindrot.org>
Co-authored-by: Darren Tucker <dtucker@dtucker.net>
Co-authored-by: naddy@openbsd.org <naddy@openbsd.org>
Co-authored-by: dtucker@openbsd.org <dtucker@openbsd.org>
Co-authored-by: tj@openbsd.org <tj@openbsd.org>
Co-authored-by: millert@openbsd.org <millert@openbsd.org>
Co-authored-by: jmc@openbsd.org <jmc@openbsd.org>
Co-authored-by: florian@openbsd.org <florian@openbsd.org>
Co-authored-by: markus@openbsd.org <markus@openbsd.org>
Co-authored-by: Tobias Heider <me@tobhe.de>
Co-authored-by: anton@openbsd.org <anton@openbsd.org>
Co-authored-by: Tim Rice <tim@multitalents.net>
Co-authored-by: tobhe@openbsd.org <tobhe@openbsd.org>
Co-authored-by: Corinna Vinschen <vinschen@redhat.com>
Co-authored-by: Sam James <sam@gentoo.org>
Co-authored-by: jsg@openbsd.org <jsg@openbsd.org>
2022-11-02 17:06:45 +01:00
|
|
|
int is_agent = 0, retried = 0, r = SSH_ERR_INTERNAL_ERROR;
|
2019-11-12 23:36:44 +01:00
|
|
|
struct notifier_ctx *notifier = NULL;
|
2020-08-27 03:06:18 +02:00
|
|
|
char *fp = NULL, *pin = NULL, *prompt = NULL;
|
2019-10-31 22:18:28 +01:00
|
|
|
|
|
|
|
*sigp = NULL;
|
|
|
|
*lenp = 0;
|
2003-05-14 05:46:00 +02:00
|
|
|
|
2018-07-03 13:39:54 +02:00
|
|
|
/* The agent supports this key. */
|
2018-03-03 04:01:50 +01:00
|
|
|
if (id->key != NULL && id->agent_fd != -1) {
|
2018-07-03 13:39:54 +02:00
|
|
|
return ssh_agent_sign(id->agent_fd, id->key, sigp, lenp,
|
|
|
|
data, datalen, alg, compat);
|
2018-03-03 04:01:50 +01:00
|
|
|
}
|
2015-01-14 21:05:27 +01:00
|
|
|
|
2003-05-14 05:46:00 +02:00
|
|
|
/*
|
2018-07-03 13:39:54 +02:00
|
|
|
* We have already loaded the private key or the private key is
|
|
|
|
* stored in external hardware.
|
2003-05-14 05:46:00 +02:00
|
|
|
*/
|
2017-03-12 00:40:26 +01:00
|
|
|
if (id->key != NULL &&
|
2018-07-03 13:39:54 +02:00
|
|
|
(id->isprivate || (id->key->flags & SSHKEY_FLAG_EXT))) {
|
2019-10-31 22:18:28 +01:00
|
|
|
sign_key = id->key;
|
Merge 9.1 (#626)
* upstream: fix poll() spin when a channel's output fd closes without
data in the channel buffer. Introduce more exact packing of channel fds into
the pollfd array. fixes bz3405 and bz3411; ok deraadt@ markus@
OpenBSD-Commit-ID: 06740737849c9047785622ad5d472cb6a3907d10
* upstream: select post-quantum KEX
sntrup761x25519-sha512@openssh.com as the default; ok markus@
OpenBSD-Commit-ID: f02d99cbfce22dffec2e2ab1b60905fbddf48fb9
* upstream: add support for the "corp-data" protocol extension to
allow server-side copies to be performed without having to go via the client.
Patch by Mike Frysinger, ok dtucker@
OpenBSD-Commit-ID: 00aa510940fedd66dab1843b58682de4eb7156d5
* upstream: add a sftp client "cp" command that supports server-side
copying of files. Useful for this task and for testing the copy-data
extension. Patch from Mike Frysinger; ok dtucker@
OpenBSD-Commit-ID: 1bb1b950af0d49f0d5425b1f267e197aa1b57444
* depend
* Skip slow tests on (very) slow test targets.
* Set Makefile SHELL as determined by configure.
This should improve compatibility for users with non-POSIX shells. If
using Makefile.in directly (eg make -f Makefile.in distprep) then SHELL
will need to be specified on the command line (along with MANFMT in that
particular case). ok djm@
* Use bash or ksh if available for SH in Makefile.
* Increase test timeout to allow slow VMs to finish
* Only run regression tests on slow VMs.
* Only return events from ppoll that were requested.
If the underlying system's select() returns bits that were not in the
request set, our ppoll() implementation can return revents for events
not requested, which can apparently cause a hang. Only return revents
for activity in the requested event set. bz#3416, analysis and fix by
yaroslav.kuzmin at vmssoftware com, ok djm@
* Specify TEST_SHELL=bash on AIX.
The system shells cause the agent-restrict test to fail due to some
quoting so explicitly specify bash until we can get configure to
autmatically work around that.
* Disable security key on fbsd6 test host.
* upstream: man pages: add missing commas between subordinate and
main clauses
jmc@ dislikes a comma before "then" in a conditional, so leave those
untouched.
ok jmc@
OpenBSD-Commit-ID: 9520801729bebcb3c9fe43ad7f9776ab4dd05ea3
* upstream: ssh: document sntrup761x25519-sha512@openssh.com as
default KEX
OpenBSD-Commit-ID: 12545bfa10bcbf552d04d9d9520d0f4e98b0e171
* upstream: openssh-9.0
OpenBSD-Commit-ID: 0dfb461188f4513ec024c1534da8c1ce14c20b64
* update version numbers for release
* update build-aux files to match autoconf-2.71
i.e. config.guess, config.sub and install-sh
* Revert "update build-aux files to match autoconf-2.71"
This reverts commit 0a8ca39fac6ad19096b6c263436f8b2dd51606f2.
It turns out that the checked-in copies of these files are actually newer
than autoconf-2.71's copies, so this was effectively a downgrade.
Spotted by Bo Anderson via github
* upstream: two defensive changes from Tobias Stoeckmann via GHPR287
enforce stricter invarient for sshbuf_set_parent() - never allow
a buffer to have a previously-set parent changed.
In sshbuf_reset(), if the reallocation fails, then zero the entire
buffer and not the (potentially smaller) default initial alloc size.
OpenBSD-Commit-ID: 14583203aa5d50ad38d2e209ae10abaf8955e6a9
* upstream: Note that curve25519-sha256 was later published in
RFC8731. ok djm@
OpenBSD-Commit-ID: 2ac2b5d642d4cf5918eaec8653cad9a4460b2743
* upstream: clear io_want/io_ready flags at start of poll() cycle;
avoids plausible spin during rekeying if channel io_want flags are reused
across cycles. ok markus@ deraadt@
OpenBSD-Commit-ID: 91034f855b7c73cd2591657c49ac30f10322b967
* Retire fbsd6 test VM.
It's long since out of support, relatively slow (it's i686) and the
compiler has trouble with PIE.
* Resync moduli.5 with upstream.
1.18: remove duplicate publication year; carsten dot kunze at arcor dot de
1.19: ssh-keygen's -G/-T have been replaced with -M generate/screen.
* upstream: Correct path for system known hosts file in description
of IgnoreUserKnownHosts. Patch from Martin Vahlensieck via tech@
OpenBSD-Commit-ID: 9b7784f054fa5aa4d63cb36bd563889477127215
* upstream: list the correct version number
for when usage of the sftp protocol became default and fix a typo
from ed maste
OpenBSD-Commit-ID: 24e1795ed2283fdeacf16413c2f07503bcdebb31
* upstream: Import regenerated moduli
OpenBSD-Commit-ID: f9a0726d957cf10692a231996a1f34e7f9cdfeb0
* upstream: Try to continue running local I/O for channels in state
OPEN during SSH transport rekeying. The most visible benefit is that it
should make ~-escapes work in the client (e.g. to exit) if the connection
happened to have stalled during a rekey event. Based work by and ok dtucker@
OpenBSD-Commit-ID: a66e8f254e92edd4ce09c9f750883ec8f1ea5f45
* upstream: Import regenerated moduli
OpenBSD-Commit-ID: f9a0726d957cf10692a231996a1f34e7f9cdfeb0
* upstream: regression test for sftp cp command
OpenBSD-Regress-ID: c96bea9edde3a384b254785e7f9b2b24a81cdf82
* upstream: Simplify forward-control test.
Since we no longer need to support SSH1 we don't need to run shell
commands on the other end of the connection and can use ssh -N instead.
This also makes the test less racy.
OpenBSD-Regress-ID: 32e94ce272820cc398f30b848b2b0f080d10302c
* upstream: Use ssh -f and ControlPersist ..
to start up test forwards and ssh -O stop to shut them down intead of
sleep loops. This speeds up the test by an order of magnitude.
OpenBSD-Regress-ID: eb3db5f805100919b092a3b2579c611fba3e83e7
* upstream: It looks like we can't completely avoid
waiting for processes to exit so retrieve the pid via controlmaster and
use that.
OpenBSD-Regress-ID: 8246f00f22b14e49d2ff1744c94897ead33d457b
* Cache timezone data in capsicum sandbox.
From emaste at freebsd.org, originally part of FreeBSD commit r339216
/ fc3c19a9 with autoconf bits added by me.
* Include stdlib.h for free() prototype.
... which is used inside the CUSTOM_SYS_AUTH_GET_LASTLOGIN_MSG block.
* Update OpenSSL and LibreSSL versions in tests.
* Add debian-riscv64 test target.
* upstream: Avoid an unnecessary xstrdup in rm_env() when matching
patterns. Since match_pattern() doesn't modify its arguments (they are
const), there is no need to make an extra copy of the strings in
options->send_env. From Martin Vahlensieck
OpenBSD-Commit-ID: 2c9db31e3f4d3403b49642c64ee048b2a0a39351
* upstream: Add missing includes of stdlib.h and stdint.h. We need
stdlib.h for malloc(3) and stdint.h for SIZE_MAX. Unlike the other xmss
files, ssh-xmss.c does not include xmss_commons.h so ssh-xmss.c must include
those headers itself. From Martin Vahlensieck
OpenBSD-Commit-ID: 70e28a9818cee3da1be2ef6503d4b396dd421e6b
* upstream: Remove unnecessary includes: openssl/hmac.h and
openssl/evp.h. From Martin Vahlensieck.
OpenBSD-Commit-ID: a6debb5fb0c8a44e43e8d5ca7cc70ad2f3ea31c3
* upstream: Check sshauthopt_new() for NULL. bz#3425, from
tessgauthier at microsoft.com. ok djm@
OpenBSD-Commit-ID: af0315bc3e44aa406daa7e0ae7c2d719a974483f
* upstream: Add authfd path to debug output. ok markus@
OpenBSD-Commit-ID: f735a17d1a6f2bee63bfc609d76ef8db8c090890
* upstream: avoid printing hash algorithm twice; from lucas AT sexy.is
OpenBSD-Commit-ID: 9d24671e10a84141b7c504396cabad600e47a941
* upstream: fix memleak on session-bind path; from Pedro Martelletto, ok
dtucker@
OpenBSD-Commit-ID: e85899a26ba402b4c0717b531317e8fc258f0a7e
* upstream: Don't leak SK device. Patch from Pedro Martelletto via
github PR#316. ok djm@
OpenBSD-Commit-ID: 17d11327545022e727d95fd08b213171c5a4585d
* upstream: mention that the helpers are used by ssh(1), ssh-agent(1)
and ssh-keygen(1). Previously only ssh(1) was mentioned. From Pedro
Martelletto
OpenBSD-Commit-ID: 30f880f989d4b329589c1c404315685960a5f153
* Remove now-empty int32_minmax.inc.
* Only run tests when source files change.
Also run tests on changes to V_9_0 branch.
* Add Mac OS X 12 test target.
* upstream: be stricter in which characters will be accepted in
specifying a mask length; allow only 0-9. From khaleesicodes via GHPR#278; ok
dtucker@
OpenBSD-Commit-ID: e267746c047ea86665cdeccef795a8a56082eeb2
* upstream: fix some integer overflows in sieve_large() that show up when
trying to generate modp groups > 16k bits. Reported via GHPR#306 by Bertram
Felgenhauer, but fixed in a different way. feedback/ok tb@
OpenBSD-Commit-ID: 81cbc6dd3a21c57bd6fadea10e44afe37bca558e
* upstream: remove an obsolete rsa1 format example from an example;
from megan batty
ok djm
OpenBSD-Commit-ID: db2c89879c29bf083df996bd830abfb1e70d62bf
* upstream: Add FIDO AUTHENTICATOR section and explain a bit how FIDO
works. The wording came mostly from the 8.2 OpenSSH release notes, addapted
to fit the man page. Then move the -O bits into the new section as is already
done for CERTIFICATES and MODULI GENERATION. Finally we can explain the
trade-offs of resident keys. While here, consistently refer to the FIDO
thingies as "FIDO authenticators", not "FIDO tokens".
input & OK jmc, naddy
OpenBSD-Commit-ID: dd98748d7644df048f78dcf793b3b63db9ab1d25
* upstream: make sure stdout is non-blocking; ok djm@
OpenBSD-Commit-ID: 64940fffbd1b882eda2d7c8c7a43c79368309c0d
* upstream: mux.c: mark argument as const; from Martin Vahlensieck
OpenBSD-Commit-ID: 69a1a93a55986c7c2ad9f733c093b46a47184341
* upstream: channel_new no longer frees remote_name. So update the
comment accordingly. As remote_name is not modified, it can be const as
well. From Martin Vahlensieck
OpenBSD-Commit-ID: e4e10dc8dc9f40c166ea5a8e991942bedc75a76a
* upstream: sshkey_unshield_private() contains a exact duplicate of
the code in private2_check_padding(). Pull private2_check_padding() up so the
code can be reused. From Martin Vahlensieck, ok deraadt@
OpenBSD-Commit-ID: 876884c3f0e62e8fd8d1594bab06900f971c9c85
* Add ubsan minimal testcase on OpenBSD.
As suggested by djm@.
* Note that, for now, we need variadic macros.
* Also retest OpenBSD upstream on .yml changes.
* upstream: When performing operations that glob(3) a remote path, ensure
that the implicit working directory used to construct that path escapes
glob(3) characters.
This prevents glob characters from being processed in places they
shouldn't, e.g. "cd /tmp/a*/", "get *.txt" should have the get operation
treat the path "/tmp/a*" literally and not attempt to expand it.
Reported by Lusia Kundel; ok markus@
OpenBSD-Commit-ID: 4f647f58482cbad3d58b1eab7f6a1691433deeef
* Remove duplicate bcrypt_pbkdf.o from Makefile
bcrypt_pbkdf.o is duplicated in the openbsd-compat Makefile's object
file list.
* upstream: improve error message when 'ssh-keygen -Y sign' is unable to
load a private key; bz3429, reported by Adam Szkoda ok dtucker@
OpenBSD-Commit-ID: bb57b285e67bea536ef81b1055467be2fc380e74
* upstream: Allow existing -U (use agent) flag to work with "-Y sign"
operations, where it will be interpreted to require that the private keys is
hosted in an agent; bz3429, suggested by Adam Szkoda; ok dtucker@
OpenBSD-Commit-ID: a7bc69873b99c32c42c7628ed9ea91565ba08c2f
* upstream: Remove errant apostrophe. From haruyama at queen-ml org.
OpenBSD-Commit-ID: dc6b294567cb84b384ad6ced9ca469f2bbf0bd10
* upstream: arrange for scp, when in sftp mode, to not ftruncate(3) files
early
previous behavious of unconditionally truncating the destination file
would cause "scp ~/foo localhost:" and "scp localhost:foo ~/" to
delete all the contents of their destination.
spotted by solene@ sthen@, also bz3431; ok dtucker@
OpenBSD-Commit-ID: ca39fdd39e0ec1466b9666f15cbcfddea6aaa179
* upstream: fix in-place copies; r1.163 incorrectly skipped truncation in
all cases, not just at the start of a transfer. This could cause overwrites
of larger files to leave junk at the end. Spotted by tb@
OpenBSD-Commit-ID: b189f19cd68119548c8e24e39c79f61e115bf92c
* upstream: Only run agent-ptrace.sh if gdb is available as all
architectures do not ship with gdb.
OpenBSD-Regress-ID: ec53e928803e6b87f9ac142d38888ca79a45348d
* upstream: regress test for in-place transfers and clobbering larger
files with smaller ones; would have caught last regression in scp(1)
OpenBSD-Regress-ID: 19de4e88dd3a4f7e5c1618c9be3c32415bd93bc2
* configure.ac: Add missing AC_DEFINE for caph_cache_tzdata test causing
HAVE_CAPH_CACHE_TZDATA to be missing from config.h.in.
Spotted by Bryan Drewery
* upstream: make SSHBUF_DBG/SSHBUF_TELL (off by default and only enabled
via #define) dump to stderr rather than stdout
OpenBSD-Commit-ID: 10298513ee32db8390aecb0397d782d68cb14318
* upstream: revert previous; it was broken (spotted by Theo)
OpenBSD-Commit-ID: 457c79afaca2f89ec2606405c1059b98b30d8b0d
* upstream: Note that ProxyJump also accepts the same tokens as
ProxyCommand. From pallxk via github PR#305.
OpenBSD-Commit-ID: 7115ac351b129205f1f1ffa6bbfd62abd76be7c5
* upstream: Avoid kill with -1 argument. The out_ctx label can be
reached before fork has been called. If this happens, then kill -1 would be
called, sending SIGTERM to all processes reachable by the current process.
From tobias@ and c3h2_ctf via github PR#286, ok djm@
OpenBSD-Commit-ID: 6277af1207d81202f5daffdccfeeaed4c763b1a8
* upstream: f sshpkt functions fail, then password is not cleared
with freezero. Unconditionally call freezero to guarantee that password is
removed from RAM.
From tobias@ and c3h2_ctf via github PR#286, ok djm@
OpenBSD-Commit-ID: 6b093619c9515328e25b0f8093779c52402c89cd
* upstream: refactor authorized_keys/principals handling
remove "struct ssh *" from arguments - this was only used to pass the
remote host/address. These can be passed in instead and the resulting
code is less tightly coupled to ssh_api.[ch]
ok dtucker@
OpenBSD-Commit-ID: 9d4373d013edc4cc4b5c21a599e1837ac31dda0d
* upstream: split the low-level file handling functions out from
auth2-pubkey.c
Put them in a new auth2-pubkeyfile.c to make it easier to refer to them
(e.g. in unit/fuzz tests) without having to refer to everything else
pubkey auth brings in.
ok dtucker@
OpenBSD-Commit-ID: 3fdca2c61ad97dc1b8d4a7346816f83dc4ce2217
* fuzzer for authorized_keys parsing
mostly redundant to authopt_fuzz, but it's sensitive code so IMO it
makes sense to test this layer too
* Test against LibreSSL 3.5.3.
* Test against OpenSSL 1.1.1o and 3.0.3.
* fix some bugs in the fuzzer
* upstream: keywords ref ssh_config.5;
from caspar schutijser
OpenBSD-Commit-ID: f146a19d7d5c9374c3b9c520da43b2732d7d1a4e
* upstream: ssh-keygen: implement "verify-required" certificate option.
This was already documented when support for user-verified FIDO
keys was added, but the ssh-keygen(1) code was missing.
ok djm@
OpenBSD-Commit-ID: f660f973391b593fea4b7b25913c9a15c3eb8a06
* upstream: ssh-keygen -A: do not generate DSA keys by default.
Based on github PR#303 from jsegitz with man page text from jmc@, ok markus@
djm@
OpenBSD-Commit-ID: 5c4c57bdd7063ff03381cfb6696659dd3f9f5b9f
* upstream: Add period at end of "not known by any other names"
message. github PR#320 from jschauma, ok djm@
OpenBSD-Commit-ID: bd60809803c4bfd3ebb7c5c4d918b10e275266f2
* upstream: Add missing *-sk types to ssh-keyscan manpage. From
skazi0 via github PR#294.
OpenBSD-Commit-ID: fda2c869cdb871f3c90a89fb3f985370bb5d25c0
* upstream: Make SetEnv directives first-match-wins in both
sshd_config and sshd_config; previously if the same name was reused then the
last would win (which is the opposite to how the config is supposed to work).
While there, make the ssh_config parsing more like sshd_config.
bz3438, ok dtucker
OpenBSD-Commit-ID: 797909c1e0262c0d00e09280459d7ab00f18273b
* upstream: test setenv in both client and server, test first-match-wins
too
OpenBSD-Regress-ID: 4c8804f9db38a02db480b9923317457b377fe34b
* upstream: move auth_openprincipals() and auth_openkeyfile() over to
auth2-pubkeyfile.c too; they make more sense there.
OpenBSD-Commit-ID: 9970d99f900e1117fdaab13e9e910a621b7c60ee
* upstream: make sure that UseDNS hostname lookup happens in the monitor
and not in the pledge(2)'d unprivileged process; fixes regression caused by
recent refactoring spotted by henning@
OpenBSD-Commit-ID: a089870b95101cd8881a2dff65b2f1627d13e88d
* fix possible NULL deref when built without FIDO
Analysis/fix from kircher in bz3443; ok dtucker@
* automatically enable built-in FIDO support
If libfido2 is found and usable, then enable the built-in
security key support unless --without-security-key-builtin
was requested.
ok dtucker@
* upstream: Log an error if pipe() fails while accepting a
connection. bz#3447, from vincent-openssh at vinc17 net, ok djm@
OpenBSD-Commit-ID: 9d59f19872b94900a5c79da2d57850241ac5df94
* upstream: Don't attempt to fprintf a null identity comment. From
Martin Vahlensieck via tech@.
OpenBSD-Commit-ID: 4c54d20a8e8e4e9912c38a7b4ef5bfc5ca2e05c2
* upstream: Make sure not to fclose() the same fd twice in case of an
error.
ok dtucker@
OpenBSD-Commit-ID: e384c4e05d5521e7866b3d53ca59acd2a86eef99
* upstream: make it clear that RekeyLimit applies to both transmitted
and received data. GHPR#328 from Jan Pazdziora
OpenBSD-Commit-ID: d180a905fec9ff418a75c07bb96ea41c9308c3f9
* request 1.1x API compatibility for OpenSSL >=3.x
idea/patch from Pedro Martelletto via GHPR#322; ok dtucker@
* fix broken case statement in previous
* Disable SK support if FIDO libs not found.
* Zero out LIBFIDO2 when SK support not usable.
Prevents us from trying to link them into ssh-sk-helper and failing to
build.
* upstream: Don't leak the strings allocated by order_hostkeyalgs()
and list_hostkey_types() that are passed to compat_pkalg_proposal(). Part of
github PR#324 from ZoltanFridrich, ok djm@
OpenBSD-Commit-ID: b2f6e5f60f2bba293b831654328a8a0035ef4a1b
* upstream: Roll back previous KEX changes as they aren't safe until
compat_pkalg_proposal and friends always allocate their returned strings.
Reported by Qualys.
OpenBSD-Commit-ID: 1c7a88a0d5033f42f88ab9bec58ef1cf72c81ad0
* upstream: allow arguments to sftp -D option, e.g. sftp -D
"/usr/libexec/sftp-server -el debug3"
ok markus@
OpenBSD-Commit-ID: 5a002b9f3a7aef2731fc0ffa9c921cf15f38ecce
* Update OpenSSL tests to the most recent releases.
* upstream: reflect the update to -D arg name in usage();
OpenBSD-Commit-ID: abdcde4f92b1ef094ae44210ee99d3b0155aad9c
* upstream: ignore SIGPIPE earlier in main(), specifically before
muxclient() which performs operations that could cause one; Reported by Noam
Lewis via bz3454, ok dtucker@
OpenBSD-Commit-ID: 63d8e13276869eebac6d7a05d5a96307f9026e47
* upstream: Always return allocated strings from the kex filtering so
that we can free them later. Fix one leak in compat_kex_proposal. Based on
github PR#324 from ZoltanFridrich with some simplications by me. ok djm@
OpenBSD-Commit-ID: 9171616da3307612d0ede086fd511142f91246e4
* upstream: Don't leak the strings allocated by order_hostkeyalgs()
and list_hostkey_types() that are passed to compat_pkalg_proposal(). Part of
github PR#324 from ZoltanFridrich, ok djm@
This is a roll-forward of the previous rollback now that the required
changes in compat.c have been done.
OpenBSD-Commit-ID: c7cd93730b3b9f53cdad3ae32462922834ef73eb
* upstream: bump up loglevel from debug to info when unable to open
authorized keys/principals file for errno != ENOENT; bz2042 ok dtucker
OpenBSD-Commit-ID: e79aa550d91ade6a80f081bda689da24c086d66b
* Skip select+rlimit check if sandboxing is disabled
It's not needed in that case, and the test can fail when being built
with some compiler memory sanitizer flags. bz#3441
* upstream: use consistent field names (s/char/byte)
in format description
OpenBSD-Commit-ID: 3de33572733ee7fcfd7db33d37db23d2280254f0
* upstream: Remove leftover line.
Remove extra line leftover from merge conflict. ok djm@
OpenBSD-Commit-ID: 460e2290875d7ae64971a7e669c244b1d1c0ae2e
* Move checks for pollfd.fd and nfds_t.
Move the checks for struct pollfd.fd and nfds_t to before the sandboxing
checks. This groups all the sandbox checks together so we can skip them
all when sandboxing is disabled.
* Skip all rlimit tests when sandboxing disabled.
The rlimit tests can hang when being run with some compiler sanitizers
so skip all of them if sandbox=no.
* Add clang sanitizer tests.
* upstream: Add TEST_REGRESS_CACHE_DIR.
If set, it is used to cache regress test names that have succeeded and
skip those on a re-run.
OpenBSD-Regress-ID: a7570dd29a58df59f2cca647c3c2ec989b49f247
* Move sanitizer logs into regress for collection.
* Add GCC address sanitizer build/test.
* Update sanitizer test targets:
- remove clang-sanitize-memory for now. It takes so long that the test
times out.
- add gcc sanitize-address and sanitize-undefined test targets.
* Test against openssl-3.0.5.
* Move unset to before we set anything.
* Refuse to use OpenSSL 3.0.4 due to potential RCE.
OpenSSL has a potential RCE in its RSA implementation (CVE-2022-2274)
so refuse to use that specific version.
* Capture stderr output from configure.
* Only refuse to use OpenSSL 3.0.4 on x86_64.
The potential RCE only impacts x86_64, so only refuse to use it if we're
targetting a potentially impacted architecture. ok djm@
* Remove special casing of crypt().
Configure goes to some lengths to pick crypt() from either libcrypt
or OpenSSL's libcrypto because they can more or less featureful (eg
supporting md5-style passwords).
OpenSSL removed its crypt() interface in 2002:
https://github.com/openssl/openssl/commit/69deec58 so these hijinks
should no longer be necessary. This also only links sshd with libcrypt
which is the only thing that needs it. ok djm@
* Clarify README.md text.
Clarify the text about the implications of building without OpenSSL, and
prefix the "configure --help" example command with a "./" so it's likely
to work as-is in more shells. From bz#3461.
* Split README.platform into its own line.
README.platform has general platform-specific information, having it
following text about FIDO2 on the same line could imply that it only
has information about FIDO2.
* Return ERANGE from getcwd() if buffer size is 1.
If getcwd() is supplied a buffer size of exactly 1 and a path of "/", it
could result in a nul byte being written out of array bounds. POSIX says
it should return ERANGE if the path will not fit in the available buffer
(with terminating nul). 1 byte cannot fit any possible path with its nul,
so immediately return ERANGE in that case.
OpenSSH never uses getcwd() with this buffer size, and all current
(and even quite old) platforms that we are currently known to work
on have a native getcwd() so this code is not used on those anyway.
Reported by Qualys, ok djm@
* Remove unintended changes.
I inadvertently included a couple of local changes with the OpenSSL
3.0.4 change. Revert, anything that should be there will be committed
separately.
* Add AUDIT_ARCH_PPC to supported seccomp arches.
Patch from dries.deschout at dodeco.eu.
* Rename bbone test target to ARM.
* Move vmshutdown to first step.
If a previous run on a physical runner has failed to clean up, the next
run will fail because it'll try to check out the code to a broken
directory mount. Make cleanup the first step.
* upstream: pull passphrase reading and confirmation into a separate
function so it can be used for FIDO2 PINs; no functional change
OpenBSD-Commit-ID: bf34f76b8283cc1d3f54633e0d4f13613d87bb2f
* upstream: when enrolling a resident key on a security token, check
if a credential with matching application and user ID strings already exists.
if so, prompt the user for confirmation before overwriting the credential.
patch from Pedro Martelletto via GHPR329
NB. cranks SSH_SK_VERSION_MAJOR, so any third-party FIDO middleware
implementations will need to adjust
OpenBSD-Commit-ID: e45e9f1bf2b2f32d9850669e7a8dbd64acc5fca4
* upstream: sk-usbhid: preserve error code returned by key_lookup()
it conveys useful information, such as the supplied pin being wrong.
Part of GHPR329 from Pedro Martelletto
OpenBSD-Commit-ID: c0647eb9290f793add363d81378439b273756c1b
* upstream: ssh-keygen: fix touch prompt, pin retries;
part of GHPR329 from Pedro Martelletto
OpenBSD-Commit-ID: 75d1005bd2ef8f29fa834c90d2684e73556fffe8
* crank SSH_SK_VERSION_MAJOR in sk-dummy.so
* Skip scp3 test if there's no scp on remote path.
scp -3 ends up using the scp that's in the remote path and will fail if
one is not available. Based on a patch from rapier at psc.edu.
* Convert "have_prog" function into "which".
"which" and its behaviour is not standardized, so convert the existing
have_prog function into "which" so we can rely on it being available
and what its semantics are. Add a have_prog wrapper that maintains the
existing behaviour.
* upstream: Test TEST_SSH_ELAPSED_TIMES for empty string not
executable. No-op on most platforms but should prevent warnings in -portable
on systems that don't have 'date %s'.
OpenBSD-Regress-ID: e39d79867b8065e33d0c5926fa1a31f85659d2a4
* upstream: Restore missing "!" in TEST_SSH_ELAPSED_TIMES test.
OpenBSD-Regress-ID: 38783f9676ec348c5a792caecee9a16e354b37b0
* Remove workarounds for OpenSSL missing AES-GCM.
We have some compatibility hacks that were added to support OpenSSL
versions that do not support AES GCM mode. Since that time, however,
the minimum OpenSSL version that we support has moved to 1.0.1 which
*does* have GCM, so this is no longer needed. ok djm@
* Remove workarounds for OpenSSL missing AES-CTR.
We have some compatibility hacks that were added to support OpenSSL
versions that do not support AES CTR mode. Since that time, however,
the minimum OpenSSL version that we support has moved to 1.0.1 which
*does* have CTR, so this is no longer needed. ok djm@
* Do not link scp, sftp and sftp-server w/ zlib.
Some of our binaries (eg sftp, sftp-server, scp) do not interact with
the channels code and thus do use libraries such as zlib and libcrypto
although they are linked with them. This adds a CHANNELLIBS and starts
by moving zlib into it, which means the aformentioned binaries are no
longer linked against zlib. ok djm@
* Group libcrypto and PRNGD checks together.
They're related more than the libcrypt or libiaf checks which are
currently between them. ok djm@
* Remove seed_rng calls from scp, sftp, sftp-server.
These binaries don't use OpenSSL's random functions. The next step
will be to stop linking them against libcrypto. ok djm@
* Move libcrypto into CHANNELLIBS.
This will result in sftp, sftp-server and scp no longer being linked
against libcrypto. ok djm@
* Move stale-configure check as early as possible.
We added a check in Makefile to catch the case where configure needs to
be rebuilt, however this did not happen until a build was attempted in
which case all of the work done by configure was wasted. Move this check
to the start of configure to catch it as early as possible. ok djm@
* Remove deprecated MacOS 10.15 runners.
* upstream: avoid double-free in error path introduced in r1.70; report
and fix based on GHPR#332 by v-rzh ok dtucker@
OpenBSD-Commit-ID: 3d21aa127b1f37cfc5bdc21461db369a663a951f
* Include CHANNEL and FIDO2 libs in configure output
* Factor out getrnd() and rename to getentropy().
Factor out the arc4random seeding into its own file and change the
interface to match getentropy. Use native getentropy if available.
This will make it easier to resync OpenBSD changes to arc4random.
Prompted by bz#3467, ok djm@.
* compat code for fido_dev_is_winhello()
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* check_sk_options: add temporary WinHello workaround
Up to libfido 1.10.0, WinHello advertises "clientPin" rather
than "uv" capability. This is fixed in 1.11.0. For the time
being, workaround it here.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* sk_sign: set FIDO2 uv attribute explicitely for WinHello
WinHello via libfido2 performs user verification by default.
However, if we stick to that, there's no way to differentiate
between keys created with or without "-O verify-required".
Set FIDO2 uv attribute explicitely to FIDO_OPT_FALSE, then check
if user verification has been requested.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* upstream: don't prompt for FIDO passphrase before attempting to enroll
the credential, just let the enroll operating fail and we'll attempt to get a
PIN anyway. Might avoid some unneccessary PIN prompts.
Part of GHPR#302 from Corinna Vinschen; ok dtucker@
OpenBSD-Commit-ID: bd5342ffc353ee37d39617906867c305564d1ce2
* Give unused param a name.
Fixes builds on platforms that do have fido2 but don't have
fido_dev_is_winhello.
* Actually put HAVE_STDINT_H around the stdint.h.
* Rename our getentropy to prevent possible loops.
Since arc4random seeds from getentropy, and we use OpenSSL for that
if enabled, there's the possibility that if we build on a system that
does not have getentropy then run on a system that does have it, then
OpenSSL could end up calling our getentropy and getting stuck in a loop.
Pointed out by deraadt@, ok djm@
* Test hostbased auth on github runners.
* fix SANDBOX_SECCOMP_FILTER_DEBUG
* Fix conditional for running hostbased tests.
* upstream: allow certificate validity intervals, sshsig verification
times and authorized_keys expiry-time options to accept dates in the UTC time
zone in addition to the default of interpreting them in the system time zone.
YYYYMMDD and YYMMDDHHMM[SS] dates/times will be interpreted as UTC if
suffixed with a 'Z' character.
Also allow certificate validity intervals to be specified in raw
seconds-since-epoch as hex value, e.g. -V 0x1234:0x4567890. This
is intended for use by regress tests and other tools that call
ssh-keygen as part of a CA workflow.
bz3468 ok dtucker
OpenBSD-Commit-ID: 454db1cdffa9fa346aea5211223a2ce0588dfe13
* upstream: add some tests for parse_absolute_time(), including cases
where it is forced to the UTC timezone. bz3468 ok dtucker
OpenBSD-Regress-ID: ea07ca31c2f3847a38df028ca632763ae44e8759
* Skip hostbased during Valgrind tests.
Valgrind doesn't let ssh exec ssh-keysign (because it's setuid) so skip
it during the Valgrind based tests.
See https://bugs.kde.org/show_bug.cgi?id=119404 for a discussion of this
(ironically there the problematic binary was ssh(1) back when it could
still be setuid).
* Rerun tests if any .github config file changes.
* Add a timegm implementation from Heimdal via Samba.
Fixes build on (at least Solaris 10).
* Replace deprecated ubuntu-18.04 runners with 22.04
* upstream: sftp-server: support home-directory request
Add support to the sftp-server for the home-directory extension defined
in draft-ietf-secsh-filexfer-extensions-00. This overlaps a bit with the
existing expand-path@openssh.com, but uses a more official protocol name,
and so is a bit more likely to be implemented by non-OpenSSH clients.
From Mike Frysinger, ok dtucker@
OpenBSD-Commit-ID: bfc580d05cc0c817831ae7ecbac4a481c23566ab
* fido_dev_is_winhello: return 0, not "false"
"false" is not used anywhere in OpenSSH, so return 0 like
everywhere else.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* Revert "check_sk_options: add temporary WinHello workaround"
Cygwin now comes with libfido2 1.11.0, so this workaround
isn't required anymore.
This reverts commit 242c044ab111a37aad3b0775727c36a4c5f0102c.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* upstream: use .Cm for "sign"; from josiah frentsos
OpenBSD-Commit-ID: 7f80a53d54857ac6ae49ea6ad93c5bd12231d1e4
* upstream: add an extra flag to sk_probe() to indicate whether we're
probing for a FIDO resident key or not. Unused here, but will make like
easier for portable
OpenBSD-Commit-ID: 432c8ff70e270378df9dbceb9bdeaa5b43b5a832
* on Cygwin, prefer WinHello FIDO device
If no FIDO device was explictly specified, then prefer the
windows://hello FIDO device. An exception to this is when
probing resident FIDO keys, in which case hardware FIDO
devices are preferred.
* Check for perms to run agent-getpeereid test.
Ubuntu 22.04 defaults to private home dirs which prevents "nobody"
running ssh-add during the agent-getpeereid test. Check for this and
add the necessary permissions.
* upstream: double free() in error path; from Eusgor via GHPR333
OpenBSD-Commit-ID: 39f35e16ba878c8d02b4d01d8826d9b321be26d4
* Add Cygwin (on windows-2019) test target.
In addition to installing the requisite Cygwin packages, we also need to
explicitly invoke "sh" for steps that run other scripts since the runner
environment doesn't understand #! paths.
* Add a bit more debug output.
* Fix cygwin conditional steps.
* upstream: Strictly enforce the maximum allowed SSH2 banner size in
ssh-keyscan and prevent a one-byte buffer overflow. Patch from Qualys, ok
djm@
OpenBSD-Commit-ID: 6ae664f9f4db6e8a0589425f74cd0bbf3aeef4e4
* upstream: remove incorrect check that can break enrolling a
resident key (introduced in r1.40)
OpenBSD-Commit-ID: 4cab364d518470e29e624af3d3f9ffa9c92b6f01
* upstream: attemp FIDO key signing without PIN and use the error
code returned to fall back only if necessary. Avoids PIN prompts for FIDO
tokens that don't require them; part of GHPR#302
OpenBSD-Commit-ID: 4f752aaf9f2e7c28bcaaf3d4f8fc290131bd038e
* Install Cygwin packages based on OS not config.
* initial list of allowed signers
* upstream: whitespace
OpenBSD-Commit-ID: d297e4387935d4aef091c5e9432578c2e513f538
* upstream: whitespace
OpenBSD-Commit-ID: a5d015efbfd228dc598ffdef612d2da3a579e5d8
* Add cygwin-release test target.
This also moves the cygwin package install from the workflow file to
setup_ci.sh so that we can install different sets of Cygwin packages
for different test configs.
* Add Windows 2022 test targets.
* Add libcrypt-devel to cygwin-release deps.
Based on feedback from vinschen at redhat.com.
* cross-sign allowed_signers with PGP key
Provides continuity of trust from legacy PGP release key to
the SSHSIG signing keys that we will use henceforth for git
signing.
* additional keys
* upstream: whitespace
OpenBSD-Commit-ID: c2bcbf93610d3d62ed206cdf9bf9ff98c6aaf232
* Move sftp from valgrind-2 to 3 to rebalance.
* upstream: sk-usbhid: fix key_lookup() on tokens with built-in UV
explicitly test whether the token performs built-in UV (e.g. biometric
tokens) and enable UV in that case. From Pedro Martelletto via GHPR#388
OpenBSD-Commit-ID: 007eb7e387d27cf3029ab06b88224e03eca62ccd
* Remove arc4random_uniform from arc4random.c
This was previously moved into its own file (matching OpenBSD) which
prematurely committed in commit 73541f2.
* Move OPENBSD ORIGINAL marker.
Putting this after the copyright statement (which doesn't change)
instead of before the version identifier (which does) prevents merge
conflicts when resyncing changes.
* Resync arc4random with OpenBSD.
This brings us up to current, including djm's random-reseeding change,
as prompted by logan at cyberstorm.mu in bz#3467. It brings the
platform-specific hooks from LibreSSL Portable, simplified to match our
use case. ok djm@.
* Remove DEF_WEAK, it's already in defines.h.
* openbsd-compat/bsd-asprintf: add <stdio.h> include for vsnprintf
Fixes the following build failure with Clang 15 on musl:
```
bsd-asprintf.c:51:8: error: call to undeclared library function 'vsnprintf' with type 'int (char *, unsigned long, const char *, struct __va_list_tag *)'; ISO C99 and laterclang -O2 -pipe -fdiagnostics-color=always -frecord-gcc-switches -pipe -Wunknown-warning-option -Qunused-arguments -Wall -Wpointer-arith -Wuninitialized -Wsign-compare -Wformat-security -Wsizeof-pointer-memaccess -Wno-pointer-sign -Wno-unused-result -Wmisleading-indentation -Wbitwise-instead-of-logical -fno-strict-aliasing -mretpoline -ftrapv -fzero-call-used-regs=all -fno-builtin-memset -fstack-protector-strong -fPIE -I. -I. -D_XOPEN_SOURCE=600 -D_BSD_SOURCE -D_DEFAULT_SOURCE -DSSHDIR=\"/etc/ssh\" -D_PATH_SSH_PROGRAM=\"/usr/bin/ssh\" -D_PATH_SSH_ASKPASS_DEFAULT=\"/usr/lib/misc/ssh-askpass\" -D_PATH_SFTP_SERVER=\"/usr/lib/misc/sftp-server\" -D_PATH_SSH_KEY_SIGN=\"/usr/lib/misc/ssh-keysign\" -D_PATH_SSH_PKCS11_HELPER=\"/usr/lib/misc/ssh-pkcs11-helper\" -D_PATH_SSH_SK_HELPER=\"/usr/lib/misc/ssh-sk-helper\" -D_PATH_SSH_PIDDIR=\"/run\" -D_PATH_PRIVSEP_CHROOT_DIR=\"/var/empty\" -DHAVE_CONFIG_H -c cipher-aes.c -o cipher-aes.o
do not support
implicit function declarations [-Wimplicit-function-declaration]
ret = vsnprintf(string, INIT_SZ, fmt, ap2);
^
bsd-asprintf.c:51:8: note: include the header <stdio.h> or explicitly provide a declaration for 'vsnprintf'
1 error generated.
```
* upstream: notifier_complete(NULL, ...) is a noop, so no need to test
that ctx!=NULL; from Corinna Vinschen
OpenBSD-Commit-ID: ade2f2e9cc519d01a586800c25621d910bce384a
* fix pester test failures
* upstream: fix repeated words ok miod@ jmc@
OpenBSD-Commit-ID: 6765daefe26a6b648cc15cadbbe337596af709b7
* upstream: .Li -> .Vt where appropriate; from josiah frentsos,
tweaked by schwarze
ok schwarze
OpenBSD-Commit-ID: 565046e3ce68b46c2f440a93d67c2a92726de8ed
* upstream: ssh-agent: attempt FIDO key signing without PIN and use
the error to determine whether a PIN is required and prompt only if
necessary. from Corinna Vinschen
OpenBSD-Commit-ID: dd6be6a0b7148608e834ee737c3479b3270b00dd
* upstream: a little extra debugging
OpenBSD-Commit-ID: edf1601c1d0905f6da4c713f4d9cecc7d1c0295a
* upstream: sk_enroll: never drop SSH_SK_USER_VERIFICATION_REQD flag
from response
Now that all FIDO signing calls attempt first without PIN and then
fall back to trying PIN only if that attempt fails, we can remove the
hack^wtrick that removed the UV flag from the keys returned during
enroll.
By Corinna Vinschen
OpenBSD-Commit-ID: 684517608c8491503bf80cd175425f0178d91d7f
* define HAVE_KILLPG
* upstream: sftp: Don't attempt to complete arguments for
non-existent commands
If user entered a non-existent command (e.g. because they made a
typo) there is no point in trying to complete its arguments. Skip
calling complete_match() if that's the case.
From Michal Privoznik
OpenBSD-Commit-ID: cf39c811a68cde2aeb98fc85addea4000ef6b07a
* upstream: sftp: Be a bit more clever about completions
There are commands (e.g. "get" or "put") that accept two
arguments, a local path and a remote path. However, the way
current completion is written doesn't take this distinction into
account and always completes remote or local paths.
By expanding CMD struct and "cmds" array this distinction can be
reflected and with small adjustment to completer code the correct
path can be completed.
By Michal Privoznik, ok dtucker@
OpenBSD-Commit-ID: 1396d921c4eb1befd531f5c4a8ab47e7a74b610b
* upstream: correct error value
OpenBSD-Commit-ID: 780efcbad76281f11f14b2a5ff04eb6db3dfdad4
* upstream: actually hook up restrict_websafe; the command-line flag
was never actually used. Spotted by Matthew Garrett
OpenBSD-Commit-ID: 0b363518ac4c2819dbaa3dfad4028633ab9cdff1
* upstream: Add a sshkey_check_rsa_length() call for checking the
length of an RSA key; ok markus@
OpenBSD-Commit-ID: de77cd5b11594297eda82edc594b0d32b8535134
* upstream: add a RequiredRSASize for checking RSA key length in
ssh(1). User authentication keys that fall beneath this limit will be
ignored. If a host presents a host key beneath this limit then the connection
will be terminated (unfortunately there are no fallbacks in the protocol for
host authentication).
feedback deraadt, Dmitry Belyavskiy; ok markus@
OpenBSD-Commit-ID: 430e339b2a79fa9ecc63f2837b06fdd88a7da13a
* upstream: Add RequiredRSASize for sshd(8); RSA keys that fall
beneath this limit will be ignored for user and host-based authentication.
Feedback deraadt@ ok markus@
OpenBSD-Commit-ID: 187931dfc19d51873df5930a04f2d972adf1f7f1
* upstream: better debugging for connect_next()
OpenBSD-Commit-ID: d16a307a0711499c971807f324484ed3a6036640
* upstream: sftp-server(8): add a "users-groups-by-id@openssh.com"
extension request that allows the client to obtain user/group names that
correspond to a set of uids/gids.
Will be used to make directory listings more useful and consistent
in sftp(1).
ok markus@
OpenBSD-Commit-ID: 7ebabde0bcb95ef949c4840fe89e697e30df47d3
* upstream: extend sftp-common.c:extend ls_file() to support supplied
user/group names; ok markus@
OpenBSD-Commit-ID: c70c70498b1fdcf158531117e405b6245863bfb0
* upstream: sftp client library support for
users-groups-by-id@openssh.com; ok markus@
OpenBSD-Commit-ID: ddb2f33a2da6349a9a89a8b5bcb9ca7c999394de
* upstream: use users-groups-by-id@openssh.com sftp-server extension
(when available) to fill in user/group names for directory listings.
Implement a client-side cache of see uid/gid=>user/group names. ok markus@
OpenBSD-Commit-ID: f239aeeadfa925a37ceee36ee8b256b8ccf4466e
* avoid Wuninitialized false positive in gcc-12ish
* no need for glob.h here
it also causes portability problems
* add debug on appveyor
* add sleep to pester test
* upstream: add RequiredRSASize to the list of keywords accepted by
-o; spotted by jmc@
OpenBSD-Commit-ID: fe871408cf6f9d3699afeda876f8adbac86a035e
* upstream: Fix typo. From AlexanderStohr via github PR#343.
OpenBSD-Commit-ID: a134c9b4039e48803fc6a87f955b0f4a03181497
* upstream: openssh-9.1
OpenBSD-Commit-ID: 5a467b2ee81da01a86adf1ad93b62b1728494e56
* crank versions in RPM spec files
* update release notes URL
* update .depend
* fix 9.1 compilation errors
* disable -p pester tests due to unreliability on older Windows versions
* remove extra sleep time from debugging scp pester tests
* modify -p tests to only run for Windows OS version 10 and above
* add windows specific code back into method moved from auth.c to auth2-pubkeyfile.c
* add preprocessor for WinHello
* revert preprocessor definition for winhello
* add windows preprocessor definition in key_lookup
* remove rdp block from appveyor since we are no longer debugging
* add ifdef to sftp-server.c
* make key_lookup compatible with winhello
* appveyor.yml
* increase debug of failing pester test
* add #ifdef SUPPORT_CRLF back into auth_check_principals_line method that was moved/renamed
* modify new scp.sh tests for windows
* remove in place tests from scp.sh
* remove rdp debug from appveyor
* retrigger appveyor
* change check of OS version in scp test
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
Co-authored-by: djm@openbsd.org <djm@openbsd.org>
Co-authored-by: Damien Miller <djm@mindrot.org>
Co-authored-by: Darren Tucker <dtucker@dtucker.net>
Co-authored-by: naddy@openbsd.org <naddy@openbsd.org>
Co-authored-by: dtucker@openbsd.org <dtucker@openbsd.org>
Co-authored-by: tj@openbsd.org <tj@openbsd.org>
Co-authored-by: millert@openbsd.org <millert@openbsd.org>
Co-authored-by: jmc@openbsd.org <jmc@openbsd.org>
Co-authored-by: florian@openbsd.org <florian@openbsd.org>
Co-authored-by: markus@openbsd.org <markus@openbsd.org>
Co-authored-by: Tobias Heider <me@tobhe.de>
Co-authored-by: anton@openbsd.org <anton@openbsd.org>
Co-authored-by: Tim Rice <tim@multitalents.net>
Co-authored-by: tobhe@openbsd.org <tobhe@openbsd.org>
Co-authored-by: Corinna Vinschen <vinschen@redhat.com>
Co-authored-by: Sam James <sam@gentoo.org>
Co-authored-by: jsg@openbsd.org <jsg@openbsd.org>
2022-11-02 17:06:45 +01:00
|
|
|
is_agent = 1;
|
2019-10-31 22:18:28 +01:00
|
|
|
} else {
|
|
|
|
/* Load the private key from the file. */
|
|
|
|
if ((prv = load_identity_file(id)) == NULL)
|
|
|
|
return SSH_ERR_KEY_NOT_FOUND;
|
|
|
|
if (id->key != NULL && !sshkey_equal_public(prv, id->key)) {
|
2020-10-18 13:32:01 +02:00
|
|
|
error_f("private key %s contents do not match public",
|
2021-04-03 08:18:40 +02:00
|
|
|
id->filename);
|
2019-10-31 22:18:28 +01:00
|
|
|
r = SSH_ERR_KEY_NOT_FOUND;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
sign_key = prv;
|
Merge 9.1 (#626)
* upstream: fix poll() spin when a channel's output fd closes without
data in the channel buffer. Introduce more exact packing of channel fds into
the pollfd array. fixes bz3405 and bz3411; ok deraadt@ markus@
OpenBSD-Commit-ID: 06740737849c9047785622ad5d472cb6a3907d10
* upstream: select post-quantum KEX
sntrup761x25519-sha512@openssh.com as the default; ok markus@
OpenBSD-Commit-ID: f02d99cbfce22dffec2e2ab1b60905fbddf48fb9
* upstream: add support for the "corp-data" protocol extension to
allow server-side copies to be performed without having to go via the client.
Patch by Mike Frysinger, ok dtucker@
OpenBSD-Commit-ID: 00aa510940fedd66dab1843b58682de4eb7156d5
* upstream: add a sftp client "cp" command that supports server-side
copying of files. Useful for this task and for testing the copy-data
extension. Patch from Mike Frysinger; ok dtucker@
OpenBSD-Commit-ID: 1bb1b950af0d49f0d5425b1f267e197aa1b57444
* depend
* Skip slow tests on (very) slow test targets.
* Set Makefile SHELL as determined by configure.
This should improve compatibility for users with non-POSIX shells. If
using Makefile.in directly (eg make -f Makefile.in distprep) then SHELL
will need to be specified on the command line (along with MANFMT in that
particular case). ok djm@
* Use bash or ksh if available for SH in Makefile.
* Increase test timeout to allow slow VMs to finish
* Only run regression tests on slow VMs.
* Only return events from ppoll that were requested.
If the underlying system's select() returns bits that were not in the
request set, our ppoll() implementation can return revents for events
not requested, which can apparently cause a hang. Only return revents
for activity in the requested event set. bz#3416, analysis and fix by
yaroslav.kuzmin at vmssoftware com, ok djm@
* Specify TEST_SHELL=bash on AIX.
The system shells cause the agent-restrict test to fail due to some
quoting so explicitly specify bash until we can get configure to
autmatically work around that.
* Disable security key on fbsd6 test host.
* upstream: man pages: add missing commas between subordinate and
main clauses
jmc@ dislikes a comma before "then" in a conditional, so leave those
untouched.
ok jmc@
OpenBSD-Commit-ID: 9520801729bebcb3c9fe43ad7f9776ab4dd05ea3
* upstream: ssh: document sntrup761x25519-sha512@openssh.com as
default KEX
OpenBSD-Commit-ID: 12545bfa10bcbf552d04d9d9520d0f4e98b0e171
* upstream: openssh-9.0
OpenBSD-Commit-ID: 0dfb461188f4513ec024c1534da8c1ce14c20b64
* update version numbers for release
* update build-aux files to match autoconf-2.71
i.e. config.guess, config.sub and install-sh
* Revert "update build-aux files to match autoconf-2.71"
This reverts commit 0a8ca39fac6ad19096b6c263436f8b2dd51606f2.
It turns out that the checked-in copies of these files are actually newer
than autoconf-2.71's copies, so this was effectively a downgrade.
Spotted by Bo Anderson via github
* upstream: two defensive changes from Tobias Stoeckmann via GHPR287
enforce stricter invarient for sshbuf_set_parent() - never allow
a buffer to have a previously-set parent changed.
In sshbuf_reset(), if the reallocation fails, then zero the entire
buffer and not the (potentially smaller) default initial alloc size.
OpenBSD-Commit-ID: 14583203aa5d50ad38d2e209ae10abaf8955e6a9
* upstream: Note that curve25519-sha256 was later published in
RFC8731. ok djm@
OpenBSD-Commit-ID: 2ac2b5d642d4cf5918eaec8653cad9a4460b2743
* upstream: clear io_want/io_ready flags at start of poll() cycle;
avoids plausible spin during rekeying if channel io_want flags are reused
across cycles. ok markus@ deraadt@
OpenBSD-Commit-ID: 91034f855b7c73cd2591657c49ac30f10322b967
* Retire fbsd6 test VM.
It's long since out of support, relatively slow (it's i686) and the
compiler has trouble with PIE.
* Resync moduli.5 with upstream.
1.18: remove duplicate publication year; carsten dot kunze at arcor dot de
1.19: ssh-keygen's -G/-T have been replaced with -M generate/screen.
* upstream: Correct path for system known hosts file in description
of IgnoreUserKnownHosts. Patch from Martin Vahlensieck via tech@
OpenBSD-Commit-ID: 9b7784f054fa5aa4d63cb36bd563889477127215
* upstream: list the correct version number
for when usage of the sftp protocol became default and fix a typo
from ed maste
OpenBSD-Commit-ID: 24e1795ed2283fdeacf16413c2f07503bcdebb31
* upstream: Import regenerated moduli
OpenBSD-Commit-ID: f9a0726d957cf10692a231996a1f34e7f9cdfeb0
* upstream: Try to continue running local I/O for channels in state
OPEN during SSH transport rekeying. The most visible benefit is that it
should make ~-escapes work in the client (e.g. to exit) if the connection
happened to have stalled during a rekey event. Based work by and ok dtucker@
OpenBSD-Commit-ID: a66e8f254e92edd4ce09c9f750883ec8f1ea5f45
* upstream: Import regenerated moduli
OpenBSD-Commit-ID: f9a0726d957cf10692a231996a1f34e7f9cdfeb0
* upstream: regression test for sftp cp command
OpenBSD-Regress-ID: c96bea9edde3a384b254785e7f9b2b24a81cdf82
* upstream: Simplify forward-control test.
Since we no longer need to support SSH1 we don't need to run shell
commands on the other end of the connection and can use ssh -N instead.
This also makes the test less racy.
OpenBSD-Regress-ID: 32e94ce272820cc398f30b848b2b0f080d10302c
* upstream: Use ssh -f and ControlPersist ..
to start up test forwards and ssh -O stop to shut them down intead of
sleep loops. This speeds up the test by an order of magnitude.
OpenBSD-Regress-ID: eb3db5f805100919b092a3b2579c611fba3e83e7
* upstream: It looks like we can't completely avoid
waiting for processes to exit so retrieve the pid via controlmaster and
use that.
OpenBSD-Regress-ID: 8246f00f22b14e49d2ff1744c94897ead33d457b
* Cache timezone data in capsicum sandbox.
From emaste at freebsd.org, originally part of FreeBSD commit r339216
/ fc3c19a9 with autoconf bits added by me.
* Include stdlib.h for free() prototype.
... which is used inside the CUSTOM_SYS_AUTH_GET_LASTLOGIN_MSG block.
* Update OpenSSL and LibreSSL versions in tests.
* Add debian-riscv64 test target.
* upstream: Avoid an unnecessary xstrdup in rm_env() when matching
patterns. Since match_pattern() doesn't modify its arguments (they are
const), there is no need to make an extra copy of the strings in
options->send_env. From Martin Vahlensieck
OpenBSD-Commit-ID: 2c9db31e3f4d3403b49642c64ee048b2a0a39351
* upstream: Add missing includes of stdlib.h and stdint.h. We need
stdlib.h for malloc(3) and stdint.h for SIZE_MAX. Unlike the other xmss
files, ssh-xmss.c does not include xmss_commons.h so ssh-xmss.c must include
those headers itself. From Martin Vahlensieck
OpenBSD-Commit-ID: 70e28a9818cee3da1be2ef6503d4b396dd421e6b
* upstream: Remove unnecessary includes: openssl/hmac.h and
openssl/evp.h. From Martin Vahlensieck.
OpenBSD-Commit-ID: a6debb5fb0c8a44e43e8d5ca7cc70ad2f3ea31c3
* upstream: Check sshauthopt_new() for NULL. bz#3425, from
tessgauthier at microsoft.com. ok djm@
OpenBSD-Commit-ID: af0315bc3e44aa406daa7e0ae7c2d719a974483f
* upstream: Add authfd path to debug output. ok markus@
OpenBSD-Commit-ID: f735a17d1a6f2bee63bfc609d76ef8db8c090890
* upstream: avoid printing hash algorithm twice; from lucas AT sexy.is
OpenBSD-Commit-ID: 9d24671e10a84141b7c504396cabad600e47a941
* upstream: fix memleak on session-bind path; from Pedro Martelletto, ok
dtucker@
OpenBSD-Commit-ID: e85899a26ba402b4c0717b531317e8fc258f0a7e
* upstream: Don't leak SK device. Patch from Pedro Martelletto via
github PR#316. ok djm@
OpenBSD-Commit-ID: 17d11327545022e727d95fd08b213171c5a4585d
* upstream: mention that the helpers are used by ssh(1), ssh-agent(1)
and ssh-keygen(1). Previously only ssh(1) was mentioned. From Pedro
Martelletto
OpenBSD-Commit-ID: 30f880f989d4b329589c1c404315685960a5f153
* Remove now-empty int32_minmax.inc.
* Only run tests when source files change.
Also run tests on changes to V_9_0 branch.
* Add Mac OS X 12 test target.
* upstream: be stricter in which characters will be accepted in
specifying a mask length; allow only 0-9. From khaleesicodes via GHPR#278; ok
dtucker@
OpenBSD-Commit-ID: e267746c047ea86665cdeccef795a8a56082eeb2
* upstream: fix some integer overflows in sieve_large() that show up when
trying to generate modp groups > 16k bits. Reported via GHPR#306 by Bertram
Felgenhauer, but fixed in a different way. feedback/ok tb@
OpenBSD-Commit-ID: 81cbc6dd3a21c57bd6fadea10e44afe37bca558e
* upstream: remove an obsolete rsa1 format example from an example;
from megan batty
ok djm
OpenBSD-Commit-ID: db2c89879c29bf083df996bd830abfb1e70d62bf
* upstream: Add FIDO AUTHENTICATOR section and explain a bit how FIDO
works. The wording came mostly from the 8.2 OpenSSH release notes, addapted
to fit the man page. Then move the -O bits into the new section as is already
done for CERTIFICATES and MODULI GENERATION. Finally we can explain the
trade-offs of resident keys. While here, consistently refer to the FIDO
thingies as "FIDO authenticators", not "FIDO tokens".
input & OK jmc, naddy
OpenBSD-Commit-ID: dd98748d7644df048f78dcf793b3b63db9ab1d25
* upstream: make sure stdout is non-blocking; ok djm@
OpenBSD-Commit-ID: 64940fffbd1b882eda2d7c8c7a43c79368309c0d
* upstream: mux.c: mark argument as const; from Martin Vahlensieck
OpenBSD-Commit-ID: 69a1a93a55986c7c2ad9f733c093b46a47184341
* upstream: channel_new no longer frees remote_name. So update the
comment accordingly. As remote_name is not modified, it can be const as
well. From Martin Vahlensieck
OpenBSD-Commit-ID: e4e10dc8dc9f40c166ea5a8e991942bedc75a76a
* upstream: sshkey_unshield_private() contains a exact duplicate of
the code in private2_check_padding(). Pull private2_check_padding() up so the
code can be reused. From Martin Vahlensieck, ok deraadt@
OpenBSD-Commit-ID: 876884c3f0e62e8fd8d1594bab06900f971c9c85
* Add ubsan minimal testcase on OpenBSD.
As suggested by djm@.
* Note that, for now, we need variadic macros.
* Also retest OpenBSD upstream on .yml changes.
* upstream: When performing operations that glob(3) a remote path, ensure
that the implicit working directory used to construct that path escapes
glob(3) characters.
This prevents glob characters from being processed in places they
shouldn't, e.g. "cd /tmp/a*/", "get *.txt" should have the get operation
treat the path "/tmp/a*" literally and not attempt to expand it.
Reported by Lusia Kundel; ok markus@
OpenBSD-Commit-ID: 4f647f58482cbad3d58b1eab7f6a1691433deeef
* Remove duplicate bcrypt_pbkdf.o from Makefile
bcrypt_pbkdf.o is duplicated in the openbsd-compat Makefile's object
file list.
* upstream: improve error message when 'ssh-keygen -Y sign' is unable to
load a private key; bz3429, reported by Adam Szkoda ok dtucker@
OpenBSD-Commit-ID: bb57b285e67bea536ef81b1055467be2fc380e74
* upstream: Allow existing -U (use agent) flag to work with "-Y sign"
operations, where it will be interpreted to require that the private keys is
hosted in an agent; bz3429, suggested by Adam Szkoda; ok dtucker@
OpenBSD-Commit-ID: a7bc69873b99c32c42c7628ed9ea91565ba08c2f
* upstream: Remove errant apostrophe. From haruyama at queen-ml org.
OpenBSD-Commit-ID: dc6b294567cb84b384ad6ced9ca469f2bbf0bd10
* upstream: arrange for scp, when in sftp mode, to not ftruncate(3) files
early
previous behavious of unconditionally truncating the destination file
would cause "scp ~/foo localhost:" and "scp localhost:foo ~/" to
delete all the contents of their destination.
spotted by solene@ sthen@, also bz3431; ok dtucker@
OpenBSD-Commit-ID: ca39fdd39e0ec1466b9666f15cbcfddea6aaa179
* upstream: fix in-place copies; r1.163 incorrectly skipped truncation in
all cases, not just at the start of a transfer. This could cause overwrites
of larger files to leave junk at the end. Spotted by tb@
OpenBSD-Commit-ID: b189f19cd68119548c8e24e39c79f61e115bf92c
* upstream: Only run agent-ptrace.sh if gdb is available as all
architectures do not ship with gdb.
OpenBSD-Regress-ID: ec53e928803e6b87f9ac142d38888ca79a45348d
* upstream: regress test for in-place transfers and clobbering larger
files with smaller ones; would have caught last regression in scp(1)
OpenBSD-Regress-ID: 19de4e88dd3a4f7e5c1618c9be3c32415bd93bc2
* configure.ac: Add missing AC_DEFINE for caph_cache_tzdata test causing
HAVE_CAPH_CACHE_TZDATA to be missing from config.h.in.
Spotted by Bryan Drewery
* upstream: make SSHBUF_DBG/SSHBUF_TELL (off by default and only enabled
via #define) dump to stderr rather than stdout
OpenBSD-Commit-ID: 10298513ee32db8390aecb0397d782d68cb14318
* upstream: revert previous; it was broken (spotted by Theo)
OpenBSD-Commit-ID: 457c79afaca2f89ec2606405c1059b98b30d8b0d
* upstream: Note that ProxyJump also accepts the same tokens as
ProxyCommand. From pallxk via github PR#305.
OpenBSD-Commit-ID: 7115ac351b129205f1f1ffa6bbfd62abd76be7c5
* upstream: Avoid kill with -1 argument. The out_ctx label can be
reached before fork has been called. If this happens, then kill -1 would be
called, sending SIGTERM to all processes reachable by the current process.
From tobias@ and c3h2_ctf via github PR#286, ok djm@
OpenBSD-Commit-ID: 6277af1207d81202f5daffdccfeeaed4c763b1a8
* upstream: f sshpkt functions fail, then password is not cleared
with freezero. Unconditionally call freezero to guarantee that password is
removed from RAM.
From tobias@ and c3h2_ctf via github PR#286, ok djm@
OpenBSD-Commit-ID: 6b093619c9515328e25b0f8093779c52402c89cd
* upstream: refactor authorized_keys/principals handling
remove "struct ssh *" from arguments - this was only used to pass the
remote host/address. These can be passed in instead and the resulting
code is less tightly coupled to ssh_api.[ch]
ok dtucker@
OpenBSD-Commit-ID: 9d4373d013edc4cc4b5c21a599e1837ac31dda0d
* upstream: split the low-level file handling functions out from
auth2-pubkey.c
Put them in a new auth2-pubkeyfile.c to make it easier to refer to them
(e.g. in unit/fuzz tests) without having to refer to everything else
pubkey auth brings in.
ok dtucker@
OpenBSD-Commit-ID: 3fdca2c61ad97dc1b8d4a7346816f83dc4ce2217
* fuzzer for authorized_keys parsing
mostly redundant to authopt_fuzz, but it's sensitive code so IMO it
makes sense to test this layer too
* Test against LibreSSL 3.5.3.
* Test against OpenSSL 1.1.1o and 3.0.3.
* fix some bugs in the fuzzer
* upstream: keywords ref ssh_config.5;
from caspar schutijser
OpenBSD-Commit-ID: f146a19d7d5c9374c3b9c520da43b2732d7d1a4e
* upstream: ssh-keygen: implement "verify-required" certificate option.
This was already documented when support for user-verified FIDO
keys was added, but the ssh-keygen(1) code was missing.
ok djm@
OpenBSD-Commit-ID: f660f973391b593fea4b7b25913c9a15c3eb8a06
* upstream: ssh-keygen -A: do not generate DSA keys by default.
Based on github PR#303 from jsegitz with man page text from jmc@, ok markus@
djm@
OpenBSD-Commit-ID: 5c4c57bdd7063ff03381cfb6696659dd3f9f5b9f
* upstream: Add period at end of "not known by any other names"
message. github PR#320 from jschauma, ok djm@
OpenBSD-Commit-ID: bd60809803c4bfd3ebb7c5c4d918b10e275266f2
* upstream: Add missing *-sk types to ssh-keyscan manpage. From
skazi0 via github PR#294.
OpenBSD-Commit-ID: fda2c869cdb871f3c90a89fb3f985370bb5d25c0
* upstream: Make SetEnv directives first-match-wins in both
sshd_config and sshd_config; previously if the same name was reused then the
last would win (which is the opposite to how the config is supposed to work).
While there, make the ssh_config parsing more like sshd_config.
bz3438, ok dtucker
OpenBSD-Commit-ID: 797909c1e0262c0d00e09280459d7ab00f18273b
* upstream: test setenv in both client and server, test first-match-wins
too
OpenBSD-Regress-ID: 4c8804f9db38a02db480b9923317457b377fe34b
* upstream: move auth_openprincipals() and auth_openkeyfile() over to
auth2-pubkeyfile.c too; they make more sense there.
OpenBSD-Commit-ID: 9970d99f900e1117fdaab13e9e910a621b7c60ee
* upstream: make sure that UseDNS hostname lookup happens in the monitor
and not in the pledge(2)'d unprivileged process; fixes regression caused by
recent refactoring spotted by henning@
OpenBSD-Commit-ID: a089870b95101cd8881a2dff65b2f1627d13e88d
* fix possible NULL deref when built without FIDO
Analysis/fix from kircher in bz3443; ok dtucker@
* automatically enable built-in FIDO support
If libfido2 is found and usable, then enable the built-in
security key support unless --without-security-key-builtin
was requested.
ok dtucker@
* upstream: Log an error if pipe() fails while accepting a
connection. bz#3447, from vincent-openssh at vinc17 net, ok djm@
OpenBSD-Commit-ID: 9d59f19872b94900a5c79da2d57850241ac5df94
* upstream: Don't attempt to fprintf a null identity comment. From
Martin Vahlensieck via tech@.
OpenBSD-Commit-ID: 4c54d20a8e8e4e9912c38a7b4ef5bfc5ca2e05c2
* upstream: Make sure not to fclose() the same fd twice in case of an
error.
ok dtucker@
OpenBSD-Commit-ID: e384c4e05d5521e7866b3d53ca59acd2a86eef99
* upstream: make it clear that RekeyLimit applies to both transmitted
and received data. GHPR#328 from Jan Pazdziora
OpenBSD-Commit-ID: d180a905fec9ff418a75c07bb96ea41c9308c3f9
* request 1.1x API compatibility for OpenSSL >=3.x
idea/patch from Pedro Martelletto via GHPR#322; ok dtucker@
* fix broken case statement in previous
* Disable SK support if FIDO libs not found.
* Zero out LIBFIDO2 when SK support not usable.
Prevents us from trying to link them into ssh-sk-helper and failing to
build.
* upstream: Don't leak the strings allocated by order_hostkeyalgs()
and list_hostkey_types() that are passed to compat_pkalg_proposal(). Part of
github PR#324 from ZoltanFridrich, ok djm@
OpenBSD-Commit-ID: b2f6e5f60f2bba293b831654328a8a0035ef4a1b
* upstream: Roll back previous KEX changes as they aren't safe until
compat_pkalg_proposal and friends always allocate their returned strings.
Reported by Qualys.
OpenBSD-Commit-ID: 1c7a88a0d5033f42f88ab9bec58ef1cf72c81ad0
* upstream: allow arguments to sftp -D option, e.g. sftp -D
"/usr/libexec/sftp-server -el debug3"
ok markus@
OpenBSD-Commit-ID: 5a002b9f3a7aef2731fc0ffa9c921cf15f38ecce
* Update OpenSSL tests to the most recent releases.
* upstream: reflect the update to -D arg name in usage();
OpenBSD-Commit-ID: abdcde4f92b1ef094ae44210ee99d3b0155aad9c
* upstream: ignore SIGPIPE earlier in main(), specifically before
muxclient() which performs operations that could cause one; Reported by Noam
Lewis via bz3454, ok dtucker@
OpenBSD-Commit-ID: 63d8e13276869eebac6d7a05d5a96307f9026e47
* upstream: Always return allocated strings from the kex filtering so
that we can free them later. Fix one leak in compat_kex_proposal. Based on
github PR#324 from ZoltanFridrich with some simplications by me. ok djm@
OpenBSD-Commit-ID: 9171616da3307612d0ede086fd511142f91246e4
* upstream: Don't leak the strings allocated by order_hostkeyalgs()
and list_hostkey_types() that are passed to compat_pkalg_proposal(). Part of
github PR#324 from ZoltanFridrich, ok djm@
This is a roll-forward of the previous rollback now that the required
changes in compat.c have been done.
OpenBSD-Commit-ID: c7cd93730b3b9f53cdad3ae32462922834ef73eb
* upstream: bump up loglevel from debug to info when unable to open
authorized keys/principals file for errno != ENOENT; bz2042 ok dtucker
OpenBSD-Commit-ID: e79aa550d91ade6a80f081bda689da24c086d66b
* Skip select+rlimit check if sandboxing is disabled
It's not needed in that case, and the test can fail when being built
with some compiler memory sanitizer flags. bz#3441
* upstream: use consistent field names (s/char/byte)
in format description
OpenBSD-Commit-ID: 3de33572733ee7fcfd7db33d37db23d2280254f0
* upstream: Remove leftover line.
Remove extra line leftover from merge conflict. ok djm@
OpenBSD-Commit-ID: 460e2290875d7ae64971a7e669c244b1d1c0ae2e
* Move checks for pollfd.fd and nfds_t.
Move the checks for struct pollfd.fd and nfds_t to before the sandboxing
checks. This groups all the sandbox checks together so we can skip them
all when sandboxing is disabled.
* Skip all rlimit tests when sandboxing disabled.
The rlimit tests can hang when being run with some compiler sanitizers
so skip all of them if sandbox=no.
* Add clang sanitizer tests.
* upstream: Add TEST_REGRESS_CACHE_DIR.
If set, it is used to cache regress test names that have succeeded and
skip those on a re-run.
OpenBSD-Regress-ID: a7570dd29a58df59f2cca647c3c2ec989b49f247
* Move sanitizer logs into regress for collection.
* Add GCC address sanitizer build/test.
* Update sanitizer test targets:
- remove clang-sanitize-memory for now. It takes so long that the test
times out.
- add gcc sanitize-address and sanitize-undefined test targets.
* Test against openssl-3.0.5.
* Move unset to before we set anything.
* Refuse to use OpenSSL 3.0.4 due to potential RCE.
OpenSSL has a potential RCE in its RSA implementation (CVE-2022-2274)
so refuse to use that specific version.
* Capture stderr output from configure.
* Only refuse to use OpenSSL 3.0.4 on x86_64.
The potential RCE only impacts x86_64, so only refuse to use it if we're
targetting a potentially impacted architecture. ok djm@
* Remove special casing of crypt().
Configure goes to some lengths to pick crypt() from either libcrypt
or OpenSSL's libcrypto because they can more or less featureful (eg
supporting md5-style passwords).
OpenSSL removed its crypt() interface in 2002:
https://github.com/openssl/openssl/commit/69deec58 so these hijinks
should no longer be necessary. This also only links sshd with libcrypt
which is the only thing that needs it. ok djm@
* Clarify README.md text.
Clarify the text about the implications of building without OpenSSL, and
prefix the "configure --help" example command with a "./" so it's likely
to work as-is in more shells. From bz#3461.
* Split README.platform into its own line.
README.platform has general platform-specific information, having it
following text about FIDO2 on the same line could imply that it only
has information about FIDO2.
* Return ERANGE from getcwd() if buffer size is 1.
If getcwd() is supplied a buffer size of exactly 1 and a path of "/", it
could result in a nul byte being written out of array bounds. POSIX says
it should return ERANGE if the path will not fit in the available buffer
(with terminating nul). 1 byte cannot fit any possible path with its nul,
so immediately return ERANGE in that case.
OpenSSH never uses getcwd() with this buffer size, and all current
(and even quite old) platforms that we are currently known to work
on have a native getcwd() so this code is not used on those anyway.
Reported by Qualys, ok djm@
* Remove unintended changes.
I inadvertently included a couple of local changes with the OpenSSL
3.0.4 change. Revert, anything that should be there will be committed
separately.
* Add AUDIT_ARCH_PPC to supported seccomp arches.
Patch from dries.deschout at dodeco.eu.
* Rename bbone test target to ARM.
* Move vmshutdown to first step.
If a previous run on a physical runner has failed to clean up, the next
run will fail because it'll try to check out the code to a broken
directory mount. Make cleanup the first step.
* upstream: pull passphrase reading and confirmation into a separate
function so it can be used for FIDO2 PINs; no functional change
OpenBSD-Commit-ID: bf34f76b8283cc1d3f54633e0d4f13613d87bb2f
* upstream: when enrolling a resident key on a security token, check
if a credential with matching application and user ID strings already exists.
if so, prompt the user for confirmation before overwriting the credential.
patch from Pedro Martelletto via GHPR329
NB. cranks SSH_SK_VERSION_MAJOR, so any third-party FIDO middleware
implementations will need to adjust
OpenBSD-Commit-ID: e45e9f1bf2b2f32d9850669e7a8dbd64acc5fca4
* upstream: sk-usbhid: preserve error code returned by key_lookup()
it conveys useful information, such as the supplied pin being wrong.
Part of GHPR329 from Pedro Martelletto
OpenBSD-Commit-ID: c0647eb9290f793add363d81378439b273756c1b
* upstream: ssh-keygen: fix touch prompt, pin retries;
part of GHPR329 from Pedro Martelletto
OpenBSD-Commit-ID: 75d1005bd2ef8f29fa834c90d2684e73556fffe8
* crank SSH_SK_VERSION_MAJOR in sk-dummy.so
* Skip scp3 test if there's no scp on remote path.
scp -3 ends up using the scp that's in the remote path and will fail if
one is not available. Based on a patch from rapier at psc.edu.
* Convert "have_prog" function into "which".
"which" and its behaviour is not standardized, so convert the existing
have_prog function into "which" so we can rely on it being available
and what its semantics are. Add a have_prog wrapper that maintains the
existing behaviour.
* upstream: Test TEST_SSH_ELAPSED_TIMES for empty string not
executable. No-op on most platforms but should prevent warnings in -portable
on systems that don't have 'date %s'.
OpenBSD-Regress-ID: e39d79867b8065e33d0c5926fa1a31f85659d2a4
* upstream: Restore missing "!" in TEST_SSH_ELAPSED_TIMES test.
OpenBSD-Regress-ID: 38783f9676ec348c5a792caecee9a16e354b37b0
* Remove workarounds for OpenSSL missing AES-GCM.
We have some compatibility hacks that were added to support OpenSSL
versions that do not support AES GCM mode. Since that time, however,
the minimum OpenSSL version that we support has moved to 1.0.1 which
*does* have GCM, so this is no longer needed. ok djm@
* Remove workarounds for OpenSSL missing AES-CTR.
We have some compatibility hacks that were added to support OpenSSL
versions that do not support AES CTR mode. Since that time, however,
the minimum OpenSSL version that we support has moved to 1.0.1 which
*does* have CTR, so this is no longer needed. ok djm@
* Do not link scp, sftp and sftp-server w/ zlib.
Some of our binaries (eg sftp, sftp-server, scp) do not interact with
the channels code and thus do use libraries such as zlib and libcrypto
although they are linked with them. This adds a CHANNELLIBS and starts
by moving zlib into it, which means the aformentioned binaries are no
longer linked against zlib. ok djm@
* Group libcrypto and PRNGD checks together.
They're related more than the libcrypt or libiaf checks which are
currently between them. ok djm@
* Remove seed_rng calls from scp, sftp, sftp-server.
These binaries don't use OpenSSL's random functions. The next step
will be to stop linking them against libcrypto. ok djm@
* Move libcrypto into CHANNELLIBS.
This will result in sftp, sftp-server and scp no longer being linked
against libcrypto. ok djm@
* Move stale-configure check as early as possible.
We added a check in Makefile to catch the case where configure needs to
be rebuilt, however this did not happen until a build was attempted in
which case all of the work done by configure was wasted. Move this check
to the start of configure to catch it as early as possible. ok djm@
* Remove deprecated MacOS 10.15 runners.
* upstream: avoid double-free in error path introduced in r1.70; report
and fix based on GHPR#332 by v-rzh ok dtucker@
OpenBSD-Commit-ID: 3d21aa127b1f37cfc5bdc21461db369a663a951f
* Include CHANNEL and FIDO2 libs in configure output
* Factor out getrnd() and rename to getentropy().
Factor out the arc4random seeding into its own file and change the
interface to match getentropy. Use native getentropy if available.
This will make it easier to resync OpenBSD changes to arc4random.
Prompted by bz#3467, ok djm@.
* compat code for fido_dev_is_winhello()
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* check_sk_options: add temporary WinHello workaround
Up to libfido 1.10.0, WinHello advertises "clientPin" rather
than "uv" capability. This is fixed in 1.11.0. For the time
being, workaround it here.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* sk_sign: set FIDO2 uv attribute explicitely for WinHello
WinHello via libfido2 performs user verification by default.
However, if we stick to that, there's no way to differentiate
between keys created with or without "-O verify-required".
Set FIDO2 uv attribute explicitely to FIDO_OPT_FALSE, then check
if user verification has been requested.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* upstream: don't prompt for FIDO passphrase before attempting to enroll
the credential, just let the enroll operating fail and we'll attempt to get a
PIN anyway. Might avoid some unneccessary PIN prompts.
Part of GHPR#302 from Corinna Vinschen; ok dtucker@
OpenBSD-Commit-ID: bd5342ffc353ee37d39617906867c305564d1ce2
* Give unused param a name.
Fixes builds on platforms that do have fido2 but don't have
fido_dev_is_winhello.
* Actually put HAVE_STDINT_H around the stdint.h.
* Rename our getentropy to prevent possible loops.
Since arc4random seeds from getentropy, and we use OpenSSL for that
if enabled, there's the possibility that if we build on a system that
does not have getentropy then run on a system that does have it, then
OpenSSL could end up calling our getentropy and getting stuck in a loop.
Pointed out by deraadt@, ok djm@
* Test hostbased auth on github runners.
* fix SANDBOX_SECCOMP_FILTER_DEBUG
* Fix conditional for running hostbased tests.
* upstream: allow certificate validity intervals, sshsig verification
times and authorized_keys expiry-time options to accept dates in the UTC time
zone in addition to the default of interpreting them in the system time zone.
YYYYMMDD and YYMMDDHHMM[SS] dates/times will be interpreted as UTC if
suffixed with a 'Z' character.
Also allow certificate validity intervals to be specified in raw
seconds-since-epoch as hex value, e.g. -V 0x1234:0x4567890. This
is intended for use by regress tests and other tools that call
ssh-keygen as part of a CA workflow.
bz3468 ok dtucker
OpenBSD-Commit-ID: 454db1cdffa9fa346aea5211223a2ce0588dfe13
* upstream: add some tests for parse_absolute_time(), including cases
where it is forced to the UTC timezone. bz3468 ok dtucker
OpenBSD-Regress-ID: ea07ca31c2f3847a38df028ca632763ae44e8759
* Skip hostbased during Valgrind tests.
Valgrind doesn't let ssh exec ssh-keysign (because it's setuid) so skip
it during the Valgrind based tests.
See https://bugs.kde.org/show_bug.cgi?id=119404 for a discussion of this
(ironically there the problematic binary was ssh(1) back when it could
still be setuid).
* Rerun tests if any .github config file changes.
* Add a timegm implementation from Heimdal via Samba.
Fixes build on (at least Solaris 10).
* Replace deprecated ubuntu-18.04 runners with 22.04
* upstream: sftp-server: support home-directory request
Add support to the sftp-server for the home-directory extension defined
in draft-ietf-secsh-filexfer-extensions-00. This overlaps a bit with the
existing expand-path@openssh.com, but uses a more official protocol name,
and so is a bit more likely to be implemented by non-OpenSSH clients.
From Mike Frysinger, ok dtucker@
OpenBSD-Commit-ID: bfc580d05cc0c817831ae7ecbac4a481c23566ab
* fido_dev_is_winhello: return 0, not "false"
"false" is not used anywhere in OpenSSH, so return 0 like
everywhere else.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* Revert "check_sk_options: add temporary WinHello workaround"
Cygwin now comes with libfido2 1.11.0, so this workaround
isn't required anymore.
This reverts commit 242c044ab111a37aad3b0775727c36a4c5f0102c.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* upstream: use .Cm for "sign"; from josiah frentsos
OpenBSD-Commit-ID: 7f80a53d54857ac6ae49ea6ad93c5bd12231d1e4
* upstream: add an extra flag to sk_probe() to indicate whether we're
probing for a FIDO resident key or not. Unused here, but will make like
easier for portable
OpenBSD-Commit-ID: 432c8ff70e270378df9dbceb9bdeaa5b43b5a832
* on Cygwin, prefer WinHello FIDO device
If no FIDO device was explictly specified, then prefer the
windows://hello FIDO device. An exception to this is when
probing resident FIDO keys, in which case hardware FIDO
devices are preferred.
* Check for perms to run agent-getpeereid test.
Ubuntu 22.04 defaults to private home dirs which prevents "nobody"
running ssh-add during the agent-getpeereid test. Check for this and
add the necessary permissions.
* upstream: double free() in error path; from Eusgor via GHPR333
OpenBSD-Commit-ID: 39f35e16ba878c8d02b4d01d8826d9b321be26d4
* Add Cygwin (on windows-2019) test target.
In addition to installing the requisite Cygwin packages, we also need to
explicitly invoke "sh" for steps that run other scripts since the runner
environment doesn't understand #! paths.
* Add a bit more debug output.
* Fix cygwin conditional steps.
* upstream: Strictly enforce the maximum allowed SSH2 banner size in
ssh-keyscan and prevent a one-byte buffer overflow. Patch from Qualys, ok
djm@
OpenBSD-Commit-ID: 6ae664f9f4db6e8a0589425f74cd0bbf3aeef4e4
* upstream: remove incorrect check that can break enrolling a
resident key (introduced in r1.40)
OpenBSD-Commit-ID: 4cab364d518470e29e624af3d3f9ffa9c92b6f01
* upstream: attemp FIDO key signing without PIN and use the error
code returned to fall back only if necessary. Avoids PIN prompts for FIDO
tokens that don't require them; part of GHPR#302
OpenBSD-Commit-ID: 4f752aaf9f2e7c28bcaaf3d4f8fc290131bd038e
* Install Cygwin packages based on OS not config.
* initial list of allowed signers
* upstream: whitespace
OpenBSD-Commit-ID: d297e4387935d4aef091c5e9432578c2e513f538
* upstream: whitespace
OpenBSD-Commit-ID: a5d015efbfd228dc598ffdef612d2da3a579e5d8
* Add cygwin-release test target.
This also moves the cygwin package install from the workflow file to
setup_ci.sh so that we can install different sets of Cygwin packages
for different test configs.
* Add Windows 2022 test targets.
* Add libcrypt-devel to cygwin-release deps.
Based on feedback from vinschen at redhat.com.
* cross-sign allowed_signers with PGP key
Provides continuity of trust from legacy PGP release key to
the SSHSIG signing keys that we will use henceforth for git
signing.
* additional keys
* upstream: whitespace
OpenBSD-Commit-ID: c2bcbf93610d3d62ed206cdf9bf9ff98c6aaf232
* Move sftp from valgrind-2 to 3 to rebalance.
* upstream: sk-usbhid: fix key_lookup() on tokens with built-in UV
explicitly test whether the token performs built-in UV (e.g. biometric
tokens) and enable UV in that case. From Pedro Martelletto via GHPR#388
OpenBSD-Commit-ID: 007eb7e387d27cf3029ab06b88224e03eca62ccd
* Remove arc4random_uniform from arc4random.c
This was previously moved into its own file (matching OpenBSD) which
prematurely committed in commit 73541f2.
* Move OPENBSD ORIGINAL marker.
Putting this after the copyright statement (which doesn't change)
instead of before the version identifier (which does) prevents merge
conflicts when resyncing changes.
* Resync arc4random with OpenBSD.
This brings us up to current, including djm's random-reseeding change,
as prompted by logan at cyberstorm.mu in bz#3467. It brings the
platform-specific hooks from LibreSSL Portable, simplified to match our
use case. ok djm@.
* Remove DEF_WEAK, it's already in defines.h.
* openbsd-compat/bsd-asprintf: add <stdio.h> include for vsnprintf
Fixes the following build failure with Clang 15 on musl:
```
bsd-asprintf.c:51:8: error: call to undeclared library function 'vsnprintf' with type 'int (char *, unsigned long, const char *, struct __va_list_tag *)'; ISO C99 and laterclang -O2 -pipe -fdiagnostics-color=always -frecord-gcc-switches -pipe -Wunknown-warning-option -Qunused-arguments -Wall -Wpointer-arith -Wuninitialized -Wsign-compare -Wformat-security -Wsizeof-pointer-memaccess -Wno-pointer-sign -Wno-unused-result -Wmisleading-indentation -Wbitwise-instead-of-logical -fno-strict-aliasing -mretpoline -ftrapv -fzero-call-used-regs=all -fno-builtin-memset -fstack-protector-strong -fPIE -I. -I. -D_XOPEN_SOURCE=600 -D_BSD_SOURCE -D_DEFAULT_SOURCE -DSSHDIR=\"/etc/ssh\" -D_PATH_SSH_PROGRAM=\"/usr/bin/ssh\" -D_PATH_SSH_ASKPASS_DEFAULT=\"/usr/lib/misc/ssh-askpass\" -D_PATH_SFTP_SERVER=\"/usr/lib/misc/sftp-server\" -D_PATH_SSH_KEY_SIGN=\"/usr/lib/misc/ssh-keysign\" -D_PATH_SSH_PKCS11_HELPER=\"/usr/lib/misc/ssh-pkcs11-helper\" -D_PATH_SSH_SK_HELPER=\"/usr/lib/misc/ssh-sk-helper\" -D_PATH_SSH_PIDDIR=\"/run\" -D_PATH_PRIVSEP_CHROOT_DIR=\"/var/empty\" -DHAVE_CONFIG_H -c cipher-aes.c -o cipher-aes.o
do not support
implicit function declarations [-Wimplicit-function-declaration]
ret = vsnprintf(string, INIT_SZ, fmt, ap2);
^
bsd-asprintf.c:51:8: note: include the header <stdio.h> or explicitly provide a declaration for 'vsnprintf'
1 error generated.
```
* upstream: notifier_complete(NULL, ...) is a noop, so no need to test
that ctx!=NULL; from Corinna Vinschen
OpenBSD-Commit-ID: ade2f2e9cc519d01a586800c25621d910bce384a
* fix pester test failures
* upstream: fix repeated words ok miod@ jmc@
OpenBSD-Commit-ID: 6765daefe26a6b648cc15cadbbe337596af709b7
* upstream: .Li -> .Vt where appropriate; from josiah frentsos,
tweaked by schwarze
ok schwarze
OpenBSD-Commit-ID: 565046e3ce68b46c2f440a93d67c2a92726de8ed
* upstream: ssh-agent: attempt FIDO key signing without PIN and use
the error to determine whether a PIN is required and prompt only if
necessary. from Corinna Vinschen
OpenBSD-Commit-ID: dd6be6a0b7148608e834ee737c3479b3270b00dd
* upstream: a little extra debugging
OpenBSD-Commit-ID: edf1601c1d0905f6da4c713f4d9cecc7d1c0295a
* upstream: sk_enroll: never drop SSH_SK_USER_VERIFICATION_REQD flag
from response
Now that all FIDO signing calls attempt first without PIN and then
fall back to trying PIN only if that attempt fails, we can remove the
hack^wtrick that removed the UV flag from the keys returned during
enroll.
By Corinna Vinschen
OpenBSD-Commit-ID: 684517608c8491503bf80cd175425f0178d91d7f
* define HAVE_KILLPG
* upstream: sftp: Don't attempt to complete arguments for
non-existent commands
If user entered a non-existent command (e.g. because they made a
typo) there is no point in trying to complete its arguments. Skip
calling complete_match() if that's the case.
From Michal Privoznik
OpenBSD-Commit-ID: cf39c811a68cde2aeb98fc85addea4000ef6b07a
* upstream: sftp: Be a bit more clever about completions
There are commands (e.g. "get" or "put") that accept two
arguments, a local path and a remote path. However, the way
current completion is written doesn't take this distinction into
account and always completes remote or local paths.
By expanding CMD struct and "cmds" array this distinction can be
reflected and with small adjustment to completer code the correct
path can be completed.
By Michal Privoznik, ok dtucker@
OpenBSD-Commit-ID: 1396d921c4eb1befd531f5c4a8ab47e7a74b610b
* upstream: correct error value
OpenBSD-Commit-ID: 780efcbad76281f11f14b2a5ff04eb6db3dfdad4
* upstream: actually hook up restrict_websafe; the command-line flag
was never actually used. Spotted by Matthew Garrett
OpenBSD-Commit-ID: 0b363518ac4c2819dbaa3dfad4028633ab9cdff1
* upstream: Add a sshkey_check_rsa_length() call for checking the
length of an RSA key; ok markus@
OpenBSD-Commit-ID: de77cd5b11594297eda82edc594b0d32b8535134
* upstream: add a RequiredRSASize for checking RSA key length in
ssh(1). User authentication keys that fall beneath this limit will be
ignored. If a host presents a host key beneath this limit then the connection
will be terminated (unfortunately there are no fallbacks in the protocol for
host authentication).
feedback deraadt, Dmitry Belyavskiy; ok markus@
OpenBSD-Commit-ID: 430e339b2a79fa9ecc63f2837b06fdd88a7da13a
* upstream: Add RequiredRSASize for sshd(8); RSA keys that fall
beneath this limit will be ignored for user and host-based authentication.
Feedback deraadt@ ok markus@
OpenBSD-Commit-ID: 187931dfc19d51873df5930a04f2d972adf1f7f1
* upstream: better debugging for connect_next()
OpenBSD-Commit-ID: d16a307a0711499c971807f324484ed3a6036640
* upstream: sftp-server(8): add a "users-groups-by-id@openssh.com"
extension request that allows the client to obtain user/group names that
correspond to a set of uids/gids.
Will be used to make directory listings more useful and consistent
in sftp(1).
ok markus@
OpenBSD-Commit-ID: 7ebabde0bcb95ef949c4840fe89e697e30df47d3
* upstream: extend sftp-common.c:extend ls_file() to support supplied
user/group names; ok markus@
OpenBSD-Commit-ID: c70c70498b1fdcf158531117e405b6245863bfb0
* upstream: sftp client library support for
users-groups-by-id@openssh.com; ok markus@
OpenBSD-Commit-ID: ddb2f33a2da6349a9a89a8b5bcb9ca7c999394de
* upstream: use users-groups-by-id@openssh.com sftp-server extension
(when available) to fill in user/group names for directory listings.
Implement a client-side cache of see uid/gid=>user/group names. ok markus@
OpenBSD-Commit-ID: f239aeeadfa925a37ceee36ee8b256b8ccf4466e
* avoid Wuninitialized false positive in gcc-12ish
* no need for glob.h here
it also causes portability problems
* add debug on appveyor
* add sleep to pester test
* upstream: add RequiredRSASize to the list of keywords accepted by
-o; spotted by jmc@
OpenBSD-Commit-ID: fe871408cf6f9d3699afeda876f8adbac86a035e
* upstream: Fix typo. From AlexanderStohr via github PR#343.
OpenBSD-Commit-ID: a134c9b4039e48803fc6a87f955b0f4a03181497
* upstream: openssh-9.1
OpenBSD-Commit-ID: 5a467b2ee81da01a86adf1ad93b62b1728494e56
* crank versions in RPM spec files
* update release notes URL
* update .depend
* fix 9.1 compilation errors
* disable -p pester tests due to unreliability on older Windows versions
* remove extra sleep time from debugging scp pester tests
* modify -p tests to only run for Windows OS version 10 and above
* add windows specific code back into method moved from auth.c to auth2-pubkeyfile.c
* add preprocessor for WinHello
* revert preprocessor definition for winhello
* add windows preprocessor definition in key_lookup
* remove rdp block from appveyor since we are no longer debugging
* add ifdef to sftp-server.c
* make key_lookup compatible with winhello
* appveyor.yml
* increase debug of failing pester test
* add #ifdef SUPPORT_CRLF back into auth_check_principals_line method that was moved/renamed
* modify new scp.sh tests for windows
* remove in place tests from scp.sh
* remove rdp debug from appveyor
* retrigger appveyor
* change check of OS version in scp test
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
Co-authored-by: djm@openbsd.org <djm@openbsd.org>
Co-authored-by: Damien Miller <djm@mindrot.org>
Co-authored-by: Darren Tucker <dtucker@dtucker.net>
Co-authored-by: naddy@openbsd.org <naddy@openbsd.org>
Co-authored-by: dtucker@openbsd.org <dtucker@openbsd.org>
Co-authored-by: tj@openbsd.org <tj@openbsd.org>
Co-authored-by: millert@openbsd.org <millert@openbsd.org>
Co-authored-by: jmc@openbsd.org <jmc@openbsd.org>
Co-authored-by: florian@openbsd.org <florian@openbsd.org>
Co-authored-by: markus@openbsd.org <markus@openbsd.org>
Co-authored-by: Tobias Heider <me@tobhe.de>
Co-authored-by: anton@openbsd.org <anton@openbsd.org>
Co-authored-by: Tim Rice <tim@multitalents.net>
Co-authored-by: tobhe@openbsd.org <tobhe@openbsd.org>
Co-authored-by: Corinna Vinschen <vinschen@redhat.com>
Co-authored-by: Sam James <sam@gentoo.org>
Co-authored-by: jsg@openbsd.org <jsg@openbsd.org>
2022-11-02 17:06:45 +01:00
|
|
|
}
|
2021-01-08 03:57:24 +01:00
|
|
|
retry_pin:
|
Merge 9.1 (#626)
* upstream: fix poll() spin when a channel's output fd closes without
data in the channel buffer. Introduce more exact packing of channel fds into
the pollfd array. fixes bz3405 and bz3411; ok deraadt@ markus@
OpenBSD-Commit-ID: 06740737849c9047785622ad5d472cb6a3907d10
* upstream: select post-quantum KEX
sntrup761x25519-sha512@openssh.com as the default; ok markus@
OpenBSD-Commit-ID: f02d99cbfce22dffec2e2ab1b60905fbddf48fb9
* upstream: add support for the "corp-data" protocol extension to
allow server-side copies to be performed without having to go via the client.
Patch by Mike Frysinger, ok dtucker@
OpenBSD-Commit-ID: 00aa510940fedd66dab1843b58682de4eb7156d5
* upstream: add a sftp client "cp" command that supports server-side
copying of files. Useful for this task and for testing the copy-data
extension. Patch from Mike Frysinger; ok dtucker@
OpenBSD-Commit-ID: 1bb1b950af0d49f0d5425b1f267e197aa1b57444
* depend
* Skip slow tests on (very) slow test targets.
* Set Makefile SHELL as determined by configure.
This should improve compatibility for users with non-POSIX shells. If
using Makefile.in directly (eg make -f Makefile.in distprep) then SHELL
will need to be specified on the command line (along with MANFMT in that
particular case). ok djm@
* Use bash or ksh if available for SH in Makefile.
* Increase test timeout to allow slow VMs to finish
* Only run regression tests on slow VMs.
* Only return events from ppoll that were requested.
If the underlying system's select() returns bits that were not in the
request set, our ppoll() implementation can return revents for events
not requested, which can apparently cause a hang. Only return revents
for activity in the requested event set. bz#3416, analysis and fix by
yaroslav.kuzmin at vmssoftware com, ok djm@
* Specify TEST_SHELL=bash on AIX.
The system shells cause the agent-restrict test to fail due to some
quoting so explicitly specify bash until we can get configure to
autmatically work around that.
* Disable security key on fbsd6 test host.
* upstream: man pages: add missing commas between subordinate and
main clauses
jmc@ dislikes a comma before "then" in a conditional, so leave those
untouched.
ok jmc@
OpenBSD-Commit-ID: 9520801729bebcb3c9fe43ad7f9776ab4dd05ea3
* upstream: ssh: document sntrup761x25519-sha512@openssh.com as
default KEX
OpenBSD-Commit-ID: 12545bfa10bcbf552d04d9d9520d0f4e98b0e171
* upstream: openssh-9.0
OpenBSD-Commit-ID: 0dfb461188f4513ec024c1534da8c1ce14c20b64
* update version numbers for release
* update build-aux files to match autoconf-2.71
i.e. config.guess, config.sub and install-sh
* Revert "update build-aux files to match autoconf-2.71"
This reverts commit 0a8ca39fac6ad19096b6c263436f8b2dd51606f2.
It turns out that the checked-in copies of these files are actually newer
than autoconf-2.71's copies, so this was effectively a downgrade.
Spotted by Bo Anderson via github
* upstream: two defensive changes from Tobias Stoeckmann via GHPR287
enforce stricter invarient for sshbuf_set_parent() - never allow
a buffer to have a previously-set parent changed.
In sshbuf_reset(), if the reallocation fails, then zero the entire
buffer and not the (potentially smaller) default initial alloc size.
OpenBSD-Commit-ID: 14583203aa5d50ad38d2e209ae10abaf8955e6a9
* upstream: Note that curve25519-sha256 was later published in
RFC8731. ok djm@
OpenBSD-Commit-ID: 2ac2b5d642d4cf5918eaec8653cad9a4460b2743
* upstream: clear io_want/io_ready flags at start of poll() cycle;
avoids plausible spin during rekeying if channel io_want flags are reused
across cycles. ok markus@ deraadt@
OpenBSD-Commit-ID: 91034f855b7c73cd2591657c49ac30f10322b967
* Retire fbsd6 test VM.
It's long since out of support, relatively slow (it's i686) and the
compiler has trouble with PIE.
* Resync moduli.5 with upstream.
1.18: remove duplicate publication year; carsten dot kunze at arcor dot de
1.19: ssh-keygen's -G/-T have been replaced with -M generate/screen.
* upstream: Correct path for system known hosts file in description
of IgnoreUserKnownHosts. Patch from Martin Vahlensieck via tech@
OpenBSD-Commit-ID: 9b7784f054fa5aa4d63cb36bd563889477127215
* upstream: list the correct version number
for when usage of the sftp protocol became default and fix a typo
from ed maste
OpenBSD-Commit-ID: 24e1795ed2283fdeacf16413c2f07503bcdebb31
* upstream: Import regenerated moduli
OpenBSD-Commit-ID: f9a0726d957cf10692a231996a1f34e7f9cdfeb0
* upstream: Try to continue running local I/O for channels in state
OPEN during SSH transport rekeying. The most visible benefit is that it
should make ~-escapes work in the client (e.g. to exit) if the connection
happened to have stalled during a rekey event. Based work by and ok dtucker@
OpenBSD-Commit-ID: a66e8f254e92edd4ce09c9f750883ec8f1ea5f45
* upstream: Import regenerated moduli
OpenBSD-Commit-ID: f9a0726d957cf10692a231996a1f34e7f9cdfeb0
* upstream: regression test for sftp cp command
OpenBSD-Regress-ID: c96bea9edde3a384b254785e7f9b2b24a81cdf82
* upstream: Simplify forward-control test.
Since we no longer need to support SSH1 we don't need to run shell
commands on the other end of the connection and can use ssh -N instead.
This also makes the test less racy.
OpenBSD-Regress-ID: 32e94ce272820cc398f30b848b2b0f080d10302c
* upstream: Use ssh -f and ControlPersist ..
to start up test forwards and ssh -O stop to shut them down intead of
sleep loops. This speeds up the test by an order of magnitude.
OpenBSD-Regress-ID: eb3db5f805100919b092a3b2579c611fba3e83e7
* upstream: It looks like we can't completely avoid
waiting for processes to exit so retrieve the pid via controlmaster and
use that.
OpenBSD-Regress-ID: 8246f00f22b14e49d2ff1744c94897ead33d457b
* Cache timezone data in capsicum sandbox.
From emaste at freebsd.org, originally part of FreeBSD commit r339216
/ fc3c19a9 with autoconf bits added by me.
* Include stdlib.h for free() prototype.
... which is used inside the CUSTOM_SYS_AUTH_GET_LASTLOGIN_MSG block.
* Update OpenSSL and LibreSSL versions in tests.
* Add debian-riscv64 test target.
* upstream: Avoid an unnecessary xstrdup in rm_env() when matching
patterns. Since match_pattern() doesn't modify its arguments (they are
const), there is no need to make an extra copy of the strings in
options->send_env. From Martin Vahlensieck
OpenBSD-Commit-ID: 2c9db31e3f4d3403b49642c64ee048b2a0a39351
* upstream: Add missing includes of stdlib.h and stdint.h. We need
stdlib.h for malloc(3) and stdint.h for SIZE_MAX. Unlike the other xmss
files, ssh-xmss.c does not include xmss_commons.h so ssh-xmss.c must include
those headers itself. From Martin Vahlensieck
OpenBSD-Commit-ID: 70e28a9818cee3da1be2ef6503d4b396dd421e6b
* upstream: Remove unnecessary includes: openssl/hmac.h and
openssl/evp.h. From Martin Vahlensieck.
OpenBSD-Commit-ID: a6debb5fb0c8a44e43e8d5ca7cc70ad2f3ea31c3
* upstream: Check sshauthopt_new() for NULL. bz#3425, from
tessgauthier at microsoft.com. ok djm@
OpenBSD-Commit-ID: af0315bc3e44aa406daa7e0ae7c2d719a974483f
* upstream: Add authfd path to debug output. ok markus@
OpenBSD-Commit-ID: f735a17d1a6f2bee63bfc609d76ef8db8c090890
* upstream: avoid printing hash algorithm twice; from lucas AT sexy.is
OpenBSD-Commit-ID: 9d24671e10a84141b7c504396cabad600e47a941
* upstream: fix memleak on session-bind path; from Pedro Martelletto, ok
dtucker@
OpenBSD-Commit-ID: e85899a26ba402b4c0717b531317e8fc258f0a7e
* upstream: Don't leak SK device. Patch from Pedro Martelletto via
github PR#316. ok djm@
OpenBSD-Commit-ID: 17d11327545022e727d95fd08b213171c5a4585d
* upstream: mention that the helpers are used by ssh(1), ssh-agent(1)
and ssh-keygen(1). Previously only ssh(1) was mentioned. From Pedro
Martelletto
OpenBSD-Commit-ID: 30f880f989d4b329589c1c404315685960a5f153
* Remove now-empty int32_minmax.inc.
* Only run tests when source files change.
Also run tests on changes to V_9_0 branch.
* Add Mac OS X 12 test target.
* upstream: be stricter in which characters will be accepted in
specifying a mask length; allow only 0-9. From khaleesicodes via GHPR#278; ok
dtucker@
OpenBSD-Commit-ID: e267746c047ea86665cdeccef795a8a56082eeb2
* upstream: fix some integer overflows in sieve_large() that show up when
trying to generate modp groups > 16k bits. Reported via GHPR#306 by Bertram
Felgenhauer, but fixed in a different way. feedback/ok tb@
OpenBSD-Commit-ID: 81cbc6dd3a21c57bd6fadea10e44afe37bca558e
* upstream: remove an obsolete rsa1 format example from an example;
from megan batty
ok djm
OpenBSD-Commit-ID: db2c89879c29bf083df996bd830abfb1e70d62bf
* upstream: Add FIDO AUTHENTICATOR section and explain a bit how FIDO
works. The wording came mostly from the 8.2 OpenSSH release notes, addapted
to fit the man page. Then move the -O bits into the new section as is already
done for CERTIFICATES and MODULI GENERATION. Finally we can explain the
trade-offs of resident keys. While here, consistently refer to the FIDO
thingies as "FIDO authenticators", not "FIDO tokens".
input & OK jmc, naddy
OpenBSD-Commit-ID: dd98748d7644df048f78dcf793b3b63db9ab1d25
* upstream: make sure stdout is non-blocking; ok djm@
OpenBSD-Commit-ID: 64940fffbd1b882eda2d7c8c7a43c79368309c0d
* upstream: mux.c: mark argument as const; from Martin Vahlensieck
OpenBSD-Commit-ID: 69a1a93a55986c7c2ad9f733c093b46a47184341
* upstream: channel_new no longer frees remote_name. So update the
comment accordingly. As remote_name is not modified, it can be const as
well. From Martin Vahlensieck
OpenBSD-Commit-ID: e4e10dc8dc9f40c166ea5a8e991942bedc75a76a
* upstream: sshkey_unshield_private() contains a exact duplicate of
the code in private2_check_padding(). Pull private2_check_padding() up so the
code can be reused. From Martin Vahlensieck, ok deraadt@
OpenBSD-Commit-ID: 876884c3f0e62e8fd8d1594bab06900f971c9c85
* Add ubsan minimal testcase on OpenBSD.
As suggested by djm@.
* Note that, for now, we need variadic macros.
* Also retest OpenBSD upstream on .yml changes.
* upstream: When performing operations that glob(3) a remote path, ensure
that the implicit working directory used to construct that path escapes
glob(3) characters.
This prevents glob characters from being processed in places they
shouldn't, e.g. "cd /tmp/a*/", "get *.txt" should have the get operation
treat the path "/tmp/a*" literally and not attempt to expand it.
Reported by Lusia Kundel; ok markus@
OpenBSD-Commit-ID: 4f647f58482cbad3d58b1eab7f6a1691433deeef
* Remove duplicate bcrypt_pbkdf.o from Makefile
bcrypt_pbkdf.o is duplicated in the openbsd-compat Makefile's object
file list.
* upstream: improve error message when 'ssh-keygen -Y sign' is unable to
load a private key; bz3429, reported by Adam Szkoda ok dtucker@
OpenBSD-Commit-ID: bb57b285e67bea536ef81b1055467be2fc380e74
* upstream: Allow existing -U (use agent) flag to work with "-Y sign"
operations, where it will be interpreted to require that the private keys is
hosted in an agent; bz3429, suggested by Adam Szkoda; ok dtucker@
OpenBSD-Commit-ID: a7bc69873b99c32c42c7628ed9ea91565ba08c2f
* upstream: Remove errant apostrophe. From haruyama at queen-ml org.
OpenBSD-Commit-ID: dc6b294567cb84b384ad6ced9ca469f2bbf0bd10
* upstream: arrange for scp, when in sftp mode, to not ftruncate(3) files
early
previous behavious of unconditionally truncating the destination file
would cause "scp ~/foo localhost:" and "scp localhost:foo ~/" to
delete all the contents of their destination.
spotted by solene@ sthen@, also bz3431; ok dtucker@
OpenBSD-Commit-ID: ca39fdd39e0ec1466b9666f15cbcfddea6aaa179
* upstream: fix in-place copies; r1.163 incorrectly skipped truncation in
all cases, not just at the start of a transfer. This could cause overwrites
of larger files to leave junk at the end. Spotted by tb@
OpenBSD-Commit-ID: b189f19cd68119548c8e24e39c79f61e115bf92c
* upstream: Only run agent-ptrace.sh if gdb is available as all
architectures do not ship with gdb.
OpenBSD-Regress-ID: ec53e928803e6b87f9ac142d38888ca79a45348d
* upstream: regress test for in-place transfers and clobbering larger
files with smaller ones; would have caught last regression in scp(1)
OpenBSD-Regress-ID: 19de4e88dd3a4f7e5c1618c9be3c32415bd93bc2
* configure.ac: Add missing AC_DEFINE for caph_cache_tzdata test causing
HAVE_CAPH_CACHE_TZDATA to be missing from config.h.in.
Spotted by Bryan Drewery
* upstream: make SSHBUF_DBG/SSHBUF_TELL (off by default and only enabled
via #define) dump to stderr rather than stdout
OpenBSD-Commit-ID: 10298513ee32db8390aecb0397d782d68cb14318
* upstream: revert previous; it was broken (spotted by Theo)
OpenBSD-Commit-ID: 457c79afaca2f89ec2606405c1059b98b30d8b0d
* upstream: Note that ProxyJump also accepts the same tokens as
ProxyCommand. From pallxk via github PR#305.
OpenBSD-Commit-ID: 7115ac351b129205f1f1ffa6bbfd62abd76be7c5
* upstream: Avoid kill with -1 argument. The out_ctx label can be
reached before fork has been called. If this happens, then kill -1 would be
called, sending SIGTERM to all processes reachable by the current process.
From tobias@ and c3h2_ctf via github PR#286, ok djm@
OpenBSD-Commit-ID: 6277af1207d81202f5daffdccfeeaed4c763b1a8
* upstream: f sshpkt functions fail, then password is not cleared
with freezero. Unconditionally call freezero to guarantee that password is
removed from RAM.
From tobias@ and c3h2_ctf via github PR#286, ok djm@
OpenBSD-Commit-ID: 6b093619c9515328e25b0f8093779c52402c89cd
* upstream: refactor authorized_keys/principals handling
remove "struct ssh *" from arguments - this was only used to pass the
remote host/address. These can be passed in instead and the resulting
code is less tightly coupled to ssh_api.[ch]
ok dtucker@
OpenBSD-Commit-ID: 9d4373d013edc4cc4b5c21a599e1837ac31dda0d
* upstream: split the low-level file handling functions out from
auth2-pubkey.c
Put them in a new auth2-pubkeyfile.c to make it easier to refer to them
(e.g. in unit/fuzz tests) without having to refer to everything else
pubkey auth brings in.
ok dtucker@
OpenBSD-Commit-ID: 3fdca2c61ad97dc1b8d4a7346816f83dc4ce2217
* fuzzer for authorized_keys parsing
mostly redundant to authopt_fuzz, but it's sensitive code so IMO it
makes sense to test this layer too
* Test against LibreSSL 3.5.3.
* Test against OpenSSL 1.1.1o and 3.0.3.
* fix some bugs in the fuzzer
* upstream: keywords ref ssh_config.5;
from caspar schutijser
OpenBSD-Commit-ID: f146a19d7d5c9374c3b9c520da43b2732d7d1a4e
* upstream: ssh-keygen: implement "verify-required" certificate option.
This was already documented when support for user-verified FIDO
keys was added, but the ssh-keygen(1) code was missing.
ok djm@
OpenBSD-Commit-ID: f660f973391b593fea4b7b25913c9a15c3eb8a06
* upstream: ssh-keygen -A: do not generate DSA keys by default.
Based on github PR#303 from jsegitz with man page text from jmc@, ok markus@
djm@
OpenBSD-Commit-ID: 5c4c57bdd7063ff03381cfb6696659dd3f9f5b9f
* upstream: Add period at end of "not known by any other names"
message. github PR#320 from jschauma, ok djm@
OpenBSD-Commit-ID: bd60809803c4bfd3ebb7c5c4d918b10e275266f2
* upstream: Add missing *-sk types to ssh-keyscan manpage. From
skazi0 via github PR#294.
OpenBSD-Commit-ID: fda2c869cdb871f3c90a89fb3f985370bb5d25c0
* upstream: Make SetEnv directives first-match-wins in both
sshd_config and sshd_config; previously if the same name was reused then the
last would win (which is the opposite to how the config is supposed to work).
While there, make the ssh_config parsing more like sshd_config.
bz3438, ok dtucker
OpenBSD-Commit-ID: 797909c1e0262c0d00e09280459d7ab00f18273b
* upstream: test setenv in both client and server, test first-match-wins
too
OpenBSD-Regress-ID: 4c8804f9db38a02db480b9923317457b377fe34b
* upstream: move auth_openprincipals() and auth_openkeyfile() over to
auth2-pubkeyfile.c too; they make more sense there.
OpenBSD-Commit-ID: 9970d99f900e1117fdaab13e9e910a621b7c60ee
* upstream: make sure that UseDNS hostname lookup happens in the monitor
and not in the pledge(2)'d unprivileged process; fixes regression caused by
recent refactoring spotted by henning@
OpenBSD-Commit-ID: a089870b95101cd8881a2dff65b2f1627d13e88d
* fix possible NULL deref when built without FIDO
Analysis/fix from kircher in bz3443; ok dtucker@
* automatically enable built-in FIDO support
If libfido2 is found and usable, then enable the built-in
security key support unless --without-security-key-builtin
was requested.
ok dtucker@
* upstream: Log an error if pipe() fails while accepting a
connection. bz#3447, from vincent-openssh at vinc17 net, ok djm@
OpenBSD-Commit-ID: 9d59f19872b94900a5c79da2d57850241ac5df94
* upstream: Don't attempt to fprintf a null identity comment. From
Martin Vahlensieck via tech@.
OpenBSD-Commit-ID: 4c54d20a8e8e4e9912c38a7b4ef5bfc5ca2e05c2
* upstream: Make sure not to fclose() the same fd twice in case of an
error.
ok dtucker@
OpenBSD-Commit-ID: e384c4e05d5521e7866b3d53ca59acd2a86eef99
* upstream: make it clear that RekeyLimit applies to both transmitted
and received data. GHPR#328 from Jan Pazdziora
OpenBSD-Commit-ID: d180a905fec9ff418a75c07bb96ea41c9308c3f9
* request 1.1x API compatibility for OpenSSL >=3.x
idea/patch from Pedro Martelletto via GHPR#322; ok dtucker@
* fix broken case statement in previous
* Disable SK support if FIDO libs not found.
* Zero out LIBFIDO2 when SK support not usable.
Prevents us from trying to link them into ssh-sk-helper and failing to
build.
* upstream: Don't leak the strings allocated by order_hostkeyalgs()
and list_hostkey_types() that are passed to compat_pkalg_proposal(). Part of
github PR#324 from ZoltanFridrich, ok djm@
OpenBSD-Commit-ID: b2f6e5f60f2bba293b831654328a8a0035ef4a1b
* upstream: Roll back previous KEX changes as they aren't safe until
compat_pkalg_proposal and friends always allocate their returned strings.
Reported by Qualys.
OpenBSD-Commit-ID: 1c7a88a0d5033f42f88ab9bec58ef1cf72c81ad0
* upstream: allow arguments to sftp -D option, e.g. sftp -D
"/usr/libexec/sftp-server -el debug3"
ok markus@
OpenBSD-Commit-ID: 5a002b9f3a7aef2731fc0ffa9c921cf15f38ecce
* Update OpenSSL tests to the most recent releases.
* upstream: reflect the update to -D arg name in usage();
OpenBSD-Commit-ID: abdcde4f92b1ef094ae44210ee99d3b0155aad9c
* upstream: ignore SIGPIPE earlier in main(), specifically before
muxclient() which performs operations that could cause one; Reported by Noam
Lewis via bz3454, ok dtucker@
OpenBSD-Commit-ID: 63d8e13276869eebac6d7a05d5a96307f9026e47
* upstream: Always return allocated strings from the kex filtering so
that we can free them later. Fix one leak in compat_kex_proposal. Based on
github PR#324 from ZoltanFridrich with some simplications by me. ok djm@
OpenBSD-Commit-ID: 9171616da3307612d0ede086fd511142f91246e4
* upstream: Don't leak the strings allocated by order_hostkeyalgs()
and list_hostkey_types() that are passed to compat_pkalg_proposal(). Part of
github PR#324 from ZoltanFridrich, ok djm@
This is a roll-forward of the previous rollback now that the required
changes in compat.c have been done.
OpenBSD-Commit-ID: c7cd93730b3b9f53cdad3ae32462922834ef73eb
* upstream: bump up loglevel from debug to info when unable to open
authorized keys/principals file for errno != ENOENT; bz2042 ok dtucker
OpenBSD-Commit-ID: e79aa550d91ade6a80f081bda689da24c086d66b
* Skip select+rlimit check if sandboxing is disabled
It's not needed in that case, and the test can fail when being built
with some compiler memory sanitizer flags. bz#3441
* upstream: use consistent field names (s/char/byte)
in format description
OpenBSD-Commit-ID: 3de33572733ee7fcfd7db33d37db23d2280254f0
* upstream: Remove leftover line.
Remove extra line leftover from merge conflict. ok djm@
OpenBSD-Commit-ID: 460e2290875d7ae64971a7e669c244b1d1c0ae2e
* Move checks for pollfd.fd and nfds_t.
Move the checks for struct pollfd.fd and nfds_t to before the sandboxing
checks. This groups all the sandbox checks together so we can skip them
all when sandboxing is disabled.
* Skip all rlimit tests when sandboxing disabled.
The rlimit tests can hang when being run with some compiler sanitizers
so skip all of them if sandbox=no.
* Add clang sanitizer tests.
* upstream: Add TEST_REGRESS_CACHE_DIR.
If set, it is used to cache regress test names that have succeeded and
skip those on a re-run.
OpenBSD-Regress-ID: a7570dd29a58df59f2cca647c3c2ec989b49f247
* Move sanitizer logs into regress for collection.
* Add GCC address sanitizer build/test.
* Update sanitizer test targets:
- remove clang-sanitize-memory for now. It takes so long that the test
times out.
- add gcc sanitize-address and sanitize-undefined test targets.
* Test against openssl-3.0.5.
* Move unset to before we set anything.
* Refuse to use OpenSSL 3.0.4 due to potential RCE.
OpenSSL has a potential RCE in its RSA implementation (CVE-2022-2274)
so refuse to use that specific version.
* Capture stderr output from configure.
* Only refuse to use OpenSSL 3.0.4 on x86_64.
The potential RCE only impacts x86_64, so only refuse to use it if we're
targetting a potentially impacted architecture. ok djm@
* Remove special casing of crypt().
Configure goes to some lengths to pick crypt() from either libcrypt
or OpenSSL's libcrypto because they can more or less featureful (eg
supporting md5-style passwords).
OpenSSL removed its crypt() interface in 2002:
https://github.com/openssl/openssl/commit/69deec58 so these hijinks
should no longer be necessary. This also only links sshd with libcrypt
which is the only thing that needs it. ok djm@
* Clarify README.md text.
Clarify the text about the implications of building without OpenSSL, and
prefix the "configure --help" example command with a "./" so it's likely
to work as-is in more shells. From bz#3461.
* Split README.platform into its own line.
README.platform has general platform-specific information, having it
following text about FIDO2 on the same line could imply that it only
has information about FIDO2.
* Return ERANGE from getcwd() if buffer size is 1.
If getcwd() is supplied a buffer size of exactly 1 and a path of "/", it
could result in a nul byte being written out of array bounds. POSIX says
it should return ERANGE if the path will not fit in the available buffer
(with terminating nul). 1 byte cannot fit any possible path with its nul,
so immediately return ERANGE in that case.
OpenSSH never uses getcwd() with this buffer size, and all current
(and even quite old) platforms that we are currently known to work
on have a native getcwd() so this code is not used on those anyway.
Reported by Qualys, ok djm@
* Remove unintended changes.
I inadvertently included a couple of local changes with the OpenSSL
3.0.4 change. Revert, anything that should be there will be committed
separately.
* Add AUDIT_ARCH_PPC to supported seccomp arches.
Patch from dries.deschout at dodeco.eu.
* Rename bbone test target to ARM.
* Move vmshutdown to first step.
If a previous run on a physical runner has failed to clean up, the next
run will fail because it'll try to check out the code to a broken
directory mount. Make cleanup the first step.
* upstream: pull passphrase reading and confirmation into a separate
function so it can be used for FIDO2 PINs; no functional change
OpenBSD-Commit-ID: bf34f76b8283cc1d3f54633e0d4f13613d87bb2f
* upstream: when enrolling a resident key on a security token, check
if a credential with matching application and user ID strings already exists.
if so, prompt the user for confirmation before overwriting the credential.
patch from Pedro Martelletto via GHPR329
NB. cranks SSH_SK_VERSION_MAJOR, so any third-party FIDO middleware
implementations will need to adjust
OpenBSD-Commit-ID: e45e9f1bf2b2f32d9850669e7a8dbd64acc5fca4
* upstream: sk-usbhid: preserve error code returned by key_lookup()
it conveys useful information, such as the supplied pin being wrong.
Part of GHPR329 from Pedro Martelletto
OpenBSD-Commit-ID: c0647eb9290f793add363d81378439b273756c1b
* upstream: ssh-keygen: fix touch prompt, pin retries;
part of GHPR329 from Pedro Martelletto
OpenBSD-Commit-ID: 75d1005bd2ef8f29fa834c90d2684e73556fffe8
* crank SSH_SK_VERSION_MAJOR in sk-dummy.so
* Skip scp3 test if there's no scp on remote path.
scp -3 ends up using the scp that's in the remote path and will fail if
one is not available. Based on a patch from rapier at psc.edu.
* Convert "have_prog" function into "which".
"which" and its behaviour is not standardized, so convert the existing
have_prog function into "which" so we can rely on it being available
and what its semantics are. Add a have_prog wrapper that maintains the
existing behaviour.
* upstream: Test TEST_SSH_ELAPSED_TIMES for empty string not
executable. No-op on most platforms but should prevent warnings in -portable
on systems that don't have 'date %s'.
OpenBSD-Regress-ID: e39d79867b8065e33d0c5926fa1a31f85659d2a4
* upstream: Restore missing "!" in TEST_SSH_ELAPSED_TIMES test.
OpenBSD-Regress-ID: 38783f9676ec348c5a792caecee9a16e354b37b0
* Remove workarounds for OpenSSL missing AES-GCM.
We have some compatibility hacks that were added to support OpenSSL
versions that do not support AES GCM mode. Since that time, however,
the minimum OpenSSL version that we support has moved to 1.0.1 which
*does* have GCM, so this is no longer needed. ok djm@
* Remove workarounds for OpenSSL missing AES-CTR.
We have some compatibility hacks that were added to support OpenSSL
versions that do not support AES CTR mode. Since that time, however,
the minimum OpenSSL version that we support has moved to 1.0.1 which
*does* have CTR, so this is no longer needed. ok djm@
* Do not link scp, sftp and sftp-server w/ zlib.
Some of our binaries (eg sftp, sftp-server, scp) do not interact with
the channels code and thus do use libraries such as zlib and libcrypto
although they are linked with them. This adds a CHANNELLIBS and starts
by moving zlib into it, which means the aformentioned binaries are no
longer linked against zlib. ok djm@
* Group libcrypto and PRNGD checks together.
They're related more than the libcrypt or libiaf checks which are
currently between them. ok djm@
* Remove seed_rng calls from scp, sftp, sftp-server.
These binaries don't use OpenSSL's random functions. The next step
will be to stop linking them against libcrypto. ok djm@
* Move libcrypto into CHANNELLIBS.
This will result in sftp, sftp-server and scp no longer being linked
against libcrypto. ok djm@
* Move stale-configure check as early as possible.
We added a check in Makefile to catch the case where configure needs to
be rebuilt, however this did not happen until a build was attempted in
which case all of the work done by configure was wasted. Move this check
to the start of configure to catch it as early as possible. ok djm@
* Remove deprecated MacOS 10.15 runners.
* upstream: avoid double-free in error path introduced in r1.70; report
and fix based on GHPR#332 by v-rzh ok dtucker@
OpenBSD-Commit-ID: 3d21aa127b1f37cfc5bdc21461db369a663a951f
* Include CHANNEL and FIDO2 libs in configure output
* Factor out getrnd() and rename to getentropy().
Factor out the arc4random seeding into its own file and change the
interface to match getentropy. Use native getentropy if available.
This will make it easier to resync OpenBSD changes to arc4random.
Prompted by bz#3467, ok djm@.
* compat code for fido_dev_is_winhello()
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* check_sk_options: add temporary WinHello workaround
Up to libfido 1.10.0, WinHello advertises "clientPin" rather
than "uv" capability. This is fixed in 1.11.0. For the time
being, workaround it here.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* sk_sign: set FIDO2 uv attribute explicitely for WinHello
WinHello via libfido2 performs user verification by default.
However, if we stick to that, there's no way to differentiate
between keys created with or without "-O verify-required".
Set FIDO2 uv attribute explicitely to FIDO_OPT_FALSE, then check
if user verification has been requested.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* upstream: don't prompt for FIDO passphrase before attempting to enroll
the credential, just let the enroll operating fail and we'll attempt to get a
PIN anyway. Might avoid some unneccessary PIN prompts.
Part of GHPR#302 from Corinna Vinschen; ok dtucker@
OpenBSD-Commit-ID: bd5342ffc353ee37d39617906867c305564d1ce2
* Give unused param a name.
Fixes builds on platforms that do have fido2 but don't have
fido_dev_is_winhello.
* Actually put HAVE_STDINT_H around the stdint.h.
* Rename our getentropy to prevent possible loops.
Since arc4random seeds from getentropy, and we use OpenSSL for that
if enabled, there's the possibility that if we build on a system that
does not have getentropy then run on a system that does have it, then
OpenSSL could end up calling our getentropy and getting stuck in a loop.
Pointed out by deraadt@, ok djm@
* Test hostbased auth on github runners.
* fix SANDBOX_SECCOMP_FILTER_DEBUG
* Fix conditional for running hostbased tests.
* upstream: allow certificate validity intervals, sshsig verification
times and authorized_keys expiry-time options to accept dates in the UTC time
zone in addition to the default of interpreting them in the system time zone.
YYYYMMDD and YYMMDDHHMM[SS] dates/times will be interpreted as UTC if
suffixed with a 'Z' character.
Also allow certificate validity intervals to be specified in raw
seconds-since-epoch as hex value, e.g. -V 0x1234:0x4567890. This
is intended for use by regress tests and other tools that call
ssh-keygen as part of a CA workflow.
bz3468 ok dtucker
OpenBSD-Commit-ID: 454db1cdffa9fa346aea5211223a2ce0588dfe13
* upstream: add some tests for parse_absolute_time(), including cases
where it is forced to the UTC timezone. bz3468 ok dtucker
OpenBSD-Regress-ID: ea07ca31c2f3847a38df028ca632763ae44e8759
* Skip hostbased during Valgrind tests.
Valgrind doesn't let ssh exec ssh-keysign (because it's setuid) so skip
it during the Valgrind based tests.
See https://bugs.kde.org/show_bug.cgi?id=119404 for a discussion of this
(ironically there the problematic binary was ssh(1) back when it could
still be setuid).
* Rerun tests if any .github config file changes.
* Add a timegm implementation from Heimdal via Samba.
Fixes build on (at least Solaris 10).
* Replace deprecated ubuntu-18.04 runners with 22.04
* upstream: sftp-server: support home-directory request
Add support to the sftp-server for the home-directory extension defined
in draft-ietf-secsh-filexfer-extensions-00. This overlaps a bit with the
existing expand-path@openssh.com, but uses a more official protocol name,
and so is a bit more likely to be implemented by non-OpenSSH clients.
From Mike Frysinger, ok dtucker@
OpenBSD-Commit-ID: bfc580d05cc0c817831ae7ecbac4a481c23566ab
* fido_dev_is_winhello: return 0, not "false"
"false" is not used anywhere in OpenSSH, so return 0 like
everywhere else.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* Revert "check_sk_options: add temporary WinHello workaround"
Cygwin now comes with libfido2 1.11.0, so this workaround
isn't required anymore.
This reverts commit 242c044ab111a37aad3b0775727c36a4c5f0102c.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* upstream: use .Cm for "sign"; from josiah frentsos
OpenBSD-Commit-ID: 7f80a53d54857ac6ae49ea6ad93c5bd12231d1e4
* upstream: add an extra flag to sk_probe() to indicate whether we're
probing for a FIDO resident key or not. Unused here, but will make like
easier for portable
OpenBSD-Commit-ID: 432c8ff70e270378df9dbceb9bdeaa5b43b5a832
* on Cygwin, prefer WinHello FIDO device
If no FIDO device was explictly specified, then prefer the
windows://hello FIDO device. An exception to this is when
probing resident FIDO keys, in which case hardware FIDO
devices are preferred.
* Check for perms to run agent-getpeereid test.
Ubuntu 22.04 defaults to private home dirs which prevents "nobody"
running ssh-add during the agent-getpeereid test. Check for this and
add the necessary permissions.
* upstream: double free() in error path; from Eusgor via GHPR333
OpenBSD-Commit-ID: 39f35e16ba878c8d02b4d01d8826d9b321be26d4
* Add Cygwin (on windows-2019) test target.
In addition to installing the requisite Cygwin packages, we also need to
explicitly invoke "sh" for steps that run other scripts since the runner
environment doesn't understand #! paths.
* Add a bit more debug output.
* Fix cygwin conditional steps.
* upstream: Strictly enforce the maximum allowed SSH2 banner size in
ssh-keyscan and prevent a one-byte buffer overflow. Patch from Qualys, ok
djm@
OpenBSD-Commit-ID: 6ae664f9f4db6e8a0589425f74cd0bbf3aeef4e4
* upstream: remove incorrect check that can break enrolling a
resident key (introduced in r1.40)
OpenBSD-Commit-ID: 4cab364d518470e29e624af3d3f9ffa9c92b6f01
* upstream: attemp FIDO key signing without PIN and use the error
code returned to fall back only if necessary. Avoids PIN prompts for FIDO
tokens that don't require them; part of GHPR#302
OpenBSD-Commit-ID: 4f752aaf9f2e7c28bcaaf3d4f8fc290131bd038e
* Install Cygwin packages based on OS not config.
* initial list of allowed signers
* upstream: whitespace
OpenBSD-Commit-ID: d297e4387935d4aef091c5e9432578c2e513f538
* upstream: whitespace
OpenBSD-Commit-ID: a5d015efbfd228dc598ffdef612d2da3a579e5d8
* Add cygwin-release test target.
This also moves the cygwin package install from the workflow file to
setup_ci.sh so that we can install different sets of Cygwin packages
for different test configs.
* Add Windows 2022 test targets.
* Add libcrypt-devel to cygwin-release deps.
Based on feedback from vinschen at redhat.com.
* cross-sign allowed_signers with PGP key
Provides continuity of trust from legacy PGP release key to
the SSHSIG signing keys that we will use henceforth for git
signing.
* additional keys
* upstream: whitespace
OpenBSD-Commit-ID: c2bcbf93610d3d62ed206cdf9bf9ff98c6aaf232
* Move sftp from valgrind-2 to 3 to rebalance.
* upstream: sk-usbhid: fix key_lookup() on tokens with built-in UV
explicitly test whether the token performs built-in UV (e.g. biometric
tokens) and enable UV in that case. From Pedro Martelletto via GHPR#388
OpenBSD-Commit-ID: 007eb7e387d27cf3029ab06b88224e03eca62ccd
* Remove arc4random_uniform from arc4random.c
This was previously moved into its own file (matching OpenBSD) which
prematurely committed in commit 73541f2.
* Move OPENBSD ORIGINAL marker.
Putting this after the copyright statement (which doesn't change)
instead of before the version identifier (which does) prevents merge
conflicts when resyncing changes.
* Resync arc4random with OpenBSD.
This brings us up to current, including djm's random-reseeding change,
as prompted by logan at cyberstorm.mu in bz#3467. It brings the
platform-specific hooks from LibreSSL Portable, simplified to match our
use case. ok djm@.
* Remove DEF_WEAK, it's already in defines.h.
* openbsd-compat/bsd-asprintf: add <stdio.h> include for vsnprintf
Fixes the following build failure with Clang 15 on musl:
```
bsd-asprintf.c:51:8: error: call to undeclared library function 'vsnprintf' with type 'int (char *, unsigned long, const char *, struct __va_list_tag *)'; ISO C99 and laterclang -O2 -pipe -fdiagnostics-color=always -frecord-gcc-switches -pipe -Wunknown-warning-option -Qunused-arguments -Wall -Wpointer-arith -Wuninitialized -Wsign-compare -Wformat-security -Wsizeof-pointer-memaccess -Wno-pointer-sign -Wno-unused-result -Wmisleading-indentation -Wbitwise-instead-of-logical -fno-strict-aliasing -mretpoline -ftrapv -fzero-call-used-regs=all -fno-builtin-memset -fstack-protector-strong -fPIE -I. -I. -D_XOPEN_SOURCE=600 -D_BSD_SOURCE -D_DEFAULT_SOURCE -DSSHDIR=\"/etc/ssh\" -D_PATH_SSH_PROGRAM=\"/usr/bin/ssh\" -D_PATH_SSH_ASKPASS_DEFAULT=\"/usr/lib/misc/ssh-askpass\" -D_PATH_SFTP_SERVER=\"/usr/lib/misc/sftp-server\" -D_PATH_SSH_KEY_SIGN=\"/usr/lib/misc/ssh-keysign\" -D_PATH_SSH_PKCS11_HELPER=\"/usr/lib/misc/ssh-pkcs11-helper\" -D_PATH_SSH_SK_HELPER=\"/usr/lib/misc/ssh-sk-helper\" -D_PATH_SSH_PIDDIR=\"/run\" -D_PATH_PRIVSEP_CHROOT_DIR=\"/var/empty\" -DHAVE_CONFIG_H -c cipher-aes.c -o cipher-aes.o
do not support
implicit function declarations [-Wimplicit-function-declaration]
ret = vsnprintf(string, INIT_SZ, fmt, ap2);
^
bsd-asprintf.c:51:8: note: include the header <stdio.h> or explicitly provide a declaration for 'vsnprintf'
1 error generated.
```
* upstream: notifier_complete(NULL, ...) is a noop, so no need to test
that ctx!=NULL; from Corinna Vinschen
OpenBSD-Commit-ID: ade2f2e9cc519d01a586800c25621d910bce384a
* fix pester test failures
* upstream: fix repeated words ok miod@ jmc@
OpenBSD-Commit-ID: 6765daefe26a6b648cc15cadbbe337596af709b7
* upstream: .Li -> .Vt where appropriate; from josiah frentsos,
tweaked by schwarze
ok schwarze
OpenBSD-Commit-ID: 565046e3ce68b46c2f440a93d67c2a92726de8ed
* upstream: ssh-agent: attempt FIDO key signing without PIN and use
the error to determine whether a PIN is required and prompt only if
necessary. from Corinna Vinschen
OpenBSD-Commit-ID: dd6be6a0b7148608e834ee737c3479b3270b00dd
* upstream: a little extra debugging
OpenBSD-Commit-ID: edf1601c1d0905f6da4c713f4d9cecc7d1c0295a
* upstream: sk_enroll: never drop SSH_SK_USER_VERIFICATION_REQD flag
from response
Now that all FIDO signing calls attempt first without PIN and then
fall back to trying PIN only if that attempt fails, we can remove the
hack^wtrick that removed the UV flag from the keys returned during
enroll.
By Corinna Vinschen
OpenBSD-Commit-ID: 684517608c8491503bf80cd175425f0178d91d7f
* define HAVE_KILLPG
* upstream: sftp: Don't attempt to complete arguments for
non-existent commands
If user entered a non-existent command (e.g. because they made a
typo) there is no point in trying to complete its arguments. Skip
calling complete_match() if that's the case.
From Michal Privoznik
OpenBSD-Commit-ID: cf39c811a68cde2aeb98fc85addea4000ef6b07a
* upstream: sftp: Be a bit more clever about completions
There are commands (e.g. "get" or "put") that accept two
arguments, a local path and a remote path. However, the way
current completion is written doesn't take this distinction into
account and always completes remote or local paths.
By expanding CMD struct and "cmds" array this distinction can be
reflected and with small adjustment to completer code the correct
path can be completed.
By Michal Privoznik, ok dtucker@
OpenBSD-Commit-ID: 1396d921c4eb1befd531f5c4a8ab47e7a74b610b
* upstream: correct error value
OpenBSD-Commit-ID: 780efcbad76281f11f14b2a5ff04eb6db3dfdad4
* upstream: actually hook up restrict_websafe; the command-line flag
was never actually used. Spotted by Matthew Garrett
OpenBSD-Commit-ID: 0b363518ac4c2819dbaa3dfad4028633ab9cdff1
* upstream: Add a sshkey_check_rsa_length() call for checking the
length of an RSA key; ok markus@
OpenBSD-Commit-ID: de77cd5b11594297eda82edc594b0d32b8535134
* upstream: add a RequiredRSASize for checking RSA key length in
ssh(1). User authentication keys that fall beneath this limit will be
ignored. If a host presents a host key beneath this limit then the connection
will be terminated (unfortunately there are no fallbacks in the protocol for
host authentication).
feedback deraadt, Dmitry Belyavskiy; ok markus@
OpenBSD-Commit-ID: 430e339b2a79fa9ecc63f2837b06fdd88a7da13a
* upstream: Add RequiredRSASize for sshd(8); RSA keys that fall
beneath this limit will be ignored for user and host-based authentication.
Feedback deraadt@ ok markus@
OpenBSD-Commit-ID: 187931dfc19d51873df5930a04f2d972adf1f7f1
* upstream: better debugging for connect_next()
OpenBSD-Commit-ID: d16a307a0711499c971807f324484ed3a6036640
* upstream: sftp-server(8): add a "users-groups-by-id@openssh.com"
extension request that allows the client to obtain user/group names that
correspond to a set of uids/gids.
Will be used to make directory listings more useful and consistent
in sftp(1).
ok markus@
OpenBSD-Commit-ID: 7ebabde0bcb95ef949c4840fe89e697e30df47d3
* upstream: extend sftp-common.c:extend ls_file() to support supplied
user/group names; ok markus@
OpenBSD-Commit-ID: c70c70498b1fdcf158531117e405b6245863bfb0
* upstream: sftp client library support for
users-groups-by-id@openssh.com; ok markus@
OpenBSD-Commit-ID: ddb2f33a2da6349a9a89a8b5bcb9ca7c999394de
* upstream: use users-groups-by-id@openssh.com sftp-server extension
(when available) to fill in user/group names for directory listings.
Implement a client-side cache of see uid/gid=>user/group names. ok markus@
OpenBSD-Commit-ID: f239aeeadfa925a37ceee36ee8b256b8ccf4466e
* avoid Wuninitialized false positive in gcc-12ish
* no need for glob.h here
it also causes portability problems
* add debug on appveyor
* add sleep to pester test
* upstream: add RequiredRSASize to the list of keywords accepted by
-o; spotted by jmc@
OpenBSD-Commit-ID: fe871408cf6f9d3699afeda876f8adbac86a035e
* upstream: Fix typo. From AlexanderStohr via github PR#343.
OpenBSD-Commit-ID: a134c9b4039e48803fc6a87f955b0f4a03181497
* upstream: openssh-9.1
OpenBSD-Commit-ID: 5a467b2ee81da01a86adf1ad93b62b1728494e56
* crank versions in RPM spec files
* update release notes URL
* update .depend
* fix 9.1 compilation errors
* disable -p pester tests due to unreliability on older Windows versions
* remove extra sleep time from debugging scp pester tests
* modify -p tests to only run for Windows OS version 10 and above
* add windows specific code back into method moved from auth.c to auth2-pubkeyfile.c
* add preprocessor for WinHello
* revert preprocessor definition for winhello
* add windows preprocessor definition in key_lookup
* remove rdp block from appveyor since we are no longer debugging
* add ifdef to sftp-server.c
* make key_lookup compatible with winhello
* appveyor.yml
* increase debug of failing pester test
* add #ifdef SUPPORT_CRLF back into auth_check_principals_line method that was moved/renamed
* modify new scp.sh tests for windows
* remove in place tests from scp.sh
* remove rdp debug from appveyor
* retrigger appveyor
* change check of OS version in scp test
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
Co-authored-by: djm@openbsd.org <djm@openbsd.org>
Co-authored-by: Damien Miller <djm@mindrot.org>
Co-authored-by: Darren Tucker <dtucker@dtucker.net>
Co-authored-by: naddy@openbsd.org <naddy@openbsd.org>
Co-authored-by: dtucker@openbsd.org <dtucker@openbsd.org>
Co-authored-by: tj@openbsd.org <tj@openbsd.org>
Co-authored-by: millert@openbsd.org <millert@openbsd.org>
Co-authored-by: jmc@openbsd.org <jmc@openbsd.org>
Co-authored-by: florian@openbsd.org <florian@openbsd.org>
Co-authored-by: markus@openbsd.org <markus@openbsd.org>
Co-authored-by: Tobias Heider <me@tobhe.de>
Co-authored-by: anton@openbsd.org <anton@openbsd.org>
Co-authored-by: Tim Rice <tim@multitalents.net>
Co-authored-by: tobhe@openbsd.org <tobhe@openbsd.org>
Co-authored-by: Corinna Vinschen <vinschen@redhat.com>
Co-authored-by: Sam James <sam@gentoo.org>
Co-authored-by: jsg@openbsd.org <jsg@openbsd.org>
2022-11-02 17:06:45 +01:00
|
|
|
/* Prompt for touch for non-agent FIDO keys that request UP */
|
|
|
|
if (!is_agent && sshkey_is_sk(sign_key) &&
|
|
|
|
(sign_key->sk_flags & SSH_SK_USER_PRESENCE_REQD)) {
|
|
|
|
/* XXX should batch mode just skip these? */
|
|
|
|
if ((fp = sshkey_fingerprint(sign_key,
|
|
|
|
options.fingerprint_hash, SSH_FP_DEFAULT)) == NULL)
|
|
|
|
fatal_f("fingerprint failed");
|
|
|
|
notifier = notify_start(options.batch_mode,
|
|
|
|
"Confirm user presence for key %s %s",
|
|
|
|
sshkey_type(sign_key), fp);
|
|
|
|
free(fp);
|
2018-07-03 13:39:54 +02:00
|
|
|
}
|
2019-10-31 22:23:19 +01:00
|
|
|
if ((r = sshkey_sign(sign_key, sigp, lenp, data, datalen,
|
2020-08-27 03:06:18 +02:00
|
|
|
alg, options.sk_provider, pin, compat)) != 0) {
|
2020-10-18 13:32:01 +02:00
|
|
|
debug_fr(r, "sshkey_sign");
|
Merge 9.1 (#626)
* upstream: fix poll() spin when a channel's output fd closes without
data in the channel buffer. Introduce more exact packing of channel fds into
the pollfd array. fixes bz3405 and bz3411; ok deraadt@ markus@
OpenBSD-Commit-ID: 06740737849c9047785622ad5d472cb6a3907d10
* upstream: select post-quantum KEX
sntrup761x25519-sha512@openssh.com as the default; ok markus@
OpenBSD-Commit-ID: f02d99cbfce22dffec2e2ab1b60905fbddf48fb9
* upstream: add support for the "corp-data" protocol extension to
allow server-side copies to be performed without having to go via the client.
Patch by Mike Frysinger, ok dtucker@
OpenBSD-Commit-ID: 00aa510940fedd66dab1843b58682de4eb7156d5
* upstream: add a sftp client "cp" command that supports server-side
copying of files. Useful for this task and for testing the copy-data
extension. Patch from Mike Frysinger; ok dtucker@
OpenBSD-Commit-ID: 1bb1b950af0d49f0d5425b1f267e197aa1b57444
* depend
* Skip slow tests on (very) slow test targets.
* Set Makefile SHELL as determined by configure.
This should improve compatibility for users with non-POSIX shells. If
using Makefile.in directly (eg make -f Makefile.in distprep) then SHELL
will need to be specified on the command line (along with MANFMT in that
particular case). ok djm@
* Use bash or ksh if available for SH in Makefile.
* Increase test timeout to allow slow VMs to finish
* Only run regression tests on slow VMs.
* Only return events from ppoll that were requested.
If the underlying system's select() returns bits that were not in the
request set, our ppoll() implementation can return revents for events
not requested, which can apparently cause a hang. Only return revents
for activity in the requested event set. bz#3416, analysis and fix by
yaroslav.kuzmin at vmssoftware com, ok djm@
* Specify TEST_SHELL=bash on AIX.
The system shells cause the agent-restrict test to fail due to some
quoting so explicitly specify bash until we can get configure to
autmatically work around that.
* Disable security key on fbsd6 test host.
* upstream: man pages: add missing commas between subordinate and
main clauses
jmc@ dislikes a comma before "then" in a conditional, so leave those
untouched.
ok jmc@
OpenBSD-Commit-ID: 9520801729bebcb3c9fe43ad7f9776ab4dd05ea3
* upstream: ssh: document sntrup761x25519-sha512@openssh.com as
default KEX
OpenBSD-Commit-ID: 12545bfa10bcbf552d04d9d9520d0f4e98b0e171
* upstream: openssh-9.0
OpenBSD-Commit-ID: 0dfb461188f4513ec024c1534da8c1ce14c20b64
* update version numbers for release
* update build-aux files to match autoconf-2.71
i.e. config.guess, config.sub and install-sh
* Revert "update build-aux files to match autoconf-2.71"
This reverts commit 0a8ca39fac6ad19096b6c263436f8b2dd51606f2.
It turns out that the checked-in copies of these files are actually newer
than autoconf-2.71's copies, so this was effectively a downgrade.
Spotted by Bo Anderson via github
* upstream: two defensive changes from Tobias Stoeckmann via GHPR287
enforce stricter invarient for sshbuf_set_parent() - never allow
a buffer to have a previously-set parent changed.
In sshbuf_reset(), if the reallocation fails, then zero the entire
buffer and not the (potentially smaller) default initial alloc size.
OpenBSD-Commit-ID: 14583203aa5d50ad38d2e209ae10abaf8955e6a9
* upstream: Note that curve25519-sha256 was later published in
RFC8731. ok djm@
OpenBSD-Commit-ID: 2ac2b5d642d4cf5918eaec8653cad9a4460b2743
* upstream: clear io_want/io_ready flags at start of poll() cycle;
avoids plausible spin during rekeying if channel io_want flags are reused
across cycles. ok markus@ deraadt@
OpenBSD-Commit-ID: 91034f855b7c73cd2591657c49ac30f10322b967
* Retire fbsd6 test VM.
It's long since out of support, relatively slow (it's i686) and the
compiler has trouble with PIE.
* Resync moduli.5 with upstream.
1.18: remove duplicate publication year; carsten dot kunze at arcor dot de
1.19: ssh-keygen's -G/-T have been replaced with -M generate/screen.
* upstream: Correct path for system known hosts file in description
of IgnoreUserKnownHosts. Patch from Martin Vahlensieck via tech@
OpenBSD-Commit-ID: 9b7784f054fa5aa4d63cb36bd563889477127215
* upstream: list the correct version number
for when usage of the sftp protocol became default and fix a typo
from ed maste
OpenBSD-Commit-ID: 24e1795ed2283fdeacf16413c2f07503bcdebb31
* upstream: Import regenerated moduli
OpenBSD-Commit-ID: f9a0726d957cf10692a231996a1f34e7f9cdfeb0
* upstream: Try to continue running local I/O for channels in state
OPEN during SSH transport rekeying. The most visible benefit is that it
should make ~-escapes work in the client (e.g. to exit) if the connection
happened to have stalled during a rekey event. Based work by and ok dtucker@
OpenBSD-Commit-ID: a66e8f254e92edd4ce09c9f750883ec8f1ea5f45
* upstream: Import regenerated moduli
OpenBSD-Commit-ID: f9a0726d957cf10692a231996a1f34e7f9cdfeb0
* upstream: regression test for sftp cp command
OpenBSD-Regress-ID: c96bea9edde3a384b254785e7f9b2b24a81cdf82
* upstream: Simplify forward-control test.
Since we no longer need to support SSH1 we don't need to run shell
commands on the other end of the connection and can use ssh -N instead.
This also makes the test less racy.
OpenBSD-Regress-ID: 32e94ce272820cc398f30b848b2b0f080d10302c
* upstream: Use ssh -f and ControlPersist ..
to start up test forwards and ssh -O stop to shut them down intead of
sleep loops. This speeds up the test by an order of magnitude.
OpenBSD-Regress-ID: eb3db5f805100919b092a3b2579c611fba3e83e7
* upstream: It looks like we can't completely avoid
waiting for processes to exit so retrieve the pid via controlmaster and
use that.
OpenBSD-Regress-ID: 8246f00f22b14e49d2ff1744c94897ead33d457b
* Cache timezone data in capsicum sandbox.
From emaste at freebsd.org, originally part of FreeBSD commit r339216
/ fc3c19a9 with autoconf bits added by me.
* Include stdlib.h for free() prototype.
... which is used inside the CUSTOM_SYS_AUTH_GET_LASTLOGIN_MSG block.
* Update OpenSSL and LibreSSL versions in tests.
* Add debian-riscv64 test target.
* upstream: Avoid an unnecessary xstrdup in rm_env() when matching
patterns. Since match_pattern() doesn't modify its arguments (they are
const), there is no need to make an extra copy of the strings in
options->send_env. From Martin Vahlensieck
OpenBSD-Commit-ID: 2c9db31e3f4d3403b49642c64ee048b2a0a39351
* upstream: Add missing includes of stdlib.h and stdint.h. We need
stdlib.h for malloc(3) and stdint.h for SIZE_MAX. Unlike the other xmss
files, ssh-xmss.c does not include xmss_commons.h so ssh-xmss.c must include
those headers itself. From Martin Vahlensieck
OpenBSD-Commit-ID: 70e28a9818cee3da1be2ef6503d4b396dd421e6b
* upstream: Remove unnecessary includes: openssl/hmac.h and
openssl/evp.h. From Martin Vahlensieck.
OpenBSD-Commit-ID: a6debb5fb0c8a44e43e8d5ca7cc70ad2f3ea31c3
* upstream: Check sshauthopt_new() for NULL. bz#3425, from
tessgauthier at microsoft.com. ok djm@
OpenBSD-Commit-ID: af0315bc3e44aa406daa7e0ae7c2d719a974483f
* upstream: Add authfd path to debug output. ok markus@
OpenBSD-Commit-ID: f735a17d1a6f2bee63bfc609d76ef8db8c090890
* upstream: avoid printing hash algorithm twice; from lucas AT sexy.is
OpenBSD-Commit-ID: 9d24671e10a84141b7c504396cabad600e47a941
* upstream: fix memleak on session-bind path; from Pedro Martelletto, ok
dtucker@
OpenBSD-Commit-ID: e85899a26ba402b4c0717b531317e8fc258f0a7e
* upstream: Don't leak SK device. Patch from Pedro Martelletto via
github PR#316. ok djm@
OpenBSD-Commit-ID: 17d11327545022e727d95fd08b213171c5a4585d
* upstream: mention that the helpers are used by ssh(1), ssh-agent(1)
and ssh-keygen(1). Previously only ssh(1) was mentioned. From Pedro
Martelletto
OpenBSD-Commit-ID: 30f880f989d4b329589c1c404315685960a5f153
* Remove now-empty int32_minmax.inc.
* Only run tests when source files change.
Also run tests on changes to V_9_0 branch.
* Add Mac OS X 12 test target.
* upstream: be stricter in which characters will be accepted in
specifying a mask length; allow only 0-9. From khaleesicodes via GHPR#278; ok
dtucker@
OpenBSD-Commit-ID: e267746c047ea86665cdeccef795a8a56082eeb2
* upstream: fix some integer overflows in sieve_large() that show up when
trying to generate modp groups > 16k bits. Reported via GHPR#306 by Bertram
Felgenhauer, but fixed in a different way. feedback/ok tb@
OpenBSD-Commit-ID: 81cbc6dd3a21c57bd6fadea10e44afe37bca558e
* upstream: remove an obsolete rsa1 format example from an example;
from megan batty
ok djm
OpenBSD-Commit-ID: db2c89879c29bf083df996bd830abfb1e70d62bf
* upstream: Add FIDO AUTHENTICATOR section and explain a bit how FIDO
works. The wording came mostly from the 8.2 OpenSSH release notes, addapted
to fit the man page. Then move the -O bits into the new section as is already
done for CERTIFICATES and MODULI GENERATION. Finally we can explain the
trade-offs of resident keys. While here, consistently refer to the FIDO
thingies as "FIDO authenticators", not "FIDO tokens".
input & OK jmc, naddy
OpenBSD-Commit-ID: dd98748d7644df048f78dcf793b3b63db9ab1d25
* upstream: make sure stdout is non-blocking; ok djm@
OpenBSD-Commit-ID: 64940fffbd1b882eda2d7c8c7a43c79368309c0d
* upstream: mux.c: mark argument as const; from Martin Vahlensieck
OpenBSD-Commit-ID: 69a1a93a55986c7c2ad9f733c093b46a47184341
* upstream: channel_new no longer frees remote_name. So update the
comment accordingly. As remote_name is not modified, it can be const as
well. From Martin Vahlensieck
OpenBSD-Commit-ID: e4e10dc8dc9f40c166ea5a8e991942bedc75a76a
* upstream: sshkey_unshield_private() contains a exact duplicate of
the code in private2_check_padding(). Pull private2_check_padding() up so the
code can be reused. From Martin Vahlensieck, ok deraadt@
OpenBSD-Commit-ID: 876884c3f0e62e8fd8d1594bab06900f971c9c85
* Add ubsan minimal testcase on OpenBSD.
As suggested by djm@.
* Note that, for now, we need variadic macros.
* Also retest OpenBSD upstream on .yml changes.
* upstream: When performing operations that glob(3) a remote path, ensure
that the implicit working directory used to construct that path escapes
glob(3) characters.
This prevents glob characters from being processed in places they
shouldn't, e.g. "cd /tmp/a*/", "get *.txt" should have the get operation
treat the path "/tmp/a*" literally and not attempt to expand it.
Reported by Lusia Kundel; ok markus@
OpenBSD-Commit-ID: 4f647f58482cbad3d58b1eab7f6a1691433deeef
* Remove duplicate bcrypt_pbkdf.o from Makefile
bcrypt_pbkdf.o is duplicated in the openbsd-compat Makefile's object
file list.
* upstream: improve error message when 'ssh-keygen -Y sign' is unable to
load a private key; bz3429, reported by Adam Szkoda ok dtucker@
OpenBSD-Commit-ID: bb57b285e67bea536ef81b1055467be2fc380e74
* upstream: Allow existing -U (use agent) flag to work with "-Y sign"
operations, where it will be interpreted to require that the private keys is
hosted in an agent; bz3429, suggested by Adam Szkoda; ok dtucker@
OpenBSD-Commit-ID: a7bc69873b99c32c42c7628ed9ea91565ba08c2f
* upstream: Remove errant apostrophe. From haruyama at queen-ml org.
OpenBSD-Commit-ID: dc6b294567cb84b384ad6ced9ca469f2bbf0bd10
* upstream: arrange for scp, when in sftp mode, to not ftruncate(3) files
early
previous behavious of unconditionally truncating the destination file
would cause "scp ~/foo localhost:" and "scp localhost:foo ~/" to
delete all the contents of their destination.
spotted by solene@ sthen@, also bz3431; ok dtucker@
OpenBSD-Commit-ID: ca39fdd39e0ec1466b9666f15cbcfddea6aaa179
* upstream: fix in-place copies; r1.163 incorrectly skipped truncation in
all cases, not just at the start of a transfer. This could cause overwrites
of larger files to leave junk at the end. Spotted by tb@
OpenBSD-Commit-ID: b189f19cd68119548c8e24e39c79f61e115bf92c
* upstream: Only run agent-ptrace.sh if gdb is available as all
architectures do not ship with gdb.
OpenBSD-Regress-ID: ec53e928803e6b87f9ac142d38888ca79a45348d
* upstream: regress test for in-place transfers and clobbering larger
files with smaller ones; would have caught last regression in scp(1)
OpenBSD-Regress-ID: 19de4e88dd3a4f7e5c1618c9be3c32415bd93bc2
* configure.ac: Add missing AC_DEFINE for caph_cache_tzdata test causing
HAVE_CAPH_CACHE_TZDATA to be missing from config.h.in.
Spotted by Bryan Drewery
* upstream: make SSHBUF_DBG/SSHBUF_TELL (off by default and only enabled
via #define) dump to stderr rather than stdout
OpenBSD-Commit-ID: 10298513ee32db8390aecb0397d782d68cb14318
* upstream: revert previous; it was broken (spotted by Theo)
OpenBSD-Commit-ID: 457c79afaca2f89ec2606405c1059b98b30d8b0d
* upstream: Note that ProxyJump also accepts the same tokens as
ProxyCommand. From pallxk via github PR#305.
OpenBSD-Commit-ID: 7115ac351b129205f1f1ffa6bbfd62abd76be7c5
* upstream: Avoid kill with -1 argument. The out_ctx label can be
reached before fork has been called. If this happens, then kill -1 would be
called, sending SIGTERM to all processes reachable by the current process.
From tobias@ and c3h2_ctf via github PR#286, ok djm@
OpenBSD-Commit-ID: 6277af1207d81202f5daffdccfeeaed4c763b1a8
* upstream: f sshpkt functions fail, then password is not cleared
with freezero. Unconditionally call freezero to guarantee that password is
removed from RAM.
From tobias@ and c3h2_ctf via github PR#286, ok djm@
OpenBSD-Commit-ID: 6b093619c9515328e25b0f8093779c52402c89cd
* upstream: refactor authorized_keys/principals handling
remove "struct ssh *" from arguments - this was only used to pass the
remote host/address. These can be passed in instead and the resulting
code is less tightly coupled to ssh_api.[ch]
ok dtucker@
OpenBSD-Commit-ID: 9d4373d013edc4cc4b5c21a599e1837ac31dda0d
* upstream: split the low-level file handling functions out from
auth2-pubkey.c
Put them in a new auth2-pubkeyfile.c to make it easier to refer to them
(e.g. in unit/fuzz tests) without having to refer to everything else
pubkey auth brings in.
ok dtucker@
OpenBSD-Commit-ID: 3fdca2c61ad97dc1b8d4a7346816f83dc4ce2217
* fuzzer for authorized_keys parsing
mostly redundant to authopt_fuzz, but it's sensitive code so IMO it
makes sense to test this layer too
* Test against LibreSSL 3.5.3.
* Test against OpenSSL 1.1.1o and 3.0.3.
* fix some bugs in the fuzzer
* upstream: keywords ref ssh_config.5;
from caspar schutijser
OpenBSD-Commit-ID: f146a19d7d5c9374c3b9c520da43b2732d7d1a4e
* upstream: ssh-keygen: implement "verify-required" certificate option.
This was already documented when support for user-verified FIDO
keys was added, but the ssh-keygen(1) code was missing.
ok djm@
OpenBSD-Commit-ID: f660f973391b593fea4b7b25913c9a15c3eb8a06
* upstream: ssh-keygen -A: do not generate DSA keys by default.
Based on github PR#303 from jsegitz with man page text from jmc@, ok markus@
djm@
OpenBSD-Commit-ID: 5c4c57bdd7063ff03381cfb6696659dd3f9f5b9f
* upstream: Add period at end of "not known by any other names"
message. github PR#320 from jschauma, ok djm@
OpenBSD-Commit-ID: bd60809803c4bfd3ebb7c5c4d918b10e275266f2
* upstream: Add missing *-sk types to ssh-keyscan manpage. From
skazi0 via github PR#294.
OpenBSD-Commit-ID: fda2c869cdb871f3c90a89fb3f985370bb5d25c0
* upstream: Make SetEnv directives first-match-wins in both
sshd_config and sshd_config; previously if the same name was reused then the
last would win (which is the opposite to how the config is supposed to work).
While there, make the ssh_config parsing more like sshd_config.
bz3438, ok dtucker
OpenBSD-Commit-ID: 797909c1e0262c0d00e09280459d7ab00f18273b
* upstream: test setenv in both client and server, test first-match-wins
too
OpenBSD-Regress-ID: 4c8804f9db38a02db480b9923317457b377fe34b
* upstream: move auth_openprincipals() and auth_openkeyfile() over to
auth2-pubkeyfile.c too; they make more sense there.
OpenBSD-Commit-ID: 9970d99f900e1117fdaab13e9e910a621b7c60ee
* upstream: make sure that UseDNS hostname lookup happens in the monitor
and not in the pledge(2)'d unprivileged process; fixes regression caused by
recent refactoring spotted by henning@
OpenBSD-Commit-ID: a089870b95101cd8881a2dff65b2f1627d13e88d
* fix possible NULL deref when built without FIDO
Analysis/fix from kircher in bz3443; ok dtucker@
* automatically enable built-in FIDO support
If libfido2 is found and usable, then enable the built-in
security key support unless --without-security-key-builtin
was requested.
ok dtucker@
* upstream: Log an error if pipe() fails while accepting a
connection. bz#3447, from vincent-openssh at vinc17 net, ok djm@
OpenBSD-Commit-ID: 9d59f19872b94900a5c79da2d57850241ac5df94
* upstream: Don't attempt to fprintf a null identity comment. From
Martin Vahlensieck via tech@.
OpenBSD-Commit-ID: 4c54d20a8e8e4e9912c38a7b4ef5bfc5ca2e05c2
* upstream: Make sure not to fclose() the same fd twice in case of an
error.
ok dtucker@
OpenBSD-Commit-ID: e384c4e05d5521e7866b3d53ca59acd2a86eef99
* upstream: make it clear that RekeyLimit applies to both transmitted
and received data. GHPR#328 from Jan Pazdziora
OpenBSD-Commit-ID: d180a905fec9ff418a75c07bb96ea41c9308c3f9
* request 1.1x API compatibility for OpenSSL >=3.x
idea/patch from Pedro Martelletto via GHPR#322; ok dtucker@
* fix broken case statement in previous
* Disable SK support if FIDO libs not found.
* Zero out LIBFIDO2 when SK support not usable.
Prevents us from trying to link them into ssh-sk-helper and failing to
build.
* upstream: Don't leak the strings allocated by order_hostkeyalgs()
and list_hostkey_types() that are passed to compat_pkalg_proposal(). Part of
github PR#324 from ZoltanFridrich, ok djm@
OpenBSD-Commit-ID: b2f6e5f60f2bba293b831654328a8a0035ef4a1b
* upstream: Roll back previous KEX changes as they aren't safe until
compat_pkalg_proposal and friends always allocate their returned strings.
Reported by Qualys.
OpenBSD-Commit-ID: 1c7a88a0d5033f42f88ab9bec58ef1cf72c81ad0
* upstream: allow arguments to sftp -D option, e.g. sftp -D
"/usr/libexec/sftp-server -el debug3"
ok markus@
OpenBSD-Commit-ID: 5a002b9f3a7aef2731fc0ffa9c921cf15f38ecce
* Update OpenSSL tests to the most recent releases.
* upstream: reflect the update to -D arg name in usage();
OpenBSD-Commit-ID: abdcde4f92b1ef094ae44210ee99d3b0155aad9c
* upstream: ignore SIGPIPE earlier in main(), specifically before
muxclient() which performs operations that could cause one; Reported by Noam
Lewis via bz3454, ok dtucker@
OpenBSD-Commit-ID: 63d8e13276869eebac6d7a05d5a96307f9026e47
* upstream: Always return allocated strings from the kex filtering so
that we can free them later. Fix one leak in compat_kex_proposal. Based on
github PR#324 from ZoltanFridrich with some simplications by me. ok djm@
OpenBSD-Commit-ID: 9171616da3307612d0ede086fd511142f91246e4
* upstream: Don't leak the strings allocated by order_hostkeyalgs()
and list_hostkey_types() that are passed to compat_pkalg_proposal(). Part of
github PR#324 from ZoltanFridrich, ok djm@
This is a roll-forward of the previous rollback now that the required
changes in compat.c have been done.
OpenBSD-Commit-ID: c7cd93730b3b9f53cdad3ae32462922834ef73eb
* upstream: bump up loglevel from debug to info when unable to open
authorized keys/principals file for errno != ENOENT; bz2042 ok dtucker
OpenBSD-Commit-ID: e79aa550d91ade6a80f081bda689da24c086d66b
* Skip select+rlimit check if sandboxing is disabled
It's not needed in that case, and the test can fail when being built
with some compiler memory sanitizer flags. bz#3441
* upstream: use consistent field names (s/char/byte)
in format description
OpenBSD-Commit-ID: 3de33572733ee7fcfd7db33d37db23d2280254f0
* upstream: Remove leftover line.
Remove extra line leftover from merge conflict. ok djm@
OpenBSD-Commit-ID: 460e2290875d7ae64971a7e669c244b1d1c0ae2e
* Move checks for pollfd.fd and nfds_t.
Move the checks for struct pollfd.fd and nfds_t to before the sandboxing
checks. This groups all the sandbox checks together so we can skip them
all when sandboxing is disabled.
* Skip all rlimit tests when sandboxing disabled.
The rlimit tests can hang when being run with some compiler sanitizers
so skip all of them if sandbox=no.
* Add clang sanitizer tests.
* upstream: Add TEST_REGRESS_CACHE_DIR.
If set, it is used to cache regress test names that have succeeded and
skip those on a re-run.
OpenBSD-Regress-ID: a7570dd29a58df59f2cca647c3c2ec989b49f247
* Move sanitizer logs into regress for collection.
* Add GCC address sanitizer build/test.
* Update sanitizer test targets:
- remove clang-sanitize-memory for now. It takes so long that the test
times out.
- add gcc sanitize-address and sanitize-undefined test targets.
* Test against openssl-3.0.5.
* Move unset to before we set anything.
* Refuse to use OpenSSL 3.0.4 due to potential RCE.
OpenSSL has a potential RCE in its RSA implementation (CVE-2022-2274)
so refuse to use that specific version.
* Capture stderr output from configure.
* Only refuse to use OpenSSL 3.0.4 on x86_64.
The potential RCE only impacts x86_64, so only refuse to use it if we're
targetting a potentially impacted architecture. ok djm@
* Remove special casing of crypt().
Configure goes to some lengths to pick crypt() from either libcrypt
or OpenSSL's libcrypto because they can more or less featureful (eg
supporting md5-style passwords).
OpenSSL removed its crypt() interface in 2002:
https://github.com/openssl/openssl/commit/69deec58 so these hijinks
should no longer be necessary. This also only links sshd with libcrypt
which is the only thing that needs it. ok djm@
* Clarify README.md text.
Clarify the text about the implications of building without OpenSSL, and
prefix the "configure --help" example command with a "./" so it's likely
to work as-is in more shells. From bz#3461.
* Split README.platform into its own line.
README.platform has general platform-specific information, having it
following text about FIDO2 on the same line could imply that it only
has information about FIDO2.
* Return ERANGE from getcwd() if buffer size is 1.
If getcwd() is supplied a buffer size of exactly 1 and a path of "/", it
could result in a nul byte being written out of array bounds. POSIX says
it should return ERANGE if the path will not fit in the available buffer
(with terminating nul). 1 byte cannot fit any possible path with its nul,
so immediately return ERANGE in that case.
OpenSSH never uses getcwd() with this buffer size, and all current
(and even quite old) platforms that we are currently known to work
on have a native getcwd() so this code is not used on those anyway.
Reported by Qualys, ok djm@
* Remove unintended changes.
I inadvertently included a couple of local changes with the OpenSSL
3.0.4 change. Revert, anything that should be there will be committed
separately.
* Add AUDIT_ARCH_PPC to supported seccomp arches.
Patch from dries.deschout at dodeco.eu.
* Rename bbone test target to ARM.
* Move vmshutdown to first step.
If a previous run on a physical runner has failed to clean up, the next
run will fail because it'll try to check out the code to a broken
directory mount. Make cleanup the first step.
* upstream: pull passphrase reading and confirmation into a separate
function so it can be used for FIDO2 PINs; no functional change
OpenBSD-Commit-ID: bf34f76b8283cc1d3f54633e0d4f13613d87bb2f
* upstream: when enrolling a resident key on a security token, check
if a credential with matching application and user ID strings already exists.
if so, prompt the user for confirmation before overwriting the credential.
patch from Pedro Martelletto via GHPR329
NB. cranks SSH_SK_VERSION_MAJOR, so any third-party FIDO middleware
implementations will need to adjust
OpenBSD-Commit-ID: e45e9f1bf2b2f32d9850669e7a8dbd64acc5fca4
* upstream: sk-usbhid: preserve error code returned by key_lookup()
it conveys useful information, such as the supplied pin being wrong.
Part of GHPR329 from Pedro Martelletto
OpenBSD-Commit-ID: c0647eb9290f793add363d81378439b273756c1b
* upstream: ssh-keygen: fix touch prompt, pin retries;
part of GHPR329 from Pedro Martelletto
OpenBSD-Commit-ID: 75d1005bd2ef8f29fa834c90d2684e73556fffe8
* crank SSH_SK_VERSION_MAJOR in sk-dummy.so
* Skip scp3 test if there's no scp on remote path.
scp -3 ends up using the scp that's in the remote path and will fail if
one is not available. Based on a patch from rapier at psc.edu.
* Convert "have_prog" function into "which".
"which" and its behaviour is not standardized, so convert the existing
have_prog function into "which" so we can rely on it being available
and what its semantics are. Add a have_prog wrapper that maintains the
existing behaviour.
* upstream: Test TEST_SSH_ELAPSED_TIMES for empty string not
executable. No-op on most platforms but should prevent warnings in -portable
on systems that don't have 'date %s'.
OpenBSD-Regress-ID: e39d79867b8065e33d0c5926fa1a31f85659d2a4
* upstream: Restore missing "!" in TEST_SSH_ELAPSED_TIMES test.
OpenBSD-Regress-ID: 38783f9676ec348c5a792caecee9a16e354b37b0
* Remove workarounds for OpenSSL missing AES-GCM.
We have some compatibility hacks that were added to support OpenSSL
versions that do not support AES GCM mode. Since that time, however,
the minimum OpenSSL version that we support has moved to 1.0.1 which
*does* have GCM, so this is no longer needed. ok djm@
* Remove workarounds for OpenSSL missing AES-CTR.
We have some compatibility hacks that were added to support OpenSSL
versions that do not support AES CTR mode. Since that time, however,
the minimum OpenSSL version that we support has moved to 1.0.1 which
*does* have CTR, so this is no longer needed. ok djm@
* Do not link scp, sftp and sftp-server w/ zlib.
Some of our binaries (eg sftp, sftp-server, scp) do not interact with
the channels code and thus do use libraries such as zlib and libcrypto
although they are linked with them. This adds a CHANNELLIBS and starts
by moving zlib into it, which means the aformentioned binaries are no
longer linked against zlib. ok djm@
* Group libcrypto and PRNGD checks together.
They're related more than the libcrypt or libiaf checks which are
currently between them. ok djm@
* Remove seed_rng calls from scp, sftp, sftp-server.
These binaries don't use OpenSSL's random functions. The next step
will be to stop linking them against libcrypto. ok djm@
* Move libcrypto into CHANNELLIBS.
This will result in sftp, sftp-server and scp no longer being linked
against libcrypto. ok djm@
* Move stale-configure check as early as possible.
We added a check in Makefile to catch the case where configure needs to
be rebuilt, however this did not happen until a build was attempted in
which case all of the work done by configure was wasted. Move this check
to the start of configure to catch it as early as possible. ok djm@
* Remove deprecated MacOS 10.15 runners.
* upstream: avoid double-free in error path introduced in r1.70; report
and fix based on GHPR#332 by v-rzh ok dtucker@
OpenBSD-Commit-ID: 3d21aa127b1f37cfc5bdc21461db369a663a951f
* Include CHANNEL and FIDO2 libs in configure output
* Factor out getrnd() and rename to getentropy().
Factor out the arc4random seeding into its own file and change the
interface to match getentropy. Use native getentropy if available.
This will make it easier to resync OpenBSD changes to arc4random.
Prompted by bz#3467, ok djm@.
* compat code for fido_dev_is_winhello()
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* check_sk_options: add temporary WinHello workaround
Up to libfido 1.10.0, WinHello advertises "clientPin" rather
than "uv" capability. This is fixed in 1.11.0. For the time
being, workaround it here.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* sk_sign: set FIDO2 uv attribute explicitely for WinHello
WinHello via libfido2 performs user verification by default.
However, if we stick to that, there's no way to differentiate
between keys created with or without "-O verify-required".
Set FIDO2 uv attribute explicitely to FIDO_OPT_FALSE, then check
if user verification has been requested.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* upstream: don't prompt for FIDO passphrase before attempting to enroll
the credential, just let the enroll operating fail and we'll attempt to get a
PIN anyway. Might avoid some unneccessary PIN prompts.
Part of GHPR#302 from Corinna Vinschen; ok dtucker@
OpenBSD-Commit-ID: bd5342ffc353ee37d39617906867c305564d1ce2
* Give unused param a name.
Fixes builds on platforms that do have fido2 but don't have
fido_dev_is_winhello.
* Actually put HAVE_STDINT_H around the stdint.h.
* Rename our getentropy to prevent possible loops.
Since arc4random seeds from getentropy, and we use OpenSSL for that
if enabled, there's the possibility that if we build on a system that
does not have getentropy then run on a system that does have it, then
OpenSSL could end up calling our getentropy and getting stuck in a loop.
Pointed out by deraadt@, ok djm@
* Test hostbased auth on github runners.
* fix SANDBOX_SECCOMP_FILTER_DEBUG
* Fix conditional for running hostbased tests.
* upstream: allow certificate validity intervals, sshsig verification
times and authorized_keys expiry-time options to accept dates in the UTC time
zone in addition to the default of interpreting them in the system time zone.
YYYYMMDD and YYMMDDHHMM[SS] dates/times will be interpreted as UTC if
suffixed with a 'Z' character.
Also allow certificate validity intervals to be specified in raw
seconds-since-epoch as hex value, e.g. -V 0x1234:0x4567890. This
is intended for use by regress tests and other tools that call
ssh-keygen as part of a CA workflow.
bz3468 ok dtucker
OpenBSD-Commit-ID: 454db1cdffa9fa346aea5211223a2ce0588dfe13
* upstream: add some tests for parse_absolute_time(), including cases
where it is forced to the UTC timezone. bz3468 ok dtucker
OpenBSD-Regress-ID: ea07ca31c2f3847a38df028ca632763ae44e8759
* Skip hostbased during Valgrind tests.
Valgrind doesn't let ssh exec ssh-keysign (because it's setuid) so skip
it during the Valgrind based tests.
See https://bugs.kde.org/show_bug.cgi?id=119404 for a discussion of this
(ironically there the problematic binary was ssh(1) back when it could
still be setuid).
* Rerun tests if any .github config file changes.
* Add a timegm implementation from Heimdal via Samba.
Fixes build on (at least Solaris 10).
* Replace deprecated ubuntu-18.04 runners with 22.04
* upstream: sftp-server: support home-directory request
Add support to the sftp-server for the home-directory extension defined
in draft-ietf-secsh-filexfer-extensions-00. This overlaps a bit with the
existing expand-path@openssh.com, but uses a more official protocol name,
and so is a bit more likely to be implemented by non-OpenSSH clients.
From Mike Frysinger, ok dtucker@
OpenBSD-Commit-ID: bfc580d05cc0c817831ae7ecbac4a481c23566ab
* fido_dev_is_winhello: return 0, not "false"
"false" is not used anywhere in OpenSSH, so return 0 like
everywhere else.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* Revert "check_sk_options: add temporary WinHello workaround"
Cygwin now comes with libfido2 1.11.0, so this workaround
isn't required anymore.
This reverts commit 242c044ab111a37aad3b0775727c36a4c5f0102c.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* upstream: use .Cm for "sign"; from josiah frentsos
OpenBSD-Commit-ID: 7f80a53d54857ac6ae49ea6ad93c5bd12231d1e4
* upstream: add an extra flag to sk_probe() to indicate whether we're
probing for a FIDO resident key or not. Unused here, but will make like
easier for portable
OpenBSD-Commit-ID: 432c8ff70e270378df9dbceb9bdeaa5b43b5a832
* on Cygwin, prefer WinHello FIDO device
If no FIDO device was explictly specified, then prefer the
windows://hello FIDO device. An exception to this is when
probing resident FIDO keys, in which case hardware FIDO
devices are preferred.
* Check for perms to run agent-getpeereid test.
Ubuntu 22.04 defaults to private home dirs which prevents "nobody"
running ssh-add during the agent-getpeereid test. Check for this and
add the necessary permissions.
* upstream: double free() in error path; from Eusgor via GHPR333
OpenBSD-Commit-ID: 39f35e16ba878c8d02b4d01d8826d9b321be26d4
* Add Cygwin (on windows-2019) test target.
In addition to installing the requisite Cygwin packages, we also need to
explicitly invoke "sh" for steps that run other scripts since the runner
environment doesn't understand #! paths.
* Add a bit more debug output.
* Fix cygwin conditional steps.
* upstream: Strictly enforce the maximum allowed SSH2 banner size in
ssh-keyscan and prevent a one-byte buffer overflow. Patch from Qualys, ok
djm@
OpenBSD-Commit-ID: 6ae664f9f4db6e8a0589425f74cd0bbf3aeef4e4
* upstream: remove incorrect check that can break enrolling a
resident key (introduced in r1.40)
OpenBSD-Commit-ID: 4cab364d518470e29e624af3d3f9ffa9c92b6f01
* upstream: attemp FIDO key signing without PIN and use the error
code returned to fall back only if necessary. Avoids PIN prompts for FIDO
tokens that don't require them; part of GHPR#302
OpenBSD-Commit-ID: 4f752aaf9f2e7c28bcaaf3d4f8fc290131bd038e
* Install Cygwin packages based on OS not config.
* initial list of allowed signers
* upstream: whitespace
OpenBSD-Commit-ID: d297e4387935d4aef091c5e9432578c2e513f538
* upstream: whitespace
OpenBSD-Commit-ID: a5d015efbfd228dc598ffdef612d2da3a579e5d8
* Add cygwin-release test target.
This also moves the cygwin package install from the workflow file to
setup_ci.sh so that we can install different sets of Cygwin packages
for different test configs.
* Add Windows 2022 test targets.
* Add libcrypt-devel to cygwin-release deps.
Based on feedback from vinschen at redhat.com.
* cross-sign allowed_signers with PGP key
Provides continuity of trust from legacy PGP release key to
the SSHSIG signing keys that we will use henceforth for git
signing.
* additional keys
* upstream: whitespace
OpenBSD-Commit-ID: c2bcbf93610d3d62ed206cdf9bf9ff98c6aaf232
* Move sftp from valgrind-2 to 3 to rebalance.
* upstream: sk-usbhid: fix key_lookup() on tokens with built-in UV
explicitly test whether the token performs built-in UV (e.g. biometric
tokens) and enable UV in that case. From Pedro Martelletto via GHPR#388
OpenBSD-Commit-ID: 007eb7e387d27cf3029ab06b88224e03eca62ccd
* Remove arc4random_uniform from arc4random.c
This was previously moved into its own file (matching OpenBSD) which
prematurely committed in commit 73541f2.
* Move OPENBSD ORIGINAL marker.
Putting this after the copyright statement (which doesn't change)
instead of before the version identifier (which does) prevents merge
conflicts when resyncing changes.
* Resync arc4random with OpenBSD.
This brings us up to current, including djm's random-reseeding change,
as prompted by logan at cyberstorm.mu in bz#3467. It brings the
platform-specific hooks from LibreSSL Portable, simplified to match our
use case. ok djm@.
* Remove DEF_WEAK, it's already in defines.h.
* openbsd-compat/bsd-asprintf: add <stdio.h> include for vsnprintf
Fixes the following build failure with Clang 15 on musl:
```
bsd-asprintf.c:51:8: error: call to undeclared library function 'vsnprintf' with type 'int (char *, unsigned long, const char *, struct __va_list_tag *)'; ISO C99 and laterclang -O2 -pipe -fdiagnostics-color=always -frecord-gcc-switches -pipe -Wunknown-warning-option -Qunused-arguments -Wall -Wpointer-arith -Wuninitialized -Wsign-compare -Wformat-security -Wsizeof-pointer-memaccess -Wno-pointer-sign -Wno-unused-result -Wmisleading-indentation -Wbitwise-instead-of-logical -fno-strict-aliasing -mretpoline -ftrapv -fzero-call-used-regs=all -fno-builtin-memset -fstack-protector-strong -fPIE -I. -I. -D_XOPEN_SOURCE=600 -D_BSD_SOURCE -D_DEFAULT_SOURCE -DSSHDIR=\"/etc/ssh\" -D_PATH_SSH_PROGRAM=\"/usr/bin/ssh\" -D_PATH_SSH_ASKPASS_DEFAULT=\"/usr/lib/misc/ssh-askpass\" -D_PATH_SFTP_SERVER=\"/usr/lib/misc/sftp-server\" -D_PATH_SSH_KEY_SIGN=\"/usr/lib/misc/ssh-keysign\" -D_PATH_SSH_PKCS11_HELPER=\"/usr/lib/misc/ssh-pkcs11-helper\" -D_PATH_SSH_SK_HELPER=\"/usr/lib/misc/ssh-sk-helper\" -D_PATH_SSH_PIDDIR=\"/run\" -D_PATH_PRIVSEP_CHROOT_DIR=\"/var/empty\" -DHAVE_CONFIG_H -c cipher-aes.c -o cipher-aes.o
do not support
implicit function declarations [-Wimplicit-function-declaration]
ret = vsnprintf(string, INIT_SZ, fmt, ap2);
^
bsd-asprintf.c:51:8: note: include the header <stdio.h> or explicitly provide a declaration for 'vsnprintf'
1 error generated.
```
* upstream: notifier_complete(NULL, ...) is a noop, so no need to test
that ctx!=NULL; from Corinna Vinschen
OpenBSD-Commit-ID: ade2f2e9cc519d01a586800c25621d910bce384a
* fix pester test failures
* upstream: fix repeated words ok miod@ jmc@
OpenBSD-Commit-ID: 6765daefe26a6b648cc15cadbbe337596af709b7
* upstream: .Li -> .Vt where appropriate; from josiah frentsos,
tweaked by schwarze
ok schwarze
OpenBSD-Commit-ID: 565046e3ce68b46c2f440a93d67c2a92726de8ed
* upstream: ssh-agent: attempt FIDO key signing without PIN and use
the error to determine whether a PIN is required and prompt only if
necessary. from Corinna Vinschen
OpenBSD-Commit-ID: dd6be6a0b7148608e834ee737c3479b3270b00dd
* upstream: a little extra debugging
OpenBSD-Commit-ID: edf1601c1d0905f6da4c713f4d9cecc7d1c0295a
* upstream: sk_enroll: never drop SSH_SK_USER_VERIFICATION_REQD flag
from response
Now that all FIDO signing calls attempt first without PIN and then
fall back to trying PIN only if that attempt fails, we can remove the
hack^wtrick that removed the UV flag from the keys returned during
enroll.
By Corinna Vinschen
OpenBSD-Commit-ID: 684517608c8491503bf80cd175425f0178d91d7f
* define HAVE_KILLPG
* upstream: sftp: Don't attempt to complete arguments for
non-existent commands
If user entered a non-existent command (e.g. because they made a
typo) there is no point in trying to complete its arguments. Skip
calling complete_match() if that's the case.
From Michal Privoznik
OpenBSD-Commit-ID: cf39c811a68cde2aeb98fc85addea4000ef6b07a
* upstream: sftp: Be a bit more clever about completions
There are commands (e.g. "get" or "put") that accept two
arguments, a local path and a remote path. However, the way
current completion is written doesn't take this distinction into
account and always completes remote or local paths.
By expanding CMD struct and "cmds" array this distinction can be
reflected and with small adjustment to completer code the correct
path can be completed.
By Michal Privoznik, ok dtucker@
OpenBSD-Commit-ID: 1396d921c4eb1befd531f5c4a8ab47e7a74b610b
* upstream: correct error value
OpenBSD-Commit-ID: 780efcbad76281f11f14b2a5ff04eb6db3dfdad4
* upstream: actually hook up restrict_websafe; the command-line flag
was never actually used. Spotted by Matthew Garrett
OpenBSD-Commit-ID: 0b363518ac4c2819dbaa3dfad4028633ab9cdff1
* upstream: Add a sshkey_check_rsa_length() call for checking the
length of an RSA key; ok markus@
OpenBSD-Commit-ID: de77cd5b11594297eda82edc594b0d32b8535134
* upstream: add a RequiredRSASize for checking RSA key length in
ssh(1). User authentication keys that fall beneath this limit will be
ignored. If a host presents a host key beneath this limit then the connection
will be terminated (unfortunately there are no fallbacks in the protocol for
host authentication).
feedback deraadt, Dmitry Belyavskiy; ok markus@
OpenBSD-Commit-ID: 430e339b2a79fa9ecc63f2837b06fdd88a7da13a
* upstream: Add RequiredRSASize for sshd(8); RSA keys that fall
beneath this limit will be ignored for user and host-based authentication.
Feedback deraadt@ ok markus@
OpenBSD-Commit-ID: 187931dfc19d51873df5930a04f2d972adf1f7f1
* upstream: better debugging for connect_next()
OpenBSD-Commit-ID: d16a307a0711499c971807f324484ed3a6036640
* upstream: sftp-server(8): add a "users-groups-by-id@openssh.com"
extension request that allows the client to obtain user/group names that
correspond to a set of uids/gids.
Will be used to make directory listings more useful and consistent
in sftp(1).
ok markus@
OpenBSD-Commit-ID: 7ebabde0bcb95ef949c4840fe89e697e30df47d3
* upstream: extend sftp-common.c:extend ls_file() to support supplied
user/group names; ok markus@
OpenBSD-Commit-ID: c70c70498b1fdcf158531117e405b6245863bfb0
* upstream: sftp client library support for
users-groups-by-id@openssh.com; ok markus@
OpenBSD-Commit-ID: ddb2f33a2da6349a9a89a8b5bcb9ca7c999394de
* upstream: use users-groups-by-id@openssh.com sftp-server extension
(when available) to fill in user/group names for directory listings.
Implement a client-side cache of see uid/gid=>user/group names. ok markus@
OpenBSD-Commit-ID: f239aeeadfa925a37ceee36ee8b256b8ccf4466e
* avoid Wuninitialized false positive in gcc-12ish
* no need for glob.h here
it also causes portability problems
* add debug on appveyor
* add sleep to pester test
* upstream: add RequiredRSASize to the list of keywords accepted by
-o; spotted by jmc@
OpenBSD-Commit-ID: fe871408cf6f9d3699afeda876f8adbac86a035e
* upstream: Fix typo. From AlexanderStohr via github PR#343.
OpenBSD-Commit-ID: a134c9b4039e48803fc6a87f955b0f4a03181497
* upstream: openssh-9.1
OpenBSD-Commit-ID: 5a467b2ee81da01a86adf1ad93b62b1728494e56
* crank versions in RPM spec files
* update release notes URL
* update .depend
* fix 9.1 compilation errors
* disable -p pester tests due to unreliability on older Windows versions
* remove extra sleep time from debugging scp pester tests
* modify -p tests to only run for Windows OS version 10 and above
* add windows specific code back into method moved from auth.c to auth2-pubkeyfile.c
* add preprocessor for WinHello
* revert preprocessor definition for winhello
* add windows preprocessor definition in key_lookup
* remove rdp block from appveyor since we are no longer debugging
* add ifdef to sftp-server.c
* make key_lookup compatible with winhello
* appveyor.yml
* increase debug of failing pester test
* add #ifdef SUPPORT_CRLF back into auth_check_principals_line method that was moved/renamed
* modify new scp.sh tests for windows
* remove in place tests from scp.sh
* remove rdp debug from appveyor
* retrigger appveyor
* change check of OS version in scp test
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
Co-authored-by: djm@openbsd.org <djm@openbsd.org>
Co-authored-by: Damien Miller <djm@mindrot.org>
Co-authored-by: Darren Tucker <dtucker@dtucker.net>
Co-authored-by: naddy@openbsd.org <naddy@openbsd.org>
Co-authored-by: dtucker@openbsd.org <dtucker@openbsd.org>
Co-authored-by: tj@openbsd.org <tj@openbsd.org>
Co-authored-by: millert@openbsd.org <millert@openbsd.org>
Co-authored-by: jmc@openbsd.org <jmc@openbsd.org>
Co-authored-by: florian@openbsd.org <florian@openbsd.org>
Co-authored-by: markus@openbsd.org <markus@openbsd.org>
Co-authored-by: Tobias Heider <me@tobhe.de>
Co-authored-by: anton@openbsd.org <anton@openbsd.org>
Co-authored-by: Tim Rice <tim@multitalents.net>
Co-authored-by: tobhe@openbsd.org <tobhe@openbsd.org>
Co-authored-by: Corinna Vinschen <vinschen@redhat.com>
Co-authored-by: Sam James <sam@gentoo.org>
Co-authored-by: jsg@openbsd.org <jsg@openbsd.org>
2022-11-02 17:06:45 +01:00
|
|
|
if (!retried && pin == NULL && !is_agent &&
|
|
|
|
sshkey_is_sk(sign_key) &&
|
2021-01-08 03:57:24 +01:00
|
|
|
r == SSH_ERR_KEY_WRONG_PASSPHRASE) {
|
|
|
|
notify_complete(notifier, NULL);
|
|
|
|
notifier = NULL;
|
Merge 9.1 (#626)
* upstream: fix poll() spin when a channel's output fd closes without
data in the channel buffer. Introduce more exact packing of channel fds into
the pollfd array. fixes bz3405 and bz3411; ok deraadt@ markus@
OpenBSD-Commit-ID: 06740737849c9047785622ad5d472cb6a3907d10
* upstream: select post-quantum KEX
sntrup761x25519-sha512@openssh.com as the default; ok markus@
OpenBSD-Commit-ID: f02d99cbfce22dffec2e2ab1b60905fbddf48fb9
* upstream: add support for the "corp-data" protocol extension to
allow server-side copies to be performed without having to go via the client.
Patch by Mike Frysinger, ok dtucker@
OpenBSD-Commit-ID: 00aa510940fedd66dab1843b58682de4eb7156d5
* upstream: add a sftp client "cp" command that supports server-side
copying of files. Useful for this task and for testing the copy-data
extension. Patch from Mike Frysinger; ok dtucker@
OpenBSD-Commit-ID: 1bb1b950af0d49f0d5425b1f267e197aa1b57444
* depend
* Skip slow tests on (very) slow test targets.
* Set Makefile SHELL as determined by configure.
This should improve compatibility for users with non-POSIX shells. If
using Makefile.in directly (eg make -f Makefile.in distprep) then SHELL
will need to be specified on the command line (along with MANFMT in that
particular case). ok djm@
* Use bash or ksh if available for SH in Makefile.
* Increase test timeout to allow slow VMs to finish
* Only run regression tests on slow VMs.
* Only return events from ppoll that were requested.
If the underlying system's select() returns bits that were not in the
request set, our ppoll() implementation can return revents for events
not requested, which can apparently cause a hang. Only return revents
for activity in the requested event set. bz#3416, analysis and fix by
yaroslav.kuzmin at vmssoftware com, ok djm@
* Specify TEST_SHELL=bash on AIX.
The system shells cause the agent-restrict test to fail due to some
quoting so explicitly specify bash until we can get configure to
autmatically work around that.
* Disable security key on fbsd6 test host.
* upstream: man pages: add missing commas between subordinate and
main clauses
jmc@ dislikes a comma before "then" in a conditional, so leave those
untouched.
ok jmc@
OpenBSD-Commit-ID: 9520801729bebcb3c9fe43ad7f9776ab4dd05ea3
* upstream: ssh: document sntrup761x25519-sha512@openssh.com as
default KEX
OpenBSD-Commit-ID: 12545bfa10bcbf552d04d9d9520d0f4e98b0e171
* upstream: openssh-9.0
OpenBSD-Commit-ID: 0dfb461188f4513ec024c1534da8c1ce14c20b64
* update version numbers for release
* update build-aux files to match autoconf-2.71
i.e. config.guess, config.sub and install-sh
* Revert "update build-aux files to match autoconf-2.71"
This reverts commit 0a8ca39fac6ad19096b6c263436f8b2dd51606f2.
It turns out that the checked-in copies of these files are actually newer
than autoconf-2.71's copies, so this was effectively a downgrade.
Spotted by Bo Anderson via github
* upstream: two defensive changes from Tobias Stoeckmann via GHPR287
enforce stricter invarient for sshbuf_set_parent() - never allow
a buffer to have a previously-set parent changed.
In sshbuf_reset(), if the reallocation fails, then zero the entire
buffer and not the (potentially smaller) default initial alloc size.
OpenBSD-Commit-ID: 14583203aa5d50ad38d2e209ae10abaf8955e6a9
* upstream: Note that curve25519-sha256 was later published in
RFC8731. ok djm@
OpenBSD-Commit-ID: 2ac2b5d642d4cf5918eaec8653cad9a4460b2743
* upstream: clear io_want/io_ready flags at start of poll() cycle;
avoids plausible spin during rekeying if channel io_want flags are reused
across cycles. ok markus@ deraadt@
OpenBSD-Commit-ID: 91034f855b7c73cd2591657c49ac30f10322b967
* Retire fbsd6 test VM.
It's long since out of support, relatively slow (it's i686) and the
compiler has trouble with PIE.
* Resync moduli.5 with upstream.
1.18: remove duplicate publication year; carsten dot kunze at arcor dot de
1.19: ssh-keygen's -G/-T have been replaced with -M generate/screen.
* upstream: Correct path for system known hosts file in description
of IgnoreUserKnownHosts. Patch from Martin Vahlensieck via tech@
OpenBSD-Commit-ID: 9b7784f054fa5aa4d63cb36bd563889477127215
* upstream: list the correct version number
for when usage of the sftp protocol became default and fix a typo
from ed maste
OpenBSD-Commit-ID: 24e1795ed2283fdeacf16413c2f07503bcdebb31
* upstream: Import regenerated moduli
OpenBSD-Commit-ID: f9a0726d957cf10692a231996a1f34e7f9cdfeb0
* upstream: Try to continue running local I/O for channels in state
OPEN during SSH transport rekeying. The most visible benefit is that it
should make ~-escapes work in the client (e.g. to exit) if the connection
happened to have stalled during a rekey event. Based work by and ok dtucker@
OpenBSD-Commit-ID: a66e8f254e92edd4ce09c9f750883ec8f1ea5f45
* upstream: Import regenerated moduli
OpenBSD-Commit-ID: f9a0726d957cf10692a231996a1f34e7f9cdfeb0
* upstream: regression test for sftp cp command
OpenBSD-Regress-ID: c96bea9edde3a384b254785e7f9b2b24a81cdf82
* upstream: Simplify forward-control test.
Since we no longer need to support SSH1 we don't need to run shell
commands on the other end of the connection and can use ssh -N instead.
This also makes the test less racy.
OpenBSD-Regress-ID: 32e94ce272820cc398f30b848b2b0f080d10302c
* upstream: Use ssh -f and ControlPersist ..
to start up test forwards and ssh -O stop to shut them down intead of
sleep loops. This speeds up the test by an order of magnitude.
OpenBSD-Regress-ID: eb3db5f805100919b092a3b2579c611fba3e83e7
* upstream: It looks like we can't completely avoid
waiting for processes to exit so retrieve the pid via controlmaster and
use that.
OpenBSD-Regress-ID: 8246f00f22b14e49d2ff1744c94897ead33d457b
* Cache timezone data in capsicum sandbox.
From emaste at freebsd.org, originally part of FreeBSD commit r339216
/ fc3c19a9 with autoconf bits added by me.
* Include stdlib.h for free() prototype.
... which is used inside the CUSTOM_SYS_AUTH_GET_LASTLOGIN_MSG block.
* Update OpenSSL and LibreSSL versions in tests.
* Add debian-riscv64 test target.
* upstream: Avoid an unnecessary xstrdup in rm_env() when matching
patterns. Since match_pattern() doesn't modify its arguments (they are
const), there is no need to make an extra copy of the strings in
options->send_env. From Martin Vahlensieck
OpenBSD-Commit-ID: 2c9db31e3f4d3403b49642c64ee048b2a0a39351
* upstream: Add missing includes of stdlib.h and stdint.h. We need
stdlib.h for malloc(3) and stdint.h for SIZE_MAX. Unlike the other xmss
files, ssh-xmss.c does not include xmss_commons.h so ssh-xmss.c must include
those headers itself. From Martin Vahlensieck
OpenBSD-Commit-ID: 70e28a9818cee3da1be2ef6503d4b396dd421e6b
* upstream: Remove unnecessary includes: openssl/hmac.h and
openssl/evp.h. From Martin Vahlensieck.
OpenBSD-Commit-ID: a6debb5fb0c8a44e43e8d5ca7cc70ad2f3ea31c3
* upstream: Check sshauthopt_new() for NULL. bz#3425, from
tessgauthier at microsoft.com. ok djm@
OpenBSD-Commit-ID: af0315bc3e44aa406daa7e0ae7c2d719a974483f
* upstream: Add authfd path to debug output. ok markus@
OpenBSD-Commit-ID: f735a17d1a6f2bee63bfc609d76ef8db8c090890
* upstream: avoid printing hash algorithm twice; from lucas AT sexy.is
OpenBSD-Commit-ID: 9d24671e10a84141b7c504396cabad600e47a941
* upstream: fix memleak on session-bind path; from Pedro Martelletto, ok
dtucker@
OpenBSD-Commit-ID: e85899a26ba402b4c0717b531317e8fc258f0a7e
* upstream: Don't leak SK device. Patch from Pedro Martelletto via
github PR#316. ok djm@
OpenBSD-Commit-ID: 17d11327545022e727d95fd08b213171c5a4585d
* upstream: mention that the helpers are used by ssh(1), ssh-agent(1)
and ssh-keygen(1). Previously only ssh(1) was mentioned. From Pedro
Martelletto
OpenBSD-Commit-ID: 30f880f989d4b329589c1c404315685960a5f153
* Remove now-empty int32_minmax.inc.
* Only run tests when source files change.
Also run tests on changes to V_9_0 branch.
* Add Mac OS X 12 test target.
* upstream: be stricter in which characters will be accepted in
specifying a mask length; allow only 0-9. From khaleesicodes via GHPR#278; ok
dtucker@
OpenBSD-Commit-ID: e267746c047ea86665cdeccef795a8a56082eeb2
* upstream: fix some integer overflows in sieve_large() that show up when
trying to generate modp groups > 16k bits. Reported via GHPR#306 by Bertram
Felgenhauer, but fixed in a different way. feedback/ok tb@
OpenBSD-Commit-ID: 81cbc6dd3a21c57bd6fadea10e44afe37bca558e
* upstream: remove an obsolete rsa1 format example from an example;
from megan batty
ok djm
OpenBSD-Commit-ID: db2c89879c29bf083df996bd830abfb1e70d62bf
* upstream: Add FIDO AUTHENTICATOR section and explain a bit how FIDO
works. The wording came mostly from the 8.2 OpenSSH release notes, addapted
to fit the man page. Then move the -O bits into the new section as is already
done for CERTIFICATES and MODULI GENERATION. Finally we can explain the
trade-offs of resident keys. While here, consistently refer to the FIDO
thingies as "FIDO authenticators", not "FIDO tokens".
input & OK jmc, naddy
OpenBSD-Commit-ID: dd98748d7644df048f78dcf793b3b63db9ab1d25
* upstream: make sure stdout is non-blocking; ok djm@
OpenBSD-Commit-ID: 64940fffbd1b882eda2d7c8c7a43c79368309c0d
* upstream: mux.c: mark argument as const; from Martin Vahlensieck
OpenBSD-Commit-ID: 69a1a93a55986c7c2ad9f733c093b46a47184341
* upstream: channel_new no longer frees remote_name. So update the
comment accordingly. As remote_name is not modified, it can be const as
well. From Martin Vahlensieck
OpenBSD-Commit-ID: e4e10dc8dc9f40c166ea5a8e991942bedc75a76a
* upstream: sshkey_unshield_private() contains a exact duplicate of
the code in private2_check_padding(). Pull private2_check_padding() up so the
code can be reused. From Martin Vahlensieck, ok deraadt@
OpenBSD-Commit-ID: 876884c3f0e62e8fd8d1594bab06900f971c9c85
* Add ubsan minimal testcase on OpenBSD.
As suggested by djm@.
* Note that, for now, we need variadic macros.
* Also retest OpenBSD upstream on .yml changes.
* upstream: When performing operations that glob(3) a remote path, ensure
that the implicit working directory used to construct that path escapes
glob(3) characters.
This prevents glob characters from being processed in places they
shouldn't, e.g. "cd /tmp/a*/", "get *.txt" should have the get operation
treat the path "/tmp/a*" literally and not attempt to expand it.
Reported by Lusia Kundel; ok markus@
OpenBSD-Commit-ID: 4f647f58482cbad3d58b1eab7f6a1691433deeef
* Remove duplicate bcrypt_pbkdf.o from Makefile
bcrypt_pbkdf.o is duplicated in the openbsd-compat Makefile's object
file list.
* upstream: improve error message when 'ssh-keygen -Y sign' is unable to
load a private key; bz3429, reported by Adam Szkoda ok dtucker@
OpenBSD-Commit-ID: bb57b285e67bea536ef81b1055467be2fc380e74
* upstream: Allow existing -U (use agent) flag to work with "-Y sign"
operations, where it will be interpreted to require that the private keys is
hosted in an agent; bz3429, suggested by Adam Szkoda; ok dtucker@
OpenBSD-Commit-ID: a7bc69873b99c32c42c7628ed9ea91565ba08c2f
* upstream: Remove errant apostrophe. From haruyama at queen-ml org.
OpenBSD-Commit-ID: dc6b294567cb84b384ad6ced9ca469f2bbf0bd10
* upstream: arrange for scp, when in sftp mode, to not ftruncate(3) files
early
previous behavious of unconditionally truncating the destination file
would cause "scp ~/foo localhost:" and "scp localhost:foo ~/" to
delete all the contents of their destination.
spotted by solene@ sthen@, also bz3431; ok dtucker@
OpenBSD-Commit-ID: ca39fdd39e0ec1466b9666f15cbcfddea6aaa179
* upstream: fix in-place copies; r1.163 incorrectly skipped truncation in
all cases, not just at the start of a transfer. This could cause overwrites
of larger files to leave junk at the end. Spotted by tb@
OpenBSD-Commit-ID: b189f19cd68119548c8e24e39c79f61e115bf92c
* upstream: Only run agent-ptrace.sh if gdb is available as all
architectures do not ship with gdb.
OpenBSD-Regress-ID: ec53e928803e6b87f9ac142d38888ca79a45348d
* upstream: regress test for in-place transfers and clobbering larger
files with smaller ones; would have caught last regression in scp(1)
OpenBSD-Regress-ID: 19de4e88dd3a4f7e5c1618c9be3c32415bd93bc2
* configure.ac: Add missing AC_DEFINE for caph_cache_tzdata test causing
HAVE_CAPH_CACHE_TZDATA to be missing from config.h.in.
Spotted by Bryan Drewery
* upstream: make SSHBUF_DBG/SSHBUF_TELL (off by default and only enabled
via #define) dump to stderr rather than stdout
OpenBSD-Commit-ID: 10298513ee32db8390aecb0397d782d68cb14318
* upstream: revert previous; it was broken (spotted by Theo)
OpenBSD-Commit-ID: 457c79afaca2f89ec2606405c1059b98b30d8b0d
* upstream: Note that ProxyJump also accepts the same tokens as
ProxyCommand. From pallxk via github PR#305.
OpenBSD-Commit-ID: 7115ac351b129205f1f1ffa6bbfd62abd76be7c5
* upstream: Avoid kill with -1 argument. The out_ctx label can be
reached before fork has been called. If this happens, then kill -1 would be
called, sending SIGTERM to all processes reachable by the current process.
From tobias@ and c3h2_ctf via github PR#286, ok djm@
OpenBSD-Commit-ID: 6277af1207d81202f5daffdccfeeaed4c763b1a8
* upstream: f sshpkt functions fail, then password is not cleared
with freezero. Unconditionally call freezero to guarantee that password is
removed from RAM.
From tobias@ and c3h2_ctf via github PR#286, ok djm@
OpenBSD-Commit-ID: 6b093619c9515328e25b0f8093779c52402c89cd
* upstream: refactor authorized_keys/principals handling
remove "struct ssh *" from arguments - this was only used to pass the
remote host/address. These can be passed in instead and the resulting
code is less tightly coupled to ssh_api.[ch]
ok dtucker@
OpenBSD-Commit-ID: 9d4373d013edc4cc4b5c21a599e1837ac31dda0d
* upstream: split the low-level file handling functions out from
auth2-pubkey.c
Put them in a new auth2-pubkeyfile.c to make it easier to refer to them
(e.g. in unit/fuzz tests) without having to refer to everything else
pubkey auth brings in.
ok dtucker@
OpenBSD-Commit-ID: 3fdca2c61ad97dc1b8d4a7346816f83dc4ce2217
* fuzzer for authorized_keys parsing
mostly redundant to authopt_fuzz, but it's sensitive code so IMO it
makes sense to test this layer too
* Test against LibreSSL 3.5.3.
* Test against OpenSSL 1.1.1o and 3.0.3.
* fix some bugs in the fuzzer
* upstream: keywords ref ssh_config.5;
from caspar schutijser
OpenBSD-Commit-ID: f146a19d7d5c9374c3b9c520da43b2732d7d1a4e
* upstream: ssh-keygen: implement "verify-required" certificate option.
This was already documented when support for user-verified FIDO
keys was added, but the ssh-keygen(1) code was missing.
ok djm@
OpenBSD-Commit-ID: f660f973391b593fea4b7b25913c9a15c3eb8a06
* upstream: ssh-keygen -A: do not generate DSA keys by default.
Based on github PR#303 from jsegitz with man page text from jmc@, ok markus@
djm@
OpenBSD-Commit-ID: 5c4c57bdd7063ff03381cfb6696659dd3f9f5b9f
* upstream: Add period at end of "not known by any other names"
message. github PR#320 from jschauma, ok djm@
OpenBSD-Commit-ID: bd60809803c4bfd3ebb7c5c4d918b10e275266f2
* upstream: Add missing *-sk types to ssh-keyscan manpage. From
skazi0 via github PR#294.
OpenBSD-Commit-ID: fda2c869cdb871f3c90a89fb3f985370bb5d25c0
* upstream: Make SetEnv directives first-match-wins in both
sshd_config and sshd_config; previously if the same name was reused then the
last would win (which is the opposite to how the config is supposed to work).
While there, make the ssh_config parsing more like sshd_config.
bz3438, ok dtucker
OpenBSD-Commit-ID: 797909c1e0262c0d00e09280459d7ab00f18273b
* upstream: test setenv in both client and server, test first-match-wins
too
OpenBSD-Regress-ID: 4c8804f9db38a02db480b9923317457b377fe34b
* upstream: move auth_openprincipals() and auth_openkeyfile() over to
auth2-pubkeyfile.c too; they make more sense there.
OpenBSD-Commit-ID: 9970d99f900e1117fdaab13e9e910a621b7c60ee
* upstream: make sure that UseDNS hostname lookup happens in the monitor
and not in the pledge(2)'d unprivileged process; fixes regression caused by
recent refactoring spotted by henning@
OpenBSD-Commit-ID: a089870b95101cd8881a2dff65b2f1627d13e88d
* fix possible NULL deref when built without FIDO
Analysis/fix from kircher in bz3443; ok dtucker@
* automatically enable built-in FIDO support
If libfido2 is found and usable, then enable the built-in
security key support unless --without-security-key-builtin
was requested.
ok dtucker@
* upstream: Log an error if pipe() fails while accepting a
connection. bz#3447, from vincent-openssh at vinc17 net, ok djm@
OpenBSD-Commit-ID: 9d59f19872b94900a5c79da2d57850241ac5df94
* upstream: Don't attempt to fprintf a null identity comment. From
Martin Vahlensieck via tech@.
OpenBSD-Commit-ID: 4c54d20a8e8e4e9912c38a7b4ef5bfc5ca2e05c2
* upstream: Make sure not to fclose() the same fd twice in case of an
error.
ok dtucker@
OpenBSD-Commit-ID: e384c4e05d5521e7866b3d53ca59acd2a86eef99
* upstream: make it clear that RekeyLimit applies to both transmitted
and received data. GHPR#328 from Jan Pazdziora
OpenBSD-Commit-ID: d180a905fec9ff418a75c07bb96ea41c9308c3f9
* request 1.1x API compatibility for OpenSSL >=3.x
idea/patch from Pedro Martelletto via GHPR#322; ok dtucker@
* fix broken case statement in previous
* Disable SK support if FIDO libs not found.
* Zero out LIBFIDO2 when SK support not usable.
Prevents us from trying to link them into ssh-sk-helper and failing to
build.
* upstream: Don't leak the strings allocated by order_hostkeyalgs()
and list_hostkey_types() that are passed to compat_pkalg_proposal(). Part of
github PR#324 from ZoltanFridrich, ok djm@
OpenBSD-Commit-ID: b2f6e5f60f2bba293b831654328a8a0035ef4a1b
* upstream: Roll back previous KEX changes as they aren't safe until
compat_pkalg_proposal and friends always allocate their returned strings.
Reported by Qualys.
OpenBSD-Commit-ID: 1c7a88a0d5033f42f88ab9bec58ef1cf72c81ad0
* upstream: allow arguments to sftp -D option, e.g. sftp -D
"/usr/libexec/sftp-server -el debug3"
ok markus@
OpenBSD-Commit-ID: 5a002b9f3a7aef2731fc0ffa9c921cf15f38ecce
* Update OpenSSL tests to the most recent releases.
* upstream: reflect the update to -D arg name in usage();
OpenBSD-Commit-ID: abdcde4f92b1ef094ae44210ee99d3b0155aad9c
* upstream: ignore SIGPIPE earlier in main(), specifically before
muxclient() which performs operations that could cause one; Reported by Noam
Lewis via bz3454, ok dtucker@
OpenBSD-Commit-ID: 63d8e13276869eebac6d7a05d5a96307f9026e47
* upstream: Always return allocated strings from the kex filtering so
that we can free them later. Fix one leak in compat_kex_proposal. Based on
github PR#324 from ZoltanFridrich with some simplications by me. ok djm@
OpenBSD-Commit-ID: 9171616da3307612d0ede086fd511142f91246e4
* upstream: Don't leak the strings allocated by order_hostkeyalgs()
and list_hostkey_types() that are passed to compat_pkalg_proposal(). Part of
github PR#324 from ZoltanFridrich, ok djm@
This is a roll-forward of the previous rollback now that the required
changes in compat.c have been done.
OpenBSD-Commit-ID: c7cd93730b3b9f53cdad3ae32462922834ef73eb
* upstream: bump up loglevel from debug to info when unable to open
authorized keys/principals file for errno != ENOENT; bz2042 ok dtucker
OpenBSD-Commit-ID: e79aa550d91ade6a80f081bda689da24c086d66b
* Skip select+rlimit check if sandboxing is disabled
It's not needed in that case, and the test can fail when being built
with some compiler memory sanitizer flags. bz#3441
* upstream: use consistent field names (s/char/byte)
in format description
OpenBSD-Commit-ID: 3de33572733ee7fcfd7db33d37db23d2280254f0
* upstream: Remove leftover line.
Remove extra line leftover from merge conflict. ok djm@
OpenBSD-Commit-ID: 460e2290875d7ae64971a7e669c244b1d1c0ae2e
* Move checks for pollfd.fd and nfds_t.
Move the checks for struct pollfd.fd and nfds_t to before the sandboxing
checks. This groups all the sandbox checks together so we can skip them
all when sandboxing is disabled.
* Skip all rlimit tests when sandboxing disabled.
The rlimit tests can hang when being run with some compiler sanitizers
so skip all of them if sandbox=no.
* Add clang sanitizer tests.
* upstream: Add TEST_REGRESS_CACHE_DIR.
If set, it is used to cache regress test names that have succeeded and
skip those on a re-run.
OpenBSD-Regress-ID: a7570dd29a58df59f2cca647c3c2ec989b49f247
* Move sanitizer logs into regress for collection.
* Add GCC address sanitizer build/test.
* Update sanitizer test targets:
- remove clang-sanitize-memory for now. It takes so long that the test
times out.
- add gcc sanitize-address and sanitize-undefined test targets.
* Test against openssl-3.0.5.
* Move unset to before we set anything.
* Refuse to use OpenSSL 3.0.4 due to potential RCE.
OpenSSL has a potential RCE in its RSA implementation (CVE-2022-2274)
so refuse to use that specific version.
* Capture stderr output from configure.
* Only refuse to use OpenSSL 3.0.4 on x86_64.
The potential RCE only impacts x86_64, so only refuse to use it if we're
targetting a potentially impacted architecture. ok djm@
* Remove special casing of crypt().
Configure goes to some lengths to pick crypt() from either libcrypt
or OpenSSL's libcrypto because they can more or less featureful (eg
supporting md5-style passwords).
OpenSSL removed its crypt() interface in 2002:
https://github.com/openssl/openssl/commit/69deec58 so these hijinks
should no longer be necessary. This also only links sshd with libcrypt
which is the only thing that needs it. ok djm@
* Clarify README.md text.
Clarify the text about the implications of building without OpenSSL, and
prefix the "configure --help" example command with a "./" so it's likely
to work as-is in more shells. From bz#3461.
* Split README.platform into its own line.
README.platform has general platform-specific information, having it
following text about FIDO2 on the same line could imply that it only
has information about FIDO2.
* Return ERANGE from getcwd() if buffer size is 1.
If getcwd() is supplied a buffer size of exactly 1 and a path of "/", it
could result in a nul byte being written out of array bounds. POSIX says
it should return ERANGE if the path will not fit in the available buffer
(with terminating nul). 1 byte cannot fit any possible path with its nul,
so immediately return ERANGE in that case.
OpenSSH never uses getcwd() with this buffer size, and all current
(and even quite old) platforms that we are currently known to work
on have a native getcwd() so this code is not used on those anyway.
Reported by Qualys, ok djm@
* Remove unintended changes.
I inadvertently included a couple of local changes with the OpenSSL
3.0.4 change. Revert, anything that should be there will be committed
separately.
* Add AUDIT_ARCH_PPC to supported seccomp arches.
Patch from dries.deschout at dodeco.eu.
* Rename bbone test target to ARM.
* Move vmshutdown to first step.
If a previous run on a physical runner has failed to clean up, the next
run will fail because it'll try to check out the code to a broken
directory mount. Make cleanup the first step.
* upstream: pull passphrase reading and confirmation into a separate
function so it can be used for FIDO2 PINs; no functional change
OpenBSD-Commit-ID: bf34f76b8283cc1d3f54633e0d4f13613d87bb2f
* upstream: when enrolling a resident key on a security token, check
if a credential with matching application and user ID strings already exists.
if so, prompt the user for confirmation before overwriting the credential.
patch from Pedro Martelletto via GHPR329
NB. cranks SSH_SK_VERSION_MAJOR, so any third-party FIDO middleware
implementations will need to adjust
OpenBSD-Commit-ID: e45e9f1bf2b2f32d9850669e7a8dbd64acc5fca4
* upstream: sk-usbhid: preserve error code returned by key_lookup()
it conveys useful information, such as the supplied pin being wrong.
Part of GHPR329 from Pedro Martelletto
OpenBSD-Commit-ID: c0647eb9290f793add363d81378439b273756c1b
* upstream: ssh-keygen: fix touch prompt, pin retries;
part of GHPR329 from Pedro Martelletto
OpenBSD-Commit-ID: 75d1005bd2ef8f29fa834c90d2684e73556fffe8
* crank SSH_SK_VERSION_MAJOR in sk-dummy.so
* Skip scp3 test if there's no scp on remote path.
scp -3 ends up using the scp that's in the remote path and will fail if
one is not available. Based on a patch from rapier at psc.edu.
* Convert "have_prog" function into "which".
"which" and its behaviour is not standardized, so convert the existing
have_prog function into "which" so we can rely on it being available
and what its semantics are. Add a have_prog wrapper that maintains the
existing behaviour.
* upstream: Test TEST_SSH_ELAPSED_TIMES for empty string not
executable. No-op on most platforms but should prevent warnings in -portable
on systems that don't have 'date %s'.
OpenBSD-Regress-ID: e39d79867b8065e33d0c5926fa1a31f85659d2a4
* upstream: Restore missing "!" in TEST_SSH_ELAPSED_TIMES test.
OpenBSD-Regress-ID: 38783f9676ec348c5a792caecee9a16e354b37b0
* Remove workarounds for OpenSSL missing AES-GCM.
We have some compatibility hacks that were added to support OpenSSL
versions that do not support AES GCM mode. Since that time, however,
the minimum OpenSSL version that we support has moved to 1.0.1 which
*does* have GCM, so this is no longer needed. ok djm@
* Remove workarounds for OpenSSL missing AES-CTR.
We have some compatibility hacks that were added to support OpenSSL
versions that do not support AES CTR mode. Since that time, however,
the minimum OpenSSL version that we support has moved to 1.0.1 which
*does* have CTR, so this is no longer needed. ok djm@
* Do not link scp, sftp and sftp-server w/ zlib.
Some of our binaries (eg sftp, sftp-server, scp) do not interact with
the channels code and thus do use libraries such as zlib and libcrypto
although they are linked with them. This adds a CHANNELLIBS and starts
by moving zlib into it, which means the aformentioned binaries are no
longer linked against zlib. ok djm@
* Group libcrypto and PRNGD checks together.
They're related more than the libcrypt or libiaf checks which are
currently between them. ok djm@
* Remove seed_rng calls from scp, sftp, sftp-server.
These binaries don't use OpenSSL's random functions. The next step
will be to stop linking them against libcrypto. ok djm@
* Move libcrypto into CHANNELLIBS.
This will result in sftp, sftp-server and scp no longer being linked
against libcrypto. ok djm@
* Move stale-configure check as early as possible.
We added a check in Makefile to catch the case where configure needs to
be rebuilt, however this did not happen until a build was attempted in
which case all of the work done by configure was wasted. Move this check
to the start of configure to catch it as early as possible. ok djm@
* Remove deprecated MacOS 10.15 runners.
* upstream: avoid double-free in error path introduced in r1.70; report
and fix based on GHPR#332 by v-rzh ok dtucker@
OpenBSD-Commit-ID: 3d21aa127b1f37cfc5bdc21461db369a663a951f
* Include CHANNEL and FIDO2 libs in configure output
* Factor out getrnd() and rename to getentropy().
Factor out the arc4random seeding into its own file and change the
interface to match getentropy. Use native getentropy if available.
This will make it easier to resync OpenBSD changes to arc4random.
Prompted by bz#3467, ok djm@.
* compat code for fido_dev_is_winhello()
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* check_sk_options: add temporary WinHello workaround
Up to libfido 1.10.0, WinHello advertises "clientPin" rather
than "uv" capability. This is fixed in 1.11.0. For the time
being, workaround it here.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* sk_sign: set FIDO2 uv attribute explicitely for WinHello
WinHello via libfido2 performs user verification by default.
However, if we stick to that, there's no way to differentiate
between keys created with or without "-O verify-required".
Set FIDO2 uv attribute explicitely to FIDO_OPT_FALSE, then check
if user verification has been requested.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* upstream: don't prompt for FIDO passphrase before attempting to enroll
the credential, just let the enroll operating fail and we'll attempt to get a
PIN anyway. Might avoid some unneccessary PIN prompts.
Part of GHPR#302 from Corinna Vinschen; ok dtucker@
OpenBSD-Commit-ID: bd5342ffc353ee37d39617906867c305564d1ce2
* Give unused param a name.
Fixes builds on platforms that do have fido2 but don't have
fido_dev_is_winhello.
* Actually put HAVE_STDINT_H around the stdint.h.
* Rename our getentropy to prevent possible loops.
Since arc4random seeds from getentropy, and we use OpenSSL for that
if enabled, there's the possibility that if we build on a system that
does not have getentropy then run on a system that does have it, then
OpenSSL could end up calling our getentropy and getting stuck in a loop.
Pointed out by deraadt@, ok djm@
* Test hostbased auth on github runners.
* fix SANDBOX_SECCOMP_FILTER_DEBUG
* Fix conditional for running hostbased tests.
* upstream: allow certificate validity intervals, sshsig verification
times and authorized_keys expiry-time options to accept dates in the UTC time
zone in addition to the default of interpreting them in the system time zone.
YYYYMMDD and YYMMDDHHMM[SS] dates/times will be interpreted as UTC if
suffixed with a 'Z' character.
Also allow certificate validity intervals to be specified in raw
seconds-since-epoch as hex value, e.g. -V 0x1234:0x4567890. This
is intended for use by regress tests and other tools that call
ssh-keygen as part of a CA workflow.
bz3468 ok dtucker
OpenBSD-Commit-ID: 454db1cdffa9fa346aea5211223a2ce0588dfe13
* upstream: add some tests for parse_absolute_time(), including cases
where it is forced to the UTC timezone. bz3468 ok dtucker
OpenBSD-Regress-ID: ea07ca31c2f3847a38df028ca632763ae44e8759
* Skip hostbased during Valgrind tests.
Valgrind doesn't let ssh exec ssh-keysign (because it's setuid) so skip
it during the Valgrind based tests.
See https://bugs.kde.org/show_bug.cgi?id=119404 for a discussion of this
(ironically there the problematic binary was ssh(1) back when it could
still be setuid).
* Rerun tests if any .github config file changes.
* Add a timegm implementation from Heimdal via Samba.
Fixes build on (at least Solaris 10).
* Replace deprecated ubuntu-18.04 runners with 22.04
* upstream: sftp-server: support home-directory request
Add support to the sftp-server for the home-directory extension defined
in draft-ietf-secsh-filexfer-extensions-00. This overlaps a bit with the
existing expand-path@openssh.com, but uses a more official protocol name,
and so is a bit more likely to be implemented by non-OpenSSH clients.
From Mike Frysinger, ok dtucker@
OpenBSD-Commit-ID: bfc580d05cc0c817831ae7ecbac4a481c23566ab
* fido_dev_is_winhello: return 0, not "false"
"false" is not used anywhere in OpenSSH, so return 0 like
everywhere else.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* Revert "check_sk_options: add temporary WinHello workaround"
Cygwin now comes with libfido2 1.11.0, so this workaround
isn't required anymore.
This reverts commit 242c044ab111a37aad3b0775727c36a4c5f0102c.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* upstream: use .Cm for "sign"; from josiah frentsos
OpenBSD-Commit-ID: 7f80a53d54857ac6ae49ea6ad93c5bd12231d1e4
* upstream: add an extra flag to sk_probe() to indicate whether we're
probing for a FIDO resident key or not. Unused here, but will make like
easier for portable
OpenBSD-Commit-ID: 432c8ff70e270378df9dbceb9bdeaa5b43b5a832
* on Cygwin, prefer WinHello FIDO device
If no FIDO device was explictly specified, then prefer the
windows://hello FIDO device. An exception to this is when
probing resident FIDO keys, in which case hardware FIDO
devices are preferred.
* Check for perms to run agent-getpeereid test.
Ubuntu 22.04 defaults to private home dirs which prevents "nobody"
running ssh-add during the agent-getpeereid test. Check for this and
add the necessary permissions.
* upstream: double free() in error path; from Eusgor via GHPR333
OpenBSD-Commit-ID: 39f35e16ba878c8d02b4d01d8826d9b321be26d4
* Add Cygwin (on windows-2019) test target.
In addition to installing the requisite Cygwin packages, we also need to
explicitly invoke "sh" for steps that run other scripts since the runner
environment doesn't understand #! paths.
* Add a bit more debug output.
* Fix cygwin conditional steps.
* upstream: Strictly enforce the maximum allowed SSH2 banner size in
ssh-keyscan and prevent a one-byte buffer overflow. Patch from Qualys, ok
djm@
OpenBSD-Commit-ID: 6ae664f9f4db6e8a0589425f74cd0bbf3aeef4e4
* upstream: remove incorrect check that can break enrolling a
resident key (introduced in r1.40)
OpenBSD-Commit-ID: 4cab364d518470e29e624af3d3f9ffa9c92b6f01
* upstream: attemp FIDO key signing without PIN and use the error
code returned to fall back only if necessary. Avoids PIN prompts for FIDO
tokens that don't require them; part of GHPR#302
OpenBSD-Commit-ID: 4f752aaf9f2e7c28bcaaf3d4f8fc290131bd038e
* Install Cygwin packages based on OS not config.
* initial list of allowed signers
* upstream: whitespace
OpenBSD-Commit-ID: d297e4387935d4aef091c5e9432578c2e513f538
* upstream: whitespace
OpenBSD-Commit-ID: a5d015efbfd228dc598ffdef612d2da3a579e5d8
* Add cygwin-release test target.
This also moves the cygwin package install from the workflow file to
setup_ci.sh so that we can install different sets of Cygwin packages
for different test configs.
* Add Windows 2022 test targets.
* Add libcrypt-devel to cygwin-release deps.
Based on feedback from vinschen at redhat.com.
* cross-sign allowed_signers with PGP key
Provides continuity of trust from legacy PGP release key to
the SSHSIG signing keys that we will use henceforth for git
signing.
* additional keys
* upstream: whitespace
OpenBSD-Commit-ID: c2bcbf93610d3d62ed206cdf9bf9ff98c6aaf232
* Move sftp from valgrind-2 to 3 to rebalance.
* upstream: sk-usbhid: fix key_lookup() on tokens with built-in UV
explicitly test whether the token performs built-in UV (e.g. biometric
tokens) and enable UV in that case. From Pedro Martelletto via GHPR#388
OpenBSD-Commit-ID: 007eb7e387d27cf3029ab06b88224e03eca62ccd
* Remove arc4random_uniform from arc4random.c
This was previously moved into its own file (matching OpenBSD) which
prematurely committed in commit 73541f2.
* Move OPENBSD ORIGINAL marker.
Putting this after the copyright statement (which doesn't change)
instead of before the version identifier (which does) prevents merge
conflicts when resyncing changes.
* Resync arc4random with OpenBSD.
This brings us up to current, including djm's random-reseeding change,
as prompted by logan at cyberstorm.mu in bz#3467. It brings the
platform-specific hooks from LibreSSL Portable, simplified to match our
use case. ok djm@.
* Remove DEF_WEAK, it's already in defines.h.
* openbsd-compat/bsd-asprintf: add <stdio.h> include for vsnprintf
Fixes the following build failure with Clang 15 on musl:
```
bsd-asprintf.c:51:8: error: call to undeclared library function 'vsnprintf' with type 'int (char *, unsigned long, const char *, struct __va_list_tag *)'; ISO C99 and laterclang -O2 -pipe -fdiagnostics-color=always -frecord-gcc-switches -pipe -Wunknown-warning-option -Qunused-arguments -Wall -Wpointer-arith -Wuninitialized -Wsign-compare -Wformat-security -Wsizeof-pointer-memaccess -Wno-pointer-sign -Wno-unused-result -Wmisleading-indentation -Wbitwise-instead-of-logical -fno-strict-aliasing -mretpoline -ftrapv -fzero-call-used-regs=all -fno-builtin-memset -fstack-protector-strong -fPIE -I. -I. -D_XOPEN_SOURCE=600 -D_BSD_SOURCE -D_DEFAULT_SOURCE -DSSHDIR=\"/etc/ssh\" -D_PATH_SSH_PROGRAM=\"/usr/bin/ssh\" -D_PATH_SSH_ASKPASS_DEFAULT=\"/usr/lib/misc/ssh-askpass\" -D_PATH_SFTP_SERVER=\"/usr/lib/misc/sftp-server\" -D_PATH_SSH_KEY_SIGN=\"/usr/lib/misc/ssh-keysign\" -D_PATH_SSH_PKCS11_HELPER=\"/usr/lib/misc/ssh-pkcs11-helper\" -D_PATH_SSH_SK_HELPER=\"/usr/lib/misc/ssh-sk-helper\" -D_PATH_SSH_PIDDIR=\"/run\" -D_PATH_PRIVSEP_CHROOT_DIR=\"/var/empty\" -DHAVE_CONFIG_H -c cipher-aes.c -o cipher-aes.o
do not support
implicit function declarations [-Wimplicit-function-declaration]
ret = vsnprintf(string, INIT_SZ, fmt, ap2);
^
bsd-asprintf.c:51:8: note: include the header <stdio.h> or explicitly provide a declaration for 'vsnprintf'
1 error generated.
```
* upstream: notifier_complete(NULL, ...) is a noop, so no need to test
that ctx!=NULL; from Corinna Vinschen
OpenBSD-Commit-ID: ade2f2e9cc519d01a586800c25621d910bce384a
* fix pester test failures
* upstream: fix repeated words ok miod@ jmc@
OpenBSD-Commit-ID: 6765daefe26a6b648cc15cadbbe337596af709b7
* upstream: .Li -> .Vt where appropriate; from josiah frentsos,
tweaked by schwarze
ok schwarze
OpenBSD-Commit-ID: 565046e3ce68b46c2f440a93d67c2a92726de8ed
* upstream: ssh-agent: attempt FIDO key signing without PIN and use
the error to determine whether a PIN is required and prompt only if
necessary. from Corinna Vinschen
OpenBSD-Commit-ID: dd6be6a0b7148608e834ee737c3479b3270b00dd
* upstream: a little extra debugging
OpenBSD-Commit-ID: edf1601c1d0905f6da4c713f4d9cecc7d1c0295a
* upstream: sk_enroll: never drop SSH_SK_USER_VERIFICATION_REQD flag
from response
Now that all FIDO signing calls attempt first without PIN and then
fall back to trying PIN only if that attempt fails, we can remove the
hack^wtrick that removed the UV flag from the keys returned during
enroll.
By Corinna Vinschen
OpenBSD-Commit-ID: 684517608c8491503bf80cd175425f0178d91d7f
* define HAVE_KILLPG
* upstream: sftp: Don't attempt to complete arguments for
non-existent commands
If user entered a non-existent command (e.g. because they made a
typo) there is no point in trying to complete its arguments. Skip
calling complete_match() if that's the case.
From Michal Privoznik
OpenBSD-Commit-ID: cf39c811a68cde2aeb98fc85addea4000ef6b07a
* upstream: sftp: Be a bit more clever about completions
There are commands (e.g. "get" or "put") that accept two
arguments, a local path and a remote path. However, the way
current completion is written doesn't take this distinction into
account and always completes remote or local paths.
By expanding CMD struct and "cmds" array this distinction can be
reflected and with small adjustment to completer code the correct
path can be completed.
By Michal Privoznik, ok dtucker@
OpenBSD-Commit-ID: 1396d921c4eb1befd531f5c4a8ab47e7a74b610b
* upstream: correct error value
OpenBSD-Commit-ID: 780efcbad76281f11f14b2a5ff04eb6db3dfdad4
* upstream: actually hook up restrict_websafe; the command-line flag
was never actually used. Spotted by Matthew Garrett
OpenBSD-Commit-ID: 0b363518ac4c2819dbaa3dfad4028633ab9cdff1
* upstream: Add a sshkey_check_rsa_length() call for checking the
length of an RSA key; ok markus@
OpenBSD-Commit-ID: de77cd5b11594297eda82edc594b0d32b8535134
* upstream: add a RequiredRSASize for checking RSA key length in
ssh(1). User authentication keys that fall beneath this limit will be
ignored. If a host presents a host key beneath this limit then the connection
will be terminated (unfortunately there are no fallbacks in the protocol for
host authentication).
feedback deraadt, Dmitry Belyavskiy; ok markus@
OpenBSD-Commit-ID: 430e339b2a79fa9ecc63f2837b06fdd88a7da13a
* upstream: Add RequiredRSASize for sshd(8); RSA keys that fall
beneath this limit will be ignored for user and host-based authentication.
Feedback deraadt@ ok markus@
OpenBSD-Commit-ID: 187931dfc19d51873df5930a04f2d972adf1f7f1
* upstream: better debugging for connect_next()
OpenBSD-Commit-ID: d16a307a0711499c971807f324484ed3a6036640
* upstream: sftp-server(8): add a "users-groups-by-id@openssh.com"
extension request that allows the client to obtain user/group names that
correspond to a set of uids/gids.
Will be used to make directory listings more useful and consistent
in sftp(1).
ok markus@
OpenBSD-Commit-ID: 7ebabde0bcb95ef949c4840fe89e697e30df47d3
* upstream: extend sftp-common.c:extend ls_file() to support supplied
user/group names; ok markus@
OpenBSD-Commit-ID: c70c70498b1fdcf158531117e405b6245863bfb0
* upstream: sftp client library support for
users-groups-by-id@openssh.com; ok markus@
OpenBSD-Commit-ID: ddb2f33a2da6349a9a89a8b5bcb9ca7c999394de
* upstream: use users-groups-by-id@openssh.com sftp-server extension
(when available) to fill in user/group names for directory listings.
Implement a client-side cache of see uid/gid=>user/group names. ok markus@
OpenBSD-Commit-ID: f239aeeadfa925a37ceee36ee8b256b8ccf4466e
* avoid Wuninitialized false positive in gcc-12ish
* no need for glob.h here
it also causes portability problems
* add debug on appveyor
* add sleep to pester test
* upstream: add RequiredRSASize to the list of keywords accepted by
-o; spotted by jmc@
OpenBSD-Commit-ID: fe871408cf6f9d3699afeda876f8adbac86a035e
* upstream: Fix typo. From AlexanderStohr via github PR#343.
OpenBSD-Commit-ID: a134c9b4039e48803fc6a87f955b0f4a03181497
* upstream: openssh-9.1
OpenBSD-Commit-ID: 5a467b2ee81da01a86adf1ad93b62b1728494e56
* crank versions in RPM spec files
* update release notes URL
* update .depend
* fix 9.1 compilation errors
* disable -p pester tests due to unreliability on older Windows versions
* remove extra sleep time from debugging scp pester tests
* modify -p tests to only run for Windows OS version 10 and above
* add windows specific code back into method moved from auth.c to auth2-pubkeyfile.c
* add preprocessor for WinHello
* revert preprocessor definition for winhello
* add windows preprocessor definition in key_lookup
* remove rdp block from appveyor since we are no longer debugging
* add ifdef to sftp-server.c
* make key_lookup compatible with winhello
* appveyor.yml
* increase debug of failing pester test
* add #ifdef SUPPORT_CRLF back into auth_check_principals_line method that was moved/renamed
* modify new scp.sh tests for windows
* remove in place tests from scp.sh
* remove rdp debug from appveyor
* retrigger appveyor
* change check of OS version in scp test
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
Co-authored-by: djm@openbsd.org <djm@openbsd.org>
Co-authored-by: Damien Miller <djm@mindrot.org>
Co-authored-by: Darren Tucker <dtucker@dtucker.net>
Co-authored-by: naddy@openbsd.org <naddy@openbsd.org>
Co-authored-by: dtucker@openbsd.org <dtucker@openbsd.org>
Co-authored-by: tj@openbsd.org <tj@openbsd.org>
Co-authored-by: millert@openbsd.org <millert@openbsd.org>
Co-authored-by: jmc@openbsd.org <jmc@openbsd.org>
Co-authored-by: florian@openbsd.org <florian@openbsd.org>
Co-authored-by: markus@openbsd.org <markus@openbsd.org>
Co-authored-by: Tobias Heider <me@tobhe.de>
Co-authored-by: anton@openbsd.org <anton@openbsd.org>
Co-authored-by: Tim Rice <tim@multitalents.net>
Co-authored-by: tobhe@openbsd.org <tobhe@openbsd.org>
Co-authored-by: Corinna Vinschen <vinschen@redhat.com>
Co-authored-by: Sam James <sam@gentoo.org>
Co-authored-by: jsg@openbsd.org <jsg@openbsd.org>
2022-11-02 17:06:45 +01:00
|
|
|
xasprintf(&prompt, "Enter PIN for %s key %s: ",
|
|
|
|
sshkey_type(sign_key), id->filename);
|
|
|
|
pin = read_passphrase(prompt, 0);
|
2021-01-08 03:57:24 +01:00
|
|
|
retried = 1;
|
|
|
|
goto retry_pin;
|
|
|
|
}
|
2019-10-31 22:18:28 +01:00
|
|
|
goto out;
|
|
|
|
}
|
2021-01-08 03:57:24 +01:00
|
|
|
|
2019-10-31 22:18:28 +01:00
|
|
|
/*
|
|
|
|
* PKCS#11 tokens may not support all signature algorithms,
|
|
|
|
* so check what we get back.
|
|
|
|
*/
|
|
|
|
if ((r = sshkey_check_sigtype(*sigp, *lenp, alg)) != 0) {
|
2020-10-18 13:32:01 +02:00
|
|
|
debug_fr(r, "sshkey_check_sigtype");
|
2019-10-31 22:18:28 +01:00
|
|
|
goto out;
|
2017-08-11 06:47:12 +02:00
|
|
|
}
|
2019-10-31 22:18:28 +01:00
|
|
|
/* success */
|
|
|
|
r = 0;
|
|
|
|
out:
|
2020-08-27 03:06:18 +02:00
|
|
|
free(prompt);
|
|
|
|
if (pin != NULL)
|
|
|
|
freezero(pin, strlen(pin));
|
2021-01-08 03:57:24 +01:00
|
|
|
notify_complete(notifier, r == 0 ? "User presence confirmed" : NULL);
|
2015-01-14 21:05:27 +01:00
|
|
|
sshkey_free(prv);
|
2018-03-03 04:01:50 +01:00
|
|
|
return r;
|
2001-03-09 01:12:22 +01:00
|
|
|
}
|
|
|
|
|
2017-03-12 00:40:26 +01:00
|
|
|
static int
|
|
|
|
id_filename_matches(Identity *id, Identity *private_id)
|
|
|
|
{
|
2022-02-02 00:32:51 +01:00
|
|
|
static const char * const suffixes[] = { ".pub", "-cert.pub", NULL };
|
2017-03-12 00:40:26 +01:00
|
|
|
size_t len = strlen(id->filename), plen = strlen(private_id->filename);
|
|
|
|
size_t i, slen;
|
|
|
|
|
|
|
|
if (strcmp(id->filename, private_id->filename) == 0)
|
|
|
|
return 1;
|
|
|
|
for (i = 0; suffixes[i]; i++) {
|
|
|
|
slen = strlen(suffixes[i]);
|
|
|
|
if (len > slen && plen == len - slen &&
|
|
|
|
strcmp(id->filename + (len - slen), suffixes[i]) == 0 &&
|
|
|
|
memcmp(id->filename, private_id->filename, plen) == 0)
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2001-06-25 07:01:22 +02:00
|
|
|
static int
|
2019-01-19 22:40:21 +01:00
|
|
|
sign_and_send_pubkey(struct ssh *ssh, Identity *id)
|
2000-04-29 15:57:08 +02:00
|
|
|
{
|
2019-01-19 22:40:21 +01:00
|
|
|
Authctxt *authctxt = (Authctxt *)ssh->authctxt;
|
2018-07-03 13:39:54 +02:00
|
|
|
struct sshbuf *b = NULL;
|
|
|
|
Identity *private_id, *sign_id = NULL;
|
|
|
|
u_char *signature = NULL;
|
|
|
|
size_t slen = 0, skip = 0;
|
|
|
|
int r, fallback_sigtype, sent = 0;
|
|
|
|
char *alg = NULL, *fp = NULL;
|
2021-12-19 23:12:54 +01:00
|
|
|
const char *loc = "", *method = "publickey";
|
2021-12-19 23:14:47 +01:00
|
|
|
int hostbound = 0;
|
2021-12-19 23:12:54 +01:00
|
|
|
|
|
|
|
/* prefer host-bound pubkey signatures if supported by server */
|
2021-12-19 23:14:47 +01:00
|
|
|
if ((ssh->kex->flags & KEX_HAS_PUBKEY_HOSTBOUND) != 0 &&
|
|
|
|
(options.pubkey_authentication & SSH_PUBKEY_AUTH_HBOUND) != 0) {
|
|
|
|
hostbound = 1;
|
2021-12-19 23:12:54 +01:00
|
|
|
method = "publickey-hostbound-v00@openssh.com";
|
2021-12-19 23:14:47 +01:00
|
|
|
}
|
2000-04-29 15:57:08 +02:00
|
|
|
|
2015-01-28 23:36:00 +01:00
|
|
|
if ((fp = sshkey_fingerprint(id->key, options.fingerprint_hash,
|
|
|
|
SSH_FP_DEFAULT)) == NULL)
|
|
|
|
return 0;
|
2001-03-09 01:12:22 +01:00
|
|
|
|
2021-12-19 23:14:47 +01:00
|
|
|
debug3_f("using %s with %s %s", method, sshkey_type(id->key), fp);
|
2000-04-29 15:57:08 +02:00
|
|
|
|
2015-09-24 08:15:11 +02:00
|
|
|
/*
|
|
|
|
* If the key is an certificate, try to find a matching private key
|
|
|
|
* and use it to complete the signature.
|
2016-04-28 16:30:21 +02:00
|
|
|
* If no such private key exists, fall back to trying the certificate
|
|
|
|
* key itself in case it has a private half already loaded.
|
2018-07-03 13:39:54 +02:00
|
|
|
* This will try to set sign_id to the private key that will perform
|
|
|
|
* the signature.
|
2015-09-24 08:15:11 +02:00
|
|
|
*/
|
2018-07-03 13:39:54 +02:00
|
|
|
if (sshkey_is_cert(id->key)) {
|
2015-09-24 08:15:11 +02:00
|
|
|
TAILQ_FOREACH(private_id, &authctxt->keys, next) {
|
|
|
|
if (sshkey_equal_public(id->key, private_id->key) &&
|
|
|
|
id->key->type != private_id->key->type) {
|
2018-07-03 13:39:54 +02:00
|
|
|
sign_id = private_id;
|
2015-09-24 08:15:11 +02:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
2017-03-12 00:40:26 +01:00
|
|
|
/*
|
|
|
|
* Exact key matches are preferred, but also allow
|
|
|
|
* filename matches for non-PKCS#11/agent keys that
|
|
|
|
* didn't load public keys. This supports the case
|
|
|
|
* of keeping just a private key file and public
|
|
|
|
* certificate on disk.
|
|
|
|
*/
|
2018-07-03 13:39:54 +02:00
|
|
|
if (sign_id == NULL &&
|
|
|
|
!id->isprivate && id->agent_fd == -1 &&
|
2017-03-12 00:40:26 +01:00
|
|
|
(id->key->flags & SSHKEY_FLAG_EXT) == 0) {
|
|
|
|
TAILQ_FOREACH(private_id, &authctxt->keys, next) {
|
|
|
|
if (private_id->key == NULL &&
|
|
|
|
id_filename_matches(id, private_id)) {
|
2018-07-03 13:39:54 +02:00
|
|
|
sign_id = private_id;
|
2017-03-12 00:40:26 +01:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
2018-07-03 13:39:54 +02:00
|
|
|
if (sign_id != NULL) {
|
2020-10-18 13:32:01 +02:00
|
|
|
debug2_f("using private key \"%s\"%s for "
|
2021-06-07 05:38:38 +02:00
|
|
|
"certificate", sign_id->filename,
|
|
|
|
sign_id->agent_fd != -1 ? " from agent" : "");
|
2015-09-24 08:15:11 +02:00
|
|
|
} else {
|
2020-10-18 13:32:01 +02:00
|
|
|
debug_f("no separate private key for certificate "
|
|
|
|
"\"%s\"", id->filename);
|
2015-09-24 08:15:11 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-07-03 13:39:54 +02:00
|
|
|
/*
|
|
|
|
* If the above didn't select another identity to do the signing
|
|
|
|
* then default to the one we started with.
|
|
|
|
*/
|
|
|
|
if (sign_id == NULL)
|
|
|
|
sign_id = id;
|
|
|
|
|
|
|
|
/* assemble and sign data */
|
|
|
|
for (fallback_sigtype = 0; fallback_sigtype <= 1; fallback_sigtype++) {
|
|
|
|
free(alg);
|
|
|
|
slen = 0;
|
|
|
|
signature = NULL;
|
|
|
|
if ((alg = key_sig_algorithm(fallback_sigtype ? NULL : ssh,
|
|
|
|
id->key)) == NULL) {
|
2021-10-11 23:46:32 +02:00
|
|
|
#ifdef WINDOWS
|
|
|
|
send_pubkey_sign_telemetry("no mutual signature supported");
|
|
|
|
#endif
|
2020-10-18 13:32:01 +02:00
|
|
|
error_f("no mutual signature supported");
|
2018-07-03 13:39:54 +02:00
|
|
|
goto out;
|
|
|
|
}
|
2020-10-18 13:32:01 +02:00
|
|
|
debug3_f("signing using %s %s", alg, fp);
|
2018-07-03 13:39:54 +02:00
|
|
|
|
|
|
|
sshbuf_free(b);
|
|
|
|
if ((b = sshbuf_new()) == NULL)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_f("sshbuf_new failed");
|
2021-01-27 10:26:53 +01:00
|
|
|
if (ssh->compat & SSH_OLD_SESSIONID) {
|
2021-01-27 11:05:28 +01:00
|
|
|
if ((r = sshbuf_putb(b, ssh->kex->session_id)) != 0)
|
|
|
|
fatal_fr(r, "sshbuf_putb");
|
2018-07-03 13:39:54 +02:00
|
|
|
} else {
|
2021-01-27 11:05:28 +01:00
|
|
|
if ((r = sshbuf_put_stringb(b,
|
|
|
|
ssh->kex->session_id)) != 0)
|
|
|
|
fatal_fr(r, "sshbuf_put_stringb");
|
2018-07-03 13:39:54 +02:00
|
|
|
}
|
2018-07-09 23:03:30 +02:00
|
|
|
skip = sshbuf_len(b);
|
2018-07-03 13:39:54 +02:00
|
|
|
if ((r = sshbuf_put_u8(b, SSH2_MSG_USERAUTH_REQUEST)) != 0 ||
|
|
|
|
(r = sshbuf_put_cstring(b, authctxt->server_user)) != 0 ||
|
|
|
|
(r = sshbuf_put_cstring(b, authctxt->service)) != 0 ||
|
2021-12-19 23:12:54 +01:00
|
|
|
(r = sshbuf_put_cstring(b, method)) != 0 ||
|
2018-07-03 13:39:54 +02:00
|
|
|
(r = sshbuf_put_u8(b, 1)) != 0 ||
|
|
|
|
(r = sshbuf_put_cstring(b, alg)) != 0 ||
|
|
|
|
(r = sshkey_puts(id->key, b)) != 0) {
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_fr(r, "assemble signed data");
|
2018-07-03 13:39:54 +02:00
|
|
|
}
|
2021-12-19 23:14:47 +01:00
|
|
|
if (hostbound) {
|
2021-12-19 23:12:54 +01:00
|
|
|
if (ssh->kex->initial_hostkey == NULL) {
|
|
|
|
fatal_f("internal error: initial hostkey "
|
|
|
|
"not recorded");
|
|
|
|
}
|
|
|
|
if ((r = sshkey_puts(ssh->kex->initial_hostkey, b)) != 0)
|
|
|
|
fatal_fr(r, "assemble %s hostkey", method);
|
|
|
|
}
|
2018-07-03 13:39:54 +02:00
|
|
|
/* generate signature */
|
|
|
|
r = identity_sign(sign_id, &signature, &slen,
|
2021-01-27 10:26:53 +01:00
|
|
|
sshbuf_ptr(b), sshbuf_len(b), ssh->compat, alg);
|
2018-07-03 13:39:54 +02:00
|
|
|
if (r == 0)
|
|
|
|
break;
|
|
|
|
else if (r == SSH_ERR_KEY_NOT_FOUND)
|
|
|
|
goto out; /* soft failure */
|
|
|
|
else if (r == SSH_ERR_SIGN_ALG_UNSUPPORTED &&
|
|
|
|
!fallback_sigtype) {
|
|
|
|
if (sign_id->agent_fd != -1)
|
|
|
|
loc = "agent ";
|
|
|
|
else if ((sign_id->key->flags & SSHKEY_FLAG_EXT) != 0)
|
|
|
|
loc = "token ";
|
|
|
|
logit("%skey %s %s returned incorrect signature type",
|
|
|
|
loc, sshkey_type(id->key), fp);
|
|
|
|
continue;
|
|
|
|
}
|
2021-10-11 23:46:32 +02:00
|
|
|
#ifdef WINDOWS
|
|
|
|
send_pubkey_sign_telemetry("signing failed");
|
|
|
|
#endif
|
2020-10-18 13:32:01 +02:00
|
|
|
error_fr(r, "signing failed for %s \"%s\"%s",
|
2020-01-21 06:56:27 +01:00
|
|
|
sshkey_type(sign_id->key), sign_id->filename,
|
2020-10-18 13:32:01 +02:00
|
|
|
id->agent_fd != -1 ? " from agent" : "");
|
2018-07-03 13:39:54 +02:00
|
|
|
goto out;
|
2000-08-23 02:46:23 +02:00
|
|
|
}
|
2018-07-03 13:39:54 +02:00
|
|
|
if (slen == 0 || signature == NULL) /* shouldn't happen */
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_f("no signature");
|
2001-03-09 01:12:22 +01:00
|
|
|
|
2000-04-29 15:57:08 +02:00
|
|
|
/* append signature */
|
2018-07-03 13:39:54 +02:00
|
|
|
if ((r = sshbuf_put_string(b, signature, slen)) != 0)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_fr(r, "append signature");
|
2000-04-29 15:57:08 +02:00
|
|
|
|
2018-07-03 13:39:54 +02:00
|
|
|
#ifdef DEBUG_PK
|
|
|
|
sshbuf_dump(b, stderr);
|
|
|
|
#endif
|
2000-04-29 15:57:08 +02:00
|
|
|
/* skip session id and packet type */
|
2018-07-03 13:39:54 +02:00
|
|
|
if ((r = sshbuf_consume(b, skip + 1)) != 0)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_fr(r, "consume");
|
2000-04-29 15:57:08 +02:00
|
|
|
|
|
|
|
/* put remaining data from buffer into packet */
|
2018-07-03 13:39:54 +02:00
|
|
|
if ((r = sshpkt_start(ssh, SSH2_MSG_USERAUTH_REQUEST)) != 0 ||
|
|
|
|
(r = sshpkt_putb(ssh, b)) != 0 ||
|
|
|
|
(r = sshpkt_send(ssh)) != 0)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_fr(r, "enqueue request");
|
2000-08-23 02:46:23 +02:00
|
|
|
|
2018-07-03 13:39:54 +02:00
|
|
|
/* success */
|
|
|
|
sent = 1;
|
2021-10-11 23:46:32 +02:00
|
|
|
#ifdef WINDOWS
|
|
|
|
send_pubkey_sign_telemetry("success");
|
|
|
|
#endif
|
2018-07-03 13:39:54 +02:00
|
|
|
|
|
|
|
out:
|
|
|
|
free(fp);
|
|
|
|
free(alg);
|
|
|
|
sshbuf_free(b);
|
|
|
|
freezero(signature, slen);
|
|
|
|
return sent;
|
2000-07-21 02:19:44 +02:00
|
|
|
}
|
|
|
|
|
2001-06-25 07:01:22 +02:00
|
|
|
static int
|
2019-01-19 22:40:21 +01:00
|
|
|
send_pubkey_test(struct ssh *ssh, Identity *id)
|
2000-07-21 02:19:44 +02:00
|
|
|
{
|
2019-01-19 22:40:21 +01:00
|
|
|
Authctxt *authctxt = (Authctxt *)ssh->authctxt;
|
2018-07-03 13:39:54 +02:00
|
|
|
u_char *blob = NULL;
|
|
|
|
char *alg = NULL;
|
2018-07-09 23:03:30 +02:00
|
|
|
size_t bloblen;
|
|
|
|
u_int have_sig = 0;
|
|
|
|
int sent = 0, r;
|
2000-07-21 02:19:44 +02:00
|
|
|
|
2018-07-03 13:39:54 +02:00
|
|
|
if ((alg = key_sig_algorithm(ssh, id->key)) == NULL) {
|
2020-10-18 13:32:01 +02:00
|
|
|
debug_f("no mutual signature algorithm");
|
2018-07-03 13:39:54 +02:00
|
|
|
goto out;
|
|
|
|
}
|
2001-03-09 01:12:22 +01:00
|
|
|
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshkey_to_blob(id->key, &blob, &bloblen)) != 0) {
|
2001-03-09 01:12:22 +01:00
|
|
|
/* we cannot handle this key */
|
2020-10-18 13:32:01 +02:00
|
|
|
debug3_f("cannot handle key");
|
2018-07-03 13:39:54 +02:00
|
|
|
goto out;
|
2000-07-21 02:19:44 +02:00
|
|
|
}
|
2001-03-09 01:12:22 +01:00
|
|
|
/* register callback for USERAUTH_PK_OK message */
|
2018-07-09 23:03:30 +02:00
|
|
|
ssh_dispatch_set(ssh, SSH2_MSG_USERAUTH_PK_OK, &input_userauth_pk_ok);
|
|
|
|
|
|
|
|
if ((r = sshpkt_start(ssh, SSH2_MSG_USERAUTH_REQUEST)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->server_user)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->service)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->method->name)) != 0 ||
|
|
|
|
(r = sshpkt_put_u8(ssh, have_sig)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, alg)) != 0 ||
|
|
|
|
(r = sshpkt_put_string(ssh, blob, bloblen)) != 0 ||
|
|
|
|
(r = sshpkt_send(ssh)) != 0)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_fr(r, "send packet");
|
2018-07-03 13:39:54 +02:00
|
|
|
sent = 1;
|
2018-07-09 23:03:30 +02:00
|
|
|
|
|
|
|
out:
|
2018-07-03 13:39:54 +02:00
|
|
|
free(alg);
|
|
|
|
free(blob);
|
|
|
|
return sent;
|
2001-03-09 01:12:22 +01:00
|
|
|
}
|
|
|
|
|
2017-05-30 10:52:19 +02:00
|
|
|
static struct sshkey *
|
2015-11-15 23:26:49 +01:00
|
|
|
load_identity_file(Identity *id)
|
2001-03-09 01:12:22 +01:00
|
|
|
{
|
2017-05-30 10:52:19 +02:00
|
|
|
struct sshkey *private = NULL;
|
2015-11-15 23:26:49 +01:00
|
|
|
char prompt[300], *passphrase, *comment;
|
2019-08-05 13:50:33 +02:00
|
|
|
int r, quit = 0, i;
|
2001-03-10 18:08:59 +01:00
|
|
|
struct stat st;
|
2001-03-09 01:12:22 +01:00
|
|
|
|
2019-06-28 15:35:04 +02:00
|
|
|
if (stat(id->filename, &st) == -1) {
|
2020-10-16 04:37:12 +02:00
|
|
|
do_log2(id->userprovided ?
|
|
|
|
SYSLOG_LEVEL_INFO : SYSLOG_LEVEL_DEBUG3,
|
|
|
|
"no such identity: %s: %s", id->filename, strerror(errno));
|
2001-03-10 18:08:59 +01:00
|
|
|
return NULL;
|
|
|
|
}
|
2015-01-14 21:05:27 +01:00
|
|
|
snprintf(prompt, sizeof prompt,
|
2015-11-15 23:26:49 +01:00
|
|
|
"Enter passphrase for key '%.100s': ", id->filename);
|
2015-01-14 21:05:27 +01:00
|
|
|
for (i = 0; i <= options.number_of_password_prompts; i++) {
|
|
|
|
if (i == 0)
|
|
|
|
passphrase = "";
|
|
|
|
else {
|
2000-09-23 08:15:56 +02:00
|
|
|
passphrase = read_passphrase(prompt, 0);
|
2015-01-14 21:05:27 +01:00
|
|
|
if (*passphrase == '\0') {
|
2000-09-23 08:15:56 +02:00
|
|
|
debug2("no passphrase given, try next key");
|
2015-01-14 21:05:27 +01:00
|
|
|
free(passphrase);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
2015-11-15 23:26:49 +01:00
|
|
|
switch ((r = sshkey_load_private_type(KEY_UNSPEC, id->filename,
|
2019-08-05 13:50:33 +02:00
|
|
|
passphrase, &private, &comment))) {
|
2015-01-14 21:05:27 +01:00
|
|
|
case 0:
|
|
|
|
break;
|
|
|
|
case SSH_ERR_KEY_WRONG_PASSPHRASE:
|
|
|
|
if (options.batch_mode) {
|
|
|
|
quit = 1;
|
|
|
|
break;
|
|
|
|
}
|
2015-01-15 12:04:36 +01:00
|
|
|
if (i != 0)
|
|
|
|
debug2("bad passphrase given, try again...");
|
2015-01-14 21:05:27 +01:00
|
|
|
break;
|
|
|
|
case SSH_ERR_SYSTEM_ERROR:
|
|
|
|
if (errno == ENOENT) {
|
2020-10-18 13:32:01 +02:00
|
|
|
debug2_r(r, "Load key \"%s\"", id->filename);
|
2001-03-09 01:12:22 +01:00
|
|
|
quit = 1;
|
2015-01-14 21:05:27 +01:00
|
|
|
break;
|
2000-09-23 08:15:56 +02:00
|
|
|
}
|
2015-01-14 21:05:27 +01:00
|
|
|
/* FALLTHROUGH */
|
|
|
|
default:
|
2020-10-18 13:32:01 +02:00
|
|
|
error_r(r, "Load key \"%s\"", id->filename);
|
2015-01-14 21:05:27 +01:00
|
|
|
quit = 1;
|
|
|
|
break;
|
|
|
|
}
|
2019-11-12 20:33:08 +01:00
|
|
|
if (private != NULL && sshkey_is_sk(private) &&
|
2019-10-31 22:18:28 +01:00
|
|
|
options.sk_provider == NULL) {
|
2020-02-06 23:30:54 +01:00
|
|
|
debug("key \"%s\" is an authenticator-hosted key, "
|
|
|
|
"but no provider specified", id->filename);
|
2019-10-31 22:18:28 +01:00
|
|
|
sshkey_free(private);
|
|
|
|
private = NULL;
|
|
|
|
quit = 1;
|
|
|
|
}
|
Merge 9.1 (#626)
* upstream: fix poll() spin when a channel's output fd closes without
data in the channel buffer. Introduce more exact packing of channel fds into
the pollfd array. fixes bz3405 and bz3411; ok deraadt@ markus@
OpenBSD-Commit-ID: 06740737849c9047785622ad5d472cb6a3907d10
* upstream: select post-quantum KEX
sntrup761x25519-sha512@openssh.com as the default; ok markus@
OpenBSD-Commit-ID: f02d99cbfce22dffec2e2ab1b60905fbddf48fb9
* upstream: add support for the "corp-data" protocol extension to
allow server-side copies to be performed without having to go via the client.
Patch by Mike Frysinger, ok dtucker@
OpenBSD-Commit-ID: 00aa510940fedd66dab1843b58682de4eb7156d5
* upstream: add a sftp client "cp" command that supports server-side
copying of files. Useful for this task and for testing the copy-data
extension. Patch from Mike Frysinger; ok dtucker@
OpenBSD-Commit-ID: 1bb1b950af0d49f0d5425b1f267e197aa1b57444
* depend
* Skip slow tests on (very) slow test targets.
* Set Makefile SHELL as determined by configure.
This should improve compatibility for users with non-POSIX shells. If
using Makefile.in directly (eg make -f Makefile.in distprep) then SHELL
will need to be specified on the command line (along with MANFMT in that
particular case). ok djm@
* Use bash or ksh if available for SH in Makefile.
* Increase test timeout to allow slow VMs to finish
* Only run regression tests on slow VMs.
* Only return events from ppoll that were requested.
If the underlying system's select() returns bits that were not in the
request set, our ppoll() implementation can return revents for events
not requested, which can apparently cause a hang. Only return revents
for activity in the requested event set. bz#3416, analysis and fix by
yaroslav.kuzmin at vmssoftware com, ok djm@
* Specify TEST_SHELL=bash on AIX.
The system shells cause the agent-restrict test to fail due to some
quoting so explicitly specify bash until we can get configure to
autmatically work around that.
* Disable security key on fbsd6 test host.
* upstream: man pages: add missing commas between subordinate and
main clauses
jmc@ dislikes a comma before "then" in a conditional, so leave those
untouched.
ok jmc@
OpenBSD-Commit-ID: 9520801729bebcb3c9fe43ad7f9776ab4dd05ea3
* upstream: ssh: document sntrup761x25519-sha512@openssh.com as
default KEX
OpenBSD-Commit-ID: 12545bfa10bcbf552d04d9d9520d0f4e98b0e171
* upstream: openssh-9.0
OpenBSD-Commit-ID: 0dfb461188f4513ec024c1534da8c1ce14c20b64
* update version numbers for release
* update build-aux files to match autoconf-2.71
i.e. config.guess, config.sub and install-sh
* Revert "update build-aux files to match autoconf-2.71"
This reverts commit 0a8ca39fac6ad19096b6c263436f8b2dd51606f2.
It turns out that the checked-in copies of these files are actually newer
than autoconf-2.71's copies, so this was effectively a downgrade.
Spotted by Bo Anderson via github
* upstream: two defensive changes from Tobias Stoeckmann via GHPR287
enforce stricter invarient for sshbuf_set_parent() - never allow
a buffer to have a previously-set parent changed.
In sshbuf_reset(), if the reallocation fails, then zero the entire
buffer and not the (potentially smaller) default initial alloc size.
OpenBSD-Commit-ID: 14583203aa5d50ad38d2e209ae10abaf8955e6a9
* upstream: Note that curve25519-sha256 was later published in
RFC8731. ok djm@
OpenBSD-Commit-ID: 2ac2b5d642d4cf5918eaec8653cad9a4460b2743
* upstream: clear io_want/io_ready flags at start of poll() cycle;
avoids plausible spin during rekeying if channel io_want flags are reused
across cycles. ok markus@ deraadt@
OpenBSD-Commit-ID: 91034f855b7c73cd2591657c49ac30f10322b967
* Retire fbsd6 test VM.
It's long since out of support, relatively slow (it's i686) and the
compiler has trouble with PIE.
* Resync moduli.5 with upstream.
1.18: remove duplicate publication year; carsten dot kunze at arcor dot de
1.19: ssh-keygen's -G/-T have been replaced with -M generate/screen.
* upstream: Correct path for system known hosts file in description
of IgnoreUserKnownHosts. Patch from Martin Vahlensieck via tech@
OpenBSD-Commit-ID: 9b7784f054fa5aa4d63cb36bd563889477127215
* upstream: list the correct version number
for when usage of the sftp protocol became default and fix a typo
from ed maste
OpenBSD-Commit-ID: 24e1795ed2283fdeacf16413c2f07503bcdebb31
* upstream: Import regenerated moduli
OpenBSD-Commit-ID: f9a0726d957cf10692a231996a1f34e7f9cdfeb0
* upstream: Try to continue running local I/O for channels in state
OPEN during SSH transport rekeying. The most visible benefit is that it
should make ~-escapes work in the client (e.g. to exit) if the connection
happened to have stalled during a rekey event. Based work by and ok dtucker@
OpenBSD-Commit-ID: a66e8f254e92edd4ce09c9f750883ec8f1ea5f45
* upstream: Import regenerated moduli
OpenBSD-Commit-ID: f9a0726d957cf10692a231996a1f34e7f9cdfeb0
* upstream: regression test for sftp cp command
OpenBSD-Regress-ID: c96bea9edde3a384b254785e7f9b2b24a81cdf82
* upstream: Simplify forward-control test.
Since we no longer need to support SSH1 we don't need to run shell
commands on the other end of the connection and can use ssh -N instead.
This also makes the test less racy.
OpenBSD-Regress-ID: 32e94ce272820cc398f30b848b2b0f080d10302c
* upstream: Use ssh -f and ControlPersist ..
to start up test forwards and ssh -O stop to shut them down intead of
sleep loops. This speeds up the test by an order of magnitude.
OpenBSD-Regress-ID: eb3db5f805100919b092a3b2579c611fba3e83e7
* upstream: It looks like we can't completely avoid
waiting for processes to exit so retrieve the pid via controlmaster and
use that.
OpenBSD-Regress-ID: 8246f00f22b14e49d2ff1744c94897ead33d457b
* Cache timezone data in capsicum sandbox.
From emaste at freebsd.org, originally part of FreeBSD commit r339216
/ fc3c19a9 with autoconf bits added by me.
* Include stdlib.h for free() prototype.
... which is used inside the CUSTOM_SYS_AUTH_GET_LASTLOGIN_MSG block.
* Update OpenSSL and LibreSSL versions in tests.
* Add debian-riscv64 test target.
* upstream: Avoid an unnecessary xstrdup in rm_env() when matching
patterns. Since match_pattern() doesn't modify its arguments (they are
const), there is no need to make an extra copy of the strings in
options->send_env. From Martin Vahlensieck
OpenBSD-Commit-ID: 2c9db31e3f4d3403b49642c64ee048b2a0a39351
* upstream: Add missing includes of stdlib.h and stdint.h. We need
stdlib.h for malloc(3) and stdint.h for SIZE_MAX. Unlike the other xmss
files, ssh-xmss.c does not include xmss_commons.h so ssh-xmss.c must include
those headers itself. From Martin Vahlensieck
OpenBSD-Commit-ID: 70e28a9818cee3da1be2ef6503d4b396dd421e6b
* upstream: Remove unnecessary includes: openssl/hmac.h and
openssl/evp.h. From Martin Vahlensieck.
OpenBSD-Commit-ID: a6debb5fb0c8a44e43e8d5ca7cc70ad2f3ea31c3
* upstream: Check sshauthopt_new() for NULL. bz#3425, from
tessgauthier at microsoft.com. ok djm@
OpenBSD-Commit-ID: af0315bc3e44aa406daa7e0ae7c2d719a974483f
* upstream: Add authfd path to debug output. ok markus@
OpenBSD-Commit-ID: f735a17d1a6f2bee63bfc609d76ef8db8c090890
* upstream: avoid printing hash algorithm twice; from lucas AT sexy.is
OpenBSD-Commit-ID: 9d24671e10a84141b7c504396cabad600e47a941
* upstream: fix memleak on session-bind path; from Pedro Martelletto, ok
dtucker@
OpenBSD-Commit-ID: e85899a26ba402b4c0717b531317e8fc258f0a7e
* upstream: Don't leak SK device. Patch from Pedro Martelletto via
github PR#316. ok djm@
OpenBSD-Commit-ID: 17d11327545022e727d95fd08b213171c5a4585d
* upstream: mention that the helpers are used by ssh(1), ssh-agent(1)
and ssh-keygen(1). Previously only ssh(1) was mentioned. From Pedro
Martelletto
OpenBSD-Commit-ID: 30f880f989d4b329589c1c404315685960a5f153
* Remove now-empty int32_minmax.inc.
* Only run tests when source files change.
Also run tests on changes to V_9_0 branch.
* Add Mac OS X 12 test target.
* upstream: be stricter in which characters will be accepted in
specifying a mask length; allow only 0-9. From khaleesicodes via GHPR#278; ok
dtucker@
OpenBSD-Commit-ID: e267746c047ea86665cdeccef795a8a56082eeb2
* upstream: fix some integer overflows in sieve_large() that show up when
trying to generate modp groups > 16k bits. Reported via GHPR#306 by Bertram
Felgenhauer, but fixed in a different way. feedback/ok tb@
OpenBSD-Commit-ID: 81cbc6dd3a21c57bd6fadea10e44afe37bca558e
* upstream: remove an obsolete rsa1 format example from an example;
from megan batty
ok djm
OpenBSD-Commit-ID: db2c89879c29bf083df996bd830abfb1e70d62bf
* upstream: Add FIDO AUTHENTICATOR section and explain a bit how FIDO
works. The wording came mostly from the 8.2 OpenSSH release notes, addapted
to fit the man page. Then move the -O bits into the new section as is already
done for CERTIFICATES and MODULI GENERATION. Finally we can explain the
trade-offs of resident keys. While here, consistently refer to the FIDO
thingies as "FIDO authenticators", not "FIDO tokens".
input & OK jmc, naddy
OpenBSD-Commit-ID: dd98748d7644df048f78dcf793b3b63db9ab1d25
* upstream: make sure stdout is non-blocking; ok djm@
OpenBSD-Commit-ID: 64940fffbd1b882eda2d7c8c7a43c79368309c0d
* upstream: mux.c: mark argument as const; from Martin Vahlensieck
OpenBSD-Commit-ID: 69a1a93a55986c7c2ad9f733c093b46a47184341
* upstream: channel_new no longer frees remote_name. So update the
comment accordingly. As remote_name is not modified, it can be const as
well. From Martin Vahlensieck
OpenBSD-Commit-ID: e4e10dc8dc9f40c166ea5a8e991942bedc75a76a
* upstream: sshkey_unshield_private() contains a exact duplicate of
the code in private2_check_padding(). Pull private2_check_padding() up so the
code can be reused. From Martin Vahlensieck, ok deraadt@
OpenBSD-Commit-ID: 876884c3f0e62e8fd8d1594bab06900f971c9c85
* Add ubsan minimal testcase on OpenBSD.
As suggested by djm@.
* Note that, for now, we need variadic macros.
* Also retest OpenBSD upstream on .yml changes.
* upstream: When performing operations that glob(3) a remote path, ensure
that the implicit working directory used to construct that path escapes
glob(3) characters.
This prevents glob characters from being processed in places they
shouldn't, e.g. "cd /tmp/a*/", "get *.txt" should have the get operation
treat the path "/tmp/a*" literally and not attempt to expand it.
Reported by Lusia Kundel; ok markus@
OpenBSD-Commit-ID: 4f647f58482cbad3d58b1eab7f6a1691433deeef
* Remove duplicate bcrypt_pbkdf.o from Makefile
bcrypt_pbkdf.o is duplicated in the openbsd-compat Makefile's object
file list.
* upstream: improve error message when 'ssh-keygen -Y sign' is unable to
load a private key; bz3429, reported by Adam Szkoda ok dtucker@
OpenBSD-Commit-ID: bb57b285e67bea536ef81b1055467be2fc380e74
* upstream: Allow existing -U (use agent) flag to work with "-Y sign"
operations, where it will be interpreted to require that the private keys is
hosted in an agent; bz3429, suggested by Adam Szkoda; ok dtucker@
OpenBSD-Commit-ID: a7bc69873b99c32c42c7628ed9ea91565ba08c2f
* upstream: Remove errant apostrophe. From haruyama at queen-ml org.
OpenBSD-Commit-ID: dc6b294567cb84b384ad6ced9ca469f2bbf0bd10
* upstream: arrange for scp, when in sftp mode, to not ftruncate(3) files
early
previous behavious of unconditionally truncating the destination file
would cause "scp ~/foo localhost:" and "scp localhost:foo ~/" to
delete all the contents of their destination.
spotted by solene@ sthen@, also bz3431; ok dtucker@
OpenBSD-Commit-ID: ca39fdd39e0ec1466b9666f15cbcfddea6aaa179
* upstream: fix in-place copies; r1.163 incorrectly skipped truncation in
all cases, not just at the start of a transfer. This could cause overwrites
of larger files to leave junk at the end. Spotted by tb@
OpenBSD-Commit-ID: b189f19cd68119548c8e24e39c79f61e115bf92c
* upstream: Only run agent-ptrace.sh if gdb is available as all
architectures do not ship with gdb.
OpenBSD-Regress-ID: ec53e928803e6b87f9ac142d38888ca79a45348d
* upstream: regress test for in-place transfers and clobbering larger
files with smaller ones; would have caught last regression in scp(1)
OpenBSD-Regress-ID: 19de4e88dd3a4f7e5c1618c9be3c32415bd93bc2
* configure.ac: Add missing AC_DEFINE for caph_cache_tzdata test causing
HAVE_CAPH_CACHE_TZDATA to be missing from config.h.in.
Spotted by Bryan Drewery
* upstream: make SSHBUF_DBG/SSHBUF_TELL (off by default and only enabled
via #define) dump to stderr rather than stdout
OpenBSD-Commit-ID: 10298513ee32db8390aecb0397d782d68cb14318
* upstream: revert previous; it was broken (spotted by Theo)
OpenBSD-Commit-ID: 457c79afaca2f89ec2606405c1059b98b30d8b0d
* upstream: Note that ProxyJump also accepts the same tokens as
ProxyCommand. From pallxk via github PR#305.
OpenBSD-Commit-ID: 7115ac351b129205f1f1ffa6bbfd62abd76be7c5
* upstream: Avoid kill with -1 argument. The out_ctx label can be
reached before fork has been called. If this happens, then kill -1 would be
called, sending SIGTERM to all processes reachable by the current process.
From tobias@ and c3h2_ctf via github PR#286, ok djm@
OpenBSD-Commit-ID: 6277af1207d81202f5daffdccfeeaed4c763b1a8
* upstream: f sshpkt functions fail, then password is not cleared
with freezero. Unconditionally call freezero to guarantee that password is
removed from RAM.
From tobias@ and c3h2_ctf via github PR#286, ok djm@
OpenBSD-Commit-ID: 6b093619c9515328e25b0f8093779c52402c89cd
* upstream: refactor authorized_keys/principals handling
remove "struct ssh *" from arguments - this was only used to pass the
remote host/address. These can be passed in instead and the resulting
code is less tightly coupled to ssh_api.[ch]
ok dtucker@
OpenBSD-Commit-ID: 9d4373d013edc4cc4b5c21a599e1837ac31dda0d
* upstream: split the low-level file handling functions out from
auth2-pubkey.c
Put them in a new auth2-pubkeyfile.c to make it easier to refer to them
(e.g. in unit/fuzz tests) without having to refer to everything else
pubkey auth brings in.
ok dtucker@
OpenBSD-Commit-ID: 3fdca2c61ad97dc1b8d4a7346816f83dc4ce2217
* fuzzer for authorized_keys parsing
mostly redundant to authopt_fuzz, but it's sensitive code so IMO it
makes sense to test this layer too
* Test against LibreSSL 3.5.3.
* Test against OpenSSL 1.1.1o and 3.0.3.
* fix some bugs in the fuzzer
* upstream: keywords ref ssh_config.5;
from caspar schutijser
OpenBSD-Commit-ID: f146a19d7d5c9374c3b9c520da43b2732d7d1a4e
* upstream: ssh-keygen: implement "verify-required" certificate option.
This was already documented when support for user-verified FIDO
keys was added, but the ssh-keygen(1) code was missing.
ok djm@
OpenBSD-Commit-ID: f660f973391b593fea4b7b25913c9a15c3eb8a06
* upstream: ssh-keygen -A: do not generate DSA keys by default.
Based on github PR#303 from jsegitz with man page text from jmc@, ok markus@
djm@
OpenBSD-Commit-ID: 5c4c57bdd7063ff03381cfb6696659dd3f9f5b9f
* upstream: Add period at end of "not known by any other names"
message. github PR#320 from jschauma, ok djm@
OpenBSD-Commit-ID: bd60809803c4bfd3ebb7c5c4d918b10e275266f2
* upstream: Add missing *-sk types to ssh-keyscan manpage. From
skazi0 via github PR#294.
OpenBSD-Commit-ID: fda2c869cdb871f3c90a89fb3f985370bb5d25c0
* upstream: Make SetEnv directives first-match-wins in both
sshd_config and sshd_config; previously if the same name was reused then the
last would win (which is the opposite to how the config is supposed to work).
While there, make the ssh_config parsing more like sshd_config.
bz3438, ok dtucker
OpenBSD-Commit-ID: 797909c1e0262c0d00e09280459d7ab00f18273b
* upstream: test setenv in both client and server, test first-match-wins
too
OpenBSD-Regress-ID: 4c8804f9db38a02db480b9923317457b377fe34b
* upstream: move auth_openprincipals() and auth_openkeyfile() over to
auth2-pubkeyfile.c too; they make more sense there.
OpenBSD-Commit-ID: 9970d99f900e1117fdaab13e9e910a621b7c60ee
* upstream: make sure that UseDNS hostname lookup happens in the monitor
and not in the pledge(2)'d unprivileged process; fixes regression caused by
recent refactoring spotted by henning@
OpenBSD-Commit-ID: a089870b95101cd8881a2dff65b2f1627d13e88d
* fix possible NULL deref when built without FIDO
Analysis/fix from kircher in bz3443; ok dtucker@
* automatically enable built-in FIDO support
If libfido2 is found and usable, then enable the built-in
security key support unless --without-security-key-builtin
was requested.
ok dtucker@
* upstream: Log an error if pipe() fails while accepting a
connection. bz#3447, from vincent-openssh at vinc17 net, ok djm@
OpenBSD-Commit-ID: 9d59f19872b94900a5c79da2d57850241ac5df94
* upstream: Don't attempt to fprintf a null identity comment. From
Martin Vahlensieck via tech@.
OpenBSD-Commit-ID: 4c54d20a8e8e4e9912c38a7b4ef5bfc5ca2e05c2
* upstream: Make sure not to fclose() the same fd twice in case of an
error.
ok dtucker@
OpenBSD-Commit-ID: e384c4e05d5521e7866b3d53ca59acd2a86eef99
* upstream: make it clear that RekeyLimit applies to both transmitted
and received data. GHPR#328 from Jan Pazdziora
OpenBSD-Commit-ID: d180a905fec9ff418a75c07bb96ea41c9308c3f9
* request 1.1x API compatibility for OpenSSL >=3.x
idea/patch from Pedro Martelletto via GHPR#322; ok dtucker@
* fix broken case statement in previous
* Disable SK support if FIDO libs not found.
* Zero out LIBFIDO2 when SK support not usable.
Prevents us from trying to link them into ssh-sk-helper and failing to
build.
* upstream: Don't leak the strings allocated by order_hostkeyalgs()
and list_hostkey_types() that are passed to compat_pkalg_proposal(). Part of
github PR#324 from ZoltanFridrich, ok djm@
OpenBSD-Commit-ID: b2f6e5f60f2bba293b831654328a8a0035ef4a1b
* upstream: Roll back previous KEX changes as they aren't safe until
compat_pkalg_proposal and friends always allocate their returned strings.
Reported by Qualys.
OpenBSD-Commit-ID: 1c7a88a0d5033f42f88ab9bec58ef1cf72c81ad0
* upstream: allow arguments to sftp -D option, e.g. sftp -D
"/usr/libexec/sftp-server -el debug3"
ok markus@
OpenBSD-Commit-ID: 5a002b9f3a7aef2731fc0ffa9c921cf15f38ecce
* Update OpenSSL tests to the most recent releases.
* upstream: reflect the update to -D arg name in usage();
OpenBSD-Commit-ID: abdcde4f92b1ef094ae44210ee99d3b0155aad9c
* upstream: ignore SIGPIPE earlier in main(), specifically before
muxclient() which performs operations that could cause one; Reported by Noam
Lewis via bz3454, ok dtucker@
OpenBSD-Commit-ID: 63d8e13276869eebac6d7a05d5a96307f9026e47
* upstream: Always return allocated strings from the kex filtering so
that we can free them later. Fix one leak in compat_kex_proposal. Based on
github PR#324 from ZoltanFridrich with some simplications by me. ok djm@
OpenBSD-Commit-ID: 9171616da3307612d0ede086fd511142f91246e4
* upstream: Don't leak the strings allocated by order_hostkeyalgs()
and list_hostkey_types() that are passed to compat_pkalg_proposal(). Part of
github PR#324 from ZoltanFridrich, ok djm@
This is a roll-forward of the previous rollback now that the required
changes in compat.c have been done.
OpenBSD-Commit-ID: c7cd93730b3b9f53cdad3ae32462922834ef73eb
* upstream: bump up loglevel from debug to info when unable to open
authorized keys/principals file for errno != ENOENT; bz2042 ok dtucker
OpenBSD-Commit-ID: e79aa550d91ade6a80f081bda689da24c086d66b
* Skip select+rlimit check if sandboxing is disabled
It's not needed in that case, and the test can fail when being built
with some compiler memory sanitizer flags. bz#3441
* upstream: use consistent field names (s/char/byte)
in format description
OpenBSD-Commit-ID: 3de33572733ee7fcfd7db33d37db23d2280254f0
* upstream: Remove leftover line.
Remove extra line leftover from merge conflict. ok djm@
OpenBSD-Commit-ID: 460e2290875d7ae64971a7e669c244b1d1c0ae2e
* Move checks for pollfd.fd and nfds_t.
Move the checks for struct pollfd.fd and nfds_t to before the sandboxing
checks. This groups all the sandbox checks together so we can skip them
all when sandboxing is disabled.
* Skip all rlimit tests when sandboxing disabled.
The rlimit tests can hang when being run with some compiler sanitizers
so skip all of them if sandbox=no.
* Add clang sanitizer tests.
* upstream: Add TEST_REGRESS_CACHE_DIR.
If set, it is used to cache regress test names that have succeeded and
skip those on a re-run.
OpenBSD-Regress-ID: a7570dd29a58df59f2cca647c3c2ec989b49f247
* Move sanitizer logs into regress for collection.
* Add GCC address sanitizer build/test.
* Update sanitizer test targets:
- remove clang-sanitize-memory for now. It takes so long that the test
times out.
- add gcc sanitize-address and sanitize-undefined test targets.
* Test against openssl-3.0.5.
* Move unset to before we set anything.
* Refuse to use OpenSSL 3.0.4 due to potential RCE.
OpenSSL has a potential RCE in its RSA implementation (CVE-2022-2274)
so refuse to use that specific version.
* Capture stderr output from configure.
* Only refuse to use OpenSSL 3.0.4 on x86_64.
The potential RCE only impacts x86_64, so only refuse to use it if we're
targetting a potentially impacted architecture. ok djm@
* Remove special casing of crypt().
Configure goes to some lengths to pick crypt() from either libcrypt
or OpenSSL's libcrypto because they can more or less featureful (eg
supporting md5-style passwords).
OpenSSL removed its crypt() interface in 2002:
https://github.com/openssl/openssl/commit/69deec58 so these hijinks
should no longer be necessary. This also only links sshd with libcrypt
which is the only thing that needs it. ok djm@
* Clarify README.md text.
Clarify the text about the implications of building without OpenSSL, and
prefix the "configure --help" example command with a "./" so it's likely
to work as-is in more shells. From bz#3461.
* Split README.platform into its own line.
README.platform has general platform-specific information, having it
following text about FIDO2 on the same line could imply that it only
has information about FIDO2.
* Return ERANGE from getcwd() if buffer size is 1.
If getcwd() is supplied a buffer size of exactly 1 and a path of "/", it
could result in a nul byte being written out of array bounds. POSIX says
it should return ERANGE if the path will not fit in the available buffer
(with terminating nul). 1 byte cannot fit any possible path with its nul,
so immediately return ERANGE in that case.
OpenSSH never uses getcwd() with this buffer size, and all current
(and even quite old) platforms that we are currently known to work
on have a native getcwd() so this code is not used on those anyway.
Reported by Qualys, ok djm@
* Remove unintended changes.
I inadvertently included a couple of local changes with the OpenSSL
3.0.4 change. Revert, anything that should be there will be committed
separately.
* Add AUDIT_ARCH_PPC to supported seccomp arches.
Patch from dries.deschout at dodeco.eu.
* Rename bbone test target to ARM.
* Move vmshutdown to first step.
If a previous run on a physical runner has failed to clean up, the next
run will fail because it'll try to check out the code to a broken
directory mount. Make cleanup the first step.
* upstream: pull passphrase reading and confirmation into a separate
function so it can be used for FIDO2 PINs; no functional change
OpenBSD-Commit-ID: bf34f76b8283cc1d3f54633e0d4f13613d87bb2f
* upstream: when enrolling a resident key on a security token, check
if a credential with matching application and user ID strings already exists.
if so, prompt the user for confirmation before overwriting the credential.
patch from Pedro Martelletto via GHPR329
NB. cranks SSH_SK_VERSION_MAJOR, so any third-party FIDO middleware
implementations will need to adjust
OpenBSD-Commit-ID: e45e9f1bf2b2f32d9850669e7a8dbd64acc5fca4
* upstream: sk-usbhid: preserve error code returned by key_lookup()
it conveys useful information, such as the supplied pin being wrong.
Part of GHPR329 from Pedro Martelletto
OpenBSD-Commit-ID: c0647eb9290f793add363d81378439b273756c1b
* upstream: ssh-keygen: fix touch prompt, pin retries;
part of GHPR329 from Pedro Martelletto
OpenBSD-Commit-ID: 75d1005bd2ef8f29fa834c90d2684e73556fffe8
* crank SSH_SK_VERSION_MAJOR in sk-dummy.so
* Skip scp3 test if there's no scp on remote path.
scp -3 ends up using the scp that's in the remote path and will fail if
one is not available. Based on a patch from rapier at psc.edu.
* Convert "have_prog" function into "which".
"which" and its behaviour is not standardized, so convert the existing
have_prog function into "which" so we can rely on it being available
and what its semantics are. Add a have_prog wrapper that maintains the
existing behaviour.
* upstream: Test TEST_SSH_ELAPSED_TIMES for empty string not
executable. No-op on most platforms but should prevent warnings in -portable
on systems that don't have 'date %s'.
OpenBSD-Regress-ID: e39d79867b8065e33d0c5926fa1a31f85659d2a4
* upstream: Restore missing "!" in TEST_SSH_ELAPSED_TIMES test.
OpenBSD-Regress-ID: 38783f9676ec348c5a792caecee9a16e354b37b0
* Remove workarounds for OpenSSL missing AES-GCM.
We have some compatibility hacks that were added to support OpenSSL
versions that do not support AES GCM mode. Since that time, however,
the minimum OpenSSL version that we support has moved to 1.0.1 which
*does* have GCM, so this is no longer needed. ok djm@
* Remove workarounds for OpenSSL missing AES-CTR.
We have some compatibility hacks that were added to support OpenSSL
versions that do not support AES CTR mode. Since that time, however,
the minimum OpenSSL version that we support has moved to 1.0.1 which
*does* have CTR, so this is no longer needed. ok djm@
* Do not link scp, sftp and sftp-server w/ zlib.
Some of our binaries (eg sftp, sftp-server, scp) do not interact with
the channels code and thus do use libraries such as zlib and libcrypto
although they are linked with them. This adds a CHANNELLIBS and starts
by moving zlib into it, which means the aformentioned binaries are no
longer linked against zlib. ok djm@
* Group libcrypto and PRNGD checks together.
They're related more than the libcrypt or libiaf checks which are
currently between them. ok djm@
* Remove seed_rng calls from scp, sftp, sftp-server.
These binaries don't use OpenSSL's random functions. The next step
will be to stop linking them against libcrypto. ok djm@
* Move libcrypto into CHANNELLIBS.
This will result in sftp, sftp-server and scp no longer being linked
against libcrypto. ok djm@
* Move stale-configure check as early as possible.
We added a check in Makefile to catch the case where configure needs to
be rebuilt, however this did not happen until a build was attempted in
which case all of the work done by configure was wasted. Move this check
to the start of configure to catch it as early as possible. ok djm@
* Remove deprecated MacOS 10.15 runners.
* upstream: avoid double-free in error path introduced in r1.70; report
and fix based on GHPR#332 by v-rzh ok dtucker@
OpenBSD-Commit-ID: 3d21aa127b1f37cfc5bdc21461db369a663a951f
* Include CHANNEL and FIDO2 libs in configure output
* Factor out getrnd() and rename to getentropy().
Factor out the arc4random seeding into its own file and change the
interface to match getentropy. Use native getentropy if available.
This will make it easier to resync OpenBSD changes to arc4random.
Prompted by bz#3467, ok djm@.
* compat code for fido_dev_is_winhello()
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* check_sk_options: add temporary WinHello workaround
Up to libfido 1.10.0, WinHello advertises "clientPin" rather
than "uv" capability. This is fixed in 1.11.0. For the time
being, workaround it here.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* sk_sign: set FIDO2 uv attribute explicitely for WinHello
WinHello via libfido2 performs user verification by default.
However, if we stick to that, there's no way to differentiate
between keys created with or without "-O verify-required".
Set FIDO2 uv attribute explicitely to FIDO_OPT_FALSE, then check
if user verification has been requested.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* upstream: don't prompt for FIDO passphrase before attempting to enroll
the credential, just let the enroll operating fail and we'll attempt to get a
PIN anyway. Might avoid some unneccessary PIN prompts.
Part of GHPR#302 from Corinna Vinschen; ok dtucker@
OpenBSD-Commit-ID: bd5342ffc353ee37d39617906867c305564d1ce2
* Give unused param a name.
Fixes builds on platforms that do have fido2 but don't have
fido_dev_is_winhello.
* Actually put HAVE_STDINT_H around the stdint.h.
* Rename our getentropy to prevent possible loops.
Since arc4random seeds from getentropy, and we use OpenSSL for that
if enabled, there's the possibility that if we build on a system that
does not have getentropy then run on a system that does have it, then
OpenSSL could end up calling our getentropy and getting stuck in a loop.
Pointed out by deraadt@, ok djm@
* Test hostbased auth on github runners.
* fix SANDBOX_SECCOMP_FILTER_DEBUG
* Fix conditional for running hostbased tests.
* upstream: allow certificate validity intervals, sshsig verification
times and authorized_keys expiry-time options to accept dates in the UTC time
zone in addition to the default of interpreting them in the system time zone.
YYYYMMDD and YYMMDDHHMM[SS] dates/times will be interpreted as UTC if
suffixed with a 'Z' character.
Also allow certificate validity intervals to be specified in raw
seconds-since-epoch as hex value, e.g. -V 0x1234:0x4567890. This
is intended for use by regress tests and other tools that call
ssh-keygen as part of a CA workflow.
bz3468 ok dtucker
OpenBSD-Commit-ID: 454db1cdffa9fa346aea5211223a2ce0588dfe13
* upstream: add some tests for parse_absolute_time(), including cases
where it is forced to the UTC timezone. bz3468 ok dtucker
OpenBSD-Regress-ID: ea07ca31c2f3847a38df028ca632763ae44e8759
* Skip hostbased during Valgrind tests.
Valgrind doesn't let ssh exec ssh-keysign (because it's setuid) so skip
it during the Valgrind based tests.
See https://bugs.kde.org/show_bug.cgi?id=119404 for a discussion of this
(ironically there the problematic binary was ssh(1) back when it could
still be setuid).
* Rerun tests if any .github config file changes.
* Add a timegm implementation from Heimdal via Samba.
Fixes build on (at least Solaris 10).
* Replace deprecated ubuntu-18.04 runners with 22.04
* upstream: sftp-server: support home-directory request
Add support to the sftp-server for the home-directory extension defined
in draft-ietf-secsh-filexfer-extensions-00. This overlaps a bit with the
existing expand-path@openssh.com, but uses a more official protocol name,
and so is a bit more likely to be implemented by non-OpenSSH clients.
From Mike Frysinger, ok dtucker@
OpenBSD-Commit-ID: bfc580d05cc0c817831ae7ecbac4a481c23566ab
* fido_dev_is_winhello: return 0, not "false"
"false" is not used anywhere in OpenSSH, so return 0 like
everywhere else.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* Revert "check_sk_options: add temporary WinHello workaround"
Cygwin now comes with libfido2 1.11.0, so this workaround
isn't required anymore.
This reverts commit 242c044ab111a37aad3b0775727c36a4c5f0102c.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* upstream: use .Cm for "sign"; from josiah frentsos
OpenBSD-Commit-ID: 7f80a53d54857ac6ae49ea6ad93c5bd12231d1e4
* upstream: add an extra flag to sk_probe() to indicate whether we're
probing for a FIDO resident key or not. Unused here, but will make like
easier for portable
OpenBSD-Commit-ID: 432c8ff70e270378df9dbceb9bdeaa5b43b5a832
* on Cygwin, prefer WinHello FIDO device
If no FIDO device was explictly specified, then prefer the
windows://hello FIDO device. An exception to this is when
probing resident FIDO keys, in which case hardware FIDO
devices are preferred.
* Check for perms to run agent-getpeereid test.
Ubuntu 22.04 defaults to private home dirs which prevents "nobody"
running ssh-add during the agent-getpeereid test. Check for this and
add the necessary permissions.
* upstream: double free() in error path; from Eusgor via GHPR333
OpenBSD-Commit-ID: 39f35e16ba878c8d02b4d01d8826d9b321be26d4
* Add Cygwin (on windows-2019) test target.
In addition to installing the requisite Cygwin packages, we also need to
explicitly invoke "sh" for steps that run other scripts since the runner
environment doesn't understand #! paths.
* Add a bit more debug output.
* Fix cygwin conditional steps.
* upstream: Strictly enforce the maximum allowed SSH2 banner size in
ssh-keyscan and prevent a one-byte buffer overflow. Patch from Qualys, ok
djm@
OpenBSD-Commit-ID: 6ae664f9f4db6e8a0589425f74cd0bbf3aeef4e4
* upstream: remove incorrect check that can break enrolling a
resident key (introduced in r1.40)
OpenBSD-Commit-ID: 4cab364d518470e29e624af3d3f9ffa9c92b6f01
* upstream: attemp FIDO key signing without PIN and use the error
code returned to fall back only if necessary. Avoids PIN prompts for FIDO
tokens that don't require them; part of GHPR#302
OpenBSD-Commit-ID: 4f752aaf9f2e7c28bcaaf3d4f8fc290131bd038e
* Install Cygwin packages based on OS not config.
* initial list of allowed signers
* upstream: whitespace
OpenBSD-Commit-ID: d297e4387935d4aef091c5e9432578c2e513f538
* upstream: whitespace
OpenBSD-Commit-ID: a5d015efbfd228dc598ffdef612d2da3a579e5d8
* Add cygwin-release test target.
This also moves the cygwin package install from the workflow file to
setup_ci.sh so that we can install different sets of Cygwin packages
for different test configs.
* Add Windows 2022 test targets.
* Add libcrypt-devel to cygwin-release deps.
Based on feedback from vinschen at redhat.com.
* cross-sign allowed_signers with PGP key
Provides continuity of trust from legacy PGP release key to
the SSHSIG signing keys that we will use henceforth for git
signing.
* additional keys
* upstream: whitespace
OpenBSD-Commit-ID: c2bcbf93610d3d62ed206cdf9bf9ff98c6aaf232
* Move sftp from valgrind-2 to 3 to rebalance.
* upstream: sk-usbhid: fix key_lookup() on tokens with built-in UV
explicitly test whether the token performs built-in UV (e.g. biometric
tokens) and enable UV in that case. From Pedro Martelletto via GHPR#388
OpenBSD-Commit-ID: 007eb7e387d27cf3029ab06b88224e03eca62ccd
* Remove arc4random_uniform from arc4random.c
This was previously moved into its own file (matching OpenBSD) which
prematurely committed in commit 73541f2.
* Move OPENBSD ORIGINAL marker.
Putting this after the copyright statement (which doesn't change)
instead of before the version identifier (which does) prevents merge
conflicts when resyncing changes.
* Resync arc4random with OpenBSD.
This brings us up to current, including djm's random-reseeding change,
as prompted by logan at cyberstorm.mu in bz#3467. It brings the
platform-specific hooks from LibreSSL Portable, simplified to match our
use case. ok djm@.
* Remove DEF_WEAK, it's already in defines.h.
* openbsd-compat/bsd-asprintf: add <stdio.h> include for vsnprintf
Fixes the following build failure with Clang 15 on musl:
```
bsd-asprintf.c:51:8: error: call to undeclared library function 'vsnprintf' with type 'int (char *, unsigned long, const char *, struct __va_list_tag *)'; ISO C99 and laterclang -O2 -pipe -fdiagnostics-color=always -frecord-gcc-switches -pipe -Wunknown-warning-option -Qunused-arguments -Wall -Wpointer-arith -Wuninitialized -Wsign-compare -Wformat-security -Wsizeof-pointer-memaccess -Wno-pointer-sign -Wno-unused-result -Wmisleading-indentation -Wbitwise-instead-of-logical -fno-strict-aliasing -mretpoline -ftrapv -fzero-call-used-regs=all -fno-builtin-memset -fstack-protector-strong -fPIE -I. -I. -D_XOPEN_SOURCE=600 -D_BSD_SOURCE -D_DEFAULT_SOURCE -DSSHDIR=\"/etc/ssh\" -D_PATH_SSH_PROGRAM=\"/usr/bin/ssh\" -D_PATH_SSH_ASKPASS_DEFAULT=\"/usr/lib/misc/ssh-askpass\" -D_PATH_SFTP_SERVER=\"/usr/lib/misc/sftp-server\" -D_PATH_SSH_KEY_SIGN=\"/usr/lib/misc/ssh-keysign\" -D_PATH_SSH_PKCS11_HELPER=\"/usr/lib/misc/ssh-pkcs11-helper\" -D_PATH_SSH_SK_HELPER=\"/usr/lib/misc/ssh-sk-helper\" -D_PATH_SSH_PIDDIR=\"/run\" -D_PATH_PRIVSEP_CHROOT_DIR=\"/var/empty\" -DHAVE_CONFIG_H -c cipher-aes.c -o cipher-aes.o
do not support
implicit function declarations [-Wimplicit-function-declaration]
ret = vsnprintf(string, INIT_SZ, fmt, ap2);
^
bsd-asprintf.c:51:8: note: include the header <stdio.h> or explicitly provide a declaration for 'vsnprintf'
1 error generated.
```
* upstream: notifier_complete(NULL, ...) is a noop, so no need to test
that ctx!=NULL; from Corinna Vinschen
OpenBSD-Commit-ID: ade2f2e9cc519d01a586800c25621d910bce384a
* fix pester test failures
* upstream: fix repeated words ok miod@ jmc@
OpenBSD-Commit-ID: 6765daefe26a6b648cc15cadbbe337596af709b7
* upstream: .Li -> .Vt where appropriate; from josiah frentsos,
tweaked by schwarze
ok schwarze
OpenBSD-Commit-ID: 565046e3ce68b46c2f440a93d67c2a92726de8ed
* upstream: ssh-agent: attempt FIDO key signing without PIN and use
the error to determine whether a PIN is required and prompt only if
necessary. from Corinna Vinschen
OpenBSD-Commit-ID: dd6be6a0b7148608e834ee737c3479b3270b00dd
* upstream: a little extra debugging
OpenBSD-Commit-ID: edf1601c1d0905f6da4c713f4d9cecc7d1c0295a
* upstream: sk_enroll: never drop SSH_SK_USER_VERIFICATION_REQD flag
from response
Now that all FIDO signing calls attempt first without PIN and then
fall back to trying PIN only if that attempt fails, we can remove the
hack^wtrick that removed the UV flag from the keys returned during
enroll.
By Corinna Vinschen
OpenBSD-Commit-ID: 684517608c8491503bf80cd175425f0178d91d7f
* define HAVE_KILLPG
* upstream: sftp: Don't attempt to complete arguments for
non-existent commands
If user entered a non-existent command (e.g. because they made a
typo) there is no point in trying to complete its arguments. Skip
calling complete_match() if that's the case.
From Michal Privoznik
OpenBSD-Commit-ID: cf39c811a68cde2aeb98fc85addea4000ef6b07a
* upstream: sftp: Be a bit more clever about completions
There are commands (e.g. "get" or "put") that accept two
arguments, a local path and a remote path. However, the way
current completion is written doesn't take this distinction into
account and always completes remote or local paths.
By expanding CMD struct and "cmds" array this distinction can be
reflected and with small adjustment to completer code the correct
path can be completed.
By Michal Privoznik, ok dtucker@
OpenBSD-Commit-ID: 1396d921c4eb1befd531f5c4a8ab47e7a74b610b
* upstream: correct error value
OpenBSD-Commit-ID: 780efcbad76281f11f14b2a5ff04eb6db3dfdad4
* upstream: actually hook up restrict_websafe; the command-line flag
was never actually used. Spotted by Matthew Garrett
OpenBSD-Commit-ID: 0b363518ac4c2819dbaa3dfad4028633ab9cdff1
* upstream: Add a sshkey_check_rsa_length() call for checking the
length of an RSA key; ok markus@
OpenBSD-Commit-ID: de77cd5b11594297eda82edc594b0d32b8535134
* upstream: add a RequiredRSASize for checking RSA key length in
ssh(1). User authentication keys that fall beneath this limit will be
ignored. If a host presents a host key beneath this limit then the connection
will be terminated (unfortunately there are no fallbacks in the protocol for
host authentication).
feedback deraadt, Dmitry Belyavskiy; ok markus@
OpenBSD-Commit-ID: 430e339b2a79fa9ecc63f2837b06fdd88a7da13a
* upstream: Add RequiredRSASize for sshd(8); RSA keys that fall
beneath this limit will be ignored for user and host-based authentication.
Feedback deraadt@ ok markus@
OpenBSD-Commit-ID: 187931dfc19d51873df5930a04f2d972adf1f7f1
* upstream: better debugging for connect_next()
OpenBSD-Commit-ID: d16a307a0711499c971807f324484ed3a6036640
* upstream: sftp-server(8): add a "users-groups-by-id@openssh.com"
extension request that allows the client to obtain user/group names that
correspond to a set of uids/gids.
Will be used to make directory listings more useful and consistent
in sftp(1).
ok markus@
OpenBSD-Commit-ID: 7ebabde0bcb95ef949c4840fe89e697e30df47d3
* upstream: extend sftp-common.c:extend ls_file() to support supplied
user/group names; ok markus@
OpenBSD-Commit-ID: c70c70498b1fdcf158531117e405b6245863bfb0
* upstream: sftp client library support for
users-groups-by-id@openssh.com; ok markus@
OpenBSD-Commit-ID: ddb2f33a2da6349a9a89a8b5bcb9ca7c999394de
* upstream: use users-groups-by-id@openssh.com sftp-server extension
(when available) to fill in user/group names for directory listings.
Implement a client-side cache of see uid/gid=>user/group names. ok markus@
OpenBSD-Commit-ID: f239aeeadfa925a37ceee36ee8b256b8ccf4466e
* avoid Wuninitialized false positive in gcc-12ish
* no need for glob.h here
it also causes portability problems
* add debug on appveyor
* add sleep to pester test
* upstream: add RequiredRSASize to the list of keywords accepted by
-o; spotted by jmc@
OpenBSD-Commit-ID: fe871408cf6f9d3699afeda876f8adbac86a035e
* upstream: Fix typo. From AlexanderStohr via github PR#343.
OpenBSD-Commit-ID: a134c9b4039e48803fc6a87f955b0f4a03181497
* upstream: openssh-9.1
OpenBSD-Commit-ID: 5a467b2ee81da01a86adf1ad93b62b1728494e56
* crank versions in RPM spec files
* update release notes URL
* update .depend
* fix 9.1 compilation errors
* disable -p pester tests due to unreliability on older Windows versions
* remove extra sleep time from debugging scp pester tests
* modify -p tests to only run for Windows OS version 10 and above
* add windows specific code back into method moved from auth.c to auth2-pubkeyfile.c
* add preprocessor for WinHello
* revert preprocessor definition for winhello
* add windows preprocessor definition in key_lookup
* remove rdp block from appveyor since we are no longer debugging
* add ifdef to sftp-server.c
* make key_lookup compatible with winhello
* appveyor.yml
* increase debug of failing pester test
* add #ifdef SUPPORT_CRLF back into auth_check_principals_line method that was moved/renamed
* modify new scp.sh tests for windows
* remove in place tests from scp.sh
* remove rdp debug from appveyor
* retrigger appveyor
* change check of OS version in scp test
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
Co-authored-by: djm@openbsd.org <djm@openbsd.org>
Co-authored-by: Damien Miller <djm@mindrot.org>
Co-authored-by: Darren Tucker <dtucker@dtucker.net>
Co-authored-by: naddy@openbsd.org <naddy@openbsd.org>
Co-authored-by: dtucker@openbsd.org <dtucker@openbsd.org>
Co-authored-by: tj@openbsd.org <tj@openbsd.org>
Co-authored-by: millert@openbsd.org <millert@openbsd.org>
Co-authored-by: jmc@openbsd.org <jmc@openbsd.org>
Co-authored-by: florian@openbsd.org <florian@openbsd.org>
Co-authored-by: markus@openbsd.org <markus@openbsd.org>
Co-authored-by: Tobias Heider <me@tobhe.de>
Co-authored-by: anton@openbsd.org <anton@openbsd.org>
Co-authored-by: Tim Rice <tim@multitalents.net>
Co-authored-by: tobhe@openbsd.org <tobhe@openbsd.org>
Co-authored-by: Corinna Vinschen <vinschen@redhat.com>
Co-authored-by: Sam James <sam@gentoo.org>
Co-authored-by: jsg@openbsd.org <jsg@openbsd.org>
2022-11-02 17:06:45 +01:00
|
|
|
if (!quit && (r = sshkey_check_rsa_length(private,
|
|
|
|
options.required_rsa_size)) != 0) {
|
|
|
|
debug_fr(r, "Skipping key %s", id->filename);
|
|
|
|
sshkey_free(private);
|
|
|
|
private = NULL;
|
|
|
|
quit = 1;
|
|
|
|
}
|
2015-12-04 01:24:55 +01:00
|
|
|
if (!quit && private != NULL && id->agent_fd == -1 &&
|
2015-11-15 23:26:49 +01:00
|
|
|
!(id->key && id->isprivate))
|
|
|
|
maybe_add_key_to_agent(id->filename, private, comment,
|
|
|
|
passphrase);
|
2018-07-09 23:03:30 +02:00
|
|
|
if (i > 0)
|
|
|
|
freezero(passphrase, strlen(passphrase));
|
2015-12-10 18:08:40 +01:00
|
|
|
free(comment);
|
2015-01-14 21:05:27 +01:00
|
|
|
if (private != NULL || quit)
|
|
|
|
break;
|
2000-07-21 02:19:44 +02:00
|
|
|
}
|
2001-03-09 01:12:22 +01:00
|
|
|
return private;
|
|
|
|
}
|
|
|
|
|
2018-07-03 13:39:54 +02:00
|
|
|
static int
|
|
|
|
key_type_allowed_by_config(struct sshkey *key)
|
|
|
|
{
|
|
|
|
if (match_pattern_list(sshkey_ssh_name(key),
|
2021-01-22 03:44:58 +01:00
|
|
|
options.pubkey_accepted_algos, 0) == 1)
|
2018-07-03 13:39:54 +02:00
|
|
|
return 1;
|
|
|
|
|
|
|
|
/* RSA keys/certs might be allowed by alternate signature types */
|
|
|
|
switch (key->type) {
|
|
|
|
case KEY_RSA:
|
|
|
|
if (match_pattern_list("rsa-sha2-512",
|
2021-01-22 03:44:58 +01:00
|
|
|
options.pubkey_accepted_algos, 0) == 1)
|
2018-07-03 13:39:54 +02:00
|
|
|
return 1;
|
|
|
|
if (match_pattern_list("rsa-sha2-256",
|
2021-01-22 03:44:58 +01:00
|
|
|
options.pubkey_accepted_algos, 0) == 1)
|
2018-07-03 13:39:54 +02:00
|
|
|
return 1;
|
|
|
|
break;
|
|
|
|
case KEY_RSA_CERT:
|
|
|
|
if (match_pattern_list("rsa-sha2-512-cert-v01@openssh.com",
|
2021-01-22 03:44:58 +01:00
|
|
|
options.pubkey_accepted_algos, 0) == 1)
|
2018-07-03 13:39:54 +02:00
|
|
|
return 1;
|
|
|
|
if (match_pattern_list("rsa-sha2-256-cert-v01@openssh.com",
|
2021-01-22 03:44:58 +01:00
|
|
|
options.pubkey_accepted_algos, 0) == 1)
|
2018-07-03 13:39:54 +02:00
|
|
|
return 1;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2021-12-19 23:08:48 +01:00
|
|
|
/* obtain a list of keys from the agent */
|
|
|
|
static int
|
|
|
|
get_agent_identities(struct ssh *ssh, int *agent_fdp,
|
|
|
|
struct ssh_identitylist **idlistp)
|
|
|
|
{
|
|
|
|
int r, agent_fd;
|
|
|
|
struct ssh_identitylist *idlist;
|
|
|
|
|
|
|
|
if ((r = ssh_get_authentication_socket(&agent_fd)) != 0) {
|
|
|
|
if (r != SSH_ERR_AGENT_NOT_PRESENT)
|
|
|
|
debug_fr(r, "ssh_get_authentication_socket");
|
|
|
|
return r;
|
|
|
|
}
|
|
|
|
if ((r = ssh_agent_bind_hostkey(agent_fd, ssh->kex->initial_hostkey,
|
|
|
|
ssh->kex->session_id, ssh->kex->initial_sig, 0)) == 0)
|
|
|
|
debug_f("bound agent to hostkey");
|
|
|
|
else
|
|
|
|
debug2_fr(r, "ssh_agent_bind_hostkey");
|
|
|
|
|
|
|
|
if ((r = ssh_fetch_identitylist(agent_fd, &idlist)) != 0) {
|
|
|
|
debug_fr(r, "ssh_fetch_identitylist");
|
|
|
|
close(agent_fd);
|
|
|
|
return r;
|
|
|
|
}
|
|
|
|
/* success */
|
|
|
|
*agent_fdp = agent_fd;
|
|
|
|
*idlistp = idlist;
|
|
|
|
debug_f("agent returned %zu keys", idlist->nkeys);
|
|
|
|
return 0;
|
|
|
|
}
|
2018-07-03 13:39:54 +02:00
|
|
|
|
2003-05-14 05:46:00 +02:00
|
|
|
/*
|
|
|
|
* try keys in the following order:
|
2020-10-29 03:52:43 +01:00
|
|
|
* 1. certificates listed in the config file
|
|
|
|
* 2. other input certificates
|
2015-09-24 08:15:11 +02:00
|
|
|
* 3. agent keys that are found in the config file
|
|
|
|
* 4. other agent keys
|
|
|
|
* 5. keys that are only listed in the config file
|
2003-05-14 05:46:00 +02:00
|
|
|
*/
|
|
|
|
static void
|
2021-12-19 23:08:48 +01:00
|
|
|
pubkey_prepare(struct ssh *ssh, Authctxt *authctxt)
|
2001-03-09 01:12:22 +01:00
|
|
|
{
|
2015-01-14 21:05:27 +01:00
|
|
|
struct identity *id, *id2, *tmp;
|
|
|
|
struct idlist agent, files, *preferred;
|
|
|
|
struct sshkey *key;
|
2023-12-18 15:45:49 +01:00
|
|
|
int disallowed, agent_fd = -1, i, r, found;
|
2015-01-14 21:05:27 +01:00
|
|
|
size_t j;
|
|
|
|
struct ssh_identitylist *idlist;
|
2023-12-18 15:45:49 +01:00
|
|
|
char *cp, *ident;
|
2003-05-14 05:46:00 +02:00
|
|
|
|
|
|
|
TAILQ_INIT(&agent); /* keys from the agent */
|
|
|
|
TAILQ_INIT(&files); /* keys from the config file */
|
|
|
|
preferred = &authctxt->keys;
|
|
|
|
TAILQ_INIT(preferred); /* preferred order of keys */
|
|
|
|
|
2012-12-02 23:49:52 +01:00
|
|
|
/* list of keys stored in the filesystem and PKCS#11 */
|
2003-05-14 05:46:00 +02:00
|
|
|
for (i = 0; i < options.num_identity_files; i++) {
|
|
|
|
key = options.identity_keys[i];
|
2019-10-31 22:18:28 +01:00
|
|
|
if (key && key->cert &&
|
|
|
|
key->cert->type != SSH2_CERT_TYPE_USER) {
|
2020-10-18 13:32:01 +02:00
|
|
|
debug_f("ignoring certificate %s: not a user "
|
|
|
|
"certificate", options.identity_files[i]);
|
2019-10-31 22:18:28 +01:00
|
|
|
continue;
|
|
|
|
}
|
2019-11-12 20:33:08 +01:00
|
|
|
if (key && sshkey_is_sk(key) && options.sk_provider == NULL) {
|
2020-10-18 13:32:01 +02:00
|
|
|
debug_f("ignoring authenticator-hosted key %s as no "
|
2019-10-31 22:18:28 +01:00
|
|
|
"SecurityKeyProvider has been specified",
|
2020-10-18 13:32:01 +02:00
|
|
|
options.identity_files[i]);
|
2010-02-26 21:55:05 +01:00
|
|
|
continue;
|
2019-10-31 22:18:28 +01:00
|
|
|
}
|
2003-05-14 05:46:00 +02:00
|
|
|
options.identity_keys[i] = NULL;
|
2006-03-26 05:19:21 +02:00
|
|
|
id = xcalloc(1, sizeof(*id));
|
2015-12-04 01:24:55 +01:00
|
|
|
id->agent_fd = -1;
|
2003-05-14 05:46:00 +02:00
|
|
|
id->key = key;
|
|
|
|
id->filename = xstrdup(options.identity_files[i]);
|
2013-04-05 02:13:08 +02:00
|
|
|
id->userprovided = options.identity_file_userprovided[i];
|
2003-05-14 05:46:00 +02:00
|
|
|
TAILQ_INSERT_TAIL(&files, id, next);
|
|
|
|
}
|
2015-09-24 08:15:11 +02:00
|
|
|
/* list of certificates specified by user */
|
|
|
|
for (i = 0; i < options.num_certificate_files; i++) {
|
|
|
|
key = options.certificates[i];
|
2018-07-09 23:03:30 +02:00
|
|
|
if (!sshkey_is_cert(key) || key->cert == NULL ||
|
2019-10-31 22:18:28 +01:00
|
|
|
key->cert->type != SSH2_CERT_TYPE_USER) {
|
2020-10-18 13:32:01 +02:00
|
|
|
debug_f("ignoring certificate %s: not a user "
|
|
|
|
"certificate", options.identity_files[i]);
|
2015-09-24 08:15:11 +02:00
|
|
|
continue;
|
2019-10-31 22:18:28 +01:00
|
|
|
}
|
2019-11-12 20:33:08 +01:00
|
|
|
if (key && sshkey_is_sk(key) && options.sk_provider == NULL) {
|
2020-10-18 13:32:01 +02:00
|
|
|
debug_f("ignoring authenticator-hosted key "
|
2020-02-06 23:30:54 +01:00
|
|
|
"certificate %s as no "
|
2019-10-31 22:18:28 +01:00
|
|
|
"SecurityKeyProvider has been specified",
|
2020-10-18 13:32:01 +02:00
|
|
|
options.identity_files[i]);
|
2019-10-31 22:18:28 +01:00
|
|
|
continue;
|
|
|
|
}
|
2015-09-24 08:15:11 +02:00
|
|
|
id = xcalloc(1, sizeof(*id));
|
2015-12-04 01:24:55 +01:00
|
|
|
id->agent_fd = -1;
|
2015-09-24 08:15:11 +02:00
|
|
|
id->key = key;
|
|
|
|
id->filename = xstrdup(options.certificate_files[i]);
|
|
|
|
id->userprovided = options.certificate_file_userprovided[i];
|
|
|
|
TAILQ_INSERT_TAIL(preferred, id, next);
|
|
|
|
}
|
2003-05-14 05:46:00 +02:00
|
|
|
/* list of keys supported by the agent */
|
2021-12-19 23:08:48 +01:00
|
|
|
if ((r = get_agent_identities(ssh, &agent_fd, &idlist)) == 0) {
|
2015-01-14 21:05:27 +01:00
|
|
|
for (j = 0; j < idlist->nkeys; j++) {
|
Merge 9.1 (#626)
* upstream: fix poll() spin when a channel's output fd closes without
data in the channel buffer. Introduce more exact packing of channel fds into
the pollfd array. fixes bz3405 and bz3411; ok deraadt@ markus@
OpenBSD-Commit-ID: 06740737849c9047785622ad5d472cb6a3907d10
* upstream: select post-quantum KEX
sntrup761x25519-sha512@openssh.com as the default; ok markus@
OpenBSD-Commit-ID: f02d99cbfce22dffec2e2ab1b60905fbddf48fb9
* upstream: add support for the "corp-data" protocol extension to
allow server-side copies to be performed without having to go via the client.
Patch by Mike Frysinger, ok dtucker@
OpenBSD-Commit-ID: 00aa510940fedd66dab1843b58682de4eb7156d5
* upstream: add a sftp client "cp" command that supports server-side
copying of files. Useful for this task and for testing the copy-data
extension. Patch from Mike Frysinger; ok dtucker@
OpenBSD-Commit-ID: 1bb1b950af0d49f0d5425b1f267e197aa1b57444
* depend
* Skip slow tests on (very) slow test targets.
* Set Makefile SHELL as determined by configure.
This should improve compatibility for users with non-POSIX shells. If
using Makefile.in directly (eg make -f Makefile.in distprep) then SHELL
will need to be specified on the command line (along with MANFMT in that
particular case). ok djm@
* Use bash or ksh if available for SH in Makefile.
* Increase test timeout to allow slow VMs to finish
* Only run regression tests on slow VMs.
* Only return events from ppoll that were requested.
If the underlying system's select() returns bits that were not in the
request set, our ppoll() implementation can return revents for events
not requested, which can apparently cause a hang. Only return revents
for activity in the requested event set. bz#3416, analysis and fix by
yaroslav.kuzmin at vmssoftware com, ok djm@
* Specify TEST_SHELL=bash on AIX.
The system shells cause the agent-restrict test to fail due to some
quoting so explicitly specify bash until we can get configure to
autmatically work around that.
* Disable security key on fbsd6 test host.
* upstream: man pages: add missing commas between subordinate and
main clauses
jmc@ dislikes a comma before "then" in a conditional, so leave those
untouched.
ok jmc@
OpenBSD-Commit-ID: 9520801729bebcb3c9fe43ad7f9776ab4dd05ea3
* upstream: ssh: document sntrup761x25519-sha512@openssh.com as
default KEX
OpenBSD-Commit-ID: 12545bfa10bcbf552d04d9d9520d0f4e98b0e171
* upstream: openssh-9.0
OpenBSD-Commit-ID: 0dfb461188f4513ec024c1534da8c1ce14c20b64
* update version numbers for release
* update build-aux files to match autoconf-2.71
i.e. config.guess, config.sub and install-sh
* Revert "update build-aux files to match autoconf-2.71"
This reverts commit 0a8ca39fac6ad19096b6c263436f8b2dd51606f2.
It turns out that the checked-in copies of these files are actually newer
than autoconf-2.71's copies, so this was effectively a downgrade.
Spotted by Bo Anderson via github
* upstream: two defensive changes from Tobias Stoeckmann via GHPR287
enforce stricter invarient for sshbuf_set_parent() - never allow
a buffer to have a previously-set parent changed.
In sshbuf_reset(), if the reallocation fails, then zero the entire
buffer and not the (potentially smaller) default initial alloc size.
OpenBSD-Commit-ID: 14583203aa5d50ad38d2e209ae10abaf8955e6a9
* upstream: Note that curve25519-sha256 was later published in
RFC8731. ok djm@
OpenBSD-Commit-ID: 2ac2b5d642d4cf5918eaec8653cad9a4460b2743
* upstream: clear io_want/io_ready flags at start of poll() cycle;
avoids plausible spin during rekeying if channel io_want flags are reused
across cycles. ok markus@ deraadt@
OpenBSD-Commit-ID: 91034f855b7c73cd2591657c49ac30f10322b967
* Retire fbsd6 test VM.
It's long since out of support, relatively slow (it's i686) and the
compiler has trouble with PIE.
* Resync moduli.5 with upstream.
1.18: remove duplicate publication year; carsten dot kunze at arcor dot de
1.19: ssh-keygen's -G/-T have been replaced with -M generate/screen.
* upstream: Correct path for system known hosts file in description
of IgnoreUserKnownHosts. Patch from Martin Vahlensieck via tech@
OpenBSD-Commit-ID: 9b7784f054fa5aa4d63cb36bd563889477127215
* upstream: list the correct version number
for when usage of the sftp protocol became default and fix a typo
from ed maste
OpenBSD-Commit-ID: 24e1795ed2283fdeacf16413c2f07503bcdebb31
* upstream: Import regenerated moduli
OpenBSD-Commit-ID: f9a0726d957cf10692a231996a1f34e7f9cdfeb0
* upstream: Try to continue running local I/O for channels in state
OPEN during SSH transport rekeying. The most visible benefit is that it
should make ~-escapes work in the client (e.g. to exit) if the connection
happened to have stalled during a rekey event. Based work by and ok dtucker@
OpenBSD-Commit-ID: a66e8f254e92edd4ce09c9f750883ec8f1ea5f45
* upstream: Import regenerated moduli
OpenBSD-Commit-ID: f9a0726d957cf10692a231996a1f34e7f9cdfeb0
* upstream: regression test for sftp cp command
OpenBSD-Regress-ID: c96bea9edde3a384b254785e7f9b2b24a81cdf82
* upstream: Simplify forward-control test.
Since we no longer need to support SSH1 we don't need to run shell
commands on the other end of the connection and can use ssh -N instead.
This also makes the test less racy.
OpenBSD-Regress-ID: 32e94ce272820cc398f30b848b2b0f080d10302c
* upstream: Use ssh -f and ControlPersist ..
to start up test forwards and ssh -O stop to shut them down intead of
sleep loops. This speeds up the test by an order of magnitude.
OpenBSD-Regress-ID: eb3db5f805100919b092a3b2579c611fba3e83e7
* upstream: It looks like we can't completely avoid
waiting for processes to exit so retrieve the pid via controlmaster and
use that.
OpenBSD-Regress-ID: 8246f00f22b14e49d2ff1744c94897ead33d457b
* Cache timezone data in capsicum sandbox.
From emaste at freebsd.org, originally part of FreeBSD commit r339216
/ fc3c19a9 with autoconf bits added by me.
* Include stdlib.h for free() prototype.
... which is used inside the CUSTOM_SYS_AUTH_GET_LASTLOGIN_MSG block.
* Update OpenSSL and LibreSSL versions in tests.
* Add debian-riscv64 test target.
* upstream: Avoid an unnecessary xstrdup in rm_env() when matching
patterns. Since match_pattern() doesn't modify its arguments (they are
const), there is no need to make an extra copy of the strings in
options->send_env. From Martin Vahlensieck
OpenBSD-Commit-ID: 2c9db31e3f4d3403b49642c64ee048b2a0a39351
* upstream: Add missing includes of stdlib.h and stdint.h. We need
stdlib.h for malloc(3) and stdint.h for SIZE_MAX. Unlike the other xmss
files, ssh-xmss.c does not include xmss_commons.h so ssh-xmss.c must include
those headers itself. From Martin Vahlensieck
OpenBSD-Commit-ID: 70e28a9818cee3da1be2ef6503d4b396dd421e6b
* upstream: Remove unnecessary includes: openssl/hmac.h and
openssl/evp.h. From Martin Vahlensieck.
OpenBSD-Commit-ID: a6debb5fb0c8a44e43e8d5ca7cc70ad2f3ea31c3
* upstream: Check sshauthopt_new() for NULL. bz#3425, from
tessgauthier at microsoft.com. ok djm@
OpenBSD-Commit-ID: af0315bc3e44aa406daa7e0ae7c2d719a974483f
* upstream: Add authfd path to debug output. ok markus@
OpenBSD-Commit-ID: f735a17d1a6f2bee63bfc609d76ef8db8c090890
* upstream: avoid printing hash algorithm twice; from lucas AT sexy.is
OpenBSD-Commit-ID: 9d24671e10a84141b7c504396cabad600e47a941
* upstream: fix memleak on session-bind path; from Pedro Martelletto, ok
dtucker@
OpenBSD-Commit-ID: e85899a26ba402b4c0717b531317e8fc258f0a7e
* upstream: Don't leak SK device. Patch from Pedro Martelletto via
github PR#316. ok djm@
OpenBSD-Commit-ID: 17d11327545022e727d95fd08b213171c5a4585d
* upstream: mention that the helpers are used by ssh(1), ssh-agent(1)
and ssh-keygen(1). Previously only ssh(1) was mentioned. From Pedro
Martelletto
OpenBSD-Commit-ID: 30f880f989d4b329589c1c404315685960a5f153
* Remove now-empty int32_minmax.inc.
* Only run tests when source files change.
Also run tests on changes to V_9_0 branch.
* Add Mac OS X 12 test target.
* upstream: be stricter in which characters will be accepted in
specifying a mask length; allow only 0-9. From khaleesicodes via GHPR#278; ok
dtucker@
OpenBSD-Commit-ID: e267746c047ea86665cdeccef795a8a56082eeb2
* upstream: fix some integer overflows in sieve_large() that show up when
trying to generate modp groups > 16k bits. Reported via GHPR#306 by Bertram
Felgenhauer, but fixed in a different way. feedback/ok tb@
OpenBSD-Commit-ID: 81cbc6dd3a21c57bd6fadea10e44afe37bca558e
* upstream: remove an obsolete rsa1 format example from an example;
from megan batty
ok djm
OpenBSD-Commit-ID: db2c89879c29bf083df996bd830abfb1e70d62bf
* upstream: Add FIDO AUTHENTICATOR section and explain a bit how FIDO
works. The wording came mostly from the 8.2 OpenSSH release notes, addapted
to fit the man page. Then move the -O bits into the new section as is already
done for CERTIFICATES and MODULI GENERATION. Finally we can explain the
trade-offs of resident keys. While here, consistently refer to the FIDO
thingies as "FIDO authenticators", not "FIDO tokens".
input & OK jmc, naddy
OpenBSD-Commit-ID: dd98748d7644df048f78dcf793b3b63db9ab1d25
* upstream: make sure stdout is non-blocking; ok djm@
OpenBSD-Commit-ID: 64940fffbd1b882eda2d7c8c7a43c79368309c0d
* upstream: mux.c: mark argument as const; from Martin Vahlensieck
OpenBSD-Commit-ID: 69a1a93a55986c7c2ad9f733c093b46a47184341
* upstream: channel_new no longer frees remote_name. So update the
comment accordingly. As remote_name is not modified, it can be const as
well. From Martin Vahlensieck
OpenBSD-Commit-ID: e4e10dc8dc9f40c166ea5a8e991942bedc75a76a
* upstream: sshkey_unshield_private() contains a exact duplicate of
the code in private2_check_padding(). Pull private2_check_padding() up so the
code can be reused. From Martin Vahlensieck, ok deraadt@
OpenBSD-Commit-ID: 876884c3f0e62e8fd8d1594bab06900f971c9c85
* Add ubsan minimal testcase on OpenBSD.
As suggested by djm@.
* Note that, for now, we need variadic macros.
* Also retest OpenBSD upstream on .yml changes.
* upstream: When performing operations that glob(3) a remote path, ensure
that the implicit working directory used to construct that path escapes
glob(3) characters.
This prevents glob characters from being processed in places they
shouldn't, e.g. "cd /tmp/a*/", "get *.txt" should have the get operation
treat the path "/tmp/a*" literally and not attempt to expand it.
Reported by Lusia Kundel; ok markus@
OpenBSD-Commit-ID: 4f647f58482cbad3d58b1eab7f6a1691433deeef
* Remove duplicate bcrypt_pbkdf.o from Makefile
bcrypt_pbkdf.o is duplicated in the openbsd-compat Makefile's object
file list.
* upstream: improve error message when 'ssh-keygen -Y sign' is unable to
load a private key; bz3429, reported by Adam Szkoda ok dtucker@
OpenBSD-Commit-ID: bb57b285e67bea536ef81b1055467be2fc380e74
* upstream: Allow existing -U (use agent) flag to work with "-Y sign"
operations, where it will be interpreted to require that the private keys is
hosted in an agent; bz3429, suggested by Adam Szkoda; ok dtucker@
OpenBSD-Commit-ID: a7bc69873b99c32c42c7628ed9ea91565ba08c2f
* upstream: Remove errant apostrophe. From haruyama at queen-ml org.
OpenBSD-Commit-ID: dc6b294567cb84b384ad6ced9ca469f2bbf0bd10
* upstream: arrange for scp, when in sftp mode, to not ftruncate(3) files
early
previous behavious of unconditionally truncating the destination file
would cause "scp ~/foo localhost:" and "scp localhost:foo ~/" to
delete all the contents of their destination.
spotted by solene@ sthen@, also bz3431; ok dtucker@
OpenBSD-Commit-ID: ca39fdd39e0ec1466b9666f15cbcfddea6aaa179
* upstream: fix in-place copies; r1.163 incorrectly skipped truncation in
all cases, not just at the start of a transfer. This could cause overwrites
of larger files to leave junk at the end. Spotted by tb@
OpenBSD-Commit-ID: b189f19cd68119548c8e24e39c79f61e115bf92c
* upstream: Only run agent-ptrace.sh if gdb is available as all
architectures do not ship with gdb.
OpenBSD-Regress-ID: ec53e928803e6b87f9ac142d38888ca79a45348d
* upstream: regress test for in-place transfers and clobbering larger
files with smaller ones; would have caught last regression in scp(1)
OpenBSD-Regress-ID: 19de4e88dd3a4f7e5c1618c9be3c32415bd93bc2
* configure.ac: Add missing AC_DEFINE for caph_cache_tzdata test causing
HAVE_CAPH_CACHE_TZDATA to be missing from config.h.in.
Spotted by Bryan Drewery
* upstream: make SSHBUF_DBG/SSHBUF_TELL (off by default and only enabled
via #define) dump to stderr rather than stdout
OpenBSD-Commit-ID: 10298513ee32db8390aecb0397d782d68cb14318
* upstream: revert previous; it was broken (spotted by Theo)
OpenBSD-Commit-ID: 457c79afaca2f89ec2606405c1059b98b30d8b0d
* upstream: Note that ProxyJump also accepts the same tokens as
ProxyCommand. From pallxk via github PR#305.
OpenBSD-Commit-ID: 7115ac351b129205f1f1ffa6bbfd62abd76be7c5
* upstream: Avoid kill with -1 argument. The out_ctx label can be
reached before fork has been called. If this happens, then kill -1 would be
called, sending SIGTERM to all processes reachable by the current process.
From tobias@ and c3h2_ctf via github PR#286, ok djm@
OpenBSD-Commit-ID: 6277af1207d81202f5daffdccfeeaed4c763b1a8
* upstream: f sshpkt functions fail, then password is not cleared
with freezero. Unconditionally call freezero to guarantee that password is
removed from RAM.
From tobias@ and c3h2_ctf via github PR#286, ok djm@
OpenBSD-Commit-ID: 6b093619c9515328e25b0f8093779c52402c89cd
* upstream: refactor authorized_keys/principals handling
remove "struct ssh *" from arguments - this was only used to pass the
remote host/address. These can be passed in instead and the resulting
code is less tightly coupled to ssh_api.[ch]
ok dtucker@
OpenBSD-Commit-ID: 9d4373d013edc4cc4b5c21a599e1837ac31dda0d
* upstream: split the low-level file handling functions out from
auth2-pubkey.c
Put them in a new auth2-pubkeyfile.c to make it easier to refer to them
(e.g. in unit/fuzz tests) without having to refer to everything else
pubkey auth brings in.
ok dtucker@
OpenBSD-Commit-ID: 3fdca2c61ad97dc1b8d4a7346816f83dc4ce2217
* fuzzer for authorized_keys parsing
mostly redundant to authopt_fuzz, but it's sensitive code so IMO it
makes sense to test this layer too
* Test against LibreSSL 3.5.3.
* Test against OpenSSL 1.1.1o and 3.0.3.
* fix some bugs in the fuzzer
* upstream: keywords ref ssh_config.5;
from caspar schutijser
OpenBSD-Commit-ID: f146a19d7d5c9374c3b9c520da43b2732d7d1a4e
* upstream: ssh-keygen: implement "verify-required" certificate option.
This was already documented when support for user-verified FIDO
keys was added, but the ssh-keygen(1) code was missing.
ok djm@
OpenBSD-Commit-ID: f660f973391b593fea4b7b25913c9a15c3eb8a06
* upstream: ssh-keygen -A: do not generate DSA keys by default.
Based on github PR#303 from jsegitz with man page text from jmc@, ok markus@
djm@
OpenBSD-Commit-ID: 5c4c57bdd7063ff03381cfb6696659dd3f9f5b9f
* upstream: Add period at end of "not known by any other names"
message. github PR#320 from jschauma, ok djm@
OpenBSD-Commit-ID: bd60809803c4bfd3ebb7c5c4d918b10e275266f2
* upstream: Add missing *-sk types to ssh-keyscan manpage. From
skazi0 via github PR#294.
OpenBSD-Commit-ID: fda2c869cdb871f3c90a89fb3f985370bb5d25c0
* upstream: Make SetEnv directives first-match-wins in both
sshd_config and sshd_config; previously if the same name was reused then the
last would win (which is the opposite to how the config is supposed to work).
While there, make the ssh_config parsing more like sshd_config.
bz3438, ok dtucker
OpenBSD-Commit-ID: 797909c1e0262c0d00e09280459d7ab00f18273b
* upstream: test setenv in both client and server, test first-match-wins
too
OpenBSD-Regress-ID: 4c8804f9db38a02db480b9923317457b377fe34b
* upstream: move auth_openprincipals() and auth_openkeyfile() over to
auth2-pubkeyfile.c too; they make more sense there.
OpenBSD-Commit-ID: 9970d99f900e1117fdaab13e9e910a621b7c60ee
* upstream: make sure that UseDNS hostname lookup happens in the monitor
and not in the pledge(2)'d unprivileged process; fixes regression caused by
recent refactoring spotted by henning@
OpenBSD-Commit-ID: a089870b95101cd8881a2dff65b2f1627d13e88d
* fix possible NULL deref when built without FIDO
Analysis/fix from kircher in bz3443; ok dtucker@
* automatically enable built-in FIDO support
If libfido2 is found and usable, then enable the built-in
security key support unless --without-security-key-builtin
was requested.
ok dtucker@
* upstream: Log an error if pipe() fails while accepting a
connection. bz#3447, from vincent-openssh at vinc17 net, ok djm@
OpenBSD-Commit-ID: 9d59f19872b94900a5c79da2d57850241ac5df94
* upstream: Don't attempt to fprintf a null identity comment. From
Martin Vahlensieck via tech@.
OpenBSD-Commit-ID: 4c54d20a8e8e4e9912c38a7b4ef5bfc5ca2e05c2
* upstream: Make sure not to fclose() the same fd twice in case of an
error.
ok dtucker@
OpenBSD-Commit-ID: e384c4e05d5521e7866b3d53ca59acd2a86eef99
* upstream: make it clear that RekeyLimit applies to both transmitted
and received data. GHPR#328 from Jan Pazdziora
OpenBSD-Commit-ID: d180a905fec9ff418a75c07bb96ea41c9308c3f9
* request 1.1x API compatibility for OpenSSL >=3.x
idea/patch from Pedro Martelletto via GHPR#322; ok dtucker@
* fix broken case statement in previous
* Disable SK support if FIDO libs not found.
* Zero out LIBFIDO2 when SK support not usable.
Prevents us from trying to link them into ssh-sk-helper and failing to
build.
* upstream: Don't leak the strings allocated by order_hostkeyalgs()
and list_hostkey_types() that are passed to compat_pkalg_proposal(). Part of
github PR#324 from ZoltanFridrich, ok djm@
OpenBSD-Commit-ID: b2f6e5f60f2bba293b831654328a8a0035ef4a1b
* upstream: Roll back previous KEX changes as they aren't safe until
compat_pkalg_proposal and friends always allocate their returned strings.
Reported by Qualys.
OpenBSD-Commit-ID: 1c7a88a0d5033f42f88ab9bec58ef1cf72c81ad0
* upstream: allow arguments to sftp -D option, e.g. sftp -D
"/usr/libexec/sftp-server -el debug3"
ok markus@
OpenBSD-Commit-ID: 5a002b9f3a7aef2731fc0ffa9c921cf15f38ecce
* Update OpenSSL tests to the most recent releases.
* upstream: reflect the update to -D arg name in usage();
OpenBSD-Commit-ID: abdcde4f92b1ef094ae44210ee99d3b0155aad9c
* upstream: ignore SIGPIPE earlier in main(), specifically before
muxclient() which performs operations that could cause one; Reported by Noam
Lewis via bz3454, ok dtucker@
OpenBSD-Commit-ID: 63d8e13276869eebac6d7a05d5a96307f9026e47
* upstream: Always return allocated strings from the kex filtering so
that we can free them later. Fix one leak in compat_kex_proposal. Based on
github PR#324 from ZoltanFridrich with some simplications by me. ok djm@
OpenBSD-Commit-ID: 9171616da3307612d0ede086fd511142f91246e4
* upstream: Don't leak the strings allocated by order_hostkeyalgs()
and list_hostkey_types() that are passed to compat_pkalg_proposal(). Part of
github PR#324 from ZoltanFridrich, ok djm@
This is a roll-forward of the previous rollback now that the required
changes in compat.c have been done.
OpenBSD-Commit-ID: c7cd93730b3b9f53cdad3ae32462922834ef73eb
* upstream: bump up loglevel from debug to info when unable to open
authorized keys/principals file for errno != ENOENT; bz2042 ok dtucker
OpenBSD-Commit-ID: e79aa550d91ade6a80f081bda689da24c086d66b
* Skip select+rlimit check if sandboxing is disabled
It's not needed in that case, and the test can fail when being built
with some compiler memory sanitizer flags. bz#3441
* upstream: use consistent field names (s/char/byte)
in format description
OpenBSD-Commit-ID: 3de33572733ee7fcfd7db33d37db23d2280254f0
* upstream: Remove leftover line.
Remove extra line leftover from merge conflict. ok djm@
OpenBSD-Commit-ID: 460e2290875d7ae64971a7e669c244b1d1c0ae2e
* Move checks for pollfd.fd and nfds_t.
Move the checks for struct pollfd.fd and nfds_t to before the sandboxing
checks. This groups all the sandbox checks together so we can skip them
all when sandboxing is disabled.
* Skip all rlimit tests when sandboxing disabled.
The rlimit tests can hang when being run with some compiler sanitizers
so skip all of them if sandbox=no.
* Add clang sanitizer tests.
* upstream: Add TEST_REGRESS_CACHE_DIR.
If set, it is used to cache regress test names that have succeeded and
skip those on a re-run.
OpenBSD-Regress-ID: a7570dd29a58df59f2cca647c3c2ec989b49f247
* Move sanitizer logs into regress for collection.
* Add GCC address sanitizer build/test.
* Update sanitizer test targets:
- remove clang-sanitize-memory for now. It takes so long that the test
times out.
- add gcc sanitize-address and sanitize-undefined test targets.
* Test against openssl-3.0.5.
* Move unset to before we set anything.
* Refuse to use OpenSSL 3.0.4 due to potential RCE.
OpenSSL has a potential RCE in its RSA implementation (CVE-2022-2274)
so refuse to use that specific version.
* Capture stderr output from configure.
* Only refuse to use OpenSSL 3.0.4 on x86_64.
The potential RCE only impacts x86_64, so only refuse to use it if we're
targetting a potentially impacted architecture. ok djm@
* Remove special casing of crypt().
Configure goes to some lengths to pick crypt() from either libcrypt
or OpenSSL's libcrypto because they can more or less featureful (eg
supporting md5-style passwords).
OpenSSL removed its crypt() interface in 2002:
https://github.com/openssl/openssl/commit/69deec58 so these hijinks
should no longer be necessary. This also only links sshd with libcrypt
which is the only thing that needs it. ok djm@
* Clarify README.md text.
Clarify the text about the implications of building without OpenSSL, and
prefix the "configure --help" example command with a "./" so it's likely
to work as-is in more shells. From bz#3461.
* Split README.platform into its own line.
README.platform has general platform-specific information, having it
following text about FIDO2 on the same line could imply that it only
has information about FIDO2.
* Return ERANGE from getcwd() if buffer size is 1.
If getcwd() is supplied a buffer size of exactly 1 and a path of "/", it
could result in a nul byte being written out of array bounds. POSIX says
it should return ERANGE if the path will not fit in the available buffer
(with terminating nul). 1 byte cannot fit any possible path with its nul,
so immediately return ERANGE in that case.
OpenSSH never uses getcwd() with this buffer size, and all current
(and even quite old) platforms that we are currently known to work
on have a native getcwd() so this code is not used on those anyway.
Reported by Qualys, ok djm@
* Remove unintended changes.
I inadvertently included a couple of local changes with the OpenSSL
3.0.4 change. Revert, anything that should be there will be committed
separately.
* Add AUDIT_ARCH_PPC to supported seccomp arches.
Patch from dries.deschout at dodeco.eu.
* Rename bbone test target to ARM.
* Move vmshutdown to first step.
If a previous run on a physical runner has failed to clean up, the next
run will fail because it'll try to check out the code to a broken
directory mount. Make cleanup the first step.
* upstream: pull passphrase reading and confirmation into a separate
function so it can be used for FIDO2 PINs; no functional change
OpenBSD-Commit-ID: bf34f76b8283cc1d3f54633e0d4f13613d87bb2f
* upstream: when enrolling a resident key on a security token, check
if a credential with matching application and user ID strings already exists.
if so, prompt the user for confirmation before overwriting the credential.
patch from Pedro Martelletto via GHPR329
NB. cranks SSH_SK_VERSION_MAJOR, so any third-party FIDO middleware
implementations will need to adjust
OpenBSD-Commit-ID: e45e9f1bf2b2f32d9850669e7a8dbd64acc5fca4
* upstream: sk-usbhid: preserve error code returned by key_lookup()
it conveys useful information, such as the supplied pin being wrong.
Part of GHPR329 from Pedro Martelletto
OpenBSD-Commit-ID: c0647eb9290f793add363d81378439b273756c1b
* upstream: ssh-keygen: fix touch prompt, pin retries;
part of GHPR329 from Pedro Martelletto
OpenBSD-Commit-ID: 75d1005bd2ef8f29fa834c90d2684e73556fffe8
* crank SSH_SK_VERSION_MAJOR in sk-dummy.so
* Skip scp3 test if there's no scp on remote path.
scp -3 ends up using the scp that's in the remote path and will fail if
one is not available. Based on a patch from rapier at psc.edu.
* Convert "have_prog" function into "which".
"which" and its behaviour is not standardized, so convert the existing
have_prog function into "which" so we can rely on it being available
and what its semantics are. Add a have_prog wrapper that maintains the
existing behaviour.
* upstream: Test TEST_SSH_ELAPSED_TIMES for empty string not
executable. No-op on most platforms but should prevent warnings in -portable
on systems that don't have 'date %s'.
OpenBSD-Regress-ID: e39d79867b8065e33d0c5926fa1a31f85659d2a4
* upstream: Restore missing "!" in TEST_SSH_ELAPSED_TIMES test.
OpenBSD-Regress-ID: 38783f9676ec348c5a792caecee9a16e354b37b0
* Remove workarounds for OpenSSL missing AES-GCM.
We have some compatibility hacks that were added to support OpenSSL
versions that do not support AES GCM mode. Since that time, however,
the minimum OpenSSL version that we support has moved to 1.0.1 which
*does* have GCM, so this is no longer needed. ok djm@
* Remove workarounds for OpenSSL missing AES-CTR.
We have some compatibility hacks that were added to support OpenSSL
versions that do not support AES CTR mode. Since that time, however,
the minimum OpenSSL version that we support has moved to 1.0.1 which
*does* have CTR, so this is no longer needed. ok djm@
* Do not link scp, sftp and sftp-server w/ zlib.
Some of our binaries (eg sftp, sftp-server, scp) do not interact with
the channels code and thus do use libraries such as zlib and libcrypto
although they are linked with them. This adds a CHANNELLIBS and starts
by moving zlib into it, which means the aformentioned binaries are no
longer linked against zlib. ok djm@
* Group libcrypto and PRNGD checks together.
They're related more than the libcrypt or libiaf checks which are
currently between them. ok djm@
* Remove seed_rng calls from scp, sftp, sftp-server.
These binaries don't use OpenSSL's random functions. The next step
will be to stop linking them against libcrypto. ok djm@
* Move libcrypto into CHANNELLIBS.
This will result in sftp, sftp-server and scp no longer being linked
against libcrypto. ok djm@
* Move stale-configure check as early as possible.
We added a check in Makefile to catch the case where configure needs to
be rebuilt, however this did not happen until a build was attempted in
which case all of the work done by configure was wasted. Move this check
to the start of configure to catch it as early as possible. ok djm@
* Remove deprecated MacOS 10.15 runners.
* upstream: avoid double-free in error path introduced in r1.70; report
and fix based on GHPR#332 by v-rzh ok dtucker@
OpenBSD-Commit-ID: 3d21aa127b1f37cfc5bdc21461db369a663a951f
* Include CHANNEL and FIDO2 libs in configure output
* Factor out getrnd() and rename to getentropy().
Factor out the arc4random seeding into its own file and change the
interface to match getentropy. Use native getentropy if available.
This will make it easier to resync OpenBSD changes to arc4random.
Prompted by bz#3467, ok djm@.
* compat code for fido_dev_is_winhello()
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* check_sk_options: add temporary WinHello workaround
Up to libfido 1.10.0, WinHello advertises "clientPin" rather
than "uv" capability. This is fixed in 1.11.0. For the time
being, workaround it here.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* sk_sign: set FIDO2 uv attribute explicitely for WinHello
WinHello via libfido2 performs user verification by default.
However, if we stick to that, there's no way to differentiate
between keys created with or without "-O verify-required".
Set FIDO2 uv attribute explicitely to FIDO_OPT_FALSE, then check
if user verification has been requested.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* upstream: don't prompt for FIDO passphrase before attempting to enroll
the credential, just let the enroll operating fail and we'll attempt to get a
PIN anyway. Might avoid some unneccessary PIN prompts.
Part of GHPR#302 from Corinna Vinschen; ok dtucker@
OpenBSD-Commit-ID: bd5342ffc353ee37d39617906867c305564d1ce2
* Give unused param a name.
Fixes builds on platforms that do have fido2 but don't have
fido_dev_is_winhello.
* Actually put HAVE_STDINT_H around the stdint.h.
* Rename our getentropy to prevent possible loops.
Since arc4random seeds from getentropy, and we use OpenSSL for that
if enabled, there's the possibility that if we build on a system that
does not have getentropy then run on a system that does have it, then
OpenSSL could end up calling our getentropy and getting stuck in a loop.
Pointed out by deraadt@, ok djm@
* Test hostbased auth on github runners.
* fix SANDBOX_SECCOMP_FILTER_DEBUG
* Fix conditional for running hostbased tests.
* upstream: allow certificate validity intervals, sshsig verification
times and authorized_keys expiry-time options to accept dates in the UTC time
zone in addition to the default of interpreting them in the system time zone.
YYYYMMDD and YYMMDDHHMM[SS] dates/times will be interpreted as UTC if
suffixed with a 'Z' character.
Also allow certificate validity intervals to be specified in raw
seconds-since-epoch as hex value, e.g. -V 0x1234:0x4567890. This
is intended for use by regress tests and other tools that call
ssh-keygen as part of a CA workflow.
bz3468 ok dtucker
OpenBSD-Commit-ID: 454db1cdffa9fa346aea5211223a2ce0588dfe13
* upstream: add some tests for parse_absolute_time(), including cases
where it is forced to the UTC timezone. bz3468 ok dtucker
OpenBSD-Regress-ID: ea07ca31c2f3847a38df028ca632763ae44e8759
* Skip hostbased during Valgrind tests.
Valgrind doesn't let ssh exec ssh-keysign (because it's setuid) so skip
it during the Valgrind based tests.
See https://bugs.kde.org/show_bug.cgi?id=119404 for a discussion of this
(ironically there the problematic binary was ssh(1) back when it could
still be setuid).
* Rerun tests if any .github config file changes.
* Add a timegm implementation from Heimdal via Samba.
Fixes build on (at least Solaris 10).
* Replace deprecated ubuntu-18.04 runners with 22.04
* upstream: sftp-server: support home-directory request
Add support to the sftp-server for the home-directory extension defined
in draft-ietf-secsh-filexfer-extensions-00. This overlaps a bit with the
existing expand-path@openssh.com, but uses a more official protocol name,
and so is a bit more likely to be implemented by non-OpenSSH clients.
From Mike Frysinger, ok dtucker@
OpenBSD-Commit-ID: bfc580d05cc0c817831ae7ecbac4a481c23566ab
* fido_dev_is_winhello: return 0, not "false"
"false" is not used anywhere in OpenSSH, so return 0 like
everywhere else.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* Revert "check_sk_options: add temporary WinHello workaround"
Cygwin now comes with libfido2 1.11.0, so this workaround
isn't required anymore.
This reverts commit 242c044ab111a37aad3b0775727c36a4c5f0102c.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
* upstream: use .Cm for "sign"; from josiah frentsos
OpenBSD-Commit-ID: 7f80a53d54857ac6ae49ea6ad93c5bd12231d1e4
* upstream: add an extra flag to sk_probe() to indicate whether we're
probing for a FIDO resident key or not. Unused here, but will make like
easier for portable
OpenBSD-Commit-ID: 432c8ff70e270378df9dbceb9bdeaa5b43b5a832
* on Cygwin, prefer WinHello FIDO device
If no FIDO device was explictly specified, then prefer the
windows://hello FIDO device. An exception to this is when
probing resident FIDO keys, in which case hardware FIDO
devices are preferred.
* Check for perms to run agent-getpeereid test.
Ubuntu 22.04 defaults to private home dirs which prevents "nobody"
running ssh-add during the agent-getpeereid test. Check for this and
add the necessary permissions.
* upstream: double free() in error path; from Eusgor via GHPR333
OpenBSD-Commit-ID: 39f35e16ba878c8d02b4d01d8826d9b321be26d4
* Add Cygwin (on windows-2019) test target.
In addition to installing the requisite Cygwin packages, we also need to
explicitly invoke "sh" for steps that run other scripts since the runner
environment doesn't understand #! paths.
* Add a bit more debug output.
* Fix cygwin conditional steps.
* upstream: Strictly enforce the maximum allowed SSH2 banner size in
ssh-keyscan and prevent a one-byte buffer overflow. Patch from Qualys, ok
djm@
OpenBSD-Commit-ID: 6ae664f9f4db6e8a0589425f74cd0bbf3aeef4e4
* upstream: remove incorrect check that can break enrolling a
resident key (introduced in r1.40)
OpenBSD-Commit-ID: 4cab364d518470e29e624af3d3f9ffa9c92b6f01
* upstream: attemp FIDO key signing without PIN and use the error
code returned to fall back only if necessary. Avoids PIN prompts for FIDO
tokens that don't require them; part of GHPR#302
OpenBSD-Commit-ID: 4f752aaf9f2e7c28bcaaf3d4f8fc290131bd038e
* Install Cygwin packages based on OS not config.
* initial list of allowed signers
* upstream: whitespace
OpenBSD-Commit-ID: d297e4387935d4aef091c5e9432578c2e513f538
* upstream: whitespace
OpenBSD-Commit-ID: a5d015efbfd228dc598ffdef612d2da3a579e5d8
* Add cygwin-release test target.
This also moves the cygwin package install from the workflow file to
setup_ci.sh so that we can install different sets of Cygwin packages
for different test configs.
* Add Windows 2022 test targets.
* Add libcrypt-devel to cygwin-release deps.
Based on feedback from vinschen at redhat.com.
* cross-sign allowed_signers with PGP key
Provides continuity of trust from legacy PGP release key to
the SSHSIG signing keys that we will use henceforth for git
signing.
* additional keys
* upstream: whitespace
OpenBSD-Commit-ID: c2bcbf93610d3d62ed206cdf9bf9ff98c6aaf232
* Move sftp from valgrind-2 to 3 to rebalance.
* upstream: sk-usbhid: fix key_lookup() on tokens with built-in UV
explicitly test whether the token performs built-in UV (e.g. biometric
tokens) and enable UV in that case. From Pedro Martelletto via GHPR#388
OpenBSD-Commit-ID: 007eb7e387d27cf3029ab06b88224e03eca62ccd
* Remove arc4random_uniform from arc4random.c
This was previously moved into its own file (matching OpenBSD) which
prematurely committed in commit 73541f2.
* Move OPENBSD ORIGINAL marker.
Putting this after the copyright statement (which doesn't change)
instead of before the version identifier (which does) prevents merge
conflicts when resyncing changes.
* Resync arc4random with OpenBSD.
This brings us up to current, including djm's random-reseeding change,
as prompted by logan at cyberstorm.mu in bz#3467. It brings the
platform-specific hooks from LibreSSL Portable, simplified to match our
use case. ok djm@.
* Remove DEF_WEAK, it's already in defines.h.
* openbsd-compat/bsd-asprintf: add <stdio.h> include for vsnprintf
Fixes the following build failure with Clang 15 on musl:
```
bsd-asprintf.c:51:8: error: call to undeclared library function 'vsnprintf' with type 'int (char *, unsigned long, const char *, struct __va_list_tag *)'; ISO C99 and laterclang -O2 -pipe -fdiagnostics-color=always -frecord-gcc-switches -pipe -Wunknown-warning-option -Qunused-arguments -Wall -Wpointer-arith -Wuninitialized -Wsign-compare -Wformat-security -Wsizeof-pointer-memaccess -Wno-pointer-sign -Wno-unused-result -Wmisleading-indentation -Wbitwise-instead-of-logical -fno-strict-aliasing -mretpoline -ftrapv -fzero-call-used-regs=all -fno-builtin-memset -fstack-protector-strong -fPIE -I. -I. -D_XOPEN_SOURCE=600 -D_BSD_SOURCE -D_DEFAULT_SOURCE -DSSHDIR=\"/etc/ssh\" -D_PATH_SSH_PROGRAM=\"/usr/bin/ssh\" -D_PATH_SSH_ASKPASS_DEFAULT=\"/usr/lib/misc/ssh-askpass\" -D_PATH_SFTP_SERVER=\"/usr/lib/misc/sftp-server\" -D_PATH_SSH_KEY_SIGN=\"/usr/lib/misc/ssh-keysign\" -D_PATH_SSH_PKCS11_HELPER=\"/usr/lib/misc/ssh-pkcs11-helper\" -D_PATH_SSH_SK_HELPER=\"/usr/lib/misc/ssh-sk-helper\" -D_PATH_SSH_PIDDIR=\"/run\" -D_PATH_PRIVSEP_CHROOT_DIR=\"/var/empty\" -DHAVE_CONFIG_H -c cipher-aes.c -o cipher-aes.o
do not support
implicit function declarations [-Wimplicit-function-declaration]
ret = vsnprintf(string, INIT_SZ, fmt, ap2);
^
bsd-asprintf.c:51:8: note: include the header <stdio.h> or explicitly provide a declaration for 'vsnprintf'
1 error generated.
```
* upstream: notifier_complete(NULL, ...) is a noop, so no need to test
that ctx!=NULL; from Corinna Vinschen
OpenBSD-Commit-ID: ade2f2e9cc519d01a586800c25621d910bce384a
* fix pester test failures
* upstream: fix repeated words ok miod@ jmc@
OpenBSD-Commit-ID: 6765daefe26a6b648cc15cadbbe337596af709b7
* upstream: .Li -> .Vt where appropriate; from josiah frentsos,
tweaked by schwarze
ok schwarze
OpenBSD-Commit-ID: 565046e3ce68b46c2f440a93d67c2a92726de8ed
* upstream: ssh-agent: attempt FIDO key signing without PIN and use
the error to determine whether a PIN is required and prompt only if
necessary. from Corinna Vinschen
OpenBSD-Commit-ID: dd6be6a0b7148608e834ee737c3479b3270b00dd
* upstream: a little extra debugging
OpenBSD-Commit-ID: edf1601c1d0905f6da4c713f4d9cecc7d1c0295a
* upstream: sk_enroll: never drop SSH_SK_USER_VERIFICATION_REQD flag
from response
Now that all FIDO signing calls attempt first without PIN and then
fall back to trying PIN only if that attempt fails, we can remove the
hack^wtrick that removed the UV flag from the keys returned during
enroll.
By Corinna Vinschen
OpenBSD-Commit-ID: 684517608c8491503bf80cd175425f0178d91d7f
* define HAVE_KILLPG
* upstream: sftp: Don't attempt to complete arguments for
non-existent commands
If user entered a non-existent command (e.g. because they made a
typo) there is no point in trying to complete its arguments. Skip
calling complete_match() if that's the case.
From Michal Privoznik
OpenBSD-Commit-ID: cf39c811a68cde2aeb98fc85addea4000ef6b07a
* upstream: sftp: Be a bit more clever about completions
There are commands (e.g. "get" or "put") that accept two
arguments, a local path and a remote path. However, the way
current completion is written doesn't take this distinction into
account and always completes remote or local paths.
By expanding CMD struct and "cmds" array this distinction can be
reflected and with small adjustment to completer code the correct
path can be completed.
By Michal Privoznik, ok dtucker@
OpenBSD-Commit-ID: 1396d921c4eb1befd531f5c4a8ab47e7a74b610b
* upstream: correct error value
OpenBSD-Commit-ID: 780efcbad76281f11f14b2a5ff04eb6db3dfdad4
* upstream: actually hook up restrict_websafe; the command-line flag
was never actually used. Spotted by Matthew Garrett
OpenBSD-Commit-ID: 0b363518ac4c2819dbaa3dfad4028633ab9cdff1
* upstream: Add a sshkey_check_rsa_length() call for checking the
length of an RSA key; ok markus@
OpenBSD-Commit-ID: de77cd5b11594297eda82edc594b0d32b8535134
* upstream: add a RequiredRSASize for checking RSA key length in
ssh(1). User authentication keys that fall beneath this limit will be
ignored. If a host presents a host key beneath this limit then the connection
will be terminated (unfortunately there are no fallbacks in the protocol for
host authentication).
feedback deraadt, Dmitry Belyavskiy; ok markus@
OpenBSD-Commit-ID: 430e339b2a79fa9ecc63f2837b06fdd88a7da13a
* upstream: Add RequiredRSASize for sshd(8); RSA keys that fall
beneath this limit will be ignored for user and host-based authentication.
Feedback deraadt@ ok markus@
OpenBSD-Commit-ID: 187931dfc19d51873df5930a04f2d972adf1f7f1
* upstream: better debugging for connect_next()
OpenBSD-Commit-ID: d16a307a0711499c971807f324484ed3a6036640
* upstream: sftp-server(8): add a "users-groups-by-id@openssh.com"
extension request that allows the client to obtain user/group names that
correspond to a set of uids/gids.
Will be used to make directory listings more useful and consistent
in sftp(1).
ok markus@
OpenBSD-Commit-ID: 7ebabde0bcb95ef949c4840fe89e697e30df47d3
* upstream: extend sftp-common.c:extend ls_file() to support supplied
user/group names; ok markus@
OpenBSD-Commit-ID: c70c70498b1fdcf158531117e405b6245863bfb0
* upstream: sftp client library support for
users-groups-by-id@openssh.com; ok markus@
OpenBSD-Commit-ID: ddb2f33a2da6349a9a89a8b5bcb9ca7c999394de
* upstream: use users-groups-by-id@openssh.com sftp-server extension
(when available) to fill in user/group names for directory listings.
Implement a client-side cache of see uid/gid=>user/group names. ok markus@
OpenBSD-Commit-ID: f239aeeadfa925a37ceee36ee8b256b8ccf4466e
* avoid Wuninitialized false positive in gcc-12ish
* no need for glob.h here
it also causes portability problems
* add debug on appveyor
* add sleep to pester test
* upstream: add RequiredRSASize to the list of keywords accepted by
-o; spotted by jmc@
OpenBSD-Commit-ID: fe871408cf6f9d3699afeda876f8adbac86a035e
* upstream: Fix typo. From AlexanderStohr via github PR#343.
OpenBSD-Commit-ID: a134c9b4039e48803fc6a87f955b0f4a03181497
* upstream: openssh-9.1
OpenBSD-Commit-ID: 5a467b2ee81da01a86adf1ad93b62b1728494e56
* crank versions in RPM spec files
* update release notes URL
* update .depend
* fix 9.1 compilation errors
* disable -p pester tests due to unreliability on older Windows versions
* remove extra sleep time from debugging scp pester tests
* modify -p tests to only run for Windows OS version 10 and above
* add windows specific code back into method moved from auth.c to auth2-pubkeyfile.c
* add preprocessor for WinHello
* revert preprocessor definition for winhello
* add windows preprocessor definition in key_lookup
* remove rdp block from appveyor since we are no longer debugging
* add ifdef to sftp-server.c
* make key_lookup compatible with winhello
* appveyor.yml
* increase debug of failing pester test
* add #ifdef SUPPORT_CRLF back into auth_check_principals_line method that was moved/renamed
* modify new scp.sh tests for windows
* remove in place tests from scp.sh
* remove rdp debug from appveyor
* retrigger appveyor
* change check of OS version in scp test
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
Co-authored-by: djm@openbsd.org <djm@openbsd.org>
Co-authored-by: Damien Miller <djm@mindrot.org>
Co-authored-by: Darren Tucker <dtucker@dtucker.net>
Co-authored-by: naddy@openbsd.org <naddy@openbsd.org>
Co-authored-by: dtucker@openbsd.org <dtucker@openbsd.org>
Co-authored-by: tj@openbsd.org <tj@openbsd.org>
Co-authored-by: millert@openbsd.org <millert@openbsd.org>
Co-authored-by: jmc@openbsd.org <jmc@openbsd.org>
Co-authored-by: florian@openbsd.org <florian@openbsd.org>
Co-authored-by: markus@openbsd.org <markus@openbsd.org>
Co-authored-by: Tobias Heider <me@tobhe.de>
Co-authored-by: anton@openbsd.org <anton@openbsd.org>
Co-authored-by: Tim Rice <tim@multitalents.net>
Co-authored-by: tobhe@openbsd.org <tobhe@openbsd.org>
Co-authored-by: Corinna Vinschen <vinschen@redhat.com>
Co-authored-by: Sam James <sam@gentoo.org>
Co-authored-by: jsg@openbsd.org <jsg@openbsd.org>
2022-11-02 17:06:45 +01:00
|
|
|
if ((r = sshkey_check_rsa_length(idlist->keys[j],
|
|
|
|
options.required_rsa_size)) != 0) {
|
|
|
|
debug_fr(r, "ignoring %s agent key",
|
|
|
|
sshkey_ssh_name(idlist->keys[j]));
|
|
|
|
continue;
|
|
|
|
}
|
2003-05-14 05:46:00 +02:00
|
|
|
found = 0;
|
|
|
|
TAILQ_FOREACH(id, &files, next) {
|
2015-01-14 21:05:27 +01:00
|
|
|
/*
|
|
|
|
* agent keys from the config file are
|
|
|
|
* preferred
|
|
|
|
*/
|
|
|
|
if (sshkey_equal(idlist->keys[j], id->key)) {
|
2003-05-14 05:46:00 +02:00
|
|
|
TAILQ_REMOVE(&files, id, next);
|
|
|
|
TAILQ_INSERT_TAIL(preferred, id, next);
|
2015-01-14 21:05:27 +01:00
|
|
|
id->agent_fd = agent_fd;
|
2003-05-14 05:46:00 +02:00
|
|
|
found = 1;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
2004-03-08 13:12:36 +01:00
|
|
|
if (!found && !options.identities_only) {
|
2006-03-26 05:19:21 +02:00
|
|
|
id = xcalloc(1, sizeof(*id));
|
2015-01-14 21:05:27 +01:00
|
|
|
/* XXX "steals" key/comment from idlist */
|
|
|
|
id->key = idlist->keys[j];
|
|
|
|
id->filename = idlist->comments[j];
|
|
|
|
idlist->keys[j] = NULL;
|
|
|
|
idlist->comments[j] = NULL;
|
|
|
|
id->agent_fd = agent_fd;
|
2003-05-14 05:46:00 +02:00
|
|
|
TAILQ_INSERT_TAIL(&agent, id, next);
|
|
|
|
}
|
|
|
|
}
|
2015-01-14 21:05:27 +01:00
|
|
|
ssh_free_identitylist(idlist);
|
2003-05-14 05:46:00 +02:00
|
|
|
/* append remaining agent keys */
|
2020-06-27 15:39:09 +02:00
|
|
|
TAILQ_CONCAT(preferred, &agent, next);
|
2015-01-14 21:05:27 +01:00
|
|
|
authctxt->agent_fd = agent_fd;
|
2003-05-14 05:46:00 +02:00
|
|
|
}
|
2016-05-24 01:30:50 +02:00
|
|
|
/* Prefer PKCS11 keys that are explicitly listed */
|
|
|
|
TAILQ_FOREACH_SAFE(id, &files, next, tmp) {
|
|
|
|
if (id->key == NULL || (id->key->flags & SSHKEY_FLAG_EXT) == 0)
|
|
|
|
continue;
|
|
|
|
found = 0;
|
|
|
|
TAILQ_FOREACH(id2, &files, next) {
|
|
|
|
if (id2->key == NULL ||
|
2020-04-17 05:38:47 +02:00
|
|
|
(id2->key->flags & SSHKEY_FLAG_EXT) != 0)
|
2016-05-24 01:30:50 +02:00
|
|
|
continue;
|
|
|
|
if (sshkey_equal(id->key, id2->key)) {
|
|
|
|
TAILQ_REMOVE(&files, id, next);
|
|
|
|
TAILQ_INSERT_TAIL(preferred, id, next);
|
|
|
|
found = 1;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
/* If IdentitiesOnly set and key not found then don't use it */
|
|
|
|
if (!found && options.identities_only) {
|
|
|
|
TAILQ_REMOVE(&files, id, next);
|
2018-07-09 23:03:30 +02:00
|
|
|
freezero(id, sizeof(*id));
|
2016-05-24 01:30:50 +02:00
|
|
|
}
|
|
|
|
}
|
2003-05-14 05:46:00 +02:00
|
|
|
/* append remaining keys from the config file */
|
2020-06-27 15:39:09 +02:00
|
|
|
TAILQ_CONCAT(preferred, &files, next);
|
2021-01-22 03:44:58 +01:00
|
|
|
/* finally, filter by PubkeyAcceptedAlgorithms */
|
2015-10-13 18:15:21 +02:00
|
|
|
TAILQ_FOREACH_SAFE(id, preferred, next, id2) {
|
2023-12-18 15:45:49 +01:00
|
|
|
disallowed = 0;
|
|
|
|
cp = NULL;
|
|
|
|
if (id->key == NULL)
|
2015-10-13 18:15:21 +02:00
|
|
|
continue;
|
2023-12-18 15:45:49 +01:00
|
|
|
if (!key_type_allowed_by_config(id->key)) {
|
|
|
|
debug("Skipping %s key %s - corresponding algorithm "
|
|
|
|
"not in PubkeyAcceptedAlgorithms",
|
|
|
|
sshkey_ssh_name(id->key), id->filename);
|
|
|
|
disallowed = 1;
|
|
|
|
} else if (ssh->kex->server_sig_algs != NULL &&
|
|
|
|
(cp = key_sig_algorithm(ssh, id->key)) == NULL) {
|
|
|
|
debug("Skipping %s key %s - corresponding algorithm "
|
|
|
|
"not supported by server",
|
|
|
|
sshkey_ssh_name(id->key), id->filename);
|
|
|
|
disallowed = 1;
|
2015-10-13 18:15:21 +02:00
|
|
|
}
|
2023-12-18 15:45:49 +01:00
|
|
|
free(cp);
|
|
|
|
if (!disallowed)
|
|
|
|
continue;
|
|
|
|
/* remove key */
|
|
|
|
TAILQ_REMOVE(preferred, id, next);
|
|
|
|
sshkey_free(id->key);
|
|
|
|
free(id->filename);
|
|
|
|
memset(id, 0, sizeof(*id));
|
2003-05-14 05:46:00 +02:00
|
|
|
}
|
2018-09-14 07:26:27 +02:00
|
|
|
/* List the keys we plan on using */
|
|
|
|
TAILQ_FOREACH_SAFE(id, preferred, next, id2) {
|
|
|
|
ident = format_identity(id);
|
|
|
|
debug("Will attempt key: %s", ident);
|
|
|
|
free(ident);
|
|
|
|
}
|
2020-10-18 13:32:01 +02:00
|
|
|
debug2_f("done");
|
2001-03-09 01:12:22 +01:00
|
|
|
}
|
|
|
|
|
2003-05-14 05:46:00 +02:00
|
|
|
static void
|
2019-02-13 00:53:10 +01:00
|
|
|
pubkey_cleanup(struct ssh *ssh)
|
2000-08-23 02:46:23 +02:00
|
|
|
{
|
2019-02-11 10:44:42 +01:00
|
|
|
Authctxt *authctxt = (Authctxt *)ssh->authctxt;
|
2003-05-14 05:46:00 +02:00
|
|
|
Identity *id;
|
|
|
|
|
2018-11-16 03:46:20 +01:00
|
|
|
if (authctxt->agent_fd != -1) {
|
2015-01-14 21:05:27 +01:00
|
|
|
ssh_close_authentication_socket(authctxt->agent_fd);
|
2018-11-16 03:46:20 +01:00
|
|
|
authctxt->agent_fd = -1;
|
|
|
|
}
|
2003-05-14 05:46:00 +02:00
|
|
|
for (id = TAILQ_FIRST(&authctxt->keys); id;
|
|
|
|
id = TAILQ_FIRST(&authctxt->keys)) {
|
|
|
|
TAILQ_REMOVE(&authctxt->keys, id, next);
|
2015-12-11 03:31:47 +01:00
|
|
|
sshkey_free(id->key);
|
2013-06-01 23:31:17 +02:00
|
|
|
free(id->filename);
|
|
|
|
free(id);
|
2000-08-23 02:46:23 +02:00
|
|
|
}
|
2000-04-29 15:57:08 +02:00
|
|
|
}
|
|
|
|
|
2016-12-05 00:54:02 +01:00
|
|
|
static void
|
|
|
|
pubkey_reset(Authctxt *authctxt)
|
|
|
|
{
|
|
|
|
Identity *id;
|
|
|
|
|
|
|
|
TAILQ_FOREACH(id, &authctxt->keys, next)
|
|
|
|
id->tried = 0;
|
|
|
|
}
|
|
|
|
|
2019-02-11 10:44:42 +01:00
|
|
|
static int
|
2019-01-19 22:40:21 +01:00
|
|
|
userauth_pubkey(struct ssh *ssh)
|
2000-04-29 15:57:08 +02:00
|
|
|
{
|
2019-01-19 22:40:21 +01:00
|
|
|
Authctxt *authctxt = (Authctxt *)ssh->authctxt;
|
2003-05-14 05:46:00 +02:00
|
|
|
Identity *id;
|
2000-09-23 08:15:56 +02:00
|
|
|
int sent = 0;
|
2018-09-14 07:26:27 +02:00
|
|
|
char *ident;
|
2023-12-18 15:45:49 +01:00
|
|
|
static int prepared;
|
|
|
|
|
|
|
|
if (!prepared) {
|
|
|
|
pubkey_prepare(ssh, authctxt);
|
|
|
|
prepared = 1;
|
|
|
|
}
|
2000-09-23 08:15:56 +02:00
|
|
|
|
2003-05-14 05:46:00 +02:00
|
|
|
while ((id = TAILQ_FIRST(&authctxt->keys))) {
|
|
|
|
if (id->tried++)
|
|
|
|
return (0);
|
2003-10-15 07:55:59 +02:00
|
|
|
/* move key to the end of the queue */
|
2003-05-14 05:46:00 +02:00
|
|
|
TAILQ_REMOVE(&authctxt->keys, id, next);
|
|
|
|
TAILQ_INSERT_TAIL(&authctxt->keys, id, next);
|
|
|
|
/*
|
|
|
|
* send a test message if we have the public key. for
|
|
|
|
* encrypted keys we cannot do this and have to load the
|
|
|
|
* private key instead
|
|
|
|
*/
|
2013-12-31 02:25:40 +01:00
|
|
|
if (id->key != NULL) {
|
2023-08-01 10:15:04 +02:00
|
|
|
ident = format_identity(id);
|
|
|
|
debug("Offering public key: %s", ident);
|
|
|
|
free(ident);
|
|
|
|
sent = send_pubkey_test(ssh, id);
|
2013-12-31 02:25:40 +01:00
|
|
|
} else {
|
2003-05-14 05:46:00 +02:00
|
|
|
debug("Trying private key: %s", id->filename);
|
2015-11-15 23:26:49 +01:00
|
|
|
id->key = load_identity_file(id);
|
2003-05-14 05:46:00 +02:00
|
|
|
if (id->key != NULL) {
|
2023-02-17 05:22:50 +01:00
|
|
|
if (id->key != NULL) {
|
2015-07-10 08:21:53 +02:00
|
|
|
id->isprivate = 1;
|
2019-01-19 22:40:21 +01:00
|
|
|
sent = sign_and_send_pubkey(ssh, id);
|
2013-12-31 02:25:40 +01:00
|
|
|
}
|
2018-07-09 23:03:30 +02:00
|
|
|
sshkey_free(id->key);
|
2003-05-14 05:46:00 +02:00
|
|
|
id->key = NULL;
|
2016-12-05 00:54:02 +01:00
|
|
|
id->isprivate = 0;
|
2001-03-09 01:12:22 +01:00
|
|
|
}
|
|
|
|
}
|
2003-05-14 05:46:00 +02:00
|
|
|
if (sent)
|
|
|
|
return (sent);
|
2000-11-13 12:57:25 +01:00
|
|
|
}
|
2003-05-14 05:46:00 +02:00
|
|
|
return (0);
|
2000-09-23 08:15:56 +02:00
|
|
|
}
|
2000-04-29 15:57:08 +02:00
|
|
|
|
2000-10-14 07:23:11 +02:00
|
|
|
/*
|
|
|
|
* Send userauth request message specifying keyboard-interactive method.
|
|
|
|
*/
|
2019-02-11 10:44:42 +01:00
|
|
|
static int
|
2019-01-19 22:40:21 +01:00
|
|
|
userauth_kbdint(struct ssh *ssh)
|
2000-10-14 07:23:11 +02:00
|
|
|
{
|
2019-01-19 22:40:21 +01:00
|
|
|
Authctxt *authctxt = (Authctxt *)ssh->authctxt;
|
2018-07-09 23:03:30 +02:00
|
|
|
int r;
|
2000-10-14 07:23:11 +02:00
|
|
|
|
2019-01-04 04:27:50 +01:00
|
|
|
if (authctxt->attempt_kbdint++ >= options.number_of_password_prompts)
|
2000-10-14 07:23:11 +02:00
|
|
|
return 0;
|
2001-09-12 20:29:00 +02:00
|
|
|
/* disable if no SSH2_MSG_USERAUTH_INFO_REQUEST has been seen */
|
2019-01-04 04:27:50 +01:00
|
|
|
if (authctxt->attempt_kbdint > 1 && !authctxt->info_req_seen) {
|
2001-09-12 20:29:00 +02:00
|
|
|
debug3("userauth_kbdint: disable: no info_req_seen");
|
2018-07-09 23:03:30 +02:00
|
|
|
ssh_dispatch_set(ssh, SSH2_MSG_USERAUTH_INFO_REQUEST, NULL);
|
2001-09-12 20:29:00 +02:00
|
|
|
return 0;
|
|
|
|
}
|
2000-10-14 07:23:11 +02:00
|
|
|
|
|
|
|
debug2("userauth_kbdint");
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshpkt_start(ssh, SSH2_MSG_USERAUTH_REQUEST)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->server_user)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->service)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->method->name)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, "")) != 0 || /* lang */
|
|
|
|
(r = sshpkt_put_cstring(ssh, options.kbd_interactive_devices ?
|
|
|
|
options.kbd_interactive_devices : "")) != 0 ||
|
|
|
|
(r = sshpkt_send(ssh)) != 0)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_fr(r, "send packet");
|
2018-07-09 23:03:30 +02:00
|
|
|
|
|
|
|
ssh_dispatch_set(ssh, SSH2_MSG_USERAUTH_INFO_REQUEST, &input_userauth_info_req);
|
2000-10-14 07:23:11 +02:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2001-02-10 23:16:41 +01:00
|
|
|
* parse INFO_REQUEST, prompt user and send INFO_RESPONSE
|
2000-10-14 07:23:11 +02:00
|
|
|
*/
|
2019-02-11 10:44:42 +01:00
|
|
|
static int
|
2017-05-30 16:23:52 +02:00
|
|
|
input_userauth_info_req(int type, u_int32_t seq, struct ssh *ssh)
|
2000-10-14 07:23:11 +02:00
|
|
|
{
|
2017-05-30 16:19:15 +02:00
|
|
|
Authctxt *authctxt = ssh->authctxt;
|
2018-07-09 23:03:30 +02:00
|
|
|
char *name = NULL, *inst = NULL, *lang = NULL, *prompt = NULL;
|
2020-11-13 05:53:12 +01:00
|
|
|
char *display_prompt = NULL, *response = NULL;
|
2018-07-09 23:03:30 +02:00
|
|
|
u_char echo = 0;
|
2000-12-22 02:43:59 +01:00
|
|
|
u_int num_prompts, i;
|
2018-07-09 23:03:30 +02:00
|
|
|
int r;
|
2000-10-14 07:23:11 +02:00
|
|
|
|
2020-11-13 05:53:12 +01:00
|
|
|
debug2_f("entering");
|
2000-10-14 07:23:11 +02:00
|
|
|
|
|
|
|
if (authctxt == NULL)
|
2020-11-13 05:53:12 +01:00
|
|
|
fatal_f("no authentication context");
|
2000-10-14 07:23:11 +02:00
|
|
|
|
2001-09-12 20:29:00 +02:00
|
|
|
authctxt->info_req_seen = 1;
|
|
|
|
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshpkt_get_cstring(ssh, &name, NULL)) != 0 ||
|
|
|
|
(r = sshpkt_get_cstring(ssh, &inst, NULL)) != 0 ||
|
|
|
|
(r = sshpkt_get_cstring(ssh, &lang, NULL)) != 0)
|
|
|
|
goto out;
|
2000-10-14 07:23:11 +02:00
|
|
|
if (strlen(name) > 0)
|
2003-04-09 12:59:48 +02:00
|
|
|
logit("%s", name);
|
2000-10-14 07:23:11 +02:00
|
|
|
if (strlen(inst) > 0)
|
2003-04-09 12:59:48 +02:00
|
|
|
logit("%s", inst);
|
2000-10-14 07:23:11 +02:00
|
|
|
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshpkt_get_u32(ssh, &num_prompts)) != 0)
|
|
|
|
goto out;
|
2000-10-14 07:23:11 +02:00
|
|
|
/*
|
|
|
|
* Begin to build info response packet based on prompts requested.
|
|
|
|
* We commit to providing the correct number of responses, so if
|
|
|
|
* further on we run into a problem that prevents this, we have to
|
|
|
|
* be sure and clean this up and send a correct error response.
|
|
|
|
*/
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshpkt_start(ssh, SSH2_MSG_USERAUTH_INFO_RESPONSE)) != 0 ||
|
|
|
|
(r = sshpkt_put_u32(ssh, num_prompts)) != 0)
|
|
|
|
goto out;
|
2000-10-14 07:23:11 +02:00
|
|
|
|
2020-11-13 05:53:12 +01:00
|
|
|
debug2_f("num_prompts %d", num_prompts);
|
2000-10-14 07:23:11 +02:00
|
|
|
for (i = 0; i < num_prompts; i++) {
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshpkt_get_cstring(ssh, &prompt, NULL)) != 0 ||
|
|
|
|
(r = sshpkt_get_u8(ssh, &echo)) != 0)
|
|
|
|
goto out;
|
2020-11-13 08:30:44 +01:00
|
|
|
if (asmprintf(&display_prompt, INT_MAX, NULL, "(%s@%s) %s",
|
2020-11-13 05:53:12 +01:00
|
|
|
authctxt->server_user, options.host_key_alias ?
|
2020-11-13 08:30:44 +01:00
|
|
|
options.host_key_alias : authctxt->host, prompt) == -1)
|
|
|
|
fatal_f("asmprintf failed");
|
2020-11-13 05:53:12 +01:00
|
|
|
response = read_passphrase(display_prompt, echo ? RP_ECHO : 0);
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshpkt_put_cstring(ssh, response)) != 0)
|
|
|
|
goto out;
|
|
|
|
freezero(response, strlen(response));
|
2013-06-01 23:31:17 +02:00
|
|
|
free(prompt);
|
2020-11-13 05:53:12 +01:00
|
|
|
free(display_prompt);
|
|
|
|
display_prompt = response = prompt = NULL;
|
2000-10-14 07:23:11 +02:00
|
|
|
}
|
2018-07-09 23:03:30 +02:00
|
|
|
/* done with parsing incoming message. */
|
|
|
|
if ((r = sshpkt_get_end(ssh)) != 0 ||
|
|
|
|
(r = sshpkt_add_padding(ssh, 64)) != 0)
|
|
|
|
goto out;
|
|
|
|
r = sshpkt_send(ssh);
|
|
|
|
out:
|
|
|
|
if (response)
|
|
|
|
freezero(response, strlen(response));
|
|
|
|
free(prompt);
|
2020-11-13 05:53:12 +01:00
|
|
|
free(display_prompt);
|
2018-07-09 23:03:30 +02:00
|
|
|
free(name);
|
|
|
|
free(inst);
|
|
|
|
free(lang);
|
|
|
|
return r;
|
2000-10-14 07:23:11 +02:00
|
|
|
}
|
2000-04-29 15:57:08 +02:00
|
|
|
|
2002-06-06 21:57:33 +02:00
|
|
|
static int
|
2019-01-19 22:34:45 +01:00
|
|
|
ssh_keysign(struct ssh *ssh, struct sshkey *key, u_char **sigp, size_t *lenp,
|
2015-01-30 12:43:14 +01:00
|
|
|
const u_char *data, size_t datalen)
|
2002-06-06 21:57:33 +02:00
|
|
|
{
|
2015-01-30 12:43:14 +01:00
|
|
|
struct sshbuf *b;
|
2002-06-06 21:59:29 +02:00
|
|
|
struct stat st;
|
2002-06-06 21:57:33 +02:00
|
|
|
pid_t pid;
|
2019-05-15 06:43:31 +02:00
|
|
|
int r, to[2], from[2], status;
|
2019-01-19 22:40:21 +01:00
|
|
|
int sock = ssh_packet_get_connection_in(ssh);
|
2015-01-30 12:43:14 +01:00
|
|
|
u_char rversion = 0, version = 2;
|
|
|
|
void (*osigchld)(int);
|
2002-06-06 21:57:33 +02:00
|
|
|
|
2015-01-30 12:43:14 +01:00
|
|
|
*sigp = NULL;
|
|
|
|
*lenp = 0;
|
2002-06-06 21:57:33 +02:00
|
|
|
|
2019-06-28 15:35:04 +02:00
|
|
|
if (stat(_PATH_SSH_KEY_SIGN, &st) == -1) {
|
2020-10-18 13:32:01 +02:00
|
|
|
error_f("not installed: %s", strerror(errno));
|
2015-01-30 12:43:14 +01:00
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
if (fflush(stdout) != 0) {
|
2020-10-18 13:32:01 +02:00
|
|
|
error_f("fflush: %s", strerror(errno));
|
2002-06-06 21:59:29 +02:00
|
|
|
return -1;
|
|
|
|
}
|
2019-06-28 15:35:04 +02:00
|
|
|
if (pipe(to) == -1) {
|
2020-10-18 13:32:01 +02:00
|
|
|
error_f("pipe: %s", strerror(errno));
|
2002-06-06 21:57:33 +02:00
|
|
|
return -1;
|
|
|
|
}
|
2019-06-28 15:35:04 +02:00
|
|
|
if (pipe(from) == -1) {
|
2020-10-18 13:32:01 +02:00
|
|
|
error_f("pipe: %s", strerror(errno));
|
2002-06-06 21:57:33 +02:00
|
|
|
return -1;
|
|
|
|
}
|
2019-06-28 15:35:04 +02:00
|
|
|
if ((pid = fork()) == -1) {
|
2020-10-18 13:32:01 +02:00
|
|
|
error_f("fork: %s", strerror(errno));
|
2002-06-06 21:57:33 +02:00
|
|
|
return -1;
|
|
|
|
}
|
2020-01-23 08:10:22 +01:00
|
|
|
osigchld = ssh_signal(SIGCHLD, SIG_DFL);
|
2002-06-06 21:57:33 +02:00
|
|
|
if (pid == 0) {
|
|
|
|
close(from[0]);
|
2019-06-28 15:35:04 +02:00
|
|
|
if (dup2(from[1], STDOUT_FILENO) == -1)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_f("dup2: %s", strerror(errno));
|
2002-06-06 21:57:33 +02:00
|
|
|
close(to[1]);
|
2019-06-28 15:35:04 +02:00
|
|
|
if (dup2(to[0], STDIN_FILENO) == -1)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_f("dup2: %s", strerror(errno));
|
2002-06-06 22:51:04 +02:00
|
|
|
close(from[1]);
|
|
|
|
close(to[0]);
|
2019-05-15 06:43:31 +02:00
|
|
|
|
2019-06-28 15:35:04 +02:00
|
|
|
if (dup2(sock, STDERR_FILENO + 1) == -1)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_f("dup2: %s", strerror(errno));
|
2019-05-15 06:43:31 +02:00
|
|
|
sock = STDERR_FILENO + 1;
|
2023-03-03 03:34:29 +01:00
|
|
|
if (fcntl(sock, F_SETFD, 0) == -1) /* keep the socket on exec */
|
|
|
|
debug3_f("fcntl F_SETFD: %s", strerror(errno));
|
2015-01-30 12:43:14 +01:00
|
|
|
closefrom(sock + 1);
|
2019-05-15 06:43:31 +02:00
|
|
|
|
2020-10-18 13:32:01 +02:00
|
|
|
debug3_f("[child] pid=%ld, exec %s",
|
|
|
|
(long)getpid(), _PATH_SSH_KEY_SIGN);
|
2015-12-11 01:20:04 +01:00
|
|
|
execl(_PATH_SSH_KEY_SIGN, _PATH_SSH_KEY_SIGN, (char *)NULL);
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_f("exec(%s): %s", _PATH_SSH_KEY_SIGN,
|
2002-06-06 21:57:33 +02:00
|
|
|
strerror(errno));
|
|
|
|
}
|
|
|
|
close(from[1]);
|
|
|
|
close(to[0]);
|
2019-05-31 05:20:07 +02:00
|
|
|
sock = STDERR_FILENO + 1;
|
2002-06-06 21:57:33 +02:00
|
|
|
|
2015-01-30 12:43:14 +01:00
|
|
|
if ((b = sshbuf_new()) == NULL)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_f("sshbuf_new failed");
|
2015-01-30 12:43:14 +01:00
|
|
|
/* send # of sock, data to be signed */
|
2017-01-30 01:32:28 +01:00
|
|
|
if ((r = sshbuf_put_u32(b, sock)) != 0 ||
|
2015-01-30 12:43:14 +01:00
|
|
|
(r = sshbuf_put_string(b, data, datalen)) != 0)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_fr(r, "buffer error");
|
2015-01-30 12:43:14 +01:00
|
|
|
if (ssh_msg_send(to[1], version, b) == -1)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_f("couldn't send request");
|
2015-01-30 12:43:14 +01:00
|
|
|
sshbuf_reset(b);
|
|
|
|
r = ssh_msg_recv(from[0], b);
|
2002-06-06 21:57:33 +02:00
|
|
|
close(from[0]);
|
|
|
|
close(to[1]);
|
2015-01-30 12:43:14 +01:00
|
|
|
if (r < 0) {
|
2020-10-18 13:32:01 +02:00
|
|
|
error_f("no reply");
|
2015-01-30 12:43:14 +01:00
|
|
|
goto fail;
|
|
|
|
}
|
2002-06-06 21:57:33 +02:00
|
|
|
|
2015-01-30 12:43:14 +01:00
|
|
|
errno = 0;
|
2019-06-28 15:35:04 +02:00
|
|
|
while (waitpid(pid, &status, 0) == -1) {
|
2015-01-30 12:43:14 +01:00
|
|
|
if (errno != EINTR) {
|
2020-10-18 13:32:01 +02:00
|
|
|
error_f("waitpid %ld: %s", (long)pid, strerror(errno));
|
2015-01-30 12:43:14 +01:00
|
|
|
goto fail;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if (!WIFEXITED(status)) {
|
2020-10-18 13:32:01 +02:00
|
|
|
error_f("exited abnormally");
|
2015-01-30 12:43:14 +01:00
|
|
|
goto fail;
|
|
|
|
}
|
|
|
|
if (WEXITSTATUS(status) != 0) {
|
2020-10-18 13:32:01 +02:00
|
|
|
error_f("exited with status %d", WEXITSTATUS(status));
|
2015-01-30 12:43:14 +01:00
|
|
|
goto fail;
|
|
|
|
}
|
|
|
|
if ((r = sshbuf_get_u8(b, &rversion)) != 0) {
|
2020-10-18 13:32:01 +02:00
|
|
|
error_fr(r, "buffer error");
|
2015-01-30 12:43:14 +01:00
|
|
|
goto fail;
|
|
|
|
}
|
|
|
|
if (rversion != version) {
|
2020-10-18 13:32:01 +02:00
|
|
|
error_f("bad version");
|
2015-01-30 12:43:14 +01:00
|
|
|
goto fail;
|
|
|
|
}
|
|
|
|
if ((r = sshbuf_get_string(b, sigp, lenp)) != 0) {
|
2020-10-18 13:32:01 +02:00
|
|
|
error_fr(r, "buffer error");
|
2015-01-30 12:43:14 +01:00
|
|
|
fail:
|
2020-01-23 08:10:22 +01:00
|
|
|
ssh_signal(SIGCHLD, osigchld);
|
2015-01-30 12:43:14 +01:00
|
|
|
sshbuf_free(b);
|
2002-06-06 21:59:29 +02:00
|
|
|
return -1;
|
|
|
|
}
|
2020-01-23 08:10:22 +01:00
|
|
|
ssh_signal(SIGCHLD, osigchld);
|
2015-01-30 12:43:14 +01:00
|
|
|
sshbuf_free(b);
|
2002-06-06 21:59:29 +02:00
|
|
|
|
2002-06-06 21:57:33 +02:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2019-02-11 10:44:42 +01:00
|
|
|
static int
|
2019-01-19 22:40:21 +01:00
|
|
|
userauth_hostbased(struct ssh *ssh)
|
2001-04-13 01:34:34 +02:00
|
|
|
{
|
2019-01-19 22:40:21 +01:00
|
|
|
Authctxt *authctxt = (Authctxt *)ssh->authctxt;
|
2015-01-30 12:43:14 +01:00
|
|
|
struct sshkey *private = NULL;
|
|
|
|
struct sshbuf *b = NULL;
|
|
|
|
u_char *sig = NULL, *keyblob = NULL;
|
|
|
|
char *fp = NULL, *chost = NULL, *lname = NULL;
|
|
|
|
size_t siglen = 0, keylen = 0;
|
|
|
|
int i, r, success = 0;
|
|
|
|
|
|
|
|
if (authctxt->ktypes == NULL) {
|
2021-01-26 06:32:21 +01:00
|
|
|
authctxt->oktypes = xstrdup(options.hostbased_accepted_algos);
|
2015-01-30 12:43:14 +01:00
|
|
|
authctxt->ktypes = authctxt->oktypes;
|
|
|
|
}
|
2015-01-08 11:14:08 +01:00
|
|
|
|
2015-01-30 12:43:14 +01:00
|
|
|
/*
|
2021-01-26 06:32:21 +01:00
|
|
|
* Work through each listed type pattern in HostbasedAcceptedAlgorithms,
|
2015-01-30 12:43:14 +01:00
|
|
|
* trying each hostkey that matches the type in turn.
|
|
|
|
*/
|
|
|
|
for (;;) {
|
|
|
|
if (authctxt->active_ktype == NULL)
|
|
|
|
authctxt->active_ktype = strsep(&authctxt->ktypes, ",");
|
|
|
|
if (authctxt->active_ktype == NULL ||
|
|
|
|
*authctxt->active_ktype == '\0')
|
|
|
|
break;
|
2020-10-18 13:32:01 +02:00
|
|
|
debug3_f("trying key type %s", authctxt->active_ktype);
|
2015-01-30 12:43:14 +01:00
|
|
|
|
|
|
|
/* check for a useful key */
|
|
|
|
private = NULL;
|
|
|
|
for (i = 0; i < authctxt->sensitive->nkeys; i++) {
|
|
|
|
if (authctxt->sensitive->keys[i] == NULL ||
|
|
|
|
authctxt->sensitive->keys[i]->type == KEY_UNSPEC)
|
|
|
|
continue;
|
2022-01-06 23:06:51 +01:00
|
|
|
if (!sshkey_match_keyname_to_sigalgs(
|
2015-01-30 12:43:14 +01:00
|
|
|
sshkey_ssh_name(authctxt->sensitive->keys[i]),
|
2022-01-06 23:06:51 +01:00
|
|
|
authctxt->active_ktype))
|
2015-01-30 12:43:14 +01:00
|
|
|
continue;
|
2001-04-13 01:34:34 +02:00
|
|
|
/* we take and free the key */
|
2015-01-30 12:43:14 +01:00
|
|
|
private = authctxt->sensitive->keys[i];
|
|
|
|
authctxt->sensitive->keys[i] = NULL;
|
2001-04-13 01:34:34 +02:00
|
|
|
break;
|
|
|
|
}
|
2015-01-30 12:43:14 +01:00
|
|
|
/* Found one */
|
|
|
|
if (private != NULL)
|
|
|
|
break;
|
|
|
|
/* No more keys of this type; advance */
|
|
|
|
authctxt->active_ktype = NULL;
|
2001-04-13 01:34:34 +02:00
|
|
|
}
|
2015-01-30 12:43:14 +01:00
|
|
|
if (private == NULL) {
|
|
|
|
free(authctxt->oktypes);
|
|
|
|
authctxt->oktypes = authctxt->ktypes = NULL;
|
|
|
|
authctxt->active_ktype = NULL;
|
2002-12-23 03:44:36 +01:00
|
|
|
debug("No more client hostkeys for hostbased authentication.");
|
2015-01-30 12:43:14 +01:00
|
|
|
goto out;
|
2001-04-13 01:34:34 +02:00
|
|
|
}
|
2014-12-11 06:13:28 +01:00
|
|
|
|
2015-01-30 12:43:14 +01:00
|
|
|
if ((fp = sshkey_fingerprint(private, options.fingerprint_hash,
|
|
|
|
SSH_FP_DEFAULT)) == NULL) {
|
2020-10-18 13:32:01 +02:00
|
|
|
error_f("sshkey_fingerprint failed");
|
2015-01-30 12:43:14 +01:00
|
|
|
goto out;
|
2001-04-13 01:34:34 +02:00
|
|
|
}
|
2022-01-06 23:06:51 +01:00
|
|
|
debug_f("trying hostkey %s %s using sigalg %s",
|
|
|
|
sshkey_ssh_name(private), fp, authctxt->active_ktype);
|
2014-12-11 06:13:28 +01:00
|
|
|
|
2001-11-12 01:02:03 +01:00
|
|
|
/* figure out a name for the client host */
|
2019-01-19 22:40:21 +01:00
|
|
|
lname = get_local_name(ssh_packet_get_connection_in(ssh));
|
|
|
|
if (lname == NULL) {
|
2020-10-18 13:32:01 +02:00
|
|
|
error_f("cannot get local ipaddr/name");
|
2015-01-30 12:43:14 +01:00
|
|
|
goto out;
|
2001-11-12 01:02:03 +01:00
|
|
|
}
|
2015-01-30 12:43:14 +01:00
|
|
|
|
|
|
|
/* XXX sshbuf_put_stringf? */
|
|
|
|
xasprintf(&chost, "%s.", lname);
|
2020-10-18 13:32:01 +02:00
|
|
|
debug2_f("chost %s", chost);
|
2001-11-12 01:02:03 +01:00
|
|
|
|
2001-04-13 01:34:34 +02:00
|
|
|
/* construct data */
|
2015-01-30 12:43:14 +01:00
|
|
|
if ((b = sshbuf_new()) == NULL) {
|
2020-10-18 13:32:01 +02:00
|
|
|
error_f("sshbuf_new failed");
|
2015-01-30 12:43:14 +01:00
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
if ((r = sshkey_to_blob(private, &keyblob, &keylen)) != 0) {
|
2020-10-18 13:32:01 +02:00
|
|
|
error_fr(r, "sshkey_to_blob");
|
2015-01-30 12:43:14 +01:00
|
|
|
goto out;
|
|
|
|
}
|
2021-01-27 11:05:28 +01:00
|
|
|
if ((r = sshbuf_put_stringb(b, ssh->kex->session_id)) != 0 ||
|
2015-01-30 12:43:14 +01:00
|
|
|
(r = sshbuf_put_u8(b, SSH2_MSG_USERAUTH_REQUEST)) != 0 ||
|
|
|
|
(r = sshbuf_put_cstring(b, authctxt->server_user)) != 0 ||
|
2018-01-23 06:27:21 +01:00
|
|
|
(r = sshbuf_put_cstring(b, authctxt->service)) != 0 ||
|
2015-01-30 12:43:14 +01:00
|
|
|
(r = sshbuf_put_cstring(b, authctxt->method->name)) != 0 ||
|
2021-01-25 07:00:17 +01:00
|
|
|
(r = sshbuf_put_cstring(b, authctxt->active_ktype)) != 0 ||
|
2015-01-30 12:43:14 +01:00
|
|
|
(r = sshbuf_put_string(b, keyblob, keylen)) != 0 ||
|
|
|
|
(r = sshbuf_put_cstring(b, chost)) != 0 ||
|
|
|
|
(r = sshbuf_put_cstring(b, authctxt->local_user)) != 0) {
|
2020-10-18 13:32:01 +02:00
|
|
|
error_fr(r, "buffer error");
|
2015-01-30 12:43:14 +01:00
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
2001-04-13 01:34:34 +02:00
|
|
|
#ifdef DEBUG_PK
|
2015-01-30 12:43:14 +01:00
|
|
|
sshbuf_dump(b, stderr);
|
2001-04-13 01:34:34 +02:00
|
|
|
#endif
|
2019-01-19 22:34:45 +01:00
|
|
|
if ((r = ssh_keysign(ssh, private, &sig, &siglen,
|
|
|
|
sshbuf_ptr(b), sshbuf_len(b))) != 0) {
|
2015-01-30 12:43:14 +01:00
|
|
|
error("sign using hostkey %s %s failed",
|
|
|
|
sshkey_ssh_name(private), fp);
|
|
|
|
goto out;
|
2001-04-13 01:34:34 +02:00
|
|
|
}
|
2015-01-30 12:43:14 +01:00
|
|
|
if ((r = sshpkt_start(ssh, SSH2_MSG_USERAUTH_REQUEST)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->server_user)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->service)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->method->name)) != 0 ||
|
2021-01-25 07:00:17 +01:00
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->active_ktype)) != 0 ||
|
2015-01-30 12:43:14 +01:00
|
|
|
(r = sshpkt_put_string(ssh, keyblob, keylen)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, chost)) != 0 ||
|
|
|
|
(r = sshpkt_put_cstring(ssh, authctxt->local_user)) != 0 ||
|
|
|
|
(r = sshpkt_put_string(ssh, sig, siglen)) != 0 ||
|
|
|
|
(r = sshpkt_send(ssh)) != 0) {
|
2020-10-18 13:32:01 +02:00
|
|
|
error_fr(r, "packet error");
|
2015-01-30 12:43:14 +01:00
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
success = 1;
|
|
|
|
|
|
|
|
out:
|
2018-07-09 23:03:30 +02:00
|
|
|
if (sig != NULL)
|
|
|
|
freezero(sig, siglen);
|
2015-01-30 12:43:14 +01:00
|
|
|
free(keyblob);
|
|
|
|
free(lname);
|
|
|
|
free(fp);
|
2013-06-01 23:31:17 +02:00
|
|
|
free(chost);
|
2015-01-30 12:43:14 +01:00
|
|
|
sshkey_free(private);
|
|
|
|
sshbuf_free(b);
|
2001-04-13 01:34:34 +02:00
|
|
|
|
2015-01-30 12:43:14 +01:00
|
|
|
return success;
|
2001-04-13 01:34:34 +02:00
|
|
|
}
|
|
|
|
|
2000-09-23 08:15:56 +02:00
|
|
|
/* find auth method */
|
|
|
|
|
|
|
|
/*
|
|
|
|
* given auth method name, if configurable options permit this method fill
|
|
|
|
* in auth_ident field and return true, otherwise return false.
|
|
|
|
*/
|
2001-06-25 07:01:22 +02:00
|
|
|
static int
|
2000-09-23 08:15:56 +02:00
|
|
|
authmethod_is_enabled(Authmethod *method)
|
|
|
|
{
|
|
|
|
if (method == NULL)
|
|
|
|
return 0;
|
|
|
|
/* return false if options indicate this method is disabled */
|
|
|
|
if (method->enabled == NULL || *method->enabled == 0)
|
|
|
|
return 0;
|
|
|
|
/* return false if batch mode is enabled but method needs interactive mode */
|
|
|
|
if (method->batch_flag != NULL && *method->batch_flag != 0)
|
|
|
|
return 0;
|
|
|
|
return 1;
|
|
|
|
}
|
2000-04-29 15:57:08 +02:00
|
|
|
|
2001-06-25 07:01:22 +02:00
|
|
|
static Authmethod *
|
2000-09-23 08:15:56 +02:00
|
|
|
authmethod_lookup(const char *name)
|
|
|
|
{
|
|
|
|
Authmethod *method = NULL;
|
|
|
|
if (name != NULL)
|
|
|
|
for (method = authmethods; method->name != NULL; method++)
|
|
|
|
if (strcmp(name, method->name) == 0)
|
|
|
|
return method;
|
|
|
|
debug2("Unrecognized authentication method name: %s", name ? name : "NULL");
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2001-03-11 02:49:19 +01:00
|
|
|
/* XXX internal state */
|
|
|
|
static Authmethod *current = NULL;
|
|
|
|
static char *supported = NULL;
|
|
|
|
static char *preferred = NULL;
|
2002-06-23 23:23:20 +02:00
|
|
|
|
2000-09-23 08:15:56 +02:00
|
|
|
/*
|
|
|
|
* Given the authentication method list sent by the server, return the
|
|
|
|
* next method we should try. If the server initially sends a nil list,
|
2001-04-06 01:26:32 +02:00
|
|
|
* use a built-in default list.
|
2001-02-05 13:42:17 +01:00
|
|
|
*/
|
2001-06-25 07:01:22 +02:00
|
|
|
static Authmethod *
|
2000-09-23 08:15:56 +02:00
|
|
|
authmethod_get(char *authlist)
|
|
|
|
{
|
2001-03-11 02:49:19 +01:00
|
|
|
char *name = NULL;
|
2002-02-26 19:15:09 +01:00
|
|
|
u_int next;
|
2001-02-05 13:42:17 +01:00
|
|
|
|
2000-09-23 08:15:56 +02:00
|
|
|
/* Use a suitable default if we're passed a nil list. */
|
|
|
|
if (authlist == NULL || strlen(authlist) == 0)
|
2001-03-11 02:49:19 +01:00
|
|
|
authlist = options.preferred_authentications;
|
|
|
|
|
|
|
|
if (supported == NULL || strcmp(authlist, supported) != 0) {
|
|
|
|
debug3("start over, passed a different list %s", authlist);
|
2013-06-01 23:31:17 +02:00
|
|
|
free(supported);
|
2001-03-11 02:49:19 +01:00
|
|
|
supported = xstrdup(authlist);
|
|
|
|
preferred = options.preferred_authentications;
|
|
|
|
debug3("preferred %s", preferred);
|
|
|
|
current = NULL;
|
|
|
|
} else if (current != NULL && authmethod_is_enabled(current))
|
|
|
|
return current;
|
2000-09-23 08:15:56 +02:00
|
|
|
|
2001-03-11 02:49:19 +01:00
|
|
|
for (;;) {
|
|
|
|
if ((name = match_list(preferred, supported, &next)) == NULL) {
|
2002-12-23 03:44:36 +01:00
|
|
|
debug("No more authentication methods to try.");
|
2001-03-11 02:49:19 +01:00
|
|
|
current = NULL;
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
preferred += next;
|
2000-10-14 07:23:11 +02:00
|
|
|
debug3("authmethod_lookup %s", name);
|
2001-03-11 02:49:19 +01:00
|
|
|
debug3("remaining preferred: %s", preferred);
|
|
|
|
if ((current = authmethod_lookup(name)) != NULL &&
|
|
|
|
authmethod_is_enabled(current)) {
|
2000-10-14 07:23:11 +02:00
|
|
|
debug3("authmethod_is_enabled %s", name);
|
2002-12-23 03:44:36 +01:00
|
|
|
debug("Next authentication method: %s", name);
|
2013-06-01 23:31:17 +02:00
|
|
|
free(name);
|
2001-03-11 02:49:19 +01:00
|
|
|
return current;
|
2000-10-14 07:23:11 +02:00
|
|
|
}
|
2013-06-06 00:22:05 +02:00
|
|
|
free(name);
|
2000-04-29 15:57:08 +02:00
|
|
|
}
|
2001-03-11 02:49:19 +01:00
|
|
|
}
|
2000-10-28 05:19:58 +02:00
|
|
|
|
2001-07-04 05:42:30 +02:00
|
|
|
static char *
|
2001-03-11 02:49:19 +01:00
|
|
|
authmethods_get(void)
|
|
|
|
{
|
|
|
|
Authmethod *method = NULL;
|
2018-07-09 23:03:30 +02:00
|
|
|
struct sshbuf *b;
|
2002-01-22 13:26:38 +01:00
|
|
|
char *list;
|
2018-07-09 23:03:30 +02:00
|
|
|
int r;
|
2001-03-11 02:49:19 +01:00
|
|
|
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((b = sshbuf_new()) == NULL)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_f("sshbuf_new failed");
|
2001-03-11 02:49:19 +01:00
|
|
|
for (method = authmethods; method->name != NULL; method++) {
|
|
|
|
if (authmethod_is_enabled(method)) {
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((r = sshbuf_putf(b, "%s%s",
|
|
|
|
sshbuf_len(b) ? "," : "", method->name)) != 0)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_fr(r, "buffer error");
|
2001-03-11 02:49:19 +01:00
|
|
|
}
|
|
|
|
}
|
2018-07-09 23:03:30 +02:00
|
|
|
if ((list = sshbuf_dup_string(b)) == NULL)
|
2020-10-18 13:32:01 +02:00
|
|
|
fatal_f("sshbuf_dup_string failed");
|
2018-07-09 23:03:30 +02:00
|
|
|
sshbuf_free(b);
|
2002-01-22 13:26:38 +01:00
|
|
|
return list;
|
2000-04-29 15:57:08 +02:00
|
|
|
}
|