fully unbreak: some $SSH invocations did not have -F
specified and could pick up the ~/.ssh/config of the user running the tests
Upstream-Regress-ID: f362d1892c0d3e66212d5d3fc02d915c58ef6b89
revise keys/principals command hang fix (bz#2655) to
consume entire output, avoiding sending SIGPIPE to subprocesses early; ok
dtucker@
Upstream-ID: 7cb04b31a61f8c78c4e48ceededcd2fd5c4ee1bc
small cleanup post SSHv1 removal:
remove SSHv1-isms in commented examples
reorder token table to group deprecated and compile-time conditional tokens
better
fix config dumping code for some compile-time conditional options that
weren't being correctly skipped (SSHv1 and PKCS#11)
Upstream-ID: f2e96b3cb3158d857c5a91ad2e15925df3060105
don't dereference authctxt before testing != NULL, it
causes compilers to make assumptions; from Karsten Weiss
Upstream-ID: 794243aad1e976ebc717885b7a97a25e00c031b2
Account for timeouts in the integrity tests as failures.
If the first test in a series for a given MAC happens to modify the low
bytes of a packet length, then ssh will time out and this will be
interpreted as a test failure. Patch from cjwatson at debian.org via
bz#2658.
Upstream-Regress-ID: e7467613b0badedaa300bc6fc7495ec2f44e2fb9
Make forwarding test less racy by using unix domain
sockets instead of TCP ports where possible. Patch from cjwatson at
debian.org via bz#2659.
Upstream-Regress-ID: 4756375aac5916ef9d25452a1c1d5fa9e90299a9
Fix typo in ~C error message for bad port forward
cancellation. bz#2672, from Brad Marshall via Colin Watson and Ubuntu's
bugtracker.
Upstream-ID: 0d4a7e5ead6cc59c9a44b4c1e5435ab3aada09af
The POSIX APIs that that sockaddrs all ignore the s*_len
field in the incoming socket, so userspace doesn't need to set it unless it
has its own reasons for tracking the size along with the sockaddr.
ok phessler@ deraadt@ florian@
Upstream-ID: ca6e49e2f22f2b9e81d6d924b90ecd7e422e7437
sshd_config is documented to set
GSSAPIStrictAcceptorCheck=yes by default, so actually make it do this.
bz#2637 ok dtucker
Upstream-ID: 99ef8ac51f17f0f7aec166cb2e34228d4d72a665
Avoid confusing error message when attempting to use
ssh-keyscan built without SSH protocol v.1 to scan for v.1 keys; bz#2583
Upstream-ID: 5d214abd3a21337d67c6dcc5aa6f313298d0d165
Re-add '%k' token for AuthorizedKeysCommand which was
lost during the re-org in rev 1.235. bz#2656, from jboning at gmail.com.
Upstream-ID: 2884e203c02764d7b3fe7472710d9c24bdc73e38
https://github.com/PowerShell/Win32-OpenSSH/issues/479https://github.com/PowerShell/Win32-OpenSSH/issues/476https://github.com/PowerShell/Win32-OpenSSH/issues/474https://github.com/PowerShell/Win32-OpenSSH/issues/467
bug #479 - "ls c:" is not working
sanitized_path() is modified to handle the edge case "\x:"
2.bug #476 - "cd c:" is not working
If "c:" is passed to _fullpath() then it is returning existing path but not "c:", so if we append "\" to "c:" then it is working fine.
3.bug #474 - code cleanup MAX_PATH
Using PATH_MAX variable instead of MAX_PATH
In dirent.h, used PATH_MAX instead of hardcoding 256 characters
In readdir(), changed the pdirentry to be a static variable. Before this, we are leaking the memory.
4.bug #467 - SFTP rename failed if the newpath already exists.
The _wrename throws error if the newpath exists.
To make it consistent with the linux behavrior,
a) if the newpath is a file and if it exists then delete it so that _wrename will succeed.
b) if the newpath is a directory and if it is empty then delete it so that _wrename will succeed.
1. Reduce the console output of build
2. move to use powershell core 6.0.0.14
3. one minor fix in pester tests.
* Update the path of unittest results and build log. suppress the warning message.
* limitoutput of choco installation.
* fix of quotes
* remove redundant log
* Set x64 Release build as default
