tyler.durden/icinga2
1
0
Fork 0
mirror of https://github.com/Icinga/icinga2.git synced 2025-04-08 17:05:25 +02:00
Code Issues Packages Projects Releases Wiki Activity
11,265 Commits 421 Branches 120 Tags
Commit Graph

4884 Commits

Author SHA1 Message Date
Michael Friedrich
6cce9c0fdd API: Automatically repair broken packages 
This partially reverts #7150 and avoids exceptions
inside the flow. Each time an empty active stage
is detected, Icinga tries to repair it from the
the given directory tree.

Also, the code now takes into account that it should
create the package storage on startup, whether within
the API object, or if disabled, inside the application.

Caching the active stages for packages in memory
only is in effect with the API feature being enabled.
This is useful for other deployed config packages,
not only the internal one.

fixes #7173
refs #7150
refs #7119
fixes #6959
 2019-05-10 12:48:34 +02:00
Elias Ohm
4c86c370bb fixup errbuf length in the other files and avoid using the static buffer in one place (for thread safety and code consistency reasons) 2019-05-09 09:30:12 +02:00
Elias Ohm
e75f063552 bring some things in line 
- account for documented buffer size openssl 1.1.x for error string (>=256 bytes)
- use nullptr instead of NULL
- fix/streamline null-checks
 2019-05-09 00:22:24 +02:00
Jean Flach
9a0d894f10 Don't use deprecated RSA_generate_key 
fixes #4635
 2019-05-08 23:46:31 +02:00
Michael Friedrich
03324b2fb6 Config packages: Catch active stage exceptions in rare cases 
Typically this already is detected on startup.
 2019-05-08 16:43:27 +02:00
Michael Friedrich
704aabcb63 Avoid dead-lock with config packages and active stages 2019-05-08 16:06:46 +02:00
Michael Friedrich
736e0806d7
Merge pull request #7164 from Icinga/bugfix/notification-times-validate 
Improve validation for times.{begin,end} in notification objects
 2019-05-07 15:58:44 +02:00
Michael Friedrich
296fc06890
Merge pull request #7163 from Icinga/bugfix/db-ido-reachable 
DB IDO: Use cached reachable state
 2019-05-07 15:21:21 +02:00
Michael Friedrich
8ae206cd5d Improve validation for times.{begin,end} in notification objects 
fixes #6939
 2019-05-07 15:20:06 +02:00
Michael Friedrich
5553438249 DB IDO: Use cached reachable state 
fixes #6844
 2019-05-07 13:47:09 +02:00
Michael Friedrich
00bc0b2303 CLI: Return non-zero on unknown sub commands 
fixes #6585
 2019-05-07 12:43:53 +02:00
Michael Friedrich
4197bc9bcd CLI: Fix updates for NodeName/ZoneName constants 
fixes #7117
 2019-05-06 10:19:56 +02:00
Michael Friedrich
edaaaae1e8
Merge pull request #7155 from Elias481/bugfix/evaluatefilter-assign-this-scope-6874 
use current frame scope for permission filter function calls
 2019-05-03 16:53:40 +02:00
Michael Friedrich
78e24c53f1 DB IDO: Do not deactivate objects during application reload/restart 
This follows the same principle as with the shutdown handler,
and was introduced with the changed reload handling with 2.9.
Previously IsShuttingDown() was sufficient which got set at one
location.

SigUsr2 as handler introduced a new location where m_ShuttingDown
is not necessarily set yet. Since this handler gets called when
l_Restarting is enabled, we'll use this flag to avoid config update
events resulting in object deactivation (object->IsActive() always
returns false).

refs #5996
refs #6691
refs #6970

fixes #7125
 2019-05-03 15:40:48 +02:00
htriem
75df3879f2 Implement sleep CheckCommand in memory 
Implements a check task with Utility::Sleep and custom var parameter sleep_time (default value: 1s)

refs #6964
 2019-05-02 16:24:42 +02:00
Elias Ohm
c10ff9dd72 try without initialization of frame Locals which are not used for permissions filter and as far as I can see also not for query filters 2019-05-02 09:03:30 +02:00
Elias Ohm
53febdea81 use current frame scope for permission filter function calls 2019-05-02 07:35:19 +02:00
Elias Ohm
cdd843a998 another small adjustment by the way just to ensure the object on stack ist the same as the one serialized further in case the object does not implement locking on mutation (besides it's mor efficient to not fetch the same value twice) 2019-05-01 12:09:24 +02:00
Elias Ohm
44ac6cf1ec add some object locking to the Dump method (which could theoreticylly suffer from same reace condition as serializer) 2019-05-01 11:49:07 +02:00
Michael Friedrich
759b090f81
Merge pull request #7150 from Icinga/bugfix/api-config-package-active-stage-name 
Ensure that runtime created API objects survive a restart
 2019-04-30 14:22:13 +02:00
Michael Friedrich
f206cba394
Merge pull request #7152 from Elias481/fix/mysql8-headers-compatibility 
account for adjusted interface of mysql8
 2019-04-30 14:20:18 +02:00
Michael Friedrich
502c43fb12 Active packages: Don't try to fix broken config packages which are not cached yet 2019-04-30 12:19:35 +02:00
Michael Friedrich
8a258de9bc
Merge pull request #6734 from leeclemens/remove-redundant-indexes 
db ido: remove redundant mysql indexes
 2019-04-29 09:13:22 +02:00
Elias Ohm
615f019c2e account for adjusted interface of mysql8 (now utilizes c99 bools instead of my_bool labelled chars) 2019-04-29 00:18:15 +02:00
Elias Ohm
91296c2a25 Lock Objects during serialization 
old behaviour was to only lock arrays, dictionaries and namespaces but not other objects
 2019-04-28 22:13:19 +02:00
Michael Friedrich
2bca7a5bb5 Repair broken API config packages at runtime 
This means a new timer which checks every 5m whether the
active-stage can be read, and if not, it overwrites the
file on disk with the details from memory.
 2019-04-26 14:53:36 +02:00
Michael Friedrich
f92c134b0a Cluster: Don't try to sync objects from broken _api package 2019-04-26 14:43:38 +02:00
Michael Friedrich
0d6d48fd59 Daemon: Deal with exceptions from broken _api package 2019-04-26 14:43:10 +02:00
Michael Friedrich
c821e73364 Cache the API package stage name with a active-stage fallback 
This prevents reading the file everytime the stageName is required
for when creating a runtime object via REST API.
 2019-04-26 13:40:27 +02:00
Michael Friedrich
1078a0a824 Add --cn parameter to 'api setup' CLI command allowing hostname overrides 
fixes #6649
 2019-04-26 10:52:05 +02:00
Michael Friedrich
3dc9927284
Merge pull request #7124 from Icinga/bugfix/namespace-thread-safe 
Namespace: place ObjectLock in all methods
 2019-04-26 08:26:59 +02:00
Michael Friedrich
37de1a919b
Merge pull request #7088 from Icinga/feature/asio-event-queue 
Implement new event queue for ASIO consumers
 2019-04-25 16:54:43 +02:00
Michael Friedrich
a7873da89d Eventqueue: Remove unused code 2019-04-25 16:21:07 +02:00
Alexander A. Klimov
e86e3cc234 EventsFilter#Push(): ensure not to modify the global namespace 2019-04-25 15:56:38 +02:00
Alexander A. Klimov
c209cf830b /v1/events: don't over-consume CPU-bound threads 2019-04-25 15:56:38 +02:00
Alexander A. Klimov
5e8b4280bc New event queue: handle empty filter 2019-04-25 15:56:38 +02:00
Alexander A. Klimov
94db282fd1 /v1/events: remove anti-deadlock hack 2019-04-25 15:56:38 +02:00
Alexander A. Klimov
81713d0509 /v1/events: use new event queue 2019-04-25 15:56:38 +02:00
Alexander A. Klimov
90d9cd9257 Feed new event queue with events 2019-04-25 15:56:38 +02:00
Alexander A. Klimov
7688994601 Implement new event queue for ASIO consumers 2019-04-25 15:56:38 +02:00
Michael Friedrich
a630d0185f
Merge pull request #6722 from Icinga/feature/notification-result 
Add notification result store/sync
 2019-04-25 15:56:14 +02:00
Michael Friedrich
0438c866f8
Merge pull request #7102 from Icinga/feature/boost-fs-7101 
Replace self-written filesystem ops with boost.filesystem
 2019-04-25 15:53:55 +02:00
Alexander A. Klimov
5afef1015d Replace unlink() with boost::filesystem::remove() 
refs #7101
 2019-04-25 09:53:02 +02:00
Alexander A. Klimov
5a17722c1f Replace _unlink() + rename() with boost::filesystem::rename() 
refs #7101
 2019-04-25 09:53:02 +02:00
Alexander A. Klimov
f1f7d0c4d6 Work around boost::filesystem::path bug on VS 
refs #7101
 2019-04-25 09:53:01 +02:00
Alexander A. Klimov
af78cd6050 Use Boost.Filesystem 
refs #7101
 2019-04-25 09:53:01 +02:00
Michael Friedrich
0d9d39c64b Fix preprocessor macro comment 2019-04-25 08:25:28 +02:00
Alexander A. Klimov
ba842403ce Fix circular #include 
refs #6985
 2019-04-25 08:25:28 +02:00
Alexander A. Klimov
5151f6567e ThreadPool: use the Boost ASIO thread pool under the hood 2019-04-25 08:25:28 +02:00
Michael Friedrich
56894bea17 Buildfix 
Obviously tired.

fixes #7138
fixes #7139
 2019-04-24 12:10:57 +02:00