Jesse Glick
63a29a90ee
Merge branch 'master' into BlockedBecauseOfBuildInProgress
2015-08-06 14:58:54 -04:00
Stephen Connolly
030793f01d
[ FIXES #1777 ] Make accessing pendingLaunches lock free
...
- My objection to #1777 is that it is introducing more locking. The ultimate aim is to make
node provisioning and queue maintenance completely lock free. Thus adding more locks than
is strictly required is a bad plan.
- We cannot completely remove provisioningLock at this time as it is used to ensure that
only one provisioning activity can take place for a given label at a time.
- This change ensures that the provisioningLock is only used by the NodeProvisioner for
its internal thread guarantees and is thus no-longer used accidentally by any
external threads
2015-08-06 11:03:40 +01:00
Daniel Beck
a798aa427f
1.623 has no notable changes
2015-08-06 11:42:07 +02:00
Stephen Connolly
0ba505b60c
[JENKINS-28690] Aha! So I believe this will fully resolve any of these kinds of deadlocks
...
- Without this, then it becomes a question of find catch and release for each potential code path that
might end up restoring the interrupt flag on the current thread.
- Since standard Lock support is kind enough to restore the interrupt flag on the current thread
when blocked waiting for the lock, that would be a hiding to nothing
- I welcome others to review my logic detailed in the code comment
- I am leaving the code comment as this is IMHO too important to assume that somebody will
check the git commit history
2015-08-06 10:39:54 +01:00
Vincent Latombe
4d3285398b
Remove administrativeMonitor in loop to catch hypothetical dups
2015-08-05 12:38:13 +02:00
Vincent Latombe
7dfa765d8b
[FIXED JENKINS-29798] Failed to listen to incoming slave connection
...
Factor tcpSlaveAgentListener logic to a single method able to shutdown
existing listener if required, then launch a new one if it needs to.
Also, the administrative monitor gets cleared if we had previously a
taken port and changed to a valid port.
2015-08-05 12:11:18 +02:00
Nicolas De Loof
d331c12bbc
prefer Computer#isUnix to be abstract
...
few chances there’s another implementation
but if one exists Function.isWindows would probably be wrong implementation
2015-08-04 16:27:31 +02:00
Oliver Gondža
585f9b1a12
Merge pull request #1780 from olivergondza/defensive-getCanonicalName
...
Avoid "EmptyStackException: null" in Items#getCanonicalName
2015-08-04 15:37:38 +02:00
Jesse Glick
50c28171a2
Merge branch 'master' into BlockedBecauseOfBuildInProgress
...
Conflicts:
core/src/main/resources/hudson/model/Messages.properties
core/src/main/resources/hudson/model/Messages_de.properties
core/src/main/resources/hudson/model/Messages_it.properties
core/src/main/resources/hudson/model/Messages_ja.properties
core/src/main/resources/hudson/model/Messages_zh_TW.properties
core/src/main/resources/jenkins/model/Messages.properties
core/src/main/resources/jenkins/model/Messages_da.properties
core/src/main/resources/jenkins/model/Messages_de.properties
core/src/main/resources/jenkins/model/Messages_es.properties
core/src/main/resources/jenkins/model/Messages_fr.properties
core/src/main/resources/jenkins/model/Messages_it.properties
core/src/main/resources/jenkins/model/Messages_ja.properties
core/src/main/resources/jenkins/model/Messages_nl.properties
core/src/main/resources/jenkins/model/Messages_pt_BR.properties
core/src/main/resources/jenkins/model/Messages_ru.properties
core/src/main/resources/jenkins/model/Messages_tr.properties
core/src/main/resources/jenkins/model/Messages_zh_CN.properties
core/src/main/resources/jenkins/model/Messages_zh_TW.properties
2015-08-04 08:05:14 -04:00
Nicolas De loof
c3bc2214db
Merge pull request #1781 from ndeloof/isUnix
...
Make it easier to determine Computer is Unix or Windows
2015-08-04 11:18:29 +02:00
Nicolas De Loof
dcf8bc753b
Make it easier to determine Computer is Unix or Windows
2015-08-04 10:12:03 +02:00
Oliver Gondža
4589ce5d66
Avoid "EmptyStackException: null" in Items#getCanonicalName
2015-08-04 09:33:41 +02:00
Jesse Glick
1fbc4fdac0
Merge pull request #1779 from jglick/BUILD_NOW_TEXT-JENKINS-26147
...
[JENKINS-26147] New home for BUILD_NOW_TEXT
2015-08-03 10:04:32 -04:00
Kohsuke Kawaguchi
668cf95f26
updated changelog for release
2015-08-02 21:32:23 -07:00
Kohsuke Kawaguchi
2391c4fc43
[maven-release-plugin] prepare for next development iteration
2015-08-02 21:26:37 -07:00
Kohsuke Kawaguchi
bb37842ab4
[maven-release-plugin] prepare release jenkins-1.623
jenkins-1.623
2015-08-02 21:26:36 -07:00
Jesse Glick
6bf099d322
Moving some messages to the natural package.
2015-07-31 17:37:43 -04:00
Jesse Glick
3f7875c150
[FIXED JENKINS-26147] Move BUILD_NOW_TEXT to ParameterizedJobMixIn.
2015-07-31 17:25:45 -04:00
Jesse Glick
ac1cc266a3
Merge pull request #1778 from jglick/ExceptionInterruption
...
Deleting bogus class ExceptionInterruption
2015-07-31 10:10:27 -04:00
Jesse Glick
98c8caf51b
Revert "Adding another cause"
...
This reverts commit 68b383e0b2 .
2015-07-31 07:03:24 -04:00
Jesse Glick
90fb626496
Merge pull request #1776 from jglick/NodeProvisioner-log-level
...
Reduce logging level
2015-07-31 06:55:34 -04:00
Jesse Glick
db559bd6f5
Reduce logging level of “queue length is less than the available capacity” message.
2015-07-30 16:28:12 -04:00
Stephen Connolly
2906718ff5
[FIXED JENKINS-28844] The curious case of the Channel memory cycles
...
Pick up the remoting 2.52 release
(cherry picked from commit aac20468ff )
2015-07-30 20:46:11 +02:00
Stephen Connolly
b591a241b4
[FIXED JENKINS-28058] Don't send a reference to Computer.class over remoting channels
...
- Logging added to help diagnose was actually causing Jenkins.class to be initialized on the remoting side
(cherry picked from commit df831325b6 )
2015-07-30 20:24:42 +02:00
Oliver Gondža
132b915615
[FIXED JENKINS-29286] Use byte size to sort disk space monitor columns
...
(cherry picked from commit fabec0eef0 )
2015-07-30 19:56:17 +02:00
Oliver Gondža
8f5bdcb414
[FIXED JENKINS-29376] Rename help texts as well
...
Amends eec43983a3 .
(cherry picked from commit 1c1d5a95d9 )
2015-07-30 19:53:30 +02:00
Oliver Gondža
1fd08b1b96
Merge pull request #1773 from olivergondza/result-null-safety
...
Null safety for hudson.model.Result
2015-07-29 19:55:52 +02:00
Daniel Beck
1598a464ff
Noting #1770 in 1.622, JENKINS-29568
2015-07-29 15:35:09 +02:00
Kohsuke Kawaguchi
7613069799
[maven-release-plugin] prepare for next development iteration
2015-07-28 08:46:49 -07:00
Kohsuke Kawaguchi
88f9f4c8fc
[maven-release-plugin] prepare release jenkins-1.609.2
jenkins-1.609.2
2015-07-28 08:46:47 -07:00
Kohsuke Kawaguchi
64a389ad9d
updated changelog for release
2015-07-27 07:24:10 -07:00
Kohsuke Kawaguchi
5712881069
[maven-release-plugin] prepare for next development iteration
2015-07-26 21:39:19 -07:00
Kohsuke Kawaguchi
e82f0094a0
[maven-release-plugin] prepare release jenkins-1.622
jenkins-1.622
2015-07-26 21:39:18 -07:00
Kohsuke Kawaguchi
92afdbe660
Merge remote-tracking branch 'origin/master'
2015-07-25 13:05:09 -07:00
Oliver Gondža
90f29f8cbc
Null safety for hudson.model.Result
2015-07-23 16:35:45 +02:00
James Nord
ed2988fc86
Merge pull request #1765 from eidottermihi/log4j-over-slf4j
...
Use log4j-bridge of slf4j and dont ship log4j.
2015-07-23 15:03:13 +01:00
Stephen Connolly
4f0ca16b01
[FIXED JENKINS-29568] A better fix
...
- 252e129609 doesn't work for all cases as there are code paths
where a pending launch can be removed from the pending list and not have spent() called.
- There was no reason for iterating the list twice anyway, as all of this takes place with the locks held
- My notifying each one as we process, if there is an Error, we will not leave any stranded. The next run
through, if there is one, will cover those instances.
2015-07-23 09:08:40 +01:00
Kohsuke Kawaguchi
4ece8c06aa
Added a test case to verify the behaviour of rowSet
...
... as a control case in the master branch to ensure the changes in the
config-ui-changes do not break its behaviour.
2015-07-22 18:03:54 -07:00
Kohsuke Kawaguchi
ab133abf9f
Merge pull request #1770
2015-07-22 11:20:09 -07:00
Jesse Glick
e55f356ed0
Merged #1771 .
2015-07-22 11:13:54 -04:00
Jesse Glick
44d1d2b75f
Offering new convenience methods in ParameterizedJobMixIn.
2015-07-22 09:39:10 -04:00
Stephen Connolly
252e129609
The call to PlannedNode.spent() *must* always occur
...
- They have been removed from the pendingLaunches, so once we leave this method we have no reference to them
- Arguably the call to spent should take place when removing from the pendingLaunches, but that would change
the method contract
2015-07-22 10:34:37 +01:00
Craig Rodrigues
05c6e66f07
akuma 1.10 adds support for FreeBSD
2015-07-20 13:07:51 -04:00
Kohsuke Kawaguchi
e7f3f8b211
updated changelog for release
2015-07-19 11:50:13 -07:00
Kohsuke Kawaguchi
fc90a4cdb8
[maven-release-plugin] prepare for next development iteration
2015-07-19 11:43:59 -07:00
Kohsuke Kawaguchi
0cc691578c
[maven-release-plugin] prepare release jenkins-1.621
jenkins-1.621
2015-07-19 11:43:58 -07:00
Oleg Nenashev
4c7daf643f
Merge pull request #1768 from eitoball/fix-typo
...
Fix typo
2015-07-17 20:34:02 +03:00
Eito Katagiri
669f94f4c3
Fix typo
2015-07-18 01:59:13 +09:00
Peter Marcoen
2a75ca1ffe
[FIXED JENKINS-2327] Do not display 'No changes' if changelog calculation is still in progress
2015-07-17 14:01:57 +02:00
Daniel Beck
f6a2fc17b9
Noting #1617 , fix broken changelog
2015-07-16 15:18:27 +02:00