There are sill issues with maven2 builds and reporters

git-svn-id: https://hudson.dev.java.net/svn/hudson/branches/main-maven3-support@36806 71c3de6d-444a-0410-be80-ed276b4c234a

to be able to get in an other place (sure git could be better :-) )
some issue when using maven 2
some stuff to do more in maven 3 build (recording mojo execution etc...)



git-svn-id: https://hudson.dev.java.net/svn/hudson/branches/main-maven3-support@36643 71c3de6d-444a-0410-be80-ed276b4c234a

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@34835 71c3de6d-444a-0410-be80-ed276b4c234a

[FIXED HUDSON-3273] Increase timeout to 30 seconds since 10 seconds isn't enough for machines with poor resources.

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@30733 71c3de6d-444a-0410-be80-ed276b4c234a

The next step is to write a new bootstrap code that works with plexus classworlds, which is package-renamed.

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@26722 71c3de6d-444a-0410-be80-ed276b4c234a

[FIXED HUDSON-2932] Maven projects now properly fall back through project-defined MAVEN_OPTS to globally-defined MAVEN_OPTS to executor environment variable MAVEN_OPTS - also added globally-defined MAVEN_OPTS and related changes

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@24652 71c3de6d-444a-0410-be80-ed276b4c234a

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@19759 71c3de6d-444a-0410-be80-ed276b4c234a

    (<a href="https://hudson.dev.java.net/issues/show_bug.cgi?id=3644">issue 3644</a>)


git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@19452 71c3de6d-444a-0410-be80-ed276b4c234a

finally introduced the builder pattern for Launcher, to stop combinatorial explosion of the launch method

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@18944 71c3de6d-444a-0410-be80-ed276b4c234a

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@18266 71c3de6d-444a-0410-be80-ed276b4c234a

[FIXED HUDSON-2373] Modified to work with Maven 2.1. Sorry for a long delay, folks. This change will be in 1.305.

We basically need an ability to feed different components.xml depending on Maven's version (which will be read by DefaultComponentDiscoverer in Plexus.) The implementation in this fix is hacky, and based on placing one more jar in front of the classpath when run on Maven 2.1, while maven-interceptor still contains the default components.xml for 2.0.x.

Ideally 2.0.x components.xml should move to its own jar and so that we can switch between them, as opposed to pass in extra jar only when it's Maven 2.1.

We'll see how Maven 3.0 works with this regard.

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@18036 71c3de6d-444a-0410-be80-ed276b4c234a

Merged revisions 17324-17325,17428,17430,17473-17475,17481-17482,17555-17563,17571,17593-17595,17619-17620,17637-17644,17656-17657,17663-17665,17667-17670,17730,17753,17755-17758,17763-17768,17770-17771,17901,17906,17911-17912,17958,17961-17962,17970-17977 via svnmerge from 
https://www.dev.java.net/svn/hudson/branches/tool-autoinstallation

........
  r17324 | jglick | 2009-04-20 18:24:32 -0700 (Mon, 20 Apr 2009) | 2 lines
  
  Defensive cloning and Javadoc corrections.
  (Though it seems no one is using the default impls of these methods yet anyway.)
........
  r17325 | jglick | 2009-04-20 18:29:06 -0700 (Mon, 20 Apr 2009) | 1 line
  
  Deprecation of delegate method is intentional.
........
  r17428 | jglick | 2009-04-23 08:07:14 -0700 (Thu, 23 Apr 2009) | 1 line
  
  Noting ToolLocationTranslator more prominently.
........
  r17430 | jglick | 2009-04-23 09:16:40 -0700 (Thu, 23 Apr 2009) | 1 line
  
  Unused imports.
........
  r17473 | jglick | 2009-04-23 19:21:11 -0700 (Thu, 23 Apr 2009) | 1 line
  
  Allow Plugin.configure to see the StaplerRequest, needed for some kinds of calls.
........
  r17474 | jglick | 2009-04-23 19:25:54 -0700 (Thu, 23 Apr 2009) | 3 lines
  
  Initial sketch of plugin to manage tool installations across slaves.
  (Should really be under ../plugins/ but that was not branched, so putting here for the moment.)
........
  r17475 | jglick | 2009-04-23 20:29:36 -0700 (Thu, 23 Apr 2009) | 1 line
  
  Typo in Javadoc.
........
  r17481 | jglick | 2009-04-23 22:46:26 -0700 (Thu, 23 Apr 2009) | 1 line
  
  Unzip of an unbuffered stream is painfully slow.
........
  r17482 | jglick | 2009-04-23 22:48:01 -0700 (Thu, 23 Apr 2009) | 1 line
  
  ZIP-based tool installer now functional.
........
  r17555 | jglick | 2009-04-27 17:50:38 -0700 (Mon, 27 Apr 2009) | 1 line
  
  Build against newest trunk.
........
  r17556 | jglick | 2009-04-27 18:38:28 -0700 (Mon, 27 Apr 2009) | 1 line
  
  Adding help.
........
  r17557 | jglick | 2009-04-27 18:57:46 -0700 (Mon, 27 Apr 2009) | 1 line
  
  Moving toolName & label config GUI into generic files.
........
  r17558 | jglick | 2009-04-27 19:37:30 -0700 (Mon, 27 Apr 2009) | 1 line
  
  Typo.
........
  r17559 | jglick | 2009-04-27 20:07:47 -0700 (Mon, 27 Apr 2009) | 1 line
  
  Second tool installer type based on a freeform command.
........
  r17560 | jglick | 2009-04-27 20:24:36 -0700 (Mon, 27 Apr 2009) | 1 line
  
  installIfNecessaryFrom will now just return false in case the target dir exists but the URL cannot be opened.
........
  r17561 | jglick | 2009-04-27 20:25:22 -0700 (Mon, 27 Apr 2009) | 2 lines
  
  chmod only if something was actually installed.
  Use a+x, not o+x.
........
  r17562 | jglick | 2009-04-27 20:26:46 -0700 (Mon, 27 Apr 2009) | 1 line
  
  Comment only.
........
  r17563 | jglick | 2009-04-27 20:41:12 -0700 (Mon, 27 Apr 2009) | 1 line
  
  I18N.
........
  r17571 | jglick | 2009-04-28 08:15:31 -0700 (Tue, 28 Apr 2009) | 2 lines
  
  Wiki link.
........
  r17593 | jglick | 2009-04-28 15:38:33 -0700 (Tue, 28 Apr 2009) | 1 line
  
  Turn off echo mode, it's just noise.
........
  r17594 | jglick | 2009-04-28 15:39:29 -0700 (Tue, 28 Apr 2009) | 2 lines
  
  Synchronizing access to installers for a given node/tool combo.
  Sending log messages to Hudson log in real time.
........
  r17595 | jglick | 2009-04-28 15:46:29 -0700 (Tue, 28 Apr 2009) | 1 line
  
  Copyrights.
........
  r17619 | jglick | 2009-04-29 06:25:23 -0700 (Wed, 29 Apr 2009) | 1 line
  
  Improved Javadoc.
........
  r17620 | jglick | 2009-04-29 06:51:11 -0700 (Wed, 29 Apr 2009) | 1 line
  
  Demonstraing how to use DLJ to mechanically install official JDK releases on supported operating systems.
........
  r17637 | kohsuke | 2009-04-30 16:15:24 -0700 (Thu, 30 Apr 2009) | 1 line
  
  fixed what appears to be a C&P mistake
........
  r17638 | kohsuke | 2009-04-30 16:16:21 -0700 (Thu, 30 Apr 2009) | 1 line
  
  show YUI logger if YUI debugging mode is set.
........
  r17639 | kohsuke | 2009-04-30 16:52:28 -0700 (Thu, 30 Apr 2009) | 7 lines
  
  Fixed a UI script problem when <hetero-list> nests inside <repeatable>.
  
  The cause of the problem was that the behavior for inner <hetero-list> kicks in first, altering HTML (by removing "prototypes"), before behavior for outer <repetable> kicks in and removes the master copy from DOM tree.
  
  So <repeatable> ends up computing a master copy after some elements are removed, and hence when we try to reinsert them later by clicking "Add", it'll get incorrect copy, and re-execution of the <hetero-list> behavior bombs out.
  
  I fixed this by skipping those HTML-altering behaviors if they show up in the part of DOM tree that's supposed to be snapshotted and removed.
........
  r17640 | kohsuke | 2009-04-30 16:57:58 -0700 (Thu, 30 Apr 2009) | 7 lines
  
  Added ToolProperty to ToolInstallation (which acts like NodeProperty to Node --- basically decorates the base object with additional properties.)
  
  The original motivation is to implement ToolInstaller this way, if Jesse is OK. But decorating tools seem to have some other interesting use, such as a plugin for configuring $M2_HOME/conf/settings.xml, or a plugin to insert unlimited cryptography extension to JDK, or a plugin to add a few more jars into $ANT_HOME/lib.
  
  For this to work, ToolInstallations that want properties needs to be modified to use a new constructor. For a starter, I modified the JDK class to do this.
  
  I also took the opportunity to modernize JDK/config.jelly.
........
  r17641 | kohsuke | 2009-04-30 17:02:04 -0700 (Thu, 30 Apr 2009) | 1 line
  
  this appears to be a left-over from early days when we didn't have a means to do form validation
........
  r17642 | kohsuke | 2009-04-30 17:02:35 -0700 (Thu, 30 Apr 2009) | 1 line
  
  we don't need no.such.JDK but we need name. rebranching
........
  r17643 | kohsuke | 2009-04-30 17:03:22 -0700 (Thu, 30 Apr 2009) | 1 line
  
  rebranched from Hudson's configure.properties
........
  r17644 | kohsuke | 2009-04-30 17:05:31 -0700 (Thu, 30 Apr 2009) | 1 line
  
  removed unnecessary translations
........
  r17656 | kohsuke | 2009-05-01 11:27:36 -0700 (Fri, 01 May 2009) | 1 line
  
  Less hack for improved type safety. Asking people to write more code is not necessarily bad as long as what they need to do is obvious to them.
........
  r17657 | kohsuke | 2009-05-01 12:07:20 -0700 (Fri, 01 May 2009) | 5 lines
  
  The toolautoinst part of the change that corresponds to rev.17640.
  
  ToolInstallers are now hooked under ToolInstallation through InstallSourceProperty as a ToolProperty.
  
  This allows the relevant UIs to be shown in one place. Still pending Jesse's feedback. Making it a single commit to simplify a revert if necessary
........
  r17663 | kohsuke | 2009-05-01 13:23:51 -0700 (Fri, 01 May 2009) | 1 line
  
  doc improvement
........
  r17664 | kohsuke | 2009-05-01 13:24:10 -0700 (Fri, 01 May 2009) | 1 line
  
  needs this to allow Stapler to databind hetero-list automatically
........
  r17665 | jglick | 2009-05-01 13:31:27 -0700 (Fri, 01 May 2009) | 1 line
  
  Warnings.
........
  r17667 | kohsuke | 2009-05-01 14:20:53 -0700 (Fri, 01 May 2009) | 1 line
  
  added a method to do chmod
........
  r17668 | kohsuke | 2009-05-01 14:30:35 -0700 (Fri, 01 May 2009) | 1 line
  
  prefer JDK6 version to avoid unnecessary JNA loading
........
  r17669 | kohsuke | 2009-05-01 14:33:09 -0700 (Fri, 01 May 2009) | 1 line
  
  no, I must have confused this.
........
  r17670 | kohsuke | 2009-05-01 15:57:11 -0700 (Fri, 01 May 2009) | 2 lines
  
  - Automated JDK installer and its test.
  - Minor update to the signature of ToolInstaller for supplying the expected default location.
........
  r17730 | kohsuke | 2009-05-04 11:27:34 -0700 (Mon, 04 May 2009) | 1 line
  
  avoid NPE as reported in HUDSON-3608
........
  r17753 | jglick | 2009-05-04 19:34:08 -0700 (Mon, 04 May 2009) | 2 lines
  
  Moved toolautoinst functionality into core.
........
  r17755 | jglick | 2009-05-04 20:28:34 -0700 (Mon, 04 May 2009) | 8 lines
  
  Added TaskListener parameter and IOException and InterruptedException to several methods:
  Run.getEnvironment
  NodeSpecific.forNode
  ToolLocationTranslator.getToolHome
  ToolLocationNodeProperty.getToolHome
  ToolInstallation.translateFor
  ProcessCache.Factory.getMavenInstallation
  ProcessCache.Factory.getJava
........
  r17756 | jglick | 2009-05-04 20:36:56 -0700 (Mon, 04 May 2009) | 1 line
  
  expectedLocation was never strictly necessary, and will soon no longer be used even by JDKInstaller.
........
  r17757 | jglick | 2009-05-04 20:40:41 -0700 (Mon, 04 May 2009) | 1 line
  
  Not sure yet what @since should be.
........
  r17758 | jglick | 2009-05-04 20:41:51 -0700 (Mon, 04 May 2009) | 1 line
  
  getDefaultInstallers method added. Not yet in use.
........
  r17763 | kohsuke | 2009-05-05 10:50:36 -0700 (Tue, 05 May 2009) | 1 line
  
  deprecated redundant getJavaHome method
........
  r17764 | kohsuke | 2009-05-05 12:10:45 -0700 (Tue, 05 May 2009) | 1 line
  
  added general-purpose property defaulting mechanism, and added default tool installer for JDK
........
  r17765 | jglick | 2009-05-05 12:35:56 -0700 (Tue, 05 May 2009) | 1 line
  
  capitalize("") -> ""
........
  r17766 | jglick | 2009-05-05 12:49:52 -0700 (Tue, 05 May 2009) | 1 line
  
  Missing @since.
........
  r17767 | jglick | 2009-05-05 13:01:41 -0700 (Tue, 05 May 2009) | 1 line
  
  Unchecked warning.
........
  r17768 | jglick | 2009-05-05 13:19:49 -0700 (Tue, 05 May 2009) | 2 lines
  
  Two unsuccessful attempts at form validation.
........
  r17770 | jglick | 2009-05-05 13:48:37 -0700 (Tue, 05 May 2009) | 1 line
  
  Better class names.
........
  r17771 | kohsuke | 2009-05-05 14:12:21 -0700 (Tue, 05 May 2009) | 5 lines
  
  brought the form validation check back. The trick is to use the 'value' parameter.
  
  In the end we'd expect this check to go away anyway, as we change input field to drop-down list, but in the mean time, this is why it didn't work.
  
  Maybe we probably need more static code checkers? Like FindBugs rules?
........
  r17901 | kohsuke | 2009-05-09 19:15:40 -0700 (Sat, 09 May 2009) | 1 line
  
  having two toolHome is odd.
........
  r17906 | kohsuke | 2009-05-10 11:09:13 -0700 (Sun, 10 May 2009) | 1 line
  
  adding the generalization of UpdateCenter, but for compatibility reasons, we are leaving UpdateCenter as-is.
........
  r17911 | kohsuke | 2009-05-10 17:02:53 -0700 (Sun, 10 May 2009) | 1 line
  
  integrated json-lib with a bit of bug fix
........
  r17912 | kohsuke | 2009-05-10 17:13:05 -0700 (Sun, 10 May 2009) | 1 line
  
  JDK list is now fetched from hudson.dev.java.net
........
  r17958 | kohsuke | 2009-05-11 13:44:25 -0700 (Mon, 11 May 2009) | 48 lines
  
  Handle AbortException better to avoid error like this:
  
  [INFO] ------------------------------------------------------------------------
  [ERROR] BUILD ERROR
  [INFO] ------------------------------------------------------------------------
  [INFO] Failed to delete directory: C:\files\hudson\workspace\hudson\hudson\main\remoting\target. Reason: Unable to delete file C:\files\hudson\workspace\hudson\hudson\main\remoting\target\remoting-1.304-SNAPSHOT.jar
  
  [INFO] ------------------------------------------------------------------------
  [INFO] For more information, run Maven with the -e switch
  [INFO] ------------------------------------------------------------------------
  [INFO] Total time: 31 seconds
  [INFO] Finished at: Sun May 10 18:15:33 PDT 2009
  [INFO] Final Memory: 25M/48M
  [INFO] ------------------------------------------------------------------------
  Recording fingerprints
  Archiving artifacts
  Recording test results
  ERROR: Failed to archive JUnit reports
  hudson.util.IOException2: remote file operation failed
  	at hudson.FilePath.act(FilePath.java:596)
  	at hudson.tasks.junit.JUnitResultArchiver.perform(JUnitResultArchiver.java:82)
  	at hudson.model.AbstractBuild$AbstractRunner.performAllBuildStep(AbstractBuild.java:372)
  	at hudson.model.AbstractBuild$AbstractRunner.performAllBuildStep(AbstractBuild.java:360)
  	at hudson.model.Build$RunnerImpl.post2(Build.java:183)
  	at hudson.model.AbstractBuild$AbstractRunner.post(AbstractBuild.java:345)
  	at hudson.model.Run.run(Run.java:943)
  	at hudson.model.Build.run(Build.java:112)
  	at hudson.model.ResourceController.execute(ResourceController.java:93)
  	at hudson.model.Executor.run(Executor.java:119)
  Caused by: hudson.AbortException: Test reports were found but none of them are new. Did tests run? 
  For example, \files\hudson\workspace\hudson\hudson\main\core\target\surefire-reports\TEST-hudson.BulkChangeTest.xml is 2 days 23 hr old
  
  	at hudson.tasks.junit.TestResult.parse(TestResult.java:134)
  	at hudson.tasks.junit.TestResult.<init>(TestResult.java:95)
  	at hudson.tasks.junit.JUnitResultArchiver$1.invoke(JUnitResultArchiver.java:95)
  	at hudson.tasks.junit.JUnitResultArchiver$1.invoke(JUnitResultArchiver.java:82)
  	at hudson.FilePath$FileCallableWrapper.call(FilePath.java:1583)
  	at hudson.remoting.UserRequest.perform(UserRequest.java:92)
  	at hudson.remoting.UserRequest.perform(UserRequest.java:46)
  	at hudson.remoting.Request$2.run(Request.java:236)
  	at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:441)
  	at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:303)
  	at java.util.concurrent.FutureTask.run(FutureTask.java:138)
  	at java.util.concurrent.ThreadPoolExecutor$Worker.runTask(ThreadPoolExecutor.java:886)
  	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:908)
  	at java.lang.Thread.run(Thread.java:619)
........
  r17961 | kohsuke | 2009-05-11 15:16:39 -0700 (Mon, 11 May 2009) | 1 line
  
  adding Windows JDK silent installation
........
  r17962 | kohsuke | 2009-05-11 16:08:13 -0700 (Mon, 11 May 2009) | 1 line
  
  Fixing JDK offline installatino for Windows
........
  r17970 | kohsuke | 2009-05-11 19:37:59 -0700 (Mon, 11 May 2009) | 1 line
  
  Fixing a problem that prevents the usage of this tag in <repeatable> block, which reinserts the same fragment over and over. So statically assigning IDs won't work
........
  r17971 | kohsuke | 2009-05-11 22:17:09 -0700 (Mon, 11 May 2009) | 1 line
  
  continuing to remove static ID generation, which breaks the tag when used inside <repetable> tag.
........
  r17972 | kohsuke | 2009-05-11 22:19:11 -0700 (Mon, 11 May 2009) | 1 line
  
  I don't know when this started, but even in the hudson-dev:run mode, the versionis no longer '?', so we need a different way to force reload
........
  r17973 | kohsuke | 2009-05-11 22:26:34 -0700 (Mon, 11 May 2009) | 1 line
  
  adjusted the text a bit since multiple "add" and "delete"s were rather confusing
........
  r17974 | kohsuke | 2009-05-11 22:31:58 -0700 (Mon, 11 May 2009) | 1 line
  
  added help screen
........
  r17975 | kohsuke | 2009-05-11 22:41:39 -0700 (Mon, 11 May 2009) | 1 line
  
  form validation now works against the checkbox, too
........
  r17976 | kohsuke | 2009-05-11 22:43:00 -0700 (Mon, 11 May 2009) | 1 line
  
  with auto installation, we expect users to specify a directory that doesn't exist yet, so don't report it as an error
........
  r17977 | kohsuke | 2009-05-11 22:43:24 -0700 (Mon, 11 May 2009) | 1 line
  
  improving the text
........


git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@17985 71c3de6d-444a-0410-be80-ed276b4c234a

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@17700 71c3de6d-444a-0410-be80-ed276b4c234a

improved AbortException so that it can record the error message. Necessary for plugins like nunit and cppunit as discussed in HUDSON-3427.
This should be in 1.298.

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@16904 71c3de6d-444a-0410-be80-ed276b4c234a

https://www.dev.java.net/svn/hudson/branches/HUDSON-3251

........
  r16689 | kohsuke | 2009-03-29 08:52:25 -0700 (Sun, 29 Mar 2009) | 1 line
  
  [HUDSON-3251] (1/18) moving maven-related source to maven-plugin
........
  r16690 | kohsuke | 2009-03-29 08:57:23 -0700 (Sun, 29 Mar 2009) | 4 lines
  
  [HUDSON-3251] (3/18) moved KeptBecauseOfParent message
  
  Note I skipped (2/18) to move the logic into MavenProject.DescriptorImpl.isApplicable.
........
  r16691 | kohsuke | 2009-03-29 08:58:21 -0700 (Sun, 29 Mar 2009) | 3 lines
  
  [HUDSON-3251] (4/18) removed imports and javadoc links to Maven-related classes
........
  r16692 | kohsuke | 2009-03-29 08:59:49 -0700 (Sun, 29 Mar 2009) | 2 lines
  
  [HUDSON-3251] (5/18) moving maven-related functionality to maven plugin.
........
  r16693 | kohsuke | 2009-03-29 09:01:11 -0700 (Sun, 29 Mar 2009) | 5 lines
  
  [HUDSON-3251] (5/18) moving maven-related functionality to maven plugin
  
  Forgot to add a newly created file.
........
  r16694 | kohsuke | 2009-03-29 09:02:24 -0700 (Sun, 29 Mar 2009) | 1 line
  
  needs a copyright header on all files in the core
........
  r16695 | kohsuke | 2009-03-29 09:07:03 -0700 (Sun, 29 Mar 2009) | 1 line
  
  [HUDSON-3251] (6/18) updating poms and packaging to bundle maven plugin (with a minor adjustment to revert forkMode change)
........
  r16696 | kohsuke | 2009-03-29 09:17:58 -0700 (Sun, 29 Mar 2009) | 1 line
  
  [HUDSON-3251] (6/18) updating poms and packaging to bundle maven plugin (with a minor adjustment to revert forkMode change)
........
  r16697 | kohsuke | 2009-03-29 09:38:23 -0700 (Sun, 29 Mar 2009) | 2 lines
  
  [HUDSON-3251] (7/18) add a hudson.bundled.plugins property that can override the default bundled plugins.
........
  r16701 | kohsuke | 2009-03-29 15:14:38 -0700 (Sun, 29 Mar 2009) | 1 line
  
  [HUDSON-3251] (8/18) finding Maven jars through reflection for now. TODO: introduce a post-initialization hook
........
  r16702 | kohsuke | 2009-03-29 15:20:36 -0700 (Sun, 29 Mar 2009) | 2 lines
  
  [HUDSON-3251] (10/18) use lastModified to avoid unpacking bundled plugins unnecessarily.
........
  r16703 | kohsuke | 2009-03-29 15:21:15 -0700 (Sun, 29 Mar 2009) | 1 line
  
  [HUDSON-3251] (12/18) updating poms and packaging to bundle maven plugin
........
  r16704 | kohsuke | 2009-03-29 15:37:32 -0700 (Sun, 29 Mar 2009) | 1 line
  
  [HUDSON-3251] (15/18) resolve plugins from the classpath in tests. Slightly reworked.
........
  r16705 | kohsuke | 2009-03-29 15:38:12 -0700 (Sun, 29 Mar 2009) | 1 line
  
  [HUDSON-3251] (16/18) update hpi plugin version
........
  r16706 | kohsuke | 2009-03-29 15:38:54 -0700 (Sun, 29 Mar 2009) | 1 line
  
  [HUDSON-3251] (17/18) moved createExecutedMojoCache
........
  r16707 | kohsuke | 2009-03-29 15:42:38 -0700 (Sun, 29 Mar 2009) | 3 lines
  
  [HUDSON-3251] after looking at (18/18), I think the same effect can be achieved more easily by not setting this system property at all.
  
  Or am I missing something?
........
  r16708 | kohsuke | 2009-03-29 15:43:30 -0700 (Sun, 29 Mar 2009) | 1 line
  
  formatting change
........
  r16709 | kohsuke | 2009-03-29 15:58:16 -0700 (Sun, 29 Mar 2009) | 1 line
  
  [HUDSON-3251] ignore certain builders and publishers in Maven2 job type. This does the same thing as (2/18) in Tom's patch, but I believe this is better. Still needs to figure out how to do this for MavenModule, which doesn't have a descriptor.
........
  r16710 | kohsuke | 2009-03-29 16:01:00 -0700 (Sun, 29 Mar 2009) | 1 line
  
  ignore generated files
........
  r16712 | kohsuke | 2009-03-29 16:29:05 -0700 (Sun, 29 Mar 2009) | 3 lines
  
  [HUDSON-3251] revisited patch (8/18). ComputerListener is extended to support the preOnline step, so that some operations can be carried out whose failure will result in the node failing to become online.
  
  Maven-plugin uses this now to avoid the use of reflection.
........
  r16713 | kohsuke | 2009-03-29 16:38:48 -0700 (Sun, 29 Mar 2009) | 1 line
  
  switching to fork because otherwise the test fails with PermGen out of space
........
  r16714 | kohsuke | 2009-03-29 16:39:36 -0700 (Sun, 29 Mar 2009) | 1 line
  
  no, the next version is 1.295.
........


git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@16734 71c3de6d-444a-0410-be80-ed276b4c234a

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@16105 71c3de6d-444a-0410-be80-ed276b4c234a

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@15927 71c3de6d-444a-0410-be80-ed276b4c234a

adding a probe to better understand http://www.nabble.com/Hudson-v.1.285---maven-job-failes-at-slave-side-td22201035.html

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@15791 71c3de6d-444a-0410-be80-ed276b4c234a

[FIXED HUDSON-2394] Improved error handling when Maven fails to start correctly in the m2 project type.

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@15501 71c3de6d-444a-0410-be80-ed276b4c234a

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@15081 71c3de6d-444a-0410-be80-ed276b4c234a

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@12538 71c3de6d-444a-0410-be80-ed276b4c234a

adding probe for diagnosing http://www.nabble.com/MavenProcessFactory-building-bad-Maven-commandline-td19564224.html

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@12506 71c3de6d-444a-0410-be80-ed276b4c234a

adding probe for diagnosing http://www.nabble.com/MavenProcessFactory-building-bad-Maven-commandline-td19564224.html

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@12505 71c3de6d-444a-0410-be80-ed276b4c234a

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@11279 71c3de6d-444a-0410-be80-ed276b4c234a

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@10974 71c3de6d-444a-0410-be80-ed276b4c234a

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@9759 71c3de6d-444a-0410-be80-ed276b4c234a

I spent a good part of yesterday afternoon to track down this problem. I first attached Visual Studio to the hanging JVM. While I can see the JVM source code and Windows API symbols all right, I couldn't figure out how to relate back the code produced by JIT to its Java, so this made the debugging very hard.

What I observed was that Maven's main thread seems to be blocked in WaitForSingleObject kernel API, but I can't figure out what handle it's blocking on. This call appears to come directly from JIT-ed Java code.

The problem disappears if I enable Java debugger support, or if I attach jconsole. So this cannot be a plain Java-level deadlock.

So after that much work, I still coudldn't figure out the root cause.


But I took the advice from mriem@dev.ajva.net anyway, and modified the remoting to use a separate TCP socket, instead of stdin/stdout. This is useful anyway, because sometimes JVM writes out to stdout and that breaks the whole set up, which was one of the known issues.

This commit brings in this change, and in the limited testing so far, this appears to work. So let's hope that it fixes the problem once and for all.


git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@9615 71c3de6d-444a-0410-be80-ed276b4c234a

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@7351 71c3de6d-444a-0410-be80-ed276b4c234a

applied a patch from http://www.nabble.com/-patch--allow-M2-projects-to-use-environment-variables-in-options-tp15061026p15086318.html


git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@6997 71c3de6d-444a-0410-be80-ed276b4c234a

applied a patch from Nigel Magney (http://www.nabble.com/-patch--allow-M2-projects-to-use-environment-variables-in-options-tt15061026.html)


git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@6865 71c3de6d-444a-0410-be80-ed276b4c234a

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@6717 71c3de6d-444a-0410-be80-ed276b4c234a

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@6652 71c3de6d-444a-0410-be80-ed276b4c234a

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@6614 71c3de6d-444a-0410-be80-ed276b4c234a

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@5874 71c3de6d-444a-0410-be80-ed276b4c234a

See http://www.nabble.com/linux-slave-uses-wrong-maven-path-for-classworlds-jar-tf4648221.html


git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@5312 71c3de6d-444a-0410-be80-ed276b4c234a

git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@4436 71c3de6d-444a-0410-be80-ed276b4c234a