[Buildbot-devel] Strange problem with git-fetch

David Whittington djwhitt at gmail.com
Wed Feb 27 15:22:33 UTC 2008


I experience a rather strange problem with a buildbot slave the other
day. When doing an update the slave reported the following exception:

2008/02/26 00:05 CST [Broker,client]
SlaveBuilder.remote_print(master-mac-ppc): message from master:
ping2008/02/26 00:05 CST [Broker,client]
SlaveBuilder.remote_ping(<SlaveBuilder 'master-mac-ppc' at 16937768>)
2008/02/26 00:05 CST [Broker,client] <SlaveBuilder 'master-mac-ppc' at
16937768>.startBuild
2008/02/26 00:05 CST [Broker,client]  startCommand:git [id 14]
2008/02/26 00:05 CST [Broker,client] ShellCommand._startCommand
2008/02/26 00:05 CST [Broker,client]  git-fetch git://git.rubini.us/code master
2008/02/26 00:05 CST [Broker,client]   in dir
/Volumes/susie/cremes/developer/projects/ruby/rubinius_bot/rubinius-master-mac-ppc/build
(timeout 1200 secs)
2008/02/26 00:05 CST [Broker,client]   watching logfiles {}
2008/02/26 00:05 CST [Broker,client]   argv: ['git-fetch',
'git://git.rubini.us/code', 'master']
2008/02/26 00:05 CST [Broker,client]  environment: {'TERM':
'xterm-color', 'SHELL': '/bin/bash', 'SHLVL': '1', 'SSH_TTY':
'/dev/ttys014', 'OLDPWD':
'/Users/cremes/developer/projects/ruby/rubinius_bot/info',
'SSH_CLIENT': '64.202.118.8 64857 22', 'LIBTOOLIZE': 'glibtoolize',
'PWD': '/Users/cremes/developer/projects/ruby', 'CVS_RSH':
'/usr/bin/ssh', 'LOGNAME': 'cremes', 'USER': 'cremes', 'MANPATH':
'/usr/share/man:/usr/local/share/man:/usr/X11/man', 'MAIL':
'/var/mail/cremes', 'PATH':
'/opt/local/bin:/opt/local/apache2/bin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:/usr/X11/bin:/Developer/Tools:/Users/cremes/bin:/opt/local/sbin:/usr/local/mysql/bin:/opt/local/transcode/bin',
'SSH_CONNECTION': '64.202.118.8 64857 75.3.115.48 22', 'HOME':
'/Users/cremes', 'CVSROOT':
':ext:cremes at cvs.opendarwin.org:/Volumes/src/cvs/od', '_':
'/usr/local/bin/buildbot', 'EDITOR': 'vi'}
2008/02/26 00:05 CST [-] command finished with signal None, exit code 1
2008/02/26 00:05 CST [-] _checkAbandoned [Failure instance: Traceback:
<class 'buildbot.slave.commands.AbandonChain'>: 1
  /Library/Python/2.5/site-packages/buildbot/slave/commands.py:158:processEnded
  /Library/Python/2.5/site-packages/buildbot/slave/commands.py:466:finished
  /System/Library/Frameworks/Python.framework/Versions/2.5/Extras/lib/python/twisted/internet/defer.py:229:callback
  /System/Library/Frameworks/Python.framework/Versions/2.5/Extras/lib/python/twisted/internet/defer.py:294:_startRunCallbacks
  --- <exception caught here> ---
  /System/Library/Frameworks/Python.framework/Versions/2.5/Extras/lib/python/twisted/internet/defer.py:307:_runCallbacks
  /Library/Python/2.5/site-packages/buildbot/slave/commands.py:691:_abandonOnFailure
  ]
2008/02/26 00:05 CST [-]  abandoning chain 1
2008/02/26 00:05 CST [-] SlaveBuilder.commandComplete
<buildbot.slave.commands.Git instance at 0x102ab70>
2008/02/26 00:12 CST [-] sending app-level keepalive

The git-fetch command stdio indicated a DNS name resolution problem.
However, doing a git-fetch manually worked. A restart of the slave
appears to have fixed the problem. I'm just wondering if anyone has
experienced similar problems and if so they determined the cause.

Thanks,

David Whittington




More information about the devel mailing list