Paul Moore <p.f.mo...@gmail.com> writes: > Hmm, wasn't me :-) The reason field on the build is blank, so this > looks like it's an automatic build.
If it's got an SVN tag indicated, and you can checkout that tag on your slave, then it's probably not the scenario that I was thinking of. In my case, it's almost always me trying to initiate a test build, and forgetting the right branch tag or what its relative to (historically forgetting to put "branches/" in it). > I'm getting odd failures with some of the tests as well, at the moment > - test_smtplib and others failing but when "Re-running failed tests in > verbose mode" there were no failures. > > I wonder if it could be a flaky network connection...? I suppose. Certainly, for this specific error, it's pretty clear that it's not that precise, and various SVN checkout problems can probably be trapped by the error, so it's pretty unlikely it's actually a buildbot version error. What you might try is locating the buildbot log entries (twistd.log) on your slave that correspond to the failed build. It'll show the exact svn command being issued, and you can try it manually in a test tree. If it succeeds then you'll know it was some sort of transient issue - if it fails, it ought to provide useful data as well. > PS Is there a better list for this type of question? I'm not sure how > interesting it is to general python-dev readers. I don't think there's anywhere separate, and generally (recent history being a bit different) buildbot related stuff has been very low volume. So this list is probably appropriate (given what the buildbots are being used for). I expect if it gets too noisy we'll hear about it :-) -- David _______________________________________________ Python-Dev mailing list Python-Dev@python.org http://mail.python.org/mailman/listinfo/python-dev Unsubscribe: http://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com
