On Sun, Jun 5, 2011 at 8:57 PM, Hanno Schlichting <hanno@hannosch.eu> wrote:
On Sun, Jun 5, 2011 at 8:24 PM, Hanno Schlichting <hanno@hannosch.eu> wrote:
$ bin/alltests -s Acquisition -t test_mixed_explicit_and_explicit Running zope.testrunner.layer.UnitTests tests: Set up zope.testrunner.layer.UnitTests in 0.000 seconds. Segmentation fault
Here's some more info:
The last successful build was #256 on Wed Apr 27 02:06:59 2011 which build revision 121465 The first failing build was #257 on Wed Apr 27 15:19:46 2011 which build revision 121473
Looking at the SVN changelog of Zope trunk, there were changes in r121460 and r121502.
So the two builds used the exact same code, but still produced a different result. So it must be a change in the config of the buildslave that triggered this. There's no Python 2.7 release anywhere close to April this year, so I'm pretty sure both builds used the same Python 2.7.1.
What caused the failure is a mystery to me.
I see that Jan-Jaap did some unrelated changes to the buildslave config on that day (http://zope3.pov.lt/trac/changeset/121467/Sandbox/janjaapdriessen/buildbot). Jan-Jaap: Do you remember doing anything else on that day? Like upgrading the operating system, installing new OS packages or anything that could cause the sudden failure? Hanno