Mailman 3 June 2019 - Python-Buildbots

Windows ARM32 buildbot
by Paul Monson 10 Jul '19

10 Jul '19

Can someone merge this PR please? https://github.com/python/buildmaster-config/pull/97 I suspect that the socket tests are flaky no Windows ARM32 because of a lack of resource, and I would like to test running one test at time to see if it improves test reliability. The tests seems less flaky when not running in a buildbot worker, but it might just be that the sample size is too small when running manually Thanks, Paul

2 2

Some more Fedora and RHEL machines willing to join the fleet
by Charalampos Stratakis 07 Jul '19

07 Jul '19

Hello, I've recently got some more x86_64 machines that I'd like to add to the buildbot fleet. More specifically, a Fedora 30 (the stable branch of Fedora), a RHEL7 and a RHEL8 machine. I'm gonna add some naming suggestions, just to get things rolling :) cstratak-fedora-stable, cstratak-RHEL7-x86_64, cstratak-RHEL8-x86_64 -- Regards, Charalampos Stratakis Software Engineer Python Maintenance Team, Red Hat

2 2

New cleantest step on buildbots: remove old test_python_xxx directories
by Victor Stinner 25 Jun '19

25 Jun '19

Hi, Since the end of last week, test_regrtest started to fail on more and more buildbots, especially on x86 Gentoo Installed with X 3.x, with an error message which contains: RuntimeWarning: tests may fail, unable to create temporary directory '/buildbot/tmp/tmpdir/test_python_28144': [Errno 17] File exists: '/buildbot/tmp/tmpdir/test_python_28144'" I knew for years that our buildbots are not great to cleanup things between two builds: it's common that a failure is causing issues in following builds and may require a manual action. I added a new "cleantest" step which runs a new "./python -m test --cleanup" command: it removes test_python_* directories in the directory used by Python for tests (ex: ./build/ when building from sources on Unix). On x86 Gentoo Installed with X 3.x, my new step removed not least than 4146 old "test_python_xxx" directories! That's pretty big knowing that Linux pids are limited to 32768: 4146 is already 13% of the available ids, which makes the issue quite likely. By the way, I also enhanced regrtest when using worker subprocesses. Workers now create their sub-directory in the temporary directory of the main process. Moreover, the main process now kills immediately workers (and wait until they complete) on failure or if interrupted. The whole test directory is always removed by the main process. The last case where regrtest is when the main process does crash, which happens often when tests are re-run. Sadly, when tests are re-run, they are run sequentially in the main process, rather than spawning a subprocess for each test file. That should be fixed. Victor -- Night gathers, and now my watch begins. It shall not end until my death.

1 0

FYI: 'POWER6' bot offline for a few weeks
by Michael 23 Jun '19

23 Jun '19

Hi all, Now that the PPC64-AIX bot is also "green" for master (and I hope real soon for 3.8 and 3.7) I feel 'comfortable' with having me server in PowerOff status while I am off on vacation. Will be back soon! Michael

1 0

Finally ! Many thanks!
by Michael 19 Jun '19

19 Jun '19

Finally, got to this point that the tests are all passing on the bot - 3.X!! p.s. - now that 3.8 is beta - Will there be a separate bot for 3.8 in the near future?

3 2

Current build stuck for four days(!)
by Chris Angelico 17 Jun '19

17 Jun '19

Turns out that this isn't a massive queue of builds happening, but just one single build. https://buildbot.python.org/all/#/builders/211/builds/46 The builder is saturating one CPU core and has been doing so for literally days. I don't think it's going to get anywhere at this point. Is there a way to bomb the build and tell it to start over? ChrisA

3 4