Victor Stinner <victor.stin...@haypocalc.com> writes:
> But today I saw a a buildbot timeout without any traceback: a possible hang
> in
> test_io on "x86 FreeBSD 7.2 3.x" buildbot, "command timed out: 3900 seconds
> without output". I realized that some buildbots (all buildbots?) override the
> TESTOPTS variable ("make buildbottest TESTOPTS= TESTPYTHONOPTS=" for "x86
> FreeBSD 7.2 3.x").
(...)
> Please, don't disable the TESTTIMEOUT option for your buildbot. You may want
> to use a shorter or longer timeout, (...)
Just a note, given the phrasing above. None of this is anything that
I for example, as a buildbot operator, am actively controlling. That
command, including the environment variable overrides, is exactly as
provided by the master for a given test run. So I'd think you could
adjust it if needed through changes in the master build configuration
and probably without having to add an environment variable or
Per Paul's follow-up on Windows, buildbot under Windows seems to
impose a 1200s idle timeout (just for no output), but I'm not positive
how it's calculated. The test process itself has never, I'm pretty
sure, specified a timeout to regrtest (via the test.bat, rt.bat,
regrtest.py path). (Oh, I guess the --timeout option itself in
regrtest is fairly new)
So if there's a change in defaults for regrtest that will change
Windows behavior implicitly, and I believe at this point the buildbots
will be inconsistent, since you're only overriding the regrtest
default in a subset of buildbot types.
-- 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
