Summary: | dev-python/pbr-5.1.1 fails test pbr.tests.test_packaging.TestRequirementParsing.test_requirement_parsing | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Jonathan Lovelace <kingjon3377> |
Component: | Current packages | Assignee: | Matthew Thode ( prometheanfire ) <prometheanfire> |
Status: | RESOLVED OBSOLETE | ||
Severity: | normal | CC: | openstack, python, sam |
Priority: | Normal | Keywords: | TESTFAILURE |
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Attachments: | dev-python:pbr-5.1.1:20200610-234843.log |
Description
Jonathan Lovelace
2020-06-11 00:56:29 UTC
Tests pass fine with python3_7 here 5.4.5 passed its tests (on python3_7) fine back in July, but 5.4.5-r1 is now apparently-reliably failing this same test in the same way (also on python3_7). (In reply to Jonathan Lovelace from comment #2) > 5.4.5 passed its tests (on python3_7) fine back in July, but 5.4.5-r1 is now > apparently-reliably failing this same test in the same way (also on > python3_7). What version of dev-python/virtualenv do you have installed? dev-python/virtualenv is 16.7.10-r1, the current stable. (In reply to Jonathan Lovelace from comment #4) > dev-python/virtualenv is 16.7.10-r1, the current stable. Can you attempt to do: git clone https://github.com/openstack/pbr cd pbr git checkout 5.1.1 pytest -vv pbr/tests/test_packaging.py::TestRequirementParsing::test_requirement_parsing Does it fail too? That passes, and it also passes when I run the same command under 'git checkout 5.4.5'. With that, and noticing that the test failure message talked about timeouts, I first tried running with MAKEOPTS=-j1 (which didn't help), then with FEATURES="-sandbox -usersandbox", and with that change all the tests passed. So maybe some timeout value needs to be increased to compensate for the overhead of the sandbox? (In reply to Jonathan Lovelace from comment #6) > That passes, and it also passes when I run the same command under 'git > checkout 5.4.5'. > > With that, and noticing that the test failure message talked about timeouts, > I first tried running with MAKEOPTS=-j1 (which didn't help), then with > FEATURES="-sandbox -usersandbox", and with that change all the tests passed. > > So maybe some timeout value needs to be increased to compensate for the > overhead of the sandbox? If your failure is still the same as in the initial message, it's not a timeout. I don't really see why the sandbox would have anything to do with it. Does it also fail with sys-apps/sandbox 2.20 if you can pull it on your stable system? |