Tuesday, 2022-11-22

*** tkajinam is now known as Guest216202:04
*** tkajinam is now known as Guest217403:28
*** yadnesh|away is now known as yadnesh04:03
*** tkajinam is now known as Guest217905:36
*** yadnesh is now known as yadnesh|afk08:20
*** yadnesh|afk is now known as yadnesh08:51
tobias-urdinhow long would it take for the proposal bot to update upper-constraint? https://pypi.org/project/cursive/#history was released yesterday but can't see any 0.2.3 patch yet to UC12:16
opendevreviewTobias Urdin proposed openstack/requirements master: Bump upper-constraint for cursive to 0.2.3  https://review.opendev.org/c/openstack/requirements/+/86527613:09
fricklertobias-urdin: the fast-track changes like https://review.opendev.org/c/openstack/requirements/+/865030 are only created for releases made by the openstack release automation13:27
fricklerother updates would normally be generated every Saturday morning (via the periodic-weekly pipeline) but the job responsible for that is currently broken13:28
fricklerso in the current state the answer to your question is: forever13:29
fricklerbut https://review.opendev.org/c/openstack/requirements/+/865276 seems o.k. and I think we can fast-track it once it passes CI13:30
tobias-urdinfrickler: ack, then my change was warranted that good :)15:06
fricklertobias-urdin: you should also be able to depend-on that in the barbican patch instead of using the devstack+cursive depends15:08
frickleractually let me just do that15:09
tobias-urdincool didn't know that worked15:10
opendevreviewOpenStack Proposal Bot proposed openstack/requirements master: update constraint for sushy to new release 4.4.0  https://review.opendev.org/c/openstack/requirements/+/86529815:33
opendevreviewOpenStack Proposal Bot proposed openstack/requirements master: update constraint for metalsmith to new release 1.9.0  https://review.opendev.org/c/openstack/requirements/+/86530415:59
*** yadnesh is now known as yadnesh|away16:12
opendevreviewMerged openstack/requirements master: Bump upper-constraint for cursive to 0.2.3  https://review.opendev.org/c/openstack/requirements/+/86527619:37
clarkbHello, I've been tracking down a problem with tox on jammy affecting ceph devstack jobs. Where I've ended up is that virtualenv 20.16.5 and older are not reliably able to set paths correctly in virtualenvs for tools like tox22:26
clarkbhttps://github.com/pypa/virtualenv/pull/2415 appers to be the fix and has links to people describing the problems in much more detail.22:26
clarkbWe constrain virtualenv to 20.16.5 though and I don't see any changes to bump that constraint up. Any idea why that hasn't been bumped yet its been out for about 10 days22:27
clarkbthat said this doens't happen for all jammy jobs and isn't reproduceable in a jammy container which makes it all very confusing22:27
prometheanfireit's probably in the blacklist22:40
prometheanfirehmm, we don't block updates to that22:41
clarkbya I wasn't finding anything obvuous for why it wasn't updating22:41
clarkbmaybe the jobs have been failing?22:41
prometheanfirejobs run once a week now22:41
clarkbthe release is more than a week old. but I guess one failure could do it22:41
prometheanfireya22:42
clarkbhttps://pypi.org/project/virtualenv/#history22:42
clarkbfwiw I just managed to reproduce this in a container22:42
prometheanfireiirc, the last update had a lot of handholding22:42
prometheanfireyou can propose a non-bot update too22:42
clarkbthings work fine if you use distro setuptools but explode when using pip installed setuptools22:42
clarkbthe reason is pip installed setuptools includes distutils which overrides th edistro stuff22:43
clarkbnow that I've managed to figure that out I think a manual bump is fine22:43
prometheanfireya, https://github.com/openstack/requirements/commit/3f65964c2609f150837b35ecfa0253def2f2be03 was 15 days ago merged, after a bunch of handholding with no more bot updates22:43
clarkbI didn't want to workaround it without a better understanding of why things differed22:43
prometheanfireack22:43
prometheanfirebeing able to explain why is important22:43
clarkband actually that may lead to a more targetted workaround in devstack22:46
clarkband then I can defer to normal constraints updates to pull in newer virtualenv22:46
clarkbI'm going to try that first since I think that better captures what we want to do here22:46

Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!