| prometheanfire | is it possible to get a git-review release so the rebase-merges fix is in a tag (packaging git-review)? | 03:19 |
|---|---|---|
| *** bhagyashris is now known as bhagyashris|ruck | 04:42 | |
| fungi | prometheanfire: https://lists.opendev.org/pipermail/service-announce/2021-November/000028.html | 04:48 |
| fungi | will that suffice? | 04:49 |
| prometheanfire | fungi: it does, I missed that 2.2.0 wasn't in https://docs.opendev.org/opendev/git-review/latest/releasenotes.html | 04:50 |
| prometheanfire | I thought I remembered it being discussed around that time, so was confused :D | 04:51 |
| fungi | yeah, the docs only get republished after a change merges, and we haven't merged any new changes since that tag was pushed | 04:51 |
| fungi | we could try to address that by making it safe to also run the docs job on tag events | 04:52 |
| * prometheanfire shrugs, 2.2.0 in tree now (packaged) | 04:54 | |
| prometheanfire | I packaged 9999 (git master in the mean time) | 04:54 |
| prometheanfire | pull.ff behavour in git changed too, not possible for me to do pull.ff=only when I used to | 04:54 |
| fungi | `git pull --ff-only` has ceased working? | 04:57 |
| prometheanfire | seems like it, I wonder if it's because of signed commits | 05:00 |
| prometheanfire | will have to test that next | 05:00 |
| prometheanfire | it's not like gentoo's tree actually has merge commits | 05:01 |
| prometheanfire | looks like https://github.com/git/git/commit/7d0daf3f12f87799a3b3f524933ef6c15d1f28c3 is what I hit | 05:25 |
| opendevreview | Sunyeop Lee proposed opendev/git-review master: Fix CommandFailed.__str__ https://review.opendev.org/c/opendev/git-review/+/823130 | 06:34 |
| opendevreview | Sunyeop Lee proposed opendev/git-review master: Fix CommandFailed.__str__ https://review.opendev.org/c/opendev/git-review/+/823130 | 06:34 |
| ykarel | prometheanfire, hi | 08:32 |
| ykarel | can u check https://review.opendev.org/c/openstack/requirements/+/823128 | 08:33 |
| ykarel | i also commented on other patch about the issue https://review.opendev.org/c/openstack/requirements/+/823122 | 08:34 |
| ykarel | may be u can suggest how to clear it with devstack patch failing | 08:34 |
| *** ykarel is now known as ykarel|lunch | 09:50 | |
| *** ykarel|lunch is now known as ykarel | 11:16 | |
| ykarel | if cores around, need reviews on https://review.opendev.org/c/openstack/requirements/+/823128 to clear gates | 12:52 |
| *** ykarel is now known as ykarel|afk | 13:42 | |
| *** ykarel|afk is now known as ykarel | 13:53 | |
| opendevreview | Dr. Jens Harbott proposed openstack/diskimage-builder master: Remove extra if/then/else construct in pip element https://review.opendev.org/c/openstack/diskimage-builder/+/822224 | 15:44 |
| prometheanfire | ykarel: I see the first merged, so that's good, do you know of an upstream bug to track for paramiko breakages? | 16:43 |
| ykarel | prometheanfire, frickler created https://github.com/paramiko/paramiko/issues/1961 | 16:43 |
| prometheanfire | cool, thanks | 16:43 |
| *** ykarel is now known as ykarel|away | 17:08 | |
| frickler | more release fun, tox 3.24.5 just out and seemingly failing jobs https://pypi.org/project/tox/#history https://storage.bhs.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_5de/773574/16/check/designate-bind9/5de8e2f/job-output.txt | 19:58 |
| fungi | oh neat | 20:02 |
| frickler | likely also a bug from neutron/devstack to just install tox without enforcing a sensible lower version | 20:04 |
| fungi | tox.ConfigError: ConfigError: substitution 'env:UPPER_CONSTRAINTS_FILE:https://releases.openstack.org/constraints/upper/master': 'UPPER_CONSTRAINTS_FILE:https://releases.openstack.org/constraints/upper/master' not found in environment | 20:04 |
| frickler | fungi: yes, seems that is what you get from tox 1.11.0 instead anything recentish | 20:07 |
| fungi | ahh, so something is causing newer versions to fail to install and pip winds up settling on an older one instead? | 20:07 |
| fungi | the changelog for the new release is just "Fixed an issue where usedevelop would cause an invocation error if setup.py does not exist." | 20:09 |
| frickler | fungi: yes, see the first matches for "tox" in that log. starting with + lib/neutron_plugins/ovn_agent:install_ovn:361 : pip_install tox | 20:09 |
| fungi | could it be the new setuptools (60.1.1)? | 20:11 |
| fungi | #2980: Bypass distutils loader when setuptools module is no longer available on sys.path. | 20:12 |
| fungi | aha! virtualenv 20.11.2 Released: about 9 hours ago | 20:13 |
| fungi | 20.11.1 was also released today | 20:15 |
| fungi | 20.11.0 yesterday | 20:15 |
| fungi | which upped the vendored copy of setuptools from 58.x to 60.x | 20:15 |
| fungi | i have a feeling that's it | 20:15 |
| fungi | v60.0.0 Breaking Changes: Setuptools once again makes its local copy of distutils the default. To override, set SETUPTOOLS_USE_DISTUTILS=stdlib. | 20:16 |
| fungi | v59.0.0 Breaking Changes: Started failing on invalid inline description with line breaks ValueError | 20:17 |
| eandersson | Not able to easily reproduce it locally | 20:26 |
| frickler | there are also jobs passing that installed tox 3.24.5 successfully, so I still think our mirrors are involved | 20:28 |
| frickler | hopefully this will resolve itself once the 1h cache timeout is reached | 20:29 |
| fungi | oh, suspecting we got served a fallback view of pypi in some places which was missing some necessary recent dependency for it? | 20:38 |
| frickler | fungi: yes, in the log pip downloads the correct tox wheel, starts working through the dependencies, has a few comments about needing some time, and finally goes down through all tox versions until it finds one that has next to no dependencies | 20:50 |
| frickler | in the end it is just "Successfully installed py-1.11.0 tox-1.1" | 20:51 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!