rdogerrit | rdo-trunk rdo-trunk proposed rdoinfo master: Promote CBS tags update for caracal-9s-testing https://review.rdoproject.org/r/c/rdoinfo/+/51562 | 00:24 |
---|---|---|
rdogerrit | rdo-trunk rdo-trunk proposed rdoinfo master: Promote CBS tags update for bobcat-9s-testing https://review.rdoproject.org/r/c/rdoinfo/+/51537 | 00:25 |
apevec | tkajinam: yeah, we too hit regressions with latest published CS9 libvirt, maintainers said "The fixed version is libvirt-10.0.0-1.el9" | 00:29 |
apevec | it is built in CS Koji but still in the gate https://kojihub.stream.centos.org/koji/buildinfo?buildID=47091 | 00:31 |
apevec | you can grab RPMs from Koji | 00:31 |
rdogerrit | rdo-trunk rdo-trunk proposed rdoinfo master: Promote CBS tags update for antelope-9s-testing https://review.rdoproject.org/r/c/rdoinfo/+/51427 | 00:38 |
rdogerrit | rdo-trunk rdo-trunk proposed rdoinfo master: Promote CBS tags update for wallaby-8s-testing https://review.rdoproject.org/r/c/rdoinfo/+/51282 | 00:39 |
rdogerrit | rdo-trunk rdo-trunk proposed deps/python-httpx c9s-caracal-rdo: Update to 0.26.0 https://review.rdoproject.org/r/c/deps/python-httpx/+/51563 | 01:04 |
apevec | spotz: FYI I tried to export blogs.r.o using WP plugins for exporting to the static site (OT: as a prep, I had to upgrade WP - it was very out of date, not sure if OSPO is supposed to monitor that, main admin was jbrooks?) | 08:21 |
apevec | both Staatic and Simply Static failed with 500 while crawling, looks like they where too aggressive... I'll try external crawler which can more gentle | 08:23 |
apevec | In any case, I'd propose to not add any new content to RDO WP instance, while I figure out the best way to export content and move it to a new site | 08:24 |
jcapitao[m] | amoralejkarolinku : can you review https://review.rdoproject.org/r/c/rdoinfo/+/51549 please ? | 08:45 |
jcapitao[m] | it's blocking a FTBFS dated from Feb 3 | 08:45 |
jcapitao[m] | this one https://review.rdoproject.org/r/c/openstack/magnum-tempest-plugin-distgit/+/51540 | 08:45 |
rdogerrit | Merged rdoinfo master: Add python-oauthlib as dependency https://review.rdoproject.org/r/c/rdoinfo/+/51549 | 09:16 |
jcapitao[m] | amoralej karolinku : so new libvirt is breaking all of our CS9 releases | 09:31 |
jcapitao[m] | dunno if you noticed it | 09:32 |
jcapitao[m] | messages from apevec earlier : yeah, we too hit regressions with latest published CS9 libvirt, maintainers said "The fixed version is libvirt-10.0.0-1.el9" | 09:33 |
jcapitao[m] | and - it is built in CS Koji but still in the gate https://kojihub.stream.centos.org/koji/buildinfo?buildID=47091 | 09:33 |
amoralej | damn | 09:33 |
amoralej | where are logs? | 09:34 |
jcapitao[m] | https://logserver.rdoproject.org/37/51537/2/check/validate-buildsys-tags-bobcat-testing-puppet-scenario001-centos9/4ec77d1/logs/weirdo-project/logs/nova/nova-compute.txt.gz | 09:34 |
jcapitao[m] | for instance | 09:34 |
jcapitao[m] | from https://review.rdoproject.org/r/c/rdoinfo/+/51537?tab=change-view-tab-header-zuul-results-summary | 09:34 |
amoralej | "2024-02-06 20:32:04.596 99517 ERROR nova.compute.manager UnicodeDecodeError: 'utf-8' codec can't decode byte 0xa1 in position 0: invalid start byte" ? | 09:35 |
amoralej | that's the issue? | 09:35 |
jcapitao[m] | yes | 09:35 |
jcapitao[m] | we have same issue for puppet-promotion pipelines | 09:35 |
apevec | another CS horror story in the making 🙄 | 09:38 |
amoralej | integration sig to the rescue :) | 09:38 |
apevec | dunno why it is taking so long in the gate, I guess there are other blockers, I'll ping in RHEL Jira ticket | 09:39 |
amoralej | we may add an exclude to the repo config in worst case | 09:39 |
amoralej | but it'd be better to avoid it if possible | 09:39 |
apevec | amoralej @_oftc_amoralej:matrix.org: IntSIG would not override RHEL gate | 09:40 |
apevec | that's where the blocker is, afaict | 09:40 |
apevec | ah yes, let's try excluding 9.10 | 09:41 |
amoralej | yes, but if there is a way to label and expose only the composes that have passed openstack jobs ... | 09:41 |
amoralej | we may consume that | 09:41 |
jcapitao[m] | but only for promotion pipelines | 09:42 |
amoralej | wrt exclusion, we may do something like https://review.rdoproject.org/r/c/rdo-infra/weirdo/+/33909 | 09:45 |
amoralej | i'm preparing the patch | 09:48 |
amoralej | apevec, jcapitao[m] is there any bz or report about the libvirt issue ? | 09:54 |
apevec | yes, we have CIX with a link to RHEL Jira, sec | 09:54 |
apevec | https://issues.redhat.com/browse/RHEL-23925 was filed by OSP then closed as dup of https://issues.redhat.com/browse/RHEL-22314 by the RHEL team | 09:55 |
apevec | lemme check if those are public | 09:55 |
apevec | ok the latter is public | 09:56 |
jcapitao[m] | thx | 09:56 |
apevec | but title looks unrelated, dunno why RHEL closed previous as a dup, just b/c "it is fixed in same 10.0" ... | 09:56 |
apevec | yeah description is for a completely different issue :( | 09:57 |
amoralej | yep | 09:58 |
amoralej | not nice | 09:58 |
jcapitao[m] | maybe a misclick | 09:58 |
apevec | ah yes, actual dup is https://issues.redhat.com/browse/RHEL-20609 | 09:59 |
apevec | so it was caught by RHEL QE early Jan ... | 10:00 |
amoralej | good | 10:01 |
apevec | jcapitao[m]: I actually mis-copied, Jira filed by OSP is marked as dup of 20609 :) | 10:01 |
rdogerrit | Alfredo Moralejo proposed rdo-infra/weirdo master: Exclude libvirt-9.10.0-1 from appstream repo https://review.rdoproject.org/r/c/rdo-infra/weirdo/+/51564 | 10:03 |
amoralej | apevec, jcapitao[m] ^ | 10:04 |
amoralej | tkajinam proposed similar approach https://review.opendev.org/c/openstack/puppet-openstack-integration/+/908217 | 10:04 |
jcapitao[m] | +2 | 10:13 |
jcapitao[m] | actually, we should also report it as a bug in our RDO JIRA | 10:13 |
amoralej | yep | 10:14 |
jcapitao[m] | I'm creating it | 10:15 |
rdogerrit | Marios Andreou proposed rdo-jobs master: Remove redundant var skip_standalone_vm in DPA CI playbooks and job https://review.rdoproject.org/r/c/rdo-jobs/+/51547 | 10:20 |
tkajinam | amoralej, jcapitao I wonder if you know any timeline about the 10.0.0 release in centos ? | 10:26 |
amoralej | i don't | 10:26 |
tkajinam | it also affects a few other projects having c9s jobs. if it takes some time then I'll propose marking these jobs non-voting | 10:26 |
tkajinam | that's a serious regression they should fix asap imo, though | 10:26 |
amoralej | yes, i agree | 10:27 |
tkajinam | it happened really annoying time. it blocks all the work needed to wipe out puppet-qdr from yoga before transitioning yoga to unmaintained :-( | 10:29 |
amoralej | i've asked in https://issues.redhat.com/browse/RHEL-20609 | 10:29 |
amoralej | let's see if we can get any input | 10:30 |
tkajinam | just my own complaint | 10:30 |
tkajinam | amoralej, thanks ! | 10:30 |
amoralej | last build for cs9 is from 29-Jan but they must be having some issue because it's not in the compose under testing | 10:32 |
amoralej | https://composes.stream.centos.org/production/latest-CentOS-Stream/compose/AppStream/x86_64/os/Packages/ | 10:33 |
amoralej | apevec, ^ fyi current compose in production is shipping 9.10.0 | 10:33 |
apevec | yes I know... | 10:38 |
jcapitao[m] | https://issues.redhat.com/browse/RDO-183 | 10:39 |
apevec | thanks, I piled it on under "blocks" in RHEL-20609 | 10:41 |
apevec | and that one is blocked by libvirt rebase https://issues.redhat.com/browse/RHEL-15267 - which probably is blocked on some other issues, sigh... | 10:42 |
amoralej | yep, it's unlucky that the fix comes with a rebase | 10:43 |
apevec | what if we grab libvirt-10.0.0-2.el9 from CS Koji and put it in a side-repo? | 10:43 |
apevec | temp in RDO deps maybe? | 10:43 |
amoralej | we could, at least to test it | 10:43 |
apevec | yes, that way we could at least check there isn't another regression lurking in a rebase | 10:44 |
jcapitao[m] | agree with that proposal | 10:47 |
rdogerrit | Merged rdo-infra/weirdo master: Exclude libvirt-9.10.0-1 from appstream repo https://review.rdoproject.org/r/c/rdo-infra/weirdo/+/51564 | 10:59 |
jcapitao[m] | amoralej apevec https://review.opendev.org/c/openstack/puppet-openstack-integration/+/908291 let's see how it goes | 11:29 |
amoralej | nice | 11:30 |
apevec | ah we already had delorean-deps-test.repo or you added it now? | 11:31 |
jcapitao[m] | we used it to test Ceph Reef with EPEL enabled | 11:31 |
apevec | ah cool, I see https://trunk.rdoproject.org/centos9-master/deps/cs9-fix/ is the temp repo, let's keep that around ;) | 11:32 |
apevec | I've put click-bait to Integration SIG matrix | 11:42 |
amoralej | :) | 11:45 |
karolinku[m] | that's remind me of joining:) | 11:45 |
apevec | double-click bait! I should move to marketing ;) | 11:50 |
apevec | BTW bookmark this https://docs.fedoraproject.org/en-US/fedora-accounts/user/#pkinit - after adding 2FA to the FAS account (highly recommended, even though might be annoying) | 11:59 |
amoralej | jcapitao[m], karolinku[m] see my comment in https://review.rdoproject.org/r/c/rdoinfo/+/51549 please | 12:03 |
karolinku[m] | ack | 12:04 |
jcapitao[m] | right | 12:08 |
jcapitao[m] | I'm going to lower the requirement on kubernetes itself | 12:09 |
rdogerrit | Joel Capitao proposed rdoinfo master: Revert "Add python-oauthlib as dependency" https://review.rdoproject.org/r/c/rdoinfo/+/51574 | 12:10 |
jcapitao[m] | but I remember having checked that | 12:12 |
jcapitao[m] | weird | 12:12 |
rdogerrit | Marios Andreou proposed rdo-jobs master: Remove redundant var skip_standalone_vm in DPA CI playbooks and job https://review.rdoproject.org/r/c/rdo-jobs/+/51547 | 12:18 |
jcapitao[m] | ok I know what happened, in the spec file we have https://review.rdoproject.org/cgit/deps/python-kubernetes/tree/SPECS/python-kubernetes.spec?id=fd6569e6b4cd6127cbb11a6a78f7ca373082fb0d#n47 | 12:55 |
jcapitao[m] | Requires: %{py3}-oauthlib | 12:55 |
jcapitao[m] | our CI failed because nothing was providing python3.9dist(oauthlib) | 12:56 |
jcapitao[m] | so I assumed it was not provided at all | 12:57 |
jcapitao[m] | but actually the package is using https://review.rdoproject.org/cgit/deps/python-kubernetes/tree/SPECS/python-kubernetes.spec?id=fd6569e6b4cd6127cbb11a6a78f7ca373082fb0d#n1 | 12:57 |
jcapitao[m] | which adds a strong requirement based on requirements.txt | 12:58 |
rdogerrit | Joel Capitao proposed deps/python-kubernetes c9s-caracal-rdo: Uncap oauthlib version https://review.rdoproject.org/r/c/deps/python-kubernetes/+/51565 | 13:00 |
jcapitao[m] | amoralej karolinku should be better with that ^ | 13:00 |
jcapitao[m] | and reverting the rdoinfo patch https://review.rdoproject.org/r/c/rdoinfo/+/51574 | 13:02 |
amoralej | jcapitao[m], lgtm let's see if that passes ci | 13:04 |
jcapitao[m] | amoralej apevec CI jobs are successful with libvirt-10.0.0-2.el9 https://review.opendev.org/c/openstack/puppet-openstack-integration/+/908291?tab=change-view-tab-header-zuul-results-summary | 13:04 |
apevec | good news! | 13:05 |
jcapitao[m] | I updated the JIRA ticket https://issues.redhat.com/browse/RDO-183 with that info | 13:05 |
amoralej | great | 13:06 |
rdogerrit | Arx Cruz proposed rdo-jobs master: Removing poi scenario000 https://review.rdoproject.org/r/c/rdo-jobs/+/51566 | 13:07 |
rdogerrit | Arx Cruz proposed rdo-jobs master: Removing poi scenario000 https://review.rdoproject.org/r/c/rdo-jobs/+/51566 | 13:08 |
rdogerrit | Arx Cruz proposed rdo-jobs master: Delete tripleo-next-os project template https://review.rdoproject.org/r/c/rdo-jobs/+/51566 | 13:14 |
apevec | jcapitao: cs9-fix is in master only, can we add it to antelope too, that's what o-k-o is using | 13:14 |
jcapitao[m] | lemme do the same for antelope | 13:15 |
rdogerrit | Merged rdo-jobs master: Delete tripleo-next-os project template https://review.rdoproject.org/r/c/rdo-jobs/+/51566 | 13:26 |
jcapitao[m] | https://review.opendev.org/c/openstack/puppet-openstack-integration/+/908309 | 13:32 |
jcapitao[m] | ^ for antelope | 13:32 |
rdogerrit | Joel Capitao proposed rdoinfo master: Promote CBS tags update for caracal-9s-testing https://review.rdoproject.org/r/c/rdoinfo/+/51562 | 13:37 |
rdogerrit | Merged deps/python-kubernetes c9s-caracal-rdo: Uncap oauthlib version https://review.rdoproject.org/r/c/deps/python-kubernetes/+/51565 | 13:38 |
rdogerrit | Ronelle Landy proposed rdo-jobs master: Remove redundant var skip_standalone_vm in DPA CI playbooks and job https://review.rdoproject.org/r/c/rdo-jobs/+/51547 | 13:41 |
rdogerrit | Marios Andreou proposed rdo-jobs master: Remove redundant var skip_standalone_vm in DPA CI playbooks and job https://review.rdoproject.org/r/c/rdo-jobs/+/51547 | 13:41 |
amoralej | today is video meeting, right? | 14:01 |
karolinku[m] | looks like so! | 14:01 |
amoralej | #startmeeting RDO meeting - 2024-02-07 | 14:02 |
opendevmeet | Meeting started Wed Feb 7 14:02:59 2024 UTC and is due to finish in 60 minutes. The chair is amoralej. Information about MeetBot at http://wiki.debian.org/MeetBot. | 14:02 |
opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 14:02 |
opendevmeet | The meeting name has been set to 'rdo_meeting___2024_02_07' | 14:02 |
amoralej | meeting is in #startmeeting RDO meeting - 2024-MM-DD | 14:03 |
amoralej | meeting is in https://meet.google.com/uzo-tfkt-top | 14:03 |
amoralej | jcapitao[m] apevec ^ | 14:03 |
apevec | still at another call, will jump in a min | 14:04 |
amoralej | #topic Moving RDO from RHBZ to Jira RDO project in issues.redhat.com | 14:21 |
amoralej | #info RDO will be moving to jira to report issues and requests | 14:21 |
amoralej | #info We need to check which documentation we need to change | 14:21 |
amoralej | #action spotz[m] will check the documentation for report to jira | 14:22 |
amoralej | #action karolinku[m] jcapitao[m] amoralej to do a rhbz cleanup session on tuesday 13th at 09:00 UTC | 14:35 |
amoralej | #topic CS9 libvirt 9.10 regression https://issues.redhat.com/browse/RDO-183 | 14:36 |
amoralej | #info latest libvirt build 9.10.0 has broken all the OpenStack deployment jobs on CentOS Stream 9 | 14:36 |
amoralej | #info workarounds are being applied by excluding the known bad build | 14:37 |
amoralej | #topic collaboration with CentOS integration SIG | 14:42 |
amoralej | #info during the CentOS integration SIG meetup at CentOS connect RDO team shared our expectations and use cases | 14:43 |
amoralej | #action we will write our use case and proposal for deliverables from integration SIG to make sure there is a common understanding of it | 14:44 |
amoralej | #topic documentation changes | 14:49 |
amoralej | #info blogs content will be moved to a new location to be defined | 14:50 |
amoralej | #info No more content should be added to blogs.rdoproject.org | 14:51 |
amoralej | #topic chair next week | 14:51 |
amoralej | #action jcapitao[m] will chair next week meeting | 14:52 |
amoralej | #endmeeting | 14:53 |
opendevmeet | Meeting ended Wed Feb 7 14:53:30 2024 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 14:53 |
opendevmeet | Minutes: https://meetings.opendev.org/meetings/rdo_meeting___2024_02_07/2024/rdo_meeting___2024_02_07.2024-02-07-14.02.html | 14:53 |
opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/rdo_meeting___2024_02_07/2024/rdo_meeting___2024_02_07.2024-02-07-14.02.txt | 14:53 |
opendevmeet | Log: https://meetings.opendev.org/meetings/rdo_meeting___2024_02_07/2024/rdo_meeting___2024_02_07.2024-02-07-14.02.log.html | 14:53 |
*** d34dh0r5- is now known as d34dh0r53 | 14:57 | |
jcapitao[m] | apevec: CI jobs are ok with libvirt-10.0.0-2.el9 for Antelope as well https://review.opendev.org/c/openstack/puppet-openstack-integration/+/908309 | 15:02 |
rdogerrit | Merged rdoinfo master: Revert "Add python-oauthlib as dependency" https://review.rdoproject.org/r/c/rdoinfo/+/51574 | 16:08 |
rdogerrit | Merged rdoinfo master: Promote CBS tags update for caracal-9s-testing https://review.rdoproject.org/r/c/rdoinfo/+/51562 | 16:10 |
jcapitao[m] | amoralej: FYI I explained the DLRN issue we hit with neutron FTBFS on Bobcat | 16:16 |
jcapitao[m] | https://issues.redhat.com/browse/PCTOOLING-1212 | 16:16 |
jcapitao[m] | filed as a bug | 16:16 |
amoralej | cool | 16:16 |
jcapitao[m] | but dunno if we should fix it though | 16:16 |
amoralej | i doubt there is a way that would not break in other cases | 16:18 |
amoralej | it'd be good to check git history to see why that --after was included | 16:18 |
jcapitao[m] | you mean, if we rely on git tree order it may break ? | 16:23 |
amoralej | we may, there may be some reason why we do it as we do it | 16:26 |
amoralej | i remember hitting issues with ordering and timestamps in the past | 16:26 |
jcapitao[m] | amoralej: I did the archeology work :) | 16:29 |
jcapitao[m] | the first "--after" that appeard is from initial commit ! | 16:29 |
jcapitao[m] | https://github.com/softwarefactory-project/DLRN/blame/450aad09f7375b453c8d37be68958e46fd07ca2c/delorean/shell.py#L21 | 16:29 |
amoralej | good | 16:29 |
jcapitao[m] | it was then move to https://github.com/softwarefactory-project/DLRN/commit/4a47b45b6049ed57ba2afa6fb6fa701cc2c731c7#diff-4159879fab8dcf9947f58c889240eaf3d2d3e7d6d165fb1a4740ffc8e49058f2R54 few commits after | 16:30 |
amoralej | then it may be good to check it, but we need to think well about that | 16:30 |
jcapitao[m] | so we don't have the explicit reason ... | 16:30 |
amoralej | jcapitao[m], did you get any info about why the neutron-lib repo has those weird timestamps ? | 16:34 |
jcapitao[m] | yes, I discussed with neutron guys, TL;DR: Github UI shows the commits based on commit date | 16:36 |
jcapitao[m] | https://github.com/openstack/neutron-lib/commits/3.8.1/ | 16:37 |
jcapitao[m] | whereas gitea relies on actual git tree https://opendev.org/openstack/neutron-lib/commits/tag/3.8.1 | 16:38 |
jcapitao[m] | amoralej: ^ | 16:38 |
jcapitao[m] | and the actual patches was created on Aug 18 2023, not edited since then | 16:39 |
jcapitao[m] | Gerrit keep the commit date as-is when merging | 16:39 |
jcapitao[m] | so it's nominal case, I was mislead by Github UI yesterday | 16:41 |
jcapitao[m] | IMHO DLRN should relies on git tree order which is the source of truth | 16:42 |
jcapitao[m] | s/relies/rely/ | 16:42 |
amoralej | jcapitao[m], i don't understand anything at this point | 16:59 |
amoralej | so, the problem was that the author of the review had the commit locally for months before sending it :) | 17:03 |
amoralej | and yep, gerrit keeps it unchanged | 17:03 |
amoralej | actually, imo git or gerrit or something should check it | 17:03 |
amoralej | but, i.e. https://review.opendev.org/c/openstack/neutron-lib/+/903334 was created on Dec 11 | 17:04 |
jcapitao[m] | exactly it was written in August but merged in December | 17:04 |
amoralej | not only merged | 17:05 |
amoralej | sent | 17:05 |
amoralej | PS1 is from Dec 11 | 17:05 |
jcapitao[m] | yes but commit date was kept https://review.opendev.org/c/openstack/neutron-lib/+/903334/2//COMMIT_MSG#5 | 17:06 |
amoralej | yes | 17:06 |
amoralej | actually, i guess commit date should be newer that author date | 17:07 |
amoralej | https://review.opendev.org/c/openstack/neutron-lib/+/903334/2//COMMIT_MSG | 17:07 |
amoralej | but it's the oposite | 17:07 |
jcapitao[m] | yeah agree, I need to check Gerrit doc on that | 17:09 |
amoralej | i suspect this is more an issue in the repo side that in the dlrn side | 17:10 |
amoralej | even if 2224 was sent before than the other two, when it was merged it had to be rebased and commitdate should be updated when rebased | 17:12 |
amoralej | https://docs.github.com/en/account-and-profile/setting-up-and-managing-your-github-profile/managing-contribution-settings-on-your-profile/troubleshooting-commits-on-your-timeline | 17:12 |
jcapitao[m] | I see your point, a "normal" rebase operation would have update commit date | 17:13 |
amoralej | yes, that's my understanding | 17:14 |
amoralej | the infra team may help | 17:14 |
amoralej | he may have a better understanding | 17:14 |
jcapitao[m] | I'll ask infra team tomorrow | 17:16 |
rdogerrit | rdo-trunk rdo-trunk proposed openstack/horizon-distgit rpm-master: python-django-horizon: failed to build 8a3006756 https://review.rdoproject.org/r/c/openstack/horizon-distgit/+/51567 | 17:32 |
apevec | what is the question for the infra team? | 22:53 |
apevec | BTW I'm questioning if keeping those repos under https://repos.fedorapeople.org/openstack/EOL/ makes sense, they actually won't work, would they? | 22:56 |
apevec | for those where SIG repos got moved to vault.c.o | 22:56 |
apevec | also few are yum repos using up space with probably non-functional RPMs: | 22:57 |
apevec | 6.5MEOL/openstack-diablo... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/lsuttufcuLfGwRUKrxzbEMgV>) | 22:57 |
apevec | those are from the old times, pre-CBS, when we were copying from Fedora Koji EPEL builds... | 22:58 |
apevec | s/repos/rdo-release*.rpm/ | 23:02 |
apevec | accidentally, in http://mirror.centos.org/centos/8-stream/cloud/x86_64/openstack-train/Packages/o/ is only ovs, ovn - looks like we have this mis-tagged? | 23:12 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!