Thursday, 2026-03-19

@noonedeadpunk:matrix.orgHey folks! I was wondering, if anything was done recently regarding FIPS in CI? As started seeing md5 being unavailable in our CI jobs:09:33
https://zuul.opendev.org/t/openstack/build/a12ed42b5863490e9c51f37135daa277/log/job-output.txt#14965-15011
@noonedeadpunk:matrix.orgas I can't really reproduce that locally so far10:13
@noonedeadpunk:matrix.orgso it would be great to have a hold node, to check if I can find smth specific on the CI image10:25
@fungicide:matrix.orgDmitriy Rabotyagov: we don't have any way to enable fips mode on ubuntu, that requires special licenses added to get access to the required packages12:01
@noonedeadpunk:matrix.orgwell, there were discussions about cannonical willing to give smth out, so I was checking12:02
@fungicide:matrix.orgit does though look like something ansible is calling (openssl?) doesn't like the checksum format12:02
@fungicide:matrix.orgwhere is the cirros image checksum it's comparing to stored?12:02
@noonedeadpunk:matrix.orgbut I think I have found the issue just now: https://review.opendev.org/c/openstack/openstack-ansible-os_tempest/+/886165/8/vars/main.yml12:03
@noonedeadpunk:matrix.orgit's ansible's get_url which does checksum verification12:03
@noonedeadpunk:matrix.orgweird I haven't seen that locally, and that it took like 2 month or smth to see the problem12:05
@fungicide:matrix.orgthat's definitely changing the format of the string and not merely the hash value12:06
@fungicide:matrix.orgso i agree, seems a likely candidate12:07
@noonedeadpunk:matrix.orgI wonder if the reason why we haven't seen the issue earlier is some kind of zuul having repo in a desynced state or smth like that12:16
@noonedeadpunk:matrix.orgbut anyway12:16
@noonedeadpunk:matrix.orglet's see if reverting the format works (it should)12:16
-@gerrit:opendev.org- Zuul merged on behalf of Brian Haley: [opendev/irc-meetings] 980731: Move neutron team and CI meeting earlier by 1 hour https://review.opendev.org/c/opendev/irc-meetings/+/98073112:50
@shrews:matrix.orgfungi: Clark thx for the bindep reviews/merge! since tomorrow marks exactly a year since the last release, any chance we'll see a new release in the near term?13:06
@fungicide:matrix.orgshrews et al: there are currently 4 merged changes since bindep 2.13.0, 9a97635 is enough of a behavior change to warrant 2.14.0 as the next version (and is the only commit with a release note); there are currently 20 outstanding changes for review in bindep, are there any others we should try to get in asap before tagging?13:10
@fungicide:matrix.orgit looks like 202717, 202718 and 202238 may be obsolete with shrews's latest change13:11
@fungicide:matrix.orgor at least would need a different exit code13:12
@fungicide:matrix.orghah, those changes are mine actually, and over 8 years old at this point. think i'm just going to abandon them13:14
@shrews:matrix.orglol13:14
@fungicide:matrix.orgso old i've long ago forgotten i'd even written them13:15
@fungicide:matrix.orgokay, those are gone. skimming the rest, a bunch of them are my packaging-related updates that need to wait for us to drop support for some older python versions, and do-not-merge test changes related to them13:19
@fungicide:matrix.orgi'll also abandon 605613 since it's tox-specific adjustments and we use nox now13:21
@fungicide:matrix.orgthere are a couple of old refactors (455832 and 456071), a 6yo wip change from ianw (753198), and then a few that look like maybe legit fixes/improvements we didn't finish reviewing: 689602, 703055, 735142, 73514313:25
@fungicide:matrix.orgnone look particularly urgent to include before we release13:25
@shrews:matrix.orgto be clear, not pressuring you for an immediate release, if you'd like time to review any outstanding changes. just hoping for one "soon-ish"  :)13:27
@fungicide:matrix.org`9a97635 (HEAD, tag: 2.14.0, origin/master, origin/HEAD, gerrit/master, gerrit/HEAD) New exit code of 2 for input file parse errors.`13:28
@fungicide:matrix.orgi propose to push that ^ and then it's probably time to talk about python versions13:28
@fungicide:matrix.orgshrews: i didn't see any straight away, but encourage others to double-check that assertion13:29
@fungicide:matrix.orgafter the release, if we drop support for python 3.8 then i think we'll be able to merge 945403 and 945416, possibly also 940841 and 94397613:34
@clarkb:matrix.orgfungi that sha and tag lgtm. I think if we delay any longer then we're likely doing a release of bindep after the openstdack release just to avoid problems during release week. So no is probably good14:50
@clarkb:matrix.org* fungi that sha and tag lgtm. I think if we delay any longer then we're likely doing a release of bindep after the openstack release just to avoid problems during release week. So now is probably good14:53
@fungicide:matrix.orgthanks! i'll go ahead and push it14:53
@fungicide:matrix.organd done, zuul's wheels should be in motion14:54
@clarkb:matrix.orgif things have settled down maybe today is a good day for gitea 1.25.5: https://review.opendev.org/c/opendev/system-config/+/98080815:00
@fungicide:matrix.orgi think so, i can help keep an eye on it after lunch15:00
@clarkb:matrix.orgcool I expect to be around for most of the day but do have to do the school run later and will probably also try to get a bike ride in before the rain returns15:01
@clarkb:matrix.orgfungi: the change doesn't have any reviews yet if you're willing to do that too that would be great15:01
@fungicide:matrix.orgyep15:01
@clarkb:matrix.orgthen we can figure out approval timing based on lunch schedules if it looks good15:01
@clarkb:matrix.orglet me check that the daily run for gitea didn't fail last night either since we had that apt lock issue15:02
@clarkb:matrix.orghttps://zuul.opendev.org/t/openstack/build/81cd744542d44261b3d5fca12d8288da it succeeded15:02
@clarkb:matrix.orgfungi: the goaccess report jobs are failing since we added new static servers and I expect we need to add their ssh host keys to some job vars? I don't think this is urgent and may be best to address once we have a longer term plan for the static servers?15:03
@fungicide:matrix.orgyeah, i think we're going to have to make adjustments to the job to accommodate the split between servers anyway?15:05
@clarkb:matrix.orgHrm ya not sure how it would handle that if at all15:07
@fungicide:matrix.orgalso we have reports for some of the same sites on 3 different servers from different points in time as things were shuffled around15:08
@fungicide:matrix.orgbut also with the current swarm of bots i question even more whether those reports contain much of value15:09
@fungicide:matrix.orgheading out to lunch, should be back in a bit15:22
-@gerrit:opendev.org- Clark Boylan proposed: [opendev/system-config] 981315: Remove lxd-installer from our servers https://review.opendev.org/c/opendev/system-config/+/98131516:02
@clarkb:matrix.orgThat is a small cleanup I noticed yesterday ^ basically a straggler lxd related package that we can remove since we already remove lxd itself.16:02
-@gerrit:opendev.org- Owen Jones proposed: [openstack/diskimage-builder] 981319: mellanox: remove qperf package from element https://review.opendev.org/c/openstack/diskimage-builder/+/98131916:09
-@gerrit:opendev.org- Owen Jones proposed: [openstack/diskimage-builder] 981319: mellanox: remove qperf package from element https://review.opendev.org/c/openstack/diskimage-builder/+/98131916:14
@mnasiadka:matrix.orgAny chance for a review on https://review.opendev.org/c/opendev/system-config/+/980851 ?16:35
@mnasiadka:matrix.orgClark: do you want me to workflow the Gitea upgrade - or some other time of day is better?16:41
@clarkb:matrix.orgmnasiadka: now is good thanks16:43
@clarkb:matrix.orgoh right that apt sources change was on my list to review. I'll review it now16:44
@mnasiadka:matrix.orgThanks16:45
-@gerrit:opendev.org- Monty Taylor https://matrix.to/#/@mordred:inaugust.com proposed: [openstack/project-config] 981322: Add wandertracks-rail repo https://review.opendev.org/c/openstack/project-config/+/98132216:46
@clarkb:matrix.orgmnasiadka: also is there a change to add the new openmetal mirror to our inventory yet? I should review that one if so16:47
@mnasiadka:matrix.orgYeah, it’s there - let me find it16:48
@mnasiadka:matrix.orgI thought it is, but can’t find it - maybe it’s on my laptop - will push it in around 30 mins16:49
@mnasiadka:matrix.orghttps://review.opendev.org/c/opendev/system-config/+/98106416:52
@mnasiadka:matrix.orgThere it is16:52
@clarkb:matrix.orgperfect I suspect we can get that deployed today as well if fungi takes a look16:53
@fungicide:matrix.orgoh, yep, on it16:57
@fungicide:matrix.orglgt16:58
@fungicide:matrix.orgm16:58
-@gerrit:opendev.org- Zuul merged on behalf of Monty Taylor https://matrix.to/#/@mordred:inaugust.com: [openstack/project-config] 981322: Add wandertracks-rail repo https://review.opendev.org/c/openstack/project-config/+/98132217:02
@fungicide:matrix.org¡aventureros al trein!17:08
@mordred:waterwanders.comheck yeah! sometimes I like to aventureros al trein17:09
@fungicide:matrix.orgles aventuriers du rail17:09
@fungicide:matrix.orgzug um zug17:09
@fungicide:matrix.orgthe many international names of the "ticket to ride" board game17:10
@fungicide:matrix.orgbut you're living it!17:10
@clarkb:matrix.orgI was just talking to someone how about caltran is like 8.50 between san francisco and sunnyvale compared to the $120 waymo and faster.17:13
@fungicide:matrix.orgcrazypants17:13
@clarkb:matrix.orgwaymo would be more competitive on time if they used freeways to be fair. But even then I think you're looking at just under an hour for each and a significant price savings to take the train and then waymo the last mile17:14
@fungicide:matrix.orgflying cars... 'nuff said17:15
-@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: [opendev/system-config] 980808: Upgrade gitea to 1.25.5 https://review.opendev.org/c/opendev/system-config/+/98080817:37
-@gerrit:opendev.org- Zuul merged on behalf of Michal Nasiadka: [opendev/system-config] 981064: Add new mirror in openmetal https://review.opendev.org/c/opendev/system-config/+/98106417:37
@clarkb:matrix.orggitea09 and 10 have updated17:43
@clarkb:matrix.orgmy client and ip get gitea10 and it seems to work in the web ui (reports the new version too)17:44
@clarkb:matrix.orgI'll wait for deployment to finish and then test git clone as its a bit harder to confirm the backend with the git client17:44
@clarkb:matrix.orgall giteas should be upgraded now17:53
@clarkb:matrix.orgbuildset was successful too: https://zuul.opendev.org/t/openstack/buildset/ac3cd9b4988447829200f1ad916b15de17:53
@fungicide:matrix.orgseems to still be working normally for me17:54
@clarkb:matrix.orggit clone works for me17:54
@fungicide:matrix.orgbrowsing, cloning17:54
@clarkb:matrix.orgI'll try to find something that has replicated then I think we can consider this done17:54
@clarkb:matrix.orghttps://review.opendev.org/c/openstack/openstack-helm-images/+/981336 was just pushed a couple minutes after I verified all giteas updated. It shows up in gitea: https://opendev.org/openstack/openstack-helm-images/commit/205dcf1ac235cae6ed788a7641fcad23269e7bfc17:57
@fungicide:matrix.orglgtm!17:59
@clarkb:matrix.orgyup now we wait for the new mirror to get configured. mnasiadka once that is done the next step is verifying that the mirror serves traffic at https://mirror02.iad3.openmetal.opendev.org and if it does we can update dns to point at it. Then we can delete the old server and its dns records. I don't think there is a rush to this so I'll probably let you continue to carry that forward as a way to gain familiarity etc. Let me know if there are any questions that come up18:03
-@gerrit:opendev.org- Zuul merged on behalf of Owen Jones: [openstack/diskimage-builder] 981319: mellanox: remove qperf package from element https://review.opendev.org/c/openstack/diskimage-builder/+/98131919:13
-@gerrit:opendev.org- David Ostrovsky proposed: [zuul/zuul-jobs] 896974: Add generic bazelisk build role https://review.opendev.org/c/zuul/zuul-jobs/+/89697419:31
-@gerrit:opendev.org- David Ostrovsky proposed: [zuul/zuul-jobs] 896974: Add generic bazelisk build role https://review.opendev.org/c/zuul/zuul-jobs/+/89697419:54
-@gerrit:opendev.org- Goutham Pacha Ravi proposed: [openstack/project-config] 981404: Add Code-Review and Rollcall-Vote to governance dashboard https://review.opendev.org/c/openstack/project-config/+/98140420:07
-@gerrit:opendev.org- David Ostrovsky proposed: [zuul/zuul-jobs] 896974: Add generic bazelisk build role https://review.opendev.org/c/zuul/zuul-jobs/+/89697420:12
-@gerrit:opendev.org- David Ostrovsky proposed: [zuul/zuul-jobs] 896974: Add generic bazelisk build role https://review.opendev.org/c/zuul/zuul-jobs/+/89697420:38
@clarkb:matrix.orgfungi: I'm looking at the old infra root imap folders and I think we may want to keep the clouds/ folder contents, letsencrypt, pypi, and then services content except for transifex?20:47
@clarkb:matrix.orgalso related I'm trying to update two laptops becuse I need to configure imap for the new account(s) on them...20:47
@clarkb:matrix.orgif we feel like that is too much then maybe we start with LE, pypi, and the services?20:52
@fungicide:matrix.orgno i think that's fine. as long as i have a list to work from i'm happy to churn through those20:53
@clarkb:matrix.orgI don't think we need all the gerrit spam20:53
@clarkb:matrix.orgor jenkins etc20:53
@fungicide:matrix.orgsounds like we still have about a month until the old account goes away20:53
@clarkb:matrix.orgyup I think April 22 is the deadline20:53
@fungicide:matrix.organd at most it'll take me a few days, maybe i can knock it out over this weekend even20:54
-@gerrit:opendev.org- David Ostrovsky proposed: [zuul/zuul-jobs] 896974: Add generic bazelisk build role https://review.opendev.org/c/zuul/zuul-jobs/+/89697420:58

Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!