Thursday, 2026-02-26

-@gerrit:opendev.org- Dmitriy Rabotyagov proposed: [openstack/project-config] 978084: Stop publishing collection to PyPi https://review.opendev.org/c/openstack/project-config/+/97808414:47
@fungicide:matrix.orgas we discussed yesterday, a limnoria restart after 16z today should be safe, so i'll go ahead and recheck 975321 since it'll need at least an hour to get back through check and gate anyway15:15
-@gerrit:opendev.org- Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org proposed: [opendev/system-config] 978087: Revert "Cap django-health-check for ara installations" https://review.opendev.org/c/opendev/system-config/+/97808715:19
@clarkb:matrix.orgfungi: sounds good re limnoria15:59
@clarkb:matrix.orgI went ahead and approved the django-health-check cap revert since there is a good ara report in the test jobs for that change16:03
@clarkb:matrix.orgNo reason to hang onto that extra bit of info for any longer16:03
@fungicide:matrix.orgthanks16:12
-@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: [opendev/system-config] 975321: Upgrade limnoria and bump its container to trixie https://review.opendev.org/c/opendev/system-config/+/97532116:33
-@gerrit:opendev.org- Zuul merged on behalf of Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org: [opendev/system-config] 978087: Revert "Cap django-health-check for ara installations" https://review.opendev.org/c/opendev/system-config/+/97808716:33
@fungicide:matrix.orginfra-prod-service-eavesdrop is running now16:36
@fungicide:matrix.orgwelcome shrews! long time no see16:37
@shrews:matrix.orgohai!16:37
@shrews:matrix.orgI actually am here with purpose16:37
@shrews:matrix.orgI my try to fix something in bindep (https://storyboard.openstack.org/#!/story/2008049) if I can find a few moments, but wanted input from you most excellent of folks. 16:38
@shrews:matrix.org* I may try to fix something in bindep (https://storyboard.openstack.org/#!/story/2008049) if I can find a few moments, but wanted input from you most excellent of folks. 16:38
@fungicide:matrix.orgopendevmeet rejoined irc channels after the container restart, so that went well16:38
@shrews:matrix.orgAny specific concern with changing bindep to return additional error codes on bindep.txt parsing errors?16:39
@shrews:matrix.orgAnsible has a product that would benefit from it16:39
@shrews:matrix.organd also, hi fungi and Clark and corvus and anyone else I may have missed  :)16:40
@fungicide:matrix.orgshrews: seems useful, i would also make sure that any error message output like tracebacks go to stderr rather than stdout (only package names should be written to stdout during non-prettified operation)16:41
@fungicide:matrix.orgi think the main bug identified in story 2008049 is that it's writing error messages to stdout16:42
@shrews:matrix.orgwell, two bugs: traceback on stdout, and no able to distiquish an error by return code. but yeah, I can totally make sure we keep stdout clean.16:43
@shrews:matrix.orgthx for the feedback fungi. i'll see about regenerating my gerrit skills first. hope you all have been well!16:45
@jim:acmegating.comhi shrews !16:46
@fungicide:matrix.orgbeen doing great!16:46
@shrews:matrix.orgossum16:46
@clarkb:matrix.orgoh hey!16:55
@clarkb:matrix.orgfungi's feedback all seem great to me16:55
@clarkb:matrix.orgfungi: re eavesdrop I'll drop a message in #openstack-infra and double check channel logging16:56
@clarkb:matrix.orghttps://meetings.opendev.org/irclogs/%23openstack-infra/%23openstack-infra.2026-02-26.log and that looks good16:56
@clarkb:matrix.orgcorvus: I think that is most if not all of the images opendev is building running on trixie now. If that helps with planning for switching zuul over16:57
@jim:acmegating.comClark: thanks!  i think we were just waiting for the dust to settle, so i reckon we can just do that any time now?16:57
@clarkb:matrix.orgI suspect so. We haven't really seen any issues on the opendev side either17:00
@clarkb:matrix.orgI guess one thing that makes zuul different is the afs stuff17:00
@clarkb:matrix.orgso that would be something to check potentially17:01
-@gerrit:opendev.org- Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org proposed:17:33
- [opendev/system-config] 978110: Add WAF rules to more static sites https://review.opendev.org/c/opendev/system-config/+/978110
- [opendev/system-config] 978111: Test that WAF rules share a common block pool https://review.opendev.org/c/opendev/system-config/+/978111
@clarkb:matrix.orgfungi: that change isn't adding the html sed rewrite stuff only the robots.txt. I'm fine with that but wanted to make sure this was intentional17:36
@fungicide:matrix.orgyes, as previously discussed, well-behaved crawlers are known to cache robots.txt files, so we'd want to give them a chance to notice those sites grew some17:41
@fungicide:matrix.orgthe mod_sed directives will be coming in a separate change that we can merge a few days later17:42
@fungicide:matrix.orgi'm also working on robots.txt additions to openstack/openstack-manuals and openstack/api-site before adding waf rules to those vhosts17:43
-@gerrit:opendev.org- Joaci Otaviano de Morais proposed wip: [openstack/project-config] 978113: Add Netapp Storage App to StarlingX https://review.opendev.org/c/openstack/project-config/+/97811317:54
-@gerrit:opendev.org- Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org proposed: [opendev/system-config] 978118: Add WAF rules for developer and docs.openstack.org https://review.opendev.org/c/opendev/system-config/+/97811818:06
@fungicide:matrix.orgi honestly did not expect 978111 to work, so that's an excellent feature bonus18:21
@clarkb:matrix.orgOh cool. I had no idea if it would but agree that is a nice bit of unexpected functionality 18:26
@fungicide:matrix.orghttps://zuul.opendev.org/t/openstack/build/d5c12bb7dab4416b984ce4d45f2b8052/log/job-output.txt#10993-11207 is worrisome, i would not expect 403 responses there, we don't even have waf rules set up for the site until the change depending on that deploys18:30
-@gerrit:opendev.org- Leonardo Fagundes Luz Serrano proposed: [openstack/project-config] 978124: Add meta-packages repo to StarlingX https://review.opendev.org/c/openstack/project-config/+/97812418:34
@clarkb:matrix.orgfungi: is the manuals build doing a crawl itself and then tripping over it?18:35
@clarkb:matrix.orgfungi: maybe we need to check the logs on static for the ip address belonging to that build node?18:35
@fungicide:matrix.orgthat's what i wondered, but i can't find any evidence that it follows links in robots.txt, and also we don't block it for that site yet18:35
@fungicide:matrix.orgso far only docs.opendev.org has the waf rules, and the only hit i find in its access log is a wget from one of my own machines trying it out18:36
@clarkb:matrix.orgif I access the url I get content and a 20018:36
@fungicide:matrix.orgsame18:37
@fungicide:matrix.orgthe other possibility is that something changed in our ua filters?18:37
@clarkb:matrix.orgoh yes maybe it is the UA filter18:37
@fungicide:matrix.orgthough that job succeeded 2 days ago18:37
@fungicide:matrix.orghuh, but earlier on tuesday it errored the same exact way18:38
@clarkb:matrix.orgI did a grep against all `*_error.log` files in the apache log dir on static and that ip doesn't show up. We should record the waf rule entries in that file. Then there is only one entry in `*_access.log` with a 200 response18:39
@clarkb:matrix.orgalmost like it is talking to something else18:40
@fungicide:matrix.orgit's not provider-specific. of the three most recent builds, there was a failure and a success in ovh-gra1, and a failure in rax-ord18:41
@fungicide:matrix.orgwith both failures looking the same18:42
-@gerrit:opendev.org- Joaci Otaviano de Morais proposed wip: [openstack/project-config] 978113: Add Netapp Storage App to StarlingX https://review.opendev.org/c/openstack/project-config/+/97811318:46
@clarkb:matrix.orgis it possibly using some external service or proxy to make the requests? I don't see evidence of that in the log though18:46
@clarkb:matrix.orgoh maybe it is using v618:47
@fungicide:matrix.orgyeah, i can't find anything that sets something like that up in the job18:47
@fungicide:matrix.orgwell, we have a failure and success from the same provider though18:48
@clarkb:matrix.orgfungi: `(110)Connection timed out: [client 2001:4801:7825:103:be76:4eff:fe10:cab:58112] AH00036: access to /swift/latest/index.html failed (filesystem path '/afs/openstack.org/docs')`18:48
@clarkb:matrix.orgit was ipv6 and I read that as a failure to find the content on disk not due to our waf or ua rules18:48
@clarkb:matrix.orgthat is from /var/log/apache2/docs.openstack.org_error.log18:48
@fungicide:matrix.orgodd that a timeout would result in a 403 error18:48
@clarkb:matrix.orgdmesg -T reports `[Thu Feb 26 16:41:20 2026] afs: Waiting for busy volume 536870992 () in cell openstack.org`18:49
@clarkb:matrix.orgI can ls /afs/openstack.org/docs though18:50
@fungicide:matrix.orgusually those are fairly short-lived incidents when you just see one mention in dmesg18:50
@clarkb:matrix.orgyes looks like that was the only one from today and occurred over an hour before the 403 errors18:51
@clarkb:matrix.orgthe afs cache lives on /opt and is only 63% used18:51
@fungicide:matrix.orgi've rechecked the change for now18:52
@fungicide:matrix.orgi wonder if the workers were all busy or something18:53
@fungicide:matrix.orgthe apache worker processes on static0218:55
@clarkb:matrix.orgthat may explain the timeout18:57
@clarkb:matrix.organd maybe the path and filesystem mapping is just additional info but not directly relevant18:59
@clarkb:matrix.orgfungi: also if that tool is making a new connection rather than using a requests session or similar then it could be contributing to going over the limits18:59
@clarkb:matrix.org* fungi: also if that tool is making a new connection for each path request rather than using a requests session or similar then it could be contributing to going over the limits18:59
@fungicide:matrix.orgthe test checking those urls just seems to be doing a `requests.head()` and then looking at the `status_code` so requests was at least claiming a http/403 status19:00
@fungicide:matrix.orgbut yeah, it looks like an individual session per url: https://opendev.org/openstack/openstack-manuals/src/branch/master/tools/www-generator.py#L18619:00
@clarkb:matrix.orgmaybe we update that test/tool to use a session for the head calls and see if that helps19:00
@fungicide:matrix.orgrechecking the change did indeed succeed19:03
@fungicide:matrix.orgunrelated, the wiki server's dmesg is flooded with `nf_conntrack: table full, dropping packet` and i'm getting a mediawiki error page served up saying it can't access its database19:07
-@gerrit:opendev.org- Leonardo Fagundes Luz Serrano proposed: [openstack/project-config] 978124: Add meta-packages repo to StarlingX https://review.opendev.org/c/openstack/project-config/+/97812419:10
-@gerrit:opendev.org- Joaci Otaviano de Morais proposed wip: [openstack/project-config] 978113: Add Netapp Storage App to StarlingX https://review.opendev.org/c/openstack/project-config/+/97811319:31
-@gerrit:opendev.org- Leonardo Fagundes Luz Serrano proposed: [openstack/project-config] 978124: Add meta-packages repo to StarlingX https://review.opendev.org/c/openstack/project-config/+/97812419:38
@fungicide:matrix.org#status log Rebooted wiki.openstack.org restoring it to working order after an unrecoverable database disconnect19:54
@status:opendev.org@fungicide:matrix.org: finished logging19:54
-@gerrit:opendev.org- Joaci Otaviano de Morais proposed wip: [openstack/project-config] 978113: Add Netapp Storage App to StarlingX https://review.opendev.org/c/openstack/project-config/+/97811320:01
-@gerrit:opendev.org- Leonardo Fagundes Luz Serrano proposed: [openstack/project-config] 978124: Add metapackages repo to StarlingX https://review.opendev.org/c/openstack/project-config/+/97812420:13
-@gerrit:opendev.org- Leonardo Fagundes Luz Serrano proposed: [openstack/project-config] 978124: Add metapackages repo to StarlingX https://review.opendev.org/c/openstack/project-config/+/97812420:21

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