Wednesday, 2019-11-20

*** slaweq has quit IRC00:01
clarkbfungi: did we ever put inap back into service with a bigger mirror?00:01
*** ociuhandu has quit IRC00:02
*** jklare has joined #openstack-infra00:02
ianwi feel like i made that mirror?00:04
clarkbianw: I guess the issue was we ran out of disk for our afs + apache caches00:05
clarkbthe old server had had a cinder volume maybe or a bigger root disk?00:05
clarkb(I was sick at the tiem of this happening so memory may be fuzzy)00:05
clarkbI've confirmed that certcheck is now happy with the certs on opendev.org too00:09
clarkbemails should stop now00:09
openstackgerritDavid Moreau Simard proposed opendev/system-config master: DNM: Test ara 1.2 for bridge  https://review.opendev.org/69510800:11
*** slaweq has joined #openstack-infra00:11
openstackgerritIan Wienand proposed opendev/system-config master: vos-release: fix typo in script name  https://review.opendev.org/69511700:11
ianwclarkb: hrm, i don't remember than ...00:12
*** xek has quit IRC00:12
ianwhttps://review.opendev.org/#/c/695099/ passed, so i'd say ansible 2.9.1 on bridge is *probably* ok00:13
*** apetrich has quit IRC00:15
*** slaweq has quit IRC00:18
*** armax has joined #openstack-infra00:25
*** slaweq has joined #openstack-infra00:42
*** rlandy|rover has quit IRC00:44
*** slaweq has quit IRC00:46
*** rfolco has quit IRC00:52
*** slaweq has joined #openstack-infra01:01
clarkbianw: fungi inap is still disabled according to project-config content. Looking at server flavors on inap an easy option would be to rebuild the server with flavor A1.16. This has 320GB of disk. Another option is a cinder volume, our quota does allow us 20 volumes and plenty of GB so we'd be able to do that in inap01:01
clarkbI can give that a go tomorrow morning likely01:01
ianw++01:02
clarkblet me know if either of you have a preference for appraoch above. Maybe benj_ has thoughts too? benj_ basically our existing cloud local mirror host doesn't have enough disk for its caches. Our two options are to supplement existing server with a cinder volume or build a new server with bigger root disk.01:02
openstackgerritGhanshyam Mann proposed openstack/openstack-zuul-jobs master: Explicit mention py2 for grenade py2 jobs  https://review.opendev.org/69512401:03
*** slaweq has quit IRC01:05
*** goldyfruit_ has joined #openstack-infra01:07
fungiclarkb: ahh, yeah, i was thinking of just attaching a 200gb cinder volume like we've done on other mirrors01:16
fungicarve into 2x 100gb logical volumes and mount to the two cache dirs01:16
fungiianw: yeah, i confirmed from old graphs the previous inap mirror had separate volumes mounted on the cachedirs, the new one only had a rootfs and not nearly enough space01:17
ianwfungi: hrm, according to my notes i brought it up on 10-24 ... ?  i feel like i would have added a volume but maybe i forgot :/01:19
fungior created the volumes but didn't add them to fstab? i haven't checked the cinder inventory there01:20
clarkb/dev/disk/by-uuid only shows two devices01:21
clarkbvolume list doesn't show any volumes. I can create and attach, add to lvm and format, etc tomorrow01:21
*** ociuhandu has joined #openstack-infra01:22
ianwok can see if i can take a look too after getting through some other bits01:22
clarkbk I'll check scrollback then01:23
*** slaweq has joined #openstack-infra01:25
*** slaweq has quit IRC01:30
*** ociuhandu has quit IRC01:34
*** lifeless has quit IRC01:34
*** mriedem has quit IRC01:44
*** lifeless has joined #openstack-infra01:49
openstackgerritDavid Moreau Simard proposed opendev/system-config master: DNM: Test ara 1.2 for bridge  https://review.opendev.org/69510801:51
openstackgerritMerged opendev/system-config master: vos-release: fix typo in script name  https://review.opendev.org/69511701:58
*** goldyfruit_ has quit IRC01:58
*** jklare has quit IRC02:00
*** jklare has joined #openstack-infra02:02
*** jamesmcarthur has joined #openstack-infra02:06
*** lifeless has quit IRC02:12
*** michael-beaver has quit IRC02:12
dmsimardianw: almost there -- it's installed and is enabled but fails during the ara-report role: http://paste.openstack.org/show/786377/02:12
*** slaweq has joined #openstack-infra02:13
ianwok, it's not writing on the db?02:16
dmsimardnope, my mistake for the database config02:16
openstackgerritDavid Moreau Simard proposed opendev/system-config master: DNM: Test ara 1.2 for bridge  https://review.opendev.org/69510802:17
*** igordc has quit IRC02:18
*** slaweq has quit IRC02:18
*** lifeless has joined #openstack-infra02:24
*** slaweq has joined #openstack-infra02:24
*** slaweq has quit IRC02:30
*** roman_g has quit IRC02:33
*** ociuhandu has joined #openstack-infra02:35
*** slaweq has joined #openstack-infra02:35
*** ociuhandu has quit IRC02:39
dmsimardsame error again... hrm02:43
dmsimard"rsync: link_stat \"/var/cache/ansible/ara.sqlite\" failed: No such file or directory (2)\nrsync error: some files/attrs were not transferred (see previous errors) (code 23) at main.c(1655)02:44
ianwargh, key didn't roll out for local afs ... why ...?02:44
dmsimardI think it's because the configuration isn't being taken into account so it's at the default location instead of that one02:44
*** slaweq has quit IRC02:48
*** gyee has quit IRC02:53
dmsimardohhhh, I know what the problem is :D02:53
*** slaweq has joined #openstack-infra03:00
*** slaweq has quit IRC03:04
*** slaweq has joined #openstack-infra03:05
openstackgerritDavid Moreau Simard proposed opendev/system-config master: DNM: Test ara 1.2 for bridge  https://review.opendev.org/69510803:14
*** slaweq has quit IRC03:16
openstackgerritIan Wienand proposed opendev/system-config master: vos-release: fix key sourcing  https://review.opendev.org/69512903:23
*** slaweq has joined #openstack-infra03:26
*** goldyfruit_ has joined #openstack-infra03:28
*** slaweq has quit IRC03:31
ianwhrm, i think i found a problem; puppet overwrites the authorized_keys03:33
*** slaweq has joined #openstack-infra03:35
*** slaweq has quit IRC03:40
*** ricolin has joined #openstack-infra03:43
openstackgerritIan Wienand proposed opendev/system-config master: vos-release: fix key sourcing; disable exclusive key  https://review.opendev.org/69512903:44
openstackgerritDavid Moreau Simard proposed opendev/system-config master: DNM: Test ara 1.2 for bridge  https://review.opendev.org/69510803:48
*** slaweq has joined #openstack-infra03:55
*** slaweq has quit IRC04:02
*** liuyulong has quit IRC04:05
*** slaweq has joined #openstack-infra04:10
openstackgerritDavid Moreau Simard proposed opendev/system-config master: DNM: Test ara 1.2 for bridge  https://review.opendev.org/69510804:10
*** slaweq has quit IRC04:15
*** armax has quit IRC04:18
*** slaweq has joined #openstack-infra04:39
*** slaweq has quit IRC04:44
openstackgerritIan Wienand proposed zuul/zuul-jobs master: use-buildset-registry: Vendor pytoml and remarshal  https://review.opendev.org/69505004:46
*** goldyfruit_ has quit IRC05:00
*** slaweq has joined #openstack-infra05:01
*** rh-jelabarre has quit IRC05:07
*** slaweq has quit IRC05:08
*** slaweq has joined #openstack-infra05:11
*** slaweq has quit IRC05:20
*** soniya29 has joined #openstack-infra05:21
*** pcaruana has joined #openstack-infra05:25
*** Lucas_Gray has joined #openstack-infra05:28
*** slaweq has joined #openstack-infra05:29
*** iokiwi has quit IRC05:29
*** surpatil has joined #openstack-infra05:30
*** iokiwi has joined #openstack-infra05:30
*** ociuhandu has joined #openstack-infra05:31
openstackgerritIan Wienand proposed zuul/nodepool master: Dockerfile: add DEBUG environment flag  https://review.opendev.org/69484505:32
openstackgerritIan Wienand proposed zuul/nodepool master: [wip] move openstack testing to use containerised daemon  https://review.opendev.org/69346405:32
*** bdodd has quit IRC05:32
*** slaweq has quit IRC05:33
*** ociuhandu has quit IRC05:35
*** bdodd has joined #openstack-infra05:36
ianwcorvus: as discussed, didn't see any non-bugfix gitea items that I think would affect us on https://review.opendev.org/#/c/694894/05:38
*** bdodd has quit IRC05:41
*** slaweq has joined #openstack-infra05:42
*** Lucas_Gray has quit IRC05:45
*** bdodd has joined #openstack-infra05:46
*** slaweq has quit IRC05:47
openstackgerritMerged openstack/project-config master: Remove unused path variable from promote secrets  https://review.opendev.org/69491205:51
*** slaweq has joined #openstack-infra06:11
*** pcrews has quit IRC06:14
*** slaweq has quit IRC06:15
*** pcrews has joined #openstack-infra06:20
*** SurajPatil has joined #openstack-infra06:26
*** slaweq has joined #openstack-infra06:27
*** surpatil has quit IRC06:28
*** slaweq has quit IRC06:32
*** larainema has joined #openstack-infra06:33
*** slaweq has joined #openstack-infra06:39
*** slaweq has quit IRC06:44
*** roman_g has joined #openstack-infra07:10
*** slaweq has joined #openstack-infra07:11
*** dpawlik has joined #openstack-infra07:14
*** slaweq has quit IRC07:15
*** pgaxatte has joined #openstack-infra07:19
*** udesale has joined #openstack-infra07:25
*** tosky has joined #openstack-infra07:35
*** priteau has quit IRC07:38
*** surpatil has joined #openstack-infra07:42
*** SurajPatil has quit IRC07:44
gmannclarkb: fungi do you know why there is stable/train in master barbican grenade job inventory (any other grenade job is all master) - https://zuul.opendev.org/t/openstack/build/a3a17f8e4d054b6d9c417b44077159a1/log/zuul-info/inventory.yaml#3207:53
*** slaweq has joined #openstack-infra07:56
*** dpawlik has quit IRC07:56
*** ccamacho has joined #openstack-infra07:57
*** tesseract has joined #openstack-infra07:58
*** soniya29 has quit IRC08:04
*** roman_g has quit IRC08:06
*** ociuhandu has joined #openstack-infra08:09
*** tkajinam has quit IRC08:17
*** ociuhandu has quit IRC08:19
*** dpawlik has joined #openstack-infra08:21
*** dchen has quit IRC08:28
*** jpena|off is now known as jpena08:49
*** elod_off is now known as elod08:50
*** ralonsoh has joined #openstack-infra08:51
*** rpittau|afk is now known as rpittau08:53
*** ociuhandu has joined #openstack-infra08:55
*** ociuhandu has quit IRC08:59
*** soniya29 has joined #openstack-infra09:00
*** ociuhandu has joined #openstack-infra09:00
*** lucasagomes has joined #openstack-infra09:01
*** apetrich has joined #openstack-infra09:03
*** ociuhandu has quit IRC09:03
*** ociuhandu has joined #openstack-infra09:04
*** rfolco has joined #openstack-infra09:05
*** jamesmcarthur has quit IRC09:07
*** rfolco has quit IRC09:09
*** ociuhandu has quit IRC09:11
*** ociuhandu has joined #openstack-infra09:11
*** ociuhandu has quit IRC09:13
*** tosky has quit IRC09:13
*** SurajPatil has joined #openstack-infra09:14
*** ociuhandu has joined #openstack-infra09:14
*** tosky has joined #openstack-infra09:14
*** SurajPatil is now known as SurajPatil|lunch09:15
*** ociuhandu has quit IRC09:15
*** jamesmcarthur has joined #openstack-infra09:15
*** ociuhandu has joined #openstack-infra09:16
*** surpatil has quit IRC09:16
*** ociuhandu has quit IRC09:17
*** ociuhandu has joined #openstack-infra09:18
*** ociuhandu has quit IRC09:19
*** ociuhandu has joined #openstack-infra09:20
*** Lucas_Gray has joined #openstack-infra09:22
*** ociuhandu has quit IRC09:26
*** ociuhandu has joined #openstack-infra09:27
*** ociuhandu has quit IRC09:29
*** ociuhandu has joined #openstack-infra09:29
*** pkopec has joined #openstack-infra09:31
*** ociuhandu has quit IRC09:32
*** ociuhandu has joined #openstack-infra09:32
*** ociuhandu has quit IRC09:34
*** ociuhandu has joined #openstack-infra09:35
*** ociuhandu has quit IRC09:37
*** iurygregory has joined #openstack-infra09:37
*** ociuhandu has joined #openstack-infra09:37
*** ociuhandu has quit IRC09:39
*** ociuhandu has joined #openstack-infra09:40
*** ociuhandu has quit IRC09:43
*** ociuhandu has joined #openstack-infra09:43
*** dpawlik has quit IRC09:45
*** ociuhandu has quit IRC09:45
*** ociuhandu has joined #openstack-infra09:46
*** rcernin has quit IRC09:46
*** Wryhder has joined #openstack-infra09:46
*** derekh has joined #openstack-infra09:47
*** ociuhandu has quit IRC09:48
*** Lucas_Gray has quit IRC09:48
*** ociuhandu has joined #openstack-infra09:49
*** Wryhder is now known as Lucas_Gray09:49
*** ociuhandu has quit IRC09:50
*** ociuhandu has joined #openstack-infra09:51
*** priteau has joined #openstack-infra09:52
*** ociuhandu has quit IRC09:52
*** ociuhandu has joined #openstack-infra09:53
*** gfidente has joined #openstack-infra09:55
*** ociuhandu has quit IRC09:56
*** ociuhandu has joined #openstack-infra09:57
*** jaosorior has joined #openstack-infra09:57
*** ociuhandu has quit IRC09:58
*** ociuhandu has joined #openstack-infra09:58
*** ruffian_sheep has joined #openstack-infra09:59
*** rascasoft has joined #openstack-infra10:00
*** priteau has quit IRC10:06
*** SurajPatil|lunch is now known as SurajPatil10:10
*** dpawlik has joined #openstack-infra10:20
*** dpawlik has quit IRC10:24
*** dtantsur|afk is now known as dtantsur10:27
AJaegergmann: I think the branch matcher at https://opendev.org/openstack/barbican/src/branch/stable/train/.zuul.yaml#L138 is the problem, those are evil.10:27
AJaegerSo, https://review.opendev.org/689458 and friends might be needed10:27
AJaegergmann: answered as well via email with the above10:30
*** ricolin has quit IRC10:32
*** xek has joined #openstack-infra10:35
*** dtantsur is now known as dtantsur|brb10:48
*** dpawlik has joined #openstack-infra10:51
mordredianw: what's our current story for certs in openstack.org domains - that's not LE yet right?10:54
fricklermordred: iirc we can do LE there, too. just need the foundation folks to install a CNAME entry for the acme subdomain manually10:56
mordredah - cool10:56
* mordred needs a real cert on storyboard-dev to make forward progress on gerrit10:56
mordredfungi: ^^ when you get up, could you put in an _acme_challenge for storyboard-dev? or do I need to ping jimmy?10:58
mordreduhm11:02
mordredwe seem to be half-transitioned between storyboard.*.openstack.org and storyboard.*.opendev.org11:03
mordredthe servers and the ansible and the puppet all seem to think they are .opendev.org - but there's no DNS entry11:03
fricklermordred: I haven't found any example of a openstack.org LE cert, though the spec mentions it. possibly migrating to storyboard-dev.opendev.org will actually be easier11:07
mordredyeah - since that should really just be adding dns records and adding a redirect11:08
*** PrinzElvis has quit IRC11:08
*** dpawlik has quit IRC11:12
openstackgerritMonty Taylor proposed opendev/system-config master: Update gitea to v1.10.0  https://review.opendev.org/69087311:14
*** ociuhandu has quit IRC11:21
*** lpetrut has joined #openstack-infra11:26
*** pgaxatte has quit IRC11:42
*** Lucas_Gray has quit IRC11:47
*** dpawlik has joined #openstack-infra11:49
*** ociuhandu has joined #openstack-infra11:51
*** dpawlik has quit IRC11:53
*** hrw has joined #openstack-infra11:53
hrwmorning11:54
*** rlandy has joined #openstack-infra11:55
*** rlandy is now known as rlandy|rover11:55
*** lmiccini has joined #openstack-infra11:55
hrwcan ubuntu mirror on mirror.sjc1.vexxhost.openstack.org machine get updated? it is out of sync breaking CI jobs11:55
hrwINFO:kolla.common.utils.kolla-toolbox:The following packages have unmet dependencies:11:55
hrwINFO:kolla.common.utils.kolla-toolbox: libssl-dev : Depends: libssl1.1 (= 1.1.1-1ubuntu2.1~18.04.4) but 1.1.1-1ubuntu2.1~18.04.5 is to be installed11:55
hrwhttps://046b6520e005cc815b3d-5ba6792c7427741c3f6f06d35721722e.ssl.cf5.rackcdn.com/695170/1/check/kolla-build-ubuntu-source/808dffa/kolla/build/000_FAILED_kolla-toolbox.txt.gz job log11:56
fricklerhrw: we have an issue with afs volumes in general, not sure about the current state, but this is being handled already11:57
hrwfrickler: thanks11:57
*** SurajPatil has quit IRC12:08
*** PrinzElvis has joined #openstack-infra12:15
*** dave-mccowan has quit IRC12:17
*** larainema has quit IRC12:23
*** pcaruana has quit IRC12:24
*** ruffian_sheep has quit IRC12:29
*** jpena is now known as jpena|lunch12:29
*** soniya29 has quit IRC12:40
*** pgaxatte has joined #openstack-infra12:45
*** pcaruana has joined #openstack-infra12:50
*** derekh has quit IRC12:55
*** ociuhandu has quit IRC12:58
*** goldyfruit_ has joined #openstack-infra12:59
fungimordred: when frickler says "need the foundation folks to install a CNAME entry" what he really means is "add cnames in that domain in rackspace like we always have"12:59
fungibut sure, i can take a stab at it shortly12:59
*** Lucas_Gray has joined #openstack-infra13:01
funginot realy half-transitioned on storyboard servers, we had simply declared a while back that we would start naming servers in the opendev.org domain if they were for services opendev would be providing, whether or not the service's domain was opendev.org at the time13:01
*** derekh has joined #openstack-infra13:02
fungiso that we didn't need to rebuild them later to change their instance names13:02
fungimordred: but yeah, no reason i can see not to go ahead and set it up as storyboard-dev.opendev.org at this point13:04
fungithough i do have a to do item to reinstall it with python313:04
fungithat may wait until we dockerify the deployment13:04
*** priteau has joined #openstack-infra13:05
fungiit will break old storyboard-webcliennt draft builds until those expire out of the log archive, but we can always recheck after we update the jobs13:06
*** ricolin has joined #openstack-infra13:07
fungias for the afs mirrors, the recovery from yesterdays hung server reboot looks like it will need to deal with stale vldb locks13:09
fungiCould not lock the VLDB entry for the volume 536870949. VLDB: vldb entry is already locked13:09
fungierrors like that13:09
*** Lucas_Gray has quit IRC13:11
mordredfungi: nod to all of the above13:11
*** Lucas_Gray has joined #openstack-infra13:11
mordredfungi: so - my primary motivation here is that storyboard-dev is pointed to by review-dev and so we inject the storyboard-dev ssl cert, which currently involves running a java command, and therefore would be something that would need to be done inside the contianer13:12
mordredHOWEVER13:12
mordredwe don't do that on review.o.o - it's purely a step we do on review-dev ... so maybe just omitting it from review-dev and not blocking gerrit docker on the storyboard-dev chain (which we can also work in parallel) is fine?13:13
mordred(because the storyboard thread could quickly become _another_ fun project, albeit a worthwhile one)13:13
fungioh, yep, that's purely to deal with the self-signed snakeoil cert on storyboard-dev13:13
fungijava doesn't like connecting to those without additional trust chain13:14
*** goldyfruit_ has quit IRC13:16
mordredyeah. which is where this all just gets gnarly :)13:17
*** dave-mccowan has joined #openstack-infra13:21
*** iurygregory has quit IRC13:31
*** iurygregory has joined #openstack-infra13:32
*** mriedem has joined #openstack-infra13:34
*** jpena|lunch is now known as jpena13:36
*** iurygregory has quit IRC13:36
*** iurygregory has joined #openstack-infra13:37
*** dpawlik has joined #openstack-infra13:39
mnasiadkagood afternoon13:41
mnasiadkaIs there a way to add more logs to be pushed to logstash on a project level? I mean that list: https://opendev.org/opendev/base-jobs/src/branch/master/roles/submit-logstash-jobs/defaults/main.yaml#L913:42
*** eharney has quit IRC13:43
*** ociuhandu has joined #openstack-infra13:44
*** roman_g has joined #openstack-infra13:48
*** ociuhandu has quit IRC13:48
*** liuyulong has joined #openstack-infra13:57
*** AJaeger has quit IRC13:58
fungimnasiadka: not as far as i know, but is there a reason not to just collect the log in question from any build which creates it?14:03
mnasiadkafungi: in kolla-ansible we put deploy logs in logs/{{ hostname }}/ansible/ - we would like to use the elastic-recheck functionality, but those logs are not in logstash - so we can't ;)14:04
mnasiadkafungi: other option is to add ansible/.* to the submit-logstash-jobs role I guess14:04
*** AJaeger has joined #openstack-infra14:06
fungilogs/.*/ansible/.* presumably, though yeah i can imagine for some builds that might be a massive number of log entries14:08
fungibut maybe no moreso than devstack's logs14:08
mnasiadkafungi: so - it's around 15 megabytes per job (https://storage.gra1.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_34c/695190/1/check/kolla-ansible-ubuntu-source/34cf558/primary/logs/ansible/)14:09
mnasiadkaI don't know what is the usual devstack log size ;-)14:09
mnasiadkawe could try although to fetch only the failed tasks into some other log file and push that to logstash14:10
*** priteau has quit IRC14:11
fungior perhaps only archive them on failed builds14:12
*** priteau has joined #openstack-infra14:12
*** pgaxatte has quit IRC14:12
mnasiadkafungi: that's even better - is there some knob to turn? :)14:13
fungiyeah, it would be in your post playbook for the job defintion14:13
*** priteau has quit IRC14:13
fungithough be aware that tends to make for false positives if you match on omnipresent-but-benign log entries14:13
fungisince it's easy to think you've found the signature for all your failures because the entry only appears on build failures (because the log it's written to is only archived on failure)14:14
*** jaosorior has quit IRC14:18
*** Lucas_Gray has quit IRC14:21
mnasiadkafungi: got it14:21
*** pgaxatte has joined #openstack-infra14:29
AJaegerconfig-core, please review https://review.opendev.org/#/c/695124/ to help with the py3 grenade fun14:29
*** PrinzElvis has quit IRC14:38
*** eharney has joined #openstack-infra14:39
*** goldyfruit_ has joined #openstack-infra14:43
*** kjackal has joined #openstack-infra14:45
*** jamesmcarthur has quit IRC14:49
dmsimardclarkb, ianw: got latest version of ara working on bridge jobs: https://3753dea4ea75a3fe0692-37ad574b65e44b7a1b1058900a1af5e6.ssl.cf5.rackcdn.com/695108/7/check/system-config-run-backup/6e71391/bridge.openstack.org/ara-report/14:56
*** jklare has quit IRC14:57
*** jamesmcarthur has joined #openstack-infra14:57
*** rh-jelabarre has joined #openstack-infra14:58
*** dtantsur|brb is now known as dtantsur15:00
*** jklare has joined #openstack-infra15:01
*** dpawlik has quit IRC15:08
*** johnthetubaguy has joined #openstack-infra15:09
*** hrw has left #openstack-infra15:09
*** ociuhandu has joined #openstack-infra15:13
*** ociuhandu has quit IRC15:19
*** udesale has quit IRC15:20
*** udesale has joined #openstack-infra15:20
*** rlandy|rover is now known as rlandy|rover|mtg15:50
*** gfidente has quit IRC15:56
*** ociuhandu has joined #openstack-infra15:56
fungicorvus: clarkb: anything in particular i should do with the afs volumes which have stale vldb locks other than remove the lock and redo the vos release with -localauth? they're locked for a release operation according to vos examine, with afs02.dfw unsurprisingly showing it's still at an old release15:59
corvusfungi: i think as long as the release isn't actually running, it's safe to unlock and redo.16:00
*** kjackal has quit IRC16:00
fungiyeah, these all got killed by timeout16:00
*** rlandy|rover|mtg is now known as rlandy|rover16:01
*** ociuhandu has quit IRC16:01
clarkbfungi: does that mean the local releasechange wasnt working?16:05
*** armax has joined #openstack-infra16:05
fungiclarkb: yeah, i think based on scrollback it needs some additional hacking to get puppet to stop overwriting the authorized_keys file ansible is setting16:06
fungisounds like they're both fighting over it16:06
fungii suppose we could create a separate shell user for that purpose16:07
*** whoami-rajat_ has joined #openstack-infra16:08
*** iurygregory has quit IRC16:12
*** udesale has quit IRC16:14
*** gyee has joined #openstack-infra16:16
fungiinfra-root: i have a root screen session on mirror-update.openstack.org holding the update flock for the ubuntu mirror and have done `vos unlock mirror.ubuntu --localauth && vos release mirror.ubuntu --localauth` in a root screen session on afsdb0116:17
fungithis will likely take a while as i'm expecting it to have to perform a full release to afs02.dfw16:18
*** dpawlik has joined #openstack-infra16:18
gmannAJaeger: thanks, still not sure how that happened (or zuul magic on branch variant ) so asked the question on ML for learning to me and other community members if such issue comes again.16:18
fungionce it's done, i'll start in on any other volumes in a similar situation16:18
AJaegergmann: I just pointed corvus to it as well (on #zuul)16:19
gmannthanks16:22
fungigonna disappear to grab lunch, then return and continue16:22
*** ricolin has quit IRC16:24
AJaegergmann: want to ask stable team or barbican to review all changes? With the -1s from Zuul, you could ask an infra-root as well to force merge the changes to fix your problem. having stable team reviewed would be nice in that case16:29
*** lmiccini has quit IRC16:29
gmannAJaeger: yeah pinged barbican stable core and mriedem and tonyb  also16:29
*** dpawlik has quit IRC16:29
AJaegerthanks, gmann16:30
*** ociuhandu has joined #openstack-infra16:33
*** dpawlik has joined #openstack-infra16:35
*** ociuhandu has quit IRC16:38
*** ociuhandu has joined #openstack-infra16:38
*** jaosorior has joined #openstack-infra16:40
*** dpawlik has quit IRC16:41
*** pgaxatte has quit IRC16:51
*** goldyfruit___ has joined #openstack-infra16:53
*** goldyfruit___ is now known as goldyfruit16:54
*** goldyfruit_ has quit IRC16:56
*** gfidente has joined #openstack-infra16:57
*** lucasagomes has quit IRC16:57
*** dpawlik has joined #openstack-infra17:12
*** xek_ has joined #openstack-infra17:15
*** xek has quit IRC17:17
*** rpittau is now known as rpittau|afk17:18
*** dpawlik has quit IRC17:21
*** takamatsu has joined #openstack-infra17:23
*** ramishra has quit IRC17:23
*** michael-beaver has joined #openstack-infra17:25
*** goldyfruit has quit IRC17:27
*** gfidente is now known as gfidente|afk17:28
*** goldyfruit has joined #openstack-infra17:30
*** goldyfruit has quit IRC17:31
*** diablo_rojo has joined #openstack-infra17:35
openstackgerritJames E. Blair proposed zuul/zuul-jobs master: WIP: use-buildset-registry: add podman support  https://review.opendev.org/69505117:35
openstackgerritJames E. Blair proposed zuul/zuul-jobs master: Add build-container-image role  https://review.opendev.org/69525117:35
*** dpawlik has joined #openstack-infra17:35
*** diablo_rojo has quit IRC17:40
clarkbI've almost got inap's mirror set up with the new volume. Waiting on the old apache proxy dir to be cleaned out so that I can mount over the top of it cleanly17:47
clarkbI'll mount -a then restart apache and openafs-client and test then reboot and test. If happy at that point I think we can turn nodepool back on17:48
fungisounds good, thanks!17:49
fungithe vos release for mirror.ubuntu is, surprising to no one i'm sure, still going17:49
*** jaosorior has quit IRC17:50
*** dpawlik has quit IRC17:51
*** whoami-rajat_ is now known as whoami-rajat17:53
*** jaosorior has joined #openstack-infra17:57
*** igordc has joined #openstack-infra18:00
*** jpena is now known as jpena|off18:00
*** derekh has quit IRC18:01
clarkbhttp://mirror.mtl01.inap.opendev.org/ubuntu/ seems to be working now after a reboot18:04
clarkbnow to find the change to revert18:04
fungiclarkb: Change-Id: Iecc8e302f8d6ac374a81deaeddf14bcd997829b218:05
openstackgerritClark Boylan proposed openstack/project-config master: Revert "Revert "Enable inap-mtl01""  https://review.opendev.org/69525418:05
clarkbfungi: ^18:09
clarkbfungi: do you think we should try doing concurrent vos releases? I wonder if each release would get its own udp window size or if it is shared. If shared one at a time is probably better. If independent then running them separately will likely speed stuff up18:11
fungii suspect the bandwidth is being maxxed out by as many or as few as we run concurrently18:14
fungibut cacti would probably confirm18:14
*** tesseract has quit IRC18:18
*** mattw4 has joined #openstack-infra18:19
*** jaosorior has quit IRC18:22
*** diablo_rojo has joined #openstack-infra18:23
clarkb#status log Added 200GB cinder volume to mirror01.mtl01.inap.opendev.org to be used as openafs and apache2 cache backend. Filesystems created and mounted at the appropriate cache locations and services restarted.18:24
openstackstatusclarkb: finished logging18:24
*** ralonsoh has quit IRC18:29
openstackgerritMerged openstack/project-config master: Revert "Revert "Enable inap-mtl01""  https://review.opendev.org/69525418:30
fungiclarkb: yeah, i doubt we'll gain a lot by trying parallel runs: http://cacti.openstack.org/cacti/graph.php?action=view&local_graph_id=6405&rra_id=all18:37
fungiseems somewhat capped around 300mbps18:37
*** diablo_rojo has quit IRC18:48
*** tosky has quit IRC18:55
clarkbmnasiadka: fungi: things to keep in mind re indexing of logs. Those log files appear to be without timestamps. I've been pretty stringent about requiring those in the past if you want to index in elasticsearch. Without timestamps the timestamps yuo'll get out of logstash are meaningless (time of index rather than time of actual log occurance). We also do our best to index > DEBUG logs. This helps us18:55
clarkbavoid flooding the elasticsearch servers with records (we peak at over a billion a day) as well as ensures that the correct level of logging is exposed to users when errors occur18:55
clarkbmnasiadka: fungi I'm not sure how easy it will be to convince ansible to log sanely, btu at the very least we will want timestamps18:56
mnasiadkaclarkb: noted - timestamps should not be problematic18:56
clarkbas a general note timestamps are also a requirement of third party CIs18:56
clarkbthey make debugging jobs much easier18:56
clarkbnot having them is a fairly major bug imo18:56
*** ociuhandu has quit IRC18:57
*** ociuhandu has joined #openstack-infra18:59
clarkboh also because the format is different than one of the existing formats we parse (apache, syslog/journald, oslo, zuul console log) we may need to add a new parsing definition to the logstash rules19:01
clarkbhowever if you end up with logs that look like the zuul console log (which is basically timestamp + hostname + ansible output) then you can probably tell logstash it is a console log19:01
*** ociuhandu has quit IRC19:03
*** jamesmcarthur has quit IRC19:14
*** jamesmcarthur has joined #openstack-infra19:27
openstackgerritAlbin Vass proposed zuul/zuul-jobs master: Adds go jobs and roles.  https://review.opendev.org/69111419:28
*** lpetrut has quit IRC19:32
*** lpetrut has joined #openstack-infra19:32
openstackgerritMerged openstack/openstack-zuul-jobs master: Explicit mention py2 for grenade py2 jobs  https://review.opendev.org/69512419:32
clarkbno certcheck warning for opendev today \o/19:35
*** ociuhandu has joined #openstack-infra19:35
AJaegergreat work, clarkb !19:40
*** pkopec has quit IRC19:41
fungiexcellent!19:44
*** ociuhandu has quit IRC19:46
*** pkopec has joined #openstack-infra19:47
*** ociuhandu has joined #openstack-infra19:47
*** iurygregory has joined #openstack-infra19:48
*** ociuhandu has quit IRC19:53
*** sreejithp has joined #openstack-infra19:55
*** pkopec has quit IRC19:55
*** ociuhandu has joined #openstack-infra19:57
AJaegergmann: glad to see that the barbican job now works - but that might just be luck. We need to merge the change on all branches - so, https://review.opendev.org/689461 , https://review.opendev.org/689463 , https://review.opendev.org/689462 , https://review.opendev.org/689466 are needed as well. infra-root, should we force-merge those zuul updates to remoe a wrong branch matcher?19:58
*** ociuhandu has quit IRC20:05
*** whoami-rajat has quit IRC20:08
*** eharney has quit IRC20:08
*** gfidente|afk has quit IRC20:12
*** jamesmcarthur has quit IRC20:20
ianwfungi / clarkb : could you poke at https://review.opendev.org/#/c/695129/ to fix the half-deployed vos release key20:21
fungiabsumotutely20:21
*** jamesmcarthur has joined #openstack-infra20:23
*** jamesmcarthur has quit IRC20:24
fungiianw: i'm probably unclear on the hostvars structure there, but see inline question20:26
*** jamesmcarthur has joined #openstack-infra20:26
ianwfungi: it should be; copy-paste error :/20:27
fungino worries20:28
openstackgerritIan Wienand proposed opendev/system-config master: vos-release: fix key sourcing; disable exclusive key  https://review.opendev.org/69512920:29
tonybgmann: I must have missed that ping .... what do you need me to review?20:30
tonybgmann: Oh I see it now in the nova channel, I'll do it today20:31
*** ociuhandu has joined #openstack-infra20:36
*** iurygregory has quit IRC20:42
*** michael-beaver has quit IRC20:45
*** ociuhandu has quit IRC20:51
*** sreejithp has quit IRC20:52
*** ociuhandu has joined #openstack-infra20:53
openstackgerritJames E. Blair proposed zuul/zuul-jobs master: WIP: use-buildset-registry: add podman support  https://review.opendev.org/69505120:54
*** sreejithp has joined #openstack-infra20:57
*** ociuhandu has quit IRC21:06
*** eharney has joined #openstack-infra21:14
openstackgerritJames E. Blair proposed zuul/zuul-jobs master: WIP: use-buildset-registry: add podman support  https://review.opendev.org/69505121:22
*** xek__ has joined #openstack-infra21:30
*** xek_ has quit IRC21:32
*** dtantsur is now known as dtantsur|afk21:32
*** tosky has joined #openstack-infra21:41
*** pcaruana has quit IRC21:42
fungivos release of mirror.ubuntu finished. rerunning the cronjob for it now to make sure it completes smoothly21:47
openstackgerritJames E. Blair proposed zuul/zuul-jobs master: WIP: use-buildset-registry: add podman support  https://review.opendev.org/69505121:47
*** ociuhandu has joined #openstack-infra21:59
fungiokay, http://files.openstack.org/mirror/ubuntu/timestamp.txt is caught up. parsing the vos listvldb output, i see 13 other volumes in a similar state (all mirrors) which also need this treatment22:04
openstackgerritJames E. Blair proposed zuul/zuul-jobs master: WIP: use-buildset-registry: add podman support  https://review.opendev.org/69505122:04
fungii'll knock out the rest of the ubuntu-related ones first22:05
ianwfungi: hrm 2 hours for a full release?22:05
ianwdid you sprinkle magic dust?22:05
fungiit may not have been a full release after all22:06
fungibut it was sufficient for afs02.dfw to start showing current again22:07
fungiso i'll take it22:07
*** ociuhandu has quit IRC22:07
ianwhuh, yeah that's the way it *should* work :)22:07
fungianyway, mirror.ubuntu-ports is underway now22:07
ianwhave you thought about moving reprepro to ansible?22:08
ianwhow much magic is in the puppet?22:08
fungiwe add a handful of conffiles in /etc and some cronjobs22:08
fungiso not much, i don't think22:09
fungiand install the package for reprepro22:09
fungialso a wrapper script22:09
fungimaybe log rotation?22:09
fungibut no, i haven't given thought to it yet22:10
openstackgerritJames E. Blair proposed zuul/zuul-jobs master: WIP: use-buildset-registry: add podman support  https://review.opendev.org/69505122:18
*** slaweq has quit IRC22:26
ianwfungi: yeah, none of this vos-release via localauth stuff applies for mirror-update.openstack.org, that's what had me thinking about it22:28
ianwthose tend to lock themselves as much as the opendev.org server22:28
*** dave-mccowan has quit IRC22:29
*** rcernin has joined #openstack-infra22:31
*** xek_ has joined #openstack-infra22:34
*** slaweq has joined #openstack-infra22:35
*** xek__ has quit IRC22:36
*** slaweq has quit IRC22:40
*** kopecmartin is now known as kopecmartin|off22:42
*** sreejithp has quit IRC22:46
openstackgerritJames E. Blair proposed zuul/zuul-jobs master: use-buildset-registry: Vendor pytoml and remarshal  https://review.opendev.org/69505022:48
openstackgerritJames E. Blair proposed zuul/zuul-jobs master: Add build-container-image role  https://review.opendev.org/69525122:48
openstackgerritJames E. Blair proposed zuul/zuul-jobs master: WIP: use-buildset-registry: add podman support  https://review.opendev.org/69505122:48
*** slaweq has joined #openstack-infra22:50
*** slaweq has quit IRC22:55
*** jamesmcarthur has quit IRC22:57
clarkbianw sorry I was dealing with tye last of the plumbing (I hope). eating lunch now then back to help23:01
ianwclarkb: oh, just the mention of the word plumbing sends a chill down my spine :)23:02
ianwwhen you think about it, your whole house is surrounded by pressurised containers just waiting to explode and cost you a small fortune :)23:02
clarkbAs of fridqy I can confirm this fact23:03
clarkbour water main broke23:03
*** tkajinam has joined #openstack-infra23:05
openstackgerritMichael Johnson proposed openstack/diskimage-builder master: Add output for mis-configured element scripts  https://review.opendev.org/68078523:05
openstackgerritJames E. Blair proposed zuul/zuul-jobs master: use-buildset-registry: Vendor pytoml and remarshal  https://review.opendev.org/69505023:05
openstackgerritJames E. Blair proposed zuul/zuul-jobs master: Add build-container-image role  https://review.opendev.org/69525123:05
openstackgerritJames E. Blair proposed zuul/zuul-jobs master: WIP: use-buildset-registry: add podman support  https://review.opendev.org/69505123:05
*** slaweq has joined #openstack-infra23:06
ianwclarkb: oh dear ... on your side or the company's side?23:06
clarkbmy side23:06
clarkbmore than a meter underground23:07
clarkb1.3m or so23:07
ianw:(23:07
ianwi do know a good plumber; i don't think he works in your area though :)23:08
*** xek_ has quit IRC23:08
*** slaweq has quit IRC23:10
clarkbwhat we discovered is you have to do your own hole digging because you can get emergency weekend plumbers but not hole diggers23:11
*** slaweq has joined #openstack-infra23:11
clarkbI just finished filling the hole today as we'll get freezing temps tonight23:11
ianwat least you can skip the gym?23:13
*** rlandy|rover is now known as rlandy|rover|bbl23:13
gmannAJaeger: yeah, merging all is the target.23:14
*** slaweq has quit IRC23:15
clarkbianw: ya I haven't been on the bike in like a week. Dirt throwing seems sufficient :)23:16
clarkband now I'm back at the desktop. Anything I should be looking at?23:16
clarkbAJaeger: gmann re force merging if we get barbican to ack it first that would be ideal but I think we can do that23:16
gmannsure, let me check the status on patch.23:17
ianwclarkb: https://review.opendev.org/#/c/695129/ if you could; the role added yeterday is broken without it23:19
clarkbianw: done23:20
*** slaweq has joined #openstack-infra23:21
gmannclarkb: AJaeger seems fedora_latest job failing consistently on barbican stable/stein. something wrong so let's wait for force merge.23:23
*** slaweq has quit IRC23:26
clarkbianw: also TIL about with_inventory_hostname23:28
ianwclarkb: yeah, i learnt about it making the letsencrypt roles; each host that's renewing registers a var, then the "deploy dns records" walks across all the letsencrypt group hostnames to check for values23:30
*** slaweq has joined #openstack-infra23:30
clarkbgmann: https://review.opendev.org/#/admin/groups/1097,members looks like stable-maint-core and dave mccowan are who we need to bug?23:31
clarkbgmann: If we can't reach them in a reasonable amount of time we can still force push but it is always good to get the ack of thse responsible first23:31
gmannok.23:33
clarkbgmann: are things broken for anyone but barbican?23:33
clarkbif not maybe devstack can proceed and that will be the carrot for barbican to review these changes?23:34
gmannclarkb: i have not checked yet. let me see check23:34
*** slaweq has quit IRC23:35
gmannother projects are good. https://review.opendev.org/#/q/status:open+branch:stable/stein23:35
gmannlet me check log and ping dave mccowan23:36
clarkbsanity checking the inap mirror things continue to look good there23:36
*** dchen has joined #openstack-infra23:36
*** slaweq has joined #openstack-infra23:37
openstackgerritMerged opendev/system-config master: vos-release: fix key sourcing; disable exclusive key  https://review.opendev.org/69512923:39
gmannseems it is broken since 8 months. i pinged on barbican channel- https://review.opendev.org/#/c/650415/23:39
*** slaweq has quit IRC23:45
*** kaisers1 has quit IRC23:48
*** slaweq has joined #openstack-infra23:48
*** kaisers has joined #openstack-infra23:48
*** slaweq has quit IRC23:53
*** mattw4 has quit IRC23:59

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!