Monday, 2021-05-03

« Sunday, 2021-05-02 Index Tuesday, 2021-05-04 »
*** jamesmcarthur has joined #zuul00:46
*** jamesmcarthur has quit IRC01:00
*** jamesmcarthur has joined #zuul01:01
*** jamesmcarthur has quit IRC01:06
*** jamesmcarthur has joined #zuul01:15
*** hamalq has joined #zuul01:49
*** hamalq has quit IRC01:54
*** evrardjp has quit IRC02:33
*** evrardjp has joined #zuul02:33
*** bhavikdbavishi has joined #zuul02:59
*** bhavikdbavishi has quit IRC03:14
*** bhavikdbavishi1 has joined #zuul03:14
*** bhavikdbavishi1 has quit IRC03:18
*** ikhan has joined #zuul03:25
*** ikhan has quit IRC03:30
*** jamesmcarthur has quit IRC03:38
*** jamesmcarthur has joined #zuul03:38
*** bhavikdbavishi has joined #zuul03:42
*** ikhan has joined #zuul03:43
*** jamesmcarthur has quit IRC03:43
*** ikhan has quit IRC03:47
*** hamalq has joined #zuul03:50
*** hamalq has quit IRC03:55
*** ikhan has joined #zuul04:01
*** ikhan has quit IRC04:06
*** jamesmcarthur has joined #zuul04:08
*** ikhan has joined #zuul04:28
*** ikhan has quit IRC04:32
*** jfoufas1 has joined #zuul04:41
*** vishalmanchanda has joined #zuul04:54
*** bhavikdbavishi has quit IRC04:57
openstackgerritSimon Westphahl proposed zuul/zuul master: Optimize layout re-calculation after re-enqueue  https://review.opendev.org/c/zuul/zuul/+/78868105:23
*** ykarel has joined #zuul05:24
*** bhavikdbavishi has joined #zuul05:26
*** ikhan has joined #zuul05:29
*** ikhan has quit IRC05:34
*** hamalq has joined #zuul05:51
*** hamalq has quit IRC05:56
*** saneax has joined #zuul06:10
*** hamalq has joined #zuul06:27
*** ikhan has joined #zuul06:30
*** hamalq has quit IRC06:31
*** hamalq has joined #zuul06:31
*** ikhan has quit IRC06:34
*** jamesmcarthur has quit IRC06:36
*** jamesmcarthur has joined #zuul06:49
*** bhavikdbavishi has quit IRC06:58
*** hashar has joined #zuul07:05
*** rpittau|afk is now known as rpittau07:19
*** nils has joined #zuul07:21
*** avass has joined #zuul07:25
*** bhavikdbavishi has joined #zuul07:29
*** ikhan has joined #zuul07:30
avassfungi: I think either setting ansible_python_interpreter in the job as var or host-var, or set it on the tasks you need it for is the way to fix that. unless the host always needs that, then I suppose setting it in nodepool would be better07:32
*** ikhan has quit IRC07:35
*** jcapitao has joined #zuul07:44
*** tosky has joined #zuul07:50
*** jcapitao has quit IRC07:54
*** jcapitao has joined #zuul07:57
*** jamesmcarthur has quit IRC08:05
*** sshnaidm|off is now known as sshnaidm08:12
*** ikhan has joined #zuul08:31
*** ikhan has quit IRC08:36
*** hamalq has quit IRC08:46
*** ricolin has joined #zuul08:54
*** jcapitao has quit IRC09:02
*** jcapitao has joined #zuul09:03
*** ikhan has joined #zuul09:32
*** hamalq has joined #zuul09:36
*** ikhan has quit IRC09:37
*** hamalq has quit IRC09:40
*** ikhan has joined #zuul09:48
*** ikhan has quit IRC09:52
*** ikhan has joined #zuul10:35
*** ikhan has quit IRC10:40
*** jcapitao is now known as jcapitao_lunch10:56
*** ikhan has joined #zuul11:07
*** ykarel is now known as ykarel|lunch11:11
*** ikhan has quit IRC11:12
openstackgerritSimon Westphahl proposed zuul/zuul master: Fix exception on build started event for noop  https://review.opendev.org/c/zuul/zuul/+/78924511:17
openstackgerritSimon Westphahl proposed zuul/zuul master: Fix exception on build started event for noop jobs  https://review.opendev.org/c/zuul/zuul/+/78924511:21
*** ikhan has joined #zuul11:23
swestcorvus: ^ should fix a minor issue we introduced with 78293911:25
*** ikhan has quit IRC11:27
*** rlandy has joined #zuul11:35
*** hamalq has joined #zuul11:36
*** hamalq has quit IRC11:41
*** Shrews has joined #zuul11:46
*** ykarel|lunch is now known as ykarel11:53
*** ikhan has joined #zuul11:54
*** ikhan has quit IRC11:59
*** jcapitao_lunch is now known as jcapitao12:01
*** jamesmcarthur has joined #zuul12:03
*** ikhan has joined #zuul12:03
*** jamesmcarthur has quit IRC12:04
*** jamesmcarthur has joined #zuul12:05
*** sduthil has joined #zuul12:13
*** hamalq has joined #zuul12:17
*** hamalq has quit IRC12:22
fungiavass: thanks, and yeah i suspect it will always need it overridden (unless ansible's python autodiscovery improves to cover this case), since it only provides a python 2.7 interpreter for transitional purposes but no longer packages many/any modules for it12:22
fungii'll look into solving it in nodepool configuration12:22
*** bhavikdbavishi has quit IRC12:25
openstackgerritTobias Henkel proposed zuul/zuul master: Fix implicit role repos missing in repo-state  https://review.opendev.org/c/zuul/zuul/+/78925212:26
openstackgerritTobias Henkel proposed zuul/zuul master: Fix resolveBranch working on outdated data  https://review.opendev.org/c/zuul/zuul/+/78925312:26
openstackgerritSimon Westphahl proposed zuul/zuul master: Use real result event in event queue test  https://review.opendev.org/c/zuul/zuul/+/78925412:30
*** bhavikdbavishi has joined #zuul12:33
*** jamesmcarthur has quit IRC12:37
openstackgerritTobias Henkel proposed zuul/zuul master: Fix resolveBranch working on outdated data  https://review.opendev.org/c/zuul/zuul/+/78925312:40
*** jamesmcarthur has joined #zuul12:46
tobiashcorvus: one small fix and a regression fix for global repo state ^12:47
tobiashclarkb: ^12:49
*** jamesmcarthur has quit IRC12:51
*** jamesmcarthur has joined #zuul12:53
*** jamesmcarthur has quit IRC13:00
*** jamesmcarthur has joined #zuul13:02
*** jamesmcarthur has quit IRC13:04
*** jamesmcarthur_ has joined #zuul13:04
*** jamesmcarthur_ has quit IRC13:08
*** jamesmcarthur has joined #zuul13:21
*** saneax has quit IRC13:46
*** bhavikdbavishi has quit IRC13:55
corvustobiash, swest: ack, i'll look asap13:59
*** hamalq has joined #zuul14:18
*** hamalq has quit IRC14:22
openstackgerritJeremy Stanley proposed zuul/zuul-jobs master: Add new Debian security mirror suite pattern  https://review.opendev.org/c/zuul/zuul-jobs/+/78909014:35
corvustristanC, tobiash: i think you both use mqtt?  can you take a look at https://review.opendev.org/788101  ?14:37
corvustobiash: and https://review.opendev.org/787743 is pywinrm related, you may be interested14:38
tobiashmqtt lgtm14:39
corvusmordred: i expect you to be fascinated by https://review.opendev.org/780399 due to your delight in cloud behaviors :)14:40
avasscorvus: digitalocean also has private ipv6 :)14:40
corvusavass: so weird! :)14:40
tobiashpywinrm as well, we install it currently using ANSIBLE_EXTRA_PACKAGES :)14:41
avasstobiash: same :)14:42
corvusfungi, clarkb: ^ just fyi -- i find the idea of private ipv6 fascinating14:42
corvustobiash: yeah, it seems like something that should work out of the box :)14:42
fungicorvus: yeah, it used to be referred to as "site-local addressing"14:46
corvuszuul-maint: fyi i will be afk for part of today, and potentially much (or all) of tomorrow.14:47
clarkbtobiash: swest looks like the outdated data change did not pass ci14:47
fungithe idea was that you might want to set up a segmented ipv6 network but lack external v6 connectivity, so not have a global prefix for it14:48
fungiso not about address conservation or as a security measure, just a means of bootstrapping local v6 in advance of having real v6 external routes14:49
tobiashcorvus: looks like a test race14:50
corvusclarkb: i suspect that's a test-case race, which may not be worth fixing since it's about canceling a job and we're about to change that for zk14:50
*** zbr has quit IRC14:51
clarkbtobiash: corvus thanks for the clarification. I'll attempt to review after some breakfast14:51
*** zbr has joined #zuul14:52
*** avass has quit IRC15:04
*** avass has joined #zuul15:05
mordredcorvus: o_O ...15:14
mordredprivate ipv615:14
mordredwhat a complete farce15:14
mordred"we've adopted a broken networking model to work around ipv4 exhaustion - as part of migrating to ipv6 let's keep the broken model that ipv6 allows us to no longer need!"15:16
* mordred throws a wet goat at someone15:16
ShrewsOuch! Whose wet goat is this in my eye????15:22
clarkbolaph is probably wondering where his goat went15:22
Shrewsolaph never leaves his farm these days15:22
*** ykarel is now known as ykarel|away15:24
fungiwell, "private ipv6" (unique local addressing) is not intended to rely on nat. ipv6 lets clients make clear route selections on their own, and can prefer local addressing over global unique addressing, so the idea is that you accept global route announcements on clients which need to talk to the rest of the internet, but rely on ula for local communication15:26
fungiin ipv6, clients are expected to have multiple (and perhaps many) address assignments15:27
fungialso the absence of nat allows clients with private addresses to also talk to each other's public addresses without hairpinning/tromboning through translating routers15:30
clarkbtobiash: swest  left a couple of thoughts on https://review.opendev.org/c/zuul/zuul/+/789253 if we want to address those in a followup I think that is fine since nothing is critical there15:36
clarkbin my review I meant15:36
*** jfoufas1 has quit IRC15:38
tobiashwhoops :D15:38
*** hamalq has joined #zuul16:00
*** hamalq has quit IRC16:09
*** hamalq has joined #zuul16:14
*** hamalq_ has joined #zuul16:16
*** hamalq has quit IRC16:20
*** ykarel|away has quit IRC16:24
*** nils has quit IRC16:31
*** rpittau is now known as rpittau|afk16:32
*** hashar is now known as hasharSkating16:33
*** hasharSkating is now known as hashar16:35
*** jcapitao has quit IRC17:05
corvustobiash, swest: the error in https://review.opendev.org/789245 is observed by the unit tests, but we don't fail them because the scheduler correctly proceeds despite the error; i wonder if we could find a way to detect that.... maybe in a test logging figure we can determine if log.exception has been called?17:15
*** Shrews has quit IRC17:32
tobiashthat's an interesting idea but I guess that will also catch quite a lot unclean shutdowns17:36
*** jamesmcarthur has quit IRC17:40
*** jfoufas1 has joined #zuul17:42
*** hashar has quit IRC17:44
openstackgerritMerged zuul/zuul master: Handle private IPv6 addresses supplied by Nodepool  https://review.opendev.org/c/zuul/zuul/+/78039917:48
*** jamesmcarthur has joined #zuul17:57
*** avass has quit IRC17:58
openstackgerritDaniel Blixt proposed zuul/zuul-jobs master: Make build-sshkey handling windows compatible  https://review.opendev.org/c/zuul/zuul-jobs/+/78066218:01
openstackgerritJames E. Blair proposed zuul/zuul master: Add ExecutorApi  https://review.opendev.org/c/zuul/zuul/+/77090218:01
openstackgerritJames E. Blair proposed zuul/zuul master: Change zone handling in ExecutorApi  https://review.opendev.org/c/zuul/zuul/+/78783318:01
openstackgerritJames E. Blair proposed zuul/zuul master: Clean up Executor API build request locking and add tests  https://review.opendev.org/c/zuul/zuul/+/78862418:01
openstackgerritJames E. Blair proposed zuul/zuul master: WIP Execute builds via ZooKeeper  https://review.opendev.org/c/zuul/zuul/+/78898818:01
*** jamesmcarthur has quit IRC18:13
*** Shrews has joined #zuul18:14
*** jamesmcarthur has joined #zuul18:15
*** jamesmcarthur has quit IRC18:17
openstackgerritJames E. Blair proposed zuul/zuul master: Retry yarn package installs  https://review.opendev.org/c/zuul/zuul/+/78940118:27
corvusclarkb, fungi: ^ re convo in #opendev18:28
*** jamesmcarthur has joined #zuul18:30
openstackgerritJeremy Stanley proposed zuul/zuul-jobs master: Add new Debian security mirror suite pattern  https://review.opendev.org/c/zuul/zuul-jobs/+/78940318:31
fungithanks!18:31
openstackgerritJeremy Stanley proposed zuul/zuul-jobs master: Test Debian Buster and Bullseye  https://review.opendev.org/c/zuul/zuul-jobs/+/78940418:31
openstackgerritJeremy Stanley proposed zuul/zuul-jobs master: Add new Debian security mirror suite pattern  https://review.opendev.org/c/zuul/zuul-jobs/+/78909018:40
openstackgerritJeremy Stanley proposed zuul/zuul-jobs master: Test Debian Buster and Bullseye  https://review.opendev.org/c/zuul/zuul-jobs/+/78940418:40
*** jfoufas1 has quit IRC18:40
*** goneri has joined #zuul18:43
*** jamesmcarthur has quit IRC18:49
clarkbcool looks like it is alredy approved18:51
*** jamesmcarthur has joined #zuul18:51
fungiokay, 789090 should be safe to merge, owing to the successful zuul-jobs-test-base-roles-debian-stretch result on it and the zuul-jobs-test-base-roles-debian-{buster,bullseye} successes on 789404 (as seen in zuul status, not yet reported on the latest patchsets)18:51
fungi789404 isn't going to pass some of its other jobs until 789090 merges though18:52
openstackgerritMerged zuul/zuul master: Add artifacts to MQTT reporter  https://review.opendev.org/c/zuul/zuul/+/78810118:52
clarkbfungi: looks like the bullseye jobs may still be sad on the big bullseye addition change19:02
clarkbI guess because we need to land the first change first since it is in a trusted playbook?19:02
fungiclarkb: yes, i tried to explain that in the commit messages and comments19:04
clarkbpart of the problem with me whenI look at those is things are just similar enough I never know if we're failing because we did the wrong old thing or if the new thing isn't working19:04
fungithe ones to pay attention to on 789404 for now are zuul-jobs-test-base-roles-debian-{buster,bullseye} which succeeded19:05
fungithe others will retry_limit until 789090 is merged, then we can recheck19:05
clarkbbecause the base-roles test swaps in the mirror updates premerge. Makes sense19:06
fungiyep, and this is also why we don't need to do the base-test dance for 789090, at least in theory19:07
clarkb++19:09
fungiwhether we'll find bullseye-specific problems with one of the other roles (e.g. nim or terraform or...) is yet to be determined19:11
fungibut the fact that they'19:13
*** vishalmanchanda has quit IRC19:13
fungire all still passing for buster is a good sign19:13
*** jamesmcarthur has quit IRC19:18
*** jamesmcarthur has joined #zuul19:44
*** jamesmcarthur has quit IRC19:52
*** jamesmcarthur has joined #zuul19:52
*** SpamapS has quit IRC19:53
*** fdegir has quit IRC19:54
*** icey has quit IRC19:54
*** SpamapS has joined #zuul19:55
*** fdegir has joined #zuul19:55
*** icey has joined #zuul19:55
*** holser has quit IRC19:56
*** jamesmcarthur has quit IRC20:13
*** jamesmcarthur has joined #zuul20:14
openstackgerritMatthieu Huin proposed zuul/zuul master: REST API: Implement nodes filtering  https://review.opendev.org/c/zuul/zuul/+/73604220:17
clarkbis anyone else able to review https://review.opendev.org/c/zuul/zuul-jobs/+/789090/4 ? that will help us move some of our debian work forward20:37
corvusclarkb, fungi: i feel like im' missing something there -- won't that do the wrong thing for ubuntu?20:41
corvusfocal is not in [jessie, stretch, buster], so won't it render as "deb mirror focal-security main"20:42
clarkbthats a good question and now taht you ask it I see that the child change doesn't run base-roles tests against ubuntu. That may indicate this file isn't used by ubuntu /me double checks20:44
fungicorvus: confusingly, roles/configure-mirrors/templates/apt/etc/apt/sources.list.d/security.list.j2 is only used in roles/configure-mirrors/tasks/mirror/Debian.yaml and not in roles/configure-mirrors/tasks/mirror/Ubuntu.yaml20:44
corvusfungi: thanks' that's the missing link20:44
clarkbyup just doule checked that20:44
corvushttps://zuul.opendev.org/t/zuul/build/1f5d527729a14307a0edc61efe4c426a/console#1/3/7/ubuntu-focal20:44
corvusand incidentally, it looks like "focal-security" actually *is* correct20:45
corvusjust due to that different template i guess20:45
fungiyeah, essentially debian, as of bullseye, is organizing their security mirror the way ubuntu has been20:46
corvusubuntu does this:  roles/configure-mirrors/templates/apt/etc/apt/sources.list.j2 has: deb {% if set_apt_mirrors_trusted %}[ trusted=yes ] {% endif %}{{ package_mirror }} {{ ansible_distribution_release }}-security main univers20:46
corvusokay.  this change makes sense to me and i agree it seems unlikely to break ubuntu.  we should, erm, comment and name those files better.  :)20:47
fungiright, that's basically what i ended up copying20:47
corvusclarkb, fungi: +2; +w at will20:48
fungiand yes, the passing test-base-role results for ubuntu platforms and stretch on that change, as well as for buster and bullseye on the child change, have me fairly convinced it's not affecting ubuntu20:48
*** Shrews has quit IRC20:49
clarkbfungi: do we need the same if set_apt_mirrors_trusted thing?20:50
clarkbit doesn't seem to be in the debian specific file20:50
clarkbI guess the file predates bullseye and this change and it didn't have it there so we should be fine20:58
fungiall that does is silence warnings about mirrors lacking signed indices20:58
clarkbgotcha so warnings not errors (and I see the warnings in the child change jobs)20:58
clarkbprobably best to keep those in place (the warnings)20:58
clarkbfungi: do you want to approve it or should I?21:00
fungii can approve it in a moment, sure21:00
*** jamesmcarthur has quit IRC21:02
fungionce it merges i'll recheck 78940421:03
openstackgerritMerged zuul/zuul master: Retry yarn package installs  https://review.opendev.org/c/zuul/zuul/+/78940121:04
openstackgerritMerged zuul/zuul-jobs master: Add new Debian security mirror suite pattern  https://review.opendev.org/c/zuul/zuul-jobs/+/78909021:13
clarkbI got the recheck21:14
clarkbone bullseye job that needed the prior merge has passed.21:17
clarkbthe bindep job21:18
clarkbso that is looking good21:18
openstackgerritMatthieu Huin proposed zuul/zuul-client master: Improve logging and CLI output  https://review.opendev.org/c/zuul/zuul-client/+/78834021:19
*** goneri has quit IRC21:28
openstackgerritMerged zuul/zuul master: Fix exception on build started event for noop jobs  https://review.opendev.org/c/zuul/zuul/+/78924521:32
clarkbhttps://review.opendev.org/c/zuul/zuul-jobs/+/789404 passes now so ya all looks good there21:34
fungiawesome, that was smoother than i expected21:37
*** jamesmcarthur has joined #zuul21:54
*** jamesmcarthur has quit IRC21:59
ianwanyone have particular thoughts on the timestamp of the zuul summary tab @ https://149.202.164.129/c/x/test-project/+/3 (have to accept self signed cert)22:01
ianwit's now uses the same timestamp formatting element as everything else, and i put it in the same colour22:01
mhuLGTM22:02
corvusianw: ++consistency22:03
openstackgerritMatthieu Huin proposed zuul/zuul-client master: Add output formatters when relevant  https://review.opendev.org/c/zuul/zuul-client/+/78883222:17
clarkbianw: https://review.opendev.org/c/zuul/zuul-jobs/+/789404 is the next step in bullseye updates22:31
clarkbthough at this point I suspect that jobs may just work on it? maybe?22:32
clarkbianw: I too like the new timestamps22:32
ianwwill take a look at bullsye things after school run22:38
*** tosky has quit IRC22:50
*** jamesmcarthur has joined #zuul23:07
*** jamesmcarthur has quit IRC23:13
*** jamesmcarthur has joined #zuul23:17
*** jamesmcarthur has quit IRC23:22
*** jamesmcarthur has joined #zuul23:49
*** jamesmcarthur has quit IRC23:54
« Sunday, 2021-05-02 Index Tuesday, 2021-05-04 »

Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!