| openstackgerrit | James E. Blair proposed zuul/nodepool master: Azure: add quota support https://review.opendev.org/c/zuul/nodepool/+/780439 | 00:20 |
|---|---|---|
| openstackgerrit | James E. Blair proposed zuul/nodepool master: Azure: implement launch retries https://review.opendev.org/c/zuul/nodepool/+/780682 | 00:20 |
| *** tosky has quit IRC | 00:35 | |
| *** ajitha has quit IRC | 00:54 | |
| *** ianychoi_ has joined #zuul | 01:03 | |
| *** dpawlik6 has joined #zuul | 01:04 | |
| *** jpenag has joined #zuul | 01:10 | |
| *** ianychoi has quit IRC | 01:12 | |
| *** SotK has quit IRC | 01:12 | |
| *** dpawlik has quit IRC | 01:12 | |
| *** jpena|off has quit IRC | 01:13 | |
| *** EmilienM has quit IRC | 01:13 | |
| *** arxcruz has quit IRC | 01:13 | |
| *** kgz has quit IRC | 01:13 | |
| *** dpawlik6 is now known as dpawlik | 01:13 | |
| *** SotK has joined #zuul | 01:13 | |
| *** kgz has joined #zuul | 01:23 | |
| *** hamalq has quit IRC | 01:23 | |
| *** ajitha has joined #zuul | 01:46 | |
| *** evrardjp_ has quit IRC | 03:33 | |
| *** evrardjp has joined #zuul | 03:33 | |
| *** bhavikdbavishi has joined #zuul | 03:34 | |
| *** saneax has joined #zuul | 03:36 | |
| *** bhavikdbavishi1 has joined #zuul | 03:37 | |
| *** bhavikdbavishi has quit IRC | 03:39 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 03:39 | |
| *** kgz has quit IRC | 03:51 | |
| *** kgz has joined #zuul | 04:21 | |
| *** ykarel|away has joined #zuul | 04:32 | |
| *** vishalmanchanda has joined #zuul | 04:33 | |
| *** ykarel|away is now known as ykarel | 04:37 | |
| *** jfoufas1 has joined #zuul | 05:27 | |
| *** bhavikdbavishi has quit IRC | 05:46 | |
| *** bhavikdbavishi has joined #zuul | 05:47 | |
| *** zbr has quit IRC | 06:13 | |
| *** zbr has joined #zuul | 06:15 | |
| *** icey_ has quit IRC | 07:00 | |
| *** icey has joined #zuul | 07:06 | |
| *** rlandy|bbl has quit IRC | 07:12 | |
| *** bhavikdbavishi has quit IRC | 07:21 | |
| *** bhavikdbavishi has joined #zuul | 07:22 | |
| *** jcapitao has joined #zuul | 07:37 | |
| *** rpittau|afk is now known as rpittau | 07:58 | |
| *** hashar has joined #zuul | 08:05 | |
| *** bhavikdbavishi has quit IRC | 08:07 | |
| openstackgerrit | Daniel Blixt proposed zuul/zuul-jobs master: WIP: Make build-sshkey handling windows compatible https://review.opendev.org/c/zuul/zuul-jobs/+/780662 | 08:19 |
| *** ykarel is now known as ykarel|lunch | 08:33 | |
| *** bhavikdbavishi has joined #zuul | 08:54 | |
| *** piotrowskim has joined #zuul | 08:56 | |
| *** bhavikdbavishi1 has joined #zuul | 08:57 | |
| *** bhavikdbavishi has quit IRC | 08:58 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 08:58 | |
| *** tosky has joined #zuul | 09:00 | |
| openstackgerrit | Andy Ladjadj proposed zuul/zuul master: [web][config] move timezone component to preferences https://review.opendev.org/c/zuul/zuul/+/755929 | 09:17 |
| openstackgerrit | Albin Vass proposed zuul/zuul master: Filter events on event connection https://review.opendev.org/c/zuul/zuul/+/760907 | 09:19 |
| openstackgerrit | Simon Westphahl proposed zuul/zuul master: Implementation of Zookeeper backed event queues https://review.opendev.org/c/zuul/zuul/+/761170 | 09:26 |
| openstackgerrit | Simon Westphahl proposed zuul/zuul master: Implementation of Zookeeper event watcher https://review.opendev.org/c/zuul/zuul/+/761171 | 09:26 |
| openstackgerrit | Simon Westphahl proposed zuul/zuul master: Switch to Zookeeper backed trigger event queues https://review.opendev.org/c/zuul/zuul/+/763754 | 09:27 |
| openstackgerrit | Simon Westphahl proposed zuul/zuul master: Switch to Zookeeper backed management event queues https://review.opendev.org/c/zuul/zuul/+/761738 | 09:27 |
| *** ykarel|lunch is now known as ykarel | 09:33 | |
| *** jpenag is now known as jpena | 09:33 | |
| openstackgerrit | Albin Vass proposed zuul/zuul master: Filter events on event connection https://review.opendev.org/c/zuul/zuul/+/760907 | 09:36 |
| openstackgerrit | Albin Vass proposed zuul/zuul master: Filter events on event connection https://review.opendev.org/c/zuul/zuul/+/760907 | 09:45 |
| *** sshnaidm has quit IRC | 09:50 | |
| openstackgerrit | Simon Westphahl proposed zuul/zuul master: Implementation of Zookeeper event watcher https://review.opendev.org/c/zuul/zuul/+/761171 | 10:08 |
| openstackgerrit | Simon Westphahl proposed zuul/zuul master: Switch to Zookeeper backed trigger event queues https://review.opendev.org/c/zuul/zuul/+/763754 | 10:08 |
| openstackgerrit | Simon Westphahl proposed zuul/zuul master: Switch to Zookeeper backed management event queues https://review.opendev.org/c/zuul/zuul/+/761738 | 10:08 |
| *** arxcruz has joined #zuul | 10:18 | |
| *** sshnaidm has joined #zuul | 10:27 | |
| openstackgerrit | Clément Mondion proposed zuul/zuul master: [api][cors] Access-Control-Allow-Origin * for all routes https://review.opendev.org/c/zuul/zuul/+/767691 | 10:37 |
| *** jangutter_ has quit IRC | 10:42 | |
| *** jangutter has joined #zuul | 10:43 | |
| openstackgerrit | Andy Ladjadj proposed zuul/zuul master: [doc] fix typo matter -> master https://review.opendev.org/c/zuul/zuul/+/780783 | 11:00 |
| openstackgerrit | Daniel Blixt proposed zuul/zuul-jobs master: WIP: Make build-sshkey handling windows compatible https://review.opendev.org/c/zuul/zuul-jobs/+/780662 | 11:07 |
| *** jangutter_ has joined #zuul | 11:08 | |
| *** jangutter has quit IRC | 11:11 | |
| openstackgerrit | Albin Vass proposed zuul/zuul master: Filter events on event connection https://review.opendev.org/c/zuul/zuul/+/760907 | 11:33 |
| *** rlandy has joined #zuul | 11:37 | |
| openstackgerrit | Albin Vass proposed zuul/zuul master: Filter events on event connection https://review.opendev.org/c/zuul/zuul/+/760907 | 11:40 |
| *** jcapitao is now known as jcapitao_lunch | 11:57 | |
| *** pleia2 has joined #zuul | 12:02 | |
| *** ykarel is now known as ykarel|afk | 12:10 | |
| *** bhavikdbavishi has quit IRC | 12:19 | |
| *** EmilienM has joined #zuul | 12:24 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: gerrit: ignore replication event https://review.opendev.org/c/zuul/zuul/+/780809 | 12:30 |
| *** bhavikdbavishi has joined #zuul | 12:33 | |
| *** jpena is now known as jpena|lunch | 12:39 | |
| tristanC | after upgrading to zuul v4 and gerrit 3.2, our scheduler logs are flooded (>6GB overnight) with gerrit replication events, it seems like we'll have to apply https://review.opendev.org/c/zuul/zuul/+/780809 | 12:40 |
| *** jangutter_ has quit IRC | 12:53 | |
| *** jcapitao_lunch is now known as jcapitao | 12:58 | |
| *** jangutter has joined #zuul | 12:59 | |
| tobiash | tristanC: lgtm | 13:04 |
| *** GomathiselviS has joined #zuul | 13:05 | |
| tobiash | zuul-maint: similarly a review on https://review.opendev.org/c/zuul/zuul/+/767162 would be appreciated which ensures that zuul doesn't fail if a git-needed commit cannot be accessed (e.g. draft that is not accessible by zuul) | 13:05 |
| GomathiselviS | corvus: Hi, need review and merge of https://review.opendev.org/c/zuul/zuul-jobs/+/773474. Other local PRs are dependent on this change. | 13:06 |
| corvus | tristanC: do you know what about zuulv4 caused that change in behavior? | 13:25 |
| tristanC | corvus: i'm not sure, perhaps since the gerrit upgrade the replication events are part of the ssh event stream? | 13:27 |
| Open10K8S | corvus: I emailed the announce last friday regarding zuul proxy. | 13:28 |
| corvus | tristanC: that's been true for quite a while | 13:28 |
| tristanC | we do get those for opendev replication too | 13:28 |
| corvus | tristanC: right; the behavior you describe has been true for opendev for maybe a year or more | 13:30 |
| corvus | it's just your commit message suggested a zuul v4 behavior change and i can't figure out what it is | 13:30 |
| corvus | the last time we dealt with ref-replicated events was februrary 2017, where we squelched warnings. there's an outstanding change of https://review.opendev.org/135332 which actually would improve handling of them if someone is triggering on them | 13:33 |
| corvus | i'm hesitant to remove a trigger event that has been supported for 3 years | 13:35 |
| corvus | tristanC: maybe it's worth a zuul-discuss message before we remove it? | 13:37 |
| mordred | tristanC: did you _just_ upgrade? | 13:37 |
| mordred | like - I wonder if something about the upgrade itself caused a re-replication of all of the refs, thus the flood of events? | 13:37 |
| mordred | so - it might not be a steady-state problem, but one that exists due to the upgrade | 13:38 |
| corvus | oh yeah good point | 13:38 |
| tristanC | mordred: it has been 23 hours since the upgrade, and we still get loads of them | 13:38 |
| corvus | i mean, there are *a lot* of replication events in the normal course, but if you re-replicated everything there would obviously be a lot more | 13:39 |
| corvus | we could also look into doing something where the gerrit driver checked to see if certain trigger events were used and ignored the unused ones. | 13:40 |
| mordred | this might be too complex - but what if it were possible to log ... | 13:40 |
| mordred | yup | 13:40 |
| mordred | that | 13:40 |
| mordred | and - you could also log the ignore events to a specific logger, so that someone could make an ignored_events log if they wanted/needed | 13:41 |
| corvus | tristanC: for reference -- opendev received 1698794 replication log lines totalling 574M yesterday | 13:41 |
| *** jpena|lunch is now known as jpena | 13:42 | |
| mordred | corvus: I like the idea of ignoring unused trigger events - seems like it would improve signal-to-noise in the log file generally without removing functionality | 13:42 |
| corvus | that's including (in fact, mostly) pipelines logging that they are ignoring them | 13:42 |
| corvus | mordred: yeah, though, to be fair, it's debug level | 13:42 |
| mordred | yah | 13:42 |
| tristanC | corvus: we do have more than 70 pipelines | 13:44 |
| tristanC | corvus: mordred: ignoring unused event sounds great, and perhaps improving the queue processor logging would help too, but on the other hand, what is the use-case for supporing replication event in zuul pipelines? | 13:46 |
| *** bhavikdbavishi has quit IRC | 13:46 | |
| mordred | tristanC: someone could want to wait to run their docs publishing until the accompaning git refs have been replicated. or if you're running a CD pipeline maybe it runs from the mirrors and you want to not trigger it until the refs are there? | 13:47 |
| mordred | or maybe you're triggering a downstream system that needs to have received the refs | 13:48 |
| corvus | yeah. and maybe no one is using them (or ever will in their current state) and we could remove them. i'd be okay with that if we send out a query on zuul-discuss first. i just don't want to yank out a feature that's been there for 3 years without discussion. | 13:50 |
| mordred | I'm not sure how important those are- but - corvus makes a very good point that those have been supported refs for a while, and we do keep discovering people running zuuls who we didn't know about already | 13:50 |
| mordred | yeah - that | 13:51 |
| mordred | and - in the period it would take to do the zuul-discuss query, the other logging improvement could probably be made | 13:51 |
| corvus | about half of our pipelines were created based on gerrit events that we supported with no idea how they would be used (post, experimental, promote, all fall into that category) | 13:51 |
| tristanC | mordred: the event doesn't seem reliable for that, it seems like a ref-replicated event is emited per replication destination | 13:52 |
| corvus | there's a completed event | 13:52 |
| *** yoctozepto has quit IRC | 13:53 | |
| *** Shrews has quit IRC | 13:53 | |
| mordred | corvus: we could use ref-replication-done to drive an ro AFS git farm - have gerrit replicate to an AFS dir, have ref-replication-done trigger a vos release job :) | 13:54 |
| corvus | yep | 13:55 |
| mordred | could even be volume-per-git-repo | 13:55 |
| *** ykarel|afk is now known as ykarel | 13:58 | |
| mordred | then replace the in-node repo cache with an afs client - what could possibly go wrong? | 13:58 |
| *** Shrews has joined #zuul | 13:59 | |
| openstackgerrit | Daniel Blixt proposed zuul/zuul-jobs master: WIP: Make build-sshkey handling windows compatible https://review.opendev.org/c/zuul/zuul-jobs/+/780662 | 14:02 |
| avass | I'd appreciate some reviews on: https://review.opendev.org/c/zuul/zuul/+/760907 if anyone got time. it's a bit of a pain right now | 14:04 |
| corvus | tristanC: under the assumption that no one is going to object and we do merge the change to ignore those events; do you want to add a release note? | 14:17 |
| corvus | avass: we should try to get that merged soon, before the zk event queue changes | 14:17 |
| avass | corvus: that would be great | 14:18 |
| *** jfoufas1 has quit IRC | 15:51 | |
| *** jfoufas1 has joined #zuul | 15:56 | |
| *** ykarel is now known as ykarel|away | 15:56 | |
| *** vishalmanchanda has quit IRC | 16:03 | |
| *** jfoufas1 has quit IRC | 16:05 | |
| *** ykarel|away has quit IRC | 16:29 | |
| *** hamalq has joined #zuul | 16:48 | |
| *** hashar has quit IRC | 16:53 | |
| *** jcapitao has quit IRC | 17:05 | |
| *** rpittau is now known as rpittau|afk | 17:29 | |
| clarkb | thinking maybe way outside the box here: would it be possible to set up gerrit (and other driver?) logging such that zuul only logged the events it knows it will take action on? If the config doesn't do anything with the events we can avoid logging them and keep log verbosity down? | 17:33 |
| clarkb | but if the config does use them then log them? | 17:33 |
| corvus | clarkb: that is absolutely what we should do for info level logs. but for *debug* level logs, i think every event that could cause an action should be logged. | 17:44 |
| corvus | iow, info should include what zuul did and why; debug should include all the input zuul used to make those decisions, and also things it did not do. | 17:45 |
| *** GomathiselviS has quit IRC | 17:48 | |
| *** vishalmanchanda has joined #zuul | 17:49 | |
| *** jpena is now known as jpena|off | 18:04 | |
| *** piotrowskim has quit IRC | 18:45 | |
| *** persia has quit IRC | 19:46 | |
| *** persia has joined #zuul | 19:47 | |
| *** persia has quit IRC | 19:52 | |
| *** persia has joined #zuul | 19:54 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: runner: add playbook selection filtering https://review.opendev.org/c/zuul/zuul/+/780964 | 19:56 |
| *** persia has quit IRC | 20:01 | |
| *** persia has joined #zuul | 20:02 | |
| *** tosky has quit IRC | 20:05 | |
| *** tosky has joined #zuul | 20:05 | |
| *** vishalmanchanda has quit IRC | 20:14 | |
| *** yoctozepto has joined #zuul | 20:14 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: runner: add job parameters listing https://review.opendev.org/c/zuul/zuul/+/644795 | 20:46 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: web: add depends-on support to the freeze job API https://review.opendev.org/c/zuul/zuul/+/639022 | 20:46 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: runner: add support for depends-on https://review.opendev.org/c/zuul/zuul/+/632064 | 20:46 |
| tristanC | corvus: https://review.opendev.org/c/zuul/zuul/+/632064 is the tip of a stack that should demonstrate the freeze_job api implements the zuul-runner requirements. | 20:50 |
| *** ajitha has quit IRC | 21:22 | |
| *** pleia2 has quit IRC | 22:35 | |
| *** sshnaidm has quit IRC | 22:35 | |
| *** zbr has quit IRC | 22:37 | |
| *** sshnaidm has joined #zuul | 22:41 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!