| *** hashar has quit IRC | 00:05 | |
| *** jamesmcarthur has quit IRC | 00:16 | |
| *** jamesmcarthur has joined #zuul | 00:17 | |
| *** snapiri has quit IRC | 00:18 | |
| *** snapiri has joined #zuul | 00:18 | |
| *** fdegir5 has joined #zuul | 00:19 | |
| *** fdegir has quit IRC | 00:19 | |
| *** jamesmcarthur has quit IRC | 00:20 | |
| *** jamesmcarthur has joined #zuul | 00:20 | |
| *** tosky has quit IRC | 00:23 | |
| *** logan- has quit IRC | 00:38 | |
| *** logan- has joined #zuul | 00:50 | |
| *** mgagne has quit IRC | 01:11 | |
| *** ttx has quit IRC | 01:11 | |
| *** Eighth_Doctor has quit IRC | 01:11 | |
| *** mgagne has joined #zuul | 01:12 | |
| *** ttx has joined #zuul | 01:12 | |
| *** Eighth_Doctor has joined #zuul | 01:12 | |
| *** mnaser has quit IRC | 01:13 | |
| *** Eighth_Doctor has quit IRC | 01:14 | |
| *** masterpe has quit IRC | 01:15 | |
| *** mordred has quit IRC | 01:15 | |
| *** mnaser has joined #zuul | 01:16 | |
| openstackgerrit | Merged zuul/zuul-client master: encrypt: add --public-key argument https://review.opendev.org/c/zuul/zuul-client/+/765313 | 01:39 |
|---|---|---|
| *** mordred has joined #zuul | 01:44 | |
| *** mordred has quit IRC | 02:04 | |
| openstackgerrit | Merged zuul/zuul master: zuul-client: test encryption of large secrets https://review.opendev.org/c/zuul/zuul/+/766037 | 02:17 |
| *** sshnaidm|ruck is now known as sshnaidm|off | 02:25 | |
| *** Eighth_Doctor has joined #zuul | 02:28 | |
| *** mordred has joined #zuul | 02:48 | |
| *** johanssone has quit IRC | 03:24 | |
| *** lyr6 has joined #zuul | 03:31 | |
| *** dpawlik1 has joined #zuul | 03:32 | |
| *** ianw_ has joined #zuul | 03:32 | |
| *** freefood_ has joined #zuul | 03:33 | |
| *** tobiash_ has joined #zuul | 03:33 | |
| *** zbr has quit IRC | 03:38 | |
| *** lyr has quit IRC | 03:38 | |
| *** dpawlik has quit IRC | 03:38 | |
| *** arxcruz has quit IRC | 03:38 | |
| *** jpena|off has quit IRC | 03:38 | |
| *** ianw has quit IRC | 03:38 | |
| *** tobiash has quit IRC | 03:38 | |
| *** freefood has quit IRC | 03:38 | |
| *** ianw_ is now known as ianw | 03:38 | |
| *** lyr6 is now known as lyr | 03:39 | |
| *** jpena|off has joined #zuul | 03:39 | |
| *** mordred has quit IRC | 03:40 | |
| *** gouthamr has quit IRC | 03:43 | |
| *** zbr has joined #zuul | 03:43 | |
| *** arxcruz has joined #zuul | 03:43 | |
| *** mordred has joined #zuul | 03:44 | |
| *** ykarel has joined #zuul | 04:04 | |
| *** bhavikdbavishi has joined #zuul | 04:21 | |
| *** jamesmcarthur has quit IRC | 04:22 | |
| *** jamesmcarthur has joined #zuul | 04:24 | |
| *** bhavikdbavishi1 has joined #zuul | 04:26 | |
| *** bhavikdbavishi has quit IRC | 04:27 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 04:27 | |
| *** ykarel_ has joined #zuul | 04:28 | |
| *** ikhan has quit IRC | 04:30 | |
| *** ykarel has quit IRC | 04:30 | |
| *** ykarel_ is now known as ykarel | 04:42 | |
| *** wuchunyang has joined #zuul | 04:48 | |
| *** wuchunyang has quit IRC | 04:52 | |
| *** vishalmanchanda has joined #zuul | 05:03 | |
| *** bhavikdbavishi has quit IRC | 05:15 | |
| *** jfoufas1 has joined #zuul | 05:26 | |
| *** wuchunyang has joined #zuul | 05:31 | |
| *** evrardjp has quit IRC | 05:33 | |
| *** evrardjp has joined #zuul | 05:33 | |
| *** ykarel_ has joined #zuul | 05:50 | |
| *** ykarel has quit IRC | 05:53 | |
| *** ianw has quit IRC | 05:59 | |
| *** ykarel_ is now known as ykarel | 05:59 | |
| *** ianw has joined #zuul | 06:01 | |
| *** gouthamr has joined #zuul | 06:22 | |
| *** jamesmcarthur has quit IRC | 06:26 | |
| *** dpawlik1 is now known as dpawlik | 07:16 | |
| *** jcapitao has joined #zuul | 07:16 | |
| *** gouthamr has joined #zuul | 07:22 | |
| *** openstackgerrit has quit IRC | 07:23 | |
| *** saneax has joined #zuul | 07:50 | |
| *** hashar has joined #zuul | 07:56 | |
| *** johanssone has joined #zuul | 07:59 | |
| *** ikhan has joined #zuul | 08:03 | |
| *** jamesmcarthur has joined #zuul | 08:27 | |
| *** jamesmcarthur has quit IRC | 08:31 | |
| *** tosky has joined #zuul | 08:41 | |
| *** rpittau|afk is now known as rpittau | 08:52 | |
| *** jpena|off is now known as jpena | 08:57 | |
| *** openstackgerrit has joined #zuul | 09:14 | |
| openstackgerrit | Merged zuul/zuul master: tools: Deprecate encrypt_secret.py, document zuul-client encrypt https://review.opendev.org/c/zuul/zuul/+/765316 | 09:14 |
| mhu | corvus: thanks, appreciated! | 09:17 |
| *** nils has joined #zuul | 09:56 | |
| avass | so the problem with nodepool requiring a single pool to fulfill a node requests happened again | 10:20 |
| avass | if for example you'd want to mix static nodes and dynamic nodes | 10:20 |
| avass | but what if there was a way in nodepool to configure a group of pools that are allowed to fullfill nodes together? | 10:21 |
| tobiash_ | avass: I think there was already brainstorming about that topic in the past | 10:25 |
| avass | oh :) | 10:26 |
| tobiash_ | avass: I think what came out was that this could be solved by adding a nodepool driver to nodepool that creates node requests for all labels in the background | 10:26 |
| tobiash_ | somewhat inception like | 10:26 |
| tobiash_ | like a node request proxy that splits up a node request | 10:27 |
| avass | I also think someone mentioned that being able to match on multiple labels would be nice | 10:28 |
| avass | I mean AND labels together so a node requires two labels | 10:29 |
| *** jfoufas1 has quit IRC | 10:34 | |
| *** wuchunyang has quit IRC | 10:43 | |
| *** sshnaidm|off has quit IRC | 11:09 | |
| *** sshnaidm|off has joined #zuul | 11:10 | |
| *** jcapitao is now known as jcapitao_lunch | 11:19 | |
| *** saneax has quit IRC | 11:35 | |
| *** jpena is now known as jpena|lunch | 12:33 | |
| *** jcapitao_lunch is now known as jcapitao | 13:00 | |
| *** mnaser has quit IRC | 13:00 | |
| *** ericsysmin has quit IRC | 13:00 | |
| *** jkt has quit IRC | 13:00 | |
| *** mnaser has joined #zuul | 13:01 | |
| *** ericsysmin has joined #zuul | 13:01 | |
| *** jkt has joined #zuul | 13:01 | |
| *** mordred has quit IRC | 13:03 | |
| *** johanssone has quit IRC | 13:04 | |
| *** Eighth_Doctor has quit IRC | 13:04 | |
| *** johanssone has joined #zuul | 13:05 | |
| *** jfoufas1 has joined #zuul | 13:07 | |
| *** saneax has joined #zuul | 13:28 | |
| *** jpena|lunch is now known as jpena | 13:30 | |
| *** mordred has joined #zuul | 13:31 | |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Fix test_gerrit.TestPolling.test_config_update https://review.opendev.org/c/zuul/zuul/+/773023 | 13:35 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Move serialization helper methods to ZooKeeperBase class https://review.opendev.org/c/zuul/zuul/+/773024 | 13:35 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Implement ZooKeeper backed merge jobs https://review.opendev.org/c/zuul/zuul/+/773025 | 13:35 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Switch to ZooKeeper backed merge jobs https://review.opendev.org/c/zuul/zuul/+/773026 | 13:35 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Collect statsd information from ZooKeeper rather than gearman https://review.opendev.org/c/zuul/zuul/+/773027 | 13:35 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Remove remaining gearman parts from merger https://review.opendev.org/c/zuul/zuul/+/773028 | 13:35 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Provide zk_client to merger client rather than the whole scheduler https://review.opendev.org/c/zuul/zuul/+/773029 | 13:35 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Remove history from RecordingMergeClient https://review.opendev.org/c/zuul/zuul/+/773030 | 13:35 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: WIP Align ZooKeeper builds and merger API https://review.opendev.org/c/zuul/zuul/+/773031 | 13:36 |
| *** mordred has quit IRC | 13:48 | |
| *** masterpe has joined #zuul | 14:13 | |
| *** hashar is now known as hasharAway | 14:24 | |
| *** sshnaidm|off has quit IRC | 14:27 | |
| *** sshnaidm|off has joined #zuul | 14:27 | |
| *** Eighth_Doctor has joined #zuul | 14:35 | |
| *** mordred has joined #zuul | 14:35 | |
| *** jamesmcarthur has joined #zuul | 15:01 | |
| *** ykarel has quit IRC | 15:11 | |
| *** ykarel has joined #zuul | 15:12 | |
| *** rpittau is now known as rpittau|afk | 15:27 | |
| openstackgerrit | Matthieu Huin proposed zuul/zuul-client master: Reorganize the installation documentation, specify versions https://review.opendev.org/c/zuul/zuul-client/+/767592 | 15:28 |
| *** jfoufas1 has quit IRC | 15:48 | |
| *** sshnaidm|off has quit IRC | 16:09 | |
| *** ykarel has quit IRC | 16:14 | |
| *** sshnaidm|off has joined #zuul | 16:14 | |
| *** jamesmcarthur has quit IRC | 16:42 | |
| *** sanjayu_ has joined #zuul | 16:45 | |
| *** saneax has quit IRC | 16:46 | |
| *** hasharAway is now known as hashar | 16:48 | |
| clarkb | avass: tobiash_ what if zuul decoupled them based on attributes | 16:57 |
| clarkb | iirc the grouping is based on what zuul has decided then nodepool just fulfills that | 16:57 |
| *** tobiash_ is now known as tobiash | 16:57 | |
| clarkb | I think the current default of grouping is likely safer for most job situations, but maybe you can annotate the dependency in config to say "we don't care where these run relative to each other" | 16:57 |
| clarkb | of coures taht requires user intervention and maybe if we did it on the nodepool side we could infer it somehow from the provider and pool configs | 16:58 |
| clarkb | eg "this is not satisfiable as requested so we'll do our best instead" | 16:58 |
| fungi | yeah, it was the easiest way to ensure nodes booted in the same provider in case they needed direct layer 2 access between one another for a build. that may not always be necessary (it was for some jobs in opendev), and there may also be other ways of accomplishing it | 17:04 |
| avass | clarkb: I think (at least in some cases) they are relative to eachother | 17:06 |
| clarkb | avass: gotcha so thee is a third situation where we want to say provider X and Y should be grouped | 17:07 |
| avass | yep | 17:08 |
| *** jamesmcarthur has joined #zuul | 17:08 | |
| avass | at the moments we'd like to move those to be dynamic nodes from the same provider but because of license issues it's not always super simple to do that | 17:09 |
| fungi | yeah, license issues are tough | 17:10 |
| fungi | that's the main reason opendev doesn't offer rhel nodes | 17:10 |
| *** jamesmcarthur has quit IRC | 17:10 | |
| *** jamesmcarthur has joined #zuul | 17:11 | |
| Shrews | that's going to be a "fun" exercise when it comes to dealing with quotas among multiple providers. have fun, whomever gets tasked with implementing it! | 17:12 |
| * Shrews LOLs with an evil grin | 17:12 | |
| fungi | indeed, probably why a meta-provider which issues its own node requests to the actual providers might be the easiest approach | 17:13 |
| fungi | then they can decline based on actual quota info | 17:14 |
| avass | Shrews: :) | 17:16 |
| *** jamesmcarthur has quit IRC | 17:17 | |
| *** jamesmcarthur has joined #zuul | 17:18 | |
| fungi | trying to synthesize a virtual quota from that would indeed be a nightmare | 17:18 |
| Shrews | the architecture is not setup for easily introducing multi-provider requests. i think there will be more work there than you expect. | 17:21 |
| *** sanjayu_ has quit IRC | 17:23 | |
| fungi | almost certainly | 17:25 |
| fungi | probably it would involve hoisting a chunk of zuul's node request logic into the fake provider too | 17:25 |
| *** vishalmanchanda has quit IRC | 17:32 | |
| *** jpena is now known as jpena|off | 17:57 | |
| *** jamesmcarthur has quit IRC | 18:10 | |
| *** jcapitao has quit IRC | 18:23 | |
| *** sduthil has quit IRC | 18:24 | |
| *** sduthil has joined #zuul | 18:25 | |
| *** jamesmcarthur has joined #zuul | 18:40 | |
| *** jamesmcarthur has quit IRC | 18:45 | |
| *** hashar is now known as hasharDinner | 18:54 | |
| *** jamesmcarthur has joined #zuul | 19:02 | |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Format multi-line log entries https://review.opendev.org/c/zuul/zuul/+/772602 | 19:06 |
| corvus | guillaumec, tristanC, ianw, avass: i think i really like that version ^. i think it threads the needle on making it mostly work for most people (i don't think it makes things worse for anyone), and i think we retain 100% of the flexibility of the python logging system. | 19:09 |
| corvus | ianw: the implementation is just a little different than yours (it lets the user change the format specifier), but the output should be the same. | 19:11 |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Format multi-line log entries https://review.opendev.org/c/zuul/zuul/+/772602 | 19:11 |
| corvus | pep8 fix ^ | 19:11 |
| corvus | guillaumec, tristanC, ianw, avass: if we all like that one, let's do that and update the nodepool change to match | 19:12 |
| *** jamesmcarthur has quit IRC | 19:12 | |
| avass | corvus: ++, we've had to update the splunk handler a bit to suit our needs but we'll try to publish that somehow | 19:13 |
| corvus | avass: cool, depending on the complexity, maybe could be a doc contribution or something in tools/ | 19:14 |
| *** jamesmcarthur has joined #zuul | 19:27 | |
| *** jamesmcarthur has quit IRC | 19:31 | |
| *** jamesmcarthur has joined #zuul | 19:33 | |
| *** jamesmcarthur has quit IRC | 19:39 | |
| *** nils has quit IRC | 19:43 | |
| *** holser has quit IRC | 19:50 | |
| *** holser has joined #zuul | 19:51 | |
| tobiash | corvus: so I can opt out to keep the current format? Our elk stack keeps those currently in a single message which is kinda neat | 20:28 |
| fungi | there are ways with grok filters to fold multi-line messages into a single record, i think? | 20:29 |
| fungi | i know we do that with some ci logs in opendev anyway | 20:30 |
| clarkb | there are, but its a pita | 20:30 |
| fungi | yeah, it's nontrivial for sure | 20:31 |
| tobiash | fungi: yes our logstash does that | 20:31 |
| corvus | tobiash: how do you accomplish that? | 20:32 |
| tobiash | corvus: if I'm reading this right, systems with custom log configs don't have to change anything since this just changes the default? | 20:33 |
| corvus | tobiash: correct | 20:33 |
| tobiash | corvus: someone who knows logstash better than me configured multiline parsing | 20:33 |
| tobiash | or in filebeat, not sure | 20:35 |
| corvus | tobiash: neat; i agree that its important with elk/splunk to have all multiline entries as a single entry; i figured the "best" way to do that would be with a native logger (eg https://pypi.org/project/splunk-handler/ or https://pypi.org/project/python-logstash/ ) | 20:35 |
| corvus | but if you've figured out how to do it with parsing, great :) i expect that would be challenging based on the ambiguity (especially with interleaved lines) | 20:36 |
| tobiash | our chain is filebeat -> logstash -> elasticsearch | 20:37 |
| tobiash | I just know the config is in either filebeat or logstash with no other tool involved | 20:37 |
| corvus | but if it's working and you don't want to touch it, and you are using a custom logging config, then yeah, i think we can land that change and you won't notice. if it happens to make your parsing easier, you could opt into it in your custom config | 20:37 |
| tobiash | and so far it worked great and I also didn't notice interleaved messagea so far | 20:37 |
| corvus | incidentally, i think if someone used the multiline change, and updated the format to include the thread id, they could reliably parse out multiline messages from an interleaved stream automatically. | 20:39 |
| tobiash | actually I think the stack traces are not interleaved. We don't see that in our system | 20:41 |
| avass | tobiash: we got interleaved stacktraces :) | 20:41 |
| tobiash | I guess I just didn't notice that then :) | 20:41 |
| *** hasharDinner is now known as hashar | 21:50 | |
| *** holser has quit IRC | 22:03 | |
| *** holser has joined #zuul | 22:04 | |
| *** hashar has quit IRC | 23:13 | |
| *** cloudnull has quit IRC | 23:52 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!