| *** threestrands has joined #zuul | 00:04 | |
| *** rlandy is now known as rlandy|afk | 00:24 | |
| *** jhesketh has joined #zuul | 00:41 | |
| *** mattw4 has quit IRC | 00:43 | |
| fdegir | corvus: what you put there, exchange standardized messages over a standardized message bus, is one of the key things - at least to me personally | 01:13 |
|---|---|---|
| fdegir | corvus: nearly all projects support some type of messaging but there is no standardization | 01:14 |
| fdegir | corvus: btw, this sig is the thing i mentioned in Shanghai and like fungi, i’m also wondering where it will go | 01:15 |
| fungi | yeah, it sounds like a promising solution, but the protocol part is easy. the schema is where things get dicey (because having looked at a lot of ci systems, the overlap in contexts seems limited) | 01:15 |
| fdegir | this is the third attempt to do such collaboration... | 01:16 |
| clarkb | ya I've been lookingat the state of the world recently and the number of systems that only care about a single code review system is a lot higher than I expected | 01:16 |
| clarkb | and if you don't care about a diverse set of inputs seems like a hard sell to standardize that schema | 01:16 |
| fungi | or only care about a single execution environment (often kubernetes) | 01:16 |
| fungi | also multi-tenancy seems almost entirely the domain of commercial hosted services | 01:17 |
| fdegir | fungi: that’s what i attempted to say during the meeting | 01:17 |
| clarkb | but also I'm not entirely sure you get a ton of benefit standardizing from the CI system side tehre. You want to psuh those changes on the code review systems | 01:17 |
| fdegir | not everyone is on k8s | 01:17 |
| fdegir | clarkb: i think this is more than ci systems | 01:18 |
| fungi | yes, to some degree the ci system's context is dependent on the code review workflow or similar trigger events driving it | 01:18 |
| fdegir | scm, artifact repos, they all are in play | 01:18 |
| *** jamesmcarthur has joined #zuul | 01:18 | |
| clarkb | fdegir: I know. I was using that particular interaction as an example. And I guess I see this as trying to bail the ocean | 01:19 |
| fdegir | your ci may well depend on an artifact that gets uploaded to artifactory, nexus, etc | 01:19 |
| fdegir | yes | 01:19 |
| clarkb | because github and gerrit will never agree | 01:19 |
| clarkb | similar with $clouds or $artifact storage | 01:19 |
| fdegir | but we can at least try to find minimum common whatever | 01:20 |
| fdegir | could be commit itself or merge event | 01:20 |
| fungi | i find the idea that we could have some federating protocol compelling, but i think first order is to classify the problems and explain them. some may be intractable and that's okay, but it's at least helpful to be able to say why that's the case | 01:21 |
| fdegir | i put gerrit, github, gitlab terms to the doc as well | 01:21 |
| fdegir | if i had energy, i was going to do the same for artifact repos but so tired after browsing docs of 8 projects | 01:22 |
| fungi | if nothing else, providing a clear description of the challenges which make solving certain problems hard or impossible at least gives others who are encountering them a way to catch up and either refute the logic presented or at least not waste time coming to the same conclusions independently | 01:22 |
| fdegir | ++ | 01:23 |
| clarkb | ya, it doesn't hurt to define those things. I guess where I get lost is in what the goal is. If the goal is to have all code review systems talk to CI systems using a standard protocol I'm not sure that is viable. Nor would it be viable to have all storage systems use the same api | 01:24 |
| clarkb | if the goal is instead to articulate the differences and chip away at them then that seems more doable | 01:24 |
| fungi | i'm not sure my goals are even that ambitious. i'd be happy if we just reached some consensus on what problems people have been trying to solve, and whether they've succeeded, and if so what commonalities there are between their tactics | 01:26 |
| fungi | knowing that would inform subsequent direction | 01:26 |
| clarkb | that makes sense | 01:26 |
| fungi | because i keep hearing different people assert that this or that class of problem is hard and complicated but i've never seen the problem space mapped out | 01:27 |
| fdegir | it’s great fosdem is this weekend | 01:29 |
| fdegir | we can chat about this stuff on saturday | 01:29 |
| fungi | indeed | 01:30 |
| fungi | see you there! | 01:30 |
| *** jamesmcarthur has quit IRC | 01:43 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-jobs master: WIP: add CONTRIBUTING guidelines https://review.opendev.org/704484 | 02:02 |
| *** jamesmcarthur has joined #zuul | 02:07 | |
| *** jamesmcarthur has quit IRC | 02:19 | |
| *** jamesmcarthur has joined #zuul | 02:44 | |
| *** jamesmcarthur has quit IRC | 02:45 | |
| *** jamesmcarthur has joined #zuul | 02:50 | |
| *** jamesmcarthur_ has joined #zuul | 02:54 | |
| *** jamesmcarthur has quit IRC | 02:54 | |
| *** jamesmcarthur_ has quit IRC | 03:03 | |
| *** jamesmcarthur has joined #zuul | 03:03 | |
| *** bhavikdbavishi has joined #zuul | 03:05 | |
| *** jamesmcarthur has quit IRC | 03:08 | |
| *** bhavikdbavishi1 has joined #zuul | 03:10 | |
| *** bhavikdbavishi has quit IRC | 03:12 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 03:12 | |
| *** jamesmcarthur has joined #zuul | 03:34 | |
| *** jamesmcarthur has quit IRC | 03:41 | |
| *** zxiiro has quit IRC | 03:41 | |
| *** bhavikdbavishi has quit IRC | 03:48 | |
| *** bhavikdbavishi has joined #zuul | 03:49 | |
| *** jamesmcarthur has joined #zuul | 04:37 | |
| *** jamesmcarthur has quit IRC | 04:41 | |
| *** rlandy|afk has quit IRC | 04:48 | |
| *** jamesmcarthur has joined #zuul | 05:12 | |
| *** bolg has joined #zuul | 05:12 | |
| *** jamesmcarthur has quit IRC | 05:18 | |
| *** bolg_ has joined #zuul | 05:24 | |
| *** bolg has quit IRC | 05:24 | |
| *** raukadah is now known as chkumar|rover | 05:34 | |
| mordred | fungi, fdegir: ++ to defining terms and goals | 05:48 |
| *** toabctl has quit IRC | 06:05 | |
| *** threestrands has quit IRC | 06:08 | |
| *** toabctl has joined #zuul | 06:12 | |
| *** jamesmcarthur has joined #zuul | 06:13 | |
| *** jamesmcarthur has quit IRC | 06:18 | |
| *** saneax has joined #zuul | 06:47 | |
| *** sshnaidm|afk is now known as sshnaidm | 06:48 | |
| *** jamesmcarthur has joined #zuul | 06:51 | |
| *** jamesmcarthur has quit IRC | 06:56 | |
| *** jamesmcarthur has joined #zuul | 07:52 | |
| *** jamesmcarthur has quit IRC | 07:57 | |
| mordred | tobiash: question on https://review.opendev.org/#/c/704150 | 08:12 |
| *** tosky has joined #zuul | 08:13 | |
| *** jamesmcarthur has joined #zuul | 08:27 | |
| *** jamesmcarthur has quit IRC | 08:32 | |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Handle draft pull requests in canMerge https://review.opendev.org/704150 | 08:37 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Enforce sql connections for scheduler and web https://review.opendev.org/630472 | 08:37 |
| bolg_ | Shrews, corvus: anyone in the mood for Add BSD/Darwin support https://review.opendev.org/c/671674/? Or are there any open questions/comments? | 08:43 |
| *** bolg_ has quit IRC | 08:44 | |
| *** bolg has joined #zuul | 08:44 | |
| *** jpena|off is now known as jpena | 08:52 | |
| openstackgerrit | Merged zuul/nodepool master: Remove duplicate log info from OpenStack driver https://review.opendev.org/704379 | 09:00 |
| *** hashar has joined #zuul | 09:03 | |
| *** fbo|off is now known as fbo | 09:06 | |
| *** mhu has joined #zuul | 09:10 | |
| openstackgerrit | Antoine Musso proposed zuul/zuul master: Add client_id to RPC client https://review.opendev.org/704201 | 09:16 |
| openstackgerrit | Merged zuul/zuul master: Be more specific with remaining time https://review.opendev.org/704191 | 09:32 |
| openstackgerrit | Merged zuul/zuul master: Update screenshot for gerrit profile in quick-start https://review.opendev.org/704222 | 09:48 |
| tobiash | mordred: thanks! | 09:55 |
| *** electrofelix has joined #zuul | 10:10 | |
| openstackgerrit | Merged zuul/zuul master: Fix path in quick-start localtest https://review.opendev.org/704385 | 10:11 |
| *** hashar has quit IRC | 11:35 | |
| *** hashar has joined #zuul | 11:58 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-jobs master: Link the policy to the CONTRIBUTING.rst file https://review.opendev.org/704484 | 12:22 |
| *** jpena is now known as jpena|lunch | 12:23 | |
| *** jtanner has quit IRC | 12:55 | |
| *** jtanner has joined #zuul | 12:55 | |
| *** rlandy has joined #zuul | 12:59 | |
| *** portdirect has quit IRC | 13:01 | |
| *** portdirect has joined #zuul | 13:02 | |
| *** jamesmcarthur has joined #zuul | 13:11 | |
| *** jamesmcarthur has quit IRC | 13:13 | |
| *** jamesmcarthur has joined #zuul | 13:13 | |
| *** jpena|lunch is now known as jpena | 13:20 | |
| *** bhavikdbavishi has quit IRC | 13:28 | |
| *** saneax has quit IRC | 13:31 | |
| *** jamesmcarthur has quit IRC | 13:35 | |
| *** jamesmcarthur has joined #zuul | 13:39 | |
| *** jamesmcarthur has quit IRC | 13:45 | |
| *** jamesmcarthur has joined #zuul | 13:47 | |
| *** Goneri has joined #zuul | 13:48 | |
| *** AJaeger has quit IRC | 14:02 | |
| openstackgerrit | Antoine Musso proposed zuul/zuul master: Speed up test_inventory https://review.opendev.org/704570 | 14:03 |
| * hashar likes profiling | 14:06 | |
| *** AJaeger has joined #zuul | 14:19 | |
| *** pcaruana has quit IRC | 14:24 | |
| zbr | easy gear review: https://review.opendev.org/#/c/703425/ | 14:25 |
| zbr | also https://review.opendev.org/#/c/671674/ - very useful | 14:26 |
| hashar | zbr: one I have already +1ed, the other could benefit from python3.4 selectors ( https://docs.python.org/3/library/selectors.html ) :] | 14:36 |
| zbr | corvus: ^ lets first fix these two, and I will be glad to add more fixes, or do reviews. but for start I want it to be able to execute the code on these platforms. | 14:37 |
| zbr | due to that gear bug, I wasn't even able to lint zuul source code on mac | 14:38 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Enforce sql connections for scheduler and web https://review.opendev.org/630472 | 14:40 |
| Shrews | hashar: ooh, i like that selector lib | 14:44 |
| hashar | Shrews: that has hit me with zuul a few years ago :D | 14:44 |
| *** bhavikdbavishi has joined #zuul | 14:57 | |
| *** bhavikdbavishi1 has joined #zuul | 15:00 | |
| *** bhavikdbavishi has quit IRC | 15:01 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 15:01 | |
| *** jamesmcarthur has quit IRC | 15:07 | |
| *** michael-beaver has joined #zuul | 15:08 | |
| *** pcaruana has joined #zuul | 15:11 | |
| openstackgerrit | Antoine Musso proposed zuul/zuul master: Log duration of AnsibleJob.execute() https://review.opendev.org/704589 | 15:22 |
| *** hashar has quit IRC | 15:22 | |
| *** gundalow has quit IRC | 15:36 | |
| *** gundalow has joined #zuul | 15:37 | |
| *** jamesmcarthur has joined #zuul | 15:37 | |
| *** migi has joined #zuul | 15:41 | |
| *** Goneri has quit IRC | 15:47 | |
| *** jamesmcarthur has quit IRC | 15:52 | |
| *** jamesmcarthur has joined #zuul | 15:57 | |
| openstackgerrit | Antoine Musso proposed zuul/zuul master: Log duration of AnsibleJob.execute() https://review.opendev.org/704589 | 16:00 |
| *** chkumar|rover is now known as raukadah | 16:02 | |
| *** bolg_ has joined #zuul | 16:05 | |
| *** bolg has quit IRC | 16:07 | |
| *** bolg_ has quit IRC | 16:07 | |
| *** zxiiro has joined #zuul | 16:15 | |
| *** jpena is now known as jpena|brb | 16:45 | |
| *** rfolco is now known as rfolco|bbl | 16:46 | |
| *** jpena|brb is now known as jpena | 17:24 | |
| *** hashar has joined #zuul | 17:25 | |
| *** bhavikdbavishi has quit IRC | 17:37 | |
| Shrews | hashar: hrm, after deeper inspection, doesn't seem like the selector lib has a clear way to handle POLLERR/POLLHUP/etc, which gear explicitly adds to bitmasks | 17:38 |
| hashar | Shrews: eeeeek :-\\ | 17:38 |
| hashar | Shrews: regardless, the change you pointed earlier looks good as is :] | 17:39 |
| hashar | Shrews: oh those events selectors are dramatically simplified! They are just selectors.EVENT_READ and selectors.EVENT_WRITE :] | 17:45 |
| Shrews | yeah | 17:46 |
| hashar | but yeah one would have to look at all those | 17:46 |
| Shrews | the selector source doesn't even reference the *ERR/HUP bitmasks | 17:46 |
| hashar | and the port is not straightforward | 17:46 |
| *** igordc has joined #zuul | 17:46 | |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: DNM: experiment with pygit2 https://review.opendev.org/704644 | 17:50 |
| *** dustinc has quit IRC | 17:58 | |
| *** dustinc has joined #zuul | 17:58 | |
| *** electrofelix has quit IRC | 17:59 | |
| *** hashar has quit IRC | 18:01 | |
| *** Shrews has quit IRC | 18:09 | |
| *** Shrews has joined #zuul | 18:09 | |
| *** jpena is now known as jpena|off | 18:32 | |
| *** tributarian has quit IRC | 18:56 | |
| *** tributarian has joined #zuul | 18:57 | |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: DNM: experiment with pygit2 https://review.opendev.org/704644 | 19:06 |
| *** stevthedev has quit IRC | 19:18 | |
| *** stevthedev has joined #zuul | 19:19 | |
| *** rfolco|bbl has quit IRC | 19:19 | |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Enforce sql connections for scheduler and web https://review.opendev.org/630472 | 19:19 |
| *** jamesmcarthur has quit IRC | 19:21 | |
| openstackgerrit | Jeremy Stanley proposed zuul/zuul master: Flesh out the glossary significantly https://review.opendev.org/704391 | 19:24 |
| clarkb | The OSF Annual Report is out, including a Zuul update https://www.openstack.org/foundation/2019-openstack-foundation-annual-report | 19:27 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Enforce sql connections for scheduler and web https://review.opendev.org/630472 | 19:30 |
| *** armstrongs has joined #zuul | 19:37 | |
| *** hashar has joined #zuul | 19:42 | |
| *** armstrongs has quit IRC | 19:44 | |
| fungi | i generated the numbers for that report, so blame me if they're suspect | 20:08 |
| Shrews | only 1000+ changes for zuul in 2019? i would have guessed more | 20:10 |
| Shrews | i wonder if that included nodepool + zuul-website + zuul-registry + zuul-jobs + etc | 20:11 |
| clarkb | was on the etherpad iirc /me looks | 20:12 |
| clarkb | now I have to find the etherpad | 20:13 |
| Shrews | clarkb: don't worry about it :) | 20:13 |
| Shrews | i suspect my expectations more than fungi's maths :) | 20:14 |
| hashar | excluding merges, based on author year: 2017: 1229 - 2018: 744 - 2019: 588 - 2020: 52 | 20:15 |
| hashar | git log --no-merges |grep Date|cut -d\ -f8|sort|uniq -c | 20:15 |
| clarkb | Shrews: https://etherpad.openstack.org/p/zuul-2019-annual-report-data | 20:16 |
| hashar | but that is just zuul/zuul.git ;) | 20:16 |
| clarkb | fungi put that data together | 20:16 |
| fungi | yes, it was gerrit changes merged in 2019 across all repositories in the "zuul" git namespace on opendev | 20:17 |
| hashar | \o/ | 20:18 |
| fungi | there were 4219 change revisions, so changes which merge to zuul repos seem to average around 4 revisions each | 20:19 |
| openstackgerrit | Andrii Ostapenko proposed zuul/zuul-jobs master: Fix periodic image build jobs https://review.opendev.org/704680 | 20:19 |
| *** mhu has quit IRC | 20:22 | |
| *** rfolco has joined #zuul | 20:24 | |
| *** jamesmcarthur has joined #zuul | 20:51 | |
| pabelanger | Any thing to focus on, to maybe get new zuul release this week? A few github / UI changes that look nice for us (zuul.a.c) | 20:57 |
| *** jamesmcarthur has quit IRC | 21:17 | |
| *** jamesmcarthur has joined #zuul | 21:18 | |
| corvus | pabelanger: no, i think just time | 21:44 |
| corvus | paladox: i'm still bootstrapping the system, so it's not all working yet. but i think i got all the access issues resolved with han-wen's help. here is gerrit's zuul reporting its first successful (noop) run via the checks plugin: https://gerrit-review.googlesource.com/c/zuul/ops/+/252317 | 21:47 |
| paladox | :O \o/ | 21:47 |
| paladox | I guess the log pointing to localhost is expected? | 21:48 |
| corvus | paladox: yep, none of that is configured right yet :) | 21:48 |
| paladox | heh | 21:49 |
| corvus | zuul-maint: the gerrit installation of zuul is going to require one extra microservice -- a daemon that gets an auth token from the metadata service and writes it to disk. this is because we need to use a service account auth token to upload logs, but we can't allow the executor pod(s) to access the metadata service. this is pretty gke specific, but it's not necessarily specific to google's gerrit (ie, | 21:55 |
| corvus | this is a legit option for anyone running zuul in gke, though there are alternative ways of setting the system up that don't use service accounts). | 21:55 |
| corvus | zuul-maint: my question for you all is: are we interested in hosting the code for this microservice on opendev, or should i ask the gerrit project to host it? | 21:55 |
| fungi | if it's useful to anyone doing log publication through gke, i see no problem considering it part of zuul | 21:56 |
| Shrews | I can't immediately think of a reason *not* to host it | 21:58 |
| tristanC | i feel the same as fungi and Shrews | 21:59 |
| *** jamesmcarthur has quit IRC | 22:01 | |
| pabelanger | wfm | 22:01 |
| corvus | thanks, i'll push up a change in a bit to add a project then | 22:01 |
| *** jamesmcarthur has joined #zuul | 22:03 | |
| clarkb | the executors can't talk to the metadta service because there are other secrets hosted there? | 22:03 |
| corvus | clarkb: any job could use the uri module to grab the service account token from the metadata service. so we have to prevent that by annotating the pod in such a way that the account doesn't have any access. but that also means we can't upload logs, so we get the token into the executor via a file, so we can use the usual bwrap constraints. | 22:09 |
| clarkb | and that has to be renewed periodically I guess | 22:09 |
| corvus | yep | 22:10 |
| corvus | every hour | 22:10 |
| corvus | this gets you the default service account bearer token: curl -H Metadata-Flavor:Google http://metadata.google.internal/computeMetadata/v1/instance/service-accounts/default/token | 22:10 |
| corvus | so, not reall hard to take advantage of that :) | 22:10 |
| tristanC | zuul-maint: ftr, https://review.opendev.org/#/q/topic:zuul-crd is working as expected, and with https://review.opendev.org/#/c/704187/ it would get published to docker.io/zuul/zuul-operator | 22:13 |
| openstackgerrit | Antoine Musso proposed zuul/zuul master: ansible manager: only failed if last ansible failed https://review.opendev.org/704699 | 22:15 |
| openstackgerrit | Antoine Musso proposed zuul/zuul master: zuul-manage-ansible: add validate option https://review.opendev.org/704701 | 22:16 |
| *** jamesmcarthur has quit IRC | 22:34 | |
| *** jamesmcarthur has joined #zuul | 22:36 | |
| *** jamesmcarthur has quit IRC | 22:45 | |
| *** rfolco has quit IRC | 22:46 | |
| *** jamesmcarthur has joined #zuul | 22:50 | |
| *** hashar has quit IRC | 22:56 | |
| *** jamesmcarthur has quit IRC | 23:03 | |
| *** rlandy is now known as rlandy|bbl | 23:32 | |
| *** panda has left #zuul | 23:37 | |
| *** michael-beaver has quit IRC | 23:47 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!