| ianw | so many will never know the fun of fingering john carmack's plan | 00:03 |
|---|---|---|
| opendevreview | Merged zuul/zuul master: Report intermediate buildsets and builds https://review.opendev.org/c/zuul/zuul/+/797182 | 00:07 |
| opendevreview | Merged zuul/zuul master: Add nodeset to build table https://review.opendev.org/c/zuul/zuul/+/798207 | 00:08 |
| corvus | ianw: lol, if i google "finger john carmack" google image results sure does bring up a picture of johnc holding up all of his fingers! | 00:49 |
| opendevreview | Merged zuul/zuul master: Link to the build page from the status page https://review.opendev.org/c/zuul/zuul/+/800111 | 01:03 |
| opendevreview | Merged zuul/zuul master: Create config cache ltime before requesting files https://review.opendev.org/c/zuul/zuul/+/800659 | 02:39 |
| *** chandankumar is now known as chkumar|rover | 05:21 | |
| opendevreview | Simon Westphahl proposed zuul/zuul master: Ensure config cache stages are used correctly https://review.opendev.org/c/zuul/zuul/+/800973 | 05:41 |
| opendevreview | Simon Westphahl proposed zuul/zuul master: Ensure config cache stages are used correctly https://review.opendev.org/c/zuul/zuul/+/800660 | 05:46 |
| swest | corvus: thanks for the follow up fix re the smart reconfig. I squashed 800973 into 800660 and updated the doc string | 05:50 |
| swest | clarkb: ^ | 05:50 |
| opendevreview | Simon Westphahl proposed zuul/zuul master: Use the ZK cache for smart reconfiguration events https://review.opendev.org/c/zuul/zuul/+/800973 | 05:54 |
| *** iurygregory_ is now known as iurygregory | 06:38 | |
| *** rpittau|afk is now known as rpittau | 07:31 | |
| *** dmellado_ is now known as dmellado | 07:41 | |
| *** bhagyashris_ is now known as bhagyashris|ruck | 08:24 | |
| opendevreview | Simon Westphahl proposed zuul/zuul master: Ensure config cache stages are used correctly https://review.opendev.org/c/zuul/zuul/+/800660 | 09:21 |
| *** dviroel|out is now known as dviroel | 11:19 | |
| *** marios_ is now known as marios | 12:40 | |
| corvus | swest: thanks! i carried over clarkb's +2 and approved | 13:00 |
| *** corvus is now known as Guest1054 | 13:01 | |
| *** Guest1054 is now known as corvus | 13:04 | |
| *** kain99[m] is now known as erabotni[m] | 13:04 | |
| *** corvus is now known as notcorvus | 13:04 | |
| *** notcorvus is now known as corvus | 13:05 | |
| *** corvus is now known as Guest1055 | 13:05 | |
| *** Guest1055 is now known as corvus | 13:06 | |
| corvus | clarkb, tobiash: if you have a minute to re-review https://review.opendev.org/800112 -- i fixed up the quick-start failure (by changing the tutorial to match the new default behavior) | 13:10 |
| *** marios is now known as marios|call | 13:10 | |
| *** marios|call is now known as marios | 13:34 | |
| *** rpittau is now known as rpittau|afk | 14:08 | |
| clarkb | corvus: ya I can take a look in a bit. | 15:01 |
| opendevreview | Andy Ladjadj proposed zuul/zuul master: Improve performance of _saveRepoState https://review.opendev.org/c/zuul/zuul/+/800664 | 15:12 |
| clarkb | corvus: couple of minor things on 800112 you may want to check but I +2'd it | 15:17 |
| clarkb | corvus: also 800660 had a post failure error in the image upload job | 15:21 |
| corvus | clarkb: thx. and drat. | 15:23 |
| avass[m] | heh, now element on windows is even worse. I can't event see corvus name | 15:25 |
| * avass[m] uploaded an image: (3KiB) < https://matrix.org/_matrix/media/r0/download/vassast.org/HPxdMmqBXWKiDFPoiZFWMsRv/image.png > | 15:25 | |
| corvus | mordred: ^ did you say you ran into that? | 15:25 |
| corvus | avass: try making it wider | 15:26 |
| avass[m] | I can see it if I mouseover :) | 15:26 |
| clarkb | in the web client I have to hard refresh periodically because element loses its i18n info and puts an error around every string saying "translation not found" | 15:27 |
| corvus | heh, the slightly old version of element i'm running is rock solid :) | 15:27 |
| avass[m] | that sort of works, but your name is left aligned while every one else is right aligned, and I gotta make it wide enough that I feel like it's taking too much space | 15:28 |
| clarkb | I've mostly ignored it since the fix is so simple | 15:28 |
| avass[m] | corvus: I haven't had any issues on linux clients :) | 15:28 |
| corvus | i think this is a result of the room and irc getting slightly out of sync earlier, and now my irc puppet user is separately registered as a room member. | 15:30 |
| corvus | it's one of those minor annoyances that come along with bridging | 15:30 |
| avass[m] | oh apparently i have the same issue with ianw | 15:31 |
| corvus | (so therefore, element has to disambiguate the two corvii) | 15:31 |
| corvus | same deal | 15:31 |
| corvus | also, probably if i change either my display name or irc nick, the issue would disappear | 15:31 |
| mordred | ah - I had the issue before - but then it went away | 15:55 |
| opendevreview | Merged zuul/zuul master: Fix typo in matrix spec https://review.opendev.org/c/zuul/zuul/+/796317 | 16:22 |
| corvus | sheesh; that's the change that we managed to merge? :) | 16:24 |
| clarkb | the one we want hasn't managed to fail yet :) | 16:25 |
| *** marios is now known as marios|out | 16:35 | |
| opendevreview | Merged zuul/zuul master: Ensure config cache stages are used correctly https://review.opendev.org/c/zuul/zuul/+/800660 | 17:39 |
| opendevreview | Merged zuul/zuul master: Always report the build page https://review.opendev.org/c/zuul/zuul/+/800112 | 18:08 |
| opendevreview | James E. Blair proposed zuul/zuul master: Add a 'complete' parameter to build/buildset queries https://review.opendev.org/c/zuul/zuul/+/801139 | 21:41 |
| corvus | clarkb, mordred: i think we need an api change before we can do a sensible update to the js app | 21:42 |
| corvus | ^ | 21:42 |
| corvus | hrm, flake8 doesn't like that | 21:45 |
| corvus | gimme a min to run the test suite and i'll push an update | 21:48 |
| opendevreview | James E. Blair proposed zuul/zuul master: Add a 'complete' parameter to build/buildset queries https://review.opendev.org/c/zuul/zuul/+/801139 | 21:57 |
| corvus | clarkb, mordred: ^ okay, that passes tests locally, and is step 1. | 21:57 |
| corvus | we won't be able to evaluate step 2 against our production data until that change is running though, because we need to add the new query parameters | 21:58 |
| corvus | maybe we can go ahead and approve that, and i can see if i have some data locally with which i can work on the next step | 22:00 |
| clarkb | if we are going to approve changes to the sqlconnection I wonder if we should plan to restart on it before the gerrit work. Just to be sure we don't have unexpected bugs | 22:06 |
| clarkb | corvus: left a couple of thoughts | 22:07 |
| corvus | clarkb: yeah, i'm happy to restart over the wknd | 22:08 |
| corvus | clarkb: why hide the failed typecast? why not just let that return 500? | 22:08 |
| clarkb | I guess that should work too | 22:10 |
| clarkb | I think the one case where it might weird is if you send ?complete=None ? but you can just drop the parameter in that case | 22:10 |
| corvus | true; i think if we want to support that, maybe it'd be better to special-case none; but at least initially, i was imagining users would just omit | 22:12 |
| clarkb | ok +2'd | 22:13 |
| corvus | clarkb: i think it might be just you and me (and sometimes mordred?) at this point, and i think it'd be good to get this done before monday, so i went ahead and approved it | 22:18 |
| clarkb | ok | 22:18 |
| clarkb | I intend on being around Sunday afternoon and evening to help wtih gerrit things fwiw. But probably won't be around tomorrow | 22:20 |
| corvus | clarkb: hrm, pep8 told me to do the wrong thing | 22:22 |
| corvus | it said i had to type "complete is None" but sqlalchemy doesn't handle that | 22:23 |
| corvus | so i need to find a way to convince pep8 to let me type "complete == None" | 22:23 |
| clarkb | corvus: at the end of the line you can do # noqa iirc | 22:23 |
| clarkb | or in a comment above the line | 22:23 |
| opendevreview | James E. Blair proposed zuul/zuul master: Add a 'complete' parameter to build/buildset queries https://review.opendev.org/c/zuul/zuul/+/801139 | 22:25 |
| corvus | ugh, bool('0') is true | 22:32 |
| opendevreview | James E. Blair proposed zuul/zuul master: Add a 'complete' parameter to build/buildset queries https://review.opendev.org/c/zuul/zuul/+/801139 | 22:34 |
| opendevreview | James E. Blair proposed zuul/zuul master: Add a 'complete' parameter to build/buildset queries https://review.opendev.org/c/zuul/zuul/+/801139 | 22:37 |
| corvus | clarkb: ^ okay that works | 22:37 |
| clarkb | corvus: and I guess complete is always a string as provided by cherrypy? | 22:39 |
| corvus | yeah | 22:40 |
| corvus | there's prior art with 'final' too, so i'm matching that behavior | 22:40 |
| clarkb | I've +2d it | 22:45 |
| opendevreview | James E. Blair proposed zuul/zuul master: Don't display in-progress builds/buildsets in the web listing https://review.opendev.org/c/zuul/zuul/+/801156 | 22:59 |
| corvus | clarkb, mordred: that's step 2. i've tested it with a zuul i have laying around that has a couple of incomplete builds in its database. i'm pretty sure that change is good, but won't have 100% confidence until we can restart opendev with 801139. i think it would be equally reasonable to keep them separate and do 2 opendev zuul-web restarts, or to go ahead and merge both and do one restart. | 23:01 |
| corvus | i bet that needs test updates tho | 23:03 |
| corvus | huh, maybe not. | 23:03 |
| clarkb | corvus: can that change also remove the web ui handling of retry_builds? seems we stop writing that list but then don't stop hanlding it separately? | 23:05 |
| clarkb | though maybe it was just completely ignored before so same difference? | 23:05 |
| corvus | clarkb: yeah, it was ignored before -- nothing in the web ui displayed it | 23:14 |
| corvus | okay 801156 passes tests locally | 23:15 |
| corvus | ugh, now the nodejs lint doesn't like 801139 | 23:24 |
| opendevreview | James E. Blair proposed zuul/zuul master: Add a 'complete' parameter to build/buildset queries https://review.opendev.org/c/zuul/zuul/+/801139 | 23:25 |
| opendevreview | James E. Blair proposed zuul/zuul master: Don't display in-progress builds/buildsets in the web listing https://review.opendev.org/c/zuul/zuul/+/801156 | 23:25 |
| corvus | clarkb: what's your preference for opendev zuul-web restarts? 1 restart with both changes merged, or merge the first, restart, retest the second, then merge the second and restart? | 23:30 |
| clarkb | The risk for doing one restart is that the build and buildset listings will stop working ya? | 23:30 |
| clarkb | that seems pretty low risk | 23:30 |
| corvus | yeah, and agreed | 23:31 |
| corvus | i think i'll go ahead and approve that then | 23:32 |
| corvus | hopefully that means i can restart with both tonight. if something goes wrong, i can fix it tomorrow. | 23:33 |
| corvus | clarkb: i'm guessing the next restart will be past your EOD; i'll take care of it | 23:34 |
| clarkb | corvus: ok sounds like a plan | 23:34 |
| corvus | i am going to take a break now | 23:34 |
| clarkb | o/ | 23:35 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
