| mnaser | just switched it to run as 10001 | 00:02 |
|---|---|---|
| mnaser | and ill see how that works | 00:02 |
| *** josefwells has joined #zuul | 00:04 | |
| mnaser | clarkb: that actually did the trick! | 00:10 |
| mnaser | i believe we should USER zuul and USER nodepool in the docker iamges imho | 00:10 |
| corvus | tristanC: ^ that going to be okay with openshift? | 00:11 |
| *** josefwells__ has quit IRC | 00:11 | |
| clarkb | I think if you use USER you can still override the user | 00:13 |
| clarkb | but if you don't use USER then it will default to root | 00:13 |
| ianw | i don't know but i'm getting "dib won't work with that" vibes :) | 00:13 |
| mnaser | Doesn’t it work already with nodepool Clarke | 00:14 |
| clarkb | user: nodepool we set that on nodepool-builder containers with docker-compose | 00:16 |
| clarkb | so ya unless there is some weird not well documented behavior difference between explicitly setting USER and overriding user at runtime I would expect it to be ok | 00:16 |
| opendevreview | Matthieu Huin proposed zuul/zuul master: Test zuul-client console-stream https://review.opendev.org/c/zuul/zuul/+/794854 | 00:20 |
| mordred | mnaser: we used to set user in places and removed it | 00:20 |
| clarkb | I think the problem had to do with hardcoded uids? but I think we are doing thatn ow anyway just without the helper mode for people who aren't randomizing those? | 00:21 |
| clarkb | hrm the current Dockerfile doesn't seem to have USER in its history anywhere | 00:23 |
| clarkb | for zuul I mean | 00:23 |
| fungi | we must do some user mapping because the host system sees zuul:zuuld ownership on files written by the container processes on bindmounted dirs | 00:25 |
| clarkb | fungi: I think we hardcode the uid in ansibel for things that map across | 00:26 |
| opendevreview | Merged zuul/nodepool master: Update dib to 3.12.0 https://review.opendev.org/c/zuul/nodepool/+/795416 | 00:36 |
| *** josefwells has quit IRC | 01:29 | |
| *** bhavikdbavishi has joined #zuul | 04:19 | |
| *** marios has joined #zuul | 05:10 | |
| opendevreview | Simon Westphahl proposed zuul/zuul master: Refactor config/tenant (re-)loading https://review.opendev.org/c/zuul/zuul/+/795263 | 05:20 |
| *** zenkuro has quit IRC | 06:10 | |
| *** bhavikdbavishi has quit IRC | 06:40 | |
| *** bhavikdbavishi has joined #zuul | 06:40 | |
| *** hashar has joined #zuul | 06:58 | |
| *** jpena|off is now known as jpena | 07:31 | |
| *** bhavikdbavishi has quit IRC | 07:51 | |
| *** tosky has joined #zuul | 07:53 | |
| opendevreview | Felix Edel proposed zuul/zuul master: WIP: Lock/unlock nodes on executor server https://review.opendev.org/c/zuul/zuul/+/774610 | 08:14 |
| opendevreview | Felix Edel proposed zuul/zuul master: WIP: Lock/unlock nodes on executor server https://review.opendev.org/c/zuul/zuul/+/774610 | 08:24 |
| avass[m] | mnaser: see: https://zuul-ci.org/docs/zuul-jobs/policy.html#preservation-of-owner-between-executor-and-remote | 08:28 |
| avass[m] | mnaser: that's why that rsync fails, `invalid argument` isn't very helpful when what it means is that the user doesn't exist | 08:29 |
| opendevreview | Guillaume Chauvel proposed zuul/zuul master: [DNM] dummy test https://review.opendev.org/c/zuul/zuul/+/795501 | 08:49 |
| opendevreview | Simon Westphahl proposed zuul/zuul master: Make reporting asynchronous https://review.opendev.org/c/zuul/zuul/+/691253 | 08:58 |
| *** rpittau|afk is now known as rpittau | 09:22 | |
| *** ysandeep|ruck has joined #zuul | 09:25 | |
| opendevreview | Merged zuul/zuul-client master: Add build-info subcommand https://review.opendev.org/c/zuul/zuul-client/+/751070 | 09:26 |
| ysandeep|ruck | #zuul hello folks o/ top jobs in tripleo gate queue are stuck.. | 09:27 |
| ysandeep|ruck | example: https://zuul.opendev.org/t/openstack/status#794634 and other changes under the chain | 09:27 |
| ysandeep|ruck | https://zuul.opendev.org/t/openstack/stream/f0e55388ce7d4cf7b989315c18ca5d3d?logfile=console.log - is already at --- END OF STREAM --- | 09:27 |
| ysandeep|ruck | could someone please help with this? | 09:28 |
| tobiash[m] | corvus, clarkb I wonder if this is a regression introduced by one of the latest changes (maybe a lost result event?) ^ let me know if I can help with the analysis | 09:40 |
| opendevreview | Matthieu Huin proposed zuul/zuul master: Test zuul-client console-stream https://review.opendev.org/c/zuul/zuul/+/794854 | 09:51 |
| *** hashar has quit IRC | 09:52 | |
| opendevreview | Simon Westphahl proposed zuul/zuul master: Make reporting asynchronous https://review.opendev.org/c/zuul/zuul/+/691253 | 10:16 |
| opendevreview | Simon Westphahl proposed zuul/zuul master: Make reporting asynchronous https://review.opendev.org/c/zuul/zuul/+/691253 | 10:16 |
| opendevreview | Guillaume Chauvel proposed zuul/zuul master: [DNM] dummy test https://review.opendev.org/c/zuul/zuul/+/795501 | 10:32 |
| *** ysandeep|ruck is now known as ysandeep|afk | 10:33 | |
| opendevreview | Matthieu Huin proposed zuul/zuul master: Test zuul-client console-stream https://review.opendev.org/c/zuul/zuul/+/794854 | 10:42 |
| *** ysandeep|afk is now known as ysandeep|ruck | 11:06 | |
| opendevreview | Mohammed Naser proposed zuul/zuul-helm master: Run Zuul under the 'zuul' user. https://review.opendev.org/c/zuul/zuul-helm/+/795518 | 11:11 |
| mnaser | ianw: no idea how, no idea why -- tried fedora-container this morning and .. "diskimage_builder.element_dependencies.MissingElementException: Element 'block-device' not found" | 11:26 |
| mnaser | urgh, nevermind | 11:28 |
| mnaser | i forgot to parent to my `base` so was misssing vm element which brought in block-device | 11:29 |
| mnaser | "Error: 'overlay' is not supported over overlayfs, a mount_program is required: backing file system is unsupported for this graph driver" weee | 11:31 |
| mnaser | looks like /var/lib/containers mount time | 11:34 |
| avass[m] | I'm currently debugging why zuul seemingly isn't receiving events from one connection. The scheduler can receive events with `gerrit stream-events`, and it was able to load configuration for projects from that connection. It's just not triggering anything and not logging any events. Any ideas how that could happen? | 11:34 |
| avass[m] | Those projects worked before restarting yesterday to pick up the layout caching bugfixes | 11:35 |
| tristanC | avass[m]: perhaps there is a lock in /zuul/event zk, we had to `zkCli.sh deleteall /zuul/events` to recover from that situation | 11:36 |
| avass[m] | tristanC: I was looking at that and thinking that it may be enough to delete the keys for that connection? | 11:36 |
| avass[m] | I'm gonna see if that works | 11:37 |
| mnaser | excellent. mounting /var/lib/containers did the trick and i got a containerfile image built | 11:39 |
| avass[m] | mnaser: did you see my comment earlier? | 11:41 |
| mnaser | avass[m]: yes! wrt the chown invalid arugment thing | 11:41 |
| avass[m] | ok, just making sure :) | 11:41 |
| mnaser | avass[m]: i think it was you that uses a bunch of zuul+gitlab right? | 11:42 |
| avass[m] | mnaser: more like trying to ;) | 11:42 |
| avass[m] | mnaser: guillaumec uses gitlab actively I believe | 11:42 |
| mnaser | avass[m]: ah i see, guillaumec fyi not sure if you are an ultimate user but this might be interesting to look at -- https://docs.gitlab.com/ee/user/project/merge_requests/approvals/#notify-external-services | 11:43 |
| *** jpena is now known as jpena|lunch | 11:44 | |
| avass[m] | tristanC: did you couple that with a restart? | 11:55 |
| tristanC | avass[m]: we do it before the restart | 11:56 |
| avass[m] | shouldn't locks be deleted when the zk session is lost? | 11:57 |
| tristanC | avass[m]: it should, but in our case we needed it to happen faster than zk timeout, so deleting all the zk events node before a restart did the trick | 11:58 |
| guillaumec | mnaser: thanks, premium only | 11:59 |
| opendevreview | Guillaume Chauvel proposed zuul/zuul master: WIP gitlab status context https://review.opendev.org/c/zuul/zuul/+/795525 | 12:00 |
| avass[m] | tristanC: oh so something like: the lock is still present so the scheduler sees that and thinks there is another scheduler that is managing those events, or does restarting not create a new session somehow? In my case the lock was still there so I'd expect the current scheduler to be managing the events for that connection. | 12:01 |
| avass[m] | but I'll try and see if a deleting /zuul/events before restarting helps later. | 12:02 |
| avass[m] | and I would have expected a leader election to occur if the lock is deleted, but maybe that didn't work in my case because I deleted `/zuul/events/connection/<connection>` | 12:04 |
| tristanC | avass[m]: i haven't had time to dig into that issue, it was speculated that the scheduler was not stopped cleanly, but even with https://review.opendev.org/c/zuul/zuul/+/794035 that did not helped | 12:04 |
| avass[m] | tristanC: ok, I'll continue digging a bit | 12:05 |
| *** ysandeep|ruck is now known as ysandeep|mtg | 12:07 | |
| opendevreview | Guillaume Chauvel proposed zuul/zuul master: [DNM] dummy test https://review.opendev.org/c/zuul/zuul/+/795501 | 12:11 |
| opendevreview | Matthieu Huin proposed zuul/zuul master: Test zuul-client console-stream https://review.opendev.org/c/zuul/zuul/+/794854 | 12:31 |
| opendevreview | Guillaume Chauvel proposed zuul/zuul master: [DNM] dummy test https://review.opendev.org/c/zuul/zuul/+/795501 | 12:39 |
| *** felixedel[m] has joined #zuul | 12:40 | |
| opendevreview | Matthieu Huin proposed zuul/zuul-client master: builds: fix API queries for boolean parameters, make tenant optional https://review.opendev.org/c/zuul/zuul-client/+/794553 | 12:40 |
| *** felixedel[m] is now known as felixedel | 12:41 | |
| *** felixedel has joined #zuul | 12:41 | |
| opendevreview | Guillaume Chauvel proposed zuul/zuul master: [DNM] dummy test https://review.opendev.org/c/zuul/zuul/+/795501 | 12:43 |
| opendevreview | Benjamin Schanzel proposed zuul/zuul-jobs master: Add a meta log upload role with a failover mechanism https://review.opendev.org/c/zuul/zuul-jobs/+/795336 | 12:48 |
| *** jpena|lunch is now known as jpena | 12:49 | |
| opendevreview | Guillaume Chauvel proposed zuul/zuul master: [DNM] dummy test https://review.opendev.org/c/zuul/zuul/+/795501 | 13:02 |
| opendevreview | Guillaume Chauvel proposed zuul/zuul master: [DNM] dummy test https://review.opendev.org/c/zuul/zuul/+/795501 | 13:08 |
| *** Simon[m]1 has joined #zuul | 13:10 | |
| *** Simon[m]1 is now known as swest[m] | 13:17 | |
| *** swest[m] has quit IRC | 13:19 | |
| *** swest[m] has joined #zuul | 13:19 | |
| opendevreview | Guillaume Chauvel proposed zuul/zuul master: WIP gitlab quick-start https://review.opendev.org/c/zuul/zuul/+/795540 | 13:33 |
| opendevreview | Matthieu Huin proposed zuul/zuul-client master: Add buildsets, buildset-info to subcommands https://review.opendev.org/c/zuul/zuul-client/+/752909 | 13:35 |
| opendevreview | Matthieu Huin proposed zuul/zuul-client master: Add console-stream subcommand https://review.opendev.org/c/zuul/zuul-client/+/751238 | 13:37 |
| opendevreview | Matthieu Huin proposed zuul/zuul-client master: Add change-status subcommand https://review.opendev.org/c/zuul/zuul-client/+/759838 | 13:38 |
| *** ysandeep|mtg is now known as ysandeep | 13:43 | |
| *** Shrews has joined #zuul | 13:47 | |
| *** josefwells has joined #zuul | 14:00 | |
| opendevreview | Guillaume Chauvel proposed zuul/zuul master: WIP gitlab quick-start https://review.opendev.org/c/zuul/zuul/+/795540 | 14:11 |
| *** tosky has quit IRC | 14:16 | |
| *** tosky has joined #zuul | 14:18 | |
| *** zenkuro has joined #zuul | 14:23 | |
| *** hashar has joined #zuul | 14:57 | |
| *** ysandeep is now known as ysandeep|away | 15:50 | |
| *** ysandeep|away has quit IRC | 16:06 | |
| *** marios is now known as marios|out | 16:15 | |
| *** rpittau is now known as rpittau|afk | 16:16 | |
| *** marios|out has quit IRC | 16:23 | |
| opendevreview | Guillaume Chauvel proposed zuul/zuul master: WIP gitlab quick-start https://review.opendev.org/c/zuul/zuul/+/795540 | 16:27 |
| *** jpena is now known as jpena|off | 16:38 | |
| opendevreview | Merged zuul/zuul master: Add support for vote-deleted event https://review.opendev.org/c/zuul/zuul/+/791572 | 16:44 |
| opendevreview | James E. Blair proposed zuul/zuul master: WIP: fix unknown job detection https://review.opendev.org/c/zuul/zuul/+/795597 | 17:02 |
| corvus | tobiash, fungi, clarkb, frickler: ^ that should have caught the stuck job issue; let me see if i can make a test for that | 17:02 |
| tobiash[m] | great :) | 17:04 |
| corvus | to summarize from #opendev -- the executor VM was terminated by the cloud, geard probably detected that and returned a WORKFAIL packet to the scheduler, but since we're in a transition period between geard and zk, and result events are in zk, the scheduler ignored the WORKFAIL packet (since that's a result event). the lost job cleanup process should have caught that and submitted a real "LOST" result event to the scheduler, but that code has a | 17:05 |
| corvus | bug | 17:05 |
| corvus | i think it's best to just fix the lost cleanup event handling, and rely on that for catching this case, rather than to try to handle the WORK_FAIL case | 17:05 |
| corvus | we only need this to work for a little while longer, then we'll be putting build requests in zk and it won't be applicable any more | 17:06 |
| opendevreview | Guillaume Chauvel proposed zuul/zuul master: WIP gitlab quick-start https://review.opendev.org/c/zuul/zuul/+/795540 | 17:24 |
| opendevreview | Mohammed Naser proposed zuul/zuul-jobs master: Switch jobs to use fedora-34 nodes https://review.opendev.org/c/zuul/zuul-jobs/+/795636 | 18:15 |
| opendevreview | Mohammed Naser proposed zuul/zuul-operator master: Switch jobs to use fedora-34 nodes https://review.opendev.org/c/zuul/zuul-operator/+/795638 | 18:16 |
| opendevreview | Mohammed Naser proposed zuul/nodepool master: Switch fedora-latest to use fedora-34 https://review.opendev.org/c/zuul/nodepool/+/795642 | 18:22 |
| opendevreview | James E. Blair proposed zuul/zuul master: Fix unknown job detection https://review.opendev.org/c/zuul/zuul/+/795597 | 18:34 |
| corvus | zuul-maint: ^ would you please expedite the review of that? it fixes a regression opendev saw in production; i'd like to merge that asap and make a release | 18:35 |
| * clarkb rereviews | 18:38 | |
| clarkb | corvus: I have a couple of questions, but I thnk you can probabl approve it if my questions are not critical | 18:47 |
| opendevreview | James E. Blair proposed zuul/zuul master: Fix unknown job detection https://review.opendev.org/c/zuul/zuul/+/795597 | 18:51 |
| corvus | clarkb: replied. i ran flake8 locally and it caught two nits; that's a fixed version | 18:51 |
| corvus | i think we can carry over all the other +2s | 18:51 |
| corvus | (flake8 caught a comment whitespace issue and an usused variable) | 18:52 |
| clarkb | corvus: that helps, thanks | 18:52 |
| corvus | thanks all; assuming that merges, i think we can go ahead and cut a release with that and the 2 layout fixes | 18:54 |
| * mnaser is going to be hacking curl/curl to run inside zuul to propose replacing travis ci | 18:55 | |
| * mnaser deciphering https://github.com/curl/curl/blob/master/.travis.yml | 18:55 | |
| corvus | mnaser: oh are they thinking about taking you up on your offer? | 18:57 |
| *** timburke_ is now known as timburke | 18:58 | |
| mnaser | corvus: "We're not picky, we can move to any service that interfaces with github decently. It's more about someone doing the actual heavy lifting of converting travis jobs to other-service-job." | 18:58 |
| mnaser | and ran out of credits at travis already in the first few days | 18:58 |
| mnaser | so i figured if there's a ci service and jobs, maybe we can convince them to at least check for now.. not gate | 18:58 |
| avass[m] | mnaser: oh sounds interesting, I can help out when I have time over if needed | 19:00 |
| mnaser | avass[m]: awesome. i will get a tenant running shortly and make vexxhost/curl run inside of it for now | 19:00 |
| mnaser | the jobs seem straight forward | 19:00 |
| corvus | mnaser: neat; most of that travis file looks pretty straightforward; not sure what the cache is for | 19:00 |
| mordred | corvus: it's for caching things | 19:00 |
| avass[m] | I've done a tiny bit of travisCI when I didn't have my zuul instance set up yet. Hope i remember something :) | 19:01 |
| mordred | corvus: (I believe that's telling travis to save those dirs when the job is done and to start the next job with the contents of what was previous saved) | 19:01 |
| mnaser | yeah its saving state across jobs | 19:01 |
| mordred | I expect their build process clones those from somewhere | 19:01 |
| corvus | mnaser: i'd totally do most of that as a big shell task just to get things going; then maybe split it up a bit for improved ui | 19:01 |
| avass[m] | corvus: I think it's caching directories between builds and nothing more complicated than that, github actions has something similar and was the inspiration for zuul-cache | 19:02 |
| corvus | mordred: so maybe that happens in a makefile? | 19:02 |
| corvus | my confusion was i didn't see where it was getting populated | 19:02 |
| mnaser | yes it archives post-job-completion those folders | 19:02 |
| corvus | i was expecting "curl http://something/wolfssl" but if that happens in a makefile, then it makes sense | 19:02 |
| mnaser | and loads if they dont exist pre-job-start, but yeah | 19:02 |
| mnaser | yeah, i MIGHT just do bindep.txt though to cover the apt installs | 19:03 |
| mnaser | but that might be too much =P | 19:03 |
| corvus | in which case avass's roles could be very helpful; but also, they're probably not needed for a first pass | 19:03 |
| corvus | mnaser: a change to move all those installs to bindep might look really nice and be a good introduction to bindep for them :) | 19:04 |
| avass[m] | mnaser: ansibles apt module is als overy similar otherwise | 19:04 |
| corvus | cause i bet a bunch of bindep environments would fit very well with how they're making those package lists | 19:04 |
| avass[m] | yeah | 19:04 |
| mordred | mnaser: you should totally do a bindep.txt file | 19:05 |
| mordred | because as you're showing them - you can also show them the wonderful value of being able to machien-specify those things | 19:05 |
| mordred | I'd say do it in a patch series so it could be 2 patches - but github | 19:05 |
| corvus | well, you can still depends-on :) | 19:05 |
| mordred | corvus: ooh - good point | 19:05 |
| mnaser | https://github.com/curl/curl/blob/master/scripts/travis/script.sh | 19:06 |
| corvus | personally, i like the idea of doing the most straightforward translation first, followed by "here's how this could be improved" | 19:06 |
| mnaser | it looks like most of the work is already done i guess | 19:06 |
| mnaser | pass env variables and call that script.. | 19:06 |
| mnaser | https://github.com/curl/curl/blob/master/scripts/travis/before_script.sh | 19:06 |
| mnaser | you see git clones, i see required-projects .. look at all the potential | 19:06 |
| mnaser | :p | 19:06 |
| * corvus lunches | 19:07 | |
| avass[m] | mnaser: may wanna keep the git clones and point the remote to a required-project in ci in case they're using that locally | 19:08 |
| avass[m] | though that does not look like something that is run locally | 19:08 |
| mordred | required-projects from curl would be a great way to start spidering out into those sibling projects | 19:09 |
| mordred | "check out what you could do with a depends-on between curl and openssl..." | 19:09 |
| mnaser | yeah, i'm aiming to be like.. the very basic form | 19:10 |
| mnaser | and then once they've adopted slowly help them use some more features | 19:10 |
| avass[m] | ++ | 19:10 |
| mordred | ++ | 19:11 |
| mordred | I think we all see a bunch of clear opportunities for improvement based on their current scripts | 19:11 |
| mnaser | btw, just a heads up | 19:18 |
| mnaser | i dont have time to look into this now, but looks like a common thing on shutdown signal https://www.irccloud.com/pastebin/jtGaBjFu/ | 19:18 |
| opendevreview | Albin Vass proposed zuul/zuul master: Remove argument from gerritwatcher cancel https://review.opendev.org/c/zuul/zuul/+/795674 | 19:20 |
| mnaser | record time avass[m] :) | 19:20 |
| avass[m] | mnaser: ^ | 19:20 |
| mnaser | https://curl.zuul.vexxhost.dev/status | 19:20 |
| avass[m] | tristanC: maybe that's why gracefully stopping the scheduler wasn't working for you? | 19:21 |
| tristanC | avass[m]: perhaps, though we don't have the traceback | 19:30 |
| *** hashar has quit IRC | 19:39 | |
| mnaser | aw man, it's happening again | 19:47 |
| mnaser | https://www.irccloud.com/pastebin/26bSoijc/ | 19:48 |
| mnaser | > Error: Project github.com/vexxhost/zuul-base-jobs does not have the default branch master | 19:48 |
| mnaser | https://opendev.org/zuul/zuul/src/branch/master/zuul/executor/server.py#L1362-L1399 | 19:49 |
| mnaser | i guess i need to set `default-branch` | 19:51 |
| avass[m] | mnaser: heh, I started working on something and pretty much came up with the same config you did. except that I put package installation in a pre-run :) | 19:58 |
| mnaser | avass[m]: that actually sounds like an even better idea | 19:59 |
| mnaser | avass[m]: feel free to submit prs, itll run jobs just fine | 19:59 |
| avass[m] | also I put everything in zuul.d with playbooks in zuul.d/playbooks with a .zuul.ignore, in case they don't want to pollute their top-level directory | 20:00 |
| opendevreview | Merged zuul/zuul master: Fix unknown job detection https://review.opendev.org/c/zuul/zuul/+/795597 | 20:00 |
| avass[m] | sure | 20:00 |
| mnaser | avass[m]: if you wanna hack at this and i can start from your progress after you're done to avoid doing duplicate work? :P | 20:00 |
| avass[m] | I think you got further than me :) | 20:02 |
| avass[m] | and I'm gonna leave in a bit, it's getting late | 20:02 |
| avass[m] | I can pick it up tomorrow | 20:03 |
| mnaser | avass[m]: cool, ill update you with my progress :) | 20:03 |
| mnaser | ill borrow your ideas though :) | 20:03 |
| opendevreview | James E. Blair proposed zuul/zuul master: Add checkpoint release note https://review.opendev.org/c/zuul/zuul/+/795693 | 20:17 |
| corvus | zuul-maint: ^ can we also squeeze that in after avass's change so we have a release note for the commit? | 20:18 |
| corvus | s/commit/release/ | 20:21 |
| *** Shrews has quit IRC | 20:24 | |
| clarkb | I've approved it | 20:24 |
| corvus | thx! | 20:24 |
| opendevreview | Merged zuul/zuul master: Remove argument from gerritwatcher cancel https://review.opendev.org/c/zuul/zuul/+/795674 | 20:46 |
| opendevreview | Guillaume Chauvel proposed zuul/zuul master: WIP gitlab quick-start https://review.opendev.org/c/zuul/zuul/+/795540 | 21:28 |
| *** zenkuro has quit IRC | 21:30 | |
| *** zenkuro has joined #zuul | 21:31 | |
| opendevreview | Merged zuul/zuul master: Add checkpoint release note https://review.opendev.org/c/zuul/zuul/+/795693 | 21:40 |
| corvus | zuul-maint: commit 9bfdf43f49a952a0c2112258198e421ff7c9b501 (HEAD -> master, tag: 4.5.0, origin/master) | 21:43 |
| corvus | how's that look ? | 21:43 |
| clarkb | 9bfdf43f49a952a0c2112258198e421ff7c9b501 is the commit I see and 4.5.0 is the appropriate version. Did you want to restart opendev on that or is the delta small enough (just that bug fix really?) | 21:45 |
| corvus | a couple of bugfixes that opendev isn't running; but avass reported running the layout fix, so it's only the bugfix from today that has seen no production use | 21:46 |
| corvus | given that, i'm comfortable with a release now and opendev restart at our convenience, but would be happy to reverse that and do the normal process. | 21:47 |
| clarkb | I'm ok with just going for it. Worst case we do a 4.5.1 | 21:48 |
| corvus | ya | 21:48 |
| corvus | i'll wait a few mins for anyone else to double check that, then push it | 21:48 |
| fungi | yep, that looks sane to me. sorry my availability is unexpectedly spotty since my internet provider has been down for several hours now | 21:49 |
| corvus | okay, 4.5.0 pushed | 21:50 |
| fungi | thanks! | 21:51 |
| mnaser | curl update: i've got jobs running right now, it was all running perfectly until job timed out, bumped to 60 minutes and watching now :) | 22:03 |
| opendevreview | Guillaume Chauvel proposed zuul/zuul master: WIP gitlab quick-start https://review.opendev.org/c/zuul/zuul/+/795540 | 22:09 |
| avass[m] | mnaser: do you have any object storage available? Swift? | 22:10 |
| mnaser | avass[m]: yes we're publishing logs to swift right now | 22:12 |
| avass[m] | mnaser: cool, then caching may be an option to speed things up. Could be enough with a simple ccache that is passed around | 22:13 |
| mnaser | avass[m]: yeah or maybe just start by the simple stuff they used to cache before or something | 22:14 |
| avass[m] | That too :) | 22:14 |
| * mnaser admits they haven't done as much C in their life to know all about the build stuff :) | 22:14 | |
| avass[m] | I've done way too much C while working at an automotive company :) | 22:15 |
| fungi | i still think of c as a high-level programming language... scarred in my youth perhaps | 22:18 |
| *** tosky has quit IRC | 22:47 | |
| opendevreview | Guillaume Chauvel proposed zuul/zuul master: WIP gitlab quick-start https://review.opendev.org/c/zuul/zuul/+/795540 | 23:04 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!