| rbergeron | jlk: ay. "words" but i have to go take a package to ups before i get charged 800 dollars for not taking the package | 00:10 |
|---|---|---|
| rbergeron | like, i have a bunch of words -- but also they are related to v3 and not v4 of the api and ... that whole thing is a separate | 00:10 |
| rbergeron | um | 00:10 |
| rbergeron | event | 00:10 |
| rbergeron | bbiaw | 00:11 |
| rbergeron | lol | 00:11 |
| jlk | shoo | 00:11 |
| rbergeron | i love shoes! | 00:11 |
| jlk | SQUIRREL! | 00:11 |
| *** xinliang_ has joined #zuul | 03:40 | |
| xinliang_ | hi guys, I am running zuul 2.5.2 on arm64 ubuntu 16.04. I found that the zuul can read the gerrit event but it can't trigger the job. Any clue or how to debug this? | 03:44 |
| xinliang_ | I can't find any useful log in /var/log/zuul/ | 03:44 |
| tristanC | xinliang_: could you share the layout.yaml ? | 03:45 |
| xinliang_ | the job status is : build:noop-check-communication:debian-jessie 0 0 1 | 03:45 |
| tristanC | xinliang_: oh, and is zuul-merger working too? | 03:46 |
| xinliang_ | tristanC: sure. here is the layout.yml: https://github.com/xin3liang/project-config-linaro-ci/blob/master/zuul/layout.yaml | 03:47 |
| tristanC | xinliang_: does the noop-check-communication job exists? | 03:48 |
| xinliang_ | tristanC: yes, service zuul-merger status says it is running. | 03:49 |
| xinliang_ | The job noop-check-communication, I can see it the jenkins web page and I can run it manually | 03:49 |
| xinliang_ | it can run success | 03:50 |
| tristanC | xinliang_: hum... is the job stuck in "queued" ? | 03:52 |
| xinliang_ | jenkins's queue? no it can run success | 03:52 |
| tristanC | xinliang_: i mean in zuul status | 03:53 |
| xinliang_ | i add a patch to sandbox project and add recheck comment it has none response. https://review.openstack.org/#/c/477442/1 | 03:53 |
| xinliang_ | tristanC: how to check the zuul queue status? | 03:54 |
| tristanC | xinliang_: zuul-server has a webapp, listening on port 8001 by default, you can query its status by running curl http://localhost:8001/status.json | python -mjson.tool | 03:56 |
| xinliang_ | tristanC: it stuck in fetch the jason file. But 8001 port is listen. http://paste.openstack.org/show/613768/ | 04:00 |
| xinliang_ | here is the output of echo 'status' | nc 127.0.0.1 4730 -w 1 :http://paste.openstack.org/show/613766/ | 04:00 |
| xinliang_ | one strange thing is that i can telnet to gearman server it is stucked when running 'telnet 127.0.0.1 4730' | 04:02 |
| tristanC | xinliang_: to quit telnet, you may ^], then type quit | 04:09 |
| tristanC | xinliang_: without the status it's hard to tell what's going on, perhaps the zuul.log may help | 04:10 |
| xinliang_ | tristanC: I mean I can't telnet to the gearman server | 04:10 |
| tristanC | that's odd, well using nc is fine too | 04:11 |
| xinliang_ | tristanC: here is some of zuul.log: http://paste.openstack.org/show/613772/ | 04:14 |
| tristanC | xinliang_: the relevant part starts with "zuul.GerritSource: Updating <Change 0x..., review-number>", a complete run looks like: http://paste.openstack.org/show/613774/ | 04:17 |
| xinliang_ | So none of "GerritSource" in my zuul.log | 04:21 |
| xinliang_ | zuul is not running properly, I will check the config again | 04:23 |
| xinliang_ | tristanC: thanks :-) | 04:23 |
| tristanC | xinliang_: you're welcome :-) | 04:27 |
| *** jlk` has joined #zuul | 05:12 | |
| *** jlk has quit IRC | 05:13 | |
| *** EmilienM has quit IRC | 05:13 | |
| *** EmilienM has joined #zuul | 05:22 | |
| openstackgerrit | Tobias Henkel proposed openstack-infra/zuul feature/zuulv3: Support custom playbook locations https://review.openstack.org/477672 | 05:34 |
| *** ianw_ has joined #zuul | 05:36 | |
| *** ianw has quit IRC | 05:39 | |
| *** ianw_ is now known as ianw | 05:39 | |
| *** isaacb has joined #zuul | 05:39 | |
| *** isaacb has quit IRC | 06:14 | |
| *** isaacb has joined #zuul | 06:27 | |
| *** jeblair has quit IRC | 07:08 | |
| *** jeblair has joined #zuul | 07:09 | |
| xinliang_ | tristanC: now I can get the status of zuul: http://paste.openstack.org/show/613785/ | 07:20 |
| tristanC | xinliang_: nice, then you should try to recheck the test change and observe zuul.log and the status | 07:22 |
| xinliang_ | still no response to the gerrit event, it seems that zuul-merger is not working | 07:23 |
| tristanC | xinliang_: though, iirc, zuul-server should queue the change before asking zuul-merger to stage the change | 07:26 |
| tristanC | when a change is stuck in queued, it either mean it's waiting for the merger, or because there are no executor ready to run the job | 07:27 |
| xinliang_ | i can see something like: zuul.GerritSource: Updating <Change 0xffff30f24790 470823,3> but no other more | 07:27 |
| xinliang_ | tristanC: how to check executor then | 07:28 |
| tristanC | xinliang_: for jenkins, the job node label needs to be available | 07:29 |
| xinliang_ | it is idle now. it should be available | 07:30 |
| tristanC | on the jenkins job configuration page, at the "Label Expression" section it should say something like " | 07:31 |
| tristanC | Label is serviced by 1 node" | 07:31 |
| xinliang_ | tristanC: can't find "Label Expression" section | 07:57 |
| *** isaacb has quit IRC | 07:59 | |
| tristanC | xinliang_: did you use a "node" parameter for the job? | 08:00 |
| xinliang_ | yes, i can see | 08:01 |
| *** isaacb has joined #zuul | 08:02 | |
| xinliang_ | tristanC: I find the string: "Label debian-jessie is serviced by 1 node" | 08:03 |
| xinliang_ | in the job config | 08:03 |
| xinliang_ | i see the that merger is wait here: job = self.worker.getJob() | 08:07 |
| *** isaacb has quit IRC | 08:09 | |
| *** isaacb has joined #zuul | 08:15 | |
| *** hashar has joined #zuul | 08:20 | |
| *** smyers has quit IRC | 09:34 | |
| *** mgagne has quit IRC | 09:35 | |
| *** smyers has joined #zuul | 09:35 | |
| *** mgagne has joined #zuul | 09:37 | |
| *** mgagne is now known as Guest28796 | 09:37 | |
| xinliang_ | tristanC: there might be problem with my executor, i found that the jenkins service status is exited. | 09:58 |
| xinliang_ | with warning: http://paste.openstack.org/show/613814/ | 09:59 |
| xinliang_ | there are execptions on /var/log/jenkins/jenkins.log: http://paste.openstack.org/show/613815/ | 10:00 |
| xinliang_ | I should fix these problems | 10:01 |
| *** jkilpatr has joined #zuul | 10:57 | |
| *** dkranz has joined #zuul | 12:59 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul feature/zuulv3: DNM crappy followup patch https://review.openstack.org/478144 | 13:12 |
| *** maxamillion has joined #zuul | 13:26 | |
| maxamillion | rbergeron: o/ | 13:26 |
| Shrews | jeblair: mordred's patch there in 478144 ^^^ demonstrates that the websocket problem is likely autobahn related, as I also had suspected. | 13:39 |
| Shrews | my vote is to move away from autobahn | 13:39 |
| *** isaacb has quit IRC | 13:55 | |
| *** isaacb has joined #zuul | 14:13 | |
| *** hashar has quit IRC | 14:31 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul feature/zuulv3: Replace autobahn with aiohttp for websockets https://review.openstack.org/478144 | 14:38 |
| mordred | Shrews, jeblair: there is a new version of that patch that's more complete and correct - I definitely agree with Shrews about moving away from autobahn - there is too much magic it's doing that we do not need - the code with aiohttp is much more direct and to the point and it allows us to do the things we want to do | 14:40 |
| dmsimard | autobahn suddenly made me remember https://github.com/auchenberg/volkswagen | 14:41 |
| * dmsimard goes back to his cave | 14:41 | |
| mordred | dmsimard: hahahahahaa | 14:41 |
| jeblair | mordred: what was the reason you picked autobahn over aiohttp earlier? | 14:41 |
| dmsimard | mordred: oh man, zuul is totally not implemented in volkswagen that's a selling point. | 14:42 |
| mordred | jeblair: I'm not sure I picked it over aiohttp per-se as much as I picked it because it seemed to be a thing that had a mode that would work without eventlet or twisted | 14:43 |
| mordred | jeblair: it was more "this library works and is maintained" - I do not remember if I considered aiohttp at that time | 14:43 |
| jeblair | ok. aiohttp sounds good to me. | 14:44 |
| mordred | jeblair: yah - also - turns out this also answers the question "can you run aiohttp in its own thread" | 14:44 |
| mordred | jeblair: since, you know, that's now the reason that aiohttp is better than autobahn :) | 14:44 |
| clarkb | autobahn works with python2 and 3, and is actively maintained and I think is considered the golden standard for websockets in python land | 14:47 |
| mordred | yah - that's also likely why I started there - the python2 and python3 support | 14:48 |
| mordred | Shrews: want me to just squash the two patches? | 14:49 |
| jeblair | autobahn supports python2? what about the websocket stuff requires python3? | 14:49 |
| Shrews | mordred: yes plz | 14:50 |
| mordred | jeblair: eventlet | 14:50 |
| jeblair | ah | 14:50 |
| mordred | jeblair: it supports python2 via eventlet or twisted | 14:50 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul feature/zuulv3: WIP: Add web-based console log streaming https://review.openstack.org/463353 | 14:52 |
| mordred | Shrews: done | 14:52 |
| Shrews | mordred: awesome. now i can rework the test. thx | 14:52 |
| *** isaacb has quit IRC | 15:12 | |
| *** hashar has joined #zuul | 15:22 | |
| clarkb | Shrews: jeblair https://review.openstack.org/#/c/476677/1 is the parent of https://review.openstack.org/#/c/476678/1 which you ahve reviewed and approved but can't merge until the parent is reviewed and approved too. Would be great if you can take a look at the parent when you get a chance too | 17:11 |
| clarkb | jeblair: similar situation with https://review.openstack.org/#/c/476266/1 being the parent of the cleanup failed upload records change | 17:12 |
| jeblair | clarkb: done! | 17:27 |
| clarkb | tyty | 17:28 |
| openstackgerrit | Merged openstack-infra/nodepool feature/zuulv3: Use private attribute lookup interanlly in builder.py https://review.openstack.org/476677 | 17:30 |
| openstackgerrit | Merged openstack-infra/nodepool feature/zuulv3: Don't join image upload workers on stop() https://review.openstack.org/476678 | 17:30 |
| openstackgerrit | Merged openstack-infra/nodepool master: Disable CleanupWorker thread for test_image_upload_fail https://review.openstack.org/476266 | 17:31 |
| openstackgerrit | Merged openstack-infra/nodepool master: Cleanup failed upload records https://review.openstack.org/476223 | 17:31 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/zuul feature/zuulv3: WIP: Add web-based console log streaming https://review.openstack.org/463353 | 17:34 |
| Shrews | jeblair: https://review.openstack.org/475505 is a follow up to your comments on the parent (which could use a +A) | 17:36 |
| *** jlk` is now known as jlk | 17:43 | |
| *** jlk has quit IRC | 17:44 | |
| *** jlk has joined #zuul | 17:44 | |
| * SpamapS didn't know what autobahn was until today | 17:48 | |
| SpamapS | mordred: I think you meant the opposite: "websocket streaming will be switching to autobahn" | 18:11 |
| mordred | no, I do not | 18:11 |
| mordred | SpamapS: autobahn does not work for us | 18:12 |
| mordred | SpamapS: aiohttp does | 18:12 |
| *** Shuo has joined #zuul | 18:12 | |
| mordred | SpamapS: autobahn layers on too much extra goo that we do not need that actually gets in the way of us being able to do the things we need to do - this is understandable, since they support asyncio, eventlet and twisted backends so they obviously need to provide a compat layer | 18:13 |
| *** Shuo has quit IRC | 18:14 | |
| SpamapS | mordred: so wouldn't that mean that we want to switch to *aiohttp* | 18:14 |
| *** Shuo has joined #zuul | 18:14 | |
| mordred | SpamapS: yes. | 18:14 |
| SpamapS | you wrote the other thing | 18:14 |
| mordred | SpamapS: wait - did I say what you quoted? | 18:14 |
| mordred | dangit | 18:14 |
| mordred | thank you | 18:14 |
| SpamapS | np | 18:14 |
| SpamapS | two words, same start letter, almost same # of chars. | 18:15 |
| mordred | ++ | 18:15 |
| *** Shuo_ has joined #zuul | 18:15 | |
| jlk | o/ | 18:16 |
| mordred | SpamapS: fwiw, https://review.openstack.org/#/c/463353/19/zuul/streamer.py is the autobahn version and https://review.openstack.org/#/c/463353/21/zuul/web.py is the aiohttp version if you wanna see the differences | 18:16 |
| SpamapS | mordred: I like the aiohttp version. Reminds me of the tokio futures stuff I"ve been poking at in Rust. | 18:18 |
| clarkb | might help reviews if changes went into streamer.py then file was renamed at end? | 18:18 |
| *** Shuo has quit IRC | 18:19 | |
| mordred | clarkb: the patch originally created zuul/streamer.py and doens't have any previous reviews on streamer.py - is there a reviewing use-case I'm missing? | 18:21 |
| clarkb | mordred: oh for some reason I thought it had merged already but tahts the same change just in different patchsets. That said you could do inter patchset diffs more easily that way | 18:22 |
| clarkb | I don't know that it matters since its not merged yet | 18:22 |
| mordred | yah - I think if there had been reviews on the old versions that would be more important | 18:22 |
| mordred | SpamapS: I agree - it felt much nicer to write/hack on too | 18:23 |
| Shrews | clarkb: i would highly recommend NOT reviewing it yet. lots of changes needed still | 18:23 |
| mordred | SpamapS: the flow was very understanable, as opposed to the callback methods on the service class from the autobahn approach where you didn't really know what was going on between callbacks | 18:24 |
| Shuo_ | pabelanger: what is the status of ansible-role-zuul? could I use it as zuul demo environment setup? | 18:33 |
| mordred | Shuo_: pabelanger is on vacation - I'm not sure ansible-role-zuul is up to date as of today though | 18:39 |
| clarkb | Shuo_: mordred left some comments | 18:40 |
| clarkb | er shrews ^ | 18:41 |
| Shuo_ | mordred: and clarkb: thanks for providing the info. Right now, I am wondering that to build a demo-ish cluster, how easy/hard it would be to a perosn not familiar with Zuul? | 18:41 |
| jeblair | Shuo_: i think at this point it would still be hard, but possible. | 18:42 |
| jeblair | Shuo_: we're just working on the documentation now, it should be merging soon. | 18:43 |
| jeblair | Shuo_: i think it will be a lot better in a few weeks when that's done and pabelanger is back and able to update ansible-role-zuul | 18:43 |
| *** pbrobinson has joined #zuul | 18:49 | |
| *** Shuo_ has quit IRC | 18:55 | |
| *** Shuo has joined #zuul | 19:05 | |
| Shuo | jeblair: Regarding the last message I saw from James "it would be hard, but not impossible", my goal is not to be too hand-off from the community (we could work under guidance) | 19:07 |
| Shuo | jeblair: "i think it will be a lot better in a few weeks when that's done and pabelanger is back and able to update ansible-role-zuul" -- we are totally fine for waiting several weeks. | 19:10 |
| *** mmedvede has quit IRC | 19:25 | |
| *** mmedvede has joined #zuul | 19:26 | |
| *** mmedvede has quit IRC | 19:28 | |
| *** mmedvede has joined #zuul | 19:31 | |
| *** Shuo has quit IRC | 19:31 | |
| *** mmedvede has quit IRC | 19:42 | |
| *** mmedvede has joined #zuul | 19:42 | |
| *** rfolco_ has joined #zuul | 19:55 | |
| *** jkilpatr has quit IRC | 20:12 | |
| openstackgerrit | David Shrewsbury proposed openstack-infra/zuul feature/zuulv3: WIP: Add web-based console log streaming https://review.openstack.org/463353 | 20:14 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/zuul feature/zuulv3: WIP: Add web-based console log streaming https://review.openstack.org/463353 | 20:19 |
| *** rfolco_ is now known as rfolco | 20:19 | |
| Shrews | mordred: ok, just need to implement the gearman/finger portion now. that new test properly fetches the "Test test" artificial response. | 20:21 |
| mordred | Shrews: woot! | 20:26 |
| *** jkilpatr has joined #zuul | 20:31 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Add run-bindep role and add it to unittests pre https://review.openstack.org/478264 | 20:32 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Port in tox jobs from openstack-zuul-jobs https://review.openstack.org/478265 | 20:32 |
| *** rcarrill1 has joined #zuul | 20:33 | |
| *** rcarrillocruz has quit IRC | 20:34 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Port in tox jobs from openstack-zuul-jobs https://review.openstack.org/478265 | 21:12 |
| *** jkilpatr has quit IRC | 21:18 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Port in tox jobs from openstack-zuul-jobs https://review.openstack.org/478265 | 21:24 |
| *** hashar has quit IRC | 21:25 | |
| *** jkilpatr has joined #zuul | 21:31 | |
| *** maxamillion has left #zuul | 21:48 | |
| *** Shuo has joined #zuul | 21:50 | |
| *** jkilpatr has quit IRC | 21:53 | |
| *** jkilpatr has joined #zuul | 21:53 | |
| mordred | jeblair: I think we want to land https://review.openstack.org/#/c/477672 before we get too much further with jobs | 22:10 |
| jeblair | mordred: fair enough | 22:11 |
| jeblair | mordred: +3 | 22:12 |
| jlk | GOOD NEWS | 22:15 |
| jlk | github search lets me search for: "Depends-On" z8s-sandbox/pull/2 type:pr in:body | 22:15 |
| jlk | and it's only returning the PR with that in the body, not in a comment. | 22:15 |
| jlk | now, time for an office party. | 22:16 |
| mordred | jlk: woot! | 22:16 |
| jeblair | \o/ | 22:17 |
| jeblair | did someone (pabelanger?) push up a change with an 'add host to inventory' task? | 22:17 |
| jeblair | i could have sworn i saw that but can't find it | 22:18 |
| mordred | jeblair: I do not knkow - do we need one of those? | 22:18 |
| mordred | or do you need an example? | 22:18 |
| mordred | oh - this is for publishing the logs | 22:18 |
| jeblair | mordred: yeah, i think we need to do that to add the logserver to the inventory to use synchronize | 22:18 |
| mordred | jeblair: http://docs.ansible.com/ansible/add_host_module.html is what you want | 22:19 |
| openstackgerrit | Merged openstack-infra/zuul feature/zuulv3: Support custom playbook locations https://review.openstack.org/477672 | 22:19 |
| jeblair | i think that should be safe in the base post playbook, since nothing will run after it | 22:19 |
| mordred | jeblair: it's only valid for the given playbook invocation | 22:19 |
| mordred | jeblair: so it's safe regardless | 22:19 |
| jeblair | mordred: oh neat | 22:19 |
| mordred | it only edits the in-memory copy of the inventory | 22:19 |
| jeblair | cause i had ideas swirling in my head about accidentally running something on logs.o.o | 22:20 |
| jeblair | mordred: a role that copies stuff from all hosts to executor seems straightforward... | 22:34 |
| jeblair | mordred: a role that copies from executor to log hosts seems less so, since i want the remote host to be logs.o.o. that sounds like a second play in the post playbook. does that sound right to you? | 22:34 |
| jeblair | mordred: i guess it can be a second play with a role | 22:35 |
| jeblair | https://etherpad.openstack.org/p/WWUw5wHlEY | 22:36 |
| jeblair | something like that? | 22:36 |
| jeblair | 22:44 < openstackgerrit> James E. Blair proposed openstack-infra/openstack-zuul-roles master: Add collect and upload log roles https://review.openstack.org/478311 | 22:45 |
| jeblair | 22:45 < openstackgerrit> James E. Blair proposed openstack-infra/openstack-zuul-jobs master: Upload logs to logs.openstack.org https://review.openstack.org/478312 | 22:45 |
| jeblair | i'm going to need to move that into project-config to make it trusted, but i wanted to put it out there | 22:46 |
| jeblair | 22:58 < openstackgerrit> James E. Blair proposed openstack-infra/project-config master: Move base job to project-config https://review.openstack.org/478313 | 22:59 |
| jeblair | 22:59 < openstackgerrit> James E. Blair proposed openstack-infra/openstack-zuul-jobs master: Remove base job https://review.openstack.org/478315 | 22:59 |
| jeblair | SpamapS: would you please see my comments on PS1 of https://review.openstack.org/477168 for background on the py2/py3 aspect of that script? | 23:05 |
| SpamapS | jeblair: ACK, looking | 23:07 |
| SpamapS | I need like, a little urwid pong game to play when my gertty gets slow syncing. | 23:09 |
| jeblair | we could at least add a spinning cursor | 23:09 |
| jeblair | /-\|/-\|/-\|/-\|/-\|/-\|/-\|/-\|/-\|/-\| | 23:10 |
| SpamapS | jeblair: hah | 23:11 |
| SpamapS | jeblair: so, py2.. IMO we should just abandon it. We can likely make encrypt_secret work with python 3.4, which gets us RHEL6/CentOS6/Ubuntu 14.04. | 23:13 |
| SpamapS | anything older and I start to wonder about openssl versions ;) | 23:14 |
| SpamapS | Otherwise we need to devise a way to test encrypt_secret. :-P | 23:15 |
| clarkb | SpamapS: the aiohttp code is already python3.5 only (it uses await) | 23:15 |
| openstackgerrit | Merged openstack-infra/zuul feature/zuulv3: requirements: update ansible version https://review.openstack.org/475228 | 23:16 |
| SpamapS | clarkb: right, but encrypt_secret is a CLI tool for zuul end-users. | 23:16 |
| SpamapS | So it makes sense for it to have a little bit wider support matrix. | 23:17 |
| clarkb | maybe? it is pretty trivial to run the encyption on a modern install then deploy to eg rhel6 if you are still stuck on it | 23:17 |
| jeblair | SpamapS: it does already work with py2. i can test the result after the py3 update and make sure it still does. after that, i don't expect it to change (at least, not in a way that would trouble me about py2/3 compatibility) | 23:18 |
| clarkb | obviously if it just works then great. It just seems like a lot of effort for rhel6 which is really old now | 23:18 |
| jeblair | SpamapS: so it seems like py2 for this is pretty low cost. i agree, we're pretty darn close to the point where it's not needed though. | 23:18 |
| SpamapS | The cost to us is only in maintaining some kind of tests. | 23:19 |
| SpamapS | I don't feel that strongly about dropping py2 for it. | 23:19 |
| SpamapS | But I feel a little more strongly that it should grow tests so we don't break it. | 23:19 |
| jeblair | i'm happy to have py3 but not py2 tests for it. | 23:20 |
| SpamapS | and I was kind of enjoying being py3 only for a minute ;) | 23:20 |
| SpamapS | that's an interesting idea. Make a best effort to be py2, but only really support py3. | 23:21 |
| SpamapS | (sounds like the way py3 has been treated for the last 5 years ;) | 23:21 |
| jeblair | turnabout is fair play | 23:21 |
| SpamapS | py3 has a secret: py3 is not left handed | 23:21 |
| jeblair | rust is not left handed either? | 23:22 |
| SpamapS | rust is the immune to iocaine. | 23:22 |
| SpamapS | s/the // | 23:22 |
| jeblair | SpamapS, mordred: i note that you both write ansible task names without trailing periods | 23:30 |
| jeblair | (i know, i know, bear with me) | 23:30 |
| jeblair | SpamapS, mordred: i think pabelanger had training periods. | 23:31 |
| jeblair | should we adopt the non-training period style? | 23:31 |
| jeblair | (as i think about it, these are most similar to long entries, which usually don't have periods) | 23:32 |
| jeblair | er log entries | 23:32 |
| jeblair | mordred: 478311 478313 478315 (zuul error is expected) are the job config changes needed to get log publishing. | 23:37 |
| jeblair | after we land those, i'll try restarting the executor and see what's broke. :) | 23:38 |
| openstackgerrit | Merged openstack-infra/zuul feature/zuulv3: Add log streaming test https://review.openstack.org/471079 | 23:46 |
| openstackgerrit | Merged openstack-infra/zuul feature/zuulv3: Remove py2 SocketServer import https://review.openstack.org/475399 | 23:47 |
| *** Shuo has quit IRC | 23:48 | |
| openstackgerrit | Merged openstack-infra/zuul feature/zuulv3: Use Event in streaming test https://review.openstack.org/475505 | 23:48 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!