| *** pwhalen has quit IRC | 01:48 | |
| *** pwhalen has joined #zuul | 01:50 | |
| *** pwhalen has quit IRC | 01:50 | |
| *** pwhalen has joined #zuul | 01:50 | |
| pabelanger | Shrews: one for the morning, apparently fingergw.listen_address = 127.0.0.1 won't work for zuul.conf, get the following traceback: http://paste.openstack.org/show/694290/ | 02:20 |
|---|---|---|
| pabelanger | fingergw.listen_address = localhost does work | 02:21 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool master: Rename get_schema to getSchema https://review.openstack.org/550670 | 02:44 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool master: Extend config API with getSupportedLabels() https://review.openstack.org/550671 | 02:44 |
| Shrews | pabelanger: maybe that has something to do with whether ipv6 is being used or not | 02:45 |
| Shrews | if your host is ipv6 enabled, you should probably configure with ipv6 addresses | 02:45 |
| Shrews | clarkb: 550671 implements that config API | 02:46 |
| *** harlowja has quit IRC | 02:49 | |
| openstackgerrit | Merged openstack-infra/nodepool master: Store label info with launcher registration https://review.openstack.org/548376 | 03:01 |
| *** rlandy|bbl is now known as rlandy | 04:03 | |
| *** harlowja has joined #zuul | 04:37 | |
| *** jappleii__ has joined #zuul | 04:42 | |
| *** jappleii__ has quit IRC | 04:43 | |
| *** threestrands_ has quit IRC | 04:44 | |
| *** _ari_ has quit IRC | 04:57 | |
| *** weshay|PTO has quit IRC | 04:58 | |
| *** pabelanger has quit IRC | 04:58 | |
| *** myoung|off has quit IRC | 04:59 | |
| *** threestrands has joined #zuul | 05:00 | |
| *** threestrands has quit IRC | 05:00 | |
| *** threestrands has joined #zuul | 05:00 | |
| *** threestrands has quit IRC | 05:25 | |
| *** swest has joined #zuul | 05:51 | |
| *** snapiri has joined #zuul | 05:51 | |
| *** swest has quit IRC | 06:05 | |
| *** harlowja has quit IRC | 06:05 | |
| *** swest has joined #zuul | 06:05 | |
| *** swest has quit IRC | 06:08 | |
| *** swest has joined #zuul | 06:08 | |
| *** jaianshu has joined #zuul | 06:13 | |
| openstackgerrit | Tobias Henkel proposed openstack-infra/zuul master: Add queue size to tenant overview https://review.openstack.org/548248 | 06:27 |
| tobiash | mordred: that works ^ | 06:28 |
| openstackgerrit | Tobias Henkel proposed openstack-infra/zuul master: Fix console stream colors https://review.openstack.org/550730 | 07:26 |
| tobiash | mordred: that fixes the streaming colors ^ | 07:26 |
| *** hashar has joined #zuul | 07:47 | |
| *** jaianshu_ has joined #zuul | 09:16 | |
| *** jaianshu has quit IRC | 09:19 | |
| *** weshay has joined #zuul | 09:40 | |
| *** _ari_ has joined #zuul | 09:40 | |
| *** pabelanger has joined #zuul | 09:41 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: configloader: check for unknown shadow projects https://review.openstack.org/550763 | 09:56 |
| openstackgerrit | Merged openstack-infra/zuul-jobs master: Add known hosts for all users https://review.openstack.org/548642 | 10:05 |
| *** jpena|off is now known as jpena | 10:05 | |
| *** dmellado has quit IRC | 10:08 | |
| *** dmellado has joined #zuul | 10:13 | |
| *** jpena is now known as jpena|off | 10:19 | |
| *** electrofelix has joined #zuul | 10:28 | |
| *** jaianshu__ has joined #zuul | 10:30 | |
| *** jaianshu_ has quit IRC | 10:34 | |
| *** pabelanger_ has joined #zuul | 12:27 | |
| *** _ari__ has joined #zuul | 12:28 | |
| *** weshay_ has joined #zuul | 12:29 | |
| *** pabelanger has quit IRC | 12:29 | |
| *** _ari_ has quit IRC | 12:30 | |
| *** weshay has quit IRC | 12:31 | |
| *** jaianshu__ has quit IRC | 12:35 | |
| *** pabelanger_ is now known as pabelanger | 13:10 | |
| pabelanger | Shrews: not sure about ipv6, need to test again. But I would still think 127.0.0.1 works for a loopback device, if you only wanted ipv4. I've used localhost for now in my testing, but I can dive more into it | 13:12 |
| *** mickt has joined #zuul | 13:18 | |
| *** rlandy has joined #zuul | 13:34 | |
| pabelanger | http://paste.openstack.org/show/695224/ | 13:52 |
| pabelanger | another bug in our callback module for logging | 13:52 |
| mordred | pabelanger: if you hav ea sec, mind +Aing https://review.openstack.org/#/c/550730/ ? | 14:12 |
| pabelanger | sure +3 | 14:12 |
| *** snapiri has quit IRC | 14:16 | |
| mordred | thanks! | 14:17 |
| mrhillsman | any thoughts on collapsible pipelines? | 14:17 |
| mrhillsman | zuul-web that is | 14:18 |
| mordred | mrhillsman: on the status page? | 14:19 |
| mrhillsman | yeah | 14:20 |
| mrhillsman | ok if there is nothing in the works or if it has even been considered | 14:20 |
| mrhillsman | just wondering | 14:20 |
| mordred | seems like a good idea to me - although I think step one is refactor the status page to use angular instead of jquery so it matches the rest of the things (which hopefully should make it easier to hide the queue labels when filtering to a project too) | 14:21 |
| mrhillsman | we have quite a few right now and i can foresee it getting out of hand pretty quickly | 14:21 |
| mrhillsman | ok cool | 14:21 |
| mordred | mrhillsman: basically, I think now that the javascript stack has landed and we have previews on the build job, we're ina good position to do a bunch of refactoring/improvements | 14:21 |
| openstackgerrit | Merged openstack-infra/nodepool master: Rename get_schema to getSchema https://review.openstack.org/550670 | 14:22 |
| mordred | mrhillsman: speaking of - don't know if you've seen or not, but if you click on build-javascript-content, like on https://review.openstack.org/#/c/548248 - it shows a live preview of the zuul web content | 14:22 |
| mrhillsman | clickety click | 14:22 |
| openstackgerrit | Merged openstack-infra/nodepool master: Extend config API with getSupportedLabels() https://review.openstack.org/550671 | 14:24 |
| mrhillsman | so you know i think for this one team we are talking i think queue is pipeline for them | 14:24 |
| electrofelix | Anyone working on the nodepool plugin for Jenkins? | 14:27 |
| pabelanger | electrofelix: There has been discussion in channel, but unsure if anybody is actively working on it | 14:28 |
| pabelanger | electrofelix: even a recently ML post | 14:29 |
| electrofelix | yeah, I thought someone mentioned someone being interested in picking it up, just can't recall who | 14:29 |
| electrofelix | pabelanger: that was me, I was looking at the zuul trigger plugin | 14:29 |
| pabelanger | Ah! | 14:29 |
| electrofelix | but worked out should really help with the nodepool one first for us to migrate | 14:30 |
| pabelanger | electrofelix: I mean, I know some downstream people that are looking to use nodepool with jenkins still, but don't believe they are on IRC ATM | 14:30 |
| * Shrews deletes his ansible slack server from weechat forever | 14:33 | |
| electrofelix | well hopefully they'll see the messages today or I'll mentioned it again tomorrow (or if I can find the original discussion in eavesdrop). | 14:33 |
| pabelanger | electrofelix: agree, it would be great to get interested parties together and work on it. I'll keep an eye out | 14:33 |
| electrofelix | ah, finally found the discussion (thanks for channel logs) | 14:34 |
| electrofelix | http://eavesdrop.openstack.org/irclogs/%23zuul/%23zuul.2018-02-02.log.html | 14:34 |
| electrofelix | hughsaunders: ping me when your around | 14:35 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: configloader: check for unknown shadow projects https://review.openstack.org/550763 | 14:42 |
| *** mickt has quit IRC | 14:46 | |
| electrofelix | Would it make sense to host both zuul-trigger and nodepool plugins for jenkins on review.openstack.org? | 14:49 |
| electrofelix | similar to the gearman plugin | 14:49 |
| openstackgerrit | Merged openstack-infra/zuul master: Normalize hostname for statsd https://review.openstack.org/547309 | 14:50 |
| mordred | electrofelix: I don't see any reason why not - seems like it would enable making some integration tests too | 14:51 |
| clarkb | electrofelix: reading the email I'm a little confused why you would use grpc instead of gearman since gearman is the existing zuul run a job protocol? | 14:52 |
| clarkb | and jenkins has a gearman plugin that you could probably modify relatively easily to support the current gearman run me a job behavior? | 14:53 |
| mordred | clarkb: probably based on line 77 of https://etherpad.openstack.org/p/zuulv3-jenkins-integration | 14:54 |
| electrofelix | clarkb: I think the assumption was that didn't necessarily work well for providing access to the console output | 14:54 |
| electrofelix | but I'm not attached to doing it that way | 14:55 |
| mordred | yah - gRPC having support for streaming channels should allow the trigger plugin to provide log streaming | 14:55 |
| clarkb | but jenkins already does log strearming | 14:55 |
| clarkb | so you just point at the jenkins log streaming? | 14:55 |
| clarkb | (its a bit convoluted in that its reimplementing http partial gets using a post protcol they invented themselves but it works | 14:56 |
| mordred | clarkb: right - but I think the idea here is to keep the interface as zuul and have jenkins be an impl detail in the ansible playbooks | 14:56 |
| clarkb | mordred: ya zuul would get the jenkins log streaming | 14:56 |
| clarkb | using the existing built in log streaming in jenkins | 14:56 |
| clarkb | so rather than finger to the executors its http post to jenkins | 14:56 |
| mordred | clarkb: sure. I mean - whatever works for people | 14:56 |
| mordred | also - gearman would require the existence of a gearman server running that job content to talk to | 14:58 |
| clarkb | which zuul starts for you | 14:58 |
| mordred | you do not want playbooks in jobs to be able to submit jobs to the zuul gearman | 14:59 |
| clarkb | I'm not sure I understand why playbooks would be involved? | 14:59 |
| mordred | the zuul-side impl of this is a base job, not a zuul plugin | 14:59 |
| clarkb | oohhhhh | 14:59 |
| mordred | zuul itself won't know anyhting about jenkins | 14:59 |
| tobiash | that's neat | 14:59 |
| mordred | sorry - maybe should have started with that :) | 14:59 |
| clarkb | in that case I think I would still use gearman simply beacuse its part of the existing technoligy being used | 14:59 |
| clarkb | and jenkins has a plugin for it | 15:00 |
| mordred | yah - except we're also working towards making it go away | 15:00 |
| clarkb | mordred: but also not replacing it with grpc as far as I know? | 15:00 |
| mordred | (and I'd like to add a gRPC api to zuul as well) | 15:00 |
| mordred | clarkb: no - not replacing it with grpc - and a grpc api may never materialize | 15:00 |
| electrofelix | in any case I think basically my initial assessment came down to, work on the nodepool plugin first, then look closer what's needed for the zuul trigger plugin | 15:01 |
| mordred | but once it's not there, I think adding the requirement that someone run a gearman server in order to hook up zuul playbooks to a jenkins server is a weird requirement | 15:01 |
| clarkb | fwiw my concern here is having jammed 0mq into places in the past being a terrible decision | 15:01 |
| clarkb | and I think that using grpc to use grpc feels similar especially when existing tooling already does the job | 15:01 |
| mordred | clarkb: I guess it's just that I donmt think existing tooling does the job - someone has to spin up a gearman server and run it | 15:01 |
| mordred | whereas the grpc listener is supported well in javaand can just be a listener thread in jenkis itself | 15:02 |
| clarkb | mordred: ya I had missed that you weren't replacing the executors but running below them | 15:02 |
| mordred | it's not a use-it-just-to-use-it sort of thing. it could also be 'make a rest listerner' | 15:02 |
| clarkb | honestly in that case I still don't think I would grpc I would just jenkins rest api to trigger job and then use the console streaming api | 15:02 |
| mordred | except jenkins frequently breaks their rest api :) | 15:03 |
| clarkb | so your ansible job would be trigger job, get back console streaming url, then while job is running: http post to get log and echo back through executor | 15:03 |
| mordred | clarkb: but yah - that would potentially work too - I think the thing is the job trigger api isn't going to be sufficient, because you need a trigger, then a wait callback, then a continue | 15:03 |
| mordred | so _something_ in jenkins is going to have to be listening for a something in th emiddle of an execution flow for a pause/continue | 15:04 |
| electrofelix | you know some of that may mean a plugin in zuul for jenkins to avoid streaming the console unless someone is looking at it might be better for network usage | 15:04 |
| clarkb | mordred: so that you can cancel jobs? | 15:06 |
| mordred | clarkb: no - so that zuul can push the git repos to the nodes before jenkins actually runs the content | 15:06 |
| mordred | clarkb: I've got a flow written up on that etherpad | 15:06 |
| clarkb | this is quickly beginning to seem like far more work than its worth >_> but I'm also not sure what the motiviation is for using jenkins and zuul v3 together | 15:08 |
| mordred | clarkb: migratoin | 15:08 |
| clarkb | so you can mostly keep your old job configs? | 15:08 |
| mordred | clarkb: for people currently running a large jenkins install, doing a big-bang migration is unlikely | 15:08 |
| clarkb | I guess that does help migrating | 15:08 |
| mordred | also - some people have a ton of jobs they'd totally migrate, but one or two super compex pipelines they need to keep for $reasons | 15:09 |
| mordred | basically - taking a page from ansible/puppet - ansiblenot requiring us to rework everything just to start using ansible was super useful - even though dehann originally didn't want the puppet module in ansible because he didn't want people to think ansible couldn't handle config management | 15:10 |
| clarkb | ya, I think one difference here though is for ebtter or worse zuul has sort of been designed to not work the way jenkins works (due to trouble we've had with jenkins) so some of these interactions get tricky | 15:11 |
| mordred | yup. totally agree | 15:11 |
| mordred | it's not going to be a one-day task | 15:11 |
| openstackgerrit | Merged openstack-infra/zuul master: Fix console stream colors https://review.openstack.org/550730 | 15:11 |
| clarkb | other than zuul going back to running a git repo server I can't think of any good alternatives to a pause/resume job interrupt functionality | 15:13 |
| clarkb | and I think we've largely said no git repo server | 15:13 |
| clarkb | super cheap method to prove out a system quickly may be to have a job step that just polls a disk location and zuul can touch that when its done | 15:14 |
| clarkb | doubt this would be reliable enough for long term use | 15:14 |
| clarkb | Shrews: I didn't get a chance to review the get supported labels change before it merged but looking at it now it looks great. Thanks | 15:21 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: Update jquery to version 3 https://review.openstack.org/550871 | 15:29 |
| *** electrofelix has quit IRC | 15:52 | |
| openstackgerrit | Merged openstack-infra/zuul master: Update jquery to version 3 https://review.openstack.org/550871 | 16:11 |
| *** harlowja has joined #zuul | 16:54 | |
| *** hashar is now known as hasharDinner | 17:13 | |
| *** harlowja has quit IRC | 18:26 | |
| mordred | tristanC: I just left a comment on https://review.openstack.org/#/c/535545 - I'll help with it soon if you don't get to it first though | 18:38 |
| *** sshnaidm|ptg has joined #zuul | 18:53 | |
| *** JasonCL has joined #zuul | 19:03 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: web: add /{tenant}/pipelines route https://review.openstack.org/541521 | 19:03 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: web: add /{tenant}/jobs.html page to display jobs https://review.openstack.org/535545 | 19:03 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: web: add /{tenant}/projects web page https://review.openstack.org/537870 | 19:03 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: Fix indentation and renable the eslint rule https://review.openstack.org/545671 | 19:03 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: web: add /{tenant}/jobs/{job_name} route https://review.openstack.org/550978 | 19:03 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: web: add /{tenant}/projects routes https://review.openstack.org/550979 | 19:03 |
| mordred | tristanC: ^^ there - did the split | 19:03 |
| openstackgerrit | Merged openstack-infra/zuul-jobs master: Parameterize the test-setup role https://review.openstack.org/510907 | 19:08 |
| *** harlowja has joined #zuul | 19:11 | |
| openstackgerrit | Tobias Henkel proposed openstack-infra/nodepool master: DNM: dib test ca-certificates-java https://review.openstack.org/550982 | 19:12 |
| *** harlowja has quit IRC | 19:18 | |
| mordred | tobiash, clarkb, pabelanger: I think we're good to go on topic:javascript up through and including https://review.openstack.org/#/c/545671 - I know corvus wanted to discuss the jobs graph and pipeline rending changes more | 19:29 |
| mordred | but if we can land through https://review.openstack.org/#/c/541521/ and get a zuul-web restart in place, it should let us look at https://review.openstack.org/#/c/535545/ https://review.openstack.org/#/c/537870/14 and https://review.openstack.org/#/c/545671/ - as well as the two final rendering patches | 19:31 |
| clarkb | I feel like I'm the wrong person to review anything called topic:javascript :) | 19:31 |
| *** harlowja has joined #zuul | 19:33 | |
| tobiash | clarkb: you could at least review the routes | 19:35 |
| tobiash | they're python ;) | 19:36 |
| *** _ari__ is now known as _ari_ | 19:36 | |
| clarkb | ya I'll try to fumble my way through it this afternoon | 19:37 |
| clarkb | js is simple enough that I mostly grok it, but it is also foreign enough that I'm like wat am I doing :) | 19:37 |
| tobiash | mordred: you probably should have named that stack python-and-a-little-bit-of-javascript :-P | 19:37 |
| mordred | tobiash: :) | 19:38 |
| mordred | I mostly want to get landed up through the "fix indentation" patch - because I hate having anything like a mass-indentation patch lurking around for too long | 19:39 |
| *** elyezer has joined #zuul | 19:40 | |
| tobiash | mordred: comment on https://review.openstack.org/#/c/550979/1 | 19:42 |
| mordred | tobiash: golly. I have no idea! | 19:44 |
| * mordred goes to look | 19:44 | |
| tobiash | mordred: +2 with comment on the commit message: https://review.openstack.org/#/c/537870/14 | 19:51 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: web: add /{tenant}/projects routes https://review.openstack.org/550979 | 19:55 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: web: add /{tenant}/pipelines route https://review.openstack.org/541521 | 19:55 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: web: add /{tenant}/jobs.html page to display jobs https://review.openstack.org/535545 | 19:55 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: web: add /{tenant}/projects web page https://review.openstack.org/537870 | 19:55 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: Fix indentation and renable the eslint rule https://review.openstack.org/545671 | 19:55 |
| mordred | tobiash: good call on 550979 ... and darn, I missedyour comment on the other before I pushed ... | 19:56 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: web: add /{tenant}/projects web page https://review.openstack.org/537870 | 19:57 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: Fix indentation and renable the eslint rule https://review.openstack.org/545671 | 19:57 |
| pabelanger | Shrews: yah, so it seems using an IPv4 fingergw.listen_address on ipv6 hosts, are an issue: http://logs.openstack.org/34/550834/5/check/windmill-ubuntu-xenial/aeef056/logs/zf01/var/log/journal/zuul-fingergw.service.log as exception | 19:58 |
| pabelanger | is there some sort of hostname lookup happening there? | 19:58 |
| pabelanger | zuul.conf FYI: http://logs.openstack.org/34/550834/5/check/windmill-ubuntu-xenial/aeef056/logs/zf01/etc/zuul/zuul.conf | 19:59 |
| tobiash | mordred: +2 on 550979 | 20:00 |
| mordred | \o/ | 20:01 |
| tobiash | mordred: so the stack is +2 from me | 20:02 |
| mordred | tobiash: sweet. thanks for the reviews! | 20:02 |
| tobiash | np | 20:02 |
| openstackgerrit | Tobias Henkel proposed openstack-infra/nodepool master: DNM: dib test ca-certificates-with-fix https://review.openstack.org/550990 | 20:05 |
| Shrews | pabelanger: no. we do set the address family to AF_INET6 for the server. not sure if we need to do something different to support both at once | 20:07 |
| *** sshnaidm|ptg is now known as sshnaidm|afk | 20:08 | |
| pabelanger | Shrews: yah, let me see if I can figure it out now that I can reproduce the failure | 20:09 |
| *** JasonCL has quit IRC | 20:10 | |
| Shrews | pabelanger: did you try 0.0.0.0? | 20:10 |
| *** JasonCL has joined #zuul | 20:10 | |
| Shrews | or '' | 20:10 |
| *** JasonCL has quit IRC | 20:11 | |
| Shrews | i think v6 and v4 should be supported, but 127.0.0.1 might be "special" | 20:11 |
| pabelanger | Shrews: yah, I can test. I think that worked, but I was looking to bind to a specific interface for testing reasons. 0.0.0.0 will bind to all ipv4 interfances | 20:13 |
| pabelanger | interfaces* | 20:13 |
| pabelanger | Shrews: https://bugs.python.org/issue20215 | 20:14 |
| pabelanger | that implies ipv6 isn't supported | 20:15 |
| Shrews | pabelanger: that's 4 years old | 20:15 |
| Shrews | v6 is definitely supported | 20:15 |
| Shrews | pabelanger: https://review.openstack.org/502137 | 20:16 |
| *** JasonCL has joined #zuul | 20:16 | |
| Shrews | pabelanger: that change was necessary b/c the tests failed on ipv6 hosts | 20:17 |
| pabelanger | ah, I remember now | 20:17 |
| *** rlandy has quit IRC | 20:19 | |
| *** elyezer has quit IRC | 20:20 | |
| pabelanger | so, is the issue when you use AF_INET6, the address must be ipv6? | 20:20 |
| pabelanger | https://docs.python.org/3.6/library/socketserver.html#socketserver.BaseServer.server_address | 20:21 |
| *** elyezer has joined #zuul | 20:21 | |
| pabelanger | reads that address depends on the protocol family from socket | 20:21 |
| *** JasonCL has quit IRC | 20:21 | |
| Shrews | pabelanger: i don't think so. fairly sure if you gave the ipv4 public address, it would work, but not accept ipv6 connections. where as if you gave it an ipv6 address to listen to, it would accept connections from either | 20:26 |
| Shrews | but i'm actually not sure b/c i keep forgetting that stuff | 20:28 |
| tobiash | Shrews: normally that's right, I think I remember some situations where this didn't work though | 20:34 |
| Shrews | pabelanger: alternative, if you're determined to use the loopback, try '::1' which i think is the v6 equivalent | 20:38 |
| pabelanger | I seem to remember in nodepool, we checked the family of the ipaddress to setup the socket in paramiko to work around some ipv4 / ipv6 issues | 20:40 |
| pabelanger | let me see if I can find that | 20:40 |
| pabelanger | http://git.openstack.org/cgit/openstack-infra/nodepool/tree/nodepool/nodeutils.py?id=d0c25fc333270054699bb22b4793f659b1b7db4a#n56 | 20:46 |
| pabelanger | is what I was thinking of | 20:47 |
| pabelanger | let me try a quick test | 20:47 |
| *** JasonCL has joined #zuul | 20:53 | |
| *** JasonCL_ has joined #zuul | 20:54 | |
| *** JasonCL has quit IRC | 20:55 | |
| *** JasonCL_ has quit IRC | 20:56 | |
| *** JasonCL has joined #zuul | 21:00 | |
| *** JasonCL has quit IRC | 21:03 | |
| *** JasonCL has joined #zuul | 21:03 | |
| *** JasonCL has quit IRC | 21:05 | |
| *** elyezer has quit IRC | 21:19 | |
| *** elyezer has joined #zuul | 21:24 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul master: DNM https://review.openstack.org/551015 | 21:36 |
| *** JasonCL has joined #zuul | 21:41 | |
| *** threestrands has joined #zuul | 21:47 | |
| *** JasonCL has quit IRC | 21:52 | |
| *** rlandy has joined #zuul | 21:56 | |
| *** JasonCL has joined #zuul | 22:01 | |
| *** JasonCL has quit IRC | 22:02 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul master: DNM https://review.openstack.org/551015 | 22:08 |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul master: DNM https://review.openstack.org/551015 | 22:23 |
| *** JasonCL has joined #zuul | 22:35 | |
| *** JasonCL has quit IRC | 22:39 | |
| *** JasonCL has joined #zuul | 22:40 | |
| *** hasharDinner has quit IRC | 22:46 | |
| *** JasonCL has quit IRC | 22:52 | |
| pabelanger | Shrews: tobiash: okay, something like ^ seems to fix the socket exception from above. I'll clean up the code and see if I can expose the issue in a unit test too | 22:57 |
| pabelanger | http://logs.openstack.org/17/551017/3/check/windmill-ubuntu-bionic/4a9d62a/logs/zf01/var/log/zuul/fingergw-debug.log | 22:57 |
| pabelanger | that is on ovh, where it was failing | 22:57 |
| *** elyezer has quit IRC | 22:58 | |
| *** elyezer has joined #zuul | 22:59 | |
| *** JasonCL has joined #zuul | 23:09 | |
| *** JasonCL has quit IRC | 23:10 | |
| *** elyezer has quit IRC | 23:17 | |
| *** JasonCL has joined #zuul | 23:18 | |
| *** JasonCL has quit IRC | 23:19 | |
| *** JasonCL has joined #zuul | 23:19 | |
| *** JasonCL has quit IRC | 23:22 | |
| *** elyezer has joined #zuul | 23:28 | |
| *** threestrands_ has joined #zuul | 23:33 | |
| *** threestrands_ has quit IRC | 23:33 | |
| *** threestrands_ has joined #zuul | 23:33 | |
| *** threestrands has quit IRC | 23:36 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!