| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: zk: use kazoo retry facilities https://review.openstack.org/535537 | 00:45 |
|---|---|---|
| *** odyssey4me has quit IRC | 00:53 | |
| *** odyssey4me has joined #zuul | 00:53 | |
| *** harlowja has quit IRC | 01:20 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: mqtt: add basic reporter https://review.openstack.org/535543 | 01:21 |
| *** rlandy|afk is now known as rlandy | 01:53 | |
| *** rlandy has quit IRC | 02:06 | |
| Wei_Liu | corvus: No, I publish docs to our internal website in CI/CD, and I want to tell the author, who commit this change, the url published in zuul response to gerrit and show it in gerrit as same as the log url. | 02:07 |
| Wei_Liu | clarkb: hi, I publish docs to our internal website in CI/CD, and I want to put the url published in zuul response to gerrit and show it in gerrit as same as the log url. How can I do it? | 02:09 |
| *** myoung|snow|bbl is now known as myoung|rover | 02:15 | |
| *** myoung|rover is now known as myoung|afk | 02:17 | |
| tristanC | Wei_Liu: you could use the zuul_return like in this change: https://review.openstack.org/#/c/548425/1/roles/emit-job-report/tasks/main.yaml | 02:32 |
| tristanC | Wei_Liu: or using the job's success-url, like this: http://git.openstack.org/cgit/openstack-infra/zuul/tree/.zuul.yaml#n81 | 02:32 |
| Wei_Liu | tristanC: I did not know which field of zuul that should I use in zuul_return, zuul can handle zuul.log_url, does it handle the parameter I defined in the zuul_return? | 02:39 |
| tristanC | Wei_Liu: zuul.log_url is the one zuul comments back to the review | 02:40 |
| Wei_Liu | tristanC: I used success-url for such jobs in zuul v2.5, not in pipeline, can I do the same thing in v3? | 02:41 |
| tristanC | Wei_Liu: iirc, success-url gets appened to the log_url, you need to zuul_return another zuul.log_url to replace the url returned to gerrit | 02:44 |
| Wei_Liu | tristanC: Thanks a lot, I will try it today. | 02:46 |
| tristanC | Wei_Liu: the zuul_return likely needs to happen after the upload-log roles to over-write the value | 02:50 |
| Wei_Liu | tristanC: yes, I will append new role after it. | 02:54 |
| *** Wei_Liu1 has joined #zuul | 03:03 | |
| *** Wei_Liu has quit IRC | 03:03 | |
| *** Wei_Liu1 is now known as Wei_Liu | 03:03 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: WIP: Handle ZK session loss during node launch https://review.openstack.org/552538 | 03:04 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: zk: use kazoo retry facilities https://review.openstack.org/535537 | 03:04 |
| *** Wei_Liu has quit IRC | 03:58 | |
| *** Wei_Liu has joined #zuul | 03:59 | |
| *** harlowja has joined #zuul | 04:12 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: WIP: Handle ZK session loss during node launch https://review.openstack.org/552538 | 04:21 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: Refactor run_handler to be generic https://review.openstack.org/535554 | 04:21 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: Refactor NodeLauncher to be generic https://review.openstack.org/535555 | 04:21 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: zk: use kazoo retry facilities https://review.openstack.org/535537 | 04:22 |
| *** harlowja has quit IRC | 04:40 | |
| openstackgerrit | Merged openstack-infra/zuul master: Add zuul-tox-remote to gate pipeline https://review.openstack.org/552692 | 05:53 |
| tobiash | corvus: I've comments on 552688 (just two accidentally added files) | 06:14 |
| openstackgerrit | Tobias Henkel proposed openstack-infra/zuul master: Add zuul_json tests https://review.openstack.org/552688 | 06:15 |
| openstackgerrit | Simon Westphahl proposed openstack-infra/zuul master: Fix zuul_json callback plugin problem with loops https://review.openstack.org/552799 | 06:37 |
| openstackgerrit | Clint 'SpamapS' Byrum proposed openstack-infra/zuul master: Use re2 for change_matcher https://review.openstack.org/536389 | 07:09 |
| openstackgerrit | Clint 'SpamapS' Byrum proposed openstack-infra/zuul master: Add irrelevant-branches negative matcher https://review.openstack.org/552809 | 07:09 |
| SpamapS | corvus: ^ split | 07:09 |
| SpamapS | Haven't tested tho | 07:09 |
| * SpamapS heads to bed | 07:09 | |
| tobiash | patch and run ;) | 07:34 |
| *** hashar has joined #zuul | 07:43 | |
| *** jpena|off is now known as jpena | 08:25 | |
| *** electrofelix has joined #zuul | 08:33 | |
| openstackgerrit | Joshua Hesketh proposed openstack-infra/zuul master: Decode gearman function into utf https://review.openstack.org/552888 | 11:54 |
| *** elyezer has joined #zuul | 11:55 | |
| openstackgerrit | Joshua Hesketh proposed openstack-infra/zuul master: Decode gearman function into utf https://review.openstack.org/552888 | 12:08 |
| *** elyezer has joined #zuul | 12:08 | |
| *** elyezer has quit IRC | 12:08 | |
| openstackgerrit | Joshua Hesketh proposed openstack-infra/zuul master: Decode gearman function into utf https://review.openstack.org/552888 | 12:11 |
| openstackgerrit | Matthieu Huin proposed openstack-infra/zuul master: [WIP] Zuul Web: Add 'projects' endpoint https://review.openstack.org/552891 | 12:16 |
| *** jpena is now known as jpena|lunch | 12:27 | |
| *** rlandy has joined #zuul | 12:32 | |
| *** odyssey4me has quit IRC | 12:47 | |
| *** odyssey4me has joined #zuul | 12:47 | |
| dmsimard | Is it technically feasible for secrets to be... variable ? Like pretend we have a generic role to publish something to... DockerHub. Could projects supply their own credentials to the role somehow ? It hurts my brain. | 12:59 |
| dmsimard | I haven't used secrets much so hoping someone can chime it :) | 12:59 |
| dmsimard | s/it/in/ | 13:00 |
| tobiash | dmsimard: secrets are attached to a job, not a role | 13:01 |
| tobiash | a job's playbook can parameterize the roles it's using | 13:01 |
| dmsimard | tobiash: right .. but I mean, are you able to supply a different secret to a job ? | 13:01 |
| tobiash | so there is no problem using a generic role | 13:01 |
| dmsimard | or because of that trust thing it won't let you do that | 13:02 |
| tobiash | dmsimard: you just inherit from it and supply your own secret | 13:02 |
| tobiash | you just cannot share the playbook, but the roles | 13:02 |
| dmsimard | makes sense, was just confirming, thanks :D | 13:02 |
| * dmsimard has an idea | 13:02 | |
| tobiash | note that secrets are only available in the playbook where the job is defined | 13:03 |
| tobiash | supplying a secret to a parent job and not supplying/duplicating the run-playbook won't work | 13:03 |
| dmsimard | tobiash: yeah, I understand | 13:05 |
| dmsimard | tobiash: maybe I can tell you about my idea and you'll see what I mean | 13:05 |
| * tobiash is listening | 13:05 | |
| dmsimard | TripleO has these ARA modules that basically sends all the Ansible metrics (task duration, etc.) to graphite or influxdb, ex influxdb: http://git.openstack.org/cgit/openstack/tripleo-quickstart-extras/tree/roles/collect-logs/library/ara_influxdb.py | 13:06 |
| dmsimard | It's kind of a shame because these are /almost/ generic -- we'd take the time to wrap these in a role and run them around the time where emit-ara-html is run | 13:07 |
| dmsimard | making it available in zuul-jobs, or wrapping it under a pattern like we do with gearman things would allow us to get a ton of metrics about the jobs running | 13:08 |
| dmsimard | at the kind of scale we are running, it would be valuable to tell if something that we do has an adverse impact on performance (for example) | 13:09 |
| dmsimard | like pretend we tweak the internal poll interval.. or the amount of forks.. or just optimize $thing | 13:09 |
| dmsimard | I realize that the performance is not uniform across all the clouds (and even within the same cloud) but it sort of averages out | 13:10 |
| tobiash | yes, a generic role probably makes sense | 13:14 |
| *** Wei_Liu has quit IRC | 13:17 | |
| dmsimard | oh darn, even better .. with https://review.openstack.org/#/c/513874/ saving the database at a particular location, it would be almost trivial to use our gearman pattern to do it | 13:20 |
| *** jpena|lunch is now known as jpena | 13:27 | |
| *** myoung|afk is now known as myoung|rover | 13:35 | |
| *** electrofelix has quit IRC | 14:16 | |
| openstackgerrit | Merged openstack-infra/zuul master: Add zuul_json tests https://review.openstack.org/552688 | 14:26 |
| openstackgerrit | Merged openstack-infra/zuul master: Fix zuul_json callback plugin problem with loops https://review.openstack.org/552799 | 15:04 |
| tobiash | \o/ | 15:04 |
| *** swest has quit IRC | 15:11 | |
| *** rlandy has quit IRC | 15:24 | |
| *** rlandy has joined #zuul | 15:24 | |
| corvus | tobiash: that's probably worth a note to the list too... | 15:29 |
| tobiash | corvus: yes, will draft a mail later | 15:29 |
| dmsimard | I don't remember who and where, but someone asked about the "let zuul start with a broken configuration" patch recently, I've poked fbo about https://review.openstack.org/#/c/535511 and he'll resume his work on it | 15:31 |
| corvus | dmsimard: thanks. that was clarkb in the infra meeting yesterday in the context of making it safer to deal with project renames | 15:32 |
| dmsimard | ah, I couldn't find in my logs but I remembered someone asked about it. Thanks. | 15:32 |
| corvus | is anyone else interested in reviewing configloader changes (ie, my effort to rework things to use less memory)? https://review.openstack.org/546428 | 15:52 |
| clarkb | corvus: yes I'll take a look | 15:53 |
| Shrews | corvus: I will look after lunch if you like | 15:53 |
| corvus | fyi, that's just the first baby step -- i expect a whole series like that before we're through | 15:54 |
| clarkb | I've confirmed we have existing tests that verify secrets and approved the change | 16:00 |
| openstackgerrit | Merged openstack-infra/zuul master: Import Zuul modules at top of files https://review.openstack.org/549730 | 16:09 |
| tobiash | corvus: will look after dinner | 16:21 |
| openstackgerrit | Merged openstack-infra/zuul master: Match github model of granting admins write https://review.openstack.org/549853 | 16:26 |
| openstackgerrit | Merged openstack-infra/zuul master: Make sure we fail because of path checks https://review.openstack.org/552502 | 16:26 |
| openstackgerrit | Merged openstack-infra/zuul master: Decode gearman function into utf https://review.openstack.org/552888 | 16:26 |
| dmsimard | Oh, contributions to Zuul from Wikimedia, that's awesome :D | 16:35 |
| dmsimard | I thought they used phabricator or something like that | 16:35 |
| clarkb | dmsimard: I think they are in the middle of a transition from gerrit to phabricator and it has gone longer than expected? so zuul is still used by the gerrit projects | 16:36 |
| clarkb | also in theory we could hook gerrit to phabricator though I don't know that that is in their plans | 16:36 |
| dmsimard | I know too little about phabricator to know how Zuul could fit in there | 16:37 |
| dmsimard | Then again, OVH has a nodepool-like project so it's probably not uncommon for orgs to have their own custom/purpose built workflow | 16:38 |
| dmsimard | It would probably be interesting to reach out and ask them how Zuul could fit their needs, even if just out of curiosity to see what are the use cases out there. | 16:45 |
| clarkb | them being wikimedia? I think they have been in touch | 16:45 |
| corvus | dmsimard: who's them? | 16:45 |
| dmsimard | corvus: OVH | 16:45 |
| dmsimard | One of the components of their platform is basically nodepool https://ovh.github.io/cds/hatchery/ | 16:46 |
| openstackgerrit | Merged openstack-infra/zuul master: Skip autohold if no autohold was requested https://review.openstack.org/547302 | 16:54 |
| openstackgerrit | Merged openstack-infra/zuul master: Add queue size to tenant overview https://review.openstack.org/548248 | 16:54 |
| openstackgerrit | Merged openstack-infra/zuul master: Don't store references to secret objects from jobs https://review.openstack.org/546428 | 16:54 |
| openstackgerrit | Merged openstack-infra/zuul master: Add test for fingergw on ipv4 / ipv6 addresses https://review.openstack.org/552645 | 16:54 |
| Shrews | corvus: the test_slow_start failure in that fingergw timeout change confuses me. do you think it could be related? http://logs.openstack.org/35/546735/1/gate/zuul-tox-py35/f00479c/testr_results.html.gz | 16:57 |
| Shrews | i don't see how really since we don't start the fingergw | 17:01 |
| * Shrews going to recheck | 17:02 | |
| *** hashar is now known as hasharAway | 17:43 | |
| *** bhavik1 has joined #zuul | 17:52 | |
| *** jpena is now known as jpena|off | 18:01 | |
| openstackgerrit | Merged openstack-infra/zuul master: Unset finger client timeout after connect https://review.openstack.org/546735 | 18:05 |
| tobiash | Shrews: I've seen this test failing several times for unrelated changes | 18:09 |
| tobiash | maybe we have some race in there | 18:09 |
| *** harlowja has joined #zuul | 18:11 | |
| Shrews | i suppose so | 18:11 |
| *** bhavik1 has quit IRC | 18:17 | |
| tobiash | corvus: how's that? https://etherpad.openstack.org/p/FpVT6Dl62i | 18:22 |
| tobiash | fungi, clarkb ^ | 18:23 |
| *** myoung|rover is now known as myoung|lunch | 18:23 | |
| corvus | tobiash: lgtm | 18:24 |
| tobiash | :) | 18:24 |
| pabelanger | do we know of any project is openstack affected? | 18:26 |
| tobiash | pabelanger: you probably need to check your post playbooks that doo job upload etc | 18:27 |
| corvus | a question for #openstack-infra :) | 18:28 |
| fungi | tobiash: thanks, reviewing now | 18:28 |
| corvus | though fwiw, once that goes out, i was planning on posting a message to openstack-dev about it. i think it's important for operators to notify their downstream users who may not be watching zuul-announce for things like this. | 18:29 |
| tobiash | pabelanger: upload-logs at least seems not affected | 18:29 |
| tobiash | just checked zuul-jobs, which seems ok | 18:30 |
| tobiash | no no_log combination with loops in there | 18:30 |
| tobiash | corvus: yeah, that makes sense | 18:31 |
| fungi | tobiash: i think that second sentence could use rephrasing to be a little more clear. i'll try to propose an alternate wording in the pad | 18:32 |
| tobiash | fungi: thanks, sounds good, what do you think about adding 'despite defining no_log' to that? | 18:34 |
| fungi | what do you think now? | 18:35 |
| tobiash | lgtm | 18:36 |
| tobiash | good to send now? | 18:37 |
| corvus | tobiash: ++ | 18:37 |
| fungi | yeah, i think so unless others object | 18:37 |
| tobiash | corvus, fungi: do we want/need a worklist or dashboard for the zuul-security stories? | 18:46 |
| corvus | tobiash: that's probably a good idea | 18:46 |
| tobiash | corvus: so what do you think is better in this case, a worklist or a dashboard? | 18:47 |
| tobiash | I'm not yet that familiar with storyboard | 18:47 |
| fungi | though it'll be interesting if we try to mix tasks from public and private stories | 18:47 |
| fungi | i want to say there's still some implementation question around that | 18:47 |
| fungi | though if you have sb questions, the weekly storyboard meeting starts in ~10 minutes | 18:48 |
| corvus | tobiash: i guess a worklist? unless we wanted a board for triaging bugs or something. | 18:49 |
| corvus | worklist is the simplest "list of stories matching criteria" thing | 18:50 |
| corvus | interestingly, worklists can be private too | 18:50 |
| tobiash | corvus: https://storyboard.openstack.org/#!/worklist/258 | 18:50 |
| tobiash | I had that private, but not sure if I can add users to that | 18:50 |
| corvus | omg, when i click that link i'm logged in | 18:51 |
| tobiash | is that bad? | 18:51 |
| tobiash | I guess that worklist will be empty for non zuul-security members? | 18:52 |
| fungi | hopefully. we should have one of them test that hypothesis | 18:53 |
| Shrews | fwiw, that link is blank for me | 18:54 |
| rcarrillocruz | so is for me | 18:54 |
| fungi | perfect, that was the desired result | 18:54 |
| rcarrillocruz | i see the title | 18:55 |
| rcarrillocruz | nothing on main page | 18:55 |
| tobiash | so the filter is active + openstack-infra/zuul + label zuul-security | 18:55 |
| corvus | tobiash: sorry, it's just a storyboard bug(?) i've been struggling with. i can't generally use it in more than one tab, and... well, it's very hard to use storyboard without tabs. so, every time i open up storyboard, boartty gets just a little bit more complete. | 18:57 |
| corvus | tobiash: so i was surprised that we accidentally found a way for me to open more than one tab | 18:58 |
| tobiash | \o/ | 18:58 |
| rcarrillocruz | boartty == gertty for storyboard ? | 18:58 |
| corvus | rcarrillocruz: yep | 18:59 |
| rcarrillocruz | haha, that's awesome corvus | 18:59 |
| corvus | it's really rough. breaks a lot. kinda works though. | 18:59 |
| corvus | http://git.openstack.org/cgit/openstack/boartty | 19:00 |
| openstackgerrit | Matthieu Huin proposed openstack-infra/zuul master: web: Add /projects, /projects/{project} endpoints https://review.openstack.org/552891 | 19:13 |
| *** myoung|lunch is now known as myoung|rover | 19:18 | |
| openstackgerrit | Merged openstack-infra/zuul master: configloader: check for unknown shadow projects https://review.openstack.org/550763 | 19:38 |
| openstackgerrit | Tobias Henkel proposed openstack-infra/zuul master: Fix safe path check for directories containing symlinks https://review.openstack.org/553029 | 19:55 |
| tobiash | corvus, clarkb: ^ | 19:56 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Perform late validation of secrets https://review.openstack.org/553041 | 20:09 |
| openstackgerrit | Merged openstack-infra/zuul master: Fix safe path check for directories containing symlinks https://review.openstack.org/553029 | 20:24 |
| tobiash | \o/ | 20:25 |
| tobiash | corvus: I have a question on 553041 | 20:25 |
| tobiash | /have/posted | 20:25 |
| tobiash | corvus, fungi, clarkb: https://etherpad.openstack.org/p/j0wvUWWWH7 | 20:27 |
| clarkb | tobiash: left a small edit otehrwise lgtm | 20:28 |
| tobiash | thanks | 20:28 |
| corvus | tobiash: lgtm | 20:29 |
| tobiash | looks like I'm going to be the bad news messenger ;) | 20:30 |
| fungi | such is the life of a vulnerability coordinator | 20:32 |
| corvus | i'm looking at tobiash's next patch in storyboard, and i just realized that using <pre> is problematic -- we still need to escape <> or it gets eaten | 20:33 |
| fungi | tobiash: your announcement wording lgtm | 20:34 |
| clarkb | markdown has a block quote method that is easy to use iirc | 20:35 |
| clarkb | but I'd have to go reread how to do it | 20:35 |
| corvus | clarkb: i could only come up with "indent everything a couple of spaces" if there's something better, that'd be great | 20:35 |
| * clarkb googles around | 20:35 | |
| tobiash | corvus: oh, didn't notice | 20:36 |
| fungi | this may also explain why lp just punted to making you put patches in text/plain downloadable attachments | 20:36 |
| clarkb | corvus: wrap in ``` and ``` | 20:36 |
| clarkb | corvus: so line 0 is ``` and line EOF-1 is ``` | 20:36 |
| corvus | tobiash: would it be easy for you to try that real quick? | 20:37 |
| tobiash | ya | 20:37 |
| tobiash | just a sec | 20:37 |
| corvus | (i want to review the next one in git, so want to run it through git am) | 20:37 |
| fungi | the alternative we've talked about for years (and still have a possibly abandoned spec floating around for?) is to run a non-public shadow gerrit, but syncing projects into it and doing the access controls for all the project-specific security review teams is where we always ended up getting stuck | 20:39 |
| fungi | it's tractable, just a lot of initial and ongoing effort for only a handful (hopefully) of patches | 20:40 |
| tobiash | oh wow, with backticks it dows syntax highlighting | 20:40 |
| corvus | oh that's nice | 20:41 |
| corvus | that looks copy/pastable, so i guess that's what we should put in our docs for now | 20:41 |
| corvus | yay, git am succeeded | 20:42 |
| tobiash | :) | 20:42 |
| fungi | excellent | 20:49 |
| clarkb | ya the ``` is code block quoting which is why it does syntax highlighting. I expect that will be the most friendly version of quoting for diffs | 20:58 |
| *** dkranz has quit IRC | 21:03 | |
| corvus | note i just found a markdown trap: * is used for italicizing, so if you type "*/*" you get an italic slash. this makes writing comments about paths (eg foo*/*bar) treacherous. | 21:04 |
| tobiash | ya, marking paths or so should be done with `*/*` | 21:09 |
| dmsimard | tobiash: the last fix for looped no_log, ARA should correctly strip that out. Are you able to confirm ? | 21:40 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Perform late validation of secrets https://review.openstack.org/553041 | 22:01 |
| *** myoung|rover is now known as myoung|afk | 22:13 | |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Perform late validation of secrets https://review.openstack.org/553041 | 22:41 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Perform late validation of nodesets https://review.openstack.org/553088 | 22:41 |
| *** hasharAway has quit IRC | 23:23 | |
| *** kmalloc has quit IRC | 23:38 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!