| *** jamesmcarthur has joined #zuul | 00:05 | |
| *** jamesmcarthur has quit IRC | 00:09 | |
| *** jamesmcarthur has joined #zuul | 00:15 | |
| *** jamesmcarthur has quit IRC | 00:25 | |
| mnaser | hmm | 00:34 |
|---|---|---|
| mnaser | zuul is complaining "paramiko.ssh_exception.SSHException: No authentication methods available" | 00:34 |
| clarkb | in what context? | 00:35 |
| mnaser | but i can run successfully: ssh -i /path/to/sshkey.pem user@host -p29418 stream-events | 00:35 |
| clarkb | ansible should use openssh by default | 00:35 |
| mnaser | this is inside the scheduler | 00:35 |
| clarkb | ah for gerrit rvents | 00:35 |
| clarkb | is it possible that gerrit is rejecting the zuul auth and that is how it ends up at no remaining methods? counld be the username or other setting (host port etc) if so | 00:37 |
| mnaser | ah ffs | 00:38 |
| mnaser | <gerrit connection>.sshkey | 00:38 |
| mnaser | "ssh_key = /etc/zuul/opendev.pem" | 00:38 |
| mnaser | and thats why you dont write things on a friday i guess | 00:39 |
| mnaser | err sunday. | 00:39 |
| mnaser | and that was it lol | 00:40 |
| *** jamesmcarthur has joined #zuul | 00:59 | |
| *** jamesmcarthur has quit IRC | 01:10 | |
| *** jamesmcarthur has joined #zuul | 03:10 | |
| *** jhesketh has joined #zuul | 03:50 | |
| *** jamesmcarthur has quit IRC | 04:01 | |
| *** jamesmcarthur has joined #zuul | 04:04 | |
| *** jamesmcarthur has quit IRC | 04:17 | |
| *** bhavikdbavishi has joined #zuul | 05:10 | |
| *** pcaruana has joined #zuul | 05:11 | |
| *** roman_g has joined #zuul | 05:47 | |
| openstackgerrit | Ian Wienand proposed zuul/zuul master: Support nodes setting 'auto' python path https://review.opendev.org/682275 | 05:53 |
| *** jamesmcarthur has joined #zuul | 06:18 | |
| *** badboy has joined #zuul | 06:21 | |
| *** jamesmcarthur has quit IRC | 06:23 | |
| *** logan- has quit IRC | 06:23 | |
| *** logan- has joined #zuul | 06:26 | |
| sshnaidm_ | is there an option to set irrelevant_files for whole project-template? | 06:37 |
| *** sshnaidm_ is now known as sshnaidm|rover | 06:37 | |
| openstackgerrit | Ian Wienand proposed zuul/zuul master: Support nodes setting 'auto' python path https://review.opendev.org/682275 | 06:40 |
| *** gtema_ has joined #zuul | 06:56 | |
| AJaeger | sshnaidm|rover: no. | 06:58 |
| *** fbo has joined #zuul | 07:08 | |
| *** tosky has joined #zuul | 07:08 | |
| *** bolg has joined #zuul | 07:25 | |
| *** jangutter has joined #zuul | 07:34 | |
| *** jpena|off is now known as jpena | 07:34 | |
| *** hashar has joined #zuul | 07:38 | |
| *** sshnaidm|rover is now known as sshnaidm|afk | 07:39 | |
| *** gtema_ has quit IRC | 07:46 | |
| *** panda has quit IRC | 07:56 | |
| *** panda has joined #zuul | 07:58 | |
| *** themroc has joined #zuul | 08:22 | |
| *** saneax has joined #zuul | 08:30 | |
| *** sshnaidm|afk is now known as sshnaidm|rover | 08:43 | |
| *** hashar has quit IRC | 08:47 | |
| *** panda is now known as panda|ruck | 08:59 | |
| *** hashar has joined #zuul | 09:11 | |
| *** mgoddard has quit IRC | 09:26 | |
| *** mgoddard has joined #zuul | 09:28 | |
| *** themroc has quit IRC | 10:03 | |
| *** gtema_ has joined #zuul | 10:22 | |
| sshnaidm|rover | can I figure out when the buildset was queued? | 10:39 |
| *** jpena is now known as jpena|lunch | 11:30 | |
| openstackgerrit | Merged zuul/zuul master: Better error handling for the PagureAPIClient https://review.opendev.org/679746 | 11:35 |
| openstackgerrit | Merged zuul/zuul master: Remove auto-add of localhost from unit test inventory https://review.opendev.org/681641 | 11:45 |
| openstackgerrit | Merged zuul/zuul-jobs master: Add tests for manifest generation for missing files https://review.opendev.org/678573 | 11:56 |
| *** gtema_ has quit IRC | 11:57 | |
| openstackgerrit | Merged zuul/zuul master: Discuss executor-only jobs, add unit-test https://review.opendev.org/679184 | 12:08 |
| *** sanjayu_ has joined #zuul | 12:17 | |
| *** saneax has quit IRC | 12:18 | |
| *** rfolco has joined #zuul | 12:23 | |
| *** rlandy has joined #zuul | 12:25 | |
| *** jpena|lunch is now known as jpena | 12:33 | |
| *** themroc has joined #zuul | 12:35 | |
| *** badboy has quit IRC | 12:39 | |
| *** jamesmcarthur has joined #zuul | 12:48 | |
| openstackgerrit | David Shrewsbury proposed zuul/zuul master: Store autohold requests in zookeeper https://review.opendev.org/661114 | 12:48 |
| *** Goneri has joined #zuul | 13:11 | |
| -openstackstatus- NOTICE: The Gerrit service on review.opendev.org will be offline briefly starting at 14:00 UTC (that's roughly 30 minutes from now) for maintenance: http://lists.openstack.org/pipermail/openstack-discuss/2019-September/009064.html | 13:30 | |
| *** bhavikdbavishi has quit IRC | 13:30 | |
| *** bhavikdbavishi has joined #zuul | 13:32 | |
| *** bolg has quit IRC | 13:35 | |
| *** jangutter has quit IRC | 13:40 | |
| *** jangutter has joined #zuul | 13:42 | |
| *** pcaruana has quit IRC | 13:45 | |
| *** jangutter has quit IRC | 13:53 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-sphinx master: Switch to zuul-release-python https://review.opendev.org/682366 | 13:58 |
| *** pcaruana has joined #zuul | 14:07 | |
| -openstackstatus- NOTICE: The Gerrit service on review.opendev.org is offline briefly for maintenance: http://lists.openstack.org/pipermail/openstack-discuss/2019-September/009064.html | 14:08 | |
| *** ChanServ changes topic to "The Gerrit service on review.opendev.org is offline briefly for maintenance: http://lists.openstack.org/pipermail/openstack-discuss/2019-September/009064.html" | 14:08 | |
| *** jamesmcarthur has quit IRC | 14:31 | |
| *** sanjayu_ has quit IRC | 14:32 | |
| *** sanjayu_ has joined #zuul | 14:32 | |
| *** jamesmcarthur has joined #zuul | 14:33 | |
| *** jamesmcarthur has quit IRC | 14:39 | |
| *** ChanServ changes topic to "Discussion of the project gating system Zuul | Website: https://zuul-ci.org/ | Docs: https://zuul-ci.org/docs/ | Source: https://git.zuul-ci.org/ | Channel logs: http://eavesdrop.openstack.org/irclogs/%23zuul/ | Weekly updates: https://etherpad.openstack.org/p/zuul-update-email" | 14:39 | |
| -openstackstatus- NOTICE: The Gerrit outage portion of the current maintenance is complete and the service is back on line, however reindexing for renamed repositories is still underway and some Zuul job fixes are in the process of being applied | 14:39 | |
| openstackgerrit | David Shrewsbury proposed zuul/zuul master: Add caching of autohold requests https://review.opendev.org/663412 | 14:47 |
| openstackgerrit | David Shrewsbury proposed zuul/zuul master: Add autohold-info CLI command https://review.opendev.org/662487 | 14:47 |
| openstackgerrit | David Shrewsbury proposed zuul/zuul master: Record held node IDs with autohold request https://review.opendev.org/662498 | 14:48 |
| openstackgerrit | David Shrewsbury proposed zuul/zuul master: Auto-delete expired autohold requests https://review.opendev.org/663762 | 14:50 |
| openstackgerrit | David Shrewsbury proposed zuul/zuul master: Mark nodes as USED when deleting autohold https://review.opendev.org/664060 | 14:50 |
| openstackgerrit | David Shrewsbury proposed zuul/zuul master: Add autohold delete/info commands to web API https://review.opendev.org/679057 | 14:50 |
| openstackgerrit | David Shrewsbury proposed zuul/zuul master: Auto-delete expired autohold requests https://review.opendev.org/663762 | 14:52 |
| *** sanjayu_ has quit IRC | 14:54 | |
| openstackgerrit | David Shrewsbury proposed zuul/zuul master: Auto-delete expired autohold requests https://review.opendev.org/663762 | 15:00 |
| *** jamesmcarthur has joined #zuul | 15:02 | |
| *** hashar has quit IRC | 15:03 | |
| fungi | zuul-maint: there's another osf newsletter being drafted this week, so if there's anything anyone would like to see included i'm happy to pass it along to the newsletter editors | 15:05 |
| fungi | just let me know | 15:05 |
| openstackgerrit | David Shrewsbury proposed zuul/zuul master: Mark nodes as USED when deleting autohold https://review.opendev.org/664060 | 15:07 |
| *** jamesmcarthur has quit IRC | 15:07 | |
| *** openstackgerrit has quit IRC | 15:08 | |
| *** openstackgerrit has joined #zuul | 15:08 | |
| openstackgerrit | David Shrewsbury proposed zuul/zuul master: Add autohold delete/info commands to web API https://review.opendev.org/679057 | 15:08 |
| Shrews | Sorry for the rebase noise | 15:10 |
| * mordred hands Shrews a rebase | 15:16 | |
| Shrews | i still feel like i did something wrong with that rebase | 15:19 |
| Shrews | especially since none of those show up on the status page | 15:20 |
| Shrews | oh, wrong status page :/ | 15:21 |
| *** mattw4 has joined #zuul | 15:21 | |
| mordred | Shrews: have you tried turning if off and back on again? | 15:23 |
| *** themroc has quit IRC | 15:39 | |
| *** jamesmcarthur has joined #zuul | 15:42 | |
| *** jamesmcarthur has quit IRC | 15:45 | |
| *** jamesmcarthur has joined #zuul | 15:45 | |
| *** bhavikdbavishi has quit IRC | 15:54 | |
| *** mattw4 has quit IRC | 15:58 | |
| openstackgerrit | David Shrewsbury proposed zuul/zuul master: Auto-delete expired autohold requests https://review.opendev.org/663762 | 16:16 |
| openstackgerrit | David Shrewsbury proposed zuul/zuul master: Mark nodes as USED when deleting autohold https://review.opendev.org/664060 | 16:18 |
| openstackgerrit | David Shrewsbury proposed zuul/zuul master: Add autohold delete/info commands to web API https://review.opendev.org/679057 | 16:18 |
| *** brendangalloway has joined #zuul | 16:19 | |
| *** rlandy is now known as rlandy|brb | 16:44 | |
| *** sshnaidm|rover is now known as sshnaidm|afk | 16:49 | |
| corvus | Shrews: comment on https://review.opendev.org/661114 -- that's my only comment on the entire stack. | 16:55 |
| *** jpena is now known as jpena|off | 16:58 | |
| Shrews | corvus: oops, you are correct about that | 16:59 |
| corvus | Shrews: cool, i *thought* it felt like an omission, but wanted to check that there weren't grander plans i didn't see :) | 16:59 |
| Shrews | corvus: oh, you know what... i think that's done by the zk.deleteHoldRequest() method | 17:00 |
| Shrews | corvus: here: https://review.opendev.org/#/c/664060/14/zuul/zk.py | 17:01 |
| Shrews | so the comment is just outdated | 17:01 |
| Shrews | had to scratch my head on that one... a pretty fundamental thing to miss if i did that! :) | 17:02 |
| *** jamesmcarthur has quit IRC | 17:03 | |
| *** brendangalloway has quit IRC | 17:04 | |
| corvus | Shrews: aha, you're right. I'll +2 the stack, you want to push one more patch that removes the TODO? :) | 17:06 |
| Shrews | corvus: yeah, i'll tack it on the end of the stack | 17:07 |
| corvus | tristanC: i'll leave the +3 of the autohold-revamp stack to you ^ | 17:07 |
| tristanC | great, i'll pick the new stack on my test setup and validate it again | 17:08 |
| openstackgerrit | Merged zuul/zuul-jobs master: Retry container gets in upload-logs-swift https://review.opendev.org/682091 | 17:09 |
| openstackgerrit | David Shrewsbury proposed zuul/zuul master: Remove outdated TODO https://review.opendev.org/682421 | 17:10 |
| Shrews | tristanC: i *think* i addressed all of your comments from Friday | 17:11 |
| *** mattw4 has joined #zuul | 17:14 | |
| Shrews | we could be cruel and force mordred to review that stack since it originated from his random thoughts :) | 17:18 |
| AJaeger | he's just back from vacation and full of energy ;) | 17:18 |
| Shrews | a nice "welcome back" present | 17:18 |
| *** Goneri has quit IRC | 17:21 | |
| *** jamesmcarthur has joined #zuul | 17:27 | |
| *** rlandy|brb is now known as rlandy | 17:29 | |
| fungi | home is where the reviews are | 17:39 |
| *** themroc has joined #zuul | 17:58 | |
| *** Goneri has joined #zuul | 18:02 | |
| *** jamesmcarthur has quit IRC | 18:08 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: zk: use pathlib.Path to define ZooKeeper paths https://review.opendev.org/682450 | 18:25 |
| *** jamesmcarthur has joined #zuul | 18:25 | |
| *** jamesmcarthur has quit IRC | 18:25 | |
| mordred | Shrews: yay? | 18:30 |
| *** hashar has joined #zuul | 18:42 | |
| tristanC | Shrews: i can't figure out how to make autohold request expired based on nodepool max-hold-age | 18:49 |
| tristanC | Shrews: the zuul autohold cli help says that (default: nodepool's max-hold-age if set, or indefinitely), but it's always either infinit, or the value given on the cli | 18:50 |
| *** hashar has quit IRC | 18:51 | |
| Shrews | tristanC: default (which is infinite) is set in nodepool config: https://zuul-ci.org/docs/nodepool/configuration.html#attr-max-hold-age | 18:51 |
| *** hashar has joined #zuul | 18:51 | |
| *** openstackgerrit has quit IRC | 18:52 | |
| Shrews | tristanC: if you change the nodepool default to something else, and it doesn't default to that, that's a bug somewhere (maybe even nodepool) | 18:52 |
| tristanC | Shrews: then there may be an issue in zuul.Scheduler.autohold method, because by default it write "0" to node_expiration and the request never expires, even when the nodes have been deleted by nodepool because of max-age | 18:52 |
| tristanC | thus I'm not sure how zuul is supposed to know the nodepool max-hold-age to properly expire its hold request | 18:53 |
| tristanC | perhaps this auto expiration logic should be based on the last held node to transition from "hold" to "deleted"? | 18:54 |
| Shrews | tristanC: i forget off hand how the node expiration logic works (perhaps ask mhu if he's around?) I have to run off to a doctor appointment, but if you can supply an example of what's not working, I can help dig when I get back. | 18:56 |
| tristanC | Shrews: alright, perhaps the issue is not introduced by autohold-revamp and it could be fixed in a follow-up | 18:57 |
| *** rlandy is now known as rlandy|biab | 18:59 | |
| tristanC | Shrews: I commented an example in https://review.opendev.org/663762 | 19:01 |
| *** openstackgerrit has joined #zuul | 19:18 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: zk: use pathlib.Path to define ZooKeeper paths https://review.opendev.org/682450 | 19:18 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: Store a list of held node per held build in hold request https://review.opendev.org/682466 | 19:18 |
| *** hashar has quit IRC | 19:20 | |
| corvus | tristanC, Shrews: yeah that's a good point -- it seems like we may want to move the default expiration from nodepool into zuul. otherwise, there's no way for zuul to have a default value for the *request* expiration. | 19:29 |
| corvus | (it'll just have nodes disappear out from under it, which is fine, but the request list will get cluttered since nothing is ever removed) | 19:30 |
| *** jamesmcarthur has joined #zuul | 19:33 | |
| *** themroc has quit IRC | 19:39 | |
| *** igordc has joined #zuul | 19:51 | |
| *** mhu has quit IRC | 19:51 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-sphinx master: Switch to zuul-release-python https://review.opendev.org/682366 | 19:52 |
| *** rlandy|biab is now known as rlandy | 19:57 | |
| tristanC | I didn't +3 the base change of the request-revamp to give another zuul-maint a chance to have a look, otherwise the stack lgtm beside an issue with autohold rest endpoint route which i think should be tenant scoped | 19:58 |
| tristanC | autohold*-revamp | 19:59 |
| corvus | tristanC: good catch; i removed my +2 on that. let's see what Shrews wants to do about the autoexpire before we +w that first one | 20:00 |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Add autogenerated tag to Gerrit reviews https://review.opendev.org/682473 | 20:02 |
| corvus | mordred: ^ another on that stack; we'll probably want to have that in before we start commenting on gerrit gerrit changes or we'll look uncouth. | 20:03 |
| *** jamesmcarthur has quit IRC | 20:15 | |
| *** jamesmcarthur has joined #zuul | 20:16 | |
| *** hashar has joined #zuul | 20:17 | |
| *** hashar has quit IRC | 20:17 | |
| *** jamesmcarthur has quit IRC | 20:21 | |
| Shrews | corvus: perhaps we should consult with mhu on the node expiration issue since i believe that was his feature. should that prevent the revamp from going in? | 20:22 |
| tristanC | Shrews: the revamp doesn't seems to be introducing the issue, it seems like something to be fixed in a follow-up | 20:23 |
| Shrews | tristanC: *nod*... i can see how this effects the auto-expire of the hold request though... i'm not sure how to deal with that right now, tbh | 20:24 |
| corvus | Shrews, tristanC: yeah, i'd argue it does introduce the issue and we should resolve it before we merge the stack. | 20:28 |
| tristanC | Shrews: iiuc corvus suggested we move the max-hold-age nodepool setting into zuul | 20:28 |
| corvus | basically, i see the stack as moving the UI for dealing with holds more or less entirely into zuul, which is fine, but we should move the expiration along with it | 20:28 |
| tristanC | unless we make nodepool store the max-hold-age in zookeeper | 20:29 |
| corvus | right, that's an option too, but that's a little weird; i'm not sure if we want to pass configuration info *up* | 20:29 |
| Shrews | yeah, not sure i like the reverse config dependency either | 20:30 |
| corvus | i think it'd be okay to add a default autohold expiration option to zuul, and then deprecate the one in nodepool (we don't have to remove it from nodepool immediately -- it was only a default anyway) | 20:30 |
| Shrews | corvus: So, a new change to add the hold expiration config setting to zuul, and then parent https://review.opendev.org/663762 and its sibilings to that? | 20:33 |
| Shrews | then the nodepool deprecation can be handled independently, i suppose | 20:34 |
| tristanC | one the other hand, the current autohold-info output may lists nodes' id that are no longer available. Thus when listing or checking for request-hold expiry, we could check if the actual node still exists, and when they are all gone, then we start the expired_hold_request_ttl timer | 20:34 |
| tristanC | though, since you can't use autohold feature without zuul, it makes sense that everything autohold related is managed by zuul | 20:36 |
| Shrews | tristanC: that would be too expensive | 20:38 |
| *** pcaruana has quit IRC | 20:38 | |
| Shrews | right now, we validate hold requests after every build completed event. to add node querying for each of those might be a bit much | 20:39 |
| tristanC | Shrews: shouldn't we do that only once current_count >= max_count ? | 20:41 |
| *** jamesmcarthur has joined #zuul | 20:45 | |
| clarkb | jamesmcarthur reminds us that shanghai forum session proposals are open now and end on the 20th. Not sure if we want to do any for zuul? Off the top of my head recent emails and thinking back to denver it might be worth having a zuulv2 -> v3 upgrade session? | 20:47 |
| fungi | i'd hoped we wouldn't still be talking about zuul v2 two years later, but... | 20:55 |
| clarkb | ya I mean its a thing people keep bringing up... having a semi official "here talk about how to upgrade" might help people? | 20:56 |
| corvus | i'd be happy to participate in that in some way; not sure i should run it. | 20:57 |
| clarkb | ya I feel like our move is likely unrepresentative, but I'm also not sure there are any groups that represent typical v2 usage that have done an upgrade yet | 20:58 |
| fungi | you could have stopped at "'m also | 20:59 |
| fungi | not sure there are any groups that represent typical v2 usage" | 20:59 |
| clarkb | fungi: well in my head its people still using Jenkins | 21:00 |
| corvus | it's a range; i've certainly talked to folks who feel they can't live with jenkins plugins. i've also talked to folks who ran into exactly the same issues we did and as a result of restructuring their jobs in the obvious way, are a stones throw away from being able to just flip the switch. | 21:00 |
| clarkb | which we had stopped doing long before upgrading | 21:00 |
| corvus | er, "can't live without jenkins plugins" | 21:00 |
| corvus | (obviously, the second case are the folks who can't live with them) | 21:00 |
| fungi | like, maybe there's some cross-section of folks who all followed the exact same blog and are running essentially the same classes of jobs? that's about as far as "typical" is likely to go | 21:01 |
| fungi | yeah, i guess if you're locked into something like sonotype source code analysis via jenkins plugin or something like that where there's unlikely to be an ansible plugin | 21:02 |
| fungi | javaish stuff | 21:02 |
| fungi | then again, i can't think of any reason ansible couldn't push data into something like that | 21:04 |
| fungi | just would need someone willing to write it | 21:04 |
| clarkb | fungi: likely the biggest barrier is ya that | 21:04 |
| clarkb | anyway thought I would throw the idea out there. I also don't know how many zuulv2 or v3 operators we'll have in shanghai | 21:05 |
| clarkb | I think pabelanger said he won't be there. tristanC ^ would that interest you? | 21:05 |
| corvus | i think the most productive way to look at the summit/forum (is forum summit or ptg now? i can't recall) is in reaching new audiences; this idea seems right on point for that. | 21:08 |
| tristanC | clarkb: I won't be attending in person, but i'd be happy to join remotely if possible | 21:09 |
| *** panda|ruck has quit IRC | 21:10 | |
| jlk | bringing this up again from the weekend. IF we have a night free in ATL to grab dinner, anybody interested in grabbing dinner at http://theoptimistrestaurant.com/ one of the nights (sunday through tuesday)? | 21:11 |
| fungi | i think forum is technically still summit, but some more ptg-ish forum/summit things (like project updates) may have been pushed to the afternoon prior to the ptg. i've not been following the scheduling choices super closely | 21:12 |
| jlk | ( fungi already said yes ) | 21:12 |
| fungi | i am always in the mood to hit up a fish restaurant | 21:12 |
| clarkb | I'd be interested too and have no plans yet | 21:12 |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Use robot_comments in Gerrit https://review.opendev.org/682487 | 21:13 |
| clarkb | re optimist restaurant | 21:13 |
| clarkb | let me try put together and etherpad idea for a zuul forum session and see if others think we should go ahead and propose it | 21:13 |
| jlk | rad | 21:13 |
| corvus | jlk: ++ yes please, and i currently have no social obligations :) | 21:13 |
| jlk | this is for ansiblefest fwiw | 21:14 |
| jlk | I see there is some talk about ptg/forum | 21:14 |
| corvus | we will have different fish in shanghai :) | 21:14 |
| clarkb | ya my response to jlk was re ansiblefest. I land sunday afternoon | 21:14 |
| jlk | kk | 21:14 |
| corvus | same here | 21:14 |
| jlk | sadly I will not be making shanghai. I'd love to go but ... | 21:15 |
| corvus | well, i arrive atl at 7:15pm, so sunday dinner might be sketchy. | 21:16 |
| *** panda has joined #zuul | 21:16 | |
| jlk | sadly I will not be making shanghai. I'd love to go but ... | 21:17 |
| fungi | corvus: just fill up on those delta peanuts | 21:17 |
| jlk | I get in somewhat late on Sunday too, I should have said Monday night through | 21:17 |
| corvus | fungi: i'm on united; i'm hoping for a stroopwafel! | 21:18 |
| fungi | oh, heh | 21:18 |
| fungi | i didn't realize atl let anyone but delta land there ;) | 21:18 |
| jlk | ah I land at 18:30~ | 21:18 |
| * fungi jokes | 21:18 | |
| corvus | (though, i think they decided that's a "morning" snack) | 21:18 |
| paladox | That's going to be a long flight! | 21:19 |
| fungi | it does take a while to cross the continental usa | 21:19 |
| clarkb | https://etherpad.openstack.org/p/zuul-shanghai-forum is a short version of something we might do? | 21:19 |
| corvus | one of the folks at the gerrit hackathon brought stroopwafels. must have had a suitcase full because we ate a lot and they never ran out. | 21:20 |
| paladox | yup, around 6-7 hours i think? | 21:20 |
| fungi | i can be in paris from here faster than i can get to the west coast | 21:20 |
| corvus | (they were way better than united's) | 21:20 |
| clarkb | paladox: ya about 5.5 one direction and 6 the other | 21:20 |
| clarkb | paladox: if direct | 21:20 |
| paladox | yup | 21:20 |
| paladox | 5.5 i think is going back to the US. | 21:20 |
| fungi | granted it takes about the same time to fly across australia too | 21:21 |
| corvus | asia and europe are similar flight times from the western us. it basically means i never fly less than 10 hours. :( | 21:22 |
| clarkb | corvus: tokyo is "only" 9 hours from pdx | 21:22 |
| paladox | clarkb those times are also for london heathrow term 5 :) | 21:23 |
| clarkb | and delta is switching us to haneda from narita in the near future | 21:23 |
| corvus | clarkb: that seems like a good omnibus session. | 21:23 |
| paladox | corvus do you get premium econ? Since in economy that would be horrible sitting on that length of flight :( | 21:27 |
| * fungi is built for basic economy. short legs for the win | 21:30 | |
| paladox | heh | 21:30 |
| corvus | paladox: i haven't yet :( | 21:30 |
| paladox | oh :( | 21:31 |
| corvus | fungi: no one is built to sit that long... | 21:31 |
| fungi | fair | 21:31 |
| fungi | but business class doesn't really solve that problem either | 21:31 |
| corvus | well, if you can lie down you're a little less likely to suffer from dvt | 21:32 |
| paladox | An hour flight to ireland in august was nice, because it wasen't long :) | 21:32 |
| jlk | lying down is good. Getting up to pee due to all the free drinks is good too | 21:32 |
| fungi | oh, well yes, lie-flat business class is a thing now i guess | 21:33 |
| paladox | BA seem to charge for food/drinks depending on ticket. They use M&S simple food. | 21:33 |
| corvus | wow did the image build in https://zuul.opendev.org/t/zuul/build/73068fdb14fa43de8eb864799646bbee/log/job-output.txt#11390 just barely time out? | 21:36 |
| clarkb | looks like it was doing something for another 25 seconds? but ya looks like it | 21:36 |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Use robot_comments in Gerrit https://review.opendev.org/682487 | 21:37 |
| corvus | paladox: i checked that https://review.opendev.org/682473 will cause the "comments only" switch in polygerrit to hide zuul comments | 21:38 |
| corvus | now, in openstack, we ran into this thing where people kept asking us why the CI system wasn't doing anything on their changes, and it's because the config was wrong, and zuul reported the error to gerrit, but hideCI was masking that, so we made it so zuul comments weren't hidden... | 21:39 |
| paladox | awesome! | 21:39 |
| corvus | we may want to do something more sophisticated there, now that we have options. like maybe report normal success/failure with "autogenerated:" but report config errors as normal comments... | 21:40 |
| corvus | or, maybe the checks plugin obviates the problem | 21:40 |
| corvus | (since a config error will show up as a big failed check, rather than the absence of anything) | 21:41 |
| fungi | that seems like a reasonable solution to the problem | 21:41 |
| corvus | my inclination is to go with the intended use of autogenerated to start with and see how it works, and if we find places it doesn't, then change it | 21:42 |
| corvus | maybe not overthink it to start :) | 21:42 |
| fungi | of course | 21:42 |
| paladox | I also shared https://review.opendev.org/#/c/682487/1 in #wikimedia-releng | 21:42 |
| paladox | Come to think of, that patch is all about file comments, not actual comments you leave? | 21:46 |
| corvus | paladox: 682487 is file/line comments; 682473 is review messages | 21:47 |
| paladox | ah | 21:47 |
| clarkb | corvus: I'll go ahead and propose the forum session idea tomorrow then. Give people to provide any feedback between now and then | 21:47 |
| paladox | thanks! | 21:47 |
| * paladox shares https://review.opendev.org/#/c/682473/ | 21:47 | |
| corvus | paladox: i think they are not exclusive, so even a plain line comment that was submitted with an "autogenerated:" tag would still have that. mostly robot_comments additionally allows you to supply a url and suggested fixes (i think both of those may currently be unused, but folks have plans) | 21:49 |
| paladox | ah, yeh. | 21:49 |
| paladox | I haven't used the robot comments feature yet. | 21:49 |
| corvus | from what i gather (i'm not 100% sure about this) there was also some initial confusion about the intended use of tags; i think that's also how we ended up with 2 potential ways of identifying a CI system line comment (does it have an autogenerated: tag? does it have a robot_id?). seems like the best thing to do is both. :) | 21:50 |
| paladox | heh | 21:50 |
| corvus | i should make sure that robot comments actually show up in polygerrit.... lemme do that real quick :) | 21:51 |
| corvus | paladox: yep, this is what they look like: https://imgur.com/a/pOetBlE | 21:55 |
| corvus | it has a little picture of a robot | 21:55 |
| paladox | Awesome! :) | 21:55 |
| paladox | i think that's from gr-icon | 21:55 |
| paladox | or from iron-icon | 21:55 |
| corvus | we should totally do the same in gertty with https://emojipedia.org/robot-face/ | 21:55 |
| paladox | :D | 21:56 |
| clarkb | corvus: my fontactually has that glyph | 21:56 |
| corvus | so the really big idea with robot comments is that they can supply suggested fixes, and then the gerrit ui would have a button next to each and you can say "apply this one" and it will | 21:56 |
| clarkb | though hrm screen/weechat seem unhappy with it | 21:56 |
| clarkb | 🤖 | 21:57 |
| clarkb | weird | 21:57 |
| fungi | yeah, i don't seem to have it for my terminal | 21:57 |
| corvus | aiui that isn't implemented yet, but maybe if we start supplying those with zuul, it would be an impetus for folks to finish that in the ui | 21:57 |
| clarkb | corvus: that is a neat idea. I guess they would be batched up to cut down on patchsets? | 21:58 |
| corvus | clarkb: yeah, i think it works like editing, so you open an edit session with the first, then close it out after you're done clicking 'fix' | 21:58 |
| corvus | or rather, it *will* work like editing | 21:58 |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Use robot_comments in Gerrit https://review.opendev.org/682487 | 21:59 |
| *** tosky has quit IRC | 22:35 | |
| *** jamesmcarthur has quit IRC | 23:37 | |
| *** rlandy has quit IRC | 23:43 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!