| pabelanger | SpamapS: I don't think I've seen that yet on github | 00:02 |
|---|---|---|
| pabelanger | usually I've seen merge fail because not using merge commits | 00:02 |
| SpamapS | wee, just figured out we're spending about $65/month on zuul test nodes. mostly t3.medium's | 00:02 |
| clarkb | that is cheap | 00:03 |
| SpamapS | (not bad for 8 committers and 5 - 10 test jobs per PR) | 00:03 |
| SpamapS | Our longest test job is 17 minutes. | 00:03 |
| pabelanger | I'm guessing their work life has improved by more then $65 too :) | 00:04 |
| SpamapS | Well, they had CircleCI before | 00:04 |
| SpamapS | and it was fine | 00:04 |
| SpamapS | ;) | 00:04 |
| SpamapS | and nearly free | 00:04 |
| pabelanger | were just bringing online ansible/ansible integration testing for network appliances this week. So far longest is 90mins | 00:04 |
| pabelanger | but that is on 1vcpu / 2gb ram | 00:04 |
| *** rlandy|brb is now known as rlandy | 00:05 | |
| pabelanger | okay, time to upgrade to 3.8.1 | 00:05 |
| SpamapS | Running Zuul itself is harder to peg as far as cost. It runs on a kubernetes cluster with a bunch of other stuff.. but it's by far the biggest thing in that cluster, and that cluster costs about $300/month to run. | 00:05 |
| SpamapS | I wonder if I could get zuul-executor to work with horizontal pod autoscaler. | 00:06 |
| clarkb | re circle ci it is $50/month for 2 containers, first is free | 00:07 |
| pabelanger | I think molecule team is spending $500 a month for travis (dedicated?) | 00:08 |
| pabelanger | I think that is a lot | 00:08 |
| SpamapS | clarkb:ya we were able to fit in the 1 container when I joined. ;) | 00:09 |
| SpamapS | That wouldn't be the case now. :) | 00:09 |
| SpamapS | I'd say overall, Zuul on AWS is pretty cost effective. And I haven't even gone back and gotten custom images going yet. | 00:09 |
| clarkb | ya your numbers seem quite good (which is a good thing for zuul) | 00:10 |
| pabelanger | SpamapS: are you doing helm? | 00:10 |
| SpamapS | That would save probably 10% job execution time, and maybe $1/month in bandwidth. ;) | 00:10 |
| SpamapS | pabelanger:negative, I just have a dir of YAML and spruce that knits it together. | 00:10 |
| pabelanger | (not sure if that is the right tech) | 00:10 |
| SpamapS | I can't stand Helm. | 00:10 |
| pabelanger | ack | 00:10 |
| SpamapS | Good idea. Very frustrating to use. | 00:10 |
| SpamapS | I'd love to publish this as open source, but it's still pretty specific to us. | 00:11 |
| SpamapS | We've been looking at rewriting all our spruce/bash using the Terraform kubernetes provider actually. | 00:11 |
| pabelanger | yah, we've to look at aws / k8s in the future, but right now vexxhost is really nice. zero infra issues so far :) | 00:11 |
| pabelanger | think next step is to get aws for nodepool | 00:12 |
| pabelanger | tower / ansible-runner teams really interested in k8s for nodepool too | 00:12 |
| SpamapS | I tried to get CityCloud on our consideration list but they don't have enough NA regions. Vexxhost was also short on regions and compliance audits. | 00:12 |
| clarkb | SpamapS: you need 3 NA regions? | 00:12 |
| SpamapS | pabelanger:I had a lot of problems with kubernetes in nodepool. It needs some production hardening. | 00:13 |
| SpamapS | clarkb:ya. | 00:13 |
| SpamapS | Well, I could justify 2 | 00:13 |
| clarkb | vexxhost has 2 fwiw | 00:13 |
| SpamapS | but the cloud would have to have a lot of other advantages. | 00:13 |
| pabelanger | SpamapS: yah, signed up today to do some work with nodepool / k8s for opendev. I think if we can get that going, it will be quicker to iterate on issues. | 00:13 |
| clarkb | and citycloud intends to bring their la region back but ya currently is off | 00:13 |
| SpamapS | like, we score for compliance status (financial regulations are pretty nuts), and how green your datacenters are. | 00:13 |
| SpamapS | pabelanger: I'm pretty sure there's just a little mismatch on some things. I didn't have time to figure it out all the way. | 00:14 |
| SpamapS | And combined with the weirdness in the upstream library... | 00:14 |
| SpamapS | I just punted | 00:14 |
| SpamapS | but color me interested | 00:14 |
| SpamapS | we run a lot of tests that don't need to be on a VM. | 00:15 |
| SpamapS | and we have tons of spare capacity on our k8s clusters. | 00:15 |
| clarkb | ya cityclouds whole thing is regulation/certs based hosting | 00:15 |
| clarkb | but mostly in europe | 00:15 |
| SpamapS | clarkb:right, they were the only openstack cloud that made the final round. | 00:15 |
| pabelanger | yah, that's what I am hoping to avoid on zuul.a.c, I don't really have bandwidth to debug it all myself. But with opendev, we are a community :) | 00:15 |
| SpamapS | But lack of NA regions just put them at a crazy disadvantage. | 00:15 |
| pabelanger | and would much rather upstream be the point on new nodepool features first | 00:15 |
| SpamapS | It came down to Google and AWS. And AWS won because we were already there. | 00:16 |
| SpamapS | pabelanger:totally | 00:16 |
| clarkb | SpamapS: you should give them that feedback if you havent already | 00:18 |
| clarkb | johan and tobias aregreat | 00:18 |
| pabelanger | clarkb: is citycloud going to be donating resources again to openstack? | 00:19 |
| clarkb | pabelanger: they sent us email recently saying they would like to but cant currently. I hope they do but would be in the future | 00:19 |
| pabelanger | +1 | 00:21 |
| SpamapS | There's also the.. like.. massive bias in the general community of engineers right now that doesn't let people even consider not-AWS. | 00:22 |
| pabelanger | yup | 00:23 |
| SpamapS | Same bias against not-GitHub | 00:23 |
| SpamapS | luckily, that same bias goes the other way for not-Jenkins ;-) | 00:24 |
| clarkb | what is odd about that bias is you talk to people using aws and they have the same uptime issues and weird behavior problems of the other clouds | 00:25 |
| fungi | yes, but "nobody ever got fired for choosing aws" (safety in numbers argument) | 00:30 |
| pabelanger | zuul 3.8.1 :) | 00:41 |
| *** jamesmcarthur has joined #zuul | 00:41 | |
| pabelanger | now to fix ara-reports | 00:41 |
| SpamapS | clarkb: IAM is such a nightmare.. I almost vetoed staying on AWS because of it. | 00:48 |
| SpamapS | But uptime problems and weird behavior.. most of those are solved for us because we're "Green field" so we have been designing for that from day 1. | 00:49 |
| clarkb | right but it isnt different on another cloud is the point | 00:49 |
| pabelanger | SpamapS: clarkb: since you might still be around, mind a quick review on https://review.opendev.org/657658/ to also publish dev releases of nodepool to pypi (zuul is working properly). | 00:49 |
| clarkb | so you design for it either way | 00:49 |
| SpamapS | clarkb:what is different is that the other clouds don't have (redshift/dynamo/lambda/etc.) | 00:50 |
| SpamapS | Also.. zomg.. having multiple vendors sucks. | 00:51 |
| SpamapS | When we rotate keys, we have to rotate 9 sets of API keys * 3 (dev/stage/prod).. and some of those don't even have APIs to provision credentials... clicky click. | 00:51 |
| SpamapS | pabelanger:I'll look | 00:52 |
| SpamapS | +A | 00:52 |
| clarkb | usually you want multiple vendors for negotisting power and ha | 00:52 |
| clarkb | but ya it adds overhead | 00:52 |
| SpamapS | clarkb:no I have multiple vendors for email verification, URL shortening, captchas, email sending, analytics, cloud, oauth... etc. etc etc. | 00:53 |
| SpamapS | SAML and Oauth help a lot, but for automated API keys.. the industry is, IMO, totally and completely insane. | 00:54 |
| SpamapS | would be *great* if I could just make a service account in GSuite and the vendor would trust oauth. :-/ | 00:55 |
| clarkb | does aws prevent you from sending email or is that a simplicity choice | 00:56 |
| pabelanger | SpamapS: ty | 00:56 |
| clarkb | (seems like half the clouds block smtp and half dont) | 00:57 |
| SpamapS | IIRC they don't prevent, but nobody accepts from their ranges. | 00:59 |
| clarkb | ah | 01:00 |
| SpamapS | They have a pretty simple service for sending. But it's too simple. We use SailThru, which has a ton of tools for making your templates look good on devices, and managing campaigns. | 01:00 |
| SpamapS | And we have a backup provider with Klaviyo | 01:00 |
| SpamapS | because sometimes you just can't trust some random cloud service with your ability to talk to your customers. ;) | 01:01 |
| pabelanger | SpamapS: who are you doing monitoring again with? (I feel I have asked this before) | 01:01 |
| SpamapS | That's the one thing we are building ourselves. ;-) | 01:01 |
| SpamapS | We are using Prometheus + Loki + Thanos | 01:02 |
| SpamapS | https://github.com/grafana/loki | 01:03 |
| SpamapS | https://improbable.io/blog/thanos-prometheus-at-scale | 01:03 |
| pabelanger | bookmarked to later | 01:04 |
| SpamapS | Yeah, we looked at all the datadogs and others and they just didn't give us the single pane that we wanted. | 01:04 |
| SpamapS | Also we have some compliance things that make it hard to retain logging data with 3rd parties. | 01:05 |
| SpamapS | Speaking of logging.. I really want to figure out how to shove the console log through xterm.js. :) | 01:05 |
| pabelanger | yah, I'd love to get elastic-recheck going locally too, but no bandwidth to dive into that. | 01:08 |
| pabelanger | but also hoping to look into our logging spec too | 01:08 |
| SpamapS | We don't have a recheck problem, but we do shove things into elastic and use it for trend analysis so we can alert when something is trending toward an apocalypse, or diverges way off normal trends. | 01:09 |
| *** jamesmcarthur has quit IRC | 01:09 | |
| SpamapS | Pretty early days for that though, so I don't know how effective it will be. | 01:09 |
| openstackgerrit | Merged zuul/nodepool master: Add release-zuul-python to post pipeline https://review.opendev.org/657658 | 01:18 |
| *** rlandy has quit IRC | 02:55 | |
| *** bhavikdbavishi has joined #zuul | 03:31 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: gerrit: add support for report only connection https://review.opendev.org/568216 | 03:42 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: executor: use node python path https://review.opendev.org/637339 | 03:43 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: config: add tenant.toDict() method and REST endpoint https://review.opendev.org/621344 | 03:43 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: executor: enable zuul_return to update Ansible inventory https://review.opendev.org/590092 | 03:44 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: executor: add log_stream_port and log_stream_file settings https://review.opendev.org/535538 | 03:44 |
| openstackgerrit | Homero Pawlowski proposed zuul/zuul master: update zuul from scratch docs https://review.opendev.org/657728 | 04:10 |
| *** raukadah is now known as chandankumar | 04:40 | |
| *** quiquell|off is now known as quiquell | 06:02 | |
| *** yolanda_ has joined #zuul | 06:31 | |
| *** gtema has joined #zuul | 06:44 | |
| *** themroc has joined #zuul | 06:55 | |
| *** jpena|off is now known as jpena | 07:59 | |
| *** electrofelix has joined #zuul | 09:41 | |
| *** bhavikdbavishi has quit IRC | 09:54 | |
| *** persia has quit IRC | 10:10 | |
| *** persia has joined #zuul | 10:11 | |
| *** bhavikdbavishi has joined #zuul | 10:14 | |
| *** zbr|pto has quit IRC | 10:42 | |
| *** panda is now known as panda|lunch | 11:04 | |
| *** bhavikdbavishi has quit IRC | 11:14 | |
| *** gtema has quit IRC | 11:19 | |
| *** gtema has joined #zuul | 11:19 | |
| *** jpena is now known as jpena|lunch | 11:31 | |
| *** quiquell is now known as quiquell|lunch | 11:36 | |
| *** rlandy has joined #zuul | 11:42 | |
| *** quiquell|lunch is now known as quiquell | 12:14 | |
| *** jamesmcarthur has joined #zuul | 12:18 | |
| *** panda|lunch is now known as panda | 12:24 | |
| *** gtema has quit IRC | 12:25 | |
| *** themroc has quit IRC | 12:25 | |
| *** zbr has joined #zuul | 12:25 | |
| *** zbr has quit IRC | 12:26 | |
| *** jpena|lunch is now known as jpena | 12:31 | |
| *** themroc has joined #zuul | 12:32 | |
| *** jamesmcarthur has quit IRC | 12:35 | |
| *** abraverm has quit IRC | 12:36 | |
| *** jamesmcarthur has joined #zuul | 12:47 | |
| *** gtema has joined #zuul | 13:15 | |
| *** quiquell is now known as quiquell|brb | 13:53 | |
| *** jamesmcarthur has quit IRC | 13:53 | |
| *** jamesmcarthur has joined #zuul | 13:53 | |
| *** quiquell|brb is now known as quiquell | 14:15 | |
| *** swest has quit IRC | 14:29 | |
| *** aspiers has joined #zuul | 14:39 | |
| *** rfolco|ruck is now known as rfolco|out | 14:48 | |
| *** quiquell is now known as quiquell|off | 14:54 | |
| *** themroc has quit IRC | 15:00 | |
| Shrews | So, I'm attempting to get the tox remote tests for zuul running locally (again), but it does not seem to be passing the value of ZUUL_SSH_KEY down to ansible. Do we need to add IdentityFile to ssh_args? Wondering how this is working in opendev.... | 15:03 |
| *** markwork has joined #zuul | 15:05 | |
| openstackgerrit | Merged zuul/zuul master: Use zuul-ansible for ARA callback plugin detection https://review.opendev.org/653497 | 15:06 |
| openstackgerrit | Mark Meyer proposed zuul/zuul master: Add initial Bitbucket Server functionality https://review.opendev.org/657837 | 15:17 |
| markwork | Hey, the Bitbucket Server stuff is me :) Any questions? Right now it's just a source, no triggers or reports. Please have a look. | 15:18 |
| *** altlogbot_1 has quit IRC | 15:21 | |
| *** irclogbot_0 has quit IRC | 15:22 | |
| openstackgerrit | Merged zuul/zuul master: Bump lru_cache size to 10 https://review.opendev.org/655173 | 15:22 |
| markwork | The only thing that's really missing is the pagination support for the BitbucketClient, that's going to happen tomorrow | 15:22 |
| *** altlogbot_2 has joined #zuul | 15:23 | |
| *** altlogbot_2 has quit IRC | 15:27 | |
| *** jamesmcarthur has quit IRC | 15:27 | |
| *** altlogbot_0 has joined #zuul | 15:29 | |
| openstackgerrit | Mark Meyer proposed zuul/zuul master: Add initial Bitbucket Server functionality https://review.opendev.org/657837 | 15:30 |
| *** irclogbot_1 has joined #zuul | 15:31 | |
| openstackgerrit | Mark Meyer proposed zuul/zuul master: Add initial Bitbucket Server functionality https://review.opendev.org/657837 | 15:32 |
| aspiers | is there a way to populate the search box on http://zuul.openstack.org/status from the querystring? | 15:34 |
| aspiers | If so, Zuul's "Starting gate jobs" comments on Gerrit jobs could link directly to the relevant jobs which would be helpful | 15:34 |
| SpamapS | markwork:osssssuuummm | 15:45 |
| markwork | SpamapS: the new code got a nice 92% test coverage, I plugged it into my local installation and it basically worked | 15:50 |
| SpamapS | aspiers:yes it populates from the query string, but I forget the names of the fields and they may not be well documented | 15:51 |
| markwork | Next up is Trigger and Report support | 15:51 |
| markwork | That said, it's not Bitbucket Cloud, but Bitbucket Server, the stuff you usually have on-prem (we do). | 15:51 |
| SpamapS | markwork:sweet. I'm reviewing now as I'm excited to welcome you to the party. Please don't take the volume of comments as anything other than a sign that we're very interested in having your work be a part of Zuul. :) | 15:51 |
| markwork | SpamapS: woohoo :) | 15:52 |
| markwork | I think there's some tabs vs spaces problems, but that's easy to correct | 15:52 |
| markwork | SpamapS: how long does it usually take until the test run? | 15:53 |
| SpamapS | markwork:20-30 min | 15:55 |
| markwork | Cool | 15:55 |
| SpamapS | Sometimes longer if they sit in the queue a while. | 15:56 |
| SpamapS | markwork:do you know how to see them running? | 15:56 |
| markwork | SpamapS: no, I think they should leave a post on Gerrit, right? | 15:57 |
| SpamapS | Yeah, if you toggle CI you might get a link | 15:57 |
| SpamapS | http://zuul.openstack.org/status <-- go here and put 'zuul' in the search box | 15:57 |
| SpamapS | http://zuul.opendev.org/t/openstack/status <-- or here, same | 15:58 |
| *** hashar has joined #zuul | 15:58 | |
| *** jamesmcarthur has joined #zuul | 15:58 | |
| markwork | SpamapS: wohoo, they're running, PEP8 failure, tox still in progress | 15:59 |
| SpamapS | markwork:curious.. I've never seen tabs used in python. Are you working in a shop that does that, or just first time python and editor did w'ever? | 16:05 |
| *** chandankumar is now known as raukadah | 16:08 | |
| aspiers | SpamapS: heh, if my feature request had already been implemented, you wouldn't have had to write those 3 lines to markwork | 16:13 |
| aspiers | SpamapS: any pointers to where to find the querystring params? | 16:14 |
| *** mattw4 has joined #zuul | 16:15 | |
| SpamapS | aspiers:right? kizmet I guess. | 16:18 |
| SpamapS | aspiers:the querystring params are deep in the web UI. I think. | 16:18 |
| aspiers | https://git.zuul-ci.org/ is no longer a useful redirect | 16:20 |
| aspiers | it should redirect to https://opendev.org/zuul/zuul | 16:20 |
| aspiers | corvus: ^^^ | 16:20 |
| markwork | SpamapS: new editor, I'm usually on Emacs, now trying Sublime | 16:21 |
| * aspiers has used emacs since 1993 | 16:21 | |
| SpamapS | markwork:kk, gertty (tty interface to gerrit) does not handle the tabs well.. so I think I'll skip reviewing more until you've adjusted that. :) | 16:22 |
| openstackgerrit | Mark Meyer proposed zuul/zuul master: Add initial Bitbucket Server functionality https://review.opendev.org/657837 | 16:22 |
| * SpamapS tries Emacs every few years but the vim is strong in this one | 16:22 | |
| corvus | SpamapS: latest gertty (new release!) renders tabs at 8 space stops | 16:22 |
| markwork | SpamapS: adjusted :) | 16:22 |
| aspiers | SpamapS: kudos for trying at least ;-) | 16:23 |
| SpamapS | markwork:cool, I left some comments on patchset 3 | 16:23 |
| *** jamesmcarthur has quit IRC | 16:23 | |
| SpamapS | Actually May is usually when I try "not vim" for a week.. Been pondering Atom this year | 16:23 |
| *** jamesmcarthur has joined #zuul | 16:23 | |
| SpamapS | Tried it last year and didn't hate it. | 16:23 |
| corvus | aspiers: good point; i think we didn't do anything special for the site roots; though git.zuul-ci.org/cgit/zuul does redirect to a useful place. but i agree, we could probably adjust the site root to go to https://opendev.org/zuul | 16:23 |
| aspiers | corvus: +1 for that | 16:24 |
| aspiers | SpamapS: looks like https://opendev.org/zuul/zuul/src/branch/master/web/src/pages/Status.jsx is the status page | 16:24 |
| aspiers | SpamapS: and it seems the filter can only be set by a cookie, not from querystring :-/ | 16:24 |
| aspiers | unless I missed something | 16:24 |
| aspiers | my React-fu is a bit shakey | 16:24 |
| SpamapS | aspiers:it's possible the react rewrite lost the querystring bits | 16:24 |
| SpamapS | Last time I used them was probably with the old frontend | 16:25 |
| aspiers | corvus: In case you missed the scrollback, I was asking for a way to permalink (well, at least temp-link) to a particular zuul status filter | 16:25 |
| SpamapS | I think it would be ideal if /status/change/##### worked | 16:25 |
| aspiers | corvus: and once that is available, Zuul could use those links in its Gerrit comments | 16:25 |
| SpamapS | which should be pretty simple in React | 16:25 |
| aspiers | SpamapS: agreed | 16:26 |
| corvus | ++ | 16:26 |
| aspiers | Is https://zuul-ci.org/docs/zuul/developer/javascript.html up to date? | 16:27 |
| corvus | aspiers: afaik yes | 16:28 |
| *** gtema has quit IRC | 16:28 | |
| aspiers | corvus: would be nice if there was a docker one-liner to get the web UI up | 16:28 |
| *** hashar has quit IRC | 16:29 | |
| corvus | ++ | 16:30 |
| aspiers | I see there's a Dockerfile | 16:30 |
| aspiers | ... in fact, there's even https://hub.docker.com/r/zuul/zuul-web | 16:30 |
| aspiers | Can't I just pull that? | 16:30 |
| fungi | it may have trouble rendering some stuff if it can't get to the reports db | 16:33 |
| corvus | the zuul-web container has the server-side component; it needs to be run as part of a complete zuul system. if you want to do that locally, you can use the zuul-quickstart. | 16:39 |
| corvus | the js stuff is all static | 16:39 |
| *** jamesmcarthur has quit IRC | 16:40 | |
| corvus | the zuul-web container also has the static js build in it. but the static js is also published standalone at http://tarballs.openstack.org/zuul/zuul-content-latest.tar.gz | 16:40 |
| SpamapS | markwork:left one more comment, but ultimately, this looks like a good start. Thanks for getting us started! | 16:41 |
| corvus | markwork, SpamapS: yay! thanks! :) | 16:43 |
| tobiash | SpamapS: are you using loki alone for logs (so no elk)? | 16:48 |
| AJaeger | tobiash: did you see my comments on https://review.opendev.org/653712 ? | 16:50 |
| SpamapS | tobiash:Both, but ELK is just for searches. Also worth noting, we're just about done with the initial build of the real infra after a brief POC | 16:50 |
| tobiash | AJaeger: yes thanks, but I didn't have time yet to work on that | 16:51 |
| tobiash | AJaeger: and you're totally right | 16:51 |
| AJaeger | tobiash: ;) | 16:52 |
| tobiash | SpamapS: ah cool, we're also using prometheus + thanos and elk for logs | 16:52 |
| tobiash | SpamapS: how satisfied are you with thanos? | 16:52 |
| openstackgerrit | Mark Meyer proposed zuul/zuul master: Add initial Bitbucket Server functionality https://review.opendev.org/657837 | 16:53 |
| tobiash | I think it's awesome, but has some performance problems sometimes | 16:53 |
| SpamapS | tobiash: well we haven't pushed much into it, so it's hard to evaluate, but I like the way it works. | 16:53 |
| SpamapS | We're going to be using it with Dynamo+S3 .. what's your backend for it? | 16:53 |
| tobiash | we're using it with S3 (radosgw) on prem | 16:54 |
| SpamapS | Will look out for performance problems. We really like the single pane of glass for logs+metrics that we get from the combo. :-P | 16:57 |
| AJaeger | /join #openstack-barbican | 17:03 |
| * AJaeger hates initial spaces | 17:03 | |
| *** mattw4 has quit IRC | 17:04 | |
| *** jpena is now known as jpena|off | 17:10 | |
| *** mattw4 has joined #zuul | 17:12 | |
| *** mattw4 has quit IRC | 17:23 | |
| *** jamesmcarthur has joined #zuul | 17:29 | |
| *** ianychoi_ has joined #zuul | 17:29 | |
| *** ianychoi has quit IRC | 17:31 | |
| openstackgerrit | Paul Belanger proposed zuul/nodepool master: Fix version number when installing dev release https://review.opendev.org/657857 | 17:52 |
| pabelanger | ^noticed when I tried to use --version on dev release of nodepool | 17:52 |
| pabelanger | copypasta from zuul | 17:52 |
| openstackgerrit | Paul Belanger proposed zuul/nodepool master: Fix version number when installing dev release https://review.opendev.org/657857 | 17:55 |
| *** jamesmcarthur has quit IRC | 18:04 | |
| *** jamesmcarthur has joined #zuul | 18:06 | |
| markwork | SpamapS: most comments are folded in, I just don't know about #3 testbitbucketconnection.py line 40, I think I still need the _getBitbucketClient | 18:09 |
| openstackgerrit | Mark Meyer proposed zuul/zuul master: Add initial Bitbucket Server functionality https://review.opendev.org/657837 | 18:12 |
| *** mattw4 has joined #zuul | 18:16 | |
| openstackgerrit | Paul Belanger proposed zuul/nodepool master: Fix version number when installing dev release https://review.opendev.org/657857 | 18:18 |
| pabelanger | Shrews: fixed | 18:18 |
| openstackgerrit | Mark Meyer proposed zuul/zuul master: Add initial Bitbucket Server functionality https://review.opendev.org/657837 | 18:26 |
| *** mattw4 has quit IRC | 18:43 | |
| *** mattw4 has joined #zuul | 18:43 | |
| *** panda is now known as panda|off | 19:01 | |
| SpamapS | markwork:I'll take a look in a bit. | 19:01 |
| *** mattw4 has quit IRC | 19:03 | |
| pabelanger | tobiash: I cannot remember if it was you I we discussed the idea of reporting a pre-run failures to builds result page, not just the last for RETRY_LIMIT. Or was somebody else. This was more the idea of help an operator deal with infra-related issue, then develoeprs for job results | 19:04 |
| *** hashar has joined #zuul | 19:09 | |
| *** hashar has quit IRC | 19:15 | |
| *** jamesmcarthur has quit IRC | 19:24 | |
| openstackgerrit | Mark Meyer proposed zuul/zuul master: Add initial Bitbucket Server functionality https://review.opendev.org/657837 | 19:28 |
| markwork | SpamapS: I'll be heading home, it's already half past 9 here. I removed all the smaato.net identifiers and moved everything to the base test. | 19:29 |
| markwork | It's also pep8 clean now. | 19:29 |
| markwork | I'll be back tomorrow. | 19:29 |
| markwork | The only question that I have is: do you want to merge this now, or do you want to have a more complete implementation in a single commit? | 19:30 |
| SpamapS | markwork:we'll probably want more documentation before we merge anything. | 19:35 |
| markwork | SpamapS: I'm fine with that. | 19:36 |
| SpamapS | markwork:but don't despair. We will keep it moving forward as long as you're engaged. :) | 19:36 |
| markwork | I still need to add pagination, to complete this step. | 19:36 |
| SpamapS | Our standards are just kind of high for these sorts of things. :) | 19:36 |
| markwork | But then I would like to make a package out of it and merge this + docs + pagination :) | 19:36 |
| SpamapS | Also I don't know if you're used to gerrit, but it's pretty good about separating changes and managing things as a stack. | 19:37 |
| markwork | No, not used to gerrit | 19:37 |
| markwork | We do have a temp gerrit in-house, but want to tear it down, once the bitbucket integration is online | 19:37 |
| SpamapS | k, don't worry about it. | 19:38 |
| markwork | Cool | 19:38 |
| markwork | Ok, then I'll try to make a package out of this and stack the Triggers and Reports on top | 19:40 |
| markwork | (In different commits) | 19:40 |
| markwork | SpamapS: do you have some good docs on the require and reject filters? I didn't dig too deeply, and before I do, I would like some overview of what they're supposed to do. | 19:43 |
| SpamapS | markwork: require and reject are just places where you can apply filters to the triggers. | 19:44 |
| SpamapS | markwork:https://zuul-ci.org/docs/zuul/user/config.html#attr-pipeline.require | 19:45 |
| SpamapS | markwork:they're primarily used to be sure only items with the right conditions met are acted on. | 19:46 |
| SpamapS | Like, we use it to make sure we don't put something in the gate that doesn't already have one approved review. | 19:46 |
| SpamapS | (we trigger on a label added) | 19:46 |
| SpamapS | and we reject if the label `do-not-merge` is on | 19:47 |
| SpamapS | stuff like that | 19:47 |
| SpamapS | anything you can fetch about an item is fair game. | 19:47 |
| markwork | That sounds good, I would like to give users some ability to comment `#test-deploy:eu-west-1' to trigger a specific test pipeline | 19:48 |
| *** mattw4 has joined #zuul | 19:49 | |
| SpamapS | markwork:yeah I have something similar going on right now. That's not a reject/require though, that's typically a trigger. | 19:58 |
| SpamapS | markwork: the comment is typically a trigger, anyway. You might want to require that it have an approval first so they don't push a bitcoin miner live or something. ;) | 19:59 |
| markwork | SpamapS: ok, thanks. I'm off to bed, already half past 10. | 20:06 |
| markwork | I'll correct myself: waiting for this build and then I'll head to the subway ;) | 20:11 |
| markwork | CI/green, very nice :) | 20:11 |
| *** rf0lc0 has joined #zuul | 20:12 | |
| *** mgoddard has quit IRC | 20:13 | |
| markwork | SpamapS: we want to use parts of Bitbucket's CodeInsights API, we should be able to get most of the functionality that the Gerrit adapter has. That will probably be multiple commit stacked on top of this one. | 20:14 |
| *** yolanda_ has quit IRC | 20:15 | |
| *** rfolco|out has quit IRC | 20:15 | |
| *** mhu has quit IRC | 20:15 | |
| *** markwork has quit IRC | 20:16 | |
| *** mattw4 has quit IRC | 20:19 | |
| *** mhu has joined #zuul | 20:19 | |
| *** mattw4 has joined #zuul | 20:19 | |
| *** mgoddard has joined #zuul | 20:22 | |
| *** yolanda_ has joined #zuul | 20:28 | |
| *** mattw4 has quit IRC | 20:31 | |
| *** markwork has joined #zuul | 20:40 | |
| SpamapS | neat | 20:50 |
| SpamapS | Every once in a while I get excited and want to look in to teaching Zuul to use the GitHub checks API.. | 20:50 |
| *** mattw4 has joined #zuul | 20:51 | |
| SpamapS | but then.. | 20:51 |
| SpamapS | side-tracked | 20:51 |
| markwork | Labels with #deloy:eu-west-1 are also nice | 20:53 |
| markwork | you just have to make zuul see a label change as an event | 20:53 |
| pabelanger | SpamapS: yah, I've a request already for better checks API support from a team in ansible org. I too haven't had time to look | 20:54 |
| openstackgerrit | Dirk Mueller proposed zuul/zuul-jobs master: Switch to TOX_CONSTRAINTS_FILE https://review.opendev.org/657886 | 20:55 |
| *** markwork has quit IRC | 21:00 | |
| *** markwork has joined #zuul | 21:10 | |
| *** dkehn has joined #zuul | 21:25 | |
| *** jamesmcarthur has joined #zuul | 21:37 | |
| *** markwork has quit IRC | 21:43 | |
| *** tjgresha has joined #zuul | 22:07 | |
| *** rlandy is now known as rlandy|bbl | 22:17 | |
| *** jamesmcarthur has quit IRC | 22:20 | |
| *** jamesmcarthur has joined #zuul | 22:47 | |
| *** jamesmcarthur has quit IRC | 23:11 | |
| *** timburke has quit IRC | 23:40 | |
| *** mattw4 has quit IRC | 23:42 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!