| *** _erlon_ has quit IRC | 00:10 | |
| *** armstrongs has joined #zuul | 00:25 | |
| *** Goneri has joined #zuul | 00:34 | |
| *** armstrongs has quit IRC | 00:35 | |
| *** Goneri has quit IRC | 00:58 | |
| openstackgerrit | Akihiro Motoki proposed zuul/zuul-jobs master: nodejs-test-dependencies: Install xdg-utils https://review.opendev.org/c/zuul/zuul-jobs/+/766644 | 01:06 |
|---|---|---|
| *** zenkuro has quit IRC | 01:21 | |
| *** hamalq_ has quit IRC | 02:05 | |
| -openstackstatus- NOTICE: The Gerrit service on review.opendev.org is being restarted quickly to enable support for Git protocol v2, downtime should be less than 5 minutes | 02:55 | |
| *** wuchunyang has joined #zuul | 03:03 | |
| *** bhavikdbavishi has joined #zuul | 03:06 | |
| *** bhavikdbavishi1 has joined #zuul | 03:09 | |
| *** bhavikdbavishi has quit IRC | 03:11 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 03:11 | |
| *** rlandy has quit IRC | 03:13 | |
| *** wuchunyang has quit IRC | 04:01 | |
| *** wuchunyang has joined #zuul | 04:49 | |
| *** wuchunyang has quit IRC | 04:54 | |
| *** wuchunyang has joined #zuul | 04:55 | |
| *** bhavikdbavishi has quit IRC | 05:23 | |
| *** evrardjp_ has quit IRC | 05:33 | |
| *** evrardjp has joined #zuul | 05:33 | |
| *** vishalmanchanda has joined #zuul | 05:47 | |
| *** bhavikdbavishi has joined #zuul | 05:56 | |
| *** bhavikdbavishi has quit IRC | 06:57 | |
| *** nils has joined #zuul | 07:44 | |
| *** zenkuro has joined #zuul | 07:46 | |
| *** bhavikdbavishi has joined #zuul | 07:52 | |
| *** bhavikdbavishi1 has joined #zuul | 07:55 | |
| *** bhavikdbavishi has quit IRC | 07:56 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 07:56 | |
| *** hamalq has joined #zuul | 07:56 | |
| *** rpittau|afk is now known as rpittau | 07:57 | |
| *** hamalq_ has joined #zuul | 07:57 | |
| *** jcapitao has joined #zuul | 08:00 | |
| *** hamalq has quit IRC | 08:01 | |
| *** evrardjp has quit IRC | 08:15 | |
| *** evrardjp has joined #zuul | 08:16 | |
| *** zbr has quit IRC | 08:23 | |
| *** zbr has joined #zuul | 08:23 | |
| *** zbr has quit IRC | 08:26 | |
| *** zbr has joined #zuul | 08:26 | |
| *** tosky has joined #zuul | 08:47 | |
| *** hashar has joined #zuul | 08:51 | |
| *** jpena|off is now known as jpena | 08:56 | |
| *** hamalq_ has quit IRC | 09:15 | |
| *** hamalq has joined #zuul | 09:15 | |
| *** bhavikdbavishi has quit IRC | 09:20 | |
| *** bhavikdbavishi has joined #zuul | 09:20 | |
| *** bhavikdbavishi1 has joined #zuul | 09:53 | |
| *** bhavikdbavishi has quit IRC | 09:56 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 09:56 | |
| *** bhavikdbavishi has quit IRC | 10:02 | |
| *** bhavikdbavishi has joined #zuul | 10:08 | |
| *** hamalq has quit IRC | 10:10 | |
| *** bhavikdbavishi1 has joined #zuul | 10:24 | |
| *** bhavikdbavishi has quit IRC | 10:26 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 10:26 | |
| *** bhavikdbavishi has joined #zuul | 10:27 | |
| *** wuchunyang has quit IRC | 10:30 | |
| *** bhavikdbavishi has quit IRC | 10:46 | |
| avass | zbr: I'm starting to see teams using ansible-lint for their zuul jobs internally. So you might see more issues on github in the future :) | 11:08 |
| zbr | avass: btw, I got full control over the project, it is to be moved to community at any moment. | 11:09 |
| zbr | but i did not had much time to focus on it recently, it does require a lot of time to fix current issues and apparently raising bugs is cheap, but new PRs are not really common. | 11:10 |
| zbr | tripleo is using it in any repo, i am not afraid of lack of popularity, on the opposite... | 11:11 |
| zbr | avass: if you are interested about testing ansible code in general, keep an eye on https://github.com/ansible-community/tox-ansible -- you may find it useful. | 11:12 |
| zbr | the readme is outdated, i hope to update it over the weekend. | 11:13 |
| *** ashbullock has joined #zuul | 11:18 | |
| avass | zbr: sure I'll take a look. But we're currently just testing them by running them like zuul-jobs | 11:23 |
| *** jcapitao is now known as jcapitao_lunch | 11:26 | |
| *** sshnaidm is now known as sshnaidm|off | 11:59 | |
| *** ashbullock has quit IRC | 12:05 | |
| *** bhavikdbavishi has joined #zuul | 12:07 | |
| *** bhavikdbavishi1 has joined #zuul | 12:09 | |
| *** bhavikdbavishi has quit IRC | 12:11 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 12:11 | |
| openstackgerrit | Jonas Sticha proposed zuul/nodepool master: WIP: aws: add support for uploading diskimages https://review.opendev.org/c/zuul/nodepool/+/735217 | 12:20 |
| *** jpena is now known as jpena|lunch | 12:38 | |
| *** jcapitao_lunch is now known as jcapitao | 12:45 | |
| *** bhavikdbavishi has quit IRC | 13:07 | |
| *** rlandy has joined #zuul | 13:07 | |
| *** bhavikdbavishi has joined #zuul | 13:08 | |
| *** fbo|off is now known as fbo | 13:18 | |
| *** bhavikdbavishi has quit IRC | 13:29 | |
| *** jpena|lunch is now known as jpena | 13:36 | |
| *** Goneri has joined #zuul | 14:05 | |
| openstackgerrit | Jan Zerebecki proposed zuul/zuul-jobs master: Switch from Debian Stretch to Buster https://review.opendev.org/c/zuul/zuul-jobs/+/766496 | 14:21 |
| avass | looks like we also got hit by: https://review.opendev.org/c/zuul/zuul/+/765990 and according the gitlab issues the timestamp format is not defined at all | 14:40 |
| avass | sigh.. | 14:40 |
| corvus | avass: can you add a link to the gitlab issue as a comment? | 14:43 |
| avass | sure | 14:44 |
| corvus | (also i left a comment) | 14:45 |
| avass | or as I read it the timestamp is ISO 8601 but that can vary a lot | 14:46 |
| avass | There's also a problem that the time format is different between the api and the webhooks. But I'm not sure if that affcets zuul | 14:50 |
| avass | corvus: https://dateutil.readthedocs.io/en/stable/parser.html#dateutil.parser.isoparse looks like a good choice in that case | 14:51 |
| openstackgerrit | Albin Vass proposed zuul/zuul master: Fix errors regarding parsing timestamps https://review.opendev.org/c/zuul/zuul/+/765990 | 15:06 |
| avass | That probably works a bit better. I'll see if I can set up a test environment and makes sure that works on monday | 15:08 |
| avass | It's been extremely dark here lately. We've had 10h of sun since november and no sun at all in december. | 15:09 |
| corvus | avass: wow, i guess that explains the late-night patches :) | 15:13 |
| avass | corvus: Oh no those are just my regular side project working hours :) | 15:14 |
| openstackgerrit | Vishal Manchanda proposed zuul/zuul-jobs master: [DNM] Testing horizon nodejs job with Chrome browser https://review.opendev.org/c/zuul/zuul-jobs/+/766744 | 15:14 |
| *** jfoufas1 has joined #zuul | 15:15 | |
| *** rpittau is now known as rpittau|afk | 15:15 | |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: Gerrit: add support for patchset level comments https://review.opendev.org/c/zuul/zuul/+/766748 | 15:20 |
| guillaumec | corvus, ^ | 15:21 |
| corvus | guillaumec: awesome thanks! | 15:22 |
| corvus | guillaumec: did you build a gerrit with https://gerrit-review.googlesource.com/c/gerrit/+/291225 to test that? | 15:22 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: use opendev-buildset-registry, build tutorial image one time https://review.opendev.org/c/zuul/zuul/+/733005 | 15:26 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: [DNM] tutorials: push dep image to buildset registry https://review.opendev.org/c/zuul/zuul/+/766076 | 15:26 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: [DNM] run tutorials using gerrit-3.3.0 https://review.opendev.org/c/zuul/zuul/+/766086 | 15:26 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: [DNM] un tutorials using gerrit-3.3.0 patchset_level friendly https://review.opendev.org/c/zuul/zuul/+/766749 | 15:26 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: [DNM] un tutorials using gerrit-3.3.0 patchset_level https://review.opendev.org/c/zuul/zuul/+/766750 | 15:26 |
| guillaumec | corvus, ^, didn't had to | 15:30 |
| *** jfoufas1 has quit IRC | 15:31 | |
| corvus | guillaumec: sorry, i'm not fully awake yet -- how did you get a gerrit with the patchset_level patch applied? | 15:33 |
| guillaumec | Jenkins build output | 15:33 |
| corvus | guillaumec: ah, i see it now in 766749, thx | 15:34 |
| corvus | guillaumec: 766748+2 :) | 15:35 |
| corvus | we might wait until gerrit's 291144 lands, but i don't think there's a downside if we go ahead and merge that (even if 291144 changes and we have to correct, it won't be a big deal) | 15:36 |
| openstackgerrit | Jan Zerebecki proposed zuul/zuul-jobs master: Switch from Debian Stretch to Buster https://review.opendev.org/c/zuul/zuul-jobs/+/766496 | 15:41 |
| openstackgerrit | Jan Zerebecki proposed zuul/zuul-jobs master: Better error output for update-test-platforms.py https://review.opendev.org/c/zuul/zuul-jobs/+/766754 | 15:49 |
| openstackgerrit | Jan Zerebecki proposed zuul/zuul-jobs master: ensure-pip: install virtualenv, it is still used https://review.opendev.org/c/zuul/zuul-jobs/+/766477 | 15:54 |
| openstackgerrit | Jonas Sticha proposed zuul/nodepool master: WIP: aws: add support for uploading diskimages https://review.opendev.org/c/zuul/nodepool/+/735217 | 15:57 |
| openstackgerrit | Jonas Sticha proposed zuul/nodepool master: Replace call to deprecated Thread.isAlive https://review.opendev.org/c/zuul/nodepool/+/766757 | 15:57 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Add option to install kubernetes with kind https://review.opendev.org/c/zuul/zuul-jobs/+/740935 | 15:59 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Update ensure-helm to default to latest version https://review.opendev.org/c/zuul/zuul-jobs/+/762564 | 15:59 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Add helm role https://review.opendev.org/c/zuul/zuul-jobs/+/766758 | 15:59 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Add functional helm test jobs using kind https://review.opendev.org/c/zuul/zuul-jobs/+/766759 | 15:59 |
| avass | If everyone else is pushing their stacks of changes I guess I just have to join in :) | 16:00 |
| *** zbr3 has joined #zuul | 16:00 | |
| *** zbr has quit IRC | 16:01 | |
| *** zbr3 is now known as zbr | 16:01 | |
| openstackgerrit | Albin Vass proposed zuul/zuul master: Clarify when to use zuul_return to pause a job. https://review.opendev.org/c/zuul/zuul/+/766761 | 16:09 |
| *** bhavikdbavishi has joined #zuul | 16:33 | |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Enable git protocol v2 in merger https://review.opendev.org/c/zuul/zuul/+/766765 | 16:50 |
| corvus | clarkb, fungi, ianw: ^ that should improve things, but will require wiping out the local cache of git repos on the mergers/executors. | 16:51 |
| clarkb | corvus: any idea if that will break zuul + git installs where the git version is older? | 16:57 |
| mordred | corvus: +2 with comments | 16:57 |
| corvus | clarkb: i'm assuming that would just be ignored, but let me check | 16:57 |
| clarkb | corvus: cool, and ya if it is just ignored then I think that is a good approach as you won't need specific images or installation details | 16:58 |
| clarkb | (with the downside being clearing out the existing repos as you say, or maybe just manually updating their configs in place?) | 16:58 |
| corvus | clarkb: yeah, i think that's the main thing that warrant's mordred's suggested reno | 16:59 |
| clarkb | ++ | 17:00 |
| corvus | ugh, my assumption was wrong; it barfs if you set that on 2.17 | 17:01 |
| corvus | clarkb, mordred: i'm starting to lean toward not putting this in the zuul source code, and instead only doing it in the docker images (and otherwise, leave it up to deployment or the happenstance of git versions) | 17:03 |
| *** hashar has quit IRC | 17:03 | |
| corvus | so if you use the zuul container images, you'll get v2; otherwise you'll get it if you configure it yourself in your own deployment method, or run on a new enough system where it's default | 17:03 |
| clarkb | corvus: that seems like a good initial approach. Maybe add a bit to the docs suggesting you can enable it if producing your own images | 17:04 |
| clarkb | corvus: pretty sure other tools like github and gitlab also do v2 as well so you'd likely see benefits there too | 17:04 |
| corvus | clarkb: honestly, i don't think it's a big enough deal to document | 17:04 |
| clarkb | k | 17:05 |
| corvus | everyone will get it eventually when everyone runs newer git | 17:05 |
| mordred | GROSS | 17:06 |
| mordred | how common is 2.17 | 17:06 |
| mordred | ? | 17:06 |
| *** bhavikdbavishi has quit IRC | 17:08 | |
| fungi | surprisingly | 17:09 |
| clarkb | centos 8 has 2.18.1 | 17:09 |
| fungi | 2.17 is what ships in ubuntu 18.04 lts | 17:09 |
| fungi | unless you use backports | 17:10 |
| avass | xenial is 2.7 and it wouldn't surprise me if anyone is running that | 17:10 |
| clarkb | corvus: that si a good point about how it will autmatically happen over time | 17:10 |
| fungi | actually ubuntu's bionic-backports doesn't even have newer git like debian's buster-backports does | 17:11 |
| fungi | and even buster was already 2.20 | 17:11 |
| avass | makes me a bit sad since I really want partial clones to become stable. but even if they do it's gonna take a while until zuul would be able to use it | 17:13 |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Enable git protocol v2 in container images https://review.opendev.org/c/zuul/zuul/+/766765 | 17:18 |
| corvus | mordred, clarkb, fungi, avass: ^ maybe something like that? | 17:19 |
| clarkb | corvus: ya that should work. you can also use git config --global protocol.version 2 | 17:20 |
| clarkb | assuming git is already installed at that point in the image | 17:20 |
| clarkb | oh also if we don't bind mount that dir? | 17:20 |
| corvus | oh hrm | 17:20 |
| corvus | VOLUME /var/lib/zuul right below | 17:20 |
| clarkb | /var/lib/zuul:/var/lib/zuul | 17:21 |
| clarkb | ya | 17:21 |
| clarkb | does git do a system level config? | 17:21 |
| clarkb | it does! | 17:21 |
| clarkb | /etc/gitconfig | 17:21 |
| corvus | i reckon that's the answer then | 17:21 |
| clarkb | which will be written using git config --system too | 17:21 |
| corvus | i like that, but i'm unsure if git is installed then | 17:22 |
| corvus | i think so; i think it's installed at RUN /output/install-from-bindep | 17:23 |
| corvus | so i'll try that | 17:23 |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Enable git protocol v2 in container images https://review.opendev.org/c/zuul/zuul/+/766765 | 17:23 |
| zbr | does it makes sense to keep testing gear with py27+py35 only when we really use it in production (zuul) with py38 ? | 17:24 |
| corvus | zbr: gear is a general purpose library | 17:24 |
| corvus | it doesn't exist only for zuul or opendev, and targets an extremely wide range of python versions | 17:25 |
| zbr | that is not tested with any supported version of python. | 17:25 |
| zbr | the only jobs running on it is are py27 and py35 | 17:25 |
| fungi | is the question why not also test gear with newer python? | 17:25 |
| zbr | yeah | 17:25 |
| corvus | then the answer to that question is: go for it! | 17:25 |
| fungi | agreed | 17:25 |
| *** hamalq has joined #zuul | 17:27 | |
| avass | oh I guess partial clones are already supported by github | 17:28 |
| *** hamalq has quit IRC | 17:28 | |
| *** hamalq has joined #zuul | 17:29 | |
| avass | corvus: could be something to look into for the merger at a later point. since that could make it faster by only cloning zuul.yaml/zuul.d | 17:29 |
| avass | requires at least git 2.19 though | 17:29 |
| corvus | avass: that could be useful, but the other thing the merger does is freeze the state of the repo so that every executor gets the same content | 17:30 |
| corvus | and for that it needs to at least know what all the refs are | 17:31 |
| corvus | (and what they point to) | 17:31 |
| *** saneax has quit IRC | 17:32 | |
| avass | ah, but could that at least make the full-reconfig faster? | 17:32 |
| corvus | avass: yeah i think so | 17:33 |
| corvus | the current approach is generally for the mergers/executors to maintain a full cached copy of every repo, and therefore updates should be small deltas. the git protocol v2 will improve that for repos with lots of refs. | 17:36 |
| *** jcapitao has quit IRC | 17:38 | |
| avass | but it still has to clone and checkout a workspace to read the config files doesnt it? | 17:38 |
| avass | if it's lightweight enough with a partial clone maybe a merger could have a number of workers to do that | 17:39 |
| corvus | avass: it doesn't need to clone, just do an incremental update; and it doesn't need to checkout for a full reconfig, it reads it from the git store directly. proposed changes do end up in a workspace because they have to be merged. | 17:40 |
| corvus | avass: the executors do exactly that. i don't think the standalone mergers do, but they could fairly easily. | 17:41 |
| corvus | that == have a bunch of workers | 17:43 |
| avass | oh that makes sense | 17:45 |
| *** jcapitao has joined #zuul | 17:45 | |
| *** tosky has quit IRC | 17:46 | |
| *** hashar has joined #zuul | 17:56 | |
| *** jcapitao has quit IRC | 17:58 | |
| *** jpena is now known as jpena|off | 18:00 | |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Squash some RUN layers in Dockerfile https://review.opendev.org/c/zuul/zuul/+/766772 | 18:11 |
| corvus | tristanC: ^ a quick cleanup prompted by your comment | 18:11 |
| tristanC | corvus: that looks good to me, this is what this run function does automatically: https://softwarefactory-project.io/cgit/software-factory/dhall-containerfile/tree/Containerfile/run.dhall | 18:26 |
| *** ashbullock has joined #zuul | 18:27 | |
| ashbullock | Hey, if anyone has an interest in the bitbucketcloud driver https://review.opendev.org/c/zuul/zuul/+/759003 I was thinking it might be easier to have a quick call to go over it, just conscious it's quite a big change to review! | 18:30 |
| ashbullock | if anyone had a spare 30 minutes sometime next week I could go through it :D | 18:35 |
| *** ashbullock has quit IRC | 18:44 | |
| corvus | oh cool, didn't know that was ready :) | 18:51 |
| *** ashbullock has joined #zuul | 18:51 | |
| *** hashar is now known as hasharDinner | 19:04 | |
| openstackgerrit | Merged zuul/zuul master: Enable git protocol v2 in container images https://review.opendev.org/c/zuul/zuul/+/766765 | 19:05 |
| mordred | corvus: oops in your dockerfile followup | 19:22 |
| corvus | derp | 19:23 |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Squash some RUN layers in Dockerfile https://review.opendev.org/c/zuul/zuul/+/766772 | 19:25 |
| fungi | fungi@ze02:/etc/zuul-executor$ sudo docker-compose exec executor cat /etc/gitconfig | 19:32 |
| fungi | [protocol] | 19:32 |
| fungi | version = 2 | 19:32 |
| fungi | yay! | 19:32 |
| fungi | though that was meant for #opendev, it's still a good yay for in here too | 19:33 |
| mordred | fungi: woot | 19:33 |
| avass | huh you don't use the docker group | 19:35 |
| avass | ? | 19:35 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: Gerrit: add support for patchset level comments https://review.opendev.org/c/zuul/zuul/+/766748 | 19:42 |
| *** ashbullock has quit IRC | 19:46 | |
| fungi | avass: not to my knowledge, we have individual sysadmin accounts and sudo provides a convenient mechanism to log who ran which commands at what times | 19:55 |
| SpamapS | btw, I hope this isn't too.. spammy.. but.. if any of you want to have fun with other FOSS-y folks.. https://k8s.dev/celebration is happening today | 19:55 |
| fungi | in case we're trying to piece together a timeline later for some reason | 19:55 |
| avass | fungi: ah alright that makes sense | 19:56 |
| *** hasharDinner is now known as hashar | 20:19 | |
| *** rfolco has quit IRC | 21:05 | |
| *** tosky has joined #zuul | 21:05 | |
| *** rlandy has quit IRC | 21:05 | |
| corvus | fungi: ze02 doesn't seem to have blown up; we're probably good for a complete restart of the mergers and executors? | 21:21 |
| openstackgerrit | Tatiana Ovchinnikova proposed zuul/zuul-jobs master: [DNM] Testing horizon nodejs job with Firefox https://review.opendev.org/c/zuul/zuul-jobs/+/766793 | 21:22 |
| corvus | (sorry wrong channel) | 21:28 |
| *** josefwells has joined #zuul | 22:02 | |
| *** vishalmanchanda has quit IRC | 22:03 | |
| josefwells | Hey zuulers, trying to bring up a some test containers, using the examples (but using github and my own static nodepool) Most stuff working (and I did have this working several months ago). But nodepool-launcher isn't happy with my host-keys somehow | 22:10 |
| josefwells | 2020-12-11 16:09:09,820 DEBUG nodepool.utils: Added ssh host key: ssh-ed25519 | 22:11 |
| josefwells | I've got - name: host.fqdn labels: ubuntu-18.04, and importantly host-key: ""ssh-ed25519 AAAA....." as returned by "ssh-keyscan -t ed25519 host.fqdn" (well, I don't add in the hostname) | 22:14 |
| *** Goneri has quit IRC | 22:22 | |
| fungi | josefwells: sorry i don't have a good answer, but can you also provide the error message you're seeing from the launcher? | 22:25 |
| fungi | maybe that'll help jog someone's memory | 22:25 |
| josefwells | I see more in the logs for the executor: | 22:34 |
| josefwells | paramiko.ssh_exception.SSHException: Incompatible ssh peer (no acceptable host key) | 22:34 |
| josefwells | I do think I got some ansible logs eventually that initially pointed me to an ssh issue | 22:35 |
| josefwells | but I can't seem to find any now | 22:36 |
| *** zenkuro has quit IRC | 22:44 | |
| fungi | oh, you said nodepool-launcher isn't happy with your host-keys, i assumed that meant you were seeing an error in its logs | 23:13 |
| josefwells | yes, I'm looking at its logs from docker, but I don't have anything from an actual job-run | 23:13 |
| josefwells | I'm going to try to bring up the node as described by the docker compose example | 23:14 |
| *** hashar has quit IRC | 23:57 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!