| *** zenkuro has quit IRC | 00:09 | |
| *** zenkuro has joined #zuul | 00:10 | |
| pabelanger | I am looking for a job that uses the buildset registry (eg: pushes an image to it) and a child job uses it. | 00:17 |
|---|---|---|
| pabelanger | I am kind of getting confused where some jobs have the intermediate registry too | 00:18 |
| clarkb | pabelanger: opendev/system-config has a number of jobs like that for our various dockerized services | 00:19 |
| clarkb | pabelanger: the intermediate registry exists to store results between changes. The buildset registry stores results within a change (and may fetch from the intermediate registry) | 00:19 |
| clarkb | basically intermediate enables depends-on and buildest enables you to build an image for gitea and gerrit in separate jobs then combine them in a single integration job | 00:20 |
| pabelanger | okay, thanks. I feel I should be looking into docker-build-image role to better understand how the image is push to the buildset registry | 00:23 |
| pabelanger | and how the child job picks that info up | 00:23 |
| clarkb | hvae you seen the zuul-jobs document on the subject /me goes looking for it | 00:26 |
| pabelanger | yup, I've read thought it. I _think_ I understand the workflow, was looking to see example | 00:26 |
| clarkb | https://zuul-ci.org/docs/zuul-jobs/docker-image.html for anyone else looking for it | 00:27 |
| fungi | we also published this about our container job workflow in opendev: https://docs.opendev.org/opendev/base-jobs/latest/docker-image.html | 00:28 |
| clarkb | pabelanger: https://opendev.org/opendev/base-jobs/src/branch/master/zuul.d/jobs.yaml#L144-L233 that is our base jobs | 00:28 |
| clarkb | https://opendev.org/opendev/system-config/src/branch/master/zuul.d/docker-images/base.yaml#L18-L46 is one place we consume them (and the other files in that dir do image specific jobs | 00:28 |
| pabelanger | ++ thanks | 00:28 |
| pabelanger | I can't use the jobs directly, as I need to fit in ansible-builder (ansible new tool for creating containers) | 00:29 |
| pabelanger | much like buildx | 00:29 |
| pabelanger | so, hoping to reuses a lot of the registry bits | 00:29 |
| pabelanger | and push commands | 00:29 |
| pabelanger | I think I'll be able to reuse all execpt yoursite-build-docker-image | 00:39 |
| pabelanger | for that, I'll have to create something specific to ansible-builder | 00:39 |
| *** zenkuro has quit IRC | 00:47 | |
| *** cloudnull has quit IRC | 00:57 | |
| *** cloudnull has joined #zuul | 01:01 | |
| *** Goneri has quit IRC | 01:09 | |
| *** rf0lc0 has quit IRC | 01:23 | |
| *** cloudnull has quit IRC | 02:10 | |
| *** cloudnull has joined #zuul | 02:14 | |
| *** hamalq has quit IRC | 02:39 | |
| *** bhavikdbavishi has joined #zuul | 03:09 | |
| *** bhavikdbavishi1 has joined #zuul | 03:14 | |
| *** bhavikdbavishi has quit IRC | 03:16 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 03:16 | |
| *** bhavikdbavishi has quit IRC | 04:27 | |
| *** bhavikdbavishi has joined #zuul | 04:27 | |
| *** polls45 has quit IRC | 05:28 | |
| *** evrardjp has quit IRC | 05:33 | |
| *** evrardjp has joined #zuul | 05:33 | |
| *** wuchunyang has joined #zuul | 05:57 | |
| *** wuchunyang has quit IRC | 06:11 | |
| *** bhagyashris is now known as bhagyashris|off | 06:17 | |
| *** bhavikdbavishi has quit IRC | 07:03 | |
| *** zenkuro has joined #zuul | 07:05 | |
| *** sduthil has quit IRC | 07:20 | |
| *** zenkuro has quit IRC | 07:21 | |
| *** zenkuro has joined #zuul | 07:22 | |
| *** tosky has joined #zuul | 07:23 | |
| *** sduthil has joined #zuul | 07:28 | |
| *** bhavikdbavishi has joined #zuul | 07:47 | |
| *** bhavikdbavishi1 has joined #zuul | 07:52 | |
| *** bhavikdbavishi has quit IRC | 07:54 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 07:54 | |
| *** cloudnull has quit IRC | 08:00 | |
| *** cloudnull has joined #zuul | 08:00 | |
| *** armstrongs has joined #zuul | 08:10 | |
| *** armstrongs has quit IRC | 08:20 | |
| *** jpena|off is now known as jpena | 08:25 | |
| *** yoctozepto has quit IRC | 08:29 | |
| *** yoctozepto has joined #zuul | 08:30 | |
| openstackgerrit | Albin Vass proposed zuul/nodepool master: Reorganize drivers into separate documents https://review.opendev.org/762208 | 08:49 |
| *** bhavikdbavishi has quit IRC | 08:58 | |
| *** bhavikdbavishi has joined #zuul | 08:58 | |
| *** yolanda__ has joined #zuul | 09:03 | |
| *** yolanda__ has quit IRC | 09:29 | |
| *** yolanda__ has joined #zuul | 09:30 | |
| *** ianychoi has quit IRC | 09:38 | |
| *** mgoddard has quit IRC | 09:39 | |
| *** nils has joined #zuul | 09:57 | |
| *** iurygregory has quit IRC | 10:19 | |
| *** iurygregory has joined #zuul | 10:36 | |
| *** bhavikdbavishi has quit IRC | 10:36 | |
| *** zenkuro has quit IRC | 11:06 | |
| *** zenkuro has joined #zuul | 11:06 | |
| *** iurygregory has quit IRC | 11:11 | |
| *** iurygregory_ has joined #zuul | 11:35 | |
| *** bhavikdbavishi has joined #zuul | 11:35 | |
| *** iurygregory_ is now known as iurygregory | 11:35 | |
| *** rf0lc0 has joined #zuul | 11:55 | |
| *** wuchunyang has joined #zuul | 12:02 | |
| *** wuchunyang has quit IRC | 12:02 | |
| *** wuchunyang has joined #zuul | 12:09 | |
| *** wuchunyang has quit IRC | 12:09 | |
| *** rlandy has joined #zuul | 12:28 | |
| pabelanger | when trying to use the buildset-(container-)registry, I assume you need to use docker login? | 12:28 |
| pabelanger | I'm getting an auth error by trying to pull images from it | 12:28 |
| pabelanger | Just starting to debug now | 12:29 |
| avass | pabelanger: I think there's a role use-buildset-registry or something to set that up for you | 12:35 |
| pabelanger | yah, I thought that was it too | 12:36 |
| avass | pabelanger: https://zuul-ci.org/docs/zuul-jobs/container-roles.html#role-use-buildset-registry | 12:36 |
| *** jpena is now known as jpena|lunch | 12:36 | |
| pabelanger | something isn't working properly then | 12:36 |
| pabelanger | holding a job to debug it now | 12:37 |
| *** zenkuro has quit IRC | 12:48 | |
| *** zenkuro has joined #zuul | 12:49 | |
| pabelanger | okay, I can see config.json being setu | 12:55 |
| pabelanger | setup* | 12:55 |
| pabelanger | something else must be going on | 12:55 |
| *** jpena|lunch is now known as jpena | 13:41 | |
| *** bhavikdbavishi has quit IRC | 14:05 | |
| pabelanger | okay, I think I got it working | 14:07 |
| pabelanger | need to collect docker debug log to confirm | 14:08 |
| *** bhavikdbavishi has joined #zuul | 14:28 | |
| *** Goneri has joined #zuul | 14:48 | |
| openstackgerrit | zbr proposed zuul/zuul-jobs master: DNM: trigger ensure-docker https://review.opendev.org/762352 | 15:03 |
| pabelanger | do we have any jobs that are publishing to quay.io? | 15:04 |
| zbr | mnaser: are you aware of ERROR Project github.com/vexxhost/zuul-base-jobs does not have the default branch master in 3s issues? | 15:08 |
| zbr | i seen failures reported by VEXXHOST CI for a good number of days. | 15:09 |
| mnaser | zbr: yes. i've been trying to get to the bottom of it but no luck. that repo's default branch is `main`. it's configured inside zuul main.yaml to use `main` | 15:09 |
| zbr | haha, main strikes :D | 15:09 |
| mnaser | https://www.irccloud.com/pastebin/3cYIlcQ7/ | 15:10 |
| mnaser | i think it's a zuul bug but it hasn't really been easy to track it down :\ | 15:10 |
| *** zenkuro has quit IRC | 15:11 | |
| *** zenkuro has joined #zuul | 15:11 | |
| zbr | so basically zuul insists around using a master branch | 15:11 |
| mnaser | yeah. | 15:11 |
| corvus | no there's an option to use an alternate branch | 15:12 |
| corvus | mnaser is using that option | 15:12 |
| corvus | there's a bug | 15:12 |
| corvus | that's different than "zuul insists on using" | 15:12 |
| zbr | caching, did this repo ever had a master branch or it was with main from start? | 15:12 |
| mnaser | main from the start | 15:12 |
| mnaser | i don't see the issue when testing a project using 'main' already, so it sounds like it's ignoring that option in specific cases -- or perhaps in cross-connection cases | 15:13 |
| *** Goneri has quit IRC | 15:13 | |
| mnaser | fwiw the option works, just not .. everywhere from what i see :\ | 15:15 |
| *** bhavikdbavishi has quit IRC | 15:16 | |
| mnaser | my search lead me here: https://opendev.org/zuul/zuul/src/branch/master/zuul/configloader.py#L1855 | 15:17 |
| openstackgerrit | Paul Belanger proposed zuul/zuul-jobs master: Include docker_registry when tagging docker-image https://review.opendev.org/762353 | 15:17 |
| pabelanger | I'll need to better understand tests so I can update ^ | 15:18 |
| mnaser | oh hold on | 15:18 |
| mnaser | https://opendev.org/zuul/zuul/src/branch/master/zuul/configloader.py#L2267-L2268 | 15:18 |
| mnaser | should that be `[tpc.load_branch if tpc.load_branch else 'master']` instead? | 15:20 |
| *** zenkuro has quit IRC | 15:24 | |
| openstackgerrit | Mohammed Naser proposed zuul/zuul master: Use load-branch with trusted repository https://review.opendev.org/762354 | 15:28 |
| mnaser | i'm not sure how to test this as there was already tests that were passing :\ | 15:28 |
| *** bhavikdbavishi has joined #zuul | 15:37 | |
| tobiash | mnaser: the tests probably didn't test the dynamic path? | 15:37 |
| mnaser | tobiash: right, that would make sense. i would probably need a little help in figuring out where to build those tests out, configloader is a bit of a far leap for me :) | 15:38 |
| mnaser | oops i added whitespace | 15:38 |
| mnaser | or rather looks like i removed extra whitespace | 15:38 |
| tobiash | most editors can be configured to strip those on save ;) | 15:39 |
| openstackgerrit | Mohammed Naser proposed zuul/zuul master: Use load-branch with trusted repository https://review.opendev.org/762354 | 15:39 |
| mnaser | its too early to look at code | 15:39 |
| tobiash | mnaser: you mean that test_tenant_config_load_branch was successfull in the past right? | 15:40 |
| openstackgerrit | Mohammed Naser proposed zuul/zuul master: Use load-branch with trusted repository https://review.opendev.org/762354 | 15:40 |
| mnaser | tobiash: yes -- and it actually seems to work in an env where i am using 'main' as my branch | 15:41 |
| tobiash | probably except changes to that trusted repo? | 15:41 |
| mnaser | not even .. because if you look at the vexxhost ci failures, that is third-party-ci | 15:41 |
| mnaser | so changes to zuul-jobs fail | 15:41 |
| mnaser | (which in this case is not a trusted repo) | 15:42 |
| tobiash | oh yes, on changes on a different repo since that's the dynamic part | 15:42 |
| tobiash | the speculative layout uses the non-existing config on master branch | 15:42 |
| mnaser | but i wonder why it isn't an issue for my other projects that are using main | 15:42 |
| mnaser | i have github projects that don't have this issue, but they're all using 'main' too | 15:43 |
| tobiash | so that test case could be extended by 1. break config on master 2. add a change that modifies zuul config to an untrusted repo that relies on a correct config on the stable branch | 15:43 |
| tobiash | mnaser: that looks to me that it only breaks for changes touching zuul config | 15:44 |
| *** zenkuro has joined #zuul | 15:44 | |
| mnaser | then maybe the issue is elsewhere. the only thing i am thinking is the fact that they are using different connections tobiash | 15:44 |
| mnaser | so both gerrithub and opendev are reporting this 'complaint', but again, i havent tested a github project using 'master' | 15:45 |
| tobiash | the bugfix you're suggesting is connection agnostic | 15:45 |
| tobiash | (and really fixing a bug) | 15:45 |
| mnaser | hm | 15:46 |
| mnaser | https://opendev.org/zuul/zuul/src/branch/master/zuul/configloader.py#L1606-L1612 | 15:46 |
| mnaser | this is another hard-coded master | 15:46 |
| tobiash | this is resorting the branches so that master is the first if existing | 15:47 |
| tobiash | that's needed so that config on master takes precedence over config on other branches | 15:47 |
| mnaser | ideally i think we should use the default-branch option instead of master there, but that's probably unrelated and we have a few other cases to cover there | 15:48 |
| tobiash | (but is unrelated to load-branch because that's for untrusted repos) | 15:48 |
| mnaser | oh boy | 15:49 |
| mnaser | i really hope this isn't it | 15:49 |
| mnaser | `exclude-unprotected-branches: true` | 15:49 |
| mnaser | i wonder if it's not loading it because maybe it doesn't have protection | 15:49 |
| mnaser | but the error reports 'master' instead of 'main' | 15:49 |
| *** zenkuro has quit IRC | 15:50 | |
| mnaser | https://opendev.org/zuul/zuul/src/branch/master/zuul/executor/server.py#L1109-L1117 | 15:52 |
| mnaser | or https://opendev.org/zuul/zuul/src/branch/master/zuul/executor/server.py#L1843-L1851 | 15:53 |
| mnaser | i don't see the "Checking out" log message, so that tells me its the first iteration that is failing | 15:55 |
| mnaser | so `default-branch` is `master` by default, so it's trying to check that out? | 15:58 |
| *** rfolco has joined #zuul | 15:59 | |
| *** rf0lc0 has quit IRC | 16:01 | |
| tobiash | mnaser: I'd have a test case to your change, shall I update it? | 16:14 |
| mnaser | tobiash: go for it :) | 16:14 |
| tobiash | (although it won't solve your problem actually) | 16:14 |
| openstackgerrit | Paul Belanger proposed zuul/zuul-jobs master: Include docker_registry when tagging docker-image https://review.opendev.org/762353 | 16:15 |
| fungi | zbr: i'm testing out the latest version of your ansi color rendering feature for the webui, do you happen to know a job which runs frequently in opendev which spits out colorized text? if not, i can push up a dnm change somewhere to temporarily create one | 16:15 |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Use load-branch with trusted repository https://review.opendev.org/762354 | 16:16 |
| tobiash | mnaser: ^ | 16:16 |
| tobiash | testing showed that this is a bug but only affects validation of zuul config to the trusted repo | 16:17 |
| *** Goneri has joined #zuul | 16:18 | |
| *** jhesketh has quit IRC | 16:26 | |
| pabelanger | how are people dealing with this error from paramiko when trying to load an RSA key for gerrit | 16:28 |
| pabelanger | http://paste.openstack.org/show/799928/ | 16:28 |
| pabelanger | that is using 2.7.1 of paramiko | 16:28 |
| openstackgerrit | Jeremy Stanley proposed zuul/zuul master: DNM: ANSI Color Test https://review.opendev.org/762363 | 16:32 |
| fungi | pabelanger: if memory serves, the problem has to do with your local key file being in the wrong format | 16:33 |
| fungi | newer openssh generates and expects a new format, not the old pem format | 16:33 |
| fungi | i don't recall what paramiko supports though | 16:33 |
| pabelanger | I've confirmed it is PEM | 16:33 |
| pabelanger | but, paramiko is trying to load ed25519key | 16:34 |
| pabelanger | but, cannot use ed25519 as gerrit doesn't support | 16:34 |
| pabelanger | not sure what is happening here | 16:34 |
| openstackgerrit | Jeremy Stanley proposed zuul/zuul master: DNM: ANSI Color Test https://review.opendev.org/762363 | 16:38 |
| avass | https://review.opendev.org/#/c/762208/ is ready (split drivers into their own documents) | 16:40 |
| avass | pabelanger: that error is misleading | 16:40 |
| avass | pabelanger: I believe it usually means that you don't have access | 16:41 |
| avass | I've seen paramiko throw that when you'd usually get a 'permission denied' | 16:42 |
| avass | pabelanger: https://github.com/paramiko/paramiko/issues/1612 | 16:42 |
| pabelanger | Oh | 16:45 |
| pabelanger | I think you are right | 16:45 |
| pabelanger | I left the port off in my zuul.conf | 16:46 |
| pabelanger | how did that happen | 16:46 |
| pabelanger | hmm | 16:47 |
| pabelanger | okay, I need to take a break | 16:47 |
| avass | :) | 16:48 |
| zbr | fungi: there is one. ANSI DNM job I used at https://review.opendev.org/#/c/761175/ -- use that | 16:50 |
| zbr | fungi: sadly testing ANSI output is very inconvenient, maybe we can alter own zuul build jobs to have a task that produces ansi specially for testing purposes | 16:51 |
| zbr | if quite hard to reach the one particular job that creates the output desired for testing the UI. if we add some tasks for this purpose we may be able to make faster progress. | 16:52 |
| clarkb | shouldn't be difficult if you have a build uuid, that will take you straigt to the build page | 16:59 |
| fungi | yeah, i was able to get that displayed in the draft view | 17:00 |
| zbr | that buildid expires over time, you need to reget an unexperied id, and paste it to the search, after picking the right dropdown item. If you count total number of actions is they are not very few. | 17:21 |
| zbr | also these involve knowing some far from obvious zuul ui-changes tricks (and limitations), not very handy for new users that may want to help with feedback. | 17:22 |
| zbr | if we can improve it, we would all benefit from it, that applies to any UI changes not only that particular ANSI one. | 17:22 |
| zbr | maybe if we can post a UI testing link directly to the build, it would make it easier. i bet most people have no idea about how site preview works, or the reload limitation and that sharing links does not work. | 17:26 |
| clarkb | we do post a UI testing link directly to the build | 17:28 |
| zbr | pabelanger: i remember fixing something similar in elastic-recheck for paramiko. | 17:29 |
| zbr | clarkb: i am not sure where is direct link because I only see the 3 dashboard job where I need 3 click to reach the site preview. | 17:31 |
| clarkb | zbr: maybe I don't understand what you mean then but each of those site previews is a UI testing link | 17:32 |
| clarkb | and is directly provided on each build | 17:32 |
| zbr | no1 click job link no2. click artifacts tab no3. click site preview. | 17:32 |
| zbr | 3 clicks != directly imho | 17:32 |
| clarkb | I mean there is a lot of data we can link to directly in the gerrit comments but doing so would make the gerrit comments useless | 17:33 |
| zbr | directly to me it would be a link named site-preview posted to gerrit. | 17:33 |
| clarkb | I think what we do hits a decent balance between overwhelming the user and giving them what they need to debug jobs | 17:33 |
| clarkb | for some builds that is going to be the ansible console, for some the console stream, others the site preview, and so on | 17:33 |
| clarkb | we make each of those accessible pretty easily | 17:34 |
| clarkb | and yes if you have a human prefilter for you then we can probably provide better links initially, but I'm not sure how we'd do that with a set of machine rules | 17:35 |
| zbr | how many times does a user need something else than site preview when clicking a dashboard job? i would say 99% will want to see sitepreview and nothing else. | 17:35 |
| zbr | i know that is not something easy to address | 17:35 |
| clarkb | I often look at the console stream because js is foreign to me and you get warnings and stuff there | 17:35 |
| zbr | mainly comes to answering the question: what is the main "product" of a succesfull build | 17:36 |
| clarkb | yes, and I think what we've got now strikes a good balance beacuse that differs based on contenxt and the individual behind the keyboard | 17:36 |
| zbr | for unitests I would say is the unittest report, for dashboard or docs is site preview | 17:36 |
| clarkb | we give you an overview with links to the likely candidates | 17:37 |
| clarkb | I never look at the unittest report fwiw (and I wrote the code for it) | 17:37 |
| clarkb | I appreciate people find it to be valuable but I don't, and I think its ok to have those differences | 17:37 |
| corvus | zbr: in opendev's zuul, we're only going to report one link per build back to gerrit, and it has to be the link to the build page because if we report anything else, the user can't navigate back to the build page. | 17:37 |
| zbr | corvus: i know. the only alternative would if we change builds to allow reporting custom data in addition to link to build. | 17:38 |
| corvus | (zuul supports supplying a different url, but we've elected not to use it for the zuul project and generally in opendev) | 17:38 |
| fungi | which would seem to be redundant to the build result page since it already provides those too | 17:39 |
| zbr | maybe if we assure artifacts are visible on the summary page we cut at least one step. | 17:39 |
| pabelanger | zbr: do you remember the issue? I've downgraded paramiko to test again | 17:39 |
| zbr | pabelanger: looking.... | 17:39 |
| corvus | zbr: i think that's worth looking into. that's where i put the artifacts when i added them, but they were moved in the pf4 update. | 17:40 |
| fungi | we did once upon a time link successful docs build results to the docs preview and failures to the build logs (back before we had a build status page) and that while convenient much of the time, became a hindrance if you wanted to view the logs of a successful run | 17:40 |
| zbr | i know that pf4 didn't bring only benefits, but we should be able to address downsides one by one. | 17:40 |
| *** nils has quit IRC | 17:42 | |
| zbr | i am considering finding a way to squeeze artifacts somewhere on the header. maybe as last column, env combining first two columns into a single one. | 17:42 |
| zbr | the first column contains a lot of boilerplate which is useless most of the time. | 17:43 |
| *** jpena is now known as jpena|off | 17:43 | |
| zbr | project/branch/pipeline could be compressed as a single line maybe? | 17:43 |
| zbr | i would have found the change title (in addition to number,patch) as far more useful than these | 17:44 |
| zbr | i personally do not find number being very friendly | 17:45 |
| zbr | also UUID and EvenID have almost no visual value, they are useful only if you want to copy them to clipboard, but not as in being displayed. | 17:46 |
| zbr | i would hide them and expose them as tooltips and only provide a button to copy them to clipboard. | 17:47 |
| zbr | pipeline name could be displayed as a label with rounded corners before the one that express the SUCCESS/FAILURE> | 17:49 |
| avass | zbr: as long as they're easily accessible I wouldn't complain. but they're invaluabe for searching logs | 17:49 |
| zbr | avass: yep, that is why i mentioned ability to copy them easily. | 17:50 |
| zbr | i know that all information in the UI is there because someone found it useful, the trick is to adapt the UI in such way that we improve the experience for most use cases. | 17:51 |
| zbr | i need them too :D | 17:51 |
| zbr | another change that could improve UI is to hide/collapse skipped tasks by default and allow user to unhide them. | 17:53 |
| *** zenkuro has joined #zuul | 18:02 | |
| *** hamalq has joined #zuul | 18:09 | |
| *** hamalq has quit IRC | 18:09 | |
| *** hamalq has joined #zuul | 18:10 | |
| openstackgerrit | Merged zuul/zuul-jobs master: More E208 (final) https://review.opendev.org/761297 | 18:24 |
| *** bhavikdbavishi has quit IRC | 18:50 | |
| *** EmilienM has joined #zuul | 18:57 | |
| *** bhavikdbavishi has joined #zuul | 19:01 | |
| openstackgerrit | Albin Vass proposed zuul/nodepool master: WIP: Digitalocean driver https://review.opendev.org/759559 | 19:08 |
| *** yolanda__ is now known as yolanda | 19:08 | |
| openstackgerrit | Albin Vass proposed zuul/nodepool master: Remove unused labels in gce tests https://review.opendev.org/762392 | 19:18 |
| avass | corvus: i believe you missed removing those ^ :) | 19:19 |
| corvus | avass: yup thx | 19:19 |
| *** bhavikdbavishi has quit IRC | 19:20 | |
| avass | :) | 19:32 |
| openstackgerrit | Albin Vass proposed zuul/nodepool master: Fix loggername for Gce test driver https://review.opendev.org/762399 | 19:32 |
| corvus | i think i should just single-core approve these | 19:33 |
| avass | probably, they shouldn't be affecting any functionality | 19:34 |
| avass | if you got (or an other core) time over the doc update to split drivers into separate documents is ready: https://review.opendev.org/#/c/762208/ | 19:36 |
| *** armstrongs has joined #zuul | 19:58 | |
| *** armstrongs has quit IRC | 20:07 | |
| openstackgerrit | Merged zuul/nodepool master: Remove unused labels in gce tests https://review.opendev.org/762392 | 20:54 |
| openstackgerrit | Merged zuul/nodepool master: Fix loggername for Gce test driver https://review.opendev.org/762399 | 20:54 |
| corvus | avass: +3 with comment; also i happened to note the formatting is messed up with azure (not the fault of your patch); in case you feel like doing another fix. my guess is wrong indentation but i haven't looked. | 21:27 |
| *** rfolco has quit IRC | 21:29 | |
| pabelanger | okay, I am struggling a little bit with buildset-registry, how do namespaces work? I am building an ansible/ansible-runner container image, that works fine. But from the other job, if I pull ansible/ansible-runner, it will come from the buildset registry. If I use quay.io/ansible/ansible-runner, it comes from quay.io. Is there a way to have th buildset registry include the quay.io namespace? | 21:34 |
| *** holser has quit IRC | 21:37 | |
| *** smyers has quit IRC | 21:37 | |
| *** sshnaidm|off has quit IRC | 21:37 | |
| *** bhagyashris|off has quit IRC | 21:37 | |
| *** timburke has quit IRC | 21:37 | |
| *** lyr has quit IRC | 21:37 | |
| *** pleia2 has quit IRC | 21:37 | |
| *** holser has joined #zuul | 21:37 | |
| *** smyers has joined #zuul | 21:37 | |
| *** sshnaidm|off has joined #zuul | 21:37 | |
| *** bhagyashris|off has joined #zuul | 21:37 | |
| *** timburke has joined #zuul | 21:37 | |
| *** lyr has joined #zuul | 21:37 | |
| *** pleia2 has joined #zuul | 21:37 | |
| *** tosky has quit IRC | 21:40 | |
| *** amotoki has quit IRC | 21:40 | |
| *** pots has quit IRC | 21:40 | |
| *** freefood has quit IRC | 21:40 | |
| *** ianw has quit IRC | 21:40 | |
| *** mordred has quit IRC | 21:40 | |
| *** persia has quit IRC | 21:40 | |
| *** yolanda has quit IRC | 21:40 | |
| *** gundalow has quit IRC | 21:40 | |
| *** ericsysmin has quit IRC | 21:40 | |
| *** shanemcd has quit IRC | 21:40 | |
| *** reiterative has quit IRC | 21:40 | |
| *** zer0c00l has quit IRC | 21:40 | |
| *** sean-k-mooney has quit IRC | 21:40 | |
| *** SpamapS has quit IRC | 21:40 | |
| *** systemd has quit IRC | 21:40 | |
| *** mordred has joined #zuul | 21:41 | |
| *** persia has joined #zuul | 21:41 | |
| *** yolanda has joined #zuul | 21:41 | |
| *** gundalow has joined #zuul | 21:41 | |
| *** ericsysmin has joined #zuul | 21:41 | |
| *** shanemcd has joined #zuul | 21:41 | |
| *** reiterative has joined #zuul | 21:41 | |
| *** zer0c00l has joined #zuul | 21:41 | |
| *** sean-k-mooney has joined #zuul | 21:41 | |
| *** SpamapS has joined #zuul | 21:41 | |
| *** systemd has joined #zuul | 21:41 | |
| *** gouthamr_ has quit IRC | 21:41 | |
| *** fbo has quit IRC | 21:41 | |
| *** tosky has joined #zuul | 21:41 | |
| *** amotoki has joined #zuul | 21:41 | |
| *** pots has joined #zuul | 21:41 | |
| *** freefood has joined #zuul | 21:41 | |
| *** ianw has joined #zuul | 21:41 | |
| *** Goneri has quit IRC | 21:41 | |
| *** nhicher has quit IRC | 21:41 | |
| *** mhu has quit IRC | 21:41 | |
| *** Pilou has quit IRC | 21:41 | |
| *** Goneri has joined #zuul | 21:42 | |
| *** nhicher has joined #zuul | 21:42 | |
| *** mhu has joined #zuul | 21:42 | |
| *** fbo has joined #zuul | 21:42 | |
| *** gouthamr_ has joined #zuul | 21:42 | |
| *** Pilou has joined #zuul | 21:42 | |
| *** ericsysmin has quit IRC | 21:43 | |
| corvus | pabelanger: i think first you need to build a quay.io/ansible/ansible-runner image | 21:43 |
| *** gundalow has quit IRC | 21:44 | |
| *** masterpe has quit IRC | 21:44 | |
| *** ericsysmin has joined #zuul | 21:44 | |
| corvus | pabelanger: you'll need the https://zuul-ci.org/docs/zuul-jobs/container-roles.html#role-build-container-image role for that (not build-docker-image) | 21:45 |
| *** mordred has quit IRC | 21:45 | |
| corvus | pabelanger: note that the upload-container-image and promote-container-image (versus -docker-image) roles have not been written yet | 21:45 |
| *** aluria has quit IRC | 21:45 | |
| pabelanger | ah, I am using build-docker-image to create quay.io/ansible/ansible-runner | 21:46 |
| pabelanger | let me switch that out | 21:46 |
| *** iurygregory has quit IRC | 21:46 | |
| *** yoctozepto has quit IRC | 21:46 | |
| *** cloudnull has quit IRC | 21:46 | |
| *** dmsimard has quit IRC | 21:46 | |
| *** bschanzel has quit IRC | 21:46 | |
| *** ttx has quit IRC | 21:46 | |
| *** pabelanger has quit IRC | 21:46 | |
| *** etp has quit IRC | 21:46 | |
| *** tobberydberg_ has quit IRC | 21:46 | |
| *** Miouge has quit IRC | 21:46 | |
| *** frickler has quit IRC | 21:46 | |
| *** arxcruz has quit IRC | 21:46 | |
| *** guillaumec has quit IRC | 21:46 | |
| *** weshay|ruck has quit IRC | 21:46 | |
| *** mmedvede has quit IRC | 21:46 | |
| *** johanssone has quit IRC | 21:46 | |
| *** felixedel has quit IRC | 21:46 | |
| *** spotz has quit IRC | 21:46 | |
| *** iurygregory has joined #zuul | 21:47 | |
| *** yoctozepto has joined #zuul | 21:47 | |
| *** cloudnull has joined #zuul | 21:47 | |
| *** dmsimard has joined #zuul | 21:47 | |
| *** pabelanger has joined #zuul | 21:47 | |
| *** bschanzel has joined #zuul | 21:47 | |
| *** ttx has joined #zuul | 21:47 | |
| *** etp has joined #zuul | 21:47 | |
| *** tobberydberg_ has joined #zuul | 21:47 | |
| *** Miouge has joined #zuul | 21:47 | |
| *** frickler has joined #zuul | 21:47 | |
| *** Eighth_Doctor has quit IRC | 21:47 | |
| corvus | pabelanger: essentially "-docker" is docker-specific and "-container" is agnostic. but we don't have upload or promote for -container yet | 21:47 |
| *** aluria has joined #zuul | 21:48 | |
| pabelanger | that is okay for now | 21:48 |
| corvus | pabelanger: it's possible docker_registry could be used with build-docker-image | 21:48 |
| *** evrardjp has quit IRC | 21:48 | |
| *** johnsom has quit IRC | 21:48 | |
| *** donnyd has quit IRC | 21:48 | |
| *** samccann has quit IRC | 21:48 | |
| *** ChrisShort has quit IRC | 21:48 | |
| *** stevthedev has quit IRC | 21:48 | |
| *** iamweswilson has quit IRC | 21:48 | |
| *** mgagne has quit IRC | 21:48 | |
| *** fdegir has quit IRC | 21:48 | |
| *** fungi has quit IRC | 21:48 | |
| pabelanger | can maybe help write them | 21:48 |
| corvus | pabelanger: haven't tried it https://zuul-ci.org/docs/zuul-jobs/container-roles.html#rolevar-build-docker-image.docker_registry | 21:48 |
| pabelanger | corvus: yah, this lead me to create https://review.opendev.org/762353/ but it doesn't look like docker pull will use the registry | 21:49 |
| corvus | pabelanger: i think the other thing to know is that the docker commands basically don't support ... yeah that | 21:50 |
| pabelanger | it does tag the quay.io/ansible/ansible-runner image properly now, I think | 21:50 |
| pabelanger | k, so flipping to podman might be the next step | 21:50 |
| corvus | pabelanger: because of the way docker is implemented, there just isn't the possibility to have a speculative container registry other than dockerhub | 21:50 |
| *** gundalow has joined #zuul | 21:50 | |
| pabelanger | kk | 21:50 |
| corvus | pabelanger: but the podman tooling supports anything | 21:50 |
| pabelanger | that is likely it | 21:50 |
| *** arxcruz has joined #zuul | 21:50 | |
| *** guillaumec has joined #zuul | 21:50 | |
| *** weshay|ruck has joined #zuul | 21:50 | |
| *** mmedvede has joined #zuul | 21:50 | |
| *** johanssone has joined #zuul | 21:50 | |
| *** felixedel has joined #zuul | 21:50 | |
| *** spotz has joined #zuul | 21:50 | |
| *** evrardjp has joined #zuul | 21:51 | |
| *** johnsom has joined #zuul | 21:51 | |
| *** donnyd has joined #zuul | 21:51 | |
| *** samccann has joined #zuul | 21:51 | |
| *** ChrisShort has joined #zuul | 21:51 | |
| *** stevthedev has joined #zuul | 21:51 | |
| *** iamweswilson has joined #zuul | 21:51 | |
| *** mgagne has joined #zuul | 21:51 | |
| *** fdegir has joined #zuul | 21:51 | |
| *** fungi has joined #zuul | 21:51 | |
| *** SotK has quit IRC | 21:51 | |
| *** tristanC has quit IRC | 21:51 | |
| *** noonedeadpunk has quit IRC | 21:51 | |
| *** avass has quit IRC | 21:51 | |
| *** odyssey4me has quit IRC | 21:51 | |
| *** danpawlik has quit IRC | 21:51 | |
| *** clarkb has quit IRC | 21:51 | |
| *** EmilienM has quit IRC | 21:51 | |
| *** decimuscorvinus has quit IRC | 21:51 | |
| *** jpena|off has quit IRC | 21:51 | |
| *** jkt has quit IRC | 21:51 | |
| corvus | pabelanger: so i'm pretty sure that 'build-container-image' followed by 'podman run quay.io/...' would work; it's possible that 'build-docker-image' followed by 'podman run quay.io' might work but i'm less sure. | 21:51 |
| corvus | pabelanger: and i'm pretty sure that build-*-image followed by 'docker run quay.io' will not work | 21:52 |
| *** EmilienM has joined #zuul | 21:53 | |
| *** decimuscorvinus has joined #zuul | 21:53 | |
| *** jpena|off has joined #zuul | 21:53 | |
| *** jkt has joined #zuul | 21:53 | |
| *** SotK has joined #zuul | 21:53 | |
| *** tristanC has joined #zuul | 21:53 | |
| *** noonedeadpunk has joined #zuul | 21:53 | |
| *** avass has joined #zuul | 21:53 | |
| *** odyssey4me has joined #zuul | 21:53 | |
| *** danpawlik has joined #zuul | 21:53 | |
| *** clarkb has joined #zuul | 21:53 | |
| pabelanger | so much choice | 21:53 |
| pabelanger | :) | 21:53 |
| *** dcastellani has quit IRC | 21:53 | |
| *** dcastellani has joined #zuul | 21:53 | |
| *** dcastellani has quit IRC | 21:53 | |
| *** SotK has quit IRC | 21:53 | |
| *** tristanC has quit IRC | 21:53 | |
| *** noonedeadpunk has quit IRC | 21:53 | |
| *** avass has quit IRC | 21:53 | |
| *** odyssey4me has quit IRC | 21:53 | |
| *** danpawlik has quit IRC | 21:53 | |
| *** clarkb has quit IRC | 21:53 | |
| *** EmilienM has quit IRC | 21:53 | |
| *** decimuscorvinus has quit IRC | 21:53 | |
| *** jpena|off has quit IRC | 21:53 | |
| *** jkt has quit IRC | 21:53 | |
| *** arxcruz has quit IRC | 21:53 | |
| *** guillaumec has quit IRC | 21:53 | |
| *** weshay|ruck has quit IRC | 21:53 | |
| *** mmedvede has quit IRC | 21:53 | |
| *** johanssone has quit IRC | 21:53 | |
| *** felixedel has quit IRC | 21:53 | |
| *** spotz has quit IRC | 21:53 | |
| *** evrardjp has quit IRC | 21:53 | |
| *** johnsom has quit IRC | 21:53 | |
| *** donnyd has quit IRC | 21:53 | |
| *** samccann has quit IRC | 21:53 | |
| *** ChrisShort has quit IRC | 21:53 | |
| *** stevthedev has quit IRC | 21:53 | |
| *** iamweswilson has quit IRC | 21:53 | |
| *** mgagne has quit IRC | 21:53 | |
| *** fdegir has quit IRC | 21:53 | |
| *** fungi has quit IRC | 21:53 | |
| *** iurygregory has quit IRC | 21:53 | |
| *** yoctozepto has quit IRC | 21:53 | |
| *** cloudnull has quit IRC | 21:53 | |
| *** dmsimard has quit IRC | 21:53 | |
| *** bschanzel has quit IRC | 21:53 | |
| *** ttx has quit IRC | 21:53 | |
| *** pabelanger has quit IRC | 21:53 | |
| *** etp has quit IRC | 21:53 | |
| *** tobberydberg_ has quit IRC | 21:53 | |
| *** Miouge has quit IRC | 21:53 | |
| *** frickler has quit IRC | 21:53 | |
| *** Pilou has quit IRC | 21:53 | |
| *** Goneri has quit IRC | 21:53 | |
| *** nhicher has quit IRC | 21:53 | |
| *** mhu has quit IRC | 21:53 | |
| *** gouthamr_ has quit IRC | 21:53 | |
| *** fbo has quit IRC | 21:53 | |
| *** yolanda has quit IRC | 21:53 | |
| *** shanemcd has quit IRC | 21:53 | |
| *** reiterative has quit IRC | 21:53 | |
| *** zer0c00l has quit IRC | 21:53 | |
| *** sean-k-mooney has quit IRC | 21:53 | |
| *** SpamapS has quit IRC | 21:53 | |
| *** systemd has quit IRC | 21:53 | |
| *** persia has quit IRC | 21:53 | |
| *** tosky has quit IRC | 21:53 | |
| *** amotoki has quit IRC | 21:53 | |
| *** pots has quit IRC | 21:53 | |
| *** freefood has quit IRC | 21:53 | |
| *** ianw has quit IRC | 21:53 | |
| *** holser has quit IRC | 21:53 | |
| *** smyers has quit IRC | 21:53 | |
| *** sshnaidm|off has quit IRC | 21:53 | |
| *** bhagyashris|off has quit IRC | 21:53 | |
| *** timburke has quit IRC | 21:53 | |
| *** lyr has quit IRC | 21:53 | |
| *** pleia2 has quit IRC | 21:53 | |
| *** ChanServ has quit IRC | 21:53 | |
| *** ericsysmin has quit IRC | 21:53 | |
| *** rpittau|afk has quit IRC | 21:53 | |
| *** mnaser has quit IRC | 21:53 | |
| *** irclogbot_1 has quit IRC | 21:53 | |
| *** tflink has quit IRC | 21:53 | |
| *** maxamillion has quit IRC | 21:53 | |
| *** mwhahaha has quit IRC | 21:53 | |
| *** gmann has quit IRC | 21:53 | |
| *** ironfoot has quit IRC | 21:53 | |
| *** zbr has quit IRC | 21:53 | |
| *** aluria has quit IRC | 21:53 | |
| *** openstackgerrit has quit IRC | 21:53 | |
| *** gundalow has quit IRC | 21:53 | |
| *** hamalq has quit IRC | 21:53 | |
| *** guilhermesp has quit IRC | 21:53 | |
| *** gouthamr has quit IRC | 21:53 | |
| *** mugsie has quit IRC | 21:53 | |
| *** rlandy has quit IRC | 21:53 | |
| *** paladox has quit IRC | 21:53 | |
| *** mnasiadka has quit IRC | 21:53 | |
| *** aprice has quit IRC | 21:53 | |
| *** erbarr has quit IRC | 21:53 | |
| *** PrinzElvis has quit IRC | 21:53 | |
| *** Open10K8S has quit IRC | 21:53 | |
| *** jbryce has quit IRC | 21:53 | |
| *** webknjaz has quit IRC | 21:53 | |
| *** logan- has quit IRC | 21:53 | |
| *** leoluk has quit IRC | 21:53 | |
| *** zenkuro has quit IRC | 21:53 | |
| *** sduthil has quit IRC | 21:53 | |
| *** sugaar has quit IRC | 21:53 | |
| *** tobiash has quit IRC | 21:53 | |
| *** Tahvok has quit IRC | 21:53 | |
| *** corvus has quit IRC | 21:53 | |
| *** dcastellani has joined #zuul | 21:54 | |
| *** clarkb has joined #zuul | 21:54 | |
| *** danpawlik has joined #zuul | 21:54 | |
| *** odyssey4me has joined #zuul | 21:54 | |
| *** avass has joined #zuul | 21:54 | |
| *** noonedeadpunk has joined #zuul | 21:54 | |
| *** tristanC has joined #zuul | 21:54 | |
| *** SotK has joined #zuul | 21:54 | |
| *** jkt has joined #zuul | 21:54 | |
| *** jpena|off has joined #zuul | 21:54 | |
| *** decimuscorvinus has joined #zuul | 21:54 | |
| *** EmilienM has joined #zuul | 21:54 | |
| *** fungi has joined #zuul | 21:54 | |
| *** fdegir has joined #zuul | 21:54 | |
| *** mgagne has joined #zuul | 21:54 | |
| *** iamweswilson has joined #zuul | 21:54 | |
| *** stevthedev has joined #zuul | 21:54 | |
| *** ChrisShort has joined #zuul | 21:54 | |
| *** samccann has joined #zuul | 21:54 | |
| *** donnyd has joined #zuul | 21:54 | |
| *** johnsom has joined #zuul | 21:54 | |
| *** evrardjp has joined #zuul | 21:54 | |
| *** spotz has joined #zuul | 21:54 | |
| *** felixedel has joined #zuul | 21:54 | |
| *** johanssone has joined #zuul | 21:54 | |
| *** mmedvede has joined #zuul | 21:54 | |
| *** weshay|ruck has joined #zuul | 21:54 | |
| *** guillaumec has joined #zuul | 21:54 | |
| *** arxcruz has joined #zuul | 21:54 | |
| *** gundalow has joined #zuul | 21:54 | |
| *** aluria has joined #zuul | 21:54 | |
| *** frickler has joined #zuul | 21:54 | |
| *** Miouge has joined #zuul | 21:54 | |
| *** tobberydberg_ has joined #zuul | 21:54 | |
| *** etp has joined #zuul | 21:54 | |
| *** ttx has joined #zuul | 21:54 | |
| *** bschanzel has joined #zuul | 21:54 | |
| *** pabelanger has joined #zuul | 21:54 | |
| *** dmsimard has joined #zuul | 21:54 | |
| *** cloudnull has joined #zuul | 21:54 | |
| *** yoctozepto has joined #zuul | 21:54 | |
| *** iurygregory has joined #zuul | 21:54 | |
| *** ericsysmin has joined #zuul | 21:54 | |
| *** Pilou has joined #zuul | 21:54 | |
| *** gouthamr_ has joined #zuul | 21:54 | |
| *** fbo has joined #zuul | 21:54 | |
| *** mhu has joined #zuul | 21:54 | |
| *** nhicher has joined #zuul | 21:54 | |
| *** Goneri has joined #zuul | 21:54 | |
| *** ianw has joined #zuul | 21:54 | |
| *** freefood has joined #zuul | 21:54 | |
| *** pots has joined #zuul | 21:54 | |
| *** amotoki has joined #zuul | 21:54 | |
| *** tosky has joined #zuul | 21:54 | |
| *** systemd has joined #zuul | 21:54 | |
| *** SpamapS has joined #zuul | 21:54 | |
| *** sean-k-mooney has joined #zuul | 21:54 | |
| *** zer0c00l has joined #zuul | 21:54 | |
| *** reiterative has joined #zuul | 21:54 | |
| *** shanemcd has joined #zuul | 21:54 | |
| *** yolanda has joined #zuul | 21:54 | |
| *** persia has joined #zuul | 21:54 | |
| *** pleia2 has joined #zuul | 21:54 | |
| *** lyr has joined #zuul | 21:54 | |
| *** timburke has joined #zuul | 21:54 | |
| *** bhagyashris|off has joined #zuul | 21:54 | |
| *** sshnaidm|off has joined #zuul | 21:54 | |
| *** smyers has joined #zuul | 21:54 | |
| *** holser has joined #zuul | 21:54 | |
| *** hamalq has joined #zuul | 21:54 | |
| *** zenkuro has joined #zuul | 21:54 | |
| *** rlandy has joined #zuul | 21:54 | |
| *** sduthil has joined #zuul | 21:54 | |
| *** paladox has joined #zuul | 21:54 | |
| *** rpittau|afk has joined #zuul | 21:54 | |
| *** tflink has joined #zuul | 21:54 | |
| *** mnaser has joined #zuul | 21:54 | |
| *** guilhermesp has joined #zuul | 21:54 | |
| *** gouthamr has joined #zuul | 21:54 | |
| *** mnasiadka has joined #zuul | 21:54 | |
| *** aprice has joined #zuul | 21:54 | |
| *** erbarr has joined #zuul | 21:54 | |
| *** PrinzElvis has joined #zuul | 21:54 | |
| *** Open10K8S has joined #zuul | 21:54 | |
| *** jbryce has joined #zuul | 21:54 | |
| *** maxamillion has joined #zuul | 21:54 | |
| *** mwhahaha has joined #zuul | 21:54 | |
| *** gmann has joined #zuul | 21:54 | |
| *** webknjaz has joined #zuul | 21:54 | |
| *** sugaar has joined #zuul | 21:54 | |
| *** mugsie has joined #zuul | 21:54 | |
| *** ironfoot has joined #zuul | 21:54 | |
| *** ChanServ has joined #zuul | 21:54 | |
| *** irclogbot_1 has joined #zuul | 21:54 | |
| *** zbr has joined #zuul | 21:54 | |
| *** logan- has joined #zuul | 21:54 | |
| *** openstackgerrit has joined #zuul | 21:54 | |
| *** tobiash has joined #zuul | 21:54 | |
| *** Tahvok has joined #zuul | 21:54 | |
| *** corvus has joined #zuul | 21:54 | |
| *** leoluk has joined #zuul | 21:54 | |
| *** tepper.freenode.net sets mode: +o ChanServ | 21:54 | |
| *** kgz has quit IRC | 21:54 | |
| *** rlandy has quit IRC | 21:54 | |
| *** paladox has quit IRC | 21:54 | |
| *** mnasiadka has quit IRC | 21:54 | |
| *** aprice has quit IRC | 21:54 | |
| *** erbarr has quit IRC | 21:54 | |
| *** PrinzElvis has quit IRC | 21:54 | |
| *** Open10K8S has quit IRC | 21:54 | |
| *** jbryce has quit IRC | 21:54 | |
| *** webknjaz has quit IRC | 21:54 | |
| *** logan- has quit IRC | 21:54 | |
| *** leoluk has quit IRC | 21:54 | |
| *** zenkuro has quit IRC | 21:54 | |
| *** sduthil has quit IRC | 21:54 | |
| *** sugaar has quit IRC | 21:54 | |
| *** tobiash has quit IRC | 21:54 | |
| *** Tahvok has quit IRC | 21:54 | |
| *** corvus has quit IRC | 21:54 | |
| *** rlandy has joined #zuul | 21:54 | |
| *** paladox has joined #zuul | 21:54 | |
| *** mnasiadka has joined #zuul | 21:54 | |
| *** aprice has joined #zuul | 21:54 | |
| *** erbarr has joined #zuul | 21:54 | |
| *** PrinzElvis has joined #zuul | 21:54 | |
| *** Open10K8S has joined #zuul | 21:54 | |
| *** jbryce has joined #zuul | 21:54 | |
| *** webknjaz has joined #zuul | 21:54 | |
| *** logan- has joined #zuul | 21:54 | |
| *** leoluk has joined #zuul | 21:54 | |
| *** logan- has quit IRC | 21:54 | |
| *** zenkuro has joined #zuul | 21:55 | |
| *** sduthil has joined #zuul | 21:55 | |
| *** sugaar has joined #zuul | 21:55 | |
| *** tobiash has joined #zuul | 21:55 | |
| *** Tahvok has joined #zuul | 21:55 | |
| *** corvus has joined #zuul | 21:55 | |
| *** dcastellani has quit IRC | 21:56 | |
| *** mnasiadka has quit IRC | 21:57 | |
| *** PrinzElvis has quit IRC | 21:57 | |
| *** mnasiadka has joined #zuul | 21:58 | |
| *** erbarr has quit IRC | 21:58 | |
| *** PrinzElvis has joined #zuul | 21:58 | |
| *** kgz has joined #zuul | 21:58 | |
| *** logan- has joined #zuul | 21:59 | |
| *** erbarr has joined #zuul | 21:59 | |
| *** dcastellani has joined #zuul | 22:00 | |
| fungi | so much netsplit it all over my screen | 22:12 |
| *** Eighth_Doctor has joined #zuul | 22:15 | |
| openstackgerrit | Ian Wienand proposed zuul/nodepool master: Install podman in container for container-based builds https://review.opendev.org/756643 | 22:17 |
| *** mordred has joined #zuul | 22:25 | |
| openstackgerrit | Merged zuul/nodepool master: Reorganize drivers into separate documents https://review.opendev.org/762208 | 22:38 |
| pabelanger | 2020-11-11 22:55:42.628757 | ubuntu-bionic | Error: Error copying image to the remote destination: Error trying to reuse blob sha256:88557d3f8a58a7ca4c8b429593aede5a9b6b963905c32aae55057503893d9e28 at destination: pinging docker registry returned: Get https://zuul-jobs.buildset-registry:5000/v2/: EOF | 22:56 |
| pabelanger | corvus: have you see that before?^ | 22:56 |
| pabelanger | that is using the build-container-image | 22:56 |
| pabelanger | role* | 22:56 |
| pabelanger | so, backend is podman | 22:57 |
| mwhahaha | might be an oci vs docker format thing | 22:58 |
| mwhahaha | we ended up forcing dockerv2 format because oci got weird when we had to handle container storage issues | 22:59 |
| pabelanger | force it where? | 23:01 |
| *** tosky has quit IRC | 23:08 | |
| pabelanger | `http://paste.openstack.org/show/799935/ | 23:16 |
| pabelanger | http://paste.openstack.org/show/799935/ | 23:16 |
| pabelanger | that is the debug log from podman | 23:16 |
| clarkb | pabelanger: have you tried fetching that url manually? maybe the name isn't in dns? | 23:17 |
| clarkb | maybe the port is different etc | 23:17 |
| clarkb | my hunch is dns though given the fqdn there | 23:17 |
| pabelanger | dns looks right | 23:20 |
| clarkb | what does wget or curl say? | 23:20 |
| clarkb | they may have better debug output | 23:20 |
| pabelanger | Hmm, getting openssl error | 23:22 |
| clarkb | hrm I think the job sets up certs for the clients. May need to ignore ssl verification in wget/curl to debug further | 23:23 |
| clarkb | but maybe the certs aren't set properly for podman? | 23:23 |
| pabelanger | do we know if we have a job using build-container-images role? | 23:24 |
| clarkb | I don't know of anyone using the podman stuff. We tried for a short period of time but it turns out podman != docker and it was easier to just keep dockering | 23:24 |
| clarkb | (specifically had issues with bind mounts) | 23:24 |
| pabelanger | http://paste.openstack.org/show/799936/ | 23:25 |
| pabelanger | that is error I see | 23:25 |
| clarkb | if you do curl --insecure does it change? | 23:27 |
| clarkb | if not then maybe its ssl setup on the registry side | 23:27 |
| pabelanger | no | 23:27 |
| clarkb | ok then ya I would check the registry side's ssl setup | 23:27 |
| pabelanger | yah, that is what I am checking now | 23:27 |
| clarkb | iirc there is config for that, maybe double check the files exist and perms are such that the files can be read | 23:27 |
| corvus | pabelanger: to answer your earlier question: there should be a test job in zuul-jobs which exercises the build-container-image role | 23:28 |
| corvus | pabelanger: zuul-jobs-test-registry-podman i think | 23:30 |
| pabelanger | corvus: ack, thanks | 23:30 |
| pabelanger | http://paste.openstack.org/show/799937/ | 23:30 |
| openstackgerrit | Merged zuul/nodepool master: Ignore unparsable/empty image build ZNode data https://review.opendev.org/746349 | 23:30 |
| pabelanger | that is all I see from zuul-registry | 23:30 |
| corvus | (careful grepping there; it's actually the same playbook used for both roles) | 23:30 |
| pabelanger | it looks like tls certs are mounted into the container | 23:31 |
| *** zenkuro has quit IRC | 23:33 | |
| pabelanger | I think it might be related to ipv6 | 23:36 |
| pabelanger | yah, a ipv4 only node seem to work properly | 23:37 |
| clarkb | the ssl error implies it is talking though which would imply to me that ipv6 is carrying the traffic | 23:40 |
| pabelanger | maybe registry isn't loading certs properly when ipv6 | 23:41 |
| pabelanger | http://paste.openstack.org/show/799938/ | 23:43 |
| pabelanger | that is diff on not working / working with openssl s_client | 23:44 |
| clarkb | weird, I would still look at it from the bottom up and check the ssl configs and the files and so on | 23:44 |
| pabelanger | let me see if I can get into the container | 23:44 |
| pabelanger | the good news, is podman pull worked properly | 23:59 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!