| *** jamesmcarthur has joined #zuul | 00:09 | |
| *** EmilienM|off is now known as EmilienM | 00:46 | |
| *** evrardjp has quit IRC | 00:52 | |
| *** jhesketh has joined #zuul | 00:52 | |
| *** jamesmcarthur has quit IRC | 00:58 | |
| *** jamesmcarthur has joined #zuul | 01:23 | |
| *** swest has quit IRC | 01:39 | |
| *** yolanda has joined #zuul | 01:52 | |
| *** yolanda has quit IRC | 01:52 | |
| *** yolanda has joined #zuul | 01:53 | |
| *** swest has joined #zuul | 01:54 | |
| *** bhavikdbavishi has joined #zuul | 02:54 | |
| *** bhavikdbavishi1 has joined #zuul | 02:57 | |
| *** bhavikdbavishi has quit IRC | 02:58 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 02:58 | |
| *** jamesmcarthur has quit IRC | 03:09 | |
| *** jamesmcarthur has joined #zuul | 03:10 | |
| *** jamesmcarthur has quit IRC | 03:24 | |
| *** jamesmcarthur has joined #zuul | 03:26 | |
| *** jamesmcarthur has quit IRC | 03:27 | |
| *** jamesmcarthur has joined #zuul | 03:28 | |
| *** bhavikdbavishi has quit IRC | 03:38 | |
| *** bhavikdbavishi has joined #zuul | 03:50 | |
| *** cdearborn has quit IRC | 03:55 | |
| *** threestrands has quit IRC | 04:15 | |
| *** bhavikdbavishi has quit IRC | 04:33 | |
| *** ysandeep|away is now known as ysandeep | 05:46 | |
| *** dpawlik has joined #zuul | 06:02 | |
| openstackgerrit | Simon Westphahl proposed zuul/nodepool master: Expose image build requests in web UI and cli https://review.opendev.org/725810 | 06:19 |
|---|---|---|
| *** kmalloc has quit IRC | 06:36 | |
| *** jamesmcarthur has quit IRC | 06:49 | |
| *** jamesmcarthur has joined #zuul | 06:50 | |
| *** rpittau|afk is now known as rpittau | 06:51 | |
| *** jamesmcarthur has quit IRC | 06:55 | |
| *** hashar has joined #zuul | 07:05 | |
| *** jcapitao has joined #zuul | 07:12 | |
| *** jamesmcarthur has joined #zuul | 07:14 | |
| *** jamesmcarthur has quit IRC | 07:16 | |
| *** jamesmcarthur has joined #zuul | 07:17 | |
| *** jamesmcarthur has quit IRC | 07:22 | |
| *** jamesmcarthur has joined #zuul | 07:23 | |
| *** adam_g has quit IRC | 07:24 | |
| *** adam_g has joined #zuul | 07:24 | |
| *** mnasiadka has quit IRC | 07:25 | |
| *** mnasiadka has joined #zuul | 07:27 | |
| *** fbo|off is now known as fbo|afk | 07:30 | |
| *** evrardjp has joined #zuul | 07:31 | |
| *** evrardjp has quit IRC | 07:32 | |
| *** evrardjp has joined #zuul | 07:34 | |
| *** bhavikdbavishi has joined #zuul | 07:34 | |
| *** bhavikdbavishi1 has joined #zuul | 07:37 | |
| *** bhavikdbavishi has quit IRC | 07:38 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 07:38 | |
| openstackgerrit | Simon Westphahl proposed zuul/nodepool master: Expose image build requests in web UI and cli https://review.opendev.org/725810 | 07:48 |
| *** tosky has joined #zuul | 07:56 | |
| *** jpena|off is now known as jpena | 07:57 | |
| *** ysandeep is now known as ysandeep|lunch | 08:21 | |
| *** panda|off is now known as panda | 08:34 | |
| *** asaleh_ has joined #zuul | 08:34 | |
| *** jamesmcarthur has quit IRC | 08:35 | |
| swest | ianw: I noticed a problem with the diskimage-builder. jobs are failing now because the global openstack upper constraints require networkx 2.4, but diskimage-builder requires <2.3 because of Python 2 compatibility. Anything I can do about that? Maybe drop support for Python 2 as proposed by mordred :) | 09:08 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: [WIP] Builds: add "held" attribute https://review.opendev.org/728994 | 09:22 |
| *** hashar is now known as hasharAway | 09:25 | |
| *** sshnaidm|afk is now known as sshnaidm | 09:40 | |
| *** ysandeep|lunch is now known as ysandeep | 09:51 | |
| mhu | I keep forgetting what the use cases are for the "zuul promote" command in the CLI - can someone give me an example that I can shamelessly steal and add to the documentation? :) | 09:56 |
| *** sgw has quit IRC | 10:07 | |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul-jobs master: DNM: test https://review.opendev.org/728640 | 10:11 |
| ianw | swest: yeah, i'll look tomorrow, i guess that just means the end of python2 for dib. because it's branchless i've been holding off but seems there's probably no choice | 10:13 |
| *** sugaar has quit IRC | 10:21 | |
| avass | zuul-maint: I've found that a lot of users are confusing the executor with the remote worker nodes, I believe it's because they're not part of the diagram: https://zuul-ci.org/docs/zuul/discussion/components.html#overview | 10:22 |
| avass | should we update that to show that the executors communicate with worker nodes as well? | 10:22 |
| tobiash | ++ | 10:23 |
| zbr | apparently ensure-docker got broken centos-8: https://storage.bhs.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_ee1/728640/2/check/zuul-jobs-test-ensure-docker-centos-8/ee118f9/job-output.txt | 10:24 |
| *** asaleh_ has quit IRC | 10:26 | |
| *** jcapitao is now known as jcapitao_lunch | 10:27 | |
| *** sugaar has joined #zuul | 10:29 | |
| *** rpittau is now known as rpittau|bbl | 10:31 | |
| avass | zbr: looks like it, how is easy would it be to get it working? | 10:31 |
| zbr | i see a new release from them, but locally I was able to upgrade without any problems, so not sure what is wrong | 10:32 |
| zbr | what is weird is that the server installed is very old | 10:32 |
| zbr | yep, that is the issue, cli is newer but the server is not. | 10:33 |
| avass | zbr: yeah, I don't have time to look into it at the moment, lots of local support oday and I've got a meeting in 30min | 10:33 |
| *** bhavikdbavishi has quit IRC | 10:34 | |
| *** jamesmcarthur has joined #zuul | 10:36 | |
| *** jamesmcarthur has quit IRC | 10:40 | |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul-jobs master: ensure-docker: workaround for centos-8 conflicts https://review.opendev.org/703053 | 10:57 |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul-jobs master: ensure-docker: workaround for centos-8 conflicts https://review.opendev.org/703053 | 11:05 |
| zbr | funny, that patch was raised in january.... | 11:06 |
| *** fbo|afk is now known as fbo | 11:14 | |
| *** jpena is now known as jpena|lunch | 11:33 | |
| *** rfolco|rover|off has joined #zuul | 11:42 | |
| *** rfolco|rover|off is now known as rfolco|rover | 11:44 | |
| *** hasharAway is now known as hashar | 11:51 | |
| *** bhavikdbavishi has joined #zuul | 11:55 | |
| *** asaleh_ has joined #zuul | 11:58 | |
| *** jcapitao_lunch is now known as jcapitao | 12:01 | |
| *** armstrongs has quit IRC | 12:03 | |
| *** rpittau|bbl is now known as rpittau | 12:04 | |
| tristanC | corvus: is there a plan to tag a new zuul and nodepool version soon? | 12:23 |
| *** rlandy has joined #zuul | 12:28 | |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Don't require tox_envlist https://review.opendev.org/726829 | 12:32 |
| avass | clarkb: how about that ^ ? | 12:33 |
| *** jpena|lunch is now known as jpena | 12:33 | |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: Zuul CLI: add console-stream command https://review.opendev.org/729236 | 12:36 |
| avass | zbr: is 703053 re-implementing a previous workaround? | 12:37 |
| *** ysandeep is now known as ysandeep|brb | 12:44 | |
| openstackgerrit | Fabien Boucher proposed zuul/zuul master: Remove another shebang and remove useless exec bits https://review.opendev.org/729240 | 12:46 |
| zbr | avass: yep, the workaround was known for long time, and i suspect the hack was implemented directly on the image. | 12:49 |
| zbr | also used on ansible zuul, see https://github.com/ansible/ansible-zuul-jobs/blob/09ceb6752b46fefa7ff54752febdfded12e87b5c/playbooks/ansible-tox-molecule/pre.yaml#L16 | 12:50 |
| zbr | there is something weird because the rpm should have being found anyway, see https://download.docker.com/linux/centos/7/x86_64/stable/Packages/ | 12:51 |
| *** sgw has joined #zuul | 13:05 | |
| *** Goneri has joined #zuul | 13:09 | |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: Builds: add "held" attribute https://review.opendev.org/728994 | 13:12 |
| *** ysandeep|brb is now known as ysandeep | 13:22 | |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul-jobs master: ensure-docker: workaround for centos-8 conflicts https://review.opendev.org/703053 | 13:37 |
| *** cdearborn has joined #zuul | 13:40 | |
| *** bhavikdbavishi has quit IRC | 13:48 | |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: [WIP] Web UI: add "held" information, filtering in Builds page https://review.opendev.org/729265 | 13:53 |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul-jobs master: ensure-docker: workaround for centos-8 conflicts https://review.opendev.org/703053 | 13:57 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: Zuul CLI: add console-stream command https://review.opendev.org/729236 | 13:57 |
| *** hashar has quit IRC | 13:58 | |
| openstackgerrit | Oleksandr Kozachenko proposed zuul/zuul-jobs master: Add DaemonSet check for wait-for-pods role https://review.opendev.org/728503 | 14:00 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: Builds: add "held" attribute https://review.opendev.org/728994 | 14:13 |
| *** newbie2020 has joined #zuul | 14:14 | |
| newbie2020 | Hi guys, | 14:14 |
| newbie2020 | I have a situation where a periodic pipeline (one using the trigger driver) triggers a Zuul job executing a task in the PRE-RUN phase multiple times from *different* branches (e.g. @master and @branch_foo) | 14:16 |
| newbie2020 | Any idea about what may be going on? | 14:16 |
| fungi | if you don't limit the job to specific branches, i believe the timer trigger will add a build for each branch in the repository | 14:17 |
| newbie2020 | I added branches: master to the job, but probably that is not correct | 14:19 |
| newbie2020 | also I was not expecting having one build execution with tasks executed multiple time... That is what really puzzles me | 14:19 |
| fungi | oh, it's multiple playbook invocations in a single build? not different builds? | 14:22 |
| fungi | job inheritence is additive for pre-run and post-run phase playbooks, could the job have a parent/ancestry which is adding the extra pre-run playbooks? | 14:23 |
| newbie2020 | it does not seem to be the case, but that sounds like a good idea about what to check... | 14:26 |
| newbie2020 | they refer to different @branches | 14:26 |
| newbie2020 | though | 14:26 |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul-jobs master: WIP: ensure-docker: workaround for centos-8 conflicts https://review.opendev.org/703053 | 14:26 |
| fungi | if you're archiving the zuul-info/inventory.yaml for your jobs you should be able to look at all the variables which went into it, like this: https://zuul.opendev.org/t/openstack/build/929f7c0817e14209bd6a971521197a94/log/zuul-info/inventory.yaml | 14:30 |
| fungi | in particular all.vars.zuul._inheritance_path shows the lineage from which it inherits | 14:30 |
| fungi | including project and branch | 14:31 |
| fungi | you can see that example has 3 pre and 3 post playbooks inherited from its ancestor job definitions: https://zuul.opendev.org/t/openstack/build/929f7c0817e14209bd6a971521197a94/console | 14:33 |
| *** jamesmcarthur has joined #zuul | 14:36 | |
| corvus | newbie2020: i can't be 100% sure without seeing the definitions, but from reading your conversation with fungi, it sounds like you may have put a "master" branch matcher on a job definition in branch "branch_foo" (ie, "branches: master"), which would mean that when zuul runs a job for the master branch, it applies the configuration from the foo branch. you would be able to confirm that by looking at the | 14:38 |
| corvus | inheritance path as fungi suggests. | 14:38 |
| corvus | newbie2020: normally you don't need to add branch matchers manually if they are on jobs in a repo with branches (they should only apply to the branches they are in). if you only want the job to run for the master branch, then only add it to the project stanza in the master branch. | 14:40 |
| fungi | yeah, that seems like a plausible explanation | 14:40 |
| corvus | tristanC: we were hoping to switch to using the zk tls stuff in opendev before doing a release, but that's taking longer than expected. maybe if we haven't finished that up this week, we should go ahead and do a release next week regardless. | 14:41 |
| *** zxiiro has joined #zuul | 14:42 | |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul-jobs master: WIP: ensure-docker: workaround for centos-8 conflicts https://review.opendev.org/703053 | 15:01 |
| newbie2020 | @corvus @fungi thanks for the suggestions, I will look into it ! | 15:03 |
| *** bhavikdbavishi has joined #zuul | 15:08 | |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul-jobs master: WIP: ensure-docker: workaround for centos-8 conflicts https://review.opendev.org/703053 | 15:12 |
| *** bhavikdbavishi has quit IRC | 15:23 | |
| *** bhavikdbavishi has joined #zuul | 15:39 | |
| *** ysandeep is now known as ysandeep|afk | 15:45 | |
| mnaser | AJaeger: is https://review.opendev.org/#/c/728503/ satisfactory to you now? :) | 15:49 |
| AJaeger | mnaser: there's nothing like a good commit message;) +2A | 15:51 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: [WIP] web UI: add Autoholds Requests page https://review.opendev.org/729307 | 15:51 |
| zbr | i need advice on what to do about use of yum_repository inside configure mirrors, i need to add an option that is not supported by the module. | 15:52 |
| zbr | ... one of the reasons why most people avoid this module | 15:52 |
| zbr | another one being that is not able to edit repos, only to rewrite them | 15:53 |
| mordred | zbr: I do not feel strongly that we should use it - especially if it's deficient. I think we have plenty of code that does not use apt_repository to install a new apt repository | 15:53 |
| zbr | https://zuul.opendev.org/t/zuul/build/c0803b7f2061452b95ec29197a3ed821 | 15:53 |
| mordred | I think the best reason to use those modules is if they are easier to use / read - but if they aren't, then they are not worth it | 15:53 |
| zbr | our use of it is not very advanced, probably we could replace it with either ini_file (bit slow due to looping) or a jinja2 template. | 15:54 |
| zbr | the template has downside of being ugly too, if we want to loop over received key/value arguments. | 15:55 |
| zbr | or I can add one single custom ini_file command inside ensure-docker, in order to add the magic module_hotfixes=True | 15:55 |
| mordred | zbr: I don't see any issue with doing that | 15:56 |
| zbr | so which path to take? | 15:56 |
| mordred | zbr: you mean basically do yum_repository: blah then ini_file: module_hotfixes yeah? | 15:56 |
| zbr | i would go for ini, could same be 3-4h | 15:56 |
| zbr | save | 15:56 |
| zbr | mordred: thanks, back to patching it, i really need to fix it asap as it breaks all current jobs using docker on centos-8. | 15:57 |
| mordred | zbr: yah. I think whatever work here should be fine | 15:58 |
| *** jcapitao has quit IRC | 16:02 | |
| openstackgerrit | Merged zuul/zuul-jobs master: Add DaemonSet check for wait-for-pods role https://review.opendev.org/728503 | 16:03 |
| AJaeger | clarkb, corvus, want to review again the tox-envlist change https://review.opendev.org/#/c/726829/ ? | 16:05 |
| mhu | corvus: regarding https://review.opendev.org/#/c/728118/ I've reworked the patch to ensure most endpoints can be tenant-scoped for whitelabeling | 16:06 |
| mhu | I've kept some unscoped paths to reflect the RPC but they all have a scoped equivalent | 16:07 |
| mhu | except for connection-specific endpoints | 16:07 |
| clarkb | AJaeger: avass done. I've not approved it since corvus may want to look too. Also that comment will hopefully prevent regressions in the future when we try to simplify :) | 16:07 |
| *** yolanda has quit IRC | 16:07 | |
| clarkb | thanks for the updated patchset there | 16:08 |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul-jobs master: WIP: ensure-docker: workaround for centos-8 conflicts https://review.opendev.org/703053 | 16:11 |
| mhu | can I get some eyes on https://review.opendev.org/728061 https://review.opendev.org/728073 and https://review.opendev.org/728098 please? They fix problems in the CLI when using the REST API and add some testing | 16:14 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: Builds: add "held" attribute https://review.opendev.org/728994 | 16:15 |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul-jobs master: WIP: ensure-docker: workaround for centos-8 conflicts https://review.opendev.org/703053 | 16:16 |
| corvus | AJaeger, clarkb, avass: ack +3 thx! | 16:17 |
| AJaeger | thanks | 16:17 |
| *** hashar has joined #zuul | 16:18 | |
| corvus | mhu: why do you want the unscoped versions? | 16:18 |
| *** tumble has joined #zuul | 16:18 | |
| corvus | mhu: i think the rule is that the zuul api should be tenant-scoped (with the exception of the /info and /tenants endpoints) | 16:18 |
| corvus | mhu: just like the web interface itself | 16:19 |
| mhu | corvus, well I can live without them :) but some of these actions do not require a tenant | 16:19 |
| corvus | mhu: a different way of looking at it is that some access scopes work for all tenants | 16:20 |
| corvus | we worked really hard to keep everything separated across tenants, so i hesitate to start breaking down those barriers | 16:20 |
| mhu | corvus, understandable | 16:21 |
| mhu | would it be okay to keep connections related endpoints unscoped? | 16:21 |
| corvus | what uses the connections endpoint? | 16:22 |
| *** rpittau is now known as rpittau|afk | 16:22 | |
| tobiash | corvus: webhooks and I guess public keys | 16:22 |
| mhu | tobiash, corvus GithubWebController for example | 16:23 |
| corvus | hrm, so we made it so that you can't use github with a whitelabeled zuul install? | 16:23 |
| mhu | corvus, good question | 16:24 |
| tobiash | I thought the public keys are also available with tenant scope | 16:24 |
| tobiash | not sure about the webhooks though | 16:24 |
| corvus | tobiash: i'm pretty sure the keys are tenant-scoped | 16:25 |
| mhu | tobiash, I don't think the webhooks are scoped, according to the way they're mounted in cherrypy | 16:25 |
| corvus | '/api/tenant/{tenant}/key/{project:.*}.pub' | 16:25 |
| tobiash | so it's just the webhooks | 16:25 |
| mhu | tobiash, corvus yeah the webhooks are mounted under /api/connection/{connection_name} | 16:26 |
| mhu | everything else besides info & tenants can be tenant scoped | 16:27 |
| mordred | we could also add a tenant-scoped api call for connections (so that whitelables work) that just works the same and ignores the tenant scope | 16:28 |
| corvus | yeah, and i think mhu has that in his change | 16:28 |
| mordred | ah - cool | 16:28 |
| tobiash | yes, just the driver endpoints are not tenant scoped | 16:28 |
| mhu | mordred, corvus just for connections | 16:28 |
| corvus | oh, the list, not the driver sub-mounts | 16:28 |
| mhu | not connection/{connection_name} though | 16:28 |
| tobiash | basically only the payload mount | 16:28 |
| corvus | okay, well, easy enough to add too... | 16:29 |
| mhu | alright, I'll have a go at it, thanks! | 16:33 |
| corvus | mhu, mordred, tobiash: there are tradeoffs here: having it not be tenant scoped makes it clear that, well, it's not tenant scoped. but it does make it harder for someone running in a whitelabel setup -- it's not impossible, but we need to tell them to route /connection/ differently. it's also entirely possible we might want to tenant-scope connections in the future (but then doing that might make it | 16:33 |
| corvus | difficult to set up the webook in the first place -- should you configure the tenant-scoped url or the global one?) | 16:33 |
| tobiash | corvus: does zuul-web know if it's whitelabeled? | 16:34 |
| corvus | tobiash: i don't think so | 16:34 |
| openstackgerrit | Merged zuul/zuul-jobs master: Don't require tox_envlist https://review.opendev.org/726829 | 16:35 |
| tobiash | corvus: the connections are a global entity so I think it's correct that they're not scoped to tenants. However I see that this makes it impossible to deliver webhooks to whitelabeled deployments. What do you think about just duplicating the connections to also be available tenant scoped? | 16:36 |
| corvus | oh wow, we already documented the special configuration for api/connection | 16:36 |
| corvus | https://zuul-ci.org/docs/zuul/howtos/installation.html#white-labeled-tenant | 16:37 |
| tobiash | or that... | 16:37 |
| tobiash | :) | 16:37 |
| *** dpawlik has quit IRC | 16:37 | |
| corvus | so i guess we already decided to just do that. i'm okay leaving that the way it is; sorry i misremembered | 16:37 |
| mhu | ahah, there you go :) | 16:37 |
| *** bhavikdbavishi has quit IRC | 16:37 | |
| corvus | mhu: so i think we can just drop anything related to connections from your patch | 16:37 |
| mhu | api/connection or api/connections :) | 16:38 |
| corvus | both :) | 16:38 |
| corvus | mhu: that leaves the autohold and authorizations.... let's examine authorizations first | 16:39 |
| corvus | mhu: it looks like right now, the authz path is /api/user/authorizations | 16:39 |
| corvus | so that is also global-only | 16:40 |
| mhu | it can be tenant scoped just fine, the endpoint will just tell you whether or not you're an "admin" on the tenant | 16:40 |
| mhu | I've made that change in the latest patchset | 16:40 |
| corvus | mhu: ok, i like that idea -- it makes sense to me that authz should always happen with a tenant scope | 16:41 |
| mhu | after all, a client doesn't really need to know if you're an admin on other tenants | 16:41 |
| corvus | mhu: your latest patchset has api/tenant/{tenant}/authorizations as well as api/authorizations | 16:42 |
| corvus | mhu: do we still need the second one? | 16:42 |
| mhu | corvus, no, to my knowledge nothing uses it except my braindead UI patch for authz/authn | 16:42 |
| mhu | which needs to be reworked anyway | 16:42 |
| *** ysandeep|afk is now known as ysandeep | 16:43 | |
| corvus | mhu: ok, so we can drop that -- then i think having autohold be tenant-scoped makes sense too (yes, it's strictly not necessary, but it makes everything easier) | 16:43 |
| mhu | corvus, fair enough | 16:43 |
| tobiash | ++ | 16:43 |
| corvus | mhu: okay, cool, i think we have a way forward here, i'll jot some quick notes on the review. thanks for your patience :) | 16:44 |
| mhu | alright, thanks for clearing things up! | 16:44 |
| mhu | hey no worries, I needed the feedback! | 16:44 |
| mhu | the whitelabel thing puzzled me at first but now it's clearer | 16:45 |
| *** bhavikdbavishi has joined #zuul | 16:46 | |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: [WIP] web UI: add Autoholds Requests page https://review.opendev.org/729307 | 16:50 |
| *** jpena is now known as jpena|off | 17:04 | |
| *** jamesmcarthur has quit IRC | 17:04 | |
| *** jamesmcarthur has joined #zuul | 17:09 | |
| *** asaleh_ has quit IRC | 17:23 | |
| *** jamesmcarthur has quit IRC | 17:24 | |
| *** newbie2020 has quit IRC | 17:27 | |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul-jobs master: WIP: ensure-docker: workaround for centos-8 conflicts https://review.opendev.org/703053 | 17:28 |
| *** evrardjp has quit IRC | 17:32 | |
| *** evrardjp has joined #zuul | 17:34 | |
| *** jamesmcarthur has joined #zuul | 17:50 | |
| *** bhavikdbavishi has quit IRC | 18:06 | |
| *** bhavikdbavishi has joined #zuul | 18:07 | |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul-jobs master: ensure-docker: workaround for centos-8 conflicts https://review.opendev.org/703053 | 18:07 |
| *** jamesmcarthur has quit IRC | 18:12 | |
| *** ysandeep is now known as ysandeep|away | 18:13 | |
| *** jamesmcarthur has joined #zuul | 18:14 | |
| *** jamesmcarthur has quit IRC | 18:17 | |
| *** jamesmcarthur has joined #zuul | 18:17 | |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul-jobs master: ensure-docker: workaround for centos-8 conflicts https://review.opendev.org/703053 | 18:19 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Deprecate default tox_envlist: venv https://review.opendev.org/726830 | 18:23 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Remove unused tox_envlist in fetch-subunit-output https://review.opendev.org/729348 | 18:24 |
| *** jamesmcarthur has quit IRC | 18:29 | |
| *** jamesmcarthur has joined #zuul | 18:30 | |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Deprecate default tox_envlist: venv https://review.opendev.org/726830 | 18:35 |
| *** bhavikdbavishi has quit IRC | 18:37 | |
| *** chandankumar is now known as raukadah | 18:48 | |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: REST API: improve tenant scoping of autohold, authorizations https://review.opendev.org/728118 | 18:53 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: CLI: add autohold-info, autohold-delete via REST https://review.opendev.org/728410 | 18:54 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: REST API: add promote endpoint https://review.opendev.org/728489 | 18:54 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: CLI: support promote via the REST API https://review.opendev.org/728907 | 18:54 |
| *** hashar has quit IRC | 18:59 | |
| zbr | need reviews on https://review.opendev.org/#/c/703053/ - to fix ensure-docker, thanks. | 19:02 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: Zuul CLI: add console-stream command https://review.opendev.org/729236 | 19:08 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: add simple test runner https://review.opendev.org/728684 | 19:09 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: add simple test runner https://review.opendev.org/728684 | 19:11 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: Zuul CLI: add console-stream command https://review.opendev.org/729236 | 19:16 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: [WIP] Web UI: add "held" information, filtering in Builds page https://review.opendev.org/729265 | 19:18 |
| avass | corvus, mordred: can you take a look at: https://review.opendev.org/#/c/703053/16 ? | 19:26 |
| mordred | avass: lgtm | 19:29 |
| mordred | zbr: nice work - I think that's actually much cleaner than before too | 19:29 |
| *** guillaumec has joined #zuul | 19:39 | |
| openstackgerrit | Merged zuul/zuul-jobs master: ensure-docker: workaround for centos-8 conflicts https://review.opendev.org/703053 | 19:55 |
| mordred | corvus: sshnaidm tells me that multi-arch builder support will be in the next release of podman | 20:06 |
| mordred | corvus: so depending on where we end up with figuring out what's going on with buildx and incorrect layers - that might also be a solution | 20:07 |
| sshnaidm | I think it'll be in v2, but don't beat me if it won't.. | 20:07 |
| mordred | sshnaidm: is there any idea how far away that is? | 20:07 |
| mordred | like - is that soon? or months from now? | 20:07 |
| sshnaidm | mordred, lemme find.. | 20:07 |
| mordred | (because if I can't get buildx to behave completely properly, next step is to just use buildkitd directly) | 20:08 |
| mordred | sshnaidm: while you're asking ... does podman have (or will it get) ability to run a container with an arch override? | 20:08 |
| sshnaidm | Apr 02 17:11:50 <baude> sshnaidm, the apis are done and the front-end is being worked on | 20:08 |
| mordred | like "podman run --arch=arm64 foo" | 20:08 |
| mordred | sshnaidm: cool | 20:08 |
| sshnaidm | ttps://podman.readthedocs.io/en/latest/_static/api.html#tag/manifests | 20:08 |
| sshnaidm | well, this link worked then | 20:09 |
| sshnaidm | fast moving project (c) | 20:09 |
| mordred | heh | 20:09 |
| sshnaidm | mordred, not sure about overriding, I was asking about multi-arch manifests | 20:11 |
| sshnaidm | to push ppc containers | 20:11 |
| sshnaidm | or whatever | 20:11 |
| mordred | yeah | 20:11 |
| mordred | sshnaidm: I have 2 use cases I want to be able to do (they aren't 100% related) - one is, on an x86 host, use binfmt_misc support in the kernel to build a non-x86 container image (ppc or arm or whatever) - if I can do that, then making a manifest that references it is straightforward | 20:12 |
| mordred | sshnaidm: the other is, on an x86 host, I would like to use the same support to run a non-x86 container | 20:13 |
| sshnaidm | mordred, to run non-x86 arch on x-86? | 20:13 |
| mordred | sshnaidm: if I build a non-x86 container image and tag it (without using a manifest) - I can run it and things Just Work | 20:13 |
| mordred | yup | 20:13 |
| mordred | but if I put it into a multi-arch manifest, then host arch is what is used to select which sub-image is used - which prevents doing that | 20:14 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: WIP: Add tutorial tests https://review.opendev.org/728194 | 20:14 |
| mordred | sshnaidm: (I told you can I exercise all of the edge cases here :) ) | 20:14 |
| sshnaidm | mordred, I see :) although I'm hardly touched all that before | 20:15 |
| mordred | sshnaidm: me neither until a few weeks ago | 20:15 |
| mordred | sshnaidm: I have had to learn many things | 20:15 |
| sshnaidm | but how do you run different arch w/o emulator? not sure I understand | 20:17 |
| mordred | sshnaidm: kernel has emulation built in | 20:18 |
| * sshnaidm tried to run ppc on qemu, sucks | 20:18 | |
| mordred | sshnaidm: # docker run --rm arm64v8/alpine uname -a | 20:18 |
| mordred | Linux c265cacf8200 4.15.0-96-generic #97-Ubuntu SMP Wed Apr 1 03:25:46 UTC 2020 aarch64 Linux | 20:18 |
| mordred | sshnaidm: that's on an x86 machine | 20:18 |
| mordred | sshnaidm: no qemu involved - this isn't running an emulated VM - this is having the kernel execute foreign arch - more like cross-compiling | 20:20 |
| sshnaidm | hmm.. standard_init_linux.go:211: exec user process caused "exec format error" | 20:20 |
| mordred | sshnaidm: you;ll want to do this first: | 20:20 |
| mordred | docker run --rm --privileged docker/binfmt:a7996909642ee92942dcd6cff44b9b95f08dad64 | 20:20 |
| mordred | which will write some entries into /proc/sys/fs/binfmt_misc/ for you | 20:21 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: WIP: Add tutorial tests https://review.opendev.org/728194 | 20:21 |
| sshnaidm | a-ha! | 20:21 |
| sshnaidm | podman run --rm arm64v8/alpine uname -a | 20:21 |
| sshnaidm | worked well | 20:21 |
| mordred | sshnaidm: so - in this case, the arm64v8/alpine image isn't a multi-arch image - it's just a plain arm64 image - so this works | 20:21 |
| mordred | but what I'd like to be able to do is build a multi-arch manifest image - and then do the same thing without having to resort to tag tricks, instead beinga ble to say --arch=foo | 20:22 |
| mordred | if that won't work - I can totally just make an arm64 tag of course | 20:22 |
| sshnaidm | mordred, I see, so you need to run multi-arch with a "wrong" arch | 20:23 |
| mordred | sshnaidm: yes! | 20:23 |
| mordred | since the _hard_ parts of this all work | 20:23 |
| mordred | it's just the ui doens't exist to let me tell the front end to pick the "wrong" backend architecture | 20:24 |
| sshnaidm | I think worth to open an rfe | 20:24 |
| mordred | sshnaidm: nod. I'll see if I can do that today | 20:24 |
| sshnaidm | podman folks are pretty responsive | 20:25 |
| mordred | same thing would be nice for podman build | 20:25 |
| mordred | if you could do that - then you could do podman build --arch=arm64 then podman manifest create and you wouldn't need any of the fancy crap docker buildx does for you | 20:25 |
| sshnaidm | yep | 20:27 |
| sshnaidm | buildx seems like something new | 20:28 |
| mordred | it is | 20:32 |
| mordred | it's non-terrible except when it's terrible :) | 20:32 |
| mordred | but it's also been helpful for learning about some of the possibilities | 20:33 |
| *** Goneri has quit IRC | 21:03 | |
| openstackgerrit | Merged zuul/zuul-jobs master: Update flake8 ignore rules to match Zuul https://review.opendev.org/729010 | 21:09 |
| openstackgerrit | Merged zuul/zuul-jobs master: Allow mapping additional hostvars in write-inventory https://review.opendev.org/728952 | 21:15 |
| *** Goneri has joined #zuul | 21:20 | |
| *** jamesmcarthur has quit IRC | 21:20 | |
| *** jamesmcarthur has joined #zuul | 21:24 | |
| *** jamesmcarthur has quit IRC | 21:31 | |
| *** jamesmcarthur has joined #zuul | 21:32 | |
| *** jamesmcarthur has quit IRC | 21:34 | |
| *** jamesmcarthur has joined #zuul | 21:34 | |
| *** jamesmcarthur has quit IRC | 21:37 | |
| *** jamesmcarthur has joined #zuul | 21:39 | |
| *** jamesmcarthur has quit IRC | 21:41 | |
| *** jamesmcarthur has joined #zuul | 21:41 | |
| *** Goneri has quit IRC | 21:48 | |
| openstackgerrit | Clark Boylan proposed zuul/zuul-jobs master: Add option to prefer https/ssl in configure-mirrors https://review.opendev.org/729407 | 21:58 |
| clarkb | I feel like ^ is probably wrong mechanically | 21:58 |
| clarkb | I'm hoping testing will confirm that for me and allow me to fix it up | 21:59 |
| clarkb | corvus: mordred ^ also there is probably a better way to do it with the planned reorg of mirror things anyway? | 21:59 |
| clarkb | AJaeger: avass: corvus: https://zuul.opendev.org/t/openstack/build/64b57ff90ea34dc9b45d11f582dfe4cf I think that is a regression in the tox env change | 22:08 |
| clarkb | I don't quite understand it yet though | 22:08 |
| clarkb | python2 + tox 3.8.6 does not have this issue. tox 3.15.0 + python3.8 does | 22:11 |
| clarkb | ok its the unescaped % in the command | 22:14 |
| clarkb | doubling up the % to escape it fixes things | 22:14 |
| *** jamesmcarthur has quit IRC | 22:33 | |
| *** jamesmcarthur has joined #zuul | 22:33 | |
| *** jamesmcarthur_ has joined #zuul | 22:40 | |
| *** jamesmcarthur has quit IRC | 22:44 | |
| *** jamesmcarthur_ has quit IRC | 23:05 | |
| *** tumble has quit IRC | 23:09 | |
| *** armstrongs has joined #zuul | 23:10 | |
| *** jamesmcarthur has joined #zuul | 23:12 | |
| *** armstrongs has quit IRC | 23:19 | |
| openstackgerrit | Monty Taylor proposed zuul/zuul-preview master: Simplify build https://review.opendev.org/729066 | 23:46 |
| openstackgerrit | Monty Taylor proposed zuul/zuul-preview master: Bump to c++17 and switch boost::optional to std::optional https://review.opendev.org/729428 | 23:46 |
| *** jamesmcarthur has quit IRC | 23:56 | |
| *** jamesmcarthur has joined #zuul | 23:57 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!