| *** sanjayu_ has quit IRC | 00:13 | |
| *** armstrongs has joined #zuul | 00:15 | |
| *** armstrongs has quit IRC | 00:23 | |
| *** Goneri has quit IRC | 01:26 | |
| *** swest has quit IRC | 01:36 | |
| *** rlandy has quit IRC | 01:48 | |
| *** swest has joined #zuul | 01:51 | |
| *** ysandeep|away is now known as ysandeep|rover | 01:55 | |
| *** dangtrinhnt has joined #zuul | 02:03 | |
| *** dangtrinhnt has quit IRC | 03:03 | |
| *** dangtrinhnt has joined #zuul | 03:24 | |
| *** marvs has quit IRC | 03:33 | |
| *** marvs has joined #zuul | 03:33 | |
| *** bhavikdbavishi has joined #zuul | 03:35 | |
| *** bhavikdbavishi1 has joined #zuul | 03:38 | |
| *** threestrands has joined #zuul | 03:39 | |
| *** bhavikdbavishi has quit IRC | 03:39 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 03:39 | |
| *** toabctl has joined #zuul | 03:39 | |
| *** dangtrinhnt has quit IRC | 03:48 | |
| *** dangtrinhnt has joined #zuul | 03:48 | |
| *** toabctl has quit IRC | 04:03 | |
| *** toabctl has joined #zuul | 04:07 | |
| *** dangtrinhnt_ has joined #zuul | 04:21 | |
| *** dangtrinhnt has quit IRC | 04:25 | |
| *** bhavikdbavishi has quit IRC | 04:33 | |
| *** bhavikdbavishi has joined #zuul | 04:35 | |
| *** evrardjp has quit IRC | 04:36 | |
| *** evrardjp has joined #zuul | 04:36 | |
| *** bhavikdbavishi has quit IRC | 04:39 | |
| *** smyers has quit IRC | 04:54 | |
| *** smyers has joined #zuul | 05:01 | |
| *** bhavikdbavishi has joined #zuul | 05:50 | |
| *** sgw has quit IRC | 05:57 | |
| *** dpawlik has joined #zuul | 06:20 | |
| *** dpawlik has quit IRC | 06:28 | |
| *** dpawlik has joined #zuul | 06:47 | |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Adds roles to install and run hashicorp packer https://review.opendev.org/709292 | 06:47 |
|---|---|---|
| *** dangtrinhnt_ has quit IRC | 06:50 | |
| *** dangtrinhnt has joined #zuul | 06:50 | |
| *** bhavikdbavishi has quit IRC | 07:08 | |
| *** sanjayu_ has joined #zuul | 07:19 | |
| *** tosky has joined #zuul | 07:29 | |
| *** bhavikdbavishi has joined #zuul | 07:31 | |
| *** zxiiro has quit IRC | 07:38 | |
| *** jcapitao has joined #zuul | 07:42 | |
| *** ysandeep|rover is now known as ysandeep|rover|l | 07:47 | |
| *** jpena|off is now known as jpena | 07:53 | |
| *** threestrands has quit IRC | 08:26 | |
| *** ysandeep|rover|l is now known as ysandeep|rover | 08:34 | |
| *** bhavikdbavishi has quit IRC | 08:47 | |
| *** avass has quit IRC | 08:55 | |
| *** avass has joined #zuul | 08:55 | |
| avass | is it possible to add config to ansible.cfg somehow? | 08:56 |
| avass | We're planning to use the splunk callback for monitoring: https://docs.ansible.com/ansible/latest/plugins/callback/splunk.html and we need to be able to whitelist that callback somehow | 08:57 |
| *** jpena is now known as jpena|away | 09:00 | |
| *** hashar has joined #zuul | 09:29 | |
| *** sshnaidm|afk is now known as sshnaidm | 09:30 | |
| *** bhavikdbavishi has joined #zuul | 09:37 | |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Connect merger to Zookeeper https://review.opendev.org/716221 | 09:48 |
| *** dangtrinhnt_ has joined #zuul | 09:59 | |
| *** jpena|away is now known as jpena | 10:00 | |
| *** dangtrinhnt has quit IRC | 10:03 | |
| tobiash | avass: you might be able to add this callback to each respective callback dir in the executor after installing it | 10:07 |
| *** ysandeep|rover is now known as ysandeep|afk | 10:09 | |
| avass | tobiash: I was thinking of something like that, but I thin ansible.cfg needs to whitelist the callback as well | 10:10 |
| tobiash | oh right | 10:11 |
| tobiash | then I guess this needs to be implemented | 10:11 |
| avass | I'll take a look at that later then :) | 10:15 |
| *** dangtrinhnt_ has quit IRC | 10:16 | |
| *** dangtrinhnt has joined #zuul | 10:17 | |
| *** dangtrinhnt has quit IRC | 10:22 | |
| *** dangtrinhnt has joined #zuul | 10:29 | |
| *** bhavikdbavishi has quit IRC | 10:33 | |
| *** weshay|ruck has quit IRC | 10:35 | |
| *** weshay_ has joined #zuul | 10:35 | |
| *** bhavikdbavishi has joined #zuul | 10:36 | |
| *** nhicher has quit IRC | 10:36 | |
| *** jpena has quit IRC | 10:37 | |
| *** ysandeep|afk is now known as ysandeep|rover | 10:39 | |
| *** jpena has joined #zuul | 10:43 | |
| *** nhicher has joined #zuul | 10:51 | |
| *** jcapitao is now known as jcapitao_lunch | 10:53 | |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Adds roles to install and run hashicorp packer https://review.opendev.org/709292 | 10:57 |
| zbr | can someone give me a hint about where is the code that builds the "preview" part on Results (the stdout tail). I want to add ANSI support to it. | 11:17 |
| zbr | clearly is done in javascript but i am still digging for it.... | 11:18 |
| zbr | found it, the magic renderFailedTask | 11:20 |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul master: WIP: Enable ANSI rendering on stdout/stderr https://review.opendev.org/716251 | 11:25 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Connect merger to Zookeeper https://review.opendev.org/716221 | 11:31 |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul master: WIP: Enable ANSI rendering on stdout/stderr https://review.opendev.org/716251 | 11:40 |
| *** jpena is now known as jpena|lunch | 11:40 | |
| *** hashar has quit IRC | 11:43 | |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Connect executor to Zookeeper https://review.opendev.org/716262 | 11:45 |
| openstackgerrit | Tobias Henkel proposed zuul/zuul-jobs master: Generalize parse tox output https://review.opendev.org/716263 | 11:45 |
| openstackgerrit | Tobias Henkel proposed zuul/zuul-jobs master: Strip source dir from file comments https://review.opendev.org/716264 | 11:45 |
| bolg | zuul-maint: I quess the https://review.opendev.org/716221 and https://review.opendev.org/716262 may break some installations in case the executor or merger config does not contain [zookeeper] section, correct? | 11:50 |
| tobiash | bolg: yes, so they should go into the 4.0.0 release | 11:50 |
| *** nhicher has quit IRC | 11:51 | |
| bolg | tobiash: should it be marked somehow? E.g. topic? | 11:52 |
| *** nhicher has joined #zuul | 11:52 | |
| tobiash | we typically vote with an administrative -2 in those cases | 11:52 |
| bolg | tobiash: feel free to do so :) | 11:52 |
| tobiash | done | 11:53 |
| bolg | thx | 11:54 |
| tobiash | bolg: it would be great if you could add a release note to one of them in the section upgrade notes | 11:55 |
| bolg | tobiash: sure | 11:57 |
| *** weshay_ has quit IRC | 11:59 | |
| *** jpena|lunch has quit IRC | 11:59 | |
| *** nhicher has quit IRC | 11:59 | |
| zbr | can anyone give me some hints on yarn/react side? | 12:00 |
| zbr | mainly "yarn build" fails on https://review.opendev.org/#/c/716251/1 with "Failed to minify the code..." | 12:02 |
| *** harrymichal has joined #zuul | 12:04 | |
| mnaser | tobiash: nice changes wrt inline comments | 12:06 |
| tobiash | mnaser: just tested with a sphinx build locally :) | 12:07 |
| *** jpena has joined #zuul | 12:07 | |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Adds roles to install and run hashicorp packer https://review.opendev.org/709292 | 12:08 |
| *** nhicher has joined #zuul | 12:08 | |
| *** weshay_ has joined #zuul | 12:10 | |
| *** weshay_ is now known as weshay|ruck | 12:12 | |
| *** hashar has joined #zuul | 12:16 | |
| *** rlandy has joined #zuul | 12:16 | |
| *** fbo has joined #zuul | 12:18 | |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Connect executor to Zookeeper https://review.opendev.org/716262 | 12:19 |
| *** jcapitao_lunch is now known as jcapitao | 12:21 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-operator master: Add integration test playbook https://review.opendev.org/714165 | 12:25 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-operator master: Add nodepool launcher service initial deployment https://review.opendev.org/715310 | 12:25 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-operator master: Add nodepool external config https://review.opendev.org/715311 | 12:25 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-operator master: Adapt the integration playbook to be usable locally https://review.opendev.org/714163 | 12:25 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-operator master: Add nodepool kubernetes pod label to integration test https://review.opendev.org/715316 | 12:25 |
| *** rfolco has joined #zuul | 12:27 | |
| *** hashar has quit IRC | 12:28 | |
| *** hashar has joined #zuul | 12:29 | |
| fbo | hi @zuul-maint could we approve https://review.opendev.org/696134/ (Pagure related), a dependent change is already W+1 but stuck due to this one. | 12:29 |
| tobiash | fbo: lgtm | 12:37 |
| *** fbo has quit IRC | 12:38 | |
| *** nhicher has quit IRC | 12:39 | |
| *** nhicher has joined #zuul | 12:40 | |
| *** fbo has joined #zuul | 12:47 | |
| fbo | tobiash: thanks | 12:50 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-operator master: Add nodepool kubernetes pod label to integration test https://review.opendev.org/715316 | 12:52 |
| mnaser | tobiash: i think we are going to probably write tests at some point for that role | 13:00 |
| tobiash | mnaser: yes we should do that :) | 13:01 |
| mnaser | tobiash: i'm thinking we can just have a string of different outputs that we know it breaks/works with, that way we avoid the whole wip changes everywhere | 13:01 |
| mnaser | tobiash: do you have time to write something out at the moment? | 13:03 |
| tobiash | mnaser: I'm deep in ops topics today | 13:04 |
| *** sgw has joined #zuul | 13:09 | |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Adds roles to install and run hashicorp packer https://review.opendev.org/709292 | 13:11 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Adds roles to install and run hashicorp packer https://review.opendev.org/709292 | 13:14 |
| tristanC | here is a good progress on zuul-operator integration tests, here are the nodepool-launcher service logs spawning pod in the minikube: https://f5c97d48dcd7c256d91c-c98b0e913e593910e72c88f6f60b9c46.ssl.cf1.rackcdn.com/715316/7/check/zuul-operator-functional-k8s/ff83f89/docker/k8s_launcher_zuul-launcher-66d8cf545f-dwvhm_default_bd078005-df77-4445-84c6-dd7987144286_0.txt | 13:20 |
| tristanC | and the zuul-executor log shows that zuul-base-jobs doesn't work with kubectl because `add-build-sshkey` fails with `Failed to lookup user zuul: "getpwnam(): name not found: 'zuul'` | 13:21 |
| *** weshay has joined #zuul | 13:23 | |
| *** fbo has quit IRC | 13:24 | |
| *** fbo has joined #zuul | 13:24 | |
| *** nhicher has quit IRC | 13:24 | |
| *** jpena has quit IRC | 13:25 | |
| *** weshay|ruck has quit IRC | 13:25 | |
| *** Goneri has joined #zuul | 13:26 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-operator master: Add nodepool kubernetes pod label to integration test https://review.opendev.org/715316 | 13:26 |
| *** nhicher has joined #zuul | 13:26 | |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Adds roles to install and run hashicorp packer https://review.opendev.org/709292 | 13:27 |
| *** jpena has joined #zuul | 13:31 | |
| mordred | tristanC: hrm. do we need add-build-sshkey when doing kubectl? | 13:36 |
| tristanC | mordred: no, actually zuul-base-jobs fails with kubectl connection because of add-build-sshkey, prepare-workspace, fetch-output and remove-build-sshkey | 13:38 |
| *** bhavikdbavishi has quit IRC | 13:39 | |
| tristanC | zuul-maint: the operator could use some review please, the integration test is currently running on a stack of 8 open changes: https://review.opendev.org/715316 | 13:41 |
| openstackgerrit | Merged zuul/zuul master: Pagure: remove connectors burden and simplify code https://review.opendev.org/696134 | 13:42 |
| openstackgerrit | Merged zuul/zuul master: Pagure - Refresh token and recheck sign when token regenerated https://review.opendev.org/698208 | 13:42 |
| tobiash | tristanC: I'll have a look later today | 13:43 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Control log archive and user preservation with vars https://review.opendev.org/701381 | 13:45 |
| tristanC | tobiash: thanks! i think this test is quite valuable as it may be the first one that test using zuul jobs with a nodepool kubernetes providers | 13:45 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-base-jobs master: base: skip role incompatible with kubectl connection https://review.opendev.org/716298 | 13:46 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-operator master: Use speculative zuul-base-jobs in integration test https://review.opendev.org/716300 | 13:51 |
| tristanC | mordred: hopefully ^ should enable testing change to the zuul-base-jobs | 13:53 |
| *** Diabelko has quit IRC | 13:56 | |
| openstackgerrit | Monty Taylor proposed zuul/zuul master: WIP: Enable ANSI rendering on stdout/stderr https://review.opendev.org/716251 | 14:02 |
| openstackgerrit | Monty Taylor proposed zuul/zuul master: Update to create-react-app 3.4.1 https://review.opendev.org/716305 | 14:02 |
| mordred | tristanC: neat! | 14:02 |
| *** y2kenny has joined #zuul | 14:02 | |
| mordred | zbr: ^^ you nerd-sniped me - I couldn't figure out why your patch wouldn't build - so I poked around and learned our create-react-app was a bit old now, so I updated it and it fixed the build issue | 14:03 |
| zbr | mordred: super! thanks. | 14:03 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Adds roles to install and run hashicorp packer https://review.opendev.org/709292 | 14:03 |
| zbr | sadly my web-dev (especially around js stuff are very poort) | 14:04 |
| mordred | tristanC, corvus, tobiash : it doesn't look like anything substantive changed for the negative in that update to create-react-app - but obviously a change like that should get extra attention before merging | 14:04 |
| mordred | mnaser: ^^ | 14:04 |
| zbr | i was suprised to discover that npm was reporting no less than 88 security warnings | 14:05 |
| tobiash | mordred: .keep has hit you :D | 14:06 |
| tobiash | mordred: maybe it's time to advocate again for https://review.opendev.org/663108 ? | 14:06 |
| tobiash | ;) | 14:06 |
| zbr | we could put a try/catch on it, getting rid of .keep would be awesome | 14:08 |
| tobiash | oh I need to revise it slightly | 14:09 |
| mordred | tobiash: blerg keep | 14:21 |
| *** dangtrinhnt has quit IRC | 14:22 | |
| openstackgerrit | Monty Taylor proposed zuul/zuul master: Update to create-react-app 3.4.1 https://review.opendev.org/716305 | 14:22 |
| *** dangtrinhnt has joined #zuul | 14:22 | |
| mordred | tobiash: or instead of try could just check for os.path.islink | 14:23 |
| *** dangtrinhnt has quit IRC | 14:28 | |
| corvus | avass, tobiash: i'm in favor of a way to add callback modules; we wanted to do that for ara, but didn't have time. but we should generalize that so that ara and splunk and anything else can be added by the site admin. | 14:31 |
| tobiash | corvus: ++ | 14:35 |
| *** ysandeep|rover is now known as ysandeep|away | 14:39 | |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Revert "Revert "Create zuul/web/static on demand"" https://review.opendev.org/663108 | 14:41 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-operator master: Add nodepool kubernetes pod label to integration test https://review.opendev.org/715316 | 14:42 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-operator master: Use speculative zuul-base-jobs in integration test https://review.opendev.org/716300 | 14:42 |
| corvus | mordred: do we still need the ansible lint skip in https://review.opendev.org/715727 ? | 14:47 |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: Remove lint skip in tox iniline comment parser https://review.opendev.org/716321 | 14:50 |
| corvus | that will tell | 14:50 |
| corvus | tristanC: is this failure something to worry about? https://softwarefactory-project.io/logs/27/715727/4/third-party-check/test-job-tox-el7/f430b9d/job-output.txt.gz | 14:51 |
| tristanC | corvus: iirc i fixed it yesterday, it was because of tox not working with old virtualenv | 14:53 |
| mordred | corvus: I think we don't | 14:55 |
| *** dangtrinhnt has joined #zuul | 14:55 | |
| corvus | tristanC: thanks, sounds like we can probably merge that and ignore the el7 error | 14:56 |
| mnaser | mordred, corvus: i actaully think we're probably good without it | 15:01 |
| mnaser | since we have the noop-y zuul_return, but we'll find out indeed :) | 15:02 |
| *** dangtrinhnt has quit IRC | 15:08 | |
| *** dangtrinhnt has joined #zuul | 15:09 | |
| *** dangtrinhnt has quit IRC | 15:09 | |
| *** dangtrinhnt has joined #zuul | 15:09 | |
| *** jcapitao is now known as jcapitao_afk | 15:18 | |
| tristanC | mordred: turns out even validate-host doesn't work with a kubectl connection to the docker.io/fedora:31 image : https://e92b17c725197d560c8a-a15027182aab035aa882f99410b51a23.ssl.cf5.rackcdn.com/716300/2/check/zuul-operator-functional-k8s/f190d96/docker/k8s_executor_zuul-executor-0_default_d02ebab2-e446-406e-882b-afc0c2dcf086_0.txt | 15:22 |
| *** avass is now known as Guest32455 | 15:23 | |
| *** avass has joined #zuul | 15:23 | |
| avass | corvus: I agree, I also saw that there already was a comment from mordred about that :) | 15:24 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-base-jobs master: base: skip role incompatible with kubectl connection https://review.opendev.org/716298 | 15:24 |
| *** bhavikdbavishi has joined #zuul | 15:31 | |
| *** bhavikdbavishi1 has joined #zuul | 15:34 | |
| *** bhavikdbavishi has quit IRC | 15:35 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 15:35 | |
| *** jcapitao_afk is now known as jcapitao | 15:45 | |
| openstackgerrit | Monty Taylor proposed zuul/zuul master: Update to create-react-app 3.4.1 https://review.opendev.org/716305 | 15:46 |
| tobiash | corvus: just curious regarding cleanup runs, did you move log upload to the cleanup phase or live with not having logs from cleanup runs? | 15:54 |
| corvus | tobiash: i think log is still in post, and we don't worry about cleanup logs. | 15:56 |
| tobiash | k, like we have it atm as well | 15:56 |
| *** dangtrinhnt has quit IRC | 15:56 | |
| *** y2kenny has left #zuul | 15:57 | |
| clarkb | correct, but also cleanup for opendev is largely used as last effort debugging tool. So if we have to reach there we've already failed pretty hard elsewhere | 15:57 |
| clarkb | (specifcally we do a raw connection and dump disk usage (bytes and inodes) as well as networking data) | 15:58 |
| zbr | corvus: clarkb : ok to proceed with https://review.opendev.org/#/c/702304/ (install-docker)? | 15:58 |
| *** dangtrinhnt has joined #zuul | 15:58 | |
| *** dangtrinhnt has quit IRC | 16:05 | |
| clarkb | zbr: I've approved that change. I'm curious how the vars/ loading in tasks/main.yaml works with defaults/main.yaml being presumably loaded first? It seems to be working based on the existing code for fidning packages to isntall, but I wouldn't have expected that if I were writing it from scratch | 16:16 |
| clarkb | maybe we are updated the variable values before we actualy use them and that is good enough | 16:20 |
| *** y2kenny has joined #zuul | 16:22 | |
| zbr | clarkb: defaults is implicit, vars/ always override it, is well documented. | 16:22 |
| y2kenny | When I run a playbook, does having hosts: localhost mean it's being run on the executor? | 16:24 |
| clarkb | y2kenny: yes | 16:25 |
| y2kenny | is that the reason why I would get "Executing local code is prohibited"? | 16:25 |
| clarkb | the executor is excluded from the 'all' group but if you explicitly use 'localhost' it will run on the executor | 16:25 |
| clarkb | y2kenny: if the job is untrusted yes (there are limitations to what you can do on the executor outside of trusted jobs) | 16:25 |
| y2kenny | I was just trying to run a find with ansible to list a set of specific files. Should I just use shell or command instead? | 16:27 |
| clarkb | y2kenny: those will be restricted too. In general the rules are "no file access outside of the build directory and no arbitrary code execution" | 16:28 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-operator master: Add nodepool kubernetes pod label to integration test https://review.opendev.org/715316 | 16:28 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-operator master: Use speculative zuul-base-jobs in integration test https://review.opendev.org/716300 | 16:28 |
| clarkb | y2kenny: this means you can run ansible modules within the build directory (in the case of find are you searching outside of that subdir?) | 16:28 |
| y2kenny | clarkb: I just want to find within the work_root | 16:28 |
| y2kenny | actually executor.work_root/project.src_dir | 16:29 |
| clarkb | that I would've expected to work if using the ansible find module | 16:29 |
| fungi | also be aware that lots of the builtin ansible modules are blacklisted on the executor if they can be used to run arbitrary code or overwrite arbitrary files | 16:30 |
| clarkb | ya its possible find isn't as safe as I think | 16:30 |
| fungi | find -exec is a thing ;) | 16:30 |
| *** bhavikdbavishi has quit IRC | 16:30 | |
| clarkb | also I thought we had documentation on these restrictions but I'm not finding any | 16:31 |
| y2kenny | fungi: so in this case what would be the recommended workaround? would I need to run it under hosts: all after copying the src over? | 16:32 |
| y2kenny | fungi: potentially I can just shell out | 16:33 |
| fungi | "find" is a very generic tool with lots of functions, what are you trying to do with it? | 16:35 |
| y2kenny | literally just trying to find all the Dockerfile in the repository and get the paths to them | 16:35 |
| fungi | when you say "list a set of specific files" you mean test whether they exist? | 16:36 |
| fungi | ahh, so generate a list of anything named "Dockerfile" within the workdir | 16:36 |
| *** evrardjp has quit IRC | 16:36 | |
| y2kenny | this is a repository of a few different docker context. I want to re-generate the docker image if there are changes to the context | 16:36 |
| *** evrardjp has joined #zuul | 16:36 | |
| y2kenny | fungi: yup, that's it | 16:37 |
| mordred | y2kenny: fwiw - there's a file matcher in zuul job definitions that you can set so that zuul will only trigger jobs if they match | 16:37 |
| mordred | then with docker images, there's a nice system for re-using images built by one job in child jobs if they were built, but otherwise pulling them from published sources | 16:38 |
| * mordred gets example links | 16:38 | |
| openstackgerrit | Merged zuul/zuul-jobs master: install-docker: allow removal of conflicting packages https://review.opendev.org/702304 | 16:38 |
| corvus | y2kenny: since you won't be able to generate the image on the executor anyway, yeah, you might as well do the find on the worker node. however, what mordred is saying might be a good approach -- opendev uses it in a repo with a bunch of dockerfiles | 16:38 |
| mordred | https://opendev.org/opendev/system-config/src/branch/master/.zuul.yaml#L211-L232 | 16:38 |
| mordred | there's our gerrit-2.13 build job - which only runs if we touch things in the base gerrit image or the context for 2.13 | 16:39 |
| mordred | then we have: https://opendev.org/opendev/system-config/src/branch/master/.zuul.yaml#L1658-L1662 | 16:39 |
| y2kenny | so I've got the job to launch a kaniko pod to generate and publish the image. | 16:39 |
| mordred | which runs system-config-run-review with a soft-depend on the build job | 16:40 |
| y2kenny | I understand the image re-use thing but this is sort of the base image infrastructure | 16:40 |
| corvus | y2kenny: have you seen this doc? https://zuul-ci.org/docs/zuul-jobs/docker-image.html | 16:42 |
| y2kenny | corvus: I have not but that is very useful. I was thinking about the intermediate registry thing. Currently I am just using an internal Artifactory server. | 16:44 |
| corvus | y2kenny: this really shines with images that depend on other images (like mordred's gerrit-2.13 -> gerrit example) | 16:44 |
| y2kenny | um... I am not sure if this helps though (because I am trying to bring up the infra to support the building of images that everything else will depends on.) | 16:45 |
| corvus | yeah, it'll depend on the specifics :) | 16:46 |
| corvus | but if you think it might be useful, then it might be worth keeping in mind so you can steer your design in that direction | 16:46 |
| y2kenny | the building and publishing image I've got working. What I am trying to do organize the docker image contexts that devs might come up with | 16:48 |
| y2kenny | I am using a separate repo to store the docker context because a lot of the upstream components/projects are not container native | 16:49 |
| corvus | y2kenny: the jobs mordred pointed to use this directory: https://opendev.org/opendev/system-config/src/branch/master/docker | 16:49 |
| y2kenny | so it's not like each project's repo has a Dockerfile | 16:49 |
| corvus | y2kenny: so there's a job per docker context directory, and it only runs on changes to that dir | 16:49 |
| tristanC | corvus: tobiash: thanks a lot for the zuul-operator review! I added a comment about the current zuul restart implementation in https://review.opendev.org/715424 | 16:50 |
| y2kenny | corvus: ok... I think I get what you are getting at with the job.files thing now. | 16:50 |
| y2kenny | corvus:... yea... I am trying to build a generic job that build all images but you are suggesting separate job for separate image | 16:51 |
| corvus | y2kenny: yep; i think that'll be clearer (if a little more verbose) and also set you up well for using artifacts or speculative container images in the future | 16:52 |
| y2kenny | corvus: which make sense because jobs are so easy to add with Zuul. I am still kind of thinking with the old mindset where adding jobs are difficult. So in this case, devs will just add an image job as they add the context to the repo | 16:53 |
| corvus | y2kenny: ++ jobs are free :) | 16:53 |
| openstackgerrit | Merged zuul/zuul-operator master: Add tenant reconfiguration when main.yaml changed https://review.opendev.org/703631 | 16:54 |
| fungi | sometimes a job is just a couple lines to inherit from another job and add a role or set a variable | 16:55 |
| y2kenny | so back to the secret thing, do you guys avoid having the need to use secret for speculative image building because there's an intermediate registry? | 16:58 |
| y2kenny | (so you only really need secret on publish which is post-review task?) | 16:58 |
| *** dustinc has joined #zuul | 17:00 | |
| *** y2kenny has left #zuul | 17:01 | |
| *** y2kenny58 has joined #zuul | 17:02 | |
| *** y2kenny58 has quit IRC | 17:02 | |
| *** jpena is now known as jpena|off | 17:02 | |
| *** y2kenny5 has joined #zuul | 17:03 | |
| *** y2kenny5 has quit IRC | 17:03 | |
| *** y2kenny8 has joined #zuul | 17:04 | |
| *** y2kenny8 has left #zuul | 17:05 | |
| *** y2kenny has joined #zuul | 17:06 | |
| *** zxiiro has joined #zuul | 17:09 | |
| corvus | y2kenny: there's a secret in the config project for the speculative registry. we put that in a base job, and it handles the stuff that needs the secret there. child jobs that inherit from it (to actually do the image building) don't need the secret in their playbooks, so that part of the job is still dynamic. | 17:16 |
| tristanC | mordred: success, it seems like https://review.opendev.org/716300 validated the zuul-base-jobs change to fix kubectl usage! | 17:16 |
| mnaser | speaking of which, is there a way to get a secret into a config-project and allow certain projects to use it? i'm having to re-encrypt secrets in every job right now | 17:44 |
| mordred | mnaser: sort of - | 17:46 |
| mordred | mnaser: put the use of the secret into a base job in the config-project - so the consumer of the secret is with the secret | 17:46 |
| mordred | mnaser: then you can put allowed-projects on the base-job | 17:46 |
| mordred | we do this with the image jobs in opendev so that people can use them without needing access to the secret itself | 17:46 |
| clarkb | you do need to be careful doing that as job material could expose the secret if it is on disk from a pre playbook | 17:47 |
| mnaser | in this case it's just the docker registry credentials | 17:47 |
| mnaser | (for the existing jobs that i use in opendev) | 17:48 |
| mordred | yeah. in that case, I'd totally follow our pattern - this is for publishing images yeah? | 17:48 |
| mnaser | yeah, for upload/promote | 17:48 |
| mordred | mnaser: yeah - you just need a base job | 17:48 |
| mnaser | now in that case... | 17:49 |
| mordred | doesn't even need to be in a config-project since it's already protected via use of secret | 17:49 |
| openstackgerrit | Merged zuul/zuul-operator master: zuul-restart: change service order to prevent tenant loading failure https://review.opendev.org/715424 | 17:52 |
| openstackgerrit | Merged zuul/zuul-operator master: Set default secret mode to 0400 https://review.opendev.org/714501 | 17:52 |
| openstackgerrit | Merged zuul/zuul-operator master: Add integration test playbook https://review.opendev.org/714165 | 17:52 |
| openstackgerrit | Merged zuul/zuul-operator master: Add nodepool launcher service initial deployment https://review.opendev.org/715310 | 17:52 |
| *** hashar is now known as hasharAway | 17:52 | |
| *** jcapitao has quit IRC | 18:08 | |
| mnaser | tristanC: i like the approach at https://review.opendev.org/716298 -- it makes for a lot cleaner base jobs to be recommended to our users | 18:11 |
| mnaser | are we testing those basejobs in any way though? :x | 18:12 |
| mnaser | tristanC: looks like past you has thought of the same solution before :) https://review.opendev.org/680711 | 18:14 |
| tristanC | mnaser: this is tested in https://review.opendev.org/716300 | 18:21 |
| tristanC | mnaser: e.g. here are the executor logs that shows a successfull kubectl port-forward usage using the zuul container image: https://3605f41c7a13495ddbda-73dc500a2fd36a4e9335396c9d68e49b.ssl.cf5.rackcdn.com/716300/3/check/zuul-operator-functional-k8s/22aa6bf/docker/k8s_executor_zuul-executor-0_default_9fdb2ca6-403d-4651-8e29-caabe2d74cde_0.txt | 18:22 |
| tristanC | mnaser: and here is an integration test that verify the console-stream does contains the job output: https://zuul.opendev.org/t/zuul/build/22aa6bf72f204978adb434f4fca8cf22/console#3/0/26/ubuntu-bionic | 18:22 |
| mnaser | ok but the actual repo it self has no tests other than the zuul-operator | 18:23 |
| tristanC | mnaser: i don't know if it is possible to test base job without spawning a new zuul | 18:24 |
| *** sanjayu_ has quit IRC | 18:26 | |
| tristanC | we are actually considering adding such update jobs to our zuul, to test new images or config-project changes, using a similar nested zuul that would validate both the image and the config project | 18:28 |
| fungi | that'll be a nice and thorough testing solution | 18:29 |
| *** cloudnull has quit IRC | 18:57 | |
| openstackgerrit | Merged zuul/zuul-jobs master: Revert "Revert "Extract pep8 messages for inline comments"" https://review.opendev.org/715727 | 19:05 |
| openstackgerrit | Merged zuul/zuul-jobs master: Remove lint skip in tox iniline comment parser https://review.opendev.org/716321 | 19:05 |
| AJaeger | mordred: this zuul-jobs change of yours is 14 months old - still relevant? https://review.opendev.org/#/c/629604/2 | 19:10 |
| mordred | AJaeger: I think so? I need to ponder a little | 19:11 |
| AJaeger | fine | 19:11 |
| *** dustinc has quit IRC | 19:23 | |
| *** y2kenny has left #zuul | 19:24 | |
| openstackgerrit | Merged zuul/nodepool master: Update dib dep to 2.35.0 https://review.opendev.org/716104 | 20:05 |
| openstackgerrit | Monty Taylor proposed zuul/zuul master: Update to create-react-app 3.4.1 https://review.opendev.org/716305 | 20:06 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: local-log-download : role with script to download all log files https://review.opendev.org/715756 | 20:10 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: local-log-download : role with script to download all log files https://review.opendev.org/715756 | 20:25 |
| *** y2kenny has joined #zuul | 20:29 | |
| *** mwhahaha has joined #zuul | 20:29 | |
| y2kenny | for the rsa keypair use to encrypt the secret, is it own by the scheduler or the web component? Is there a way to save they keys so that the encryption survive zuul restart? | 20:40 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: local-log-download : role with script to download all log files https://review.opendev.org/715756 | 20:41 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: local-log-download : role with script to download all log files https://review.opendev.org/715756 | 20:41 |
| clarkb | y2kenny: the scheduler component owns them. And they are all on disk can can be preserved by preserving that path | 20:43 |
| y2kenny | in the scheduler.state_dir? | 20:43 |
| clarkb | ya on our install it is /var/lib/zuul/keys | 20:44 |
| y2kenny | great thanks. | 20:45 |
| *** hasharAway has quit IRC | 20:46 | |
| *** hashar has joined #zuul | 20:48 | |
| *** hashar has quit IRC | 21:32 | |
| *** harrymichal has quit IRC | 21:38 | |
| *** harrymichal has joined #zuul | 21:38 | |
| *** harrymichal has quit IRC | 21:45 | |
| *** harrymichal has joined #zuul | 21:57 | |
| *** armstrongs has joined #zuul | 22:01 | |
| *** armstrongs has quit IRC | 22:11 | |
| *** y2kenny has quit IRC | 22:29 | |
| openstackgerrit | Merged zuul/nodepool master: Filter active images for OpenStack provider https://review.opendev.org/713471 | 22:31 |
| *** zxiiro has quit IRC | 22:43 | |
| openstackgerrit | James E. Blair proposed zuul/zuul-registry master: Very rarely, we see an object in swift doesn't match the sha256 we expect. Nor does the Etag (swift-calculated md5sum) match the md5sum that openstacksdk calculated on upload. Something is going wrong somewhere, but nothing is raising an exception. Thi https://review.opendev.org/716444 | 22:45 |
| corvus | oh dear | 22:45 |
| openstackgerrit | James E. Blair proposed zuul/zuul-registry master: Add debug/verification for uploads https://review.opendev.org/716444 | 22:45 |
| corvus | mordred: ^ i think that may help us localize the problem | 22:46 |
| mordred | corvus: cool! also - nice commit message the first time | 22:46 |
| mordred | corvus: I'm excited to learn what's going on here | 22:48 |
| corvus | mordred: i'm going to just start doing them as stream-of-conciousness james joyce style | 22:48 |
| mordred | corvus: please | 22:49 |
| *** tosky has quit IRC | 22:54 | |
| *** ianychoi has quit IRC | 23:15 | |
| *** ianychoi has joined #zuul | 23:16 | |
| openstackgerrit | Mohammed Naser proposed zuul/zuul-jobs master: golangci-lint: add job https://review.opendev.org/716452 | 23:18 |
| mordred | mnaser: you didn't want to add inline comment reporting? :) | 23:33 |
| mnaser | mordred: that was my follow up actually :D i was trying to see what the output looked like :) | 23:34 |
| mordred | mnaser: nice | 23:37 |
| openstackgerrit | Mohammed Naser proposed zuul/zuul-jobs master: golangci-lint: add job https://review.opendev.org/716452 | 23:41 |
| corvus | mnaser, mordred: i did enough work on this to know it's not a 2-hour programming job. it's not hugely complicated, but it would probably take me the better part of a week with few distractions. | 23:51 |
| mordred | corvus: nod | 23:51 |
| mnaser | corvus: yeah, it does seem relatively non-trivial and one of those things you need to get right (or otherwise you end up giving root) | 23:53 |
| corvus | yep. still very much want to do it though. | 23:54 |
| *** Goneri has quit IRC | 23:57 | |
| mnaser | today on "i thought this will be a simple job", mnaser runs into error "[Errno 8] Exec format error: 'golangci-lint'" | 23:59 |
| mnaser | the only other time i've ever seen that is when running something on the wrong architecture? | 23:59 |
| *** shanemcd has quit IRC | 23:59 | |
| clarkb | mnaser: we do have arm64 nodes | 23:59 |
| clarkb | (though I doubt you mixed them in accidentally) | 23:59 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!