| fungi | easy enough to check... | 00:21 |
|---|---|---|
| fungi | survey says... nope: https://packages.ubuntu.com/focal/python3-pip | 00:21 |
| ianw | ok, yeah it's that we list it in the packages to install when we take that path | 00:22 |
| fungi | python3-pip and python3.8-venv both depend on python-pip-whl which provides an unpackable wheel copy of pip | 00:23 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: [wip] fix venv https://review.opendev.org/726704 | 00:24 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: Revert "Revert "ensure-tox: use venv to install"" https://review.opendev.org/726705 | 00:24 |
| fungi | i think python3-pip winds up bouncing through ensure-pip to unpack the contents of python-pip-whl into the system context, while the venv module unpacks the same into any venv it creates | 00:24 |
| fungi | mmm, i guess not on the first supposition... python3-pip actually includes pip directly: https://packages.ubuntu.com/focal/all/python3-pip/filelist | 00:26 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: [wip] fix venv https://review.opendev.org/726704 | 00:29 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: Revert "Revert "ensure-tox: use venv to install"" https://review.opendev.org/726705 | 00:29 |
| ianw | ok, that passes and i think is the solution, let clean it up | 00:39 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: ensure-pip: always check for python3-venv on Debuntu https://review.opendev.org/726704 | 00:44 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: Revert "Revert "ensure-tox: use venv to install"" https://review.opendev.org/726705 | 00:44 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: [wip] venv probe full path https://review.opendev.org/726715 | 01:00 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: ensure-pip: always check for python3-venv on Debuntu https://review.opendev.org/726704 | 01:10 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: Revert "Revert "ensure-tox: use venv to install"" https://review.opendev.org/726705 | 01:10 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: [wip] venv probe full path https://review.opendev.org/726715 | 01:10 |
| *** swest has quit IRC | 01:47 | |
| *** swest has joined #zuul | 02:02 | |
| *** cloudnull2 has joined #zuul | 02:15 | |
| *** cloudnull has quit IRC | 02:16 | |
| *** cloudnull2 is now known as cloudnull | 02:16 | |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: ensure-pip: use full python3 path https://review.opendev.org/726715 | 03:17 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: ensure-pip: use full python3 path https://review.opendev.org/726715 | 03:40 |
| *** evrardjp has quit IRC | 04:36 | |
| *** evrardjp has joined #zuul | 04:36 | |
| AJaeger | thanks, ianw ! Will review later... | 05:17 |
| AJaeger | ianw: this is all about python3 - is the python2 package fine or do we need to do anything there as well? | 05:55 |
| ianw | AJaeger: this is really just about what we're exporting as the tool to install venv's, which should just be "python3 -m venv" as every platform has that | 06:02 |
| AJaeger | ianw: Thanks, Understood now. I double checked that py27 env uses py27 as well | 06:03 |
| ianw | yeah, the *tox* should be fine installed in a python3 venv, and it can then run with python2 | 06:05 |
| ianw | like as in -e py27 | 06:05 |
| *** dpawlik has joined #zuul | 06:10 | |
| AJaeger | yep, that worked | 06:10 |
| *** dpawlik has quit IRC | 06:13 | |
| *** yolanda has joined #zuul | 06:18 | |
| *** dpawlik has joined #zuul | 06:18 | |
| *** saneax has joined #zuul | 06:31 | |
| *** fbo|off is now known as fbo|afk | 06:55 | |
| *** jcapitao has joined #zuul | 07:10 | |
| *** tosky has joined #zuul | 07:34 | |
| *** rpittau|afk is now known as rpittau | 07:36 | |
| *** hashar has joined #zuul | 07:38 | |
| *** avass has joined #zuul | 07:42 | |
| *** jpena|off is now known as jpena | 07:53 | |
| *** panda|pto is now known as panda | 08:04 | |
| *** arxcruz has quit IRC | 08:10 | |
| *** arxcruz has joined #zuul | 08:10 | |
| *** Miouge- has joined #zuul | 08:15 | |
| *** EmilienM has quit IRC | 08:15 | |
| *** Miouge has quit IRC | 08:15 | |
| *** SpamapS has quit IRC | 08:15 | |
| *** SpamapS has joined #zuul | 08:16 | |
| *** EmilienM has joined #zuul | 08:16 | |
| *** sshnaidm|off is now known as sshnaidm | 09:15 | |
| openstackgerrit | Albin Vass proposed zuul/zuul master: Enables whitelisting and configuring callbacks https://review.opendev.org/717260 | 09:16 |
| openstackgerrit | Albin Vass proposed zuul/zuul master: Enables whitelisting and configuring callbacks https://review.opendev.org/717260 | 09:18 |
| avass | tobiash: how about that? | 09:18 |
| *** pingou has joined #zuul | 09:35 | |
| *** nils has joined #zuul | 09:36 | |
| * pingou waives to nils | 09:36 | |
| * nils waves back | 09:36 | |
| nils | I just asked about testing Zuul jobs over on #fedora-ci -- I want to test a job I'm developing locally i.e. before committing it to the repo. | 09:39 |
| nils | Just to see if I use the right macros/variables and if my playbook works and so on. | 09:40 |
| nils | And the only thing I found re: doing this would be to follow the quick-start installation and tutorial and spin up the whole thing locally, only it doesn't seem to work (this is with SELinux not enforcing, just in case): https://paste.centos.org/view/e7ee4404 | 09:42 |
| nils | Does anybody around have an idea what's going wrong? Or is there a lighter-weight way how I can test my job? | 09:42 |
| avass | nils: you could test the playbooks and roles by running ansible, but you would need a zuul to test the job config as far as I know | 10:06 |
| nils | thanks | 10:07 |
| avass | nils: but if you're using any zuul: vars you need to mock them away, you can do that by passing an extra-vars file | 10:09 |
| avass | nils: you could do something like this: https://review.opendev.org/gitweb?p=zuul/zuul-jobs.git;a=blob;f=test-playbooks/artifactory/run.yaml;h=08050d80b0f852c92747e607e6a76d64d4c3b37c;hb=7004f0328b2b30fc76f2dba7778ccf8cfd551c57#l64 | 10:11 |
| nils | avass, the zuul vars are part of what I wanted to test ;) because it's the first custom job I'm writing | 10:12 |
| avass | nils: ah :) | 10:12 |
| nils | but I guess I can ask here :) | 10:12 |
| nils | So, if I wanted to do something with the diff of a pull request, would zuul.branch as the "base" of the diff and zuul.change as the "tip" be the right variables to use? | 10:14 |
| nils | The difference between zuul.change and zuul.patchset isn't obvious when reading https://zuul-ci.org/docs/zuul/reference/jobs.html#change-items | 10:14 |
| nils | I assume one of the identifies the pull request and the other the tip of the feature branch being submitted, but I'm not sure which is which. | 10:15 |
| *** rpittau is now known as rpittau|bbl | 10:15 | |
| avass | nils: hmm, I don't know how zuul.patchset behaves for something other than gerrit | 10:16 |
| nils | avass, I assume that the "change items" variables are the place to look for this information, though, would that be right? | 10:17 |
| nils | Gettting more concrete, I want to pipe the diff of a submitted PR through ansible-review (because ansible-review on the whole repo takes ages and there might be broken stuff in it anyway). I need the branch point (which would be the tip of the branch the PR is submitted agains) and the tip of the PR (branch). | 10:18 |
| avass | nils: looks like the patchset for github is the sha of HEAD on the PR | 10:20 |
| nils | That's great info, thanks again! | 10:20 |
| avass | and zuul.branch is the branch being targeted e.g master in most cases | 10:21 |
| avass | I think | 10:22 |
| avass | nils: no problem :) | 10:22 |
| nils | avass, yeah that would let me do `git diff {{ zuul.branch }}..{{ zuul.patchset }} | ansible-review ...` | 10:24 |
| avass | yeah :) | 10:24 |
| nils | Ahh maybe rather `git diff $(git merge-base {{ zuul.branch }} {{ zuul.patchset }})..{{ zuul.patchset }}` :D | 10:24 |
| nils | Not sure if this is the same as `... { zuul.branch }...{ zuul.patchset } ...` | 10:26 |
| avass | nils: wouldn't that produce the same output? | 10:26 |
| *** jcapitao is now known as jcapitao_lunch | 10:27 | |
| nils | If say master has moved on in the meantime, then the diff isn't meaningful | 10:27 |
| nils | I'm not sure if three dots achieve the same with git diff as computing the merge base manually | 10:27 |
| avass | nils: ah, but you'd want to diff against the head of master right? git diff $(git rev-parse {{ zuul.branch }}) {{ zuul.patchset }} | 10:29 |
| nils | avass, only if the head of master is the merge base, too :) | 10:30 |
| nils | since git diff can cope with branch names I don't have to use git rev-parse I guess | 10:30 |
| avass | ah yeah, heh | 10:31 |
| avass | zuul-jobs-maint: can you take a look at https://review.opendev.org/#/c/690057/ | 10:57 |
| *** fbo|afk is now known as fbo | 11:06 | |
| *** iurygregory has quit IRC | 11:37 | |
| *** jpena is now known as jpena|lunch | 11:37 | |
| *** pingou has left #zuul | 11:42 | |
| *** jcapitao_lunch is now known as jcapitao | 11:42 | |
| *** toabctl has quit IRC | 11:43 | |
| *** rfolco has joined #zuul | 11:53 | |
| *** rfolco is now known as rfolco|rover | 11:53 | |
| *** rpittau|bbl is now known as rpittau | 11:58 | |
| *** iurygregory has joined #zuul | 11:58 | |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Remove requiring tox_envlist https://review.opendev.org/726829 | 12:19 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Deprecate default tox_envlist: venv https://review.opendev.org/726830 | 12:20 |
| *** rlandy has joined #zuul | 12:22 | |
| avass | zuul-jobs-maint: I think the default behaviour of defaulting to venv in the tox role is a bit unintuitive, can we deprecate that? | 12:23 |
| *** jpena|lunch is now known as jpena | 12:33 | |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Remove requiring tox_envlist https://review.opendev.org/726829 | 12:35 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Deprecate default tox_envlist: venv https://review.opendev.org/726830 | 12:36 |
| AJaeger | mordred, mnaser , please review this stack: https://review.opendev.org/#/c/726704 | 12:37 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Do not fail if stestr or testr is not found https://review.opendev.org/726836 | 12:46 |
| *** zxiiro has joined #zuul | 12:57 | |
| *** sgw has joined #zuul | 12:58 | |
| *** iurygregory has quit IRC | 13:01 | |
| avass | mordred: why do we run tox here: https://opendev.org/zuul/zuul-jobs/src/branch/master/roles/tox/tasks/siblings.yaml#L1 ? | 13:02 |
| *** iurygregory has joined #zuul | 13:02 | |
| AJaeger | avass: to install all requirements - and then we replace the "siblings" | 13:03 |
| avass | AJaeger: ah | 13:03 |
| avass | AJaeger: that probably needs a comment | 13:04 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Deprecate default tox_envlist: venv https://review.opendev.org/726830 | 13:10 |
| AJaeger | avass: let's double check that with mordred | 13:10 |
| avass | AJaeger: if you have time, want to take a look at the upload-artifactory role? :) https://review.opendev.org/#/c/725678/ | 13:15 |
| AJaeger | avass: will try later today | 13:16 |
| avass | AJaeger: thanks! | 13:16 |
| *** nhicher has quit IRC | 13:19 | |
| *** fbo has quit IRC | 13:19 | |
| openstackgerrit | Merged zuul/zuul-jobs master: ensure-pip: always check for python3-venv on Debuntu https://review.opendev.org/726704 | 13:20 |
| *** Goneri has joined #zuul | 13:21 | |
| *** Defolos has joined #zuul | 13:25 | |
| *** fbo has joined #zuul | 13:27 | |
| *** nhicher has joined #zuul | 13:28 | |
| mordred | AJaeger, avass yes - that's right | 13:29 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Add explanatory comment to tox siblings https://review.opendev.org/726845 | 13:32 |
| avass | mordred: ^ :) | 13:32 |
| *** dpawlik has quit IRC | 13:38 | |
| *** dpawlik has joined #zuul | 13:42 | |
| fungi | nils: avass: there's this series of changes which hope to separate out a "runner" which could be used to locally (re)execute a zuul job without all the zuul daemons... still, it's a bit like saying you want to test a shell scrip without using the shell to do it, zuul not only supplies variables from the triggering context but also compiles playbooks and roles from a limitless pool of different git repositories, | 13:58 |
| fungi | expects that changes being tested could depend on other proposed changes which could themselves even contain changes or addition to job configuration, and so on | 13:58 |
| nils | fungi, that's great to hear and I understand that "running Zuul without Zuul" isn't the same thing. But I imagine that it would allow people to catch a couple of mistakes locally before submitting their changes. | 14:00 |
| avass | fungi: how's that going? I thought there was just a spec for it so far | 14:00 |
| nils | Also, a shell is slightly smaller (and easier to get running from scratch) than Zuul :) | 14:01 |
| nils | Not sure what exactly is going wrong on my end, but I don't seem to be lucky trying it here. | 14:02 |
| fungi | nils: yep, it was an imperfect analogy | 14:02 |
| openstackgerrit | Monty Taylor proposed zuul/zuul-website master: Switch website to Gatsby https://review.opendev.org/717371 | 14:02 |
| openstackgerrit | Monty Taylor proposed zuul/zuul-website master: Add blog to website https://review.opendev.org/724648 | 14:02 |
| nils | fungi, no worries, my comment was in jest | 14:02 |
| fungi | avass: the changes for that review topic are in merge conflict since mid-last year, though there's movement on the spec: https://review.opendev.org/681277 | 14:04 |
| *** tumble has joined #zuul | 14:05 | |
| *** cdearborn has joined #zuul | 14:09 | |
| avass | fungi: oh, nice :) | 14:17 |
| tobiash | tristanC: I added a question to the zuul-runner spec | 14:35 |
| *** jhesketh has quit IRC | 14:35 | |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Remove requiring tox_envlist https://review.opendev.org/726829 | 14:42 |
| openstackgerrit | Merged zuul/zuul-jobs master: Add explanatory comment to tox siblings https://review.opendev.org/726845 | 14:46 |
| corvus | nils: it's also worth noting that since zuul's configuration is dynamically evaluated, you can 'test' the job by pushing up a change that implements it. you can revise that change until it works as expected, all before merging it. | 14:53 |
| corvus | nils: for determining the "diff", the best way to do that may be "diff origin/{{zuul.branch}} {{zuul.branch}}" | 14:53 |
| corvus | nils: see the discussion at https://zuul-ci.org/docs/zuul/reference/jobs.html#git-repositories for an explanation about that | 14:54 |
| corvus | nils: (2nd paragraph) | 14:54 |
| corvus | nils: i would not recommend zuul.patchset for that -- it may not represent the actual state of the repo after any merges | 14:55 |
| *** sshnaidm is now known as sshnaidm|afk | 14:57 | |
| *** klindgren_ has joined #zuul | 15:09 | |
| *** klindgren has quit IRC | 15:09 | |
| -openstackstatus- NOTICE: Our CI mirrors in OVH BHS1 and GRA1 regions were offline between 12:55 and 14:35 UTC, any failures there due to unreachable mirrors can safely be rechecked | 15:09 | |
| *** hashar has quit IRC | 15:14 | |
| zbr | corvus: fungi: any change to look at https://review.opendev.org/#/c/690057/ ? | 15:18 |
| openstackgerrit | Andreas Jaeger proposed zuul/zuul-jobs master: Add new non-npm specific javascript jobs https://review.opendev.org/726547 | 15:22 |
| nils | corvus, aah, what's zuul.patchset then? | 15:24 |
| tristanC | tobiash: thanks, just replied | 15:25 |
| nils | Also, what do you think of "diff origin/{{ zuul.branch }} {{ zuul.branch }}" vs. "diff origin/{{ zuul.branch }}...{{ zuul.branch }}"? | 15:25 |
| openstackgerrit | Merged zuul/zuul-jobs master: Revert "Revert "ensure-tox: use venv to install"" https://review.opendev.org/726705 | 15:29 |
| openstackgerrit | Merged zuul/zuul-jobs master: ensure-pip: use full python3 path https://review.opendev.org/726715 | 15:29 |
| corvus | nils: zuul.patchset is just informative; it's use is mainly to see what version of a change or pr was tested. and i'm not really sure about the difference between " " and "..." in git diff | 15:37 |
| nils | corvus, "diff ref1 ref2" or "diff ref1..ref2" is just the plain diff between two trees, "diff ref1...ref2" is the diff between the merge point of both and ref2, it's equivalent to "diff $(git merge-base ref1 ref2) ref2" | 15:39 |
| AJaeger | mordred: I updated your change https://review.opendev.org/726547 - is that ready for review now? | 15:41 |
| *** kmalloc has joined #zuul | 15:41 | |
| corvus | nils: ah. zuul is going to make "origin/{{zuul.branch}}" be the 'previous state' -- where that might either be the current actual tip of the branch, or, in the case of dependent changes, it could be the resulting speculative merge of the dependent change(s). then "{{zuul.branch}}" or HEAD is going to be the merge of the current change onto that. if it's fast-forwardable, it'll be a ff, otherwise it'll be | 15:44 |
| corvus | a new merge commit. origin/{{zuul.branch}} should always be a parent of {{zuul.branch}}, so i think the commands should be identical in this case. | 15:44 |
| clarkb | nils: I think they will be the same in the zuul context due to how zuul works | 15:44 |
| clarkb | corvus: yup that is my understanding too | 15:44 |
| corvus | clarkb: whew :) | 15:44 |
| mordred | AJaeger: yes - I believe it is - you can see it working here: https://review.opendev.org/#/c/726554/ | 15:45 |
| mordred | AJaeger: oh - actually - I think it wants one more little update | 15:49 |
| mordred | AJaeger: no - nevermind. that one is good | 15:49 |
| nils | corvus, clarkb: thanks, I'll do it "by the book" nonetheless :) | 15:50 |
| openstackgerrit | Monty Taylor proposed zuul/zuul master: Update node to v14 and reorg the jobs a bit https://review.opendev.org/726553 | 15:50 |
| clarkb | mordred: AJaeger re ^ yall might have input on the horizon testing threads taht popped up over the weekend? | 15:51 |
| avass | corvus, nils: Oh right, I should have thought of the speculative merging going on in zuul | 15:51 |
| mordred | clarkb: actually - with the new set of jobs we've got there - we should be able to not install yarn when it's not needed | 15:52 |
| mordred | clarkb: and then we can have horizon migrate to those | 15:52 |
| mordred | let me respond | 15:53 |
| clarkb | mordred: thanks! | 15:53 |
| mordred | AJaeger: with that - let me update those jobs to handle not installing yarn | 15:55 |
| openstackgerrit | Andreas Jaeger proposed zuul/zuul-jobs master: Add new non-npm specific javascript jobs https://review.opendev.org/726547 | 16:01 |
| AJaeger | mordred: sorry, just pushed one fix - back over to you ^ | 16:02 |
| openstackgerrit | Monty Taylor proposed zuul/zuul-jobs master: Add new non-npm specific javascript jobs https://review.opendev.org/726547 | 16:02 |
| mordred | AJaeger: crap. didn't see that. let me grab your fix | 16:02 |
| AJaeger | mordred: I didn't see IRC ;) | 16:02 |
| openstackgerrit | Monty Taylor proposed zuul/zuul-jobs master: Add new non-npm specific javascript jobs https://review.opendev.org/726547 | 16:03 |
| mordred | AJaeger, clarkb, yoctozepto: ok. that should be good ^^ | 16:03 |
| yoctozepto | mordred: ack, looks fine to my eye | 16:04 |
| yoctozepto | trying out in jslib | 16:04 |
| yoctozepto | I should really find some time to push that unlucky thingy | 16:04 |
| yoctozepto | I am not really sure it will pick those job amendments though | 16:04 |
| *** tumble has quit IRC | 16:08 | |
| AJaeger | mordred: LGTM, hope it passes all tests ;) | 16:08 |
| *** rpittau is now known as rpittau|afk | 16:09 | |
| *** ianychoi_ is now known as ianychoi | 16:09 | |
| tobiash | mnaser: do you run the swift proxy we talked about a few months ago? | 16:09 |
| mnaser | tobiash: i never got around completing it :( | 16:10 |
| tobiash | ah ok | 16:10 |
| tobiash | ours broke after an update to python 3.8 because eventlet needs pyopenssl or it breaks openstacksdk | 16:11 |
| tobiash | https://github.com/eventlet/eventlet/issues/526 in case someone suffers from "wrap_socket() got an unexpected keyword argument '_context'" | 16:11 |
| avass | mordred, AJaeger: Don't we need to install npm if we don't install yarn? :) | 16:13 |
| avass | or am I just missing that? | 16:13 |
| mordred | avass: hrm. oh - good point! we were counting on ensure-yarn to ensure npm | 16:13 |
| mordred | avass: or - maybe we get npm from node in the first place? | 16:15 |
| avass | mordred: but we install node in ensure-yarn, so if we skip ensure-yarn we need to run ensure-nodejs at least | 16:16 |
| mordred | avass: yeah | 16:17 |
| mnaser | what do we think about renaming opendev-buildset-registry (or $site-buildset-registry) to 'base-buildset-registry' -- the reason i ask is because https://zuul.vexxhost.dev/t/opendev/config-errors -- unless i create a fake 'opendev-buildset-registry' job... i'll always have that config error | 16:17 |
| mnaser | i could technically _not_ load any jobs at all, but as i'm only running tox-* jobs, but, that means potentially missing out on adding jobs from zuul-jobs later | 16:18 |
| openstackgerrit | Monty Taylor proposed zuul/zuul-jobs master: Add new non-npm specific javascript jobs https://review.opendev.org/726547 | 16:18 |
| openstackgerrit | Merged zuul/zuul-jobs master: cabal-test: add build target job variable https://review.opendev.org/726266 | 16:20 |
| openstackgerrit | Merged zuul/zuul-jobs master: haskell-stack-test: add build target job variable https://review.opendev.org/726267 | 16:20 |
| avass | mordred: lgtm, except for the linting | 16:20 |
| mordred | avass: doh | 16:20 |
| *** jcapitao has quit IRC | 16:22 | |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: tox: allow tox to be upgraded https://review.opendev.org/690057 | 16:25 |
| avass | zbr: rebase that for you ^ | 16:25 |
| avass | zbr: uhm, nevermind I'll fix it | 16:26 |
| zbr | thanks! | 16:28 |
| openstackgerrit | Monty Taylor proposed zuul/zuul-jobs master: Add new non-npm specific javascript jobs https://review.opendev.org/726547 | 16:30 |
| openstackgerrit | Monty Taylor proposed zuul/zuul-jobs master: Extract ensure-javascript-build-tool role https://review.opendev.org/726900 | 16:30 |
| mordred | avass, zbr: let's make sure ianw looks at that when he's up before we land it - he's been doing a bunch of work updating the tox roles and is juggling a lot of things has all of the things paged in (it doens't look like a problem to me - but I think we should run it by ianw) | 16:31 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: tox: allow tox to be upgraded https://review.opendev.org/690057 | 16:32 |
| avass | mordred, zbr: sure | 16:33 |
| avass | zbr: that should be alright right? | 16:33 |
| avass | zbr: I'll leave it to you now | 16:33 |
| zbr | fungi: any idea about a better name for tox_constraint variable? | 16:34 |
| *** evrardjp has quit IRC | 16:36 | |
| avass | zbr: tox_required_version? | 16:36 |
| fungi | tox_requirements or tox_packages or tox_versions maybe? | 16:36 |
| *** evrardjp has joined #zuul | 16:36 | |
| avass | or that | 16:36 |
| fungi | zbr: and as for my comment about the tox-venv plugin, i don't have anything against the plugin personally (and i make extensive use of it in some of my projects), but it seems unrelated to the rest of the change and isn't mentioned in the commit message at all | 16:39 |
| fungi | and it could introduce subtle behavior changes | 16:40 |
| fungi | (though hopefully not, it still deserves its own discussion) | 16:40 |
| avass | fungi: that's just the test-playbook though :( | 16:41 |
| avass | :)* | 16:42 |
| fungi | oh, yep, i missed the test- on the front of that directory name | 16:43 |
| fungi | sorry about that | 16:43 |
| yoctozepto | mordred: honestly I don't think nodejs-test-dependencies should even install chromium ;-) | 16:46 |
| yoctozepto | this is only because some of these javascript project target browsers | 16:46 |
| mordred | yoctozepto: ah - nod | 16:46 |
| yoctozepto | I installed firefox because the original legacy jobs used them | 16:46 |
| zbr | i will put "tox_requirements" describes it better. | 16:46 |
| yoctozepto | if zuul picked the depends-on correctly, then it seems unit tests still pass :-) | 16:47 |
| yoctozepto | using the new-old jobs | 16:47 |
| yoctozepto | now could use using the new ones | 16:48 |
| yoctozepto | it actually relies on nodejs8-jobs template to pull these in | 16:49 |
| yoctozepto | mordred: are you planning updating those templates too? | 16:49 |
| mordred | yoctozepto: also - check my followup patch in https://review.opendev.org/726900 - there's a noteworthy change - which is not installing dependencies in pre but actually installing them in during run | 16:49 |
| mordred | yoctozepto: I was planning on talking with folks about it - I think we want to make sure updating them doesn't bork people | 16:50 |
| mordred | but I think it would be a good idea to | 16:50 |
| zbr | avass: if you have time look at https://github.com/ansible/ansible-lint/pull/781 as it fixes one error I seen running on zuul-jobs. | 16:51 |
| yoctozepto | mordred: I could then depend on it and see if it breaks these precious unit tests | 16:51 |
| yoctozepto | mordred: aye, why the change though? does pre have more magic than being retried? | 16:53 |
| mordred | yoctozepto: just the retries - but the idea is that pre is setup and shouldn't really ever fail in normal dev workflow, only in environmentally weird cases like an apt repo being unreachable. but devs write patches to add deps all the time - so it's not really appropriate to have a job fail in pre because of that dev action | 16:56 |
| yoctozepto | mordred: ack, good practice to remember, tradeoff of shorts but makes sense, thanks | 16:57 |
| mordred | ++ | 16:58 |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul-jobs master: tox: allow tox to be upgraded https://review.opendev.org/690057 | 16:58 |
| corvus | mordred, yoctozepto: yeah, that matches what we do with tox (install python and tox in pre; install project deps in run) | 16:59 |
| yoctozepto | corvus, mordred: well, that is because it is tox that actually does it in the first place :-) | 17:02 |
| yoctozepto | corvus, mordred: but yeah, analogy accepted | 17:02 |
| corvus | yoctozepto: (we could have put tox --notests in the pre playbook but did not) | 17:03 |
| yoctozepto | corvus: fair enough :-) | 17:03 |
| avass | zbr: lgtm | 17:07 |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul-jobs master: yamlint: EOF newlines and comments indent https://review.opendev.org/725516 | 17:12 |
| *** sshnaidm|afk is now known as sshnaidm | 17:12 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: spec: add a zuul-runner cli https://review.opendev.org/681277 | 17:18 |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul-jobs master: tox: allow tox to be upgraded https://review.opendev.org/690057 | 17:20 |
| *** jpena is now known as jpena|off | 17:25 | |
| clarkb | mnaser: mordred I sent http://lists.openstack.org/pipermail/openstack-discuss/2020-May/014809.html to openstack-discuss to mention the tox behavior that came up over the weekend | 17:35 |
| clarkb | do you think it would be helpful to also send that to the zuul-discuss list? | 17:36 |
| fbo | Hi zuul-maints, I'd like to know if that Ansible plugin dir adjacent to playbook error is expected there: https://pagure.io/fedora-infra/ansible/pull-request/54 | 17:37 |
| clarkb | fbo: my understanding is that plugins are not "safe" from a security standpoint so cannot be loaded in an untrusted (or pre merge context) | 17:38 |
| clarkb | fbo: this means you need to ship them in trusted repos | 17:38 |
| fbo | clarkb: plugins are in the root of the project but zuul job playbook in ci/ so I'm not sure about that adjacent rule.https://opendev.org/zuul/zuul/src/branch/master/zuul/executor/server.py#L1610 | 17:41 |
| clarkb | fbo: its not the adjacency its them existing at all I think | 17:42 |
| fbo | that's unfortunate that this project structure prevent the repo to handled speculatively | 17:42 |
| tristanC | iiuc the plugins are not necessary for zuul, it seems like the check may be too restrictives | 17:42 |
| clarkb | oh I see you are saying those plugins exist for ansible and not zuul + ansible | 17:42 |
| clarkb | so zuul should just ignore them? | 17:42 |
| fungi | oh, it's plugins shipped in a repository but which zuul isn't expected to use? | 17:42 |
| tristanC | fungi: clarkb: yes, we would like zuul to ignore them | 17:43 |
| fungi | i wonder if that can be excluded in the project entry for the tenant | 17:43 |
| clarkb | got it, in that case I think the issue is going to be related to paths and how zuul/ansible looks for things like playbooks and plugins | 17:43 |
| fbo | fungi: yes I guess | 17:43 |
| clarkb | fungi: I think this is controlled more by ansible than zuul | 17:43 |
| fungi | oh | 17:43 |
| clarkb | its going to depend on whether or not ansible can be told to ignore those plugins when running under zuul | 17:44 |
| clarkb | (I don't know what the existing behavior does, in theory those plugins would be loaded as is today otherwise we wouldn't be guarding against it) | 17:44 |
| *** sanjayu_ has joined #zuul | 17:45 | |
| *** saneax has quit IRC | 17:48 | |
| *** fbo is now known as fbo|off | 17:52 | |
| fungi | maybe it can be given a pluginpath override | 17:56 |
| fungi | or some such | 17:56 |
| * fungi makes vague hand-waving gesture | 17:56 | |
| avass | fungi: maybe have a playbook directory that can't be loaded by zuul? | 18:08 |
| *** iurygregory has quit IRC | 18:25 | |
| AJaeger | mordred: could you check my comment at https://review.opendev.org/#/c/726547/13/playbooks/javascript/pre.yaml ? | 18:27 |
| corvus | fbo|off, tristanC, nils: i agree it seems like zuul should only be looking under the ci/ directory, but it seems to be checking the root of the repo; at first glance that seems unexpected and i think it's worth further inspection to understand it. | 18:28 |
| nils | corvus, I'll look into it tomorrow | 18:33 |
| *** iurygregory has joined #zuul | 18:38 | |
| avass | AJaeger: yeah that should be ensure-nodejs, good catch | 18:40 |
| *** toabctl has joined #zuul | 18:57 | |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul-jobs master: tox: allow tox to be upgraded https://review.opendev.org/690057 | 18:59 |
| zbr | anyone aware of TypeError: unsupported operand type(s) for -=: 'Retry' and 'int' -- pip/urlib3? https://zuul.opendev.org/t/zuul/build/abcad91f4b044b12a695f6bdcab7b88b | 19:23 |
| zbr | this error loks quite weird, two nested exceptions | 19:23 |
| AJaeger | zbr, earlier error was "http.client.RemoteDisconnected: Remote end closed connection without response | 19:26 |
| AJaeger | zbr: so a not handled timeout? | 19:26 |
| zbr | and apparently nobody cared to fix that bug in pip, i see >2yr bugs with same behavior. | 19:27 |
| openstackgerrit | Andreas Jaeger proposed zuul/zuul-jobs master: Add new non-npm specific javascript jobs https://review.opendev.org/726547 | 19:29 |
| AJaeger | avass: thanks. fixed it ^ | 19:30 |
| *** kmalloc has quit IRC | 19:31 | |
| openstackgerrit | Andreas Jaeger proposed zuul/zuul-jobs master: Extract ensure-javascript-build-tool role https://review.opendev.org/726900 | 19:33 |
| openstackgerrit | Andreas Jaeger proposed zuul/zuul-jobs master: Extract ensure-javascript-build-tool role https://review.opendev.org/726900 | 19:35 |
| *** nils has quit IRC | 19:51 | |
| openstackgerrit | Andreas Jaeger proposed zuul/zuul-jobs master: Extract ensure-javascript-build-tool role https://review.opendev.org/726900 | 19:56 |
| mordred | AJaeger: ++ thank you! | 20:03 |
| openstackgerrit | Andreas Jaeger proposed zuul/zuul-jobs master: Extract ensure-javascript-build-tool role https://review.opendev.org/726900 | 20:05 |
| AJaeger | mordred: fixed one more problem ^ | 20:05 |
| AJaeger | mordred: testing in https://review.opendev.org/#/c/726940/1 | 20:06 |
| AJaeger | mordred: could you review my changes, please? - and take over again? I'm signing off now... | 20:07 |
| *** avass has quit IRC | 20:08 | |
| *** dpawlik has quit IRC | 20:46 | |
| *** tumble has joined #zuul | 21:20 | |
| *** rfolco|rover has quit IRC | 21:21 | |
| *** rfolco|rover has joined #zuul | 21:26 | |
| *** guillaumec has joined #zuul | 21:27 | |
| *** sshnaidm is now known as sshnaidm|afk | 22:11 | |
| *** rfolco|rover has quit IRC | 22:42 | |
| *** kmalloc has joined #zuul | 22:46 | |
| *** tosky has quit IRC | 22:57 | |
| *** sshnaidm|afk has quit IRC | 23:14 | |
| *** sshnaidm has joined #zuul | 23:15 | |
| *** sshnaidm is now known as sshnaidm|afk | 23:16 | |
| *** guillaumec has quit IRC | 23:16 | |
| *** tumble has quit IRC | 23:47 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!