| *** frenzyfriday has joined #zuul | 00:01 | |
| *** frenzyfriday has quit IRC | 00:06 | |
| *** decimuscorvinus_ has joined #zuul | 00:26 | |
| *** decimuscorvinus has quit IRC | 00:26 | |
| *** rlandy|bbl is now known as rlandy | 00:33 | |
| *** frenzyfriday has joined #zuul | 00:52 | |
| *** frenzyfriday has quit IRC | 00:56 | |
| *** rlandy has quit IRC | 01:03 | |
| zenkuro | corvus: success-url incorectly works vsriables. I copy excat value into itself | 01:40 |
|---|---|---|
| zenkuro | so, no variable value extraction happens | 01:41 |
| *** frenzyfriday has joined #zuul | 01:46 | |
| *** wuchunyang has joined #zuul | 01:46 | |
| *** frenzyfriday has quit IRC | 01:50 | |
| *** frenzyfriday has joined #zuul | 02:40 | |
| *** frenzyfriday has quit IRC | 02:45 | |
| *** frenzyfriday has joined #zuul | 03:34 | |
| *** frenzyfriday has quit IRC | 03:38 | |
| *** bhavikdbavishi has joined #zuul | 03:58 | |
| *** bhavikdbavishi1 has joined #zuul | 04:03 | |
| *** bhavikdbavishi has quit IRC | 04:05 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 04:05 | |
| *** wuchunyang has quit IRC | 04:06 | |
| *** wuchunyang has joined #zuul | 04:06 | |
| *** bhavikdbavishi has quit IRC | 04:26 | |
| *** bhavikdbavishi has joined #zuul | 04:26 | |
| *** evrardjp has quit IRC | 04:33 | |
| *** evrardjp has joined #zuul | 04:33 | |
| *** wuchunyang has joined #zuul | 04:36 | |
| *** bhagyashris is now known as bhagyashris|away | 05:23 | |
| *** hamalq has joined #zuul | 05:39 | |
| openstackgerrit | Andreas Jaeger proposed zuul/zuul-jobs master: Update hashicorp jobs file with correct title https://review.opendev.org/747853 | 05:43 |
| *** hamalq has quit IRC | 05:43 | |
| *** bhavikdbavishi has quit IRC | 05:54 | |
| *** bhavikdbavishi has joined #zuul | 05:56 | |
| *** bhavikdbavishi1 has joined #zuul | 05:59 | |
| *** hamalq has joined #zuul | 06:00 | |
| *** bhavikdbavishi has quit IRC | 06:01 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 06:01 | |
| *** mach1na has joined #zuul | 06:04 | |
| *** hamalq_ has joined #zuul | 06:04 | |
| *** hamalq has quit IRC | 06:04 | |
| *** hamalq_ has quit IRC | 06:08 | |
| *** frenzyfriday has joined #zuul | 06:16 | |
| *** bhavikdbavishi has quit IRC | 06:47 | |
| *** mach1na has quit IRC | 06:54 | |
| *** hamalq has joined #zuul | 07:13 | |
| *** bhavikdbavishi has joined #zuul | 07:13 | |
| *** mach1na has joined #zuul | 07:16 | |
| *** hamalq has quit IRC | 07:17 | |
| *** tosky has joined #zuul | 07:21 | |
| *** hamalq has joined #zuul | 07:34 | |
| *** hamalq has quit IRC | 07:38 | |
| *** hamalq has joined #zuul | 07:49 | |
| *** holser has quit IRC | 07:52 | |
| *** jpena|off is now known as jpena | 07:52 | |
| *** hamalq has quit IRC | 07:53 | |
| *** hamalq has joined #zuul | 08:05 | |
| *** hamalq has quit IRC | 08:09 | |
| *** holser has joined #zuul | 08:11 | |
| *** holser has quit IRC | 08:11 | |
| *** holser has joined #zuul | 08:12 | |
| *** hashar has joined #zuul | 08:16 | |
| openstackgerrit | Merged zuul/zuul-jobs master: Update hashicorp jobs file with correct title https://review.opendev.org/747853 | 08:21 |
| *** sshnaidm|bbl is now known as sshnaidm | 08:36 | |
| *** mach1na has quit IRC | 08:36 | |
| *** mach1na has joined #zuul | 08:39 | |
| *** AshBullock has joined #zuul | 08:49 | |
| *** nils has joined #zuul | 09:19 | |
| *** holser has quit IRC | 09:40 | |
| *** wuchunyang has quit IRC | 10:08 | |
| *** wuchunyang has joined #zuul | 10:11 | |
| *** piotrowskim has joined #zuul | 10:13 | |
| *** wuchunyang has quit IRC | 10:15 | |
| *** holser has joined #zuul | 10:29 | |
| *** mgagne has quit IRC | 10:39 | |
| *** bhavikdbavishi has quit IRC | 10:55 | |
| *** AshBullock has quit IRC | 10:56 | |
| *** holser has quit IRC | 11:09 | |
| *** bhavikdbavishi has joined #zuul | 11:12 | |
| *** mach1na has quit IRC | 11:19 | |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: gerritconnection: use generic simpleQuery over default HTTP when polling https://review.opendev.org/749246 | 11:34 |
| *** jpena is now known as jpena|lunch | 11:34 | |
| *** vishalmanchanda has joined #zuul | 11:40 | |
| *** mach1na has joined #zuul | 11:53 | |
| *** holser has joined #zuul | 11:54 | |
| *** mattd01 has joined #zuul | 12:01 | |
| *** mach1na has quit IRC | 12:01 | |
| *** rlandy has joined #zuul | 12:01 | |
| *** mach1na has joined #zuul | 12:02 | |
| *** bhavikdbavishi has quit IRC | 12:17 | |
| *** frenzyfriday has quit IRC | 12:33 | |
| *** jpena|lunch is now known as jpena | 12:37 | |
| openstackgerrit | Haïkel Guémar proposed zuul/zuul-jobs master: Add-sshkey: ensure that ssh home directory exists https://review.opendev.org/749278 | 12:43 |
| *** AshBullock has joined #zuul | 13:17 | |
| *** Goneri has joined #zuul | 13:27 | |
| AshBullock | Hey all, I've been doing some reading of the zuul codebase and documentation as I want to start contributing, particularly I'm looking at the BitBucket driver which I see some progress has already been made, I was wondering if anyone had any time for me to ask some questions around my understanding, particularly of the drivers? | 13:41 |
| openstackgerrit | Merged zuul/zuul-jobs master: Add-sshkey: ensure that ssh home directory exists https://review.opendev.org/749278 | 13:44 |
| *** frenzyfriday has joined #zuul | 13:44 | |
| *** AshBullock has quit IRC | 13:46 | |
| *** AshBullock has joined #zuul | 13:46 | |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: gerritconnection: use generic simpleQuery over default HTTP when polling https://review.opendev.org/749246 | 13:51 |
| fungi | AshBullock: ofosos was the one working on that patch series, but it doesn't look like he's touched it for roughly a year and the last time we saw him in this irc channel was three months ago | 14:03 |
| fungi | i spoke with him briefly in brussels when i was there for fosdem and he seemed to still be interested in getting it working. he said it was mostly functional but he was having trouble tracking down a constant bitbucket api disconnection problem with it and he wasn't sure if the problem was the code or his network | 14:04 |
| fungi | AshBullock: i think lots of us in here are happy to answer general questions about zuul and its source code and writing new connection drivers, but a likely first step will be trying to deploy a patched scheduler with the current state of the bitbucket driver and see what it does. maybe do it from a quickstart deployment so you've also got a test gerrit to use as a reference connection for comparison? | 14:07 |
| zbr | fungi: corvus: simple E208 fixes: https://review.opendev.org/#/c/748498/ | 14:10 |
| *** sshnaidm has quit IRC | 14:34 | |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: gerritconnection: use generic simpleQuery over default HTTP when polling https://review.opendev.org/749246 | 14:43 |
| openstackgerrit | Clark Boylan proposed zuul/nodepool master: DNM test arm64 docker image builds https://review.opendev.org/749299 | 14:48 |
| *** sshnaidm has joined #zuul | 14:53 | |
| *** hashar is now known as hasharAway | 15:02 | |
| *** yoctozepto has quit IRC | 15:02 | |
| *** yoctozepto has joined #zuul | 15:03 | |
| AshBullock | fungi Thanks, yeah I've tried to reach out to ofosos last week to show interest in contributing and see where he'd got up to with it, yep that's exactly where I am at right now, I've got a dev zuul node with the patch set up and running with a github connection, it's currently failing to start up with the bitbucket configs added, but I've got some | 15:05 |
| AshBullock | further debugging to do before I've got any useful questions :D :) | 15:05 |
| fungi | great! excited to see some potential for further progress on that connection driver | 15:08 |
| *** AshBullock has quit IRC | 15:10 | |
| openstackgerrit | Clark Boylan proposed zuul/nodepool master: Omnibus nodepool container image fixes https://review.opendev.org/741942 | 15:15 |
| openstackgerrit | Merged zuul/zuul master: Revert PF4 build page https://review.opendev.org/749098 | 15:22 |
| *** hasharAway is now known as hashar | 15:25 | |
| *** zenkuro15 has joined #zuul | 15:31 | |
| *** mach1na has quit IRC | 15:33 | |
| *** AshBullock has joined #zuul | 15:45 | |
| zenkuro15 | hi, it looks like success-url use some kind of safe url formating, and I cant figure out how to pass variables to it. I cant address local job variables | 15:46 |
| clarkb | zenkuro15: its evaluated in the context of some zuul state (I'd have to look up what state that is) but it happens within zuul not job context iirc | 15:49 |
| *** sshnaidm is now known as sshnaidm|bbl | 15:50 | |
| zenkuro15 | clarkb: /usr/lib/python3.6/site-packages/zuul/model.py:2732 formatUrlPattern | 15:51 |
| zenkuro15 | I think is somehow related to it | 15:51 |
| clarkb | ya it is let me get a link, it sows what you can interpolate | 15:52 |
| clarkb | zenkuro15: https://opendev.org/zuul/zuul/src/branch/master/zuul/model.py#L2736-L2752 | 15:53 |
| clarkb | change, pipeline, tenant, buildset, job and build seem available | 15:53 |
| zenkuro15 | yah... but hey dont... or I do not understand how to use them | 15:55 |
| clarkb | zenkuro15: from our git logs back when we used it: success-url: http://logs.openstack.org/{change.number[4]}{change.number[5]}/{change.number}/{change.patchset}/{pipeline.name}/{build.uuid[0]}{build.uuid[1]}{build.uuid[2]}{build.uuid[3]}{build.uuid[4]}{build.uuid[5]}{build.uuid[6]}/ | 15:57 |
| zenkuro15 | clarkb: dark magic! Thanks! | 15:58 |
| *** hamalq has joined #zuul | 16:00 | |
| *** bhavikdbavishi has joined #zuul | 16:04 | |
| *** bhavikdbavishi1 has joined #zuul | 16:17 | |
| *** bhavikdbavishi has quit IRC | 16:18 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 16:18 | |
| openstackgerrit | Clark Boylan proposed zuul/nodepool master: DNM test arm64 docker image builds https://review.opendev.org/749299 | 16:28 |
| openstackgerrit | Clark Boylan proposed zuul/nodepool master: Use arm64 wheels from openstack if available https://review.opendev.org/741942 | 16:28 |
| openstackgerrit | Clark Boylan proposed zuul/nodepool master: Update bindep and vhd-util install for arm64 https://review.opendev.org/749359 | 16:28 |
| clarkb | zuulians I think https://review.opendev.org/749359 is safe to land and will help arm64 builds while being covered by x86 builds (so we can land that before we do arm64 images) | 16:29 |
| clarkb | the other two are sort of exploratory and maybe we'll find we aren't ready for arm64 imges yet or maybe we will find its good to go | 16:29 |
| *** hamalq has quit IRC | 16:31 | |
| *** hamalq has joined #zuul | 16:31 | |
| *** hamalq_ has joined #zuul | 16:33 | |
| *** zenkuro15 has quit IRC | 16:35 | |
| *** hamalq has quit IRC | 16:37 | |
| clarkb | fungi: fwiw ^ the first round of that confirmed the cp35 abi3 wheel for cryptography is used by newer pythons with newer pip | 16:44 |
| *** zenkuro has quit IRC | 16:46 | |
| *** sshnaidm|bbl is now known as sshnaidm | 16:46 | |
| fungi | that's awesome, means we don't actually need to build other abis once everyone has new enough pip | 16:52 |
| fungi | so circa 2030 ;) | 16:52 |
| *** sshnaidm is now known as sshnaidm|afk | 17:11 | |
| *** jpena is now known as jpena|off | 17:17 | |
| *** mattd01 has quit IRC | 17:19 | |
| *** hashar is now known as hasharAway | 17:22 | |
| zbr | clarkb: fungi: https://review.opendev.org/#/c/740733/ ? | 17:23 |
| zbr | also https://review.opendev.org/#/c/748737/2 which is fixing at least broken jobs, like https://zuul.opendev.org/t/zuul/build/cfad19b04a2148d1bb7f778211280429 | 17:27 |
| clarkb | zbr: for that second change I don't think that is a complete normalziation | 17:28 |
| clarkb | though fungi has a lot more of that paged in | 17:28 |
| clarkb | instead we should just run ensure-pip and then pkg resources will be present | 17:28 |
| clarkb | this is what I suggested when this first came up | 17:28 |
| zbr | this seems like using a side effect from another role in order to avoid a bug | 17:29 |
| clarkb | the bug is the package isn't installed. ensure-pip intentionally installs that package | 17:29 |
| clarkb | yes its ia side effect but an intentional one | 17:29 |
| zbr | i wonder who will be able to explain me how prAPI came to exist | 17:30 |
| zbr | that import from pkg_resources makes no sense at all | 17:30 |
| zbr | the method I added is the current implementation being imported | 17:31 |
| clarkb | its probably just to shorten the identifier | 17:31 |
| zbr | adding one extra dependency for an ansible module, which we all know that they are very sensitive to imports | 17:31 |
| zbr | is like someone deliberately wanted to make it fail | 17:32 |
| clarkb | if that is a copied implementation then your change should make that clear | 17:32 |
| clarkb | (because its vendored code) | 17:32 |
| clarkb | what? | 17:32 |
| clarkb | no one wants anything todeliberately fail | 17:32 |
| clarkb | can we please not over dramatize this? | 17:32 |
| zbr | so basically i need to document hat i vendor: return replace('-', '_') | 17:32 |
| clarkb | if its another tools verbatim api then yes I would | 17:33 |
| zbr | ok in the title? | 17:33 |
| clarkb | I would just put a comment above the function with info from where it originates | 17:34 |
| clarkb | that way if the api changes we can fix it | 17:34 |
| clarkb | and its clear who that originated from | 17:34 |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul-jobs master: Remove dependency on pkg_resources https://review.opendev.org/748737 | 17:35 |
| zbr | sure. i hope I do not have to add a copyright note for copy/pasting replace('-', '_') | 17:35 |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul-jobs master: Remove dependency on pkg_resources https://review.opendev.org/748737 | 17:38 |
| zbr | clarkb: https://review.opendev.org/#/c/748498/ also has few E208 fixes | 17:44 |
| *** rlandy has quit IRC | 17:44 | |
| *** sgw has quit IRC | 17:44 | |
| *** aluria has quit IRC | 17:44 | |
| *** rlandy has joined #zuul | 17:46 | |
| *** sgw has joined #zuul | 17:46 | |
| *** aluria has joined #zuul | 17:46 | |
| *** noonedeadpunk has quit IRC | 17:48 | |
| zbr | Maybe I should call the topic presevegate, to be in sync with https://github.com/ansible/ansible/pull/71486 | 17:48 |
| clarkb | on the e208 side I wanted to seehow the opendev/base-jobs change does before merging more changes as they have overlappibg concerns | 17:50 |
| clarkb | that landed receently so maybe give it a few morehoursbefore landing more | 17:50 |
| *** noonedeadpunk has joined #zuul | 17:51 | |
| *** AshBullock has quit IRC | 17:58 | |
| *** mattd01 has joined #zuul | 18:14 | |
| *** bhavikdbavishi has quit IRC | 18:37 | |
| clarkb | corvus: ianw: on the nodepool arm64 image side of things we time out in https://review.opendev.org/749299 which has to build wheels for things that are missing. If we mix in the openstack wheel mirror (https://review.opendev.org/741942) the job passes. That was mostly expected and points out there are wheels other than cryptography that need to be built | 18:55 |
| clarkb | I think we can safely use the openstack wheel mirror except we'll have problems if/when openstack constraints lag behind upstream releases (zuul doesn't pin those deps) | 18:55 |
| clarkb | I think that means our next steps are to identify those additional wheels and see how we can build those next | 18:55 |
| clarkb | the logs from https://review.opendev.org/749299 should identify which tarballs instead of wheels we pull | 18:56 |
| clarkb | if we're willing to accept the constraints mismatch risk and maybe disable arm64 when that becomes a problem we could likely build arm64 images at least today :0 | 19:01 |
| clarkb | er :) | 19:01 |
| clarkb | might be a good experiment eithre way | 19:01 |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul-jobs master: Remove dependency on pkg_resources https://review.opendev.org/748737 | 19:03 |
| *** frenzyfriday has quit IRC | 19:16 | |
| *** openstackgerrit has quit IRC | 19:21 | |
| *** armstrongs has joined #zuul | 19:29 | |
| *** hasharAway has quit IRC | 19:59 | |
| *** tflink has quit IRC | 20:32 | |
| *** tflink has joined #zuul | 20:33 | |
| *** mattd01 has quit IRC | 20:35 | |
| *** frenzyfriday has joined #zuul | 20:43 | |
| *** frenzyfriday has quit IRC | 20:52 | |
| *** nils has quit IRC | 20:57 | |
| *** vishalmanchanda has quit IRC | 20:58 | |
| clarkb | Thinking out loud a bit more about the nodepool arm64 image situation: what if we used the openstack wheel mirror for arm64 but made the jobs nonvoting and also split x86 from arm64? Then when openstack is caught up we're working and when not caught up it fails and we ignore it. Then once we've got the wheels built for more than just openstack we merge back into a single image manifest? | 21:24 |
| *** frenzyfriday has joined #zuul | 21:42 | |
| *** frenzyfriday has quit IRC | 21:46 | |
| clarkb | ianw: PyYAML, PrettyTable, voluptuous, openshift, pynacl, dogpile.cache, netifaces, cffi, MarkupSafe, ruamel.yaml.clib these are the packages we need to build wheels for. PyYAML, PrettyTable, voluptuous, and openshift build reasonably quickly. Then we hit pynacl and timeout so unsure of time to build the others. | 21:51 |
| clarkb | ianw: it does appear that bcrypt has grown an arm64 wheel since we last did this too which is good news | 21:52 |
| clarkb | pynacl is another pyca project | 21:53 |
| clarkb | so maybe we can gently encourage them to add arm64 wheels there too :) | 21:53 |
| clarkb | tristanC: do you know why the openshift lib doesn't publish wheels? Looks like its pure python | 21:55 |
| clarkb | (thats not really a blocker, just noticing it as an oddity that maybe they want to address) | 21:56 |
| *** hamalq_ has quit IRC | 22:16 | |
| ianw | clarkb: https://github.com/pyca/pynacl/issues/601#issuecomment-684113983 on pynacl | 22:25 |
| ianw | they really want release automation; something we've of course solved :) | 22:25 |
| ianw | the problem is we can't offer windows and osx resources | 22:26 |
| clarkb | so the issue is if we cant have windows or osx then we cant have linux? its almost like one of these is freely available and accessible... | 22:27 |
| ianw | well yeah, the theme of the discussion has been they want to avoid what they're doing now of manually uploading bits and do everything in one | 22:28 |
| ianw | it's not so much we can't have linux; it's more that we can't offer zuul as the solution | 22:29 |
| ianw | ... maybe not *totally* true ... one thought was that zuul could trigger a github action to build those wheels in the release pipeline | 22:30 |
| ianw | and then collect the github artifacts, and upload that | 22:30 |
| clarkb | It just seems like we're avoiding progress because we need to go buy licenses to make more people happy | 22:31 |
| ianw | while it's certainly doable -- i don't think we have lego blocks to do that out of the box | 22:31 |
| clarkb | ya as long as there is an api it should be doable | 22:32 |
| ianw | that's what they currently do; release the sdist, then trigger their github actions to build the wheels, collect them from an artifact page and upload to pypi manually | 22:32 |
| *** Goneri has quit IRC | 22:37 | |
| *** tosky has quit IRC | 22:43 | |
| *** frenzyfriday has joined #zuul | 23:30 | |
| *** holser has quit IRC | 23:33 | |
| *** frenzyfriday has quit IRC | 23:40 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!