| *** spatel has quit IRC | 00:01 | |
| *** spatel has joined #openstack-ansible-sig | 00:22 | |
| *** spatel has quit IRC | 00:24 | |
| *** EmilienM|off is now known as EmilienM | 00:46 | |
| *** evrardjp has quit IRC | 00:52 | |
| *** gtema has joined #openstack-ansible-sig | 06:25 | |
| *** rpittau|afk is now known as rpittau | 06:51 | |
| *** spatel has joined #openstack-ansible-sig | 07:19 | |
| *** spatel has quit IRC | 07:23 | |
| *** evrardjp has joined #openstack-ansible-sig | 07:31 | |
| *** evrardjp has quit IRC | 07:32 | |
| *** evrardjp has joined #openstack-ansible-sig | 07:34 | |
| *** holser has joined #openstack-ansible-sig | 08:01 | |
| *** holser_ has joined #openstack-ansible-sig | 08:10 | |
| *** holser has quit IRC | 08:10 | |
| *** spatel has joined #openstack-ansible-sig | 09:39 | |
| *** sshnaidm|afk is now known as sshnaidm | 09:40 | |
| *** spatel has quit IRC | 09:44 | |
| *** rpittau is now known as rpittau|bbl | 10:31 | |
| *** gtema has quit IRC | 11:03 | |
| *** gtema has joined #openstack-ansible-sig | 11:27 | |
| *** rpittau|bbl is now known as rpittau | 12:04 | |
| *** gtema_ has joined #openstack-ansible-sig | 12:52 | |
| *** gtema has quit IRC | 12:53 | |
| *** holser_ has quit IRC | 13:14 | |
| *** holser has joined #openstack-ansible-sig | 13:14 | |
| openstackgerrit | Monty Taylor proposed openstack/ansible-collections-openstack master: Remove license_file entry in galaxy.yml https://review.opendev.org/729271 | 14:04 |
|---|---|---|
| mordred | sshnaidm: ^^ whack-a-mole | 14:19 |
| mordred | sshnaidm: I have uploaded current master to galaxy as another pre-release (to test that we're uploadable) | 14:20 |
| sshnaidm | mordred, ugh, right, I saw this error in ansible-test as well, forgot.. | 14:23 |
| mordred | sshnaidm: ansible-test gave us that error? | 14:23 |
| mordred | sshnaidm: why didn't the gate fail? | 14:23 |
| sshnaidm | mordred, when I pulled out galaxy.yml from .gitignore | 14:24 |
| mordred | OH | 14:24 |
| sshnaidm | because it's also ansible-test.ignore | 14:24 |
| mordred | right - because for reasons that surpass understanding ansible-test won't lint galaxy.yml if it's in .gitignore | 14:24 |
| mordred | sshnaidm: sigh. should we have our gate job remove .gitignore before running? | 14:25 |
| sshnaidm | I'll do later job with galaxy upload as a test, to local galaxy server, we'll be fine | 14:25 |
| mordred | sshnaidm: that sounds like a great idea | 14:26 |
| sshnaidm | till then I hope it won't be a lot of changes there.. | 14:26 |
| sshnaidm | mordred, it's weird, dev-125 is counted as earlier version than dev-85 | 14:28 |
| mordred | sshnaidm: they're using semver instead of python versioning | 14:33 |
| mordred | sshnaidm: which is a shame, since semver is shittastic | 14:33 |
| mordred | but - I guess I don't care as soon as 1.0 is uploaded | 14:33 |
| sshnaidm | yeah | 14:34 |
| sshnaidm | if galaxy.yml was wrong, how ansible-galaxy built it in job then.. need to check this | 14:35 |
| sshnaidm | mordred, I'm pretty sure there was nothing about this license error before, I have another collections that had the same duplicate key and were uploaded before | 14:44 |
| sshnaidm | and now it fails | 14:44 |
| openstackgerrit | Monty Taylor proposed openstack/ansible-collections-openstack master: Validate that we can upload build collection https://review.opendev.org/729294 | 14:50 |
| openstackgerrit | Monty Taylor proposed openstack/ansible-collections-openstack master: Remove license_file entry in galaxy.yml https://review.opendev.org/729271 | 14:52 |
| mordred | sshnaidm: ^^ rebased that - so we should be able to see the first fail and then the second succeed - at which point we can squash them | 14:53 |
| sshnaidm | mordred, ++ | 14:57 |
| openstackgerrit | Sagi Shnaidman proposed openstack/ansible-collections-openstack master: Redesign OpenstackModule class https://review.opendev.org/722283 | 14:58 |
| sshnaidm | odyssey4me, mordred dtantsur|afk gtema_ noonedeadpunk please take a look at review in your time ^^ | 15:01 |
| gtema_ | ok | 15:01 |
| mordred | sshnaidm: I have also submitted a PR to galaxy_importer to fix the return code thing | 15:02 |
| sshnaidm | mordred, cool, was thinking the same | 15:02 |
| sshnaidm | but I still will create local galaxy :D | 15:02 |
| sshnaidm | mordred, so, after these patches are in, are we ready to push 1.0.0? | 15:03 |
| mordred | sshnaidm: yah. I mean - once you've got that - then we should make it pause and return it as a zuul artifact (similar to docker registries) - so that subsequent jobs can configure to point to it and do galaxy install and just work | 15:04 |
| mordred | sshnaidm: yes - I discovered them because I was trying to push a 1.0.0 already | 15:04 |
| mordred | incidentally, I'm actually happy that we haven't automated uploads uet | 15:04 |
| mordred | because I would have been VERY grumpy if I'd pushed a tag and then the upload had failed | 15:04 |
| sshnaidm | yeah | 15:06 |
| sshnaidm | mordred, maybe to create a separate job for -e build? | 15:31 |
| sshnaidm | or include in linters | 15:32 |
| mordred | sshnaidm: it is included in the liters - we run the build tasks there | 15:38 |
| sshnaidm | Error while finding module specification for 'galaxy_importer.main' (ModuleNotFoundError: No module named 'galaxy_importer' | 15:41 |
| sshnaidm | mordred, need -e in script | 15:41 |
| mordred | sshnaidm: it's working for me | 15:42 |
| mordred | sshnaidm: did you rebuild your tox env? | 15:42 |
| sshnaidm | mordred, it's from job: https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_be5/729294/1/check/openstack-tox-linters-ansible-devel/be5631d/job-output.txt | 15:42 |
| sshnaidm | it passed, when it shouldn't | 15:43 |
| mordred | OH | 15:44 |
| openstackgerrit | Monty Taylor proposed openstack/ansible-collections-openstack master: Validate that we can upload build collection https://review.opendev.org/729294 | 15:44 |
| mordred | sshnaidm: I'm not sure why tox didn't consider that command failing a failure | 15:45 |
| mordred | sshnaidm: trying the linters env locally now | 15:46 |
| sshnaidm | mordred, no "-e" in script | 15:46 |
| sshnaidm | mordred, and no "ERROR", so it exists normal | 15:46 |
| mordred | oh - gotcha | 15:46 |
| mordred | makes sense | 15:46 |
| * mordred fixes | 15:46 | |
| openstackgerrit | Monty Taylor proposed openstack/ansible-collections-openstack master: Validate that we can upload build collection https://review.opendev.org/729294 | 15:48 |
| mordred | sshnaidm: that should do it | 15:48 |
| mordred | thanks! | 15:48 |
| openstackgerrit | Monty Taylor proposed openstack/ansible-collections-openstack master: Remove license_file entry in galaxy.yml https://review.opendev.org/729271 | 15:49 |
| mordred | sshnaidm: also - good to do this in 2 stages to make sure we fail properly it turns out :) | 15:49 |
| sshnaidm | totally :) | 15:50 |
| mordred | sshnaidm: https://review.opendev.org/#/c/729294/ failed - and the followup change passed | 16:12 |
| mordred | sshnaidm: shall I go ahead and squash? | 16:12 |
| sshnaidm | mordred, sure | 16:13 |
| openstackgerrit | Monty Taylor proposed openstack/ansible-collections-openstack master: Validate that we can upload build collection https://review.opendev.org/729294 | 16:13 |
| sshnaidm | ERROR: InvocationError for command /bin/bash tools/check-import.sh . (exited with code 1) | 16:13 |
| sshnaidm | we're good | 16:13 |
| mordred | \o/ | 16:13 |
| *** gtema_ has quit IRC | 16:14 | |
| *** gtema has joined #openstack-ansible-sig | 16:14 | |
| openstackgerrit | Sagi Shnaidman proposed openstack/ansible-collections-openstack master: Redesign OpenstackModule class https://review.opendev.org/722283 | 16:21 |
| *** rpittau is now known as rpittau|afk | 16:22 | |
| *** gtema_ has joined #openstack-ansible-sig | 16:40 | |
| *** gtema has quit IRC | 16:43 | |
| *** holser has quit IRC | 16:44 | |
| *** holser has joined #openstack-ansible-sig | 16:46 | |
| mordred | sshnaidm: the build_ignore is still not ignoring things | 16:47 |
| sshnaidm | mordred, do you use ansible-galaxy from devel? | 16:48 |
| sshnaidm | for me it works with latest versions | 16:48 |
| mordred | I am | 16:48 |
| mordred | and it worked a couple of days ago | 16:48 |
| mordred | I feel like it maybe broke again | 16:48 |
| sshnaidm | hmm | 16:49 |
| mordred | I can't get it to work at all | 16:49 |
| sshnaidm | mordred, do you use galaxy.yml from repo? | 16:49 |
| mordred | yeah - this is me running tox -ebuild | 16:51 |
| mordred | sshnaidm: NEVERMIND | 16:52 |
| mordred | sshnaidm: I had a virtualenv activated and it was making running tox weird | 16:52 |
| sshnaidm | k | 16:52 |
| openstackgerrit | Monty Taylor proposed openstack/ansible-collections-openstack master: Ignore a few more things https://review.opendev.org/729327 | 16:56 |
| mordred | sshnaidm: ^^ looked at the built tarball and saw a few minor things | 16:56 |
| sshnaidm | mordred, I think you can juse use "ci" instead of "ci*" | 16:57 |
| mordred | oh! you can | 16:58 |
| * sshnaidm is still wondering why wasn't it done as a whitelist ?? | 16:58 | |
| mordred | I think that was broken before | 16:58 |
| openstackgerrit | Monty Taylor proposed openstack/ansible-collections-openstack master: Ignore a few more things https://review.opendev.org/729327 | 16:59 |
| *** gtema_ has quit IRC | 17:00 | |
| *** evrardjp has quit IRC | 17:32 | |
| *** gtema has joined #openstack-ansible-sig | 17:33 | |
| *** evrardjp has joined #openstack-ansible-sig | 17:34 | |
| *** gtema has quit IRC | 17:41 | |
| mordred | sshnaidm: if you have a sec, both of those are green | 18:30 |
| mordred | sshnaidm: thanks! | 18:31 |
| mordred | sshnaidm: once they land, I'll upload 1.0 | 18:31 |
| mordred | sshnaidm: and then I'll see about getting us a galaxy upload job triggered by tags | 18:32 |
| sshnaidm | mordred, done | 18:32 |
| sshnaidm | mordred, great, thanks | 18:32 |
| openstackgerrit | Sagi Shnaidman proposed openstack/ansible-collections-openstack master: Redesign OpenstackModule class https://review.opendev.org/722283 | 18:55 |
| sshnaidm | mordred, don't we have something like "devstack in container"? Feeling wasted waiting every time for devstack to install.. | 18:55 |
| sshnaidm | just need some openstack cloud ready | 18:56 |
| mordred | sshnaidm: we do not | 19:00 |
| openstackgerrit | Sagi Shnaidman proposed openstack/ansible-collections-openstack master: Redesign OpenstackModule class https://review.opendev.org/722283 | 19:14 |
| openstackgerrit | Merged openstack/ansible-collections-openstack master: Validate that we can upload build collection https://review.opendev.org/729294 | 19:29 |
| openstackgerrit | Merged openstack/ansible-collections-openstack master: Ignore a few more things https://review.opendev.org/729327 | 19:29 |
| sshnaidm | mordred, we're good ^ | 19:33 |
| mordred | sshnaidm: \o/ | 19:36 |
| mordred | sshnaidm: https://galaxy.ansible.com/openstack/cloud | 19:46 |
| mordred | released | 19:46 |
| mordred | sshnaidm: and tag pushed | 19:47 |
| sshnaidm | mordred, +++++ | 19:47 |
| sshnaidm | \o/ \o/ | 19:48 |
| mordred | sshnaidm: we did a thing! | 19:48 |
| sshnaidm | YES | 19:48 |
| mordred | jrosser: hopefully the new collection doesn't completely break all your stuff | 19:48 |
| mordred | mnaser: ^^ | 19:48 |
| jrosser | mordred: I’ll take a look tomorrow at trying that out, also have a think about how to make some depends-on work | 19:49 |
| mordred | jrosser: yeah - that's on my todo list too | 19:50 |
| jrosser | I’m figuring “detect zuul CI and symlink” is probably first attempt | 19:51 |
| mordred | jrosser: sshnaidm was talking about setting up a job for us here that starts up a galaxy server - I'm thinking we could potentailly do something similar to how we're doing docker images, where we could have a buildset-galaxy that we publish stuff in to - and then have yoru jobs soft-depend on that so that if we've published stuff into the local galaxy you'll get it | 19:51 |
| mordred | jrosser: "detect zuul and symlink" should also work fine :) | 19:51 |
| jrosser | simple first :) | 19:52 |
| mordred | START WITH THE MOST COMPLEX ALWAYS!!! ;) | 19:52 |
| sshnaidm | you so underestimate symlinks :) | 19:53 |
| sshnaidm | they can bring a lot of fun | 19:53 |
| sshnaidm | mordred, maybe to setup galaxy server somewhere in infra for test? Like we have zuul registry | 19:54 |
| * sshnaidm is hiding | 19:55 | |
| mordred | sshnaidm: yeah - that's the sort of thing I think we're going to want if we want to _really_ do this completely solidly | 19:55 |
| mordred | sshnaidm: we should write down what it's going to look like - if docker is any indication, it's going to be "fun" :) | 19:55 |
| sshnaidm | docker is boring | 19:56 |
| sshnaidm | podman is a new fun | 19:56 |
| mordred | sshnaidm: well - speculative container registries that support both is even more fun :) | 20:00 |
| mordred | sshnaidm: turns out, sadly, there are sets of things each do that the other doesn't | 20:00 |
| sshnaidm | mordred, now we're talking! | 20:00 |
| mordred | sshnaidm: (see also multi-arch container images) | 20:00 |
| * mordred would like for buildah to grow support for building multi-arch container images ... but that is not a thing yet | 20:01 | |
| sshnaidm | mordred, should be in next podman version, btw | 20:01 |
| mordred | great | 20:01 |
| sshnaidm | asked podman folks just a few weeks ago | 20:01 |
| sshnaidm | yeah, of course fresh and untested, but.. | 20:02 |
| mordred | sshnaidm: we'll see whether it's released first or whether I figure out what's going on with docker buildx pulling layers for the wrong architecture first | 20:02 |
| sshnaidm | mordred, seriously? | 20:02 |
| mordred | yup | 20:02 |
| mordred | don't know why yet - still debugging | 20:02 |
| mordred | it's possible it has something to do with our step of copying images from the intermediate to the buildset registry with skopeo | 20:03 |
| mordred | although I think we're doing that correctly at this point | 20:03 |
| mordred | still diagnosing | 20:03 |
| mordred | sshnaidm: in any case- whenever there is something pre-release even that is potentially usable, I have a real-world use case that is complicated enough to be useful as a test | 20:04 |
| mordred | and where I will swtich to podman _immediately_ if it does the job better | 20:04 |
| * sshnaidm is going release container.podman collection today | 20:47 | |
| sshnaidm | so twice a big day! | 20:47 |
| mordred | \o/ | 22:14 |
| *** holser has quit IRC | 22:31 | |
| *** holser has joined #openstack-ansible-sig | 22:33 | |
| *** holser has quit IRC | 23:30 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!