| *** sshnaidm|afk has joined #zuul | 00:20 | |
| *** sshnaidm|afk is now known as sshnaidm|off | 00:20 | |
| *** imtiazc has joined #zuul | 00:22 | |
| *** openstackgerrit has joined #zuul | 00:27 | |
| openstackgerrit | Paul Belanger proposed zuul/zuul-jobs master: ensure-podman: Use official podman repos for ubuntu https://review.opendev.org/c/zuul/zuul-jobs/+/765177 | 00:28 |
|---|---|---|
| imtiazc | Hi, Is there any documentation or guidelines on how to deploy Zuul for production? The quick start guide works for a demo or development environment but for production, we need to have some redundancy. How is OpenStack or other Zuul users deploying Zuul? | 00:29 |
| openstackgerrit | Paul Belanger proposed zuul/zuul-jobs master: ensure-podman: Use official podman repos for ubuntu https://review.opendev.org/c/zuul/zuul-jobs/+/765177 | 00:30 |
| *** ikhan has quit IRC | 00:30 | |
| clarkb | imtiazc: https://zuul-ci.org/docs/zuul/howtos/zuul-from-scratch.html is the more in depth installation info | 00:31 |
| clarkb | imtiazc: openstack ( really opendev now) deploys zuul using ansible and docker+docker-compose. We run 8 zuul-mergers, 12 zuul-executors, 4 nodepool launchers, 2 nodepool builders. The zuul-scheduler, zuul-web, and zuul fingergw all run on the main scheduler host | 00:31 |
| clarkb | everything but the scheduler should scale out horizontally (though in some cases you need to shard the configuration, for example our nodepool launchers each talk to different cloud providers) | 00:32 |
| imtiazc | clarkb: Thanks so much! Are the Ansible playbooks for deploying Zuul available somewhere? | 00:39 |
| clarkb | imtiazc: https://opendev.org/opendev/system-config/src/branch/master/playbooks/roles they are all in there. I'm not sure we really recommend reusing them though. For reuse pabelanger's windmill roles maybe better? | 00:40 |
| clarkb | (we've tried in the past to make our config management more reconsumable but it is significant effort and few people seem to want to help make it happen so we've largely decided we're ok with stickingto what we need) | 00:40 |
| imtiazc | Fair enough. Thanks Again. | 00:44 |
| *** hamalq has quit IRC | 00:54 | |
| *** guilhermesp has quit IRC | 00:56 | |
| *** gouthamr has quit IRC | 00:57 | |
| *** mnasiadka has quit IRC | 00:57 | |
| *** gundalow has quit IRC | 00:57 | |
| *** guilhermesp has joined #zuul | 00:58 | |
| *** gouthamr has joined #zuul | 00:58 | |
| *** mnasiadka has joined #zuul | 00:58 | |
| *** gundalow has joined #zuul | 01:00 | |
| *** tosky has quit IRC | 01:01 | |
| *** rfolco has joined #zuul | 01:02 | |
| fungi | imtiazc: one other deployment option i'm aware of... zuul is now packaged as part of fedora 33, if you're looking for a packaged distribution of it | 01:05 |
| *** ikhan has joined #zuul | 01:16 | |
| *** rfolco has quit IRC | 01:18 | |
| *** ikhan has quit IRC | 01:21 | |
| *** openstackgerrit has quit IRC | 01:38 | |
| *** ikhan has joined #zuul | 02:19 | |
| *** Goneri has quit IRC | 02:19 | |
| *** bhavikdbavishi has joined #zuul | 02:53 | |
| *** zenkuro has joined #zuul | 02:56 | |
| *** wuchunyang has joined #zuul | 02:56 | |
| *** zenkuro has quit IRC | 03:05 | |
| *** zenkuro has joined #zuul | 03:06 | |
| *** bhavikdbavishi1 has joined #zuul | 03:12 | |
| *** bhavikdbavishi has quit IRC | 03:14 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 03:14 | |
| *** wuchunyang has quit IRC | 04:02 | |
| *** bhavikdbavishi has quit IRC | 04:18 | |
| *** bhavikdbavishi has joined #zuul | 04:19 | |
| *** zenkuro has quit IRC | 04:35 | |
| *** zenkuro has joined #zuul | 04:36 | |
| *** rishabhhpe has joined #zuul | 04:40 | |
| *** rishabhhpe has quit IRC | 05:02 | |
| *** rishabhhpe has joined #zuul | 05:05 | |
| *** rishabhhpe has quit IRC | 05:20 | |
| *** zenkuro has quit IRC | 05:20 | |
| *** zenkuro has joined #zuul | 05:21 | |
| *** rishabhhpe has joined #zuul | 05:21 | |
| *** ikhan has quit IRC | 05:29 | |
| *** evrardjp has quit IRC | 05:33 | |
| *** rishabhhpe has quit IRC | 05:33 | |
| *** evrardjp has joined #zuul | 05:33 | |
| *** ajitha has joined #zuul | 05:36 | |
| *** vishalmanchanda has joined #zuul | 05:38 | |
| *** bhavikdbavishi1 has joined #zuul | 06:11 | |
| *** bhavikdbavishi has quit IRC | 06:13 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 06:13 | |
| *** bhavikdbavishi has quit IRC | 06:47 | |
| *** bhavikdbavishi has joined #zuul | 07:27 | |
| *** bhavikdbavishi1 has joined #zuul | 07:29 | |
| *** bhavikdbavishi has quit IRC | 07:31 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 07:31 | |
| *** saneax has joined #zuul | 07:31 | |
| *** zenkuro has quit IRC | 07:39 | |
| *** zenkuro has joined #zuul | 07:40 | |
| *** jcapitao has joined #zuul | 07:57 | |
| *** nils has joined #zuul | 08:11 | |
| *** imtiazc has quit IRC | 08:18 | |
| *** nils has quit IRC | 08:50 | |
| *** jpena|off is now known as jpena | 08:57 | |
| *** tosky has joined #zuul | 09:05 | |
| *** msuszko10 has joined #zuul | 09:26 | |
| *** piotrowskim has joined #zuul | 09:37 | |
| *** bhavikdbavishi has quit IRC | 09:57 | |
| *** nils has joined #zuul | 10:05 | |
| *** zenkuro has quit IRC | 10:25 | |
| *** zenkuro has joined #zuul | 10:26 | |
| *** bhavikdbavishi has joined #zuul | 10:32 | |
| *** bhavikdbavishi1 has joined #zuul | 10:39 | |
| *** wuchunyang has joined #zuul | 10:42 | |
| *** bhavikdbavishi has quit IRC | 10:42 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 10:42 | |
| *** wuchunyang has quit IRC | 10:47 | |
| *** wuchunyang has joined #zuul | 10:50 | |
| *** hashar has joined #zuul | 10:51 | |
| *** wuchunyang has quit IRC | 10:55 | |
| *** wuchunyang has joined #zuul | 11:16 | |
| tobiash | zuul-client seems to not be configured in the gerrit bot | 11:30 |
| tobiash | remote: https://review.opendev.org/c/zuul/zuul-client/+/765516 Prevent override of bearer token by .netrc | 11:31 |
| tobiash | mhu, corvus: this would be needed so zuul-client works for us ^ | 11:32 |
| mhu | good catch, I wasn't aware of this behavior for requests | 11:34 |
| *** rfolco has joined #zuul | 11:36 | |
| tobiash | I neither :) | 11:38 |
| mhu | must have been fun to troubleshoot :) | 11:49 |
| tobiash | oh yes | 11:49 |
| tobiash | was like: enqueue - 401 - regenerate token - enqueue - 401 - wtf what's going on... | 11:50 |
| tobiash | avass: looking at https://review.opendev.org/c/zuul/zuul-jobs/+/764062 (windows checkout of zuul-jobs) I don't think it's necessary to do that with all fixtures since that complicates things quite a lot. What do you think about just special handling the problematic fixtures? | 11:50 |
| tobiash | there are just very few that violate windows filename patterns | 11:51 |
| tobiash | we have the same problem btw | 11:52 |
| *** jcapitao is now known as jcapitao_luncj | 11:57 | |
| *** jcapitao_luncj is now known as jcapitao_lunch | 11:57 | |
| avass | tobiash: I haven't look at it too much. I'll ping Daniel to see if he has any comments | 12:00 |
| avass | I don't really use windows so I'm not sure what the problem is. I guess it has something to do with unicode characters and in that case it could be limited I suppose | 12:01 |
| tobiash | some fixture files have ':' in its filename so a checkout complains and fails | 12:03 |
| avass | oh | 12:03 |
| avass | I'm a bit bogged down debugging why we're getting a tls handshake failure when pulling the artifactory docker image. but I'll see if I can get hold of him | 12:05 |
| *** ikhan has joined #zuul | 12:20 | |
| *** jpena is now known as jpena|lunch | 12:32 | |
| *** weshay|pto is now known as weshay|ruck | 12:49 | |
| *** jcapitao_lunch is now known as jcapitao | 12:58 | |
| *** rlandy has joined #zuul | 13:00 | |
| *** ikhan has quit IRC | 13:15 | |
| *** bhavikdbavishi has quit IRC | 13:26 | |
| *** jpena|lunch is now known as jpena | 13:29 | |
| tobiash | avass: in the past I got tls handshake failures with misconfigured mtu maybe you want to check that | 13:46 |
| *** msuszko10 has quit IRC | 13:48 | |
| *** saneax has quit IRC | 13:57 | |
| *** hashar has quit IRC | 14:08 | |
| pabelanger | Hmm, so I am trying to update ensure-podman to use a newer version of podman, but hitting an issue. It seems that .zuul-siblings is breaking for COPY command: https://zuul.opendev.org/t/zuul/build/961edebc009047518bb1f87932ac05f7/console#2/5/10/builder | 14:11 |
| pabelanger | can't figure it out | 14:11 |
| pabelanger | no items matching glob makes me think something in newer podman is ignoring hidden directories? | 14:12 |
| *** wuchunyang has quit IRC | 14:14 | |
| *** wuchunyang has joined #zuul | 14:15 | |
| *** wuchunyang has quit IRC | 14:20 | |
| tobiash | at least it says 1 filtered out | 14:32 |
| pabelanger | right, trying to figure out what that means | 14:35 |
| *** bhavikdbavishi has joined #zuul | 14:37 | |
| pabelanger | I'm guessing this is a bug in buildah: https://github.com/containers/buildah/commit/9c6969a79e1ba211e6f2e9146d68451ff5a63a07 | 14:46 |
| *** openstackgerrit has joined #zuul | 14:46 | |
| openstackgerrit | Paul Belanger proposed zuul/zuul-jobs master: ensure-podman: Use official podman repos for ubuntu https://review.opendev.org/c/zuul/zuul-jobs/+/765177 | 14:47 |
| pabelanger | if reading right, we might need to update .dockerignore | 14:47 |
| *** Goneri has joined #zuul | 14:54 | |
| *** bhavikdbavishi1 has joined #zuul | 15:01 | |
| *** bhavikdbavishi has quit IRC | 15:02 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 15:02 | |
| avass | corvus: how come the zuul-helm charts are using secrets for almost everything? I'd expect that to be configmaps | 15:03 |
| avass | anyway I'm preparing a code dump for the charts so anyone can review it if there's any changes | 15:09 |
| openstackgerrit | Tobias Henkel proposed zuul/zuul-client master: Prevent override of bearer token by .netrc https://review.opendev.org/c/zuul/zuul-client/+/765516 | 15:32 |
| openstackgerrit | Tobias Henkel proposed zuul/zuul-client master: Bump mypy to 0.790 https://review.opendev.org/c/zuul/zuul-client/+/765553 | 15:33 |
| zbr | tobiash: is zuul-client still an admin specific tool or it received features that could be used by normal zuul users? | 15:37 |
| mhu | zbr, it's meant for everybody | 15:38 |
| tobiash | zbr: zuul-client is targeted for end users | 15:38 |
| mhu | but you need a token for admin tasks | 15:38 |
| corvus | zbr: you're thinking of the "zuul" client command in the zuul repo; it's admin only. zuul-client is a new tool for all. | 15:38 |
| corvus | zbr: the "zuul" client command may be retired when we're happy zuul-client can replace it. | 15:39 |
| zbr | that was my impression too but going a google search by zuul-client got me to... https://zuul-ci.org/docs/zuul/reference/client.html | 15:39 |
| openstackgerrit | Albin Vass proposed zuul/zuul-helm master: Adds zookeeper chart to handle tls setup https://review.opendev.org/c/zuul/zuul-helm/+/765554 | 15:39 |
| openstackgerrit | Albin Vass proposed zuul/zuul-helm master: Updates scheduler config https://review.opendev.org/c/zuul/zuul-helm/+/765555 | 15:39 |
| corvus | google will catch up with https://zuul-ci.org/docs/zuul-client/ eventually | 15:40 |
| corvus | avass: i don't know the answer to your question off the top of my head; mnaser may | 15:40 |
| tobiash | corvus: we could help that maybe by renaming that page? | 15:40 |
| avass | corvus: alright | 15:40 |
| tobiash | to something like 'Zuul Admin Client' | 15:40 |
| zbr | And mentioning the zuul-client with link on first paragraph. I can make CR, ok? | 15:41 |
| corvus | tobiash: yes, or just remove it :) | 15:41 |
| mnaser | corvus, avass: the reason is because there is secrets inside those config files so from an rbac pov, you would put those as secrets so they can only be read by someone who can read secrets | 15:41 |
| corvus | zbr: ++ | 15:41 |
| tobiash | I'm not yet confident enough that I'd remove it now :) | 15:41 |
| avass | mnaser: oh but we use environment variables for the secretss | 15:41 |
| corvus | tobiash: oh yeah, sorry, i just meant leave it alone until it's removed in the not distant future | 15:42 |
| avass | since zuul supports environment interpolation | 15:42 |
| tobiash | corvus: ok, that works for me as well :) | 15:42 |
| mnaser | avass: right, but then it gets a little tricky for things like github or private keys | 15:42 |
| mnaser | but yes, actually very valid idea | 15:42 |
| avass | mnaser: those would be mounted as secrets too | 15:42 |
| avass | which is how we do it | 15:42 |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Bump mypy to 0.790 https://review.opendev.org/c/zuul/zuul/+/765557 | 15:43 |
| mnaser | avass: sure -- i was aiming for simplicity initially :) but i am not against it if we properly interpolate all that stuff out without adding complexity | 15:43 |
| avass | :) | 15:44 |
| avass | I'll try to pick out the parts that fit well enough from our setup and let you decide | 15:44 |
| tobiash | is it expected that pushing to gerrit got sometimes really slow compared with before the update? | 15:44 |
| tobiash | pushing 765557 took 42s | 15:44 |
| corvus | infra-root: ^ | 15:44 |
| corvus | tobiash: my understanding is that we probably haven't tracked down all the old zuul's not using "change:" so there may still be load issues from that | 15:45 |
| avass | separating the changes a bit for now to make it easier to review, but some parts probably makes sense together | 15:45 |
| corvus | tobiash: also, in general i do think it's expected to be a bit slower even in the best conditions; not that much slower, but we're starting from a higher baseline | 15:46 |
| tobiash | this indicates that processing on the server takes quite long: http://paste.openstack.org/show/800739/ | 15:46 |
| openstackgerrit | Jeremy Stanley proposed zuul/zuul-client master: Use general docs URL for link to zuul + z-c docs https://review.opendev.org/c/zuul/zuul-client/+/765558 | 15:46 |
| fungi | heh, it's probably not helping that https://pypi.org/project/zuul-client/ says the docs for it are https://zuul-ci.org/docs/zuul/ | 15:46 |
| tobiash | oh so the old queries for changes are expensive searches now? | 15:46 |
| corvus | fungi: ++ | 15:46 |
| tobiash | fungi: yeah just noticed today as well | 15:47 |
| tobiash | is that part of the package? | 15:47 |
| corvus | tobiash: zuul v2 and some versions of v3 use an inefficient query | 15:47 |
| tobiash | ah ok | 15:47 |
| fungi | tobiash: the readme gets reused as the package long description | 15:47 |
| fungi | if i get another minute later, i'll add package metadata with explicit urls to docs and such, for richer pypi listings | 15:48 |
| openstackgerrit | Tobias Henkel proposed zuul/zuul-client master: Fix url to docs in readme https://review.opendev.org/c/zuul/zuul-client/+/765559 | 15:49 |
| corvus | tobiash: qq on 765516 | 15:49 |
| tobiash | corvus: we have on the zuul hostname a separate basic authed endpoint for accessing zuul api for scripted access (our default api is behind saml and not scriptable). Some of our users configure their auth using a netrc for that. | 15:51 |
| tobiash | so that curl or wget works with log downloads | 15:52 |
| *** odyssey4me has quit IRC | 15:52 | |
| tobiash | I use that as well and with that the netrc matches and auth header gets overridden without this | 15:53 |
| corvus | tobiash: makes sense, thanks! | 15:53 |
| *** odyssey4me has joined #zuul | 15:53 | |
| corvus | tobiash: and https://github.com/psf/requests/issues/3929#issuecomment-386114903 is why you didn't do trust_env? | 15:56 |
| tobiash | yes, trust_env disables also proxy handling which would be a non-acceptible side effect | 15:57 |
| tobiash | (and ca bundle) | 15:57 |
| openstackgerrit | Sorin Sbârnea proposed zuul/zuul master: Deprecates old admin client https://review.opendev.org/c/zuul/zuul/+/765564 | 16:12 |
| tobiash | fungi: I've overlooked your readme fix, +w on yours and abandoned mine :) | 16:15 |
| fungi | thanks! | 16:15 |
| fungi | i'm hacking together an overhaul of the python package metadata for it now to demo, then i can do the same to our other projects once folks are happy with it | 16:15 |
| zbr | fungi: that is interesting, i was about to propose replacing pbr with setuptools-scm for zuul project. | 16:18 |
| zbr | I did had a lot of experience using the newer tools, and my impression is that pbr is no longer needed, especially for projects requiring modern pythons. | 16:19 |
| *** bhavikdbavishi has quit IRC | 16:19 | |
| * zbr but if others want to stick to pbr, i will shut up. | 16:20 | |
| tristanC | corvus: could i approve https://review.opendev.org/c/zuul/zuul-operator/+/708860 to fix promotion job? | 16:20 |
| corvus | tristanC: i'll review | 16:21 |
| corvus | tristanC: small thing on that | 16:24 |
| corvus | left comment | 16:24 |
| clarkb | tobiash: corvus: we're still also looking at some tunables. It isn't clear if the jgit tunables must go in jgit.config now or if they can go in gerrit.config too. There are contradictory statements in the documentation which I've askled questions about on gerrit slack | 16:30 |
| clarkb | if gerrit.config is ok then our tunables should already be set appropriately (or at least more so than the defaults). If they have to go into jgit.config then we need to make some changes | 16:31 |
| clarkb | but I've found a number of documentation errors post upgrade and am super confused about this particular area. Hoping the gerrit devs can clarify | 16:31 |
| dmsimard | I miss the table in gerrit where all the latest Zuul job results went -- do we always have to expand comments to find them now ? | 16:33 |
| dmsimard | sorry if this is obvious, I haven't used the new gerrit UI a lot | 16:33 |
| clarkb | dmsimard: yes, the old hidecijs hack doesn't work anymore due to the polygerrit ui rewrite. Work has started on writing a proper gerrit plugin to get that back again | 16:33 |
| zbr | dmsimard: yes, or read the mailing list for a temporary tampermonkey script. | 16:34 |
| tobiash | that would at least make sense, we've also seen very large upload pack times when gc was not configured due to a mistake (but on old gerrit) | 16:34 |
| dmsimard | clarkb, zbr: thanks, ++ | 16:34 |
| corvus | zbr: http://lists.opendev.org/pipermail/service-discuss/2020-December/000138.html | 16:34 |
| corvus | dmsimard: http://lists.opendev.org/pipermail/service-discuss/2020-December/000138.html | 16:34 |
| corvus | zbr: sorry was meant for zbr | 16:34 |
| corvus | zbr: sorry was meant for dmsimard :/ | 16:34 |
| corvus | dmsimard: that should catch you up :) | 16:35 |
| dmsimard | corvus: you're forgiven :p | 16:35 |
| openstackgerrit | Jeremy Stanley proposed zuul/zuul-client master: Overhaul Python package metadata https://review.opendev.org/c/zuul/zuul-client/+/765567 | 16:35 |
| corvus | fungi: palindrome | 16:36 |
| clarkb | as a side note looking at cacti data yesterday I'm reasonably confident that the disk io is not the bottle neck | 16:36 |
| clarkb | when we run backups and our periodic git gc in the background we see iops and bw to the disk jump significnatly compared to when gerrit is slower | 16:36 |
| clarkb | (mention that because people constantly ask " are you using ssds?") | 16:37 |
| zbr | avass: corvus: https://review.opendev.org/c/zuul/zuul/+/764471 should address some weird page reflows when scrollbar appears, which can happen with our expand/collapse. | 16:37 |
| clarkb | I suspect its largely a game of tuning at this point and we need to decipher what it is the docs are trying to tell us with help from upsteram (which we are currently trying to do) | 16:37 |
| corvus | fungi: qq on 765567 | 16:38 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-operator master: Add zuul-operator-upload-image job https://review.opendev.org/c/zuul/zuul-operator/+/708860 | 16:38 |
| *** bhavikdbavishi has joined #zuul | 16:40 | |
| openstackgerrit | Merged zuul/zuul-client master: Use general docs URL for link to zuul + z-c docs https://review.opendev.org/c/zuul/zuul-client/+/765558 | 16:41 |
| fungi | corvus: answered, i could go either way (and i did initially consider it) | 16:42 |
| *** hashar has joined #zuul | 16:50 | |
| openstackgerrit | Sorin Sbârnea proposed zuul/zuul-client master: Validate project metadata https://review.opendev.org/c/zuul/zuul-client/+/765570 | 16:53 |
| corvus | zbr: did you post your second comment on https://review.opendev.org/c/zuul/zuul/+/764471 to the wrong change? | 17:07 |
| zbr | corvus: i know :( | 17:08 |
| corvus | zbr: i'm guessing you meant that for fungi's zuul-client change 765567 ? | 17:08 |
| zbr | corvus: yes, in fact I already raised https://review.opendev.org/c/zuul/zuul-client/+/765570 to do it. | 17:10 |
| fungi | zbr: twine check doesn't check much. the real way to test it is to upload the package to test.pypi.org | 17:11 |
| zbr | if we switch to pep-517 i can even to more than this as there are extra tools available. | 17:11 |
| zbr | it does something, and yep, on my projects I upload any merge to test.pypi.org but don't tell this to pypa team ;) | 17:11 |
| fungi | there are a variety of things warehouse will reject on upload that twine check doesn't catch, and also twine check won't tell you how the changed metadata is going to get displayed by warehouse | 17:12 |
| fungi | pypa folks don't mind if you upload constantly to test.pypi.org, but they also warn that they might take it offline for extended periods or clear out the contents when needed | 17:12 |
| fungi | so just need to keep that in mind | 17:13 |
| zbr | i do not mind if it goes offline, or they start to remove packages, i never seen a downtime on test instance, but i seen on production :D | 17:14 |
| zbr | if curious about extended testing of metadata, check https://github.com/ansible-community/molecule/blob/master/tox.ini#L145-L153 | 17:14 |
| zbr | check-manifest does a better job, but it needs pep-517, so it will not work with pbr. | 17:14 |
| zbr | there is a long pending feature request to improve check, in fact they are actively working to move the code there from where it is now. | 17:15 |
| zbr | so "twine check" seams like a good future bet. afaik, it will only get better. | 17:16 |
| corvus | this stuff doesn't change very often | 17:16 |
| corvus | it's probably not worth investing a lot of time/effort to do something like test uploads; simple local twine checks seem harmless even if they aren't completely effective, but still, not exactly a priority | 17:17 |
| corvus | i'd be okay cutting a release of zuul-client with the new data, fixing it if it bombs, repeat as needed, then copy to the other repos | 17:18 |
| corvus | and then we're good until the next metadata overhaul in 8 years :) | 17:18 |
| *** rlandy is now known as rlandy|brb | 17:20 | |
| openstackgerrit | Merged zuul/zuul-operator master: Add zuul-operator-upload-image job https://review.opendev.org/c/zuul/zuul-operator/+/708860 | 17:22 |
| *** imtiazc has joined #zuul | 17:24 | |
| tristanC | corvus: thanks, the zuul-operator image is now public. I'll write an all-in-one resources to set it up in one kubectl apply command | 17:25 |
| corvus | tristanC: groovy :) | 17:30 |
| *** rlandy|brb is now known as rlandy | 17:47 | |
| *** vishalmanchanda has quit IRC | 17:58 | |
| *** jpena is now known as jpena|off | 17:59 | |
| fungi | corvus: so here's an upload test with the license metadata made more specific: https://test.pypi.org/project/zuul-client/0.0.4.dev3/ | 18:16 |
| fungi | technically asf calls it the "apache license, version 2.0" so i tested with that set in the license metadata | 18:16 |
| fungi | looks like it puts the text of the trove classifier first in the license line for the left sidebar but puts the license metadata field in parentheses after it | 18:17 |
| *** wuchunyang has joined #zuul | 18:20 | |
| *** wuchunyang has quit IRC | 18:25 | |
| dmsimard | finally caught up with the ML thread about zuul status table in gerrit (thanks for the link) and yeah https://104.130.172.52/c/openstack/diskimage-builder/+/554002 is what I meant :) | 18:26 |
| openstackgerrit | Jeremy Stanley proposed zuul/zuul-client master: Overhaul Python package metadata https://review.opendev.org/c/zuul/zuul-client/+/765567 | 18:29 |
| *** hamalq has joined #zuul | 18:36 | |
| *** nils has quit IRC | 18:36 | |
| *** zenkuro has quit IRC | 18:46 | |
| *** bhavikdbavishi has quit IRC | 18:48 | |
| *** hashar has quit IRC | 18:57 | |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Zuul Cache role https://review.opendev.org/c/zuul/zuul-jobs/+/764808 | 19:11 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Zuul Cache role https://review.opendev.org/c/zuul/zuul-jobs/+/764808 | 19:12 |
| *** jcapitao has quit IRC | 19:14 | |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Zuul Cache role https://review.opendev.org/c/zuul/zuul-jobs/+/764808 | 19:27 |
| avass | corvus: maybe it's possible to set up tests for the gcs roles with: https://github.com/fsouza/fake-gcs-server | 19:37 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Zuul Cache role https://review.opendev.org/c/zuul/zuul-jobs/+/764808 | 19:55 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Zuul Cache role https://review.opendev.org/c/zuul/zuul-jobs/+/764808 | 20:11 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Zuul Cache role https://review.opendev.org/c/zuul/zuul-jobs/+/764808 | 20:21 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Zuul Cache role https://review.opendev.org/c/zuul/zuul-jobs/+/764808 | 20:32 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Zuul Cache role https://review.opendev.org/c/zuul/zuul-jobs/+/764808 | 20:44 |
| *** rlandy has quit IRC | 21:04 | |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Zuul Cache role https://review.opendev.org/c/zuul/zuul-jobs/+/764808 | 21:06 |
| *** rfolco has quit IRC | 21:07 | |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Zuul Cache role https://review.opendev.org/c/zuul/zuul-jobs/+/764808 | 21:13 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Zuul Cache role https://review.opendev.org/c/zuul/zuul-jobs/+/764808 | 21:24 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Zuul Cache role https://review.opendev.org/c/zuul/zuul-jobs/+/764808 | 21:36 |
| *** ajitha has quit IRC | 21:45 | |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Zuul Cache role https://review.opendev.org/c/zuul/zuul-jobs/+/764808 | 21:47 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Zuul Cache role https://review.opendev.org/c/zuul/zuul-jobs/+/764808 | 21:56 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Zuul Cache role https://review.opendev.org/c/zuul/zuul-jobs/+/764808 | 22:07 |
| *** wuchunyang has joined #zuul | 22:22 | |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Zuul Cache role https://review.opendev.org/c/zuul/zuul-jobs/+/764808 | 22:26 |
| *** wuchunyang has quit IRC | 22:26 | |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Zuul Cache role https://review.opendev.org/c/zuul/zuul-jobs/+/764808 | 22:29 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Zuul Cache role https://review.opendev.org/c/zuul/zuul-jobs/+/764808 | 22:34 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Zuul Cache role https://review.opendev.org/c/zuul/zuul-jobs/+/764808 | 22:43 |
| SpamapS | avass: I did something similar for the AWS tests https://opendev.org/zuul/nodepool/src/branch/master/nodepool/tests/unit/test_driver_aws.py#L25 | 22:51 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-operator master: Add allinone deployment and documentation https://review.opendev.org/c/zuul/zuul-operator/+/765630 | 22:51 |
| tristanC | corvus: with ^ `kubectl apply -f https://opendev.org/zuul/zuul-operator/raw/branch/master/deploy/allinone.yaml` should deploy a zuul | 22:52 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Zuul Cache role https://review.opendev.org/c/zuul/zuul-jobs/+/764808 | 22:55 |
| tristanC | and by should, i mean when docker doesn't rate limit you. the same resources are also used in the integration tests | 22:55 |
| avass | SpamapS: yeah which is great | 22:55 |
| avass | would it be possible for opendev to deploy it's own docker registry instead? | 22:57 |
| avass | *container registry ;) | 22:57 |
| *** rfolco has joined #zuul | 23:01 | |
| fungi | we can easily serve container images via apache mod_autoindex... is there more to it than that? | 23:02 |
| avass | a bit :) | 23:02 |
| fungi | unfortunate. that's usually enough to serve packages for most any other distribution | 23:03 |
| avass | fungi: if you like a good read: https://github.com/opencontainers/distribution-spec/blob/master/spec.md | 23:04 |
| corvus | we do already run one registry; pretty easy to run another. :) | 23:04 |
| fungi | yeah, we can add more zuul-registry servers if that'll solve a problem | 23:05 |
| avass | yeah I don't think there would be too much work | 23:05 |
| *** rfolco has quit IRC | 23:06 | |
| avass | but I'm not sure if the rate limit is a real problem either | 23:06 |
| corvus | the upload roles are already set up to scope access by project. so a shared opendev registry would be simple. | 23:06 |
| corvus | avass: agreed | 23:07 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Zuul Cache role https://review.opendev.org/c/zuul/zuul-jobs/+/764808 | 23:10 |
| avass | has anyone experimented with running zuul with pypy? | 23:25 |
| fungi | i have not, you'll presumably want the very recent pypy with python 3.x support | 23:26 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-operator master: Add allinone deployment and documentation https://review.opendev.org/c/zuul/zuul-operator/+/765630 | 23:26 |
| avass | fungi: pamac install pypy3 got me support for 3.7.9 :) | 23:27 |
| avass | but it doesn't like fb-re2 | 23:27 |
| fungi | mmm | 23:28 |
| fungi | i wonder if there's another re2 implementation which works with pypy3? | 23:28 |
| avass | I would guess most re2 bindings for python uses the cpython c api which pypy doesn't like | 23:29 |
| *** Goneri has quit IRC | 23:50 | |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: WIP: Zuul Cache role https://review.opendev.org/c/zuul/zuul-jobs/+/764808 | 23:50 |
| *** tosky has quit IRC | 23:54 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!