| *** Goneri has quit IRC | 00:28 | |
| *** rlandy has quit IRC | 00:42 | |
| *** tosky has quit IRC | 00:54 | |
| *** openstackgerrit has quit IRC | 00:58 | |
| *** rfolco has joined #zuul | 01:10 | |
| *** msuszko has quit IRC | 01:10 | |
| *** rfolco has quit IRC | 01:28 | |
| *** msuszko has joined #zuul | 01:43 | |
| *** openstackgerrit has joined #zuul | 01:54 | |
| openstackgerrit | Ian Wienand proposed zuul/zuul master: doc: update example setup playbook https://review.opendev.org/c/zuul/zuul/+/765899 | 01:54 |
|---|---|---|
| *** zer0c00l has quit IRC | 02:08 | |
| *** paladox has quit IRC | 02:10 | |
| *** paladox has joined #zuul | 02:15 | |
| *** saneax has joined #zuul | 02:57 | |
| *** bhavikdbavishi has joined #zuul | 03:04 | |
| *** bhavikdbavishi1 has joined #zuul | 03:09 | |
| *** bhavikdbavishi has quit IRC | 03:10 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 03:10 | |
| *** saneax has quit IRC | 03:17 | |
| *** hamalq_ has quit IRC | 03:45 | |
| *** ikhan has joined #zuul | 04:16 | |
| *** ikhan has quit IRC | 04:20 | |
| *** bhavikdbavishi has quit IRC | 04:26 | |
| *** bhavikdbavishi has joined #zuul | 04:26 | |
| *** zenkuro has joined #zuul | 05:24 | |
| *** evrardjp has quit IRC | 05:33 | |
| *** evrardjp has joined #zuul | 05:33 | |
| *** bhavikdbavishi1 has joined #zuul | 06:16 | |
| *** ikhan has joined #zuul | 06:16 | |
| *** bhavikdbavishi has quit IRC | 06:17 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 06:17 | |
| *** ikhan has quit IRC | 06:21 | |
| *** saneax has joined #zuul | 06:31 | |
| *** zenkuro has quit IRC | 06:32 | |
| *** zenkuro has joined #zuul | 06:33 | |
| *** hamalq has joined #zuul | 06:44 | |
| *** bhavikdbavishi has quit IRC | 06:56 | |
| *** bhavikdbavishi has joined #zuul | 06:57 | |
| *** mach1na has joined #zuul | 06:59 | |
| *** bhavikdbavishi has quit IRC | 07:07 | |
| *** mach1na has quit IRC | 07:07 | |
| *** vishalmanchanda has joined #zuul | 07:19 | |
| *** mach1na has joined #zuul | 07:24 | |
| *** bhavikdbavishi has joined #zuul | 07:33 | |
| *** bhavikdbavishi1 has joined #zuul | 07:36 | |
| *** bhavikdbavishi has quit IRC | 07:38 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 07:38 | |
| *** jcapitao has joined #zuul | 08:00 | |
| *** hashar has joined #zuul | 08:07 | |
| *** rpittau|afk is now known as rpittau | 08:10 | |
| *** ikhan has joined #zuul | 08:16 | |
| *** ikhan has quit IRC | 08:20 | |
| *** rfolco has joined #zuul | 09:03 | |
| *** rfolco has quit IRC | 09:07 | |
| *** hamalq has quit IRC | 09:10 | |
| *** hamalq has joined #zuul | 09:14 | |
| *** hamalq_ has joined #zuul | 09:16 | |
| *** ikhan has joined #zuul | 09:16 | |
| *** hamalq has quit IRC | 09:18 | |
| *** hamalq_ has quit IRC | 09:20 | |
| *** ikhan has quit IRC | 09:20 | |
| *** rfolco has joined #zuul | 09:51 | |
| *** hamalq has joined #zuul | 10:00 | |
| *** hamalq has quit IRC | 10:04 | |
| *** bhavikdbavishi has quit IRC | 10:04 | |
| lyr | Hi there | 10:12 |
| lyr | ANyone can help me with https://paste.garrigue.re/?39cf6890b9e477a3#4sKq4PZdSSbCq77ejgBi9PZCg7An9S8GxcpLYD3A6p2X ? I don't get the issue | 10:13 |
| tobiash | lyr: looks like you're missing the debootstrap dependency | 10:18 |
| tobiash | lyr: but I think the people in #openstack-dib might be able to help better if that's not the problem | 10:19 |
| lyr | oh good, I tried joining a hypothetical #nodepool without success | 10:22 |
| lyr | tobiash: think is, I don't get why debootstrap isn't part of debian-minimal or alike elements | 10:23 |
| tobiash | lyr: it's needed on the host | 10:24 |
| lyr | hum | 10:24 |
| lyr | it's a centos 7 host | 10:24 |
| *** mach1na has quit IRC | 10:27 | |
| *** bhavikdbavishi has joined #zuul | 10:30 | |
| *** bhavikdbavishi1 has joined #zuul | 10:33 | |
| *** mach1na has joined #zuul | 10:33 | |
| *** bhavikdbavishi has quit IRC | 10:34 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 10:34 | |
| tobiash | looks like debootstrap is missing in bindep for centos/fedora: https://opendev.org/openstack/diskimage-builder/src/branch/master/bindep.txt#L30 | 10:36 |
| tobiash | ianw: is this on purpose? ^ I guess it's missing because that's part of epel and only needed to build debian based images? | 10:36 |
| *** nils has joined #zuul | 10:39 | |
| lyr | that's a problem for software factory users, sf-config (repository) is incompatible with epel-release | 10:41 |
| ianw | tobiash: yeah, i don't think it's on purpose | 10:55 |
| ianw | not sure if we test building debian/ubuntu images on centos hosts anywhere | 10:55 |
| lyr | I installed it somehow, and now I'm stuck with a "Authorization failed. The request you have made requires authentication" on a "POST http://openstack:5000/v3/auth/tokens" | 11:10 |
| lyr | (as reported by openstack logs) | 11:11 |
| *** bhavikdbavishi has quit IRC | 11:11 | |
| lyr | On nodepool side, it's | 11:12 |
| lyr | ERROR nodepool.CleanupWorker: Failure during resource cleanup for provider openstack | 11:12 |
| *** bhavikdbavishi has joined #zuul | 11:18 | |
| *** bhavikdbavishi has quit IRC | 11:22 | |
| *** bhavikdbavishi has joined #zuul | 11:22 | |
| *** hashar is now known as hasharLunch | 11:26 | |
| *** bhavikdbavishi has quit IRC | 11:32 | |
| *** bhavikdbavishi has joined #zuul | 11:33 | |
| *** wuchunyang has joined #zuul | 11:44 | |
| *** jcapitao is now known as jcapitao_lunch | 11:45 | |
| tobiash | clarkb: ping me if you need help with the openshift dep update, this issue now also blocks our nodepool updates | 11:50 |
| tobiash | lyr: sounds like you're missing clouds.yaml or secure.yaml | 11:50 |
| *** hamalq has joined #zuul | 12:01 | |
| *** wuchunyang has quit IRC | 12:02 | |
| *** hamalq has quit IRC | 12:05 | |
| *** mach1na has quit IRC | 12:16 | |
| *** ikhan has joined #zuul | 12:16 | |
| *** ikhan has quit IRC | 12:21 | |
| tobiash | corvus: judging from the responses on the ml I guess we'll be settling on [database]? | 12:28 |
| *** frenzyfriday has joined #zuul | 12:37 | |
| *** hasharLunch is now known as hashar | 12:37 | |
| openstackgerrit | Oleksandr Kovalchuk proposed zuul/zuul master: Fix errors regarding parsing timestamps https://review.opendev.org/c/zuul/zuul/+/765990 | 12:37 |
| *** frenzyfriday has quit IRC | 12:38 | |
| *** jfoufas1 has joined #zuul | 12:58 | |
| *** mach1na has joined #zuul | 13:02 | |
| *** rlandy has joined #zuul | 13:03 | |
| *** jcapitao_lunch is now known as jcapitao | 13:04 | |
| *** ikhan has joined #zuul | 13:04 | |
| *** ikhan has quit IRC | 13:04 | |
| *** piotrowskim has joined #zuul | 13:11 | |
| mhu | Hello zuul-maint, zuul-client doesn't build on Fedora 34 because python-requests is pinned to an obsolete version in zuul-client ver. 0.0.2 | 13:12 |
| zbr | mhu: feel free to make a CR! cc me. | 13:12 |
| mhu | While I could patch this in the spec, it'd be nicer to ship a new release upstream since requests is unpinned | 13:13 |
| mhu | and we've added the encrypt subcommand since | 13:13 |
| mhu | zuul-maint, could we land these patches too before the 0.0.3 release? https://review.opendev.org/q/project:zuul/zuul-client+status:open+(label:Code-Review%252B1+OR+label:Code-Review%252B2) | 13:13 |
| zbr | imho, we need to be sure the client works fine with the full range of python versions. | 13:14 |
| mhu | zbr, for what it's worth the spec builds fine on Fedora 33 | 13:14 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-operator master: Add zuul-ensure-own-config role https://review.opendev.org/c/zuul/zuul-operator/+/765882 | 13:14 |
| mhu | and the unit tests ensure zuul-client works for py3.6 and py3.8 | 13:15 |
| mhu | and worst case scenario there's a zuul-client container that should work *just* fine | 13:16 |
| *** ikhan has joined #zuul | 13:19 | |
| *** rfolco has quit IRC | 13:22 | |
| *** rfolco has joined #zuul | 13:22 | |
| *** bhavikdbavishi has quit IRC | 13:34 | |
| *** tosky has joined #zuul | 13:44 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-operator master: Add zuul-ensure-own-config role https://review.opendev.org/c/zuul/zuul-operator/+/765882 | 13:54 |
| mhu | oh, I didn't even realize there was a zuul-client 0.0.3 already | 14:01 |
| *** hamalq has joined #zuul | 14:02 | |
| openstackgerrit | Matthieu Huin proposed zuul/zuul-client master: Add more installation instructions https://review.opendev.org/c/zuul/zuul-client/+/765999 | 14:04 |
| *** hamalq has quit IRC | 14:07 | |
| *** bhavikdbavishi has joined #zuul | 14:12 | |
| *** bhavikdbavishi1 has joined #zuul | 14:14 | |
| *** bhavikdbavishi has quit IRC | 14:16 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 14:16 | |
| mhu | fungi, corvus there was a bug in zuul-client encrypt in 0.0.3 that is fixed in https://opendev.org/zuul/zuul-client/commit/b371dc2f40b1cd1b51d266ad7a4eefedfa68414e would it be possible to release 0.0.3.1 to include the fix? | 14:17 |
| *** hamalq has joined #zuul | 14:17 | |
| *** hamalq has quit IRC | 14:22 | |
| *** Goneri has joined #zuul | 14:39 | |
| openstackgerrit | Merged zuul/zuul-client master: Overhaul Python package metadata https://review.opendev.org/c/zuul/zuul-client/+/765567 | 14:49 |
| guillaumec | clarkb, fungi, ianw: about "Zuul needs updates to properly support Gerrit 3.3.0": https://etherpad.opendev.org/p/zuul_gerrt-3.3_comment_recheck | 15:04 |
| fungi | lyr: ianw: also remember that to build newer debian/ubuntu chroots you need a similarly new version of debootstrap, so if the version in epel is stale that can make it impossible | 15:07 |
| corvus | guillaumec: it looks like stream-events doesn't report the information we need, is that right? | 15:08 |
| guillaumec | corvus, right | 15:08 |
| corvus | guillaumec: i'll draft a message to the repo-discuss list | 15:09 |
| fungi | so looks like it's that gerrit 3.3.0 needs updates to be properly supportable? | 15:11 |
| corvus | guillaumec: how does this look? https://etherpad.opendev.org/p/Ctf_-EL090z4f3Y5ZvwF | 15:23 |
| lyr | fungi: thanks, noted | 15:26 |
| *** bhavikdbavishi has quit IRC | 15:26 | |
| guillaumec | corvus, lgtm | 15:28 |
| clarkb | tobiash: yes help would be appreciated. I've got too many distractions right now. My next steps were likely going to be using the functional openshift job + updates to the openshift driver to sort out if I'm using the new openshift stuff properly. Then figure out our unit test fakes once more confident the new apis are used properly. The current ps seems to work for the k8s side which is good | 15:30 |
| corvus | guillaumec: sent! thanks for doing the research | 15:30 |
| *** hashar has quit IRC | 15:31 | |
| tobiash | clarkb: k, then I'll take it over and fix it hopefully tomorrow | 15:39 |
| tobiash | thanks | 15:40 |
| corvus | tobiash: [database] sounds like a winner | 15:40 |
| tobiash | ++ | 15:40 |
| corvus | mhu: with zuul and nodepool, we try not to make a release without any release notes, but there are none pending: https://zuul-ci.org/docs/zuul-client/releasenotes.html | 15:41 |
| corvus | for zuul-client | 15:42 |
| mhu | corvus, right, shall I add one for the fix? | 15:42 |
| corvus | mhu: it looks like https://opendev.org/zuul/zuul-client/commit/b371dc2f40b1cd1b51d266ad7a4eefedfa68414e is a serious enough bug to warrant a release note (and serious enough to warrant a release) | 15:42 |
| corvus | mhu: so yeah, i think that'd be a good idea | 15:42 |
| corvus | mhu: we can merge that real quick then do a release | 15:42 |
| mhu | corvus, great, on it | 15:44 |
| *** tosky has quit IRC | 15:45 | |
| tobiash | clarkb: it would be great to get https://review.opendev.org/c/zuul/zuul-client/+/765516 into the zuul-client so we can start to use it with the next release? (fix for .netrc users) | 15:51 |
| mhu | tobiash, you should add a release note as well then | 15:52 |
| tobiash | k | 15:52 |
| corvus | i don't think we need a release note for that | 15:52 |
| corvus | i mean, if you want to add it, sure | 15:53 |
| tobiash | k | 15:53 |
| tobiash | it's a simple bugfix where we usually didn't do release notes | 15:53 |
| corvus | but it only affected some users in certain conditions | 15:53 |
| corvus | yeah | 15:53 |
| corvus | the encrypt thing was a bit more 'major' in that the encrypt command just didn't work, and it's important enough to cut a release, so it's important enough for a note | 15:54 |
| *** ikhan has quit IRC | 15:54 | |
| corvus | mhu: btw, is there testing that would catch that encrypt error now? | 15:54 |
| mhu | corvus, i could add one, there's already a functional test but with too small a value to trigger the chunk loop error | 15:55 |
| corvus | mhu: big values are pretty common and trigger special code paths (ie, the chunk loop), so i think it'd be a good idea to do that | 15:56 |
| clarkb | tobiash: I'll take a look in a bit. Juggling a few things and a meeting | 15:56 |
| tobiash | thanks :) | 15:56 |
| mhu | agreed, I'll add it next, it's trivial from the existing test | 15:56 |
| corvus | mhu: cool thx | 15:56 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-operator master: Add zuul-ensure-own-config role https://review.opendev.org/c/zuul/zuul-operator/+/765882 | 15:57 |
| *** ikhan has joined #zuul | 15:58 | |
| tobiash | I also have a weird day today with thousands of interruptions and small issues everywhere | 15:58 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul-client master: Add release note for encrypt bugfix https://review.opendev.org/c/zuul/zuul-client/+/766031 | 15:59 |
| mhu | corvus, if it's fine for you ^ | 15:59 |
| *** hashar has joined #zuul | 16:00 | |
| clarkb | tobiash: we need a support group | 16:00 |
| tobiash | today I am the support group of my support group ;) | 16:01 |
| avass | I know that feeling | 16:05 |
| tobiash | corvus: I'll look into the [database] tomorrow then | 16:09 |
| *** bhavikdbavishi has joined #zuul | 16:10 | |
| *** rlandy is now known as rlandy|bbl | 16:11 | |
| *** bhavikdbavishi1 has joined #zuul | 16:12 | |
| *** bhavikdbavishi has quit IRC | 16:14 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 16:14 | |
| *** hamalq has joined #zuul | 16:18 | |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: zuul-client: test encryption of large secrets https://review.opendev.org/c/zuul/zuul/+/766037 | 16:21 |
| lyr | Any hints to fix openstack.exceptions.NotSupported: The image service for openstack:RegionOne exists but does not have any supported versions ? | 16:23 |
| openstackgerrit | Merged zuul/zuul-client master: Add release note for encrypt bugfix https://review.opendev.org/c/zuul/zuul-client/+/766031 | 16:23 |
| *** hamalq has quit IRC | 16:23 | |
| clarkb | lyr: I think that means the version of the image service (glance) that is running isn't supported by your config (or sdk version) | 16:24 |
| clarkb | lyr: you might be able to work around this by explicitly pinning to an older image api verison in clouds.yaml or by downgrading sdk? | 16:25 |
| *** mach1na has quit IRC | 16:25 | |
| lyr | clarkb: thanks, I'll dig in my clouds.yaml | 16:31 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-operator master: Add zuul-ensure-own-config role https://review.opendev.org/c/zuul/zuul-operator/+/765882 | 16:36 |
| corvus | mhu, tobiash: commit bc0597ee54264f0c98a346ae0542c49afa3ca4df (HEAD -> master, tag: 0.0.4, origin/master, origin/HEAD, refs/changes/31/766031/1) | 16:38 |
| corvus | mhu, tobiash: that look good for zuul-client? | 16:38 |
| tobiash | I guess I'll have to wait for the next release then for the .netrc fix | 16:39 |
| corvus | tobiash: oh we can wait for that | 16:39 |
| clarkb | I'm just about in a spot where I can review taht | 16:39 |
| clarkb | give me 10 minutes? | 16:39 |
| tobiash | clarkb: awesome, thanks! | 16:40 |
| corvus | yeah sorry, no rush | 16:40 |
| corvus | that's why i asked :) | 16:40 |
| tobiash | :) | 16:40 |
| corvus | in case i forgot something :) | 16:40 |
| clarkb | approved, that didn't take as long as I feared | 16:44 |
| tobiash | :) | 16:44 |
| tobiash | it was just a small fix | 16:45 |
| clarkb | ya, and more a python issue than needing to udnerstand netrc or requests internals in a significant way | 16:45 |
| mhu | corvus, tobiash clarkb if you can go the extra mile and clear these that are +1 or +2'ed already ... https://review.opendev.org/q/project:zuul/zuul-client+status:open+(label:Code-Review%252B1+OR+label:Code-Review%252B2) | 16:45 |
| clarkb | or at least the fix was python workaround | 16:45 |
| mhu | otherwise I'm content with a fix release up to .netrc | 16:45 |
| tobiash | mhu: I've looked at them and they all have +2 from me or I'm the author | 16:46 |
| clarkb | sorry still juggling a few things, if I get a chance I'll look | 16:46 |
| mhu | tobiash, there are two from me in the list, https://review.opendev.org/c/zuul/zuul-client/+/765313 and https://review.opendev.org/c/zuul/zuul-client/+/765203 | 16:46 |
| mhu | oh yeah, you +2'ed thme already, thanks :) | 16:47 |
| tobiash | mhu: yeah, both lack a non-me review ;) | 16:47 |
| mhu | they're not essential for release but bring z-c encrypt on par with the features of encrypt_secret.py in zuul/zuul | 16:48 |
| *** sugaar4 has joined #zuul | 16:54 | |
| *** sugaar has quit IRC | 16:55 | |
| lyr | clarkb: can't find anything in clouds.yaml to enforce a glance api version | 16:56 |
| *** hamalq has joined #zuul | 16:56 | |
| lyr | clarkb: we do have an ageing openstack, including glance api v1 | 16:56 |
| clarkb | lyr: I think you can try something like image_api_version: 1 in your cloud profile in the clouds.yaml | 16:57 |
| clarkb | lyr: we more commonly use it to control identity versions using identity_api_version: 3 for example | 16:57 |
| clarkb | however if sdk isn't automatically falling back to v1 for glance then I wonder if they dropped support for it all together :/ | 16:58 |
| mordred | not to my knowledge - it should just work | 16:59 |
| lyr | clarkb: hmm... I feel I was mislead by warning message, our glance does have "enable_v2_api = true | 16:59 |
| lyr | " | 16:59 |
| mordred | lyr: there could be issues with the version discovery documents your glance is serving - I can't remember, but I think there may be old bugs in glace causing that to be auth protected. openstacksdk _should_ work around those as long as you have recent openstacksdk and recent keystoneauth | 17:00 |
| lyr | I also have "Unable to establish connection to http://192.168.32.139:8774/v2.1/REDACTED/servers/detail" on zuul server side, which is strange, this is an openstack IP | 17:00 |
| openstackgerrit | Merged zuul/zuul-client master: Prevent override of bearer token by .netrc https://review.opendev.org/c/zuul/zuul-client/+/765516 | 17:01 |
| *** hamalq has quit IRC | 17:01 | |
| mordred | yeah - v1 and v2 support are both still in sdk | 17:02 |
| *** hamalq has joined #zuul | 17:06 | |
| corvus | mhu: qq on https://review.opendev.org/765203 | 17:09 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-operator master: Add zuul-ensure-own-config role https://review.opendev.org/c/zuul/zuul-operator/+/765882 | 17:15 |
| corvus | mhu: also 203 has a merge conflict that needs resolving | 17:15 |
| tobiash | 765313 is in merge conflict as well | 17:21 |
| *** tosky has joined #zuul | 17:26 | |
| mhu | corvus, tobiash thanks I'll have a look tomorrow | 17:27 |
| lyr | So, I've this error, how can I determine a supported version & enforce it in the config ? https://paste.garrigue.re/?50ebc5e96cb62342#5F5DLNmwjGjKkHZMogPQC3JrAWhHYGREMdmGr5h82RXs | 17:30 |
| clarkb | lyr: if you fetch the root of the glance image url path it should give you version data | 17:32 |
| *** jcapitao has quit IRC | 17:36 | |
| lyr | clarkb: I don't know where to look. About version, I've nodepool 3.13.1, glance 2.9.1, glance-api 16.0.1 | 17:41 |
| clarkb | lyr: in this case it is talkinb about the api version (the protocol that glance 2.9.1 speaks for example) | 17:42 |
| clarkb | lyr: if your glance is at https://glance.mycloud.com/ curling that url should get you back a json document with version discovery info | 17:42 |
| *** rpittau is now known as rpittau|afk | 17:43 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-operator master: Add zuul-ensure-own-config role https://review.opendev.org/c/zuul/zuul-operator/+/765882 | 17:44 |
| lyr | clarkb: https://paste.garrigue.re/?57c192a5fa55971a#6yrBtD5rDEyZ8AgMrijtn25FKgFms6d1F7wNtismqy1W | 17:44 |
| lyr | gotta go for today | 17:44 |
| lyr | but if you got suggestion about config to be tried, I'll gladly test them | 17:45 |
| clarkb | mordred: ^ any idea why the sdk wouldn't like that document? | 17:46 |
| clarkb | maybe beacuse the catalog url doesn't point to the glance root? | 17:46 |
| clarkb | lyr: thats the next thing I would check (but enjoy your time away from work), do a catalog list and check what is returned for image there, then curl that url and cross check against the paste you just shared | 17:47 |
| *** cloudnull has quit IRC | 17:47 | |
| clarkb | maybe a proxy is sad or the url path isn't :9292/ but :9292/something-else | 17:47 |
| *** cloudnull has joined #zuul | 17:47 | |
| mordred | clarkb: yeah - that doc looks fine | 18:00 |
| *** hashar is now known as hasharDinner | 18:02 | |
| *** jfoufas1 has quit IRC | 18:13 | |
| *** nils has quit IRC | 18:16 | |
| *** saneax has quit IRC | 18:27 | |
| fungi | anybody happen to know why zuul doesn't consider the wip flag in gerrit as blocking merge? we tested approving a "wip" change in opendev, and zuul happily enqueued it into our gate pipeline (but could not merge it, only reported verified +2) | 18:27 |
| fungi | i guess the api call we use to check mergability requirements are met doesn't take the wip change state into account? | 18:28 |
| clarkb | fungi: that would be my hunch | 18:29 |
| clarkb | iirc its a query to gerrit that says if something can merge or not | 18:29 |
| clarkb | possible the bug is in gerrit tiself | 18:29 |
| SpamapS | In theory that's functionally equivalent to draft PR status in GitHub | 18:29 |
| fungi | yeah, basically what i'm wondering | 18:29 |
| fungi | SpamapS: yeah, i'm more just curious if gerrit is failing to report wip state as a blocking condition | 18:30 |
| fungi | but still apparently blocking on it | 18:30 |
| corvus | fungi, SpamapS: a quick "ssh review gerrit query --submit-records change:765821" shows wip is not in the submitRecords field which is what zuul currently looks at | 18:40 |
| *** bhavikdbavishi has quit IRC | 18:40 | |
| corvus | so we'll need an extra check for "wip: true" in the gerrit driver | 18:40 |
| corvus | should be easy | 18:40 |
| corvus | i think i'd add it as a pipeline requirement option | 18:40 |
| corvus | (that way you could have pipelines that only act on wip changes; dunno what for, but that's the sort of thing we usually try to accomodate for future expansion) | 18:41 |
| fungi | sounds reasonable, but also is wip something which *should* be included in submitRecords? | 18:41 |
| corvus | fungi: that's a good question i don't have an answer for; good repo-discuss topic i think | 18:41 |
| fungi | it does seem to effectively block submitting | 18:41 |
| *** piotrowskim has quit IRC | 18:44 | |
| zbr | corvus: fungi: if you can give me some hints I count try to do something tomorrow about wip. clearly we need it and may help me learn more about zuul. | 18:46 |
| zbr | am going offline now but I will read the backlog tomorrow | 18:46 |
| corvus | zbr: yep -- mimic the 'status: open' pipeline requirement | 18:47 |
| fungi | right, and once that merges and we get our scheduler restarted in opendev we can add wip as an explicit blocker in our gate pipeline definition | 18:49 |
| tristanC | fungi: if that effectively block submitting, shouldn't it be a default requirement for gerrit submit? | 18:52 |
| zbr | tristanC: i think it is. | 18:52 |
| tobiash | I think the pipeline requirement is useful but as part of a gate pipeline I think it should be handled automatically in the canMerge method in the gerrit driver | 18:52 |
| zbr | indeeed wip presence indicates inability to merge, so no gate pipelines should ever run on WIP, but check ones should. | 18:53 |
| corvus | yes, we should do both. | 18:54 |
| fungi | right, two-pronged approach: allow zuul to check wip state so it can be explicitly matched on in pipelines, but also start a discussion in the gerrit community about incorporating wip into submitRecords | 18:55 |
| tobiash | The canMerge is probably quite easy by adding wip as change attribute and check for it in canMerge | 18:55 |
| corvus | though i don't think we check open in canMerge? perhaps it is incorporated in the submit records check | 18:55 |
| corvus | tobiash: yes, but i want both | 18:55 |
| corvus | i definitely want it as a pipeline requirement/reject, so if zbr wants to write the change, then it should be included | 18:56 |
| tobiash | sure, both use cases make sense | 18:56 |
| corvus | if that's too big of a change, that's fine, i'll write the patch | 18:56 |
| tobiash | the pipeline requirement could be backed by a change attribute as well I guess | 18:57 |
| *** frickler has joined #zuul | 19:00 | |
| corvus | anyway, that should be enough for zbr to look at tomorrow, so let's see what he comes up with | 19:02 |
| tobiash | today I found an interesting bug related to dynamic layouts. Suppose repo a using extra-config-paths and repo b without extra-config-paths. There are changes x in a and y in b where y depends-on x and both touch zuul config. In this case the dynamic layout of y misses the extra-config-paths of x most likely leading to a job freeze error. | 19:07 |
| tobiash | I suppose we're not passing the extra config files to the initial merge in this case | 19:07 |
| tobiash | but haven't yet time to look deeper | 19:07 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: [DNM] tutorials: push dep image to buildset registry https://review.opendev.org/c/zuul/zuul/+/766076 | 19:17 |
| *** Goneri has quit IRC | 19:18 | |
| *** Goneri has joined #zuul | 19:37 | |
| *** ianychoi__ has quit IRC | 19:44 | |
| *** Pilou has quit IRC | 19:53 | |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: [DNM] run tutorials using gerrit-3.3.0 https://review.opendev.org/c/zuul/zuul/+/766086 | 20:20 |
| ianw | corvus: you're right, the tag is there in "recent" zuul comments (like, not from a random change years ago). | 20:21 |
| corvus | ianw: yeah, i think it's been there for ~1 year | 20:26 |
| *** rlandy|bbl is now known as rlandy|pto | 20:26 | |
| ianw | corvus: probably makes a nice carrot, if you want your results in the table, use the latest zuul :) | 20:38 |
| corvus | :) | 20:38 |
| corvus | i *think* http reporting is required for that; so that's a zuulv3 only feature | 20:39 |
| corvus | (a zuulv2 backport would not be trivial) | 20:39 |
| fungi | might also require some extra effort for jenkins and other non-zuul systems to make comments matching the same set of parameters | 20:54 |
| corvus | ianw: is there a way to customize a polygerrit plugin? ie, pass config parameters through gerrit.config? | 21:00 |
| corvus | was wondering if we could supply a regex for name-based matching (so the plugin could do tag-based and then optionally name-based) | 21:00 |
| ianw | corvus: i do not *think* so; it might need a java component to expose a rest api that the plugin would query | 21:02 |
| corvus | bummer; a generic config pass-through would be cool | 21:04 |
| guillaumec | corvus, ianw : zuul-status plugin is doing that https://gerrit.googlesource.com/plugins/zuul-status/+/refs/heads/master/src/main/resources/static/zuul-status-view.js#128 https://gerrit.googlesource.com/plugins/zuul-status/+/refs/heads/master/src/main/resources/Documentation/config.md | 21:22 |
| ianw | guillaumec: oh, nice! well looks like it can be done then :) | 21:23 |
| corvus | perfect! | 21:26 |
| *** hasharDinner has quit IRC | 21:34 | |
| *** zenkuro has quit IRC | 21:39 | |
| *** ikhan has quit IRC | 23:02 | |
| *** vishalmanchanda has quit IRC | 23:05 | |
| *** rfolco has quit IRC | 23:06 | |
| clarkb | tobiash: looking at the openshift functional job logs I'm pretty sure I've got the new api stuff not completely correct. Specifically I think I need to requets *List kinds for list results | 23:21 |
| clarkb | tobiash: I bet you'v egot an openshift to test against locally so you might end up being quicker than me | 23:21 |
| *** rfolco has joined #zuul | 23:21 | |
| ianw | corvus/peanut gallery: https://104.130.172.52/q/project:openstack%252Fdiskimage-builder is pretty much complete now for the plugin now. i imported our users so i can actually test a wide range of results | 23:22 |
| ianw | i'm taking suggestions on tweaking the table layout etc if anyone has them | 23:22 |
| ianw | every change should have a "Zuul Summary" tab | 23:23 |
| mordred | ianw: my browser just flat refuses to serve that page to me | 23:24 |
| ianw | mordred: it is self-signed | 23:24 |
| mordred | apparently chrome does not want to give me the ability to choose to view it | 23:24 |
| mordred | "You cannot visit 104.130.172.52 right now because the website sent scrambled credentials that Google Chrome cannot process. Network errors and attacks are usually temporary, so this page will probably work later." | 23:24 |
| mordred | it's not important, obvs | 23:24 |
| mordred | just talking out loud | 23:24 |
| ianw | ahh, i think i heard they had turned that on; that you can't even override by default a self-signed cert | 23:25 |
| mordred | silly me for clicking the update button | 23:26 |
| ianw | it probably makes sense to actually display the ci results in reverse chronological order, with the latest at the top | 23:26 |
| guillaumec | ianw, personal preference for 2px padding, it could be ... configurable ! it looks nice | 23:33 |
| ianw | guillaumec: 2px padding on the <td> you mean? | 23:34 |
| guillaumec | exactly | 23:35 |
| *** rlandy|pto has quit IRC | 23:35 | |
| ianw | i agree, updated that now. however i think it needs some left padding now | 23:36 |
| ianw | i guess that would be padding-left on the table | 23:36 |
| tristanC | ianw: that's looking good to me, thanks! | 23:36 |
| openstackgerrit | Clark Boylan proposed zuul/nodepool master: Bump openshift dep https://review.opendev.org/c/zuul/nodepool/+/765873 | 23:37 |
| clarkb | I haveno idea if ^ is correct, mostly using the ci to help me fumble around | 23:37 |
| clarkb | openshift's api docs and the python lib don't really align in their communication of how to query resources | 23:37 |
| fungi | mordred: strangely, clicking anywhere in the page once chrome refuses to load it and typing "thisisunsafe" will cause it to then load? | 23:40 |
| fungi | strange internet rumors | 23:41 |
| fungi | something about secret bypass sequences | 23:41 |
| mordred | fungi: oh my holy craop | 23:42 |
| mordred | that does, in fact, work | 23:42 |
| mordred | I'm a bit disappointed it wasn't up up down down left right b a | 23:42 |
| mordred | ianw: that looks awesome! | 23:43 |
| fungi | yup, hilarious | 23:44 |
| fungi | and yeah, the konami code would have been better, missed opportunities | 23:44 |
| mordred | ianw: has it been suggested already to have a gerrit be able to configure which zuul is "the" zuul? | 23:45 |
| ianw | mordred: yeah, i think we probably want to add a sort for that | 23:45 |
| mordred | cool | 23:45 |
| mordred | brilliant work | 23:46 |
| ianw | i'm not sure what the old one did? | 23:46 |
| mordred | I also don't know - but I feel like the official zuul was on top? | 23:47 |
| clarkb | ya I think it was | 23:48 |
| ianw | it's compressed down to just https://github.com/ianw/gerrit-zuul-summary-status/blob/main/gr-zuul-summary-status/gr-zuul-summary-status.js now | 23:48 |
| *** rfolco has quit IRC | 23:53 | |
| *** rfolco has joined #zuul | 23:53 | |
| *** rfolco has quit IRC | 23:58 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!