| *** bobh has joined #openstack-release | 00:15 | |
| *** wolverineav has quit IRC | 00:23 | |
| *** wolverineav has joined #openstack-release | 00:24 | |
| *** bobh has quit IRC | 00:28 | |
| *** wolverineav has quit IRC | 00:29 | |
| *** thorst has joined #openstack-release | 00:29 | |
| *** kumarmn has quit IRC | 00:31 | |
| *** thorst has quit IRC | 00:46 | |
| fungi | dhellmann: i figured out the missing signatures thing, patches up to address (it specifically impacts tagged releases for python api server projects). root cause is that some people assumed any python artifacts not published to pypi also didn't need some other release features. basically the job for branch tarballs in post was also being reused for tags for those projects | 00:47 |
|---|---|---|
| fungi | smcginnis: ^ | 00:47 |
| fungi | unrelated, we have people wanting to link to sections of http://git.openstack.org/cgit/openstack/releases/tree/README.rst | 00:47 |
| fungi | any chance we could start publishing a rendered version of that somewhere? (docs.o.o, releases.o.o...) | 00:48 |
| fungi | or is it already being published somewhere i'm just not finding? | 00:48 |
| *** ekcs has quit IRC | 00:49 | |
| *** pkovar has quit IRC | 00:56 | |
| *** mriedem has quit IRC | 00:56 | |
| *** amoralej|off has quit IRC | 01:04 | |
| *** kumarmn has joined #openstack-release | 01:05 | |
| *** amoralej has joined #openstack-release | 01:08 | |
| *** kumarmn has quit IRC | 01:12 | |
| *** amoralej has quit IRC | 01:13 | |
| *** mriedem has joined #openstack-release | 01:15 | |
| *** Qiming has quit IRC | 01:20 | |
| *** amoralej has joined #openstack-release | 01:20 | |
| *** sree has joined #openstack-release | 01:27 | |
| *** Qiming has joined #openstack-release | 01:28 | |
| *** mriedem has quit IRC | 01:28 | |
| *** thorst has joined #openstack-release | 01:29 | |
| *** sree has quit IRC | 01:32 | |
| *** Qiming has quit IRC | 01:33 | |
| *** thorst has quit IRC | 01:34 | |
| *** Qiming has joined #openstack-release | 01:35 | |
| *** Qiming has quit IRC | 01:39 | |
| *** Qiming has joined #openstack-release | 01:40 | |
| *** edmondsw has joined #openstack-release | 01:43 | |
| *** ricolin has joined #openstack-release | 01:46 | |
| *** edmondsw has quit IRC | 01:47 | |
| *** hongbin has joined #openstack-release | 01:57 | |
| *** bobh has joined #openstack-release | 02:10 | |
| *** amoralej has quit IRC | 02:12 | |
| *** yamamoto has joined #openstack-release | 02:14 | |
| *** bobh has quit IRC | 02:15 | |
| *** jhesketh_ is now known as jhesketh | 02:22 | |
| *** Qiming has quit IRC | 02:22 | |
| *** thorst has joined #openstack-release | 02:46 | |
| *** cody-somerville has joined #openstack-release | 02:48 | |
| *** cody-somerville has quit IRC | 02:48 | |
| *** cody-somerville has joined #openstack-release | 02:48 | |
| *** thorst has quit IRC | 02:50 | |
| *** Qiming has joined #openstack-release | 02:52 | |
| *** Qiming has quit IRC | 02:56 | |
| *** dave-mccowan has quit IRC | 02:59 | |
| *** iyamahat has joined #openstack-release | 03:06 | |
| *** yamahata has quit IRC | 03:08 | |
| *** iyamahat has quit IRC | 03:14 | |
| *** edmondsw has joined #openstack-release | 03:31 | |
| *** edmondsw has quit IRC | 03:35 | |
| *** hongbin has quit IRC | 04:00 | |
| *** yamahata has joined #openstack-release | 04:01 | |
| *** sree has joined #openstack-release | 04:02 | |
| *** claudiub|2 has joined #openstack-release | 04:08 | |
| *** iyamahat has joined #openstack-release | 04:23 | |
| *** coolsvap has joined #openstack-release | 04:32 | |
| *** ekcs has joined #openstack-release | 04:35 | |
| *** thorst has joined #openstack-release | 04:51 | |
| *** udesale has joined #openstack-release | 04:54 | |
| *** thorst has quit IRC | 04:56 | |
| *** iyamahat has quit IRC | 05:06 | |
| *** iyamahat has joined #openstack-release | 05:09 | |
| openstackgerrit | Tony Breeds proposed openstack/releases master: Release designate-dashboard 6.0.0.0b1 https://review.openstack.org/514800 | 05:16 |
| openstackgerrit | Tony Breeds proposed openstack/releases master: Release Designate 6.0.0.0b1 https://review.openstack.org/514798 | 05:18 |
| *** iyamahat has quit IRC | 06:29 | |
| *** pcaruana has joined #openstack-release | 06:39 | |
| *** csomerville has joined #openstack-release | 06:40 | |
| *** cody-somerville has quit IRC | 06:43 | |
| *** ekcs has quit IRC | 06:51 | |
| *** thorst has joined #openstack-release | 06:53 | |
| *** thorst has quit IRC | 06:57 | |
| *** gcb has joined #openstack-release | 07:00 | |
| ttx | fungi: <trolling>https://github.com/openstack/releases/blob/master/README.rst</trolling> | 07:05 |
| *** jtomasek has joined #openstack-release | 07:09 | |
| *** armax has quit IRC | 07:09 | |
| smcginnis | ttx: Hah, that was my first thought. :) | 07:41 |
| smcginnis | Would be interesting to know the need/what they actually want to link to. Can certainly add some docs. | 07:42 |
| ttx | I think he was baiting | 07:42 |
| smcginnis | I think I saw some conversation around it in the infra channel, but I was on the way to the airport at the time. | 07:43 |
| smcginnis | I'll have to see if I still have it in my scrollback. | 07:43 |
| *** amoralej has joined #openstack-release | 07:45 | |
| *** gcb has quit IRC | 07:57 | |
| *** pbandark has joined #openstack-release | 08:14 | |
| *** thorst has joined #openstack-release | 08:14 | |
| *** thorst has quit IRC | 08:19 | |
| smcginnis | ttx: In case you are curious on the README publishing background: https://review.openstack.org/#/c/506812/2/doc/source/migration.rst | 08:22 |
| ttx | ah! | 08:26 |
| ttx | my trolling is less funny now | 08:27 |
| smcginnis | :) | 08:27 |
| ttx | smcginnis: tempted to try releases one by one until one fails. Objections? | 08:27 |
| ttx | (it's my release day) | 08:27 |
| smcginnis | ttx: I think we may have one outstanding issue yet (that we know of) with the tarballs for service projects not getting signed. | 08:28 |
| smcginnis | ttx: But other than that, Doug and Jeremy fixed an issue with having multiple deliverables in one release yesterday. | 08:29 |
| smcginnis | And things should be working for the most part now. | 08:29 |
| ttx | hmm, I can process doc/tolling changes at least | 08:29 |
| ttx | tooling* | 08:29 |
| smcginnis | We have not fully tested non-python lib or service releases either. | 08:29 |
| smcginnis | ttx: Yes, those should be safe as far as I'm aware. | 08:29 |
| smcginnis | ttx: Would be great to get your thoughts on: https://review.openstack.org/512443 | 08:32 |
| ttx | yes I'm on it | 08:33 |
| smcginnis | Cool, thanks. | 08:33 |
| ttx | smcginnis: I'd say it's missing documentation, was that planned as a separate change ? | 08:33 |
| *** lucas-afk is now known as lucasagomes | 08:34 | |
| *** electrofelix has joined #openstack-release | 08:34 | |
| smcginnis | ttx: No, I'll add that. | 08:39 |
| ttx | otherwise looks good afaict | 08:39 |
| smcginnis | Although... maybe we should talk more about how we want this all to work long term. | 08:40 |
| smcginnis | Like, do we plan on linking to it visibly from somewhere else? I wasn't sure on that point. | 08:40 |
| ttx | I think it's fine if the link appears in README | 08:41 |
| smcginnis | An earlier take on it, I had it included in the series release page with the link to the schedule. | 08:41 |
| openstackgerrit | Merged openstack/releases master: Add keystone deadlines to queens schedule https://review.openstack.org/509835 | 08:41 |
| smcginnis | ttx: Ah, so have the README state to get the highlights, go to releases.openstack.org/$SERIES/highlights.html | 08:41 |
| ttx | or "things you put here will appear in..." | 08:42 |
| smcginnis | OK, works for me. | 08:42 |
| ttx | smcginnis: that failure on https://review.openstack.org/#/c/514844/1 is a bit weird | 08:47 |
| ttx | especially as nothing really changed around that since the check test yesterday | 08:48 |
| smcginnis | Yeah, I pulled that up but haven't dig in yet. | 08:48 |
| smcginnis | It doesn't look like it was run in the check job for some reason. | 08:48 |
| smcginnis | Yet those tests were not just added. | 08:48 |
| ttx | I'm tempted to retry it but then we might miss a corner case | 08:48 |
| ttx | smcginnis: I'll approve https://review.openstack.org/#/c/513432/3 (horizon q1) so that we see if signing was fixed or not | 08:50 |
| ttx | not a big deal if it hasn't as it's just a b1 | 08:50 |
| smcginnis | ttx: Sounds good. Let's take a look at the job logs and the artifacts after it completes to make sure it all looks right. | 08:51 |
| ttx | hmm, it's probably not fixed yet as https://review.openstack.org/#/c/514869/ was not merged | 08:51 |
| smcginnis | Yeah, there's a group of patches (linked in there) for the signing issue. | 08:54 |
| smcginnis | Sorry, I missed that part of the comment earlier. It is not fixed. | 08:55 |
| ttx | ok, let's hold a bit then | 08:55 |
| openstackgerrit | Sean McGinnis proposed openstack/releases master: Add release highlights https://review.openstack.org/512443 | 08:57 |
| *** yamahata has quit IRC | 08:58 | |
| *** yamamoto has quit IRC | 09:05 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/releases master: Release ironic-inspector 6.1.0 for Queens https://review.openstack.org/514967 | 09:09 |
| *** yamamoto has joined #openstack-release | 09:12 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/releases master: Release ironic 9.1.2 for Pike https://review.openstack.org/514395 | 09:15 |
| *** sdague has joined #openstack-release | 09:20 | |
| openstackgerrit | Ian Wienand proposed openstack-infra/release-tools master: Add some notes on eol_branch.sh https://review.openstack.org/514846 | 09:22 |
| *** yamamoto has quit IRC | 09:49 | |
| *** udesale has quit IRC | 10:13 | |
| *** ricolin has quit IRC | 10:28 | |
| *** pcaruana has quit IRC | 10:37 | |
| *** sree has quit IRC | 10:38 | |
| *** sree has joined #openstack-release | 10:38 | |
| *** sree has quit IRC | 10:43 | |
| *** yamamoto has joined #openstack-release | 10:49 | |
| *** yamamoto has quit IRC | 11:01 | |
| *** dtantsur|afk is now known as dtantsur | 11:01 | |
| *** sree has joined #openstack-release | 11:02 | |
| dtantsur | hey folks! is the py35 job broken? it complains about something wrt automaton | 11:06 |
| dtantsur | http://logs.openstack.org/67/514967/1/check/openstack-tox-py35/5df54db/testr_results.html.gz | 11:06 |
| *** sree has quit IRC | 11:07 | |
| *** yamamoto has joined #openstack-release | 11:14 | |
| *** lucasagomes is now known as lucas-hungry | 11:20 | |
| *** bobh has joined #openstack-release | 11:20 | |
| *** thorst has joined #openstack-release | 11:21 | |
| *** bobh has quit IRC | 11:23 | |
| *** thorst has quit IRC | 11:26 | |
| *** dave-mccowan has joined #openstack-release | 11:30 | |
| *** bobh has joined #openstack-release | 11:38 | |
| *** wolverineav has joined #openstack-release | 11:43 | |
| *** pcaruana has joined #openstack-release | 11:43 | |
| *** bobh has quit IRC | 11:45 | |
| *** bobh has joined #openstack-release | 11:52 | |
| fungi | smcginnis: ttx: i'm perfectly happy to generate relevant signatures with our current signing key and stick them on tarballs.o.o in the right places until the job corrections merge | 11:54 |
| fungi | also, yes, what prompted the request for publishing the instructions from the releases repo readme somewhere is that the storyboard migration docs want to link to the "deliverable files" subsection, and the current patch links to them on gh (which is less than preferable) | 11:56 |
| ttx | dtantsur: yes there is an issue there, started overnight | 11:56 |
| dtantsur | ack | 11:56 |
| ttx | waiting for dhellmann since he may know what other change triggered it | 11:56 |
| ttx | Looks like an error that used to be caught is now bubbling up | 11:57 |
| fungi | i'm also gone most of today (appointment on the mainland) but should be able to catch up a bit over wireless modem around 1500z and hoping to be home to pitch in again by 2100z | 11:57 |
| ttx | "fatal: ambiguous argument '1.4.1': unknown revision or path not in the working tree" | 11:58 |
| *** Guest3211 has quit IRC | 11:59 | |
| * ttx digs deeper | 12:00 | |
| *** yamamoto has quit IRC | 12:02 | |
| *** sree has joined #openstack-release | 12:02 | |
| *** pcaruana has quit IRC | 12:03 | |
| ttx | that error is actually expected | 12:03 |
| *** pcaruana has joined #openstack-release | 12:04 | |
| *** sree has quit IRC | 12:06 | |
| *** thorst has joined #openstack-release | 12:08 | |
| *** pkovar has joined #openstack-release | 12:12 | |
| *** amrith has joined #openstack-release | 12:14 | |
| ttx | ok so the issue is probably the newton-eol tag and removal of branch | 12:14 |
| *** amrith is now known as Guest7230 | 12:15 | |
| ttx | I'll propose something | 12:15 |
| *** kumarmn has joined #openstack-release | 12:15 | |
| *** kumarmn has quit IRC | 12:20 | |
| *** rosmaita has joined #openstack-release | 12:22 | |
| dhellmann | o/ | 12:23 |
| *** lucas-hungry is now known as lucasagomes | 12:24 | |
| ttx | dhellmann: looks like our newton-releated automaton tests trigger false negatives now that the branch is gone | 12:24 |
| ttx | trying to work up a patch | 12:24 |
| *** thorst_ has joined #openstack-release | 12:24 | |
| dhellmann | ttx: yeah, those tests are pretty bad (says the author). They rely on state in the repos they clone. We should use the release-test repo instead. If you get them fixed up for now, I'll take on rewriting them (unless you want to do that, too) | 12:25 |
| ttx | I now get two errors instead of 0 (expecting just 1) | 12:25 |
| *** thorst has quit IRC | 12:26 | |
| dhellmann | ttx, smcginnis : as far as publishing the docs, I think we should do it, but I'm not sure the best way. Simplest might be to move the job for our existing custom publishing into the repo, then change what files it expects to look for so we can use the standard documentation publishing job to build a manual. Alternately, we could add content to the project team guide, or a new release-manual, but making a new | 12:27 |
| dhellmann | document in a separate repo feels like overkill. | 12:27 |
| *** Guest7230 has quit IRC | 12:31 | |
| *** edmondsw has joined #openstack-release | 12:32 | |
| ttx | dhellmann: hmm it's non trivial to trigger "not present in pike branch" without "receiving 1.12.2 is not a descendant of 1.12.1" with the state of pike or ocata | 12:33 |
| ttx | since they both got a point release afterwards | 12:33 |
| ttx | (after branching) | 12:33 |
| ttx | original test in newton leveraged the fact that 1.4.0 was both branch point and last tag, iiuc | 12:34 |
| *** amrith_ has joined #openstack-release | 12:35 | |
| *** amrith_ is now known as Guest30707 | 12:35 | |
| ttx | so you could simulate a 1.4.1 on top of that that pointed to a commit in master that was still a descendant of 1.4.0 and trigger only "not present in newton branch" error | 12:35 |
| *** edmondsw has quit IRC | 12:37 | |
| *** ricolin has joined #openstack-release | 12:37 | |
| *** edmondsw has joined #openstack-release | 12:37 | |
| ttx | solutions include switching to another library for the hotfix (any reason why automaton in those tests?) or properly switching to releases-test (but that would require pushing tags by hand to build what we want) | 12:38 |
| dhellmann | I picked automaton because it had a few releases and started with a so it showed up early in my poking around | 12:39 |
| dhellmann | it's probably better for us to use a repo we control, like release-test | 12:39 |
| *** bobh has quit IRC | 12:39 | |
| ttx | yes | 12:39 |
| dhellmann | I'll take a look at that | 12:39 |
| ttx | ok | 12:40 |
| ttx | failing tests are http://logs.openstack.org/67/514967/1/check/openstack-tox-py35/5df54db/testr_results.html.gz | 12:40 |
| ttx | won't be able to land anything in openstack/releases until it's fixed | 12:41 |
| dhellmann | looking now | 12:41 |
| *** udesale has joined #openstack-release | 12:48 | |
| *** pkovar has quit IRC | 12:50 | |
| *** mriedem has joined #openstack-release | 12:50 | |
| *** pkovar has joined #openstack-release | 12:58 | |
| *** yamamoto has joined #openstack-release | 13:02 | |
| dhellmann | man, these tests are ugly | 13:03 |
| dhellmann | I think I have it working, though | 13:03 |
| *** yamamoto has quit IRC | 13:12 | |
| *** kumarmn has joined #openstack-release | 13:14 | |
| *** bobh has joined #openstack-release | 13:17 | |
| *** kumarmn has quit IRC | 13:18 | |
| *** kumarmn has joined #openstack-release | 13:33 | |
| openstackgerrit | Doug Hellmann proposed openstack/releases master: update tests to replace use of automaton repo with release-test https://review.openstack.org/515069 | 13:43 |
| dhellmann | ttx: ^^ | 13:43 |
| dhellmann | bbiab | 13:43 |
| *** yamahata has joined #openstack-release | 13:44 | |
| openstackgerrit | Thierry Carrez proposed openstack/releases master: add wheel checks to missing-releases https://review.openstack.org/514844 | 13:49 |
| openstackgerrit | Thierry Carrez proposed openstack/releases master: remove obsolete release job name https://review.openstack.org/514387 | 13:49 |
| ttx | Looks good, rebased a couple of other changes on top of it | 13:49 |
| *** iyamahat has joined #openstack-release | 13:53 | |
| *** armax has joined #openstack-release | 13:53 | |
| *** sree has joined #openstack-release | 13:55 | |
| *** yamahata has quit IRC | 13:55 | |
| *** sree has quit IRC | 14:00 | |
| *** yamahata has joined #openstack-release | 14:07 | |
| *** hongbin has joined #openstack-release | 14:08 | |
| *** Guest30707 is now known as amrith | 14:10 | |
| *** iyamahat has quit IRC | 14:11 | |
| openstackgerrit | Rico Lin proposed openstack/releases master: Release heat queens-1 (10.0.0.0b1) https://review.openstack.org/515083 | 14:15 |
| openstackgerrit | Rico Lin proposed openstack/releases master: Release heatclient 1.12.0 https://review.openstack.org/515089 | 14:21 |
| openstackgerrit | Rico Lin proposed openstack/releases master: Release heatclient 1.12.0 https://review.openstack.org/515089 | 14:24 |
| *** yamahata has quit IRC | 14:25 | |
| openstackgerrit | Merged openstack/releases master: update tests to replace use of automaton repo with release-test https://review.openstack.org/515069 | 14:26 |
| openstackgerrit | Merged openstack/releases master: add wheel checks to missing-releases https://review.openstack.org/514844 | 14:26 |
| openstackgerrit | Merged openstack/releases master: remove obsolete release job name https://review.openstack.org/514387 | 14:26 |
| openstackgerrit | Rico Lin proposed openstack/releases master: Release heat-agents queens-1 (1.5.0) https://review.openstack.org/515090 | 14:26 |
| *** udesale has quit IRC | 14:26 | |
| *** jeblair has joined #openstack-release | 14:28 | |
| jeblair | smcginnis, dhellmann: hi, i need to restart the zuul scheduler... can you hold off on releases for a few minutes? | 14:29 |
| smcginnis | jeblair: Sounds good. | 14:29 |
| jeblair | i'll wait until the current 3 changes in release-post make it through | 14:30 |
| *** udesale has joined #openstack-release | 14:30 | |
| *** coolsvap has quit IRC | 14:31 | |
| *** sree has joined #openstack-release | 14:42 | |
| jeblair | smcginnis: while we're waiting -- how are things going, any problems i can help with today? | 14:43 |
| smcginnis | jeblair: Maybe. We do still have a signing issue for services that do not publish to pypi. | 14:47 |
| smcginnis | jeblair: fungi started working on it yesterday, but it got late and now he's out this morning. | 14:47 |
| *** sree has quit IRC | 14:47 | |
| smcginnis | jeblair: There's these patches that I think we need: https://review.openstack.org/#/c/514855/1 | 14:47 |
| smcginnis | And the follow on: https://review.openstack.org/#/c/514869/1 | 14:48 |
| openstackgerrit | Rico Lin proposed openstack/releases master: Release heat pike 9.0.1 https://review.openstack.org/515096 | 14:54 |
| *** yamahata has joined #openstack-release | 14:54 | |
| openstackgerrit | Rico Lin proposed openstack/releases master: Release heat ocata 8.0.5 https://review.openstack.org/515098 | 14:56 |
| jeblair | the queue is clear now, i'm going to restart | 14:57 |
| openstackgerrit | Rico Lin proposed openstack/releases master: Release python-heatclient ocata 1.8.2 https://review.openstack.org/515099 | 14:59 |
| jeblair | smcginnis, dhellmann: scheduler has been restarted. all-clear now. | 15:03 |
| smcginnis | jeblair: Thanks! | 15:03 |
| openstackgerrit | Rico Lin proposed openstack/releases master: Release python-heatclient pike 1.11.1 https://review.openstack.org/515102 | 15:03 |
| openstackgerrit | Thierry Carrez proposed openstack/releases master: Release Designate 6.0.0.0b1 https://review.openstack.org/514798 | 15:19 |
| openstackgerrit | Thierry Carrez proposed openstack/releases master: Release ironic-inspector 6.1.0 for Queens https://review.openstack.org/514967 | 15:19 |
| openstackgerrit | Thierry Carrez proposed openstack/releases master: Release heat queens-1 (10.0.0.0b1) https://review.openstack.org/515083 | 15:20 |
| openstackgerrit | Thierry Carrez proposed openstack/releases master: Release heatclient 1.12.0 https://review.openstack.org/515089 | 15:20 |
| openstackgerrit | Thierry Carrez proposed openstack/releases master: Release ironic 9.1.2 for Pike https://review.openstack.org/514395 | 15:21 |
| openstackgerrit | Thierry Carrez proposed openstack/releases master: Add release highlights https://review.openstack.org/512443 | 15:21 |
| ttx | rebasing things on top of the tests fix that just merged | 15:21 |
| *** sree has joined #openstack-release | 15:29 | |
| openstackgerrit | Doug Hellmann proposed openstack/releases master: add demo of series-highlights https://review.openstack.org/515109 | 15:32 |
| smcginnis | dhellmann: :D | 15:32 |
| *** sree has quit IRC | 15:34 | |
| *** thorst_ has quit IRC | 15:34 | |
| *** thorst has joined #openstack-release | 15:35 | |
| smcginnis | dhellmann: Oh, and thanks for the update to the ML on releases. I was thinking about that yesterday but didn't get a chance to get around to it. | 15:36 |
| dhellmann | smcginnis : np. we had a couple of people drop by the channel and I thought it would be good to make sure everyone knew we're not quite out of the woods | 15:36 |
| dhellmann | now that the tests are passing again, are we ready to approve some more and see how things go? | 15:36 |
| smcginnis | dhellmann: I think we still have the signing issue. | 15:37 |
| smcginnis | dhellmann: fungi's patches are still out there with some comments on them. | 15:37 |
| dhellmann | ah, ok | 15:37 |
| smcginnis | dhellmann: We probably could go ahead and clean up later, but I was hoping it wouldn't take too long to get those through. | 15:37 |
| dhellmann | yeah, we should wait to avoid having to do a bunch of manual signing | 15:38 |
| *** thorst has quit IRC | 15:39 | |
| fungi | i'm around for a few minutes, just bandwidth-constrained in a waiting room. i'll catch up on the patch comments and try to get them updated | 15:40 |
| dhellmann | fungi : did the signing issue affect libraries, or only services? | 15:42 |
| dhellmann | I think I remember you saying only services, which would mean we could release some of the libraries | 15:43 |
| fungi | dhellmann: any projects using the openstack-server-release project-template or the puiblish-openstack-python-tarball job in pre-release/release pipelines | 15:43 |
| dhellmann | ok | 15:44 |
| fungi | the list of 4 impacted pre-releases you provided yesterday (nova, glance, cinder, ironic?) all fell into that category, which is what ultimately led me to realize the cause | 15:44 |
| smcginnis | Yeah, that looked like just them. | 15:45 |
| fungi | looks like jeblair took care of the comments. thanks jeblair! | 15:48 |
| openstackgerrit | Thierry Carrez proposed openstack/releases master: Octavia projects Queens MS1 release https://review.openstack.org/513072 | 15:48 |
| *** pcaruana has quit IRC | 15:49 | |
| openstackgerrit | Thierry Carrez proposed openstack/releases master: Release designate-dashboard 6.0.0.0b1 https://review.openstack.org/514800 | 15:49 |
| ttx | rebasing things to get them to pass tests | 15:49 |
| ttx | dhellmann, smcginnis: if you agree, I'll approve libs now? | 15:50 |
| *** yamahata has quit IRC | 15:51 | |
| smcginnis | ttx: That sounds good to me. We should be OK for those. | 15:51 |
| smcginnis | And for the rest hopefully we should be good shortly. | 15:51 |
| fungi | okay, disappearing back into a black hole again | 15:51 |
| dhellmann | ttx, smcginnis : let's go one at a time for a while until we're sure there are no other issues, but yeah, let's start approving | 15:52 |
| smcginnis | fungi: Thanks | 15:53 |
| openstackgerrit | Witold Bedyk proposed openstack/releases master: Release monasca-grafana-datasource 1.1.0 https://review.openstack.org/515115 | 15:53 |
| ttx | ok, let me pick a few | 15:53 |
| ttx | python-watcherclient in the pipe | 15:54 |
| dhellmann | smcginnis , ttx: do we have a final decision on the pypi job variant issue that is blocking neutron projects? | 15:55 |
| dhellmann | I can work on validation updates if we want to allow the variants | 15:55 |
| smcginnis | Not sure if there was really strong concensus, but I do think we want the variants. Or rather, will need the variants. | 15:56 |
| ttx | I think we could approve https://review.openstack.org/#/c/514509 to work around the issue until we fix it properly. Have a bunch of patches rebased on top of it now | 15:57 |
| dhellmann | ttx: we could. I want them to be explicit, though. I'll work on that as a separate patch. | 16:00 |
| ttx | ack | 16:00 |
| *** coolsvap has joined #openstack-release | 16:01 | |
| ttx | so the two remaining issues (that we know of) are: lack of signing in pre-release server jobs (fungi working on it) and the neutron/horizon job situation (dhellmann working on it). None should affect regular libs | 16:03 |
| smcginnis | That's my understanding as well. | 16:03 |
| ttx | Anything else inflight I missed? | 16:03 |
| smcginnis | None that I'm aware of, but I've been too distracted the last couple days. Know of anything else dhellmann? | 16:04 |
| ttx | gate is pretty backed up so we'll have to wait before getting the trial balloon around python-watcherclient | 16:04 |
| dhellmann | I think that's everything currently in progress. | 16:05 |
| dhellmann | the email sending problem was due to load on the list server; we need to update that job to use credentials to be allowed to connect even under those circumstances. smcginnis did you capture that already? | 16:05 |
| smcginnis | dhellmann: No, I missed that one. | 16:06 |
| smcginnis | Glad to hear that issue's been identified. | 16:06 |
| dhellmann | yeah, it's why we get the email sometimes but not others -- spammers clogging up the tubes | 16:07 |
| smcginnis | Makes sense. | 16:07 |
| openstackgerrit | Doug Hellmann proposed openstack/releases master: add explicit release types for neutron and horizon plugins https://review.openstack.org/515119 | 16:09 |
| *** Rockyg has joined #openstack-release | 16:15 | |
| openstackgerrit | Doug Hellmann proposed openstack/releases master: Octavia projects Queens MS1 release https://review.openstack.org/513072 | 16:15 |
| openstackgerrit | Doug Hellmann proposed openstack/releases master: set release types for ocatavia and neutron-lbaas https://review.openstack.org/515122 | 16:15 |
| *** iyamahat has joined #openstack-release | 16:26 | |
| *** thorst has joined #openstack-release | 16:27 | |
| openstackgerrit | Merged openstack/releases master: Release python-watcherclient Queens-1 https://review.openstack.org/513452 | 16:29 |
| openstackgerrit | Sean McGinnis proposed openstack/releases master: Add release highlights https://review.openstack.org/512443 | 16:30 |
| ttx | release-post for python-watcherclient in progress | 16:31 |
| smcginnis | Hopefully not just sitting queued for too long. | 16:32 |
| *** udesale has quit IRC | 16:35 | |
| *** lucasagomes is now known as lucas-afk | 16:38 | |
| openstackgerrit | Armando Migliaccio proposed openstack/releases master: Add variants of publish-to-pypi jobs https://review.openstack.org/514509 | 16:41 |
| openstackgerrit | Armando Migliaccio proposed openstack/releases master: Neutron: release Queens M-1 milestones for neutron projects https://review.openstack.org/512432 | 16:41 |
| *** iyamahat has quit IRC | 16:43 | |
| *** iyamahat has joined #openstack-release | 16:43 | |
| armax | dhellmann: reviewed https://review.openstack.org/#/c/515119/, did you miss those projects by accident? | 16:43 |
| openstackgerrit | Rico Lin proposed openstack/releases master: Release heat pike 9.0.1 https://review.openstack.org/515096 | 16:47 |
| openstackgerrit | Rico Lin proposed openstack/releases master: Release heat ocata 8.0.5 https://review.openstack.org/515098 | 16:47 |
| dhellmann | armax : they're still using publish-to-pypi so setting the release-type in the releases repo won't work. project-config needs to be updated first. | 16:52 |
| *** claudiub|2 has quit IRC | 16:52 | |
| ttx | tag in | 16:55 |
| dhellmann | armax : do you think that difference may indicate that releases for those deliverables won't work? | 16:56 |
| dhellmann | smcginnis : do you have any idea why there's a "None" at the start of the paragraph in http://logs.openstack.org/09/515109/1/check/build-openstack-sphinx-docs/e25e93a/html/queens/highlights.html ? | 16:58 |
| armax | um, I see | 16:58 |
| smcginnis | dhellmann: Hmm, no I don't. When I did some test notes the output was fine. I'll look in to it. | 16:59 |
| dhellmann | the field is just a text blob, right? it doesn't need to be a list? | 17:00 |
| ttx | publish-static job ok | 17:00 |
| ttx | now watching release pipe | 17:00 |
| smcginnis | dhellmann: Yeah, should just be treated as a text blob. | 17:01 |
| dhellmann | ok | 17:01 |
| *** dtantsur is now known as dtantsur|afk | 17:01 | |
| dhellmann | I'll add some prints locally and see what I'm getting as rst output | 17:02 |
| smcginnis | dhellmann: Great, I'm stuck in a meeting at the moment, so may be a bit before I can get to looking. | 17:03 |
| openstackgerrit | Doug Hellmann proposed openstack/releases master: Add release highlights https://review.openstack.org/512443 | 17:10 |
| openstackgerrit | Doug Hellmann proposed openstack/releases master: add demo of series-highlights https://review.openstack.org/515109 | 17:10 |
| openstackgerrit | Doug Hellmann proposed openstack/releases master: fix default initial setting for accumulating notes https://review.openstack.org/515140 | 17:10 |
| dhellmann | smcginnis : there's the fix ^^ | 17:10 |
| smcginnis | Ahhh! Oops. Pretty obvious now. :) | 17:11 |
| ttx | pypi upload ok | 17:11 |
| EmilienM | probably not a good time to bring new topics but I'll try anyway - would it be possible one day to have a periodic job that runs every week to propose a bugfix release on stable branches for some projects? | 17:13 |
| ttx | dhellmann: I need to drop to dinner, the python-watcherclient thing announce-release job hasn'e started yet | 17:15 |
| ttx | the rest worked ok so far | 17:15 |
| dhellmann | ttx: ack. I'm stepping out for lunch. I will check the python-watcherclient jobs when I return | 17:16 |
| ttx | cheers | 17:16 |
| *** ricolin has quit IRC | 17:24 | |
| armax | dhellmann: let me look at how these projects are configured | 17:29 |
| armax | dhellmann: I don’t see a reason why these should differ from the others | 17:30 |
| armax | but it’s ok for now if they are | 17:30 |
| *** bobh has quit IRC | 17:30 | |
| *** ekcs has joined #openstack-release | 17:44 | |
| *** Rockyg has quit IRC | 17:49 | |
| dhellmann | armax : it's not really clear why any of the projects need neutron installed just to build an sdist. *shrug* | 17:49 |
| johnsom | So looking at the last validate failure for the octavia projects, I guess the pbr magic for version numbers doesn't work for horizon plugins? I need to manage a version number in package.json? | 17:53 |
| dhellmann | johnsom: link? | 17:54 |
| johnsom | We didn't do that before when releasing neutron-lbaas-dashboard (octavia-dashboard is a new release) | 17:54 |
| johnsom | dhellmann https://review.openstack.org/#/c/513072/ | 17:54 |
| johnsom | http://logs.openstack.org/72/513072/3/check/openstack-tox-validate/ffacc3e/job-output.txt.gz#_2017-10-25_17_20_20_108372 | 17:55 |
| dhellmann | johnsom : there have been some changes in this area of the validation recently | 17:55 |
| dhellmann | are you actually packaging a nodejs project out of that repo? | 17:55 |
| johnsom | Yep, just trying to figure out what I need to do to help | 17:55 |
| dhellmann | ok. we don't have a way to automatically set the version inside the npm metadata files, so you have to do that by hand | 17:56 |
| johnsom | We have only been doing pypi not nodejs historically | 17:56 |
| dhellmann | so, land a patch in your repo to update that version, then tag that commit | 17:56 |
| johnsom | Ok | 17:57 |
| dhellmann | the validation can tell they're different, but the release build job for node doesn't update the file | 17:57 |
| dhellmann | we could make it do that, if you want to work on that | 17:57 |
| *** electrofelix has quit IRC | 17:57 | |
| dhellmann | oh, wait, no, I think we said we wouldn't do that because it means other folks packaging from our repos won't get the same bits | 17:58 |
| dhellmann | they'll have to do the same dance | 17:58 |
| johnsom | I will put it on my list of when I get time. Pretty backlogged in the other zuul stuffs right now | 17:58 |
| johnsom | Ok, off the list.... grin | 17:58 |
| dhellmann | :-) | 17:59 |
| *** pkovar has quit IRC | 18:07 | |
| *** amoralej is now known as amoralej|off | 18:09 | |
| *** strigazi_ has joined #openstack-release | 18:13 | |
| *** strigazi has quit IRC | 18:16 | |
| *** strigazi_ is now known as strigazi | 18:17 | |
| armax | dhellmann: that’s what the crux of the discussion between jeremy and me yesterday | 18:20 |
| armax | eventually we should be able to get rid of that I believe | 18:21 |
| *** kumarmn_ has joined #openstack-release | 18:32 | |
| *** kumarmn has quit IRC | 18:36 | |
| openstackgerrit | Michael Dong proposed openstack/releases master: Release syntribos 0.4.0 https://review.openstack.org/512446 | 18:37 |
| dhellmann | armax : ok, good, and then we can eliminate the custom release types | 18:39 |
| *** csomerville has quit IRC | 18:39 | |
| *** csomerville has joined #openstack-release | 18:39 | |
| openstackgerrit | Sean McGinnis proposed openstack/releases master: Add release highlights https://review.openstack.org/512443 | 18:41 |
| smcginnis | dhellmann: Think we're ready to try a stable branch release? | 18:41 |
| openstackgerrit | Merged openstack/releases master: Add variants of publish-to-pypi jobs https://review.openstack.org/514509 | 18:48 |
| *** wolverineav has quit IRC | 18:49 | |
| *** wolverineav has joined #openstack-release | 18:49 | |
| *** wolverineav has quit IRC | 18:50 | |
| *** wolverineav has joined #openstack-release | 18:50 | |
| smcginnis | Going to give the stable release a shot and see what happens. | 19:06 |
| *** armax has quit IRC | 19:08 | |
| dhellmann | smcginnis : yeah, if the signing thing is fixed let's give it a try | 19:14 |
| openstackgerrit | Merged openstack/releases master: nova: release pike 16.0.2 https://review.openstack.org/513496 | 19:18 |
| smcginnis | dhellmann: And here we go. ^ :) | 19:18 |
| smcginnis | Looked like everything we needed went through, so we'll see how this goes. | 19:19 |
| * dhellmann braces himself | 19:19 | |
| smcginnis | dhellmann: Were you running one of our commands to identify those deliverables that were not signed? | 19:20 |
| dhellmann | smcginnis : yes, missing-releases | 19:20 |
| dhellmann | tox -e venv -- missing-releases | 19:20 |
| smcginnis | dhellmann: Cool, didn't realize it did further validation. | 19:20 |
| dhellmann | it looks for the tarball and signature file | 19:20 |
| dhellmann | and I have a patch up to make it look for wheels, too | 19:20 |
| dhellmann | that may have merged? | 19:20 |
| smcginnis | Oh, yeah. I think that may have. | 19:21 |
| *** nikhil has joined #openstack-release | 19:21 | |
| dhellmann | I have a few other tool updates that haven't so I lost track | 19:21 |
| smcginnis | I've been so scattered with distractions lately, I've lost track of a lot of things. Then found them, then lost them again... | 19:22 |
| smcginnis | Hmm, the release job for nova doesn't appear to be running anything. | 19:22 |
| dhellmann | it can take zuul a few minutes to figure out which jobs to run | 19:23 |
| smcginnis | This last week or so has certainly been a good exercise in patience. | 19:24 |
| dhellmann | seriously | 19:24 |
| * smcginnis watches release-openstack-python-without-pypi stream... | 19:27 | |
| smcginnis | Some venv error message in there, but doesn't appear to be an actual error. | 19:30 |
| smcginnis | sign-artifacts passed | 19:30 |
| smcginnis | Normal logstash error message, but appears to have all worked otherwise. | 19:31 |
| smcginnis | dhellmann: That wheel check does not pay attention to whether it actually is something published to pypi apparently. But looks happy. | 19:34 |
| *** iyamahat has quit IRC | 19:35 | |
| smcginnis | Oh, just realized that's just the difference between the python 2 check and 2/3. | 19:37 |
| smcginnis | So just ignore me. :) | 19:37 |
| smcginnis | And announce email successful to the ML. | 19:38 |
| smcginnis | I'm afraid to try the newton ansible release. | 19:39 |
| smcginnis | Ocata ironic-python-agent on its way. | 19:39 |
| *** armax has joined #openstack-release | 19:42 | |
| *** pbandark has quit IRC | 19:42 | |
| openstackgerrit | Merged openstack/releases master: Release ironic-python-agent 2.0.1 for Ocata https://review.openstack.org/514382 | 19:49 |
| *** coolsvap has quit IRC | 19:51 | |
| smcginnis | dhellmann, ttx, dims, fungi, tonyb: Draft countdown email ready - https://etherpad.openstack.org/p/queens-email-r17 | 19:54 |
| smcginnis | Will send tomorrow morning my time. | 19:54 |
| dhellmann | smcginnis : oh, yeah, it only looks for wheels on our infra right now | 19:56 |
| smcginnis | dhellmann: Yeah, that's good. Just got thrown off by the "did not find python 2 wheel" message and of course jumped too quickly to conclusions. | 19:57 |
| dhellmann | smcginnis : email looks good | 19:57 |
| * smcginnis needs to slow down to speed up | 19:57 | |
| dhellmann | oh, right, the final output of that is meant to be reviewed the rest is just progress messages | 19:57 |
| smcginnis | ++ I like it. | 19:57 |
| openstackgerrit | Michael Johnson proposed openstack/releases master: Octavia projects Queens MS1 release https://review.openstack.org/513072 | 20:00 |
| smcginnis | dhellmann: Did we ever figure out why these jobs still do the fallback bindep installs? | 20:01 |
| smcginnis | Would be nice to not install the world. | 20:01 |
| dhellmann | there is no bindep.txt file | 20:01 |
| *** iyamahat has joined #openstack-release | 20:02 | |
| dhellmann | which job is still doing that? | 20:02 |
| smcginnis | dhellmann: In which repo? | 20:02 |
| dhellmann | depends on the job | 20:02 |
| smcginnis | The current release-openstack-python job for ironic-python-agent. | 20:02 |
| dhellmann | does ironic-python-agent have a bindep.txt? | 20:02 |
| smcginnis | Yeah, no bindep in that repo. | 20:02 |
| dhellmann | I do see one in master | 20:03 |
| dhellmann | maybe there wasn't one at the point it was being tagged | 20:03 |
| dhellmann | it dates from the 12th so maybe | 20:03 |
| smcginnis | dhellmann: Yep, this is on Ocata. No file there. | 20:04 |
| smcginnis | Guess it's just an issue that will go away over time. | 20:04 |
| openstackgerrit | Merged openstack/releases master: Release syntribos 0.4.0 https://review.openstack.org/512446 | 20:05 |
| *** edmondsw_ has joined #openstack-release | 20:06 | |
| *** edmondsw has quit IRC | 20:09 | |
| smcginnis | No wheels for ironic-python-agent. | 20:09 |
| smcginnis | But there is a whl published to pypi at least. | 20:10 |
| smcginnis | Oh, I see. It's not converting the dashes to underscores, so it doesn't find them. | 20:13 |
| dhellmann | smcginnis : oh, yeah | 20:16 |
| openstackgerrit | Sean McGinnis proposed openstack/releases master: Convert dashes to underscores for wheel checks https://review.openstack.org/515195 | 20:18 |
| smcginnis | Oops, I missed a couple there. One moment. | 20:18 |
| openstackgerrit | Sean McGinnis proposed openstack/releases master: Convert dashes to underscores for wheel checks https://review.openstack.org/515195 | 20:19 |
| dhellmann | smcginnis : project isn't a string, is it? | 20:19 |
| dhellmann | should be a dict | 20:19 |
| dhellmann | I think you want to be changing the functions in links.py | 20:20 |
| smcginnis | dhellmann: Doh, yep. | 20:20 |
| dhellmann | johnsom : it looks like the nodejs version number strings don't match the versions in the release repo | 20:24 |
| johnsom | Yeah, so I just found that the node stuff uses a custom semver | 20:24 |
| openstackgerrit | Sean McGinnis proposed openstack/releases master: onvert dashes to underscores for wheel checks https://review.openstack.org/515195 | 20:25 |
| johnsom | It won't take our 1.0.0.0b1.... | 20:25 |
| dhellmann | johnsom : oh, that hasn't been an issue for us in the past | 20:25 |
| dhellmann | I wonder if anyone else is using milestone versioning with node projects | 20:25 |
| johnsom | http://logs.openstack.org/66/515166/1/check/octavia-dashboard-nodejs4-npm-run-lint/1d19938/job-output.txt.gz#_2017-10-25_18_49_09_148094 | 20:25 |
| *** thorst has quit IRC | 20:27 | |
| johnsom | We are having our weekly irc meeting. Can I come back in a bit and discuss options? | 20:27 |
| dhellmann | johnsom : of course | 20:27 |
| johnsom | Thanks | 20:27 |
| dhellmann | smcginnis : heads up about a versioning issue ^^ | 20:32 |
| smcginnis | npm doesn't like our milestone version format? | 20:33 |
| dhellmann | yeah, too many digits probably | 20:33 |
| smcginnis | Wouldn't that be considered a lib that doesn't use that format? We don't do that for the python-$projectclient or non-client libs. | 20:34 |
| openstackgerrit | Sean McGinnis proposed openstack/releases master: Convert dashes to underscores for wheel checks https://review.openstack.org/515195 | 20:35 |
| dhellmann | smcginnis : that's a good point. switching to cycle-with-intermediary should fix the conflict, but I see several of the other dashboard projects use milestone versioning, too | 20:37 |
| dhellmann | I feel like we had a conversation about those projects last cycle, but I don't remember the outcome. maybe ttx or dims do. | 20:37 |
| smcginnis | Don't think I as around for that conversation. At least not ringing any bells. | 20:38 |
| smcginnis | s/as/was/ | 20:38 |
| dhellmann | yeah, I think it was early in the cycle | 20:38 |
| openstackgerrit | Merged openstack/releases master: Release Watcher Queens-1 https://review.openstack.org/513451 | 20:43 |
| *** thorst has joined #openstack-release | 20:48 | |
| *** thorst has quit IRC | 20:48 | |
| *** thorst has joined #openstack-release | 20:49 | |
| *** thorst has quit IRC | 20:53 | |
| *** edmondsw_ is now known as edmondsw | 20:55 | |
| *** sree has joined #openstack-release | 20:58 | |
| *** rosmaita has quit IRC | 21:00 | |
| *** sree has quit IRC | 21:03 | |
| openstackgerrit | Merged openstack/releases master: Add Barbican Queens Milestone 1 https://review.openstack.org/513571 | 21:06 |
| johnsom | dhellmann Ok, back. We switched to cycle with milestones to help the translations... | 21:16 |
| * dhellmann hangs his head | 21:16 | |
| dhellmann | how does that help with translations? | 21:16 |
| johnsom | They wanted/needed RCs to get translation updates in | 21:17 |
| johnsom | Plus the whole packaging thing. | 21:17 |
| dhellmann | I'm not sure what you mean by the packaging thing? | 21:18 |
| johnsom | Sure you could do a bunch of intermediaries, one for string freeze, etc. but I thought that was kind of the point with the milestones | 21:18 |
| dhellmann | is there a technical reason that release candidates help? | 21:18 |
| dhellmann | sure it is. | 21:19 |
| dhellmann | we're up against a set of conflicting rules about versioning and I'm just trying to understand the whole picture so we can figure out the right way to address it | 21:19 |
| johnsom | Packaging thing is dropping RCs so the packagers can do testing of the RCs to have packages ready for the release. | 21:19 |
| johnsom | Yeah, I hear you. I have no idea why the node folks intentionally went custom semver. | 21:20 |
| dhellmann | ok. that doesn't seem to stop them with the projects that don't use release candidates, so I'm not too worried about that one. | 21:20 |
| dhellmann | yeah, and we're doing python semver, so... | 21:20 |
| dhellmann | I mean, technically pre-release versions aren't semver, right? | 21:20 |
| dhellmann | this probably needs to have an email thread, because we don't have everyone here who has info to contribute | 21:21 |
| dhellmann | could you start that thread and tag it release and i18n so those teams both see it? | 21:21 |
| dhellmann | the only "obvious" solution is to stop using pre-release versions but there are issues with that so I don't want to say that's the *only* solution | 21:22 |
| johnsom | https://docs.npmjs.com/misc/semver#prerelease-tags | 21:22 |
| johnsom | Sorry, I am new to all this npm stuff | 21:23 |
| dhellmann | me, too | 21:23 |
| dhellmann | it would be useful to compare what some of the other npm projects are doing. some of the dashboards I looked at that had pre-release versions don't seem to be npm-based | 21:23 |
| dhellmann | alternatively we could update all of the release machinery to allow those sorts of version numbers | 21:24 |
| dhellmann | and just go with npm versioning | 21:24 |
| johnsom | Yeah, horizon is shifting over. I am not sure if any of the other plugins are using angular yet. | 21:24 |
| dhellmann | ok, well, there's enough work to do there that I think we'd need to get some help from the folks using the tools | 21:25 |
| smcginnis | Allowing what is the norm for npm prerelease tags could be good. | 21:25 |
| smcginnis | But yeah, I have 0 experience with that realm. | 21:25 |
| dhellmann | I can advise on what needs to be updated, but I'm not going to be able to do it all myself. | 21:25 |
| dhellmann | yeah, I agree, smcginnis | 21:25 |
| dhellmann | I believe we have hook points in place to let us do it, the variations just need to be implemented | 21:26 |
| smcginnis | Or we can start a discussion to use that format for all our stuff. | 21:27 |
| smcginnis | Looking at the docs, I actually find it more readable. | 21:27 |
| dhellmann | there are python standards in play, too | 21:27 |
| dhellmann | someone would have to verify that the npm and python rules can be made to work together | 21:27 |
| dhellmann | and then update pbr | 21:27 |
| johnsom | Yeah, one option is to go back to how it worked in pike. If it's python packaging, ignore that version field in the package.json. Care if it is packaging for the npm ecosystem. | 21:29 |
| dhellmann | we started paying attention to that field because we ended up tagging releases where the tag and contents didn't match | 21:30 |
| johnsom | Fair enough | 21:30 |
| dhellmann | if you're using python packaging and not npm packaging, then we can ignore the field. but I thought this was being packaged for npm? | 21:30 |
| fungi | skimmed scrollback, looks like everything's a-okay for release automation now as far as we know? | 21:30 |
| dhellmann | fungi : I think so. smcginnis hasn't reported any issues, and I've been doing some required compliance certification so I haven't approved any releases. | 21:31 |
| johnsom | dhellmann To date, this has only been ending up on pypi. | 21:32 |
| *** rosmaita has joined #openstack-release | 21:33 | |
| dhellmann | johnsom : ok, that may make it easier to move ahead then. let me think about how that would need to work. I need to finish this compliance thing before the course times out so it'll be tomorrow morning before I get to it. | 21:34 |
| smcginnis | fungi: So far so good. | 21:34 |
| johnsom | It's not a pure nodejs/angular code base, it's python backed. | 21:34 |
| smcginnis | fungi: Haven't tried some of them like horizon plugins or the scary ansible releases. | 21:34 |
| johnsom | dhellmann Ok, good luck. (did once of my own this morning) | 21:34 |
| fungi | smcginnis: okay, i'm not going anywhere else between now and summit travel so around to help iron out any other issues as they crop up | 21:35 |
| smcginnis | fungi: Cool, thanks. Maybe I'll get up enough nerve to try one of these other ones soon. ;) | 21:35 |
| smcginnis | Think I'll send horizon through. | 21:37 |
| * fungi braces for impact | 21:37 | |
| johnsom | Yeah, why isn't horizon having the same issue??? https://github.com/openstack/horizon/blob/master/package.json#L2 | 21:38 |
| dhellmann | johnsom : the validation rules changed in the past day or two. maybe before the horizon patch went through the check queue? | 21:43 |
| dhellmann | ok, that took me less time than expected | 21:43 |
| dhellmann | let me see about this validation bypass | 21:43 |
| smcginnis | dhellmann: Validate just failed for horizon. | 21:43 |
| dhellmann | yep | 21:44 |
| *** iyamahat has quit IRC | 21:45 | |
| *** sree has joined #openstack-release | 21:45 | |
| johnsom | https://www.irccloud.com/pastebin/COicHV9A/ | 21:49 |
| smcginnis | So looks like we'll have to figure out versioning with that one. I'm going to just move along for now to keep some of these things moving. | 21:49 |
| *** sree has quit IRC | 21:50 | |
| dhellmann | smcginnis : yeah, keep going, I'm working on the versioning | 21:51 |
| *** edmondsw has quit IRC | 21:51 | |
| smcginnis | Perfect, thanks dhellmann | 21:51 |
| johnsom | Let me know if there is something I can do to help | 21:52 |
| *** wolverineav has quit IRC | 21:52 | |
| *** wolverineav has joined #openstack-release | 21:52 | |
| *** edmondsw has joined #openstack-release | 21:56 | |
| *** sdague has quit IRC | 21:56 | |
| *** wolverineav has quit IRC | 21:57 | |
| openstackgerrit | Merged openstack/releases master: Release ceilometer-powervm Queens-1 https://review.openstack.org/513482 | 21:57 |
| kota_ | would someone review https://review.openstack.org/#/c/513325/ ? that is for queen-1 of Storlets proejct | 21:58 |
| dhellmann | kota_ : http://lists.openstack.org/pipermail/openstack-dev/2017-October/123952.html | 22:00 |
| kota_ | dhellmann: i see.thx. I thought it was resolved because gerrit bot reported some patchges got merged recently. | 22:01 |
| *** kzaitsev_pi has quit IRC | 22:02 | |
| *** kumarmn_ has quit IRC | 22:03 | |
| dhellmann | kota_ : we're taking them slowly and monitoring the jobs closely. Some are working, but we are still finding issues. | 22:03 |
| *** kzaitsev_pi has joined #openstack-release | 22:04 | |
| *** edmondsw has quit IRC | 22:04 | |
| openstackgerrit | Doug Hellmann proposed openstack/releases master: Neutron: release Queens M-1 milestones for neutron projects https://review.openstack.org/512432 | 22:04 |
| openstackgerrit | Doug Hellmann proposed openstack/releases master: add explicit release types for neutron and horizon plugins https://review.openstack.org/515119 | 22:04 |
| openstackgerrit | Doug Hellmann proposed openstack/releases master: set release types for ocatavia and neutron-lbaas https://review.openstack.org/515122 | 22:04 |
| openstackgerrit | Doug Hellmann proposed openstack/releases master: Octavia projects Queens MS1 release https://review.openstack.org/513072 | 22:04 |
| openstackgerrit | Doug Hellmann proposed openstack/releases master: apply language-specific version rules more carefully https://review.openstack.org/515217 | 22:04 |
| dhellmann | smcginnis : that series may have rebased more than necessary and made some of the other releases depend on old versions of the patches :-/ | 22:05 |
| kota_ | dhellmann: ok | 22:05 |
| dhellmann | maybe we should take neutron next so we can get some of these others landed | 22:05 |
| dhellmann | johnsom : ^^ see 515217 | 22:05 |
| johnsom | Lookiong | 22:06 |
| smcginnis | dhellmann: Neutron would be good. I have to head out for a class this evening, but I could kick it off and check on the results later. | 22:06 |
| dhellmann | smcginnis : what's the queue delay looking like right now? | 22:07 |
| smcginnis | dhellmann: Release patches seem to be going through fairly quick. | 22:07 |
| dhellmann | ok. I'll approve neutron but leave the others that have code changes for someone else | 22:07 |
| smcginnis | ++ | 22:08 |
| ying_zuo | Hello. Is there anything that I need to fix for the horizon queens-1 release patch? | 22:08 |
| ying_zuo | https://review.openstack.org/#/c/513432/ | 22:08 |
| dhellmann | ying_zuo : we're working on that right now | 22:09 |
| *** kumarmn has joined #openstack-release | 22:10 | |
| ying_zuo | dhellmann: thanks. please let me know if there's anything that I can help | 22:11 |
| *** kzaitsev_pi has quit IRC | 22:11 | |
| openstackgerrit | Doug Hellmann proposed openstack/releases master: Release Horizon Queens-1 13.0.0.0b1 https://review.openstack.org/513432 | 22:12 |
| openstackgerrit | Doug Hellmann proposed openstack/releases master: set horizon release-type explicitly https://review.openstack.org/515220 | 22:12 |
| dhellmann | ying_zuo : ^^ this should fix it | 22:13 |
| *** kzaitsev_pi has joined #openstack-release | 22:13 | |
| ying_zuo | awesome! thanks | 22:13 |
| johnsom | dhellmann That works for me | 22:14 |
| *** kumarmn has quit IRC | 22:15 | |
| openstackgerrit | Doug Hellmann proposed openstack/releases master: Release designate-dashboard 6.0.0.0b1 https://review.openstack.org/514800 | 22:17 |
| openstackgerrit | Doug Hellmann proposed openstack/releases master: set designate-dashboard release type explicitly https://review.openstack.org/515221 | 22:17 |
| dhellmann | fungi, smcginnis : some of the list-changes jobs are failing with a weird error. Both heat-agents and heat are affected so far. http://logs.openstack.org/83/515083/2/check/releases-tox-list-changes/3f40e2a/job-output.txt.gz#_2017-10-25_16_31_24_671153 | 22:19 |
| dhellmann | they don't run list-changes, but they don't actually fail either | 22:20 |
| *** iyamahat has joined #openstack-release | 22:20 | |
| openstackgerrit | Merged openstack/releases master: Neutron: release Queens M-1 milestones for neutron projects https://review.openstack.org/512432 | 22:21 |
| dhellmann | the heat client job worked | 22:22 |
| dhellmann | the thing that concerns me is not so much that the command failed but that the job didn't report the failure | 22:23 |
| fungi | dhellmann: i'll dig into that real quick and see if i can make heads or tails of it | 22:27 |
| smcginnis | That's the venv error I noticed earlier. | 22:27 |
| dhellmann | fungi : thanks. I've rechecked the patches I found with the problem so we'll have a second data point on those soon. | 22:27 |
| dhellmann | oh, I missed that, do you have more details? | 22:27 |
| smcginnis | No, just that I noticed it happening in an earlier job but it seemed to be able to get by it. | 22:28 |
| *** iyamahat has quit IRC | 22:28 | |
| smcginnis | Does that git config command look right? Or is that just the weird log formatting? | 22:28 |
| *** edmondsw has joined #openstack-release | 22:28 | |
| dhellmann | I can't tell | 22:30 |
| *** pbandark has joined #openstack-release | 22:31 | |
| *** pbandark has quit IRC | 22:33 | |
| *** edmondsw has quit IRC | 22:34 | |
| *** thorst has joined #openstack-release | 22:35 | |
| *** thorst has quit IRC | 22:42 | |
| *** pbandark has joined #openstack-release | 22:42 | |
| *** thorst has joined #openstack-release | 22:43 | |
| dhellmann | smcginnis, fungi : I think the error is coming out of pip itself. I see the --get-regexp option in the command line within the pip source code | 22:46 |
| fungi | dhellmann: oh, now that i look at the error you linked, yeah that's (theoretically benign?) fallout from installing from git repos on disk which have no "origin" remote. pip attempts to annotate git installs with git origin urls, but cannot in this case (you'll also see corresponding telltale comment lines before each affected package in pip freeze output) | 22:46 |
| *** claudiub|2 has joined #openstack-release | 22:46 | |
| *** thorst has quit IRC | 22:46 | |
| dhellmann | so what's making tox skip running the tests? | 22:46 |
| fungi | that's a very good question | 22:47 |
| dhellmann | oh, this is tox-siblings again | 22:47 |
| dhellmann | bah | 22:47 |
| dhellmann | tox-siblings runs tox --notest | 22:47 |
| dhellmann | then *later* tox is run again | 22:48 |
| dhellmann | and at that point there is actual output | 22:48 |
| fungi | "tox-siblings : Run tox without tests" | 22:48 |
| dhellmann | right | 22:48 |
| fungi | yeah, the task name should have been a giveaway i guess, but i didn't spot it either | 22:48 |
| dhellmann | my human pattern matching stopped on that ____summary____ line thinking that was the interesting tox run | 22:48 |
| *** wolverineav has joined #openstack-release | 22:48 | |
| fungi | once we get ara turned back on for successful job runs, i hope it will make some of this easier (or once we get os-loganalyze smarter about being able to auto-collapse pre/post phase tasks) | 22:49 |
| dhellmann | if I wanted the list-changes job to write the output of list-changes to its own file, where would I have it write that file to make the log copying stuff include it when copying files over to logs.o.o? | 22:49 |
| dhellmann | I find the ara output requires way too many clicks to get to anything useful | 22:49 |
| fungi | easy solution would be to write it into the tox logdir since we grab everything from there in tox-based jobs | 22:50 |
| dhellmann | though I can never remember what the little icons mean or where I click to get useful data so maybe that's why | 22:50 |
| dhellmann | ok | 22:50 |
| dhellmann | oh, I'll bet tox doesn't support shell redirection like tee, though | 22:50 |
| dhellmann | meh, I can have it run a script | 22:50 |
| fungi | dhellmann: yeah, i've finally used ara often enough to remember that i want to expand the tasks list for the playbook of interest, then click on the result column for the task i want to see output from | 22:51 |
| fungi | so maybe two clicks, but much less scrolling | 22:51 |
| fungi | except, for now, we only collect ara bits from failed jobs, until we get the database-backed wsgi frontend for it going on logs.o.o so we don't keep too many tiny json files for every job | 22:52 |
| *** hongbin has quit IRC | 23:05 | |
| openstackgerrit | Merged openstack-infra/release-tools master: Add some notes on eol_branch.sh https://review.openstack.org/514846 | 23:06 |
| openstackgerrit | Doug Hellmann proposed openstack/releases master: log the output of list-changes to a separate file for easier reading https://review.openstack.org/515227 | 23:09 |
| dhellmann | fungi : ^^ | 23:10 |
| dhellmann | and with that I'm going to call it a night | 23:10 |
| dhellmann | smcginnis , ttx : I didn't notice any of those neutron tag jobs fail while they were running but it looks like we're missing some artifacts. I don't know if that's because of the naming or if they're really not there. http://paste.openstack.org/show/624645/ If one of you hasn't looked at it before me, I'll check in the morning. | 23:12 |
| *** yamahata has joined #openstack-release | 23:27 | |
| fungi | dhellmann: i believe that change should be self-testing too (yay zuulv3!) | 23:29 |
| fungi | dhellmann: yep! http://logs.openstack.org/27/515227/1/check/releases-tox-list-changes/6b0f961/tox/list-changes-results.log | 23:31 |
| fungi | _this_ is the main reason i was so excited for v3, even with all the pain | 23:31 |
| *** yamahata has quit IRC | 23:32 | |
| *** kumarmn has joined #openstack-release | 23:32 | |
| *** kumarmn has quit IRC | 23:37 | |
| *** mriedem is now known as mriedem_away | 23:44 | |
| *** mriedem_away has quit IRC | 23:44 | |
| *** kumarmn has joined #openstack-release | 23:45 | |
| *** thorst has joined #openstack-release | 23:45 | |
| *** thorst has quit IRC | 23:46 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!