| *** packet has quit IRC | 00:00 | |
| hashar | the way I worked around the issue I described was to disconnect/reconnect the gearman client | 00:01 |
|---|---|---|
| hashar | (in jenkins) | 00:01 |
| *** bradjones has quit IRC | 00:01 | |
| hashar | then newly proposed changes would be processed | 00:01 |
| hashar | and to unlock the queues I used 'zuul promote' (but that retrigger everything :/ ) | 00:01 |
| anteaya | oooh | 00:02 |
| *** bradjones has joined #openstack-infra | 00:03 | |
| anteaya | the graphs for zm5 and zm8 are way higher for eth1 traffic compared to the rest of the zm servers | 00:08 |
| anteaya | no idea if that is useful, probably not | 00:08 |
| *** zz_sabari is now known as sabari | 00:09 | |
| *** baoli has quit IRC | 00:09 | |
| *** ChuckC has joined #openstack-infra | 00:11 | |
| *** sarob has joined #openstack-infra | 00:11 | |
| *** stevemar has quit IRC | 00:12 | |
| *** stevemar has joined #openstack-infra | 00:13 | |
| *** Rockyg has quit IRC | 00:14 | |
| *** cinerama has quit IRC | 00:15 | |
| *** SotK has quit IRC | 00:16 | |
| *** SotK has joined #openstack-infra | 00:16 | |
| *** cinerama has joined #openstack-infra | 00:17 | |
| anteaya | asselin: while you are at it can you propose a change against the spec, noting those two modules don't currently exist and should be removed from the list in the spec? | 00:20 |
| asselin | anteaya, I can do that | 00:20 |
| anteaya | asselin: thanks | 00:23 |
| *** dimtruck is now known as zz_dimtruck | 00:25 | |
| *** zz_dimtruck is now known as dimtruck | 00:26 | |
| *** hashar has quit IRC | 00:27 | |
| jeblair | clarkb: my plan is to run this until we see the error again, or as long as i can stand it, then restart geard with all the outstanding changes | 00:30 |
| *** talluri has joined #openstack-infra | 00:30 | |
| *** chuckC_ has quit IRC | 00:31 | |
| clarkb | jeblair: including mine? | 00:31 |
| *** pc_m has quit IRC | 00:32 | |
| *** dimtruck is now known as zz_dimtruck | 00:33 | |
| jeblair | clarkb: yeah | 00:34 |
| *** hdd has joined #openstack-infra | 00:34 | |
| pleia2 | nibalizer, jeblair, et al - they've renamed their tool to git-codereview https://go-review.googlesource.com/#/c/1741/ | 00:37 |
| fungi | wowiewowwow | 00:37 |
| fungi | that was fast | 00:37 |
| * pleia2 bows | 00:37 | |
| pleia2 | hehe | 00:37 |
| anteaya | pleia2: way to go | 00:38 |
| nibalizer | whahhhhhhht | 00:38 |
| jeblair | pleia2: any chance they want to help with ours? or is it not written in go enough? | 00:38 |
| anteaya | you force for change you | 00:38 |
| nibalizer | pleia2: is a superwoman | 00:38 |
| fungi | i bet you convinced them that _their_ dev community would be too confused by it | 00:38 |
| pleia2 | fungi: yep, that was my strategy | 00:38 |
| anteaya | very effective | 00:38 |
| *** Ryan_Lane has quit IRC | 00:39 | |
| pleia2 | jeblair: I doubt it, them seem to be happy to do it all in go | 00:39 |
| jeblair | we should submit patches to their repos to add .gitreview files | 00:39 |
| fungi | bwahaha | 00:40 |
| *** Ryan_Lane has joined #openstack-infra | 00:40 | |
| *** dangers is now known as dangers_away | 00:42 | |
| *** Masahiro has joined #openstack-infra | 00:43 | |
| anteaya | it took them all of 9 minutes to change... once they decided to | 00:43 |
| *** ChuckC has quit IRC | 00:44 | |
| clarkb | jeblair: its looking like zuul is near clearing out the current set of work. May be a good time to restart geard when it enters this sort of state | 00:44 |
| *** r-daneel has quit IRC | 00:46 | |
| *** hdd has quit IRC | 00:47 | |
| *** Masahiro has quit IRC | 00:47 | |
| *** yamamoto has joined #openstack-infra | 00:50 | |
| *** derekh has joined #openstack-infra | 00:52 | |
| clarkb | jeblair: did my latest gear patchset hold up under your load testing? or did you onl check it with admin status command? | 00:53 |
| clarkb | jeblair: mostly curious if you have any additional feedback | 00:53 |
| *** Masahiro has joined #openstack-infra | 00:54 | |
| *** radez is now known as radez_g0n3 | 00:54 | |
| openstackgerrit | Ramy Asselin proposed openstack-infra/infra-specs: Remove 2 modules that won't be split out https://review.openstack.org/142627 | 00:57 |
| *** salv-orlando has quit IRC | 00:57 | |
| openstackgerrit | Merged openstack/requirements: Set alembic>=0.7.1 https://review.openstack.org/140275 | 00:57 |
| *** yaguang has joined #openstack-infra | 01:01 | |
| *** esker has joined #openstack-infra | 01:02 | |
| openstackgerrit | Merged openstack-infra/elastic-recheck: Add query for nova unit test race bug 1403564 https://review.openstack.org/142485 | 01:02 |
| uvirtbot | Launchpad bug 1403564 in nova "nova.tests.unit.conductor.test_conductor._BaseTaskTestCase.test_build_instances_scheduler_failure is racy" [Medium,In progress] https://launchpad.net/bugs/1403564 | 01:03 |
| openstackgerrit | Merged openstack-infra/gear: Only wake relevant connections https://review.openstack.org/142496 | 01:03 |
| openstackgerrit | Merged openstack-infra/subunit2sql: Add --average option to sql2subunit cli https://review.openstack.org/132119 | 01:03 |
| *** unicell has joined #openstack-infra | 01:03 | |
| *** rmcall has quit IRC | 01:03 | |
| *** otter768 has joined #openstack-infra | 01:04 | |
| *** jerryz has joined #openstack-infra | 01:04 | |
| clarkb | dstufft question for you | 01:06 |
| dstufft | ok | 01:06 |
| clarkb | dstufft: if you `git clone pbr && cd pbr && git checkout master && virtualenv venv && venv/bin/python setup.py sdist` it explodes | 01:07 |
| clarkb | dstufft: the egg_info.writer does not exist on pbr master | 01:07 |
| clarkb | but for some reason distutils is looking for that. So I guess my question is "is the egg_info.writers entry point 'global' and not encapsulated by the virtualenv?" | 01:08 |
| dstufft | it should be encaspulated by the virtualenv | 01:08 |
| dstufft | moment | 01:08 |
| *** otter768 has quit IRC | 01:08 | |
| *** ZZelle_ has quit IRC | 01:08 | |
| clarkb | the othe thing I tried was maybe pbr grabs the latest version of itself to install itself (ugh) and since latest enters the entrypoint but local code doesn't have the target it fails. But I made a new virtualenv and installed pbr from pypi at version 0.10.0 which does not do the entrypoint and tried again. but it still failed | 01:09 |
| dstufft | https://bpaste.net/show/f4ad0040555f | 01:10 |
| *** zz_avozza is now known as avozza | 01:10 | |
| openstackgerrit | Merged openstack-infra/elastic-recheck: Add query for bug 1403510 affecting docutils/py33/py34 https://review.openstack.org/142513 | 01:11 |
| uvirtbot | Launchpad bug 1403510 in os-apply-config "gate tests failed with "NameError: name 'StandardError' is not defined"" [High,Confirmed] https://launchpad.net/bugs/1403510 | 01:11 |
| clarkb | dstufft: huh it worked for you | 01:11 |
| *** MarkAtwood has quit IRC | 01:11 | |
| dstufft | clarkb: if you installed a copy of pbr into that virtualenv and you were in the root of the repo it would fail as you said | 01:11 |
| *** sarob has quit IRC | 01:12 | |
| clarkb | ok thats what I did | 01:12 |
| dstufft | so my guess is you either triggered an install of pbr from pypi somehow, or you're running an old virtualenv which doesn't have --no-site-packages by default | 01:12 |
| *** ChuckC has joined #openstack-infra | 01:13 | |
| clarkb | hrm I don't have pbr globally installed | 01:13 |
| clarkb | let me fiddle a bit more | 01:13 |
| *** chuckC_ has joined #openstack-infra | 01:13 | |
| *** Sukhdev has quit IRC | 01:14 | |
| *** yamamoto has quit IRC | 01:14 | |
| *** shashankhegde has quit IRC | 01:14 | |
| clarkb | huh ok I got it to work. I think it didn't like my venv in the same dir as pbr | 01:15 |
| clarkb | dstufft: thank you for the sanity check | 01:15 |
| *** esker has quit IRC | 01:16 | |
| *** esker has joined #openstack-infra | 01:16 | |
| *** derekh has quit IRC | 01:16 | |
| dstufft | clarkb: no problem | 01:16 |
| dstufft | clarkb: I would probably suggest getting the egg writer stuff on master sooner rather than later, because it wouldn't surprise me if more than one person ran into an error like that | 01:17 |
| clarkb | dstufft: oh ya, tripleo hit it :) | 01:17 |
| clarkb | dstufft: so I think that is my next step once we get the current in flight changes happy | 01:17 |
| *** clif_h has joined #openstack-infra | 01:17 | |
| clif_h | is py33 gate testing broken atm? | 01:18 |
| clarkb | clif_h: yes | 01:18 |
| clif_h | clarkb: ty | 01:18 |
| dstufft | the fix is somewhere on the queue | 01:18 |
| dstufft | heh | 01:18 |
| *** sabari is now known as zz_sabari | 01:18 | |
| dstufft | merge-check queue is massive lol | 01:19 |
| anteaya | clif_h: go to http://status.openstack.org/zuul/ | 01:19 |
| anteaya | you will see all of our testing queues | 01:19 |
| clarkb | dstufft: its currently 666 for me :) | 01:19 |
| dstufft | me to | 01:20 |
| dstufft | evil queue | 01:20 |
| anteaya | I too see 666 | 01:20 |
| clarkb | it is a sign | 01:20 |
| jeblair | clarkb: drat, i was afk and missed the bottoming out of the queue | 01:20 |
| *** avozza is now known as zz_avozza | 01:20 | |
| clarkb | jeblair: oh well | 01:20 |
| dstufft | if something is in the gate queue, does that mean it's going to get merged assuming the gate passes? | 01:20 |
| jeblair | clarkb: i'm still not seeing jenkins01 erroring; perhaps the underlying network issue has cleared | 01:21 |
| clarkb | dstufft: yes (with the slight corner case where gerrit jgit may decide something can't merge even though C git said it could) | 01:21 |
| anteaya | dstufft: yes, it should be what it means | 01:21 |
| dstufft | so the fix for py33 is in the gate queue then | 01:21 |
| clarkb | dstufft: but our detection of that case is pretty good now and we fail fast on it | 01:21 |
| *** talluri has quit IRC | 01:21 | |
| dstufft | as is the .postN version | 01:21 |
| clarkb | dstufft: yup and once those three changes merge we will tag a new release | 01:22 |
| clarkb | jeblair: in fact maybe you can promote those three pbr changes if/when you restart things | 01:22 |
| dstufft | oh and the git thing | 01:22 |
| dstufft | forgot about it | 01:22 |
| jeblair | clarkb: ack | 01:23 |
| dstufft | oh and my requirements change is almost through the check queue :yay: | 01:24 |
| openstackgerrit | Merged openstack-infra/gear: Fix accounting bug when peeking for job https://review.openstack.org/142604 | 01:24 |
| clarkb | dstufft: I think it has been approved too so should end up in the gate once those tests pass | 01:24 |
| dstufft | clarkb: you'd think i'd have the pipeline figured out by now :/ | 01:24 |
| jeblair | we are always changing it :) it's kinda dark city around here... | 01:25 |
| dstufft | success \o/ | 01:25 |
| *** SumitNaiksatam has quit IRC | 01:26 | |
| anteaya | ha ha ha | 01:27 |
| anteaya | and I thought it was just me who never knew what was going on | 01:27 |
| anteaya | :D | 01:27 |
| dstufft | hopefully the 9 hours in the gate key doesn't mean it's gonna take 9 hours to process all this | 01:28 |
| dstufft | or is that what it means | 01:28 |
| clarkb | dstufft: that 9 hours is how long it has been in there | 01:28 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/project-config: Add nodepool script to preseed testrepository from subunit2sql https://review.openstack.org/136234 | 01:28 |
| *** timcline has joined #openstack-infra | 01:28 | |
| dstufft | ah | 01:29 |
| *** timcline has quit IRC | 01:29 | |
| *** sdake has quit IRC | 01:29 | |
| *** zz_sabari is now known as sabari | 01:29 | |
| *** timcline has joined #openstack-infra | 01:29 | |
| *** bhunter71 has joined #openstack-infra | 01:31 | |
| *** enikanorov__ has quit IRC | 01:32 | |
| *** bradjones has quit IRC | 01:32 | |
| *** kumartin has quit IRC | 01:32 | |
| jeblair | restarting geard | 01:32 |
| clarkb | rgr | 01:33 |
| *** enikanorov has joined #openstack-infra | 01:33 | |
| *** sdake has joined #openstack-infra | 01:33 | |
| *** sdake has quit IRC | 01:33 | |
| *** sdake has joined #openstack-infra | 01:33 | |
| *** shayneburgess has quit IRC | 01:33 | |
| jeblair | clarkb: hrm, i think when we do this we may end up waiting out the full 300 second timeout before zuul reconnects. that could probably be improved. | 01:34 |
| clarkb | ya should get a fin and close the socket | 01:35 |
| jeblair | clarkb: yep. we just need to pass that through to the wait event that's waiting on the job handle to be returned | 01:35 |
| *** gyee has quit IRC | 01:37 | |
| *** ivar-lazzaro has quit IRC | 01:37 | |
| jeblair | there it goes | 01:37 |
| *** sabari is now known as zz_sabari | 01:38 | |
| openstackgerrit | Stefano Maffulli 'reed' proposed openstack-infra/system-config: Remove community.openstack.org from the list of server https://review.openstack.org/142544 | 01:38 |
| jeblair | clarkb: the cpu usage of geard is _way_ better than before | 01:38 |
| *** mtanino has quit IRC | 01:38 | |
| clarkb | nice | 01:41 |
| anteaya | which server is the geard server? | 01:41 |
| jeblair | anteaya: it runs on zuul.o.o | 01:42 |
| anteaya | ah | 01:42 |
| jeblair | anteaya: you should see cpu, load average, and possibly eth0 traffic drop on that host | 01:42 |
| anteaya | so far seeing cpu graph drop | 01:43 |
| anteaya | looks like this started around 1800 | 01:43 |
| anteaya | eth0 is less but still jumping around a bit | 01:45 |
| anteaya | oh eth0 out just spiked | 01:45 |
| jeblair | clarkb: i think geard may be stuck. i'm going to restart it without your patch, but save the log output | 01:47 |
| clarkb | jeblair: ok | 01:47 |
| *** wuhg has joined #openstack-infra | 01:47 | |
| openstackgerrit | Elizabeth K. Joseph proposed openstack-infra/puppet-zanata: Add a TODO and support for configuring Wildfly https://review.openstack.org/142645 | 01:50 |
| anteaya | current system cpu usage is less than average | 01:52 |
| anteaya | and eth0 out came back down | 01:53 |
| anteaya | eth0 out spiked again | 01:56 |
| *** zz_gondoi is now known as gondoi | 01:57 | |
| jyuso | anteaya: hi,i want to ask a question here:our CI isn't based on Zuul,so there is no sandbox jenkins project which is auto generated by Zuul in my server.do i still need to create a Jenkins project to test voting on sandbox? | 01:58 |
| *** salv-orlando has joined #openstack-infra | 01:58 | |
| *** dkranz has joined #openstack-infra | 01:59 | |
| asselin | jyuso, the sandbox project is the one hosted by openstack | 02:00 |
| anteaya | eth0 out is starting to drop | 02:01 |
| asselin | jyuso, https://github.com/openstack-dev/sandbox | 02:01 |
| fungi | clarkb: remember that dhellmann's patch adds . in the deps in tox.ini because something in the pbr test requirements is dragging in a release of pbr (circular dep) | 02:02 |
| clarkb | fungi: ya I think that is a similar situation | 02:02 |
| *** salv-orlando has quit IRC | 02:03 | |
| asselin | jyuso, you don't need zuul, there are other options. see here for an alternative: http://ci.openstack.org/third_party.html#the-jenkins-gerrit-trigger-plugin-way | 02:03 |
| clarkb | ok time to step out for dinner | 02:04 |
| anteaya | clarkb: enjoy dinner | 02:04 |
| openstackgerrit | Mauro Rodrigues proposed openstack-infra/elastic-recheck: Query for bug #1253896 https://review.openstack.org/142479 | 02:04 |
| uvirtbot | Launchpad bug 1253896 in tempest/havana "Attempts to verify guests are running via SSH fails. SSH connection to guest does not work." [Critical,Confirmed] https://launchpad.net/bugs/1253896 | 02:04 |
| anteaya | asselin: you're awesome | 02:05 |
| anteaya | system cpu load and eth0 are dropping | 02:06 |
| jyuso | asselin: Thanks.I'm using gerrit trigger on my server to trigger my local jenkins project:) | 02:06 |
| jeblair | clarkb: i think i found the problem with your patch; left comment. | 02:06 |
| anteaya | what is causing the eth0 out spikes when they happen? | 02:07 |
| asselin | jyuso, gotta go now. hopefully that helps you make progress. | 02:07 |
| *** markmcclain has joined #openstack-infra | 02:08 | |
| jyuso | asselin:OK,thanks. | 02:09 |
| *** timcline has quit IRC | 02:10 | |
| *** david-lyle is now known as david-lyle_afk | 02:10 | |
| *** Longgeek has quit IRC | 02:11 | |
| *** dkranz has quit IRC | 02:13 | |
| *** ddieterly has joined #openstack-infra | 02:16 | |
| dstufft | clarkb: fungi https://hg.python.org/peps/rev/93db88237edb | 02:17 |
| dstufft | not released in setuptools yet | 02:17 |
| *** wenlock has joined #openstack-infra | 02:17 | |
| clarkb | jeblair thank you | 02:17 |
| *** zz_avozza is now known as avozza | 02:18 | |
| *** erikwilson has joined #openstack-infra | 02:18 | |
| *** erikwilson has quit IRC | 02:18 | |
| fungi | dstufft: thanks--i guess we'll want to prepare to adjust our pre-release tag pattern again | 02:18 |
| dstufft | fungi: sorry for flip flopping on y'all :( | 02:19 |
| fungi | dstufft: it happens. at least it's a change for the better | 02:19 |
| *** ryanpetrello has joined #openstack-infra | 02:20 | |
| anteaya | looks like the merge-check queue got cleared out | 02:21 |
| anteaya | that lightens a bit of load | 02:21 |
| *** timcline has joined #openstack-infra | 02:21 | |
| *** wenlock has quit IRC | 02:22 | |
| *** timcline_ has joined #openstack-infra | 02:22 | |
| *** baoli has joined #openstack-infra | 02:24 | |
| *** markmcclain1 has joined #openstack-infra | 02:24 | |
| *** markmcclain1 has quit IRC | 02:24 | |
| openstackgerrit | Merged openstack-infra/subunit2sql: Update TODO file https://review.openstack.org/141965 | 02:24 |
| *** markmcclain1 has joined #openstack-infra | 02:24 | |
| *** markmcclain has quit IRC | 02:26 | |
| *** timcline has quit IRC | 02:26 | |
| *** timcline_ has quit IRC | 02:26 | |
| anteaya | eth0 out spiked again and system cpu increased | 02:27 |
| *** ryanpetrello has quit IRC | 02:27 | |
| anteaya | hope it is of no concern | 02:27 |
| *** timcline has joined #openstack-infra | 02:27 | |
| *** yaguang has quit IRC | 02:31 | |
| *** weshay has quit IRC | 02:32 | |
| *** koolhead17 has joined #openstack-infra | 02:32 | |
| *** avozza is now known as zz_avozza | 02:33 | |
| *** armax has joined #openstack-infra | 02:34 | |
| anteaya | both are dropping again | 02:36 |
| *** Ryan_Lane has quit IRC | 02:36 | |
| openstackgerrit | Matthew Treinish proposed openstack-infra/subunit2sql: Add a db api method to extract run_times for a single test https://review.openstack.org/142309 | 02:36 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/subunit2sql: WIP: Add tool for generating line graphs of test run times https://review.openstack.org/142649 | 02:36 |
| *** amotoki has joined #openstack-infra | 02:38 | |
| anteaya | I think it is time for me to be offline | 02:40 |
| anteaya | g'night | 02:40 |
| *** yosam has joined #openstack-infra | 02:42 | |
| *** Ryan_Lane has joined #openstack-infra | 02:45 | |
| clarkb | jeblair I think my intent was to never read beyond what I could process to avoid leaking memory | 02:46 |
| clarkb | jeblair also it should go through the loop once,fail to make apacket then go through again with need bytes true. but I likely need to rethink all that | 02:47 |
| jeblair | clarkb: i understand why need_bytes is required now | 02:47 |
| jeblair | clarkb: if it reads 2 packets, it will only return one, the next time through it needs to return the second without performing another read. the third time through, it should do a read and then raise eagain | 02:48 |
| clarkb | yup | 02:48 |
| jeblair | clarkb: so i think needs_bytes needs to be saved | 02:48 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/subunit2sql: WIP: Add tool for generating line graphs of test run times https://review.openstack.org/142649 | 02:48 |
| jeblair | clarkb: (like input_buffer) | 02:48 |
| clarkb | jeblair ya I can do tbat | 02:48 |
| jeblair | clarkb: i just hacked that up and it looks promising... | 02:48 |
| clarkb | awesome feel free to push a new patchset | 02:48 |
| *** Ryan_Lane has quit IRC | 02:49 | |
| jeblair | ok cool. will do | 02:49 |
| *** bswartz has quit IRC | 02:50 | |
| *** patrickeast has quit IRC | 02:50 | |
| *** bswartz has joined #openstack-infra | 02:51 | |
| openstackgerrit | James E. Blair proposed openstack-infra/gear: Read more bytes per readPacket iteration https://review.openstack.org/140186 | 02:52 |
| jeblair | pushed and -1'd :) | 02:54 |
| jeblair | dstufft, clarkb: pbr patches merged | 02:56 |
| jeblair | fungi: ^ | 02:56 |
| dstufft | fungi: clarkb so we need a pbr release then yea? | 03:01 |
| *** shashankhegde has joined #openstack-infra | 03:02 | |
| *** timcline has quit IRC | 03:02 | |
| *** SumitNaiksatam has joined #openstack-infra | 03:02 | |
| jeblair | restarting gear with latest patchset of 140186 | 03:03 |
| *** otter768 has joined #openstack-infra | 03:05 | |
| jeblair | okay, that time we hit a registration race; i'll restart zuul and re-enqueue | 03:05 |
| *** dkranz has joined #openstack-infra | 03:06 | |
| *** otter768 has quit IRC | 03:09 | |
| *** SumitNaiksatam has quit IRC | 03:10 | |
| *** SumitNaiksatam has joined #openstack-infra | 03:10 | |
| *** talluri has joined #openstack-infra | 03:12 | |
| *** dkranz has quit IRC | 03:13 | |
| jeblair | clarkb: i think there's still a bug; some functions were not being registered; i think the can_do packets were being missed | 03:16 |
| jeblair | so we're running geard master now (which was working fairly well for the past few hours). will probably stay with that tonight. | 03:16 |
| *** koolhead17 has quit IRC | 03:17 | |
| *** dannywilson has quit IRC | 03:20 | |
| *** melwitt has quit IRC | 03:21 | |
| *** reed has quit IRC | 03:23 | |
| *** zz_avozza is now known as avozza | 03:26 | |
| *** liusheng has quit IRC | 03:27 | |
| *** yaguang_ has joined #openstack-infra | 03:27 | |
| *** julim has joined #openstack-infra | 03:29 | |
| *** liusheng has joined #openstack-infra | 03:30 | |
| openstackgerrit | Jamie Lennox proposed openstack/requirements: Update version of neutronclient https://review.openstack.org/142657 | 03:37 |
| *** yamamoto has joined #openstack-infra | 03:43 | |
| *** enikanorov_ has joined #openstack-infra | 03:43 | |
| *** enikanorov has quit IRC | 03:44 | |
| *** avozza is now known as zz_avozza | 03:44 | |
| *** jaypipes has joined #openstack-infra | 03:44 | |
| *** koolhead17 has joined #openstack-infra | 03:47 | |
| *** yaguang_ has quit IRC | 03:52 | |
| *** dims__ has quit IRC | 03:53 | |
| *** dims__ has joined #openstack-infra | 03:54 | |
| *** emagana has quit IRC | 03:56 | |
| *** emagana has joined #openstack-infra | 03:58 | |
| *** timcline has joined #openstack-infra | 04:00 | |
| *** dims__ has quit IRC | 04:01 | |
| *** dims__ has joined #openstack-infra | 04:02 | |
| *** emagana has quit IRC | 04:02 | |
| *** timcline has quit IRC | 04:04 | |
| *** yaguang_ has joined #openstack-infra | 04:05 | |
| *** timcline has joined #openstack-infra | 04:06 | |
| *** dims__ has quit IRC | 04:07 | |
| *** talluri has quit IRC | 04:09 | |
| *** talluri has joined #openstack-infra | 04:10 | |
| *** sarob has joined #openstack-infra | 04:12 | |
| *** sarob has quit IRC | 04:17 | |
| *** harlowja is now known as harlowja_away | 04:19 | |
| *** yamamoto has quit IRC | 04:21 | |
| *** Masahiro has quit IRC | 04:24 | |
| *** talluri has quit IRC | 04:25 | |
| *** timcline has quit IRC | 04:27 | |
| *** timcline has joined #openstack-infra | 04:27 | |
| *** Masahiro has joined #openstack-infra | 04:31 | |
| *** sweston has quit IRC | 04:35 | |
| *** mfink has joined #openstack-infra | 04:35 | |
| *** zz_avozza is now known as avozza | 04:36 | |
| *** wenlock has joined #openstack-infra | 04:40 | |
| *** dhp has quit IRC | 04:45 | |
| *** dhp has joined #openstack-infra | 04:46 | |
| *** yamamoto has joined #openstack-infra | 04:50 | |
| *** timcline has quit IRC | 04:50 | |
| *** esp has left #openstack-infra | 04:52 | |
| *** bhunter71 has quit IRC | 04:56 | |
| *** avozza is now known as zz_avozza | 04:56 | |
| *** sarob has joined #openstack-infra | 05:05 | |
| *** otter768 has joined #openstack-infra | 05:06 | |
| *** nosnos has joined #openstack-infra | 05:10 | |
| *** otter768 has quit IRC | 05:10 | |
| *** SumitNaiksatam has quit IRC | 05:14 | |
| *** SumitNaiksatam has joined #openstack-infra | 05:17 | |
| *** shashankhegde has quit IRC | 05:20 | |
| *** timcline has joined #openstack-infra | 05:22 | |
| *** koolhead17 has quit IRC | 05:23 | |
| *** andreykurilin_ has joined #openstack-infra | 05:25 | |
| *** timcline has quit IRC | 05:27 | |
| *** dhp has quit IRC | 05:27 | |
| *** shashankhegde has joined #openstack-infra | 05:28 | |
| *** sarob has quit IRC | 05:30 | |
| *** salv-orlando has joined #openstack-infra | 05:33 | |
| *** markmcclain1 has quit IRC | 05:34 | |
| *** nikil22 has joined #openstack-infra | 05:35 | |
| *** jerryz has quit IRC | 05:36 | |
| *** salv-orlando has quit IRC | 05:37 | |
| *** dhp has joined #openstack-infra | 05:39 | |
| *** sputnik13 has joined #openstack-infra | 05:43 | |
| yosam | I need some assistance with submitting my first git review to the openstack-dev sandbox. Is this the correct place for seeking help? | 05:43 |
| *** koolhead17 has joined #openstack-infra | 05:44 | |
| grue_pm | yosam: people round here are usually very helpful wiht anything or will point you in the right direction. It's a bit quieter now as a lot of them are asleep <-: | 05:45 |
| naggappan | yosam : let us know what issue you are facing when you try to commit | 05:46 |
| yosam | fatal: A Contributor Agreement must be completed before uploading: | 05:46 |
| naggappan | yousam : please gothrough https://wiki.openstack.org/wiki/Documentation/HowTo | 05:47 |
| yosam | I am attempting to setup an External Test Platform | 05:47 |
| naggappan | yosam : you mean third party CI system? | 05:47 |
| *** zz_avozza is now known as avozza | 05:48 | |
| *** xyang1 has quit IRC | 05:48 | |
| yosam | http://ci.openstack.org/third_party.html# | 05:48 |
| *** jerryz has joined #openstack-infra | 05:48 | |
| *** gondoi is now known as zz_gondoi | 05:49 | |
| *** yolanda has quit IRC | 05:50 | |
| naggappan | yosam : yes for this you need to create a ssh key pair then send the request with public key and other detaisl to openstack-infra mailing list | 05:50 |
| *** yolanda has joined #openstack-infra | 05:51 | |
| yosam | I have done so and have a gerrit user | 05:51 |
| naggappan | yosam: so you need help in setting up now? | 05:52 |
| *** sputnik13 has quit IRC | 05:52 | |
| *** shashankhegde has quit IRC | 05:52 | |
| yosam | I have been using http://www.joinfu.com/2014/02/setting-up-an-external-openstack-testing-system/ for guidance | 05:53 |
| *** sputnik13 has joined #openstack-infra | 05:53 | |
| naggappan | yosam : yes thats the good one. all the setup details are over there. But lots of puppet manifest have been changed . so you need to modify few instllation scripts in it. | 05:54 |
| *** jerryz_ has joined #openstack-infra | 05:55 | |
| *** Masahiro has quit IRC | 05:55 | |
| yosam | I have modified the installation scripts, however I may have missed something. Currently I have got as far as Testing Communication Between Upstream and Your Master | 05:56 |
| *** jerryz has quit IRC | 05:58 | |
| *** yfried_ has quit IRC | 05:58 | |
| naggappan | Yosam: Have you did any test by commiting a code in openstack/sandbox ? | 05:59 |
| naggappan | to just verify the test triggers or not | 05:59 |
| yosam | The git commit was successful | 06:02 |
| naggappan | yosam: so the patch submited in review.openstack.org successfully. But your test did not trigger ? | 06:03 |
| *** sdake has quit IRC | 06:04 | |
| *** sdake has joined #openstack-infra | 06:04 | |
| *** sdake has quit IRC | 06:05 | |
| *** sdake has joined #openstack-infra | 06:05 | |
| *** avozza is now known as zz_avozza | 06:08 | |
| yosam | correct the Contributor Agreement warning is generated when try git review | 06:08 |
| yosam | so no test was triggered | 06:08 |
| naggappan | yousam: when you do git review , you will get a review url. Do you get that url for your patchset | 06:09 |
| *** teran has quit IRC | 06:09 | |
| *** dims__ has joined #openstack-infra | 06:09 | |
| *** k4n0 has joined #openstack-infra | 06:10 | |
| *** andreykurilin_ has quit IRC | 06:11 | |
| naggappan | yosam : I guess you use same user to submit the patch to openstack-dev/sandbox project also. You should submit as a standard user to the openstack-dev/sandbox. Like how the developers will do the commit to projects. once the patch is submitted , not only you in any one submit the patch the test should trigger in your CI system automatically. | 06:12 |
| yosam | no | 06:12 |
| *** rmcall has joined #openstack-infra | 06:13 | |
| *** dims__ has quit IRC | 06:13 | |
| yosam | by standard user do you mean the gerrit user name | 06:13 |
| naggappan | yousam : yes gerrit user name. | 06:14 |
| *** sputnik13 has quit IRC | 06:14 | |
| *** primeministerp has quit IRC | 06:14 | |
| yosam | thank you for your assistance I will keep trying | 06:15 |
| *** hdd has joined #openstack-infra | 06:16 | |
| *** primeministerp has joined #openstack-infra | 06:17 | |
| *** rmcall has quit IRC | 06:20 | |
| *** dannywilson has joined #openstack-infra | 06:20 | |
| *** rmcall has joined #openstack-infra | 06:20 | |
| *** achanda has joined #openstack-infra | 06:22 | |
| *** sputnik13 has joined #openstack-infra | 06:22 | |
| *** timcline has joined #openstack-infra | 06:23 | |
| *** dannywilson has quit IRC | 06:24 | |
| *** timcline has quit IRC | 06:27 | |
| *** sputnik13 has quit IRC | 06:27 | |
| *** sdake has quit IRC | 06:28 | |
| *** tomoe_ has joined #openstack-infra | 06:29 | |
| *** Masahiro has joined #openstack-infra | 06:29 | |
| *** MarkAtwood has joined #openstack-infra | 06:29 | |
| *** tomoe_ has quit IRC | 06:29 | |
| *** tomoe has joined #openstack-infra | 06:29 | |
| *** sdake has joined #openstack-infra | 06:30 | |
| *** sdake has quit IRC | 06:30 | |
| *** sdake has joined #openstack-infra | 06:30 | |
| *** mugsie has quit IRC | 06:32 | |
| *** mugsie has joined #openstack-infra | 06:33 | |
| *** cody-somerville has quit IRC | 06:34 | |
| *** mrmartin has joined #openstack-infra | 06:37 | |
| jeblair | clarkb, fungi: the jenkins01 error happened again; existing logging still wasn't enough to debug so i've applied local patch to narrow it further and restarted geard. | 06:38 |
| *** sputnik13 has joined #openstack-infra | 06:43 | |
| *** pcrews has joined #openstack-infra | 06:43 | |
| *** AJaeger has joined #openstack-infra | 06:44 | |
| AJaeger | fungi, clarkb: Still awake? There are a lot NOT_REGISTERED jobs ;( | 06:45 |
| AJaeger | jeblair: see above ^ | 06:45 |
| *** zz_sabari is now known as sabari | 06:47 | |
| *** garyk has joined #openstack-infra | 06:48 | |
| AJaeger | ah, seems to be recovering now. | 06:49 |
| *** pcrews has quit IRC | 06:50 | |
| *** sdake has quit IRC | 06:50 | |
| *** vigneshvar has joined #openstack-infra | 06:52 | |
| *** MarkAtwood has quit IRC | 06:53 | |
| *** Ryan_Lane has joined #openstack-infra | 06:54 | |
| *** hdd has quit IRC | 06:55 | |
| *** AJaeger has quit IRC | 06:58 | |
| *** mrunge has joined #openstack-infra | 06:58 | |
| *** dannywilson has joined #openstack-infra | 06:59 | |
| *** zz_avozza is now known as avozza | 07:01 | |
| *** _nadya_ has joined #openstack-infra | 07:03 | |
| *** koolhead17 has quit IRC | 07:05 | |
| *** _nadya_ has quit IRC | 07:05 | |
| *** kevinbenton has quit IRC | 07:06 | |
| *** yfried_ has joined #openstack-infra | 07:06 | |
| *** otter768 has joined #openstack-infra | 07:06 | |
| *** kevinbenton has joined #openstack-infra | 07:08 | |
| *** sdake has joined #openstack-infra | 07:10 | |
| *** sdake has quit IRC | 07:10 | |
| *** sdake has joined #openstack-infra | 07:10 | |
| *** subscope has joined #openstack-infra | 07:11 | |
| *** otter768 has quit IRC | 07:11 | |
| *** wenlock has left #openstack-infra | 07:14 | |
| *** stevemar has quit IRC | 07:14 | |
| *** _nadya_ has joined #openstack-infra | 07:17 | |
| *** btran has quit IRC | 07:17 | |
| *** jcoufal has joined #openstack-infra | 07:18 | |
| *** avozza is now known as zz_avozza | 07:18 | |
| *** rmcall has quit IRC | 07:20 | |
| *** armax has quit IRC | 07:21 | |
| *** camunoz_ has joined #openstack-infra | 07:23 | |
| *** timcline has joined #openstack-infra | 07:24 | |
| *** che-arne has quit IRC | 07:24 | |
| *** nikil22 has quit IRC | 07:25 | |
| *** sdake has quit IRC | 07:26 | |
| *** _nadya_ has quit IRC | 07:27 | |
| *** timcline has quit IRC | 07:28 | |
| *** sdake has joined #openstack-infra | 07:29 | |
| *** sdake has quit IRC | 07:30 | |
| *** sdake has joined #openstack-infra | 07:30 | |
| *** achanda has quit IRC | 07:31 | |
| *** sputnik13 has quit IRC | 07:31 | |
| *** sputnik13 has joined #openstack-infra | 07:33 | |
| *** alexpilotti has joined #openstack-infra | 07:34 | |
| *** dannywilson has quit IRC | 07:35 | |
| *** dannywilson has joined #openstack-infra | 07:36 | |
| *** xyang1 has joined #openstack-infra | 07:39 | |
| *** sdake has quit IRC | 07:40 | |
| *** dannywilson has quit IRC | 07:41 | |
| *** sdake has joined #openstack-infra | 07:42 | |
| *** HeOS has quit IRC | 07:42 | |
| *** e0ne has joined #openstack-infra | 07:53 | |
| *** e0ne has quit IRC | 07:54 | |
| *** garyk has quit IRC | 07:55 | |
| *** pcaruana has joined #openstack-infra | 07:56 | |
| *** zz_avozza is now known as avozza | 07:58 | |
| *** xyang1 has quit IRC | 07:58 | |
| *** ujuc has joined #openstack-infra | 08:00 | |
| *** Ryan_Lane has quit IRC | 08:04 | |
| *** salv-orlando has joined #openstack-infra | 08:04 | |
| *** salv-orlando has quit IRC | 08:09 | |
| *** salv-orlando has joined #openstack-infra | 08:11 | |
| *** BobBall_AWOL is now known as BobBall | 08:13 | |
| *** sabari is now known as zz_sabari | 08:15 | |
| *** amotoki has quit IRC | 08:15 | |
| *** yfried_ is now known as yfried|afk | 08:17 | |
| *** hashar has joined #openstack-infra | 08:17 | |
| *** amuller has joined #openstack-infra | 08:18 | |
| *** yamamoto has quit IRC | 08:18 | |
| *** achanda has joined #openstack-infra | 08:19 | |
| *** timcline has joined #openstack-infra | 08:25 | |
| *** yfried|afk is now known as yfried_ | 08:25 | |
| *** mpaolino has joined #openstack-infra | 08:25 | |
| *** _nadya_ has joined #openstack-infra | 08:28 | |
| *** timcline has quit IRC | 08:29 | |
| *** _nadya_ has quit IRC | 08:31 | |
| *** sarob has joined #openstack-infra | 08:31 | |
| *** sarob has quit IRC | 08:36 | |
| ttx | Hmm, we might need a pbr tag to propagate fix for https://bugs.launchpad.net/os-apply-config/+bug/1403510 | 08:43 |
| uvirtbot | Launchpad bug 1403510 in os-apply-config "gate tests failed with "NameError: name 'StandardError' is not defined"" [High,Confirmed] | 08:43 |
| *** luqas has joined #openstack-infra | 08:45 | |
| *** Ala has joined #openstack-infra | 08:46 | |
| *** jedimike has joined #openstack-infra | 08:51 | |
| *** alexpilotti has quit IRC | 08:56 | |
| *** ujuc has quit IRC | 09:00 | |
| *** dizquierdo has joined #openstack-infra | 09:07 | |
| *** otter768 has joined #openstack-infra | 09:07 | |
| *** jpich has joined #openstack-infra | 09:07 | |
| *** mpaolino has quit IRC | 09:08 | |
| *** skolekonov has joined #openstack-infra | 09:09 | |
| *** camunoz_ has quit IRC | 09:09 | |
| *** salv-orlando has quit IRC | 09:11 | |
| *** nfedotov has joined #openstack-infra | 09:11 | |
| *** otter768 has quit IRC | 09:12 | |
| *** derekh has joined #openstack-infra | 09:14 | |
| *** HeOS has joined #openstack-infra | 09:17 | |
| *** MaxV has joined #openstack-infra | 09:17 | |
| *** garyk has joined #openstack-infra | 09:17 | |
| *** Guest51388 has joined #openstack-infra | 09:17 | |
| *** _nadya_ has joined #openstack-infra | 09:18 | |
| *** _nadya__ has joined #openstack-infra | 09:19 | |
| *** _nadya_ has quit IRC | 09:19 | |
| *** subscope has quit IRC | 09:20 | |
| *** liusheng has quit IRC | 09:22 | |
| *** pblaho has joined #openstack-infra | 09:23 | |
| *** liusheng has joined #openstack-infra | 09:24 | |
| *** timcline has joined #openstack-infra | 09:25 | |
| *** timcline has quit IRC | 09:27 | |
| *** timcline has joined #openstack-infra | 09:27 | |
| *** _shaps_ has joined #openstack-infra | 09:28 | |
| *** berendt has joined #openstack-infra | 09:28 | |
| openstackgerrit | Aleksey proposed openstack-infra/storyboard: Fix i18n https://review.openstack.org/142503 | 09:30 |
| *** adrian_otto has left #openstack-infra | 09:30 | |
| *** timcline has quit IRC | 09:32 | |
| *** jamielennox is now known as jamielennox|away | 09:32 | |
| *** e0ne has joined #openstack-infra | 09:39 | |
| *** zhiwei has joined #openstack-infra | 09:40 | |
| *** ssam2 has joined #openstack-infra | 09:47 | |
| *** luqas has quit IRC | 09:48 | |
| *** yaguang has joined #openstack-infra | 09:51 | |
| *** yaguang has quit IRC | 09:51 | |
| *** yaguang has joined #openstack-infra | 09:51 | |
| *** zz_johnthetubagu is now known as johnthetubaguy | 09:52 | |
| *** yaguang_ has quit IRC | 09:52 | |
| *** _nadya__ has quit IRC | 09:54 | |
| *** zhiwei has quit IRC | 09:56 | |
| *** gilliard_afk is now known as gilliard | 09:57 | |
| *** jp_at_hp has joined #openstack-infra | 10:00 | |
| *** mrmartin has quit IRC | 10:02 | |
| *** e0ne is now known as e0ne_ | 10:05 | |
| *** e0ne_ is now known as e0ne | 10:09 | |
| *** ihrachyshka has joined #openstack-infra | 10:09 | |
| openstackgerrit | Merged openstack/requirements: Ensure we have a PEP 440 compatible version of pytz https://review.openstack.org/142348 | 10:10 |
| *** salv-orlando has joined #openstack-infra | 10:12 | |
| *** dtantsur|afk is now known as dtantsur | 10:12 | |
| *** ZZelle has quit IRC | 10:14 | |
| *** yfried_ is now known as yfried|afk | 10:15 | |
| *** salv-orlando has quit IRC | 10:17 | |
| *** yfried|afk is now known as yfried_ | 10:17 | |
| *** luqas has joined #openstack-infra | 10:20 | |
| *** teran has joined #openstack-infra | 10:25 | |
| *** achanda has quit IRC | 10:26 | |
| *** achanda has joined #openstack-infra | 10:27 | |
| *** timcline has joined #openstack-infra | 10:28 | |
| *** yfried_ is now known as yfried|afk | 10:31 | |
| *** achanda has quit IRC | 10:31 | |
| *** achanda has joined #openstack-infra | 10:32 | |
| *** timcline has quit IRC | 10:32 | |
| openstackgerrit | Aleksey proposed openstack-infra/storyboard: Fix i18n https://review.openstack.org/142503 | 10:34 |
| *** alexpilotti has joined #openstack-infra | 10:35 | |
| *** rlandy has joined #openstack-infra | 10:42 | |
| *** yfried|afk is now known as yfried_ | 10:45 | |
| *** achanda has quit IRC | 10:45 | |
| *** achanda has joined #openstack-infra | 10:46 | |
| tchaypo | pleia2: are you around perchance? | 10:49 |
| *** sweston has joined #openstack-infra | 10:49 | |
| *** achanda has quit IRC | 10:50 | |
| *** subscope has joined #openstack-infra | 10:56 | |
| *** Guest51388 has quit IRC | 10:57 | |
| *** sweston has quit IRC | 10:59 | |
| *** rediskin has quit IRC | 11:00 | |
| *** mpaolino has joined #openstack-infra | 11:01 | |
| *** unicell has quit IRC | 11:02 | |
| *** yfried_ is now known as yfried|afk | 11:02 | |
| openstackgerrit | James Polley proposed stackforge/gertty: Add detailed examples and dashboards a la gerrit https://review.openstack.org/142108 | 11:02 |
| *** sputnik13 has quit IRC | 11:03 | |
| *** dizquierdo is now known as dizquierdo_afk | 11:06 | |
| *** otter768 has joined #openstack-infra | 11:08 | |
| *** Svedrin has quit IRC | 11:10 | |
| *** salv-orlando has joined #openstack-infra | 11:12 | |
| *** otter768 has quit IRC | 11:12 | |
| *** _nadya_ has joined #openstack-infra | 11:14 | |
| *** Svedrin has joined #openstack-infra | 11:16 | |
| *** yfried|afk is now known as yfried_ | 11:17 | |
| *** dims__ has joined #openstack-infra | 11:17 | |
| *** avozza is now known as zz_avozza | 11:20 | |
| *** aysyd has joined #openstack-infra | 11:21 | |
| *** dims__ has quit IRC | 11:22 | |
| openstackgerrit | Aleksey proposed openstack-infra/storyboard: Changes for calls iteritems and str to six https://review.openstack.org/142733 | 11:24 |
| *** adalbas has joined #openstack-infra | 11:25 | |
| *** nosnos has quit IRC | 11:26 | |
| *** ilyashakhat has joined #openstack-infra | 11:26 | |
| *** nosnos has joined #openstack-infra | 11:26 | |
| openstackgerrit | Aleksey proposed openstack-infra/storyboard: Fix i18n https://review.openstack.org/142503 | 11:27 |
| *** timcline has joined #openstack-infra | 11:29 | |
| *** ociuhandu has joined #openstack-infra | 11:29 | |
| *** _nadya_ has quit IRC | 11:29 | |
| *** nosnos has quit IRC | 11:30 | |
| *** yaguang has quit IRC | 11:31 | |
| *** garyk has quit IRC | 11:31 | |
| *** timcline has quit IRC | 11:33 | |
| mordred | fungi: morning. any pbr tasks for me to jump on? | 11:34 |
| *** Masahiro has quit IRC | 11:34 | |
| mordred | ttx: yah, I think I need to tag one as soon as I get airborne | 11:36 |
| *** hashar has quit IRC | 11:37 | |
| *** nosnos has joined #openstack-infra | 11:39 | |
| *** hashar has joined #openstack-infra | 11:40 | |
| Daviey | mordred: pbr? I have a waiting pull request :) | 11:46 |
| *** sdake has quit IRC | 11:46 | |
| mordred | Daviey: yah ... we've been staving off setuptools disaster with some small point releases ... but I'll review your patch today | 11:47 |
| *** sdake has joined #openstack-infra | 11:47 | |
| *** sdake has quit IRC | 11:47 | |
| *** sdake has joined #openstack-infra | 11:47 | |
| *** _nadya_ has joined #openstack-infra | 11:50 | |
| mordred | Daviey: looks good | 11:52 |
| *** dims__ has joined #openstack-infra | 11:53 | |
| *** _nadya_ has quit IRC | 11:56 | |
| openstackgerrit | Cédric de Saint Martin proposed openstack-dev/pbr: Check for git installation before using git. https://review.openstack.org/142738 | 11:56 |
| *** hashar has quit IRC | 11:59 | |
| *** salv-orlando has quit IRC | 11:59 | |
| *** _nadya_ has joined #openstack-infra | 11:59 | |
| *** hashar has joined #openstack-infra | 12:03 | |
| *** erlon has joined #openstack-infra | 12:05 | |
| *** Longgeek has joined #openstack-infra | 12:06 | |
| openstackgerrit | Tim Kelsey proposed openstack/requirements: Bumping PyKMIP version from 0.1.0 to 0.2.0 https://review.openstack.org/137016 | 12:07 |
| *** slagle_ is now known as slagle | 12:07 | |
| *** sdake has quit IRC | 12:10 | |
| *** pc_m has joined #openstack-infra | 12:13 | |
| *** andreaf_ is now known as andreaf | 12:13 | |
| *** e0ne is now known as e0ne_ | 12:13 | |
| dstufft | mordred: I think feature/0.10 needs another release | 12:14 |
| dstufft | unless someone did it already | 12:14 |
| dstufft | the .postN, fix for docutils, and fix for not having git installed landed last night | 12:14 |
| *** ZZelle has joined #openstack-infra | 12:15 | |
| mordred | dstufft: woot | 12:17 |
| dstufft | also the glboal requirements landed for pytz too | 12:17 |
| *** sdake has joined #openstack-infra | 12:18 | |
| *** sdake has quit IRC | 12:18 | |
| *** sdake has joined #openstack-infra | 12:18 | |
| dstufft | I think that means that the only thing left for setuptools 8 is the warnings, I have a PR against setuptools which will subclass RuntimeWarning to make a PEP440Warning so we can filter just that warning (as long as we're OK with failing on setuptools 8.0-8.0.4 and passing setuptools 8.0.5+ | 12:18 |
| dstufft | but that's not landed yet | 12:18 |
| mordred | cool. I think we're definitely fine with failing on those setuptoolses | 12:20 |
| mordred | since nothing else has released dependent on them, so once 8.0.5 is in the wild, it's unlikely the others will ever come up | 12:20 |
| * mordred has pushed 0.10.4 tag for pbr | 12:20 | |
| dstufft | I also have a PR that'll change the normalization rules for 1.0c1 -> 1.0rc1 | 12:21 |
| dstufft | we changed PEP 440 yesterday for it | 12:21 |
| *** e0ne_ is now known as e0ne | 12:22 | |
| mordred | dstufft: this has been a fun week hasnt it? | 12:23 |
| dstufft | mordred: just think | 12:24 |
| dstufft | soon pip 6 gets released | 12:24 |
| dstufft | :D | 12:24 |
| mordred | dstufft: yay! | 12:24 |
| dims__ | haha dstufft :) | 12:24 |
| *** zz_avozza is now known as avozza | 12:24 | |
| dstufft | although that shouldn't be bad | 12:24 |
| dstufft | PEP 440 is the biggest thing in pip 6 that'll cause breakage | 12:24 |
| *** cdent has joined #openstack-infra | 12:24 | |
| dstufft | and setuptools is already leading on that | 12:24 |
| *** alexpilotti has quit IRC | 12:26 | |
| dstufft | other than PEP 440 pip 6 should mostly be things that just make things better for people, on by default HTTP cache, randomized build directories, way less noisy output, automatic retries on HTTP requests if downloading things fails | 12:27 |
| dstufft | oh, pip 6 does deprecate accessing non HTTPS URLs without a flag, but those still _work_ on pip 6, it just yells at you until you add the flag | 12:27 |
| dstufft | (to be clear, it yells at you and works without the flag, it won't stop working till 7) | 12:27 |
| mordred | oh. we should do something about that then | 12:28 |
| mordred | since we have non-https mirrors | 12:28 |
| dstufft | yea | 12:28 |
| dstufft | you can just set --trusted-host if you want, or add certificates and do --cert | 12:28 |
| mordred | fungi: ^^ maybe we should get a cert | 12:28 |
| dstufft | or get trusted certs | 12:28 |
| mordred | yah. I mean, they're $10 | 12:28 |
| dstufft | pip 6 has config files in /etc/ now too, and inside a virtualenv | 12:29 |
| dstufft | might be useful for y'all | 12:29 |
| mordred | ooh | 12:29 |
| mordred | very much | 12:29 |
| mordred | that actually might be the easiest thing to do on our build hosts actually | 12:29 |
| mordred | can we set multiple trusted-host entries? | 12:29 |
| *** timcline has joined #openstack-infra | 12:30 | |
| mordred | (where's the best place for me to look to learn about that system?) | 12:30 |
| dstufft | yea you can set multiple | 12:31 |
| dstufft | you mean the trusted host system? I can link the PR, there's nothing special about it though, it adds a --trusted-host flag where you specify the DNS name | 12:32 |
| dstufft | like --trusted-host wahtever.example.com | 12:32 |
| *** mpaolino has quit IRC | 12:33 | |
| *** Longgeek has quit IRC | 12:33 | |
| *** Longgeek has joined #openstack-infra | 12:33 | |
| dstufft | https://github.com/pypa/pip/pull/2162 | 12:33 |
| *** sdake has quit IRC | 12:33 | |
| mordred | no - I mean the new config file stuff | 12:33 |
| dstufft | oh | 12:34 |
| dstufft | sec | 12:34 |
| *** timcline has quit IRC | 12:34 | |
| *** Masahiro has joined #openstack-infra | 12:34 | |
| dstufft | it's not on pip.pypa.io yet because it's not released, so you'll have to put up with weird formatting since github doens't grok sphinx - https://github.com/pypa/pip/blob/develop/docs/user_guide.rst#configuration | 12:35 |
| dstufft | mordred: is there any interest in making it so the tarballs emitted by pbr don't require pbr? | 12:36 |
| openstackgerrit | yolanda.robla proposed openstack-infra/storyboard-webclient: Read user preferences from backend https://review.openstack.org/137798 | 12:39 |
| mordred | dstufft: yes - I started working on having pbr emit a setup.py in tarballs that didn't use pbr | 12:39 |
| mordred | but then I got busy with other things | 12:39 |
| *** Masahiro has quit IRC | 12:39 | |
| dstufft | mordred: I may poke at that sometime | 12:39 |
| dstufft | I kind of want a pbr like thing, but without an install time dep on pbr outside of a dev environment | 12:40 |
| mordred | dstufft: https://review.openstack.org/97033 <-- non-functional work in progress | 12:40 |
| dstufft | though i'd want to turn off some things too, not sure if there's any interest in that | 12:40 |
| mordred | dstufft: I agree, the install-time requirement is ugly | 12:40 |
| *** yaguang has joined #openstack-infra | 12:40 | |
| mordred | and we have "turn off these features" flags for several things already | 12:40 |
| mordred | so I don't think those features would be unwelcome | 12:41 |
| dstufft | cool | 12:41 |
| *** yaguang has quit IRC | 12:41 | |
| mordred | dstufft: I know notmyname and the swift guys would love you for a million years if you made such a thing, btw | 12:42 |
| dstufft | mordred: I started writing something for use in pip, then realized I was more or less just recreating parts of pbr | 12:42 |
| dstufft | (still not sure if it's reasonable for pip to do something like that, I'm pretty sure it's not reasonable for pip to do something like that in a released version, I think it would probably be fine in a develop version though) | 12:44 |
| *** garyk has joined #openstack-infra | 12:44 | |
| *** jgallard_ has joined #openstack-infra | 12:47 | |
| *** unicell has joined #openstack-infra | 12:47 | |
| *** weshay has joined #openstack-infra | 12:51 | |
| *** subscope has quit IRC | 12:53 | |
| *** mfink has quit IRC | 12:54 | |
| *** sandywalsh has quit IRC | 12:57 | |
| *** sdake has joined #openstack-infra | 13:01 | |
| *** sdake has quit IRC | 13:01 | |
| *** sdake has joined #openstack-infra | 13:01 | |
| openstackgerrit | yolanda.robla proposed openstack-infra/storyboard-webclient: Read user preferences from backend https://review.openstack.org/137798 | 13:01 |
| *** nosnos has quit IRC | 13:05 | |
| *** nosnos has joined #openstack-infra | 13:06 | |
| *** yfried_ has quit IRC | 13:07 | |
| *** yfried_ has joined #openstack-infra | 13:08 | |
| *** tomoe_ has joined #openstack-infra | 13:09 | |
| *** otter768 has joined #openstack-infra | 13:09 | |
| openstackgerrit | yolanda.robla proposed openstack-infra/storyboard-webclient: Read user preferences from backend https://review.openstack.org/137798 | 13:09 |
| *** mpaolino has joined #openstack-infra | 13:10 | |
| *** nosnos has quit IRC | 13:10 | |
| *** tomoe has quit IRC | 13:11 | |
| *** kgiusti has joined #openstack-infra | 13:12 | |
| *** otter768 has quit IRC | 13:14 | |
| *** dprince has joined #openstack-infra | 13:14 | |
| *** mfink has joined #openstack-infra | 13:15 | |
| *** Daviey has quit IRC | 13:15 | |
| *** Daviey has joined #openstack-infra | 13:15 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/system-config: List PyPI mirrors as trusted hosts everywhere https://review.openstack.org/142772 | 13:15 |
| mordred | dstufft: ^^ does that look sane to you? | 13:16 |
| *** andreaf is now known as andreaf_ | 13:16 | |
| garyk | fungi: are you around? | 13:17 |
| garyk | anteaya: are you around? | 13:17 |
| dstufft | mordred: um, in my "it's 8am and I have a headache" state it looks right to me | 13:18 |
| mordred | dstufft: I know that state :) | 13:18 |
| mordred | dstufft: mine is "it's 8am, I have a headache and zomg I'm on a plane" | 13:18 |
| *** mwagner_lap has joined #openstack-infra | 13:19 | |
| dstufft | mordred: :D | 13:19 |
| *** mbacchi has joined #openstack-infra | 13:20 | |
| dstufft | mordred: so I rechecked https://review.openstack.org/#/c/142322/ and it looks like everything works now except the test cases that checkout the output of invoking a pythong script and get that warning | 13:20 |
| *** mfink__ has joined #openstack-infra | 13:21 | |
| dstufft | mordred: so either that can be fixed now by ignoring RuntimeWarnings in pkg_resources, or it can wait until setuptools 8.0.5 and ignore PEP440Warnings | 13:21 |
| *** xyang1 has joined #openstack-infra | 13:21 | |
| dstufft | oh looks like jason might have woekn up | 13:22 |
| dstufft | my setuptools PRs just started getting merged | 13:22 |
| *** mfink has quit IRC | 13:23 | |
| *** yamamoto has joined #openstack-infra | 13:23 | |
| mordred | woot! | 13:24 |
| *** fandi has joined #openstack-infra | 13:24 | |
| dims__ | mordred: in the base.pp is the source value right? seems like it is missing files/ | 13:25 |
| mordred | dims__: sadly, yes | 13:25 |
| dims__ | puppet:///modules/openstack_project/global-pip.conf -> puppet:///modules/openstack_project/files/global-pip.conf | 13:25 |
| mordred | wait - lemme look | 13:25 |
| * dims__ has no clue :) | 13:25 | |
| mordred | dims__: yeah - puppet does magic and turns puppet:///modules/openstack_project/global-pip.conf into modules/openstack_project/files/global-pip.conf | 13:26 |
| dims__ | about how puppet works | 13:26 |
| *** yfried__ has joined #openstack-infra | 13:26 | |
| mordred | dims__: because ruby | 13:26 |
| dims__ | ah ok | 13:26 |
| *** esker has quit IRC | 13:26 | |
| *** yfried_ has quit IRC | 13:27 | |
| *** wznoinsk has quit IRC | 13:27 | |
| *** _nadya_ has quit IRC | 13:28 | |
| *** timcline has joined #openstack-infra | 13:30 | |
| *** ZZelle has quit IRC | 13:31 | |
| *** Ala has quit IRC | 13:34 | |
| *** timcline has quit IRC | 13:35 | |
| *** koolhead17 has joined #openstack-infra | 13:35 | |
| *** koolhead17 has joined #openstack-infra | 13:35 | |
| *** Masahiro has joined #openstack-infra | 13:35 | |
| *** bswartz has quit IRC | 13:36 | |
| *** Ala has joined #openstack-infra | 13:38 | |
| *** ryanpetrello has joined #openstack-infra | 13:39 | |
| *** dmellado has quit IRC | 13:39 | |
| *** mrmartin has joined #openstack-infra | 13:40 | |
| *** jcoufal_ has joined #openstack-infra | 13:40 | |
| *** Masahiro has quit IRC | 13:40 | |
| openstackgerrit | Boris Pavlovic proposed openstack-infra/project-config: Add rally-py34 job https://review.openstack.org/142785 | 13:41 |
| *** jcoufal has quit IRC | 13:43 | |
| garyk | mestery: anteaya: fungi: you guys around? | 13:44 |
| *** Longgeek has quit IRC | 13:44 | |
| *** Longgeek has joined #openstack-infra | 13:45 | |
| *** yfried has joined #openstack-infra | 13:46 | |
| *** fandi has quit IRC | 13:48 | |
| *** yfried__ has quit IRC | 13:49 | |
| *** funzo has quit IRC | 13:49 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 13:50 | |
| *** fandi has joined #openstack-infra | 13:51 | |
| *** derekh has quit IRC | 13:52 | |
| dstufft | mordred: setuptools 8.1 is released | 13:52 |
| *** yamamoto has quit IRC | 13:55 | |
| mordred | dstufft: neat | 13:56 |
| *** yamamoto has joined #openstack-infra | 13:56 | |
| *** derekh has joined #openstack-infra | 13:57 | |
| *** jedimike has quit IRC | 13:58 | |
| *** isviridov_away is now known as isviridov | 13:59 | |
| mordred | ttx: ^^ we can go back to rc now | 13:59 |
| *** yfried has quit IRC | 13:59 | |
| ttx | hah! | 14:00 |
| * ttx hasn't pushed the patch to release-tools he uses for supporting c yet | 14:00 | |
| *** sputnik13 has joined #openstack-infra | 14:00 | |
| ttx | mordred: but we should still enfore x.y.z in the future, and use .0rc1 and .0b1 ? | 14:01 |
| ttx | +c | 14:01 |
| *** sandywalsh has joined #openstack-infra | 14:01 | |
| *** Adri2000 has quit IRC | 14:03 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/project-config: Go back to rc for pre-releases https://review.openstack.org/142791 | 14:05 |
| mordred | ttx: I think so | 14:05 |
| *** _nadya_ has joined #openstack-infra | 14:05 | |
| ttx | alrighty | 14:06 |
| *** _nadya_ has quit IRC | 14:07 | |
| *** e0ne_ has joined #openstack-infra | 14:07 | |
| *** _nadya_ has joined #openstack-infra | 14:07 | |
| *** Guest51388 has joined #openstack-infra | 14:07 | |
| garyk | mestery: anteaya: fungi: you guys around? | 14:08 |
| openstackgerrit | Monty Taylor proposed openstack-infra/system-config: Add rinse to nodepool https://review.openstack.org/136598 | 14:08 |
| *** e0ne has quit IRC | 14:08 | |
| openstackgerrit | Merged openstack-infra/system-config: Add debootstrap to the needed packages for dib https://review.openstack.org/124077 | 14:08 |
| openstackgerrit | Monty Taylor proposed openstack-infra/system-config: Make apt skip grabbing translations https://review.openstack.org/136837 | 14:09 |
| *** dkliban_afk is now known as dkliban | 14:10 | |
| *** _nadya_ has quit IRC | 14:10 | |
| mestery | garyk: Here sir | 14:10 |
| *** Adri2000 has joined #openstack-infra | 14:11 | |
| *** Adri2000 has quit IRC | 14:11 | |
| *** Adri2000 has joined #openstack-infra | 14:11 | |
| *** ddieterly has quit IRC | 14:11 | |
| *** ddieterly has joined #openstack-infra | 14:12 | |
| *** bswartz has joined #openstack-infra | 14:14 | |
| *** mriedem has joined #openstack-infra | 14:15 | |
| *** ddieterly has quit IRC | 14:16 | |
| *** radez_g0n3 is now known as radez | 14:18 | |
| *** garyk has quit IRC | 14:19 | |
| *** imcsk8 has quit IRC | 14:19 | |
| *** imcsk8 has joined #openstack-infra | 14:19 | |
| dstufft | mordred: where's the best place to put an environment variable to disable warnings for the tests? | 14:20 |
| dstufft | mordred: see https://bpaste.net/show/7fc02c9cbea6 | 14:20 |
| *** berendt has quit IRC | 14:22 | |
| *** pc_m has quit IRC | 14:22 | |
| *** hdd has joined #openstack-infra | 14:25 | |
| *** dizquierdo_afk is now known as dizquierdo | 14:27 | |
| mestery | Good morning infra folks! I don't seem to have permissions to add/remove third party CI accounts in gerrit, I'm apparently not in the neutron-ci group. Any help? | 14:27 |
| mestery | anteaya: ^^^^ | 14:27 |
| *** erikwilson has joined #openstack-infra | 14:27 | |
| *** esker has joined #openstack-infra | 14:28 | |
| *** k4n0 has quit IRC | 14:29 | |
| *** garyk has joined #openstack-infra | 14:30 | |
| *** pc_m has joined #openstack-infra | 14:30 | |
| *** pc_m has quit IRC | 14:31 | |
| mestery | anteaya: I think I just need to be added to the neutron-ci gerrit group so I can add/remove accounts. I was there at one time, now I'm gone. | 14:31 |
| *** timcline has joined #openstack-infra | 14:31 | |
| *** pc_m has joined #openstack-infra | 14:31 | |
| *** pc_m has quit IRC | 14:31 | |
| *** pc_m has joined #openstack-infra | 14:32 | |
| garyk | mestery: i have addressed all of the comments in https://review.openstack.org/140409 | 14:32 |
| mestery | garyk: looking | 14:32 |
| *** mfink__ has quit IRC | 14:32 | |
| garyk | it has history, tags and is ready to rock 'n roll | 14:32 |
| garyk | in actual fact it has the same properties as all of our openstack repos | 14:33 |
| *** mfink has joined #openstack-infra | 14:33 | |
| mestery | garyk: So you're ok not taking branches with you? | 14:33 |
| *** ZZelle has joined #openstack-infra | 14:33 | |
| *** mfink__ has joined #openstack-infra | 14:33 | |
| mestery | garyk: +1 | 14:34 |
| *** yaguang has joined #openstack-infra | 14:34 | |
| fungi | okay, back and caffeinated now. what's broken? | 14:35 |
| *** timcline has quit IRC | 14:36 | |
| garyk | mestery: thanks | 14:36 |
| fungi | scrollback time | 14:36 |
| mestery | fungi: lol | 14:36 |
| garyk | fungi: can you please look at https://review.openstack.org/140409 | 14:36 |
| mestery | fungi: Can you help get me into the neutron-ci gerrit group so I can enable/disable neutron third-party CI systems? | 14:37 |
| garyk | the git repo has been update to support all tags, history | 14:37 |
| fungi | jeblair: fyi once you're awake looks like the jenkins01 error is back as of ~1100 utc | 14:37 |
| *** mfink has quit IRC | 14:37 | |
| *** dkranz has joined #openstack-infra | 14:38 | |
| *** ZZelle has quit IRC | 14:38 | |
| *** ddieterly has joined #openstack-infra | 14:39 | |
| fungi | hrm, as yosam pointed out a while back, openstack-dev/ci-sandbox is set to require a signed icla. i think that should get removed. patch on the way | 14:40 |
| *** sabeen has joined #openstack-infra | 14:40 | |
| fungi | mordred: ttx: dhellmann: i fell asleep before the pbr fixes merged, so did not get anything tagged yet | 14:41 |
| *** jedimike has joined #openstack-infra | 14:41 | |
| *** naggappan has quit IRC | 14:42 | |
| *** yaguang has quit IRC | 14:42 | |
| *** yaguang has joined #openstack-infra | 14:42 | |
| garyk | mestery: not sure i follow you comment about the branches? | 14:43 |
| *** packet has joined #openstack-infra | 14:43 | |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/project-config: The ci-sandbox project should not require an ICLA https://review.openstack.org/142804 | 14:45 |
| mestery | garyk: ihar told me I should take the branches with me because it would make stable backports easier. | 14:45 |
| garyk | mestery: it has them as the tags - i think | 14:46 |
| fungi | mordred: as for pip and ssl validation, i already checked with dstufft and it's trivial to instruct pip to validate against a specific certificate (or simply add to the system trust set). i was going to turn on https with the local snakeoil cert on each of our pypi mirrors and look into the puppet we need to update the workers to trust those | 14:47 |
| mestery | garyk: OK, just wanted to be sure :) | 14:47 |
| dstufft | fungi: morded tagged pbr | 14:47 |
| garyk | mestery: maybe i missed a step. but i am really not sure | 14:47 |
| dstufft | fungi: the only thing we need for setuptools 8 now is to add PYTHONWARNINGS='ignore::RuntimeWarning:pkg_resources' when invoking the tests to suppress the warnings for the tests that subprocess to a python script and assert against the output | 14:48 |
| mestery | garyk: Did you make sure to checkout each remote branch when you cloned neutron? | 14:48 |
| dstufft | fungi: I have no idea where the best place to add that env var is tho | 14:48 |
| garyk | hmmm. | 14:49 |
| fungi | garyk: what's up? | 14:49 |
| garyk | fungi: hi, wanted to know if you had anyt houghts on https://review.openstack.org/#/c/140409/ | 14:50 |
| fungi | mestery: are you sure you can't add and remove accounts in neutron-ci? it's owned by the neutron-release group of which you're a member (being a member of neutron-ci won't help you since the group is not self-managed) | 14:52 |
| garyk | mestery: i do not recall checking out all of the remote branches - i did not see that as part of the instructions. then again i am not sure i understand the actual importnatce of that at the moment as this is meant to be a new repo | 14:52 |
| mestery | fungi: I don't see it in gerrit at all. Shoudl I see it there? | 14:53 |
| mestery | garyk: It was at the top of the service-split.sh script :) I'll make sure the instructions are updated. | 14:53 |
| bknudson | I'm trying to figure out why a bunch of my reviews went into merge conflict -- and what I have to do to get them out -- https://review.openstack.org/#/c/62275/ | 14:54 |
| bknudson | I don't think there's actually a merge conflict since I rebased | 14:54 |
| bknudson | and no conflict | 14:54 |
| fungi | mestery: the neutron-ci group is set to be visible in the gerrit groups list, and can be directly managed at https://review.openstack.org/#/admin/groups/510 | 14:55 |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/storyboard: Added cron plugin to clean old access tokens. https://review.openstack.org/140536 | 14:55 |
| mestery | fungi: I was just remembering that, I've bookmarked it now :) | 14:55 |
| garyk | mestery: this one? https://github.com/dougwig/split-repo/blob/master/service-split.sh | 14:55 |
| mestery | thanks fungi ! | 14:55 |
| sdague | bknudson: where do you think they merge conflicted ? | 14:55 |
| mestery | garyk: Yes! | 14:56 |
| bknudson | sdague: I don't think there is a merge conflict... but it says there's a merge conflict on the review page | 14:56 |
| bknudson | I guess I can just rebase them. | 14:56 |
| garyk | can you point me to the part with the branches? just want to make sure it is done | 14:56 |
| sdague | bknudson: yeh, the gerrit merge resolution seems quite conservative compared with git locally | 14:56 |
| bknudson | there's a bunch in stable/juno where nothing even merged, so not sure how those could have gone into conflict -- https://review.openstack.org/#/c/142545/ | 14:56 |
| sdague | so typically, just rebase if you see that | 14:57 |
| garyk | fek - ok, i see the issue now - i did not have the remote branches as local | 14:57 |
| bknudson | sdague: ok, I'll rebase. | 14:57 |
| garyk | mestery: let me try again | 14:57 |
| mestery | garyk: OK, cool! | 14:57 |
| sdague | bknudson: wow, that's a bunch of backports :) | 14:57 |
| bknudson | sdague: oh, I ran recheck on https://review.openstack.org/#/c/62275/ and now that one's not merge conflict anymore. | 14:57 |
| fungi | dstufft: looks like you're talking about 142322 so we can probably set that in devstack-gate. i'll take a stab at a patch | 14:58 |
| sdague | bknudson: yeh, I don't know | 14:58 |
| dstufft | fungi: yea that's what Iw as using to see if openstack is setuptools 8 ready or not | 14:58 |
| bknudson | ok, I'll just recheck... wasn't sure if I'd be required to actually rebase to clear it or if recheck would do it. | 14:58 |
| fungi | bknudson: that looks like it might have been an error on zuul's part (maybe something wrong with one of the zuul merge workers). i'll take a look at the servers in a sec | 15:01 |
| *** wenlock has joined #openstack-infra | 15:02 | |
| *** mpaolino has quit IRC | 15:03 | |
| *** wuhg has quit IRC | 15:03 | |
| *** dims__ has quit IRC | 15:04 | |
| *** dims__ has joined #openstack-infra | 15:04 | |
| *** Ala has quit IRC | 15:05 | |
| *** yosam has quit IRC | 15:05 | |
| *** Ala has joined #openstack-infra | 15:05 | |
| *** dkranz has quit IRC | 15:06 | |
| *** tomoe_ has quit IRC | 15:06 | |
| *** _nadya_ has joined #openstack-infra | 15:07 | |
| *** david-lyle_afk has quit IRC | 15:07 | |
| *** Ala has quit IRC | 15:08 | |
| *** dims_ has joined #openstack-infra | 15:10 | |
| *** otter768 has joined #openstack-infra | 15:10 | |
| jaypipes | clarkb, fungi, sdague: does Gerrit have rate-limiting on its HTTP API? | 15:10 |
| jaypipes | clarkb, fungi, sdague: for reference, trying to debug this: https://storyboard.openstack.org/#!/story/2000084 | 15:10 |
| *** dims__ has quit IRC | 15:10 | |
| jaypipes | after a little while using Gertty, I'm getting a 401 Unauthorized for some reason from Gerrit. | 15:11 |
| *** mrunge has quit IRC | 15:12 | |
| *** otter768 has quit IRC | 15:15 | |
| fungi | jaypipes: possibly... i'll check the gerrit error log from that point in time and see if it's at all enlightening | 15:15 |
| *** erikwilson has quit IRC | 15:15 | |
| *** mpaolino has joined #openstack-infra | 15:15 | |
| *** bradjones has joined #openstack-infra | 15:16 | |
| *** tonytan4ever has joined #openstack-infra | 15:18 | |
| jaypipes | fungi: thx! | 15:18 |
| *** thedodd has joined #openstack-infra | 15:18 | |
| *** timcline has joined #openstack-infra | 15:19 | |
| *** Ala has joined #openstack-infra | 15:19 | |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/project-config: Pre-release tags use rc instead of c https://review.openstack.org/142814 | 15:20 |
| *** signed8bit has joined #openstack-infra | 15:21 | |
| *** Ala has quit IRC | 15:21 | |
| *** dkranz has joined #openstack-infra | 15:21 | |
| *** emagana has joined #openstack-infra | 15:21 | |
| *** __TheDodd__ has joined #openstack-infra | 15:22 | |
| *** thedodd has quit IRC | 15:23 | |
| fungi | dstufft: thinking more about those error message matches, i think maybe tempest should be the one to set that envvar since it can be run outside the scope of devstack-gate by end users, so setting it within devstack-gate doesn't really help them, only us | 15:23 |
| dstufft | fungi: ok | 15:24 |
| dstufft | fungi: can you do that or is that something I Should do? I don't know much about tempest :) | 15:24 |
| *** Masahiro has joined #openstack-infra | 15:24 | |
| fungi | dstufft: i'll take a stab at it | 15:24 |
| *** Ala has joined #openstack-infra | 15:24 | |
| *** chuckC_ has quit IRC | 15:25 | |
| *** stevemar has joined #openstack-infra | 15:25 | |
| *** andreykurilin has joined #openstack-infra | 15:25 | |
| fungi | mtreinish: we need to set an envvar to silence errors causing problems in tests like this: http://logs.openstack.org/22/142322/1/check/check-tempest-dsvm-full/395143d/console.html#_2014-12-18_13_02_06_216 | 15:25 |
| fungi | mtreinish: quick temperature check, will setting that globally in tempest be acceptable? | 15:25 |
| dstufft | it'll silence RuntimeWarnings that occur in the pkg_resources module | 15:26 |
| fungi | mtreinish: that's one of the remaining bits of fallout from setuptools 8, fwiw | 15:26 |
| dstufft | but not runtime warnings that occur elsewhere | 15:26 |
| dstufft | or any other kind of warning | 15:26 |
| dims_ | dhellmann: fungi: is there a pbr release in the works for the sphinx py33 problems? https://review.openstack.org/#/c/142561/ | 15:27 |
| *** nelsnelson has joined #openstack-infra | 15:27 | |
| fungi | also, looks like we're already up to setuptools 8.2 now | 15:27 |
| dstufft | dims_: should already be released | 15:27 |
| dims_ | dstufft: so, i just need to do rechecks? | 15:27 |
| sdague | dstufft: that python-debian warning seems odd | 15:27 |
| dstufft | mordred did it a little bit ago | 15:27 |
| dstufft | sdague: why? | 15:28 |
| fungi | dims_: if you're still seeing the issue with pbr 0.10.4 then we're missing something else | 15:28 |
| *** amitgandhinz has joined #openstack-infra | 15:28 | |
| sdague | dstufft: because at least on my system the PKG-INFO doesn't look like that | 15:28 |
| dims_ | fungi: just checking if i should start recheck-s | 15:28 |
| *** mattfarina has joined #openstack-infra | 15:28 | |
| sdague | oh, I guess on 14.04 it does | 15:28 |
| *** Masahiro has quit IRC | 15:28 | |
| dstufft | fungi: yea, 8.1 was released instead of 8.0.5 because I added a new class, and then 8.2 was another change that was unrelated to pep 440 | 15:29 |
| fungi | sdague: that's setuptools 8 complaining about non-pep-440 conformant versions of python libraries installed from distro system packages | 15:29 |
| mtreinish | fungi: hmm, you can't just set it in devstack-gate.sh? because of tox? | 15:30 |
| sdague | so 14.04 is on setuptools 3.3 out of the box | 15:30 |
| *** __TheDodd__ is now known as thedodd | 15:31 | |
| sdague | which i guess makes sense, that was the released version in march | 15:31 |
| fungi | mtreinish: we _can_ but that doesn't help anybody else running tempest without devstack-gate | 15:32 |
| *** zz_dimtruck is now known as dimtruck | 15:32 | |
| *** wenlock has quit IRC | 15:32 | |
| fungi | mtreinish: and since it's tempest tests tripping over that warning message, it seems to make sense to address the error as close as possible to where it's occurring | 15:33 |
| fungi | mtreinish: maybe even within tests that are interfacing with pkg_resources | 15:34 |
| dstufft | there's a progrmatic API to disable it | 15:34 |
| fungi | dstufft: right, just trying to make sure we disable it in the best place for the largest number of people/systems | 15:34 |
| *** bhunter71 has joined #openstack-infra | 15:34 | |
| dstufft | oh wait, the programatic api won't help | 15:35 |
| dstufft | because it's subproccesses | 15:35 |
| *** changbl has joined #openstack-infra | 15:36 | |
| dstufft | so you'd still use the env var, you'd just pass the env var in the subprocess call if you wanted to edit the tests themselves | 15:36 |
| fungi | right | 15:36 |
| *** belmoreira has joined #openstack-infra | 15:36 | |
| fungi | maybe make an environ appending decorator | 15:36 |
| sdague | fungi: it's alos generating a ton of noise in devstack log - http://logs.openstack.org/22/142322/1/check/check-tempest-dsvm-full/395143d/logs/devstacklog.txt.gz | 15:36 |
| fungi | if we really want it to be per-test that is | 15:36 |
| sdague | http://logs.openstack.org/22/142322/1/check/check-tempest-dsvm-full/395143d/logs/devstacklog.txt.gz#_2014-12-18_12_34_46_433 | 15:37 |
| mtreinish | fungi: my concern is it's not actually an issue with tempest, it's a system level thing which tempest is just running into | 15:37 |
| sdague | yeh | 15:37 |
| sdague | basically every CLI call to a python program on a 14.04 box will now issue a warning | 15:37 |
| fungi | mtreinish: however, tempest is parsing stdout from tools not under its control. a bit of a grey area | 15:37 |
| sdague | so everything expecting to parse python output is going to run into an issue | 15:38 |
| clarkb | it shouldnt be everything right? jjst code that depends on funky versioned stuff | 15:38 |
| fungi | clarkb: pkg_resources goes through the entire index of installed python packages | 15:39 |
| sdague | clarkb: nope, it's all the thigns | 15:39 |
| fungi | clarkb: so it's complaining about versions of things even if they aren't being imported | 15:39 |
| mtreinish | clarkb: yeah, I don't think the python clients depend on python-debian | 15:39 |
| clarkb | :( | 15:39 |
| sdague | 222 occurances during a devstack run | 15:39 |
| fungi | just wondering if we fix it in devstack-gate and then people running devstack and tempest are on their own to apply the same workaround, or fix it in devstack but people running tempest outside of a devstack install still ahev to work around it, or... | 15:40 |
| dstufft | only if someone upgrades their system setuptools of course | 15:40 |
| sdague | this seems like an extremely bad place to put the deprecation message in setuptools | 15:40 |
| *** isviridov is now known as isviridov_away | 15:42 | |
| clarkb | I am slightly amused python-debian is a thing and that ubuntu munges the version | 15:42 |
| sdague | clarkb: debian munged it I think | 15:42 |
| clarkb | debian has good versions on pypi at least | 15:43 |
| sdague | https://packages.debian.org/wheezy-backports/python-debian | 15:43 |
| sdague | I think because it was an nmu | 15:43 |
| *** r-daneel has joined #openstack-infra | 15:44 | |
| *** markmcclain has joined #openstack-infra | 15:44 | |
| sdague | yeh, there was an nmu to port to python3 | 15:44 |
| *** hdd has quit IRC | 15:44 | |
| fungi | ahh, yep. lots of backports are nmus if they can't be directly pulled from testing | 15:44 |
| sdague | which is kind of a crazy nmu honestly | 15:44 |
| fungi | looks like it got several according to http://metadata.ftp-master.debian.org/changelogs//main/p/python-debian/python-debian_0.1.21+nmu2~bpo70+1_changelog | 15:45 |
| jeblair | fungi, clarkb: awake and looking into jenkins01 | 15:46 |
| fungi | py3k support and a regression fix for Deb822Dict.has_key | 15:46 |
| fungi | and then that got pulled into wheezy backports with a rebuild | 15:46 |
| garyk | mestery: anteaya: the patch has been updated - it now has branches, tags, what else do i need - i am out of wishes | 15:47 |
| clarkb | jeblair: ok I was going to house keep patch pbr if mordred had not already then look into debugging my gear patch | 15:47 |
| fungi | judging from the dates, i'm guessing it was somewhat abandoned for a year or so | 15:47 |
| anteaya | garyk: I need to take a moment and connect with fungi | 15:47 |
| anteaya | garyk: stand by | 15:47 |
| fungi | garyk: can you spin straw into gold? | 15:47 |
| pleia2 | tchaypo: sorry, west coast US so that was about 3AM my time, and have some headache triage to do this morning, so I'll be back in a little while (feel free to email or private message and I'll get back to you :)) | 15:47 |
| sdague | fungi: except there are new versions on pypi? | 15:47 |
| *** cnesa has joined #openstack-infra | 15:48 | |
| sdague | so we end up taking an updated setuptools because we update pip? | 15:48 |
| fungi | sdague: yeah, i'm guessing those weren't suitable for wheezy, and ubuntu imported from wheezy-backports for trusty | 15:48 |
| fungi | sdague: if we run with newer pip/virtualenv then it gets sucked in, or if we pip install -U anything | 15:49 |
| *** dmsimard_away is now known as dmsimard | 15:49 | |
| anteaya | fungi: so after digesting all that transpired last night with the neutron plugin split conversation, I realize that folks might have been operating under the impression that the force push workflow we did with dougwig for the advanced services split it the expected workflow from their end | 15:49 |
| fungi | sdague: though "newer pip/virtualenv" is a forward-looking statement. those don't technically exist for a few more days | 15:49 |
| fungi | anteaya: right. we only fell back on that because of a failure to plan on the part of the neutron developers | 15:50 |
| anteaya | in my mind the empty repos were created too quickly (on a weekend, I never got to reivew any of them) and the force push arrangement was meant as a special case to fix the issue that the repos were created with no content | 15:50 |
| anteaya | fungi: right | 15:50 |
| clarkb | basically devstack situation prior to our pins was latest setuptools | 15:50 |
| fungi | anteaya: well, and a failure to heed/understand the questions about why they were creating new projects without importing anything | 15:50 |
| anteaya | and I now realize that their use of the term "expriement" and not delete old code is what they mean | 15:50 |
| clarkb | now we are working to remove those pins so that everything works regardless of setuptools | 15:50 |
| anteaya | fungi: right | 15:50 |
| anteaya | so I am trying to make sure I do a good job of communicating that now | 15:51 |
| *** wenlock has joined #openstack-infra | 15:51 | |
| garyk | fungi: for you most certainly | 15:51 |
| garyk | anteaya: thanks | 15:51 |
| *** _nadya_ has quit IRC | 15:51 | |
| sdague | clarkb: ok, except now we're going to need a different work around to make it so every python program doesn't spew a pkg_resources error message regardless of if they do anything with it | 15:51 |
| * anteaya moves to -neutron | 15:51 | |
| *** cnesa has quit IRC | 15:52 | |
| ttx | the "zuul job queue" graph on http://status.openstack.org/zuul/ looks a bit funny. | 15:52 |
| fungi | ttx: yep, jeblair's still trying to pin down the cause | 15:53 |
| clarkb | sdague: yes but we dont exactly get to choose our version of setuptools... its something we will have to deal with | 15:54 |
| sdague | ok, this seems like a really bad way for setuptools to complain though, because it's blasting all the way out to the user on completely unrelated activities | 15:55 |
| *** vigneshvar has quit IRC | 15:55 | |
| sdague | I can see pip complaining about it when you try to do a package thing | 15:55 |
| *** andreaf has joined #openstack-infra | 15:55 | |
| sdague | because it's relevant there | 15:56 |
| sdague | it's completely non relevant when just running a python program | 15:56 |
| clarkb | file a bug against setuptools? | 15:56 |
| clarkb | there isnt much I can do to address that. we have been focusing on fixing things under our control | 15:57 |
| mestery | garyk: If it has branches, you're good sir! | 15:57 |
| garyk | mestery: i hope so. it could be our hanukka miracle | 15:57 |
| mestery | garyk: lol | 15:57 |
| fungi | sdague: yeah, right now setuptools is what cares about the version number patterns, so it's what has to complain, but it gets called at a low level | 15:58 |
| dstufft | there isn't a good place for setuptools or pip to really warn | 15:58 |
| dstufft | because 99% of the time an end users is goingt o see them | 15:59 |
| dstufft | who isn't the package author | 15:59 |
| dstufft | and so has no control over it | 15:59 |
| fungi | i'm assuming the hope is that these messages annoy the crap out of end users until they complain at their respective distros long enough to convince them to set different pep-440-compliant version numbers in the egg-info | 15:59 |
| *** dpyzhov has left #openstack-infra | 15:59 | |
| clarkb | I think having a warning is great (it caught pytz issues for example) | 16:00 |
| *** mtanino has joined #openstack-infra | 16:00 | |
| sdague | dstufft: pip can't interogate instead on demand? | 16:00 |
| fungi | more stick, less carrot i guess | 16:00 |
| dstufft | you can't have the pytz warning without this warning | 16:00 |
| clarkb | it is unfortunate that it is so noisy but we wont fix otherwise | 16:00 |
| sdague | clarkb: we're not going to fix it | 16:00 |
| clarkb | we arent? | 16:00 |
| *** isviridov_away is now known as isviridov | 16:00 | |
| sdague | we're going to turn it off because it will be too many complaints from users being confused by it | 16:01 |
| dstufft | afaik the only reason it's being turned off is because some of the tests execute a python subprocess and assert against the output | 16:02 |
| clarkb | we could update python debian | 16:02 |
| fungi | i'll adjust 142322 for now and see if that solves the issue for our infrastructure | 16:02 |
| *** david-lyle has joined #openstack-infra | 16:02 | |
| fungi | dstufft: well, and because it adds a lot of noise in our logs about things not necessarily under our control | 16:02 |
| dstufft | most users aren't going to see warnings from distro packages | 16:02 |
| dstufft | because most users don't force upgrade setuptools | 16:03 |
| dstufft | in the system python | 16:03 |
| *** rmcall has joined #openstack-infra | 16:03 | |
| clarkb | dstufft right being in virtualenvs "fixes" this | 16:03 |
| dstufft | and assumingly when Ubuntu pulls in setuptools 8 they'll fix it in one fashion or another as well for whatever release that is | 16:03 |
| anteaya | I have a question posted to the non existant -requests list, someone who had an infra created CI account would like to make ammendments to the email for 2 different ibm-ci accounts and is unable to self-serve | 16:04 |
| anteaya | how should I advise them to proceed | 16:04 |
| *** andreaf has quit IRC | 16:04 | |
| sdague | dstufft: unless they ever pip -U a thing? | 16:04 |
| clarkb | anteaya ya I saw that. I think ideally they create another account unless we can convert the existing one somehow | 16:05 |
| clarkb | fungi ^ | 16:05 |
| dstufft | in many cases our only way to communicate with package authors is to raise a warning for _their_ users, so that their users go and tell them it's not working | 16:05 |
| dstufft | sdague: pip install -U doesn't upgrade setuptools unless something depends on setuptools | 16:05 |
| dstufft | in the install_requires | 16:05 |
| dstufft | which most projects don't | 16:05 |
| *** _nadya_ has joined #openstack-infra | 16:05 | |
| sdague | so what are we pip install -U that is dragging this in in devstack? | 16:05 |
| clarkb | irc you arent | 16:06 |
| *** _nadya_ has quit IRC | 16:06 | |
| anteaya | clarkb: did the post to more than the non existant -requests list | 16:06 |
| clarkb | instead devstack specifically installs setuptools | 16:06 |
| dstufft | you guys force upgrade setuptools and pip on purpose afaik | 16:06 |
| anteaya | clarkb: or did you get an admin email from the -requests list too? | 16:06 |
| clarkb | dstufft ya | 16:06 |
| openstackgerrit | yolanda.robla proposed openstack-infra/storyboard-webclient: Retrieve task statuses from the backend https://review.openstack.org/138389 | 16:06 |
| *** dkranz has quit IRC | 16:06 | |
| sdague | clarkb: we do for pip | 16:07 |
| clarkb | anteaya I think admin but I didnt look that closely | 16:07 |
| sdague | so pip drags in setuptoos? | 16:07 |
| anteaya | clarkb: kk | 16:07 |
| *** pradk has joined #openstack-infra | 16:07 | |
| clarkb | sdague and the next line is setuptools | 16:07 |
| sdague | oh, right | 16:07 |
| clarkb | sdague and yes pip vendors setuptools for itself so real soon now latest pip will have latest setuptools | 16:07 |
| *** pc_m has quit IRC | 16:07 | |
| sdague | ok, so I guess we just blanket turn this off in devstack | 16:08 |
| fungi | anteaya: i think they also e-mailed the infra ml | 16:08 |
| *** pc_m has joined #openstack-infra | 16:08 | |
| anteaya | fungi: ah | 16:08 |
| tchaypo | pleia2: np, I responded on the review :) | 16:08 |
| fungi | sdague: dstufft: clarkb: 142322,2 will hopefully do it | 16:08 |
| anteaya | so I'll play whack a mole, but what do I tell them? | 16:08 |
| *** cnesa has joined #openstack-infra | 16:09 | |
| anteaya | I dont' think expecting folks with old accounts to make new ones is a good suggestion | 16:09 |
| sdague | I still think the communication should be from pypi => packagers. Spamming thousands of users and making it their problem to yell at packagers (if they even understand the issue) seems bad form. | 16:09 |
| anteaya | we will just end up with confused operators (is there any other kind?) and a boatload of abandoned accounts | 16:09 |
| clarkb | anteaya I agree except for when they want to make changes | 16:09 |
| anteaya | fair enough | 16:09 |
| fungi | anteaya: clarkb: if they don't create new accounts, they have to keep coming back to us to make changes | 16:09 |
| fungi | that | 16:10 |
| anteaya | sigh | 16:10 |
| clarkb | because that is the overhead on us that is the problem | 16:10 |
| anteaya | right | 16:10 |
| clarkb | if the existing account works no problem | 16:10 |
| dstufft | sdague: pypi -> packagers doesn't work ina lot of situations | 16:10 |
| fungi | i mean, we could associate some specific lp openid url with their existing account, but that still implies creatign an lp account which is a hair away from creating a new gerrit account | 16:10 |
| sdague | fungi: so, honestly, we need this to be sticky | 16:10 |
| fungi | sdague: top of stack.sh isn't sticky enough? | 16:11 |
| anteaya | clarkb fungi mind if we get jeblair's thoughts here too? | 16:11 |
| *** pc_m has quit IRC | 16:11 | |
| sdague | no, because stack.sh doesn't drive tempest | 16:11 |
| clarkb | anteaya I think we should | 16:11 |
| sdague | or any number of other scripts that get run | 16:11 |
| jeblair | anteaya: what needs changing? | 16:12 |
| anteaya | if we implement something I'd at least like use to be in sync with what and how | 16:12 |
| *** erikwilson has joined #openstack-infra | 16:12 | |
| *** erikwilson has quit IRC | 16:12 | |
| *** pc_m has joined #openstack-infra | 16:12 | |
| anteaya | jeblair: folks with old ci accounts can't make changes to the account | 16:12 |
| fungi | sdague: but if we don't do it in stack.sh, we still get a log full of warnings from devstack, unrelated to the tempest failures, right? | 16:12 |
| jeblair | anteaya: i mean, what change do they want to make? | 16:12 |
| anteaya | jeblair: for instance to update emails on the account | 16:12 |
| fungi | sdague: oh, i get you, we should probably set it in multiple places | 16:12 |
| *** jgallard_ has quit IRC | 16:12 | |
| anteaya | someone emailed -requests and apparently -infra asking to change the email on 2 ibm-ci accounts | 16:12 |
| sdague | fungi: right, but my very real concern is dealing with all kinds of people thinking that their devstack installs are now broken because of this | 16:12 |
| clarkb | fungi maybe we want those warnings and only suppress where tests fail | 16:13 |
| fungi | sdague: should stack.sh stick it in /etc/environment and then re-source that? | 16:13 |
| sdague | and I have to field that | 16:13 |
| sdague | or just patch setuptools and remove the warning | 16:13 |
| jeblair | anteaya: yeah, i agree with the consensus. after they create a new account we may need to do something like remove emails from the old accounts, or change their names, but that will be a decreasing burden over time. | 16:13 |
| sdague | because... different shells people use | 16:13 |
| fungi | sdague: i don't think we're going to run a custom setuptools (especially given that means a patched pip and virtualenv) | 16:13 |
| anteaya | jeblair: so what do I tell them, to create new ci accounts using the self-serve process? | 16:14 |
| sdague | this creates *so much* work around | 16:14 |
| jeblair | anteaya: yep | 16:14 |
| *** markmcclain has quit IRC | 16:14 | |
| anteaya | jeblair: so it shall be | 16:14 |
| sdague | so I think patching setuptools to delete that warning is the right option | 16:14 |
| jeblair | clarkb: so, wow, i still have no idea what's going on. i have restarted again with even more detailed logging... | 16:14 |
| sdague | or just keep the pin | 16:14 |
| anteaya | jeblair: just tell this one person or make an announcement and change the documentation? | 16:14 |
| *** scottda has joined #openstack-infra | 16:15 | |
| *** andreykurilin has quit IRC | 16:15 | |
| jeblair | anteaya: tell this one person, make a wiki page with instructions, and point the next person at the wiki page | 16:15 |
| fungi | sdague: we'll need to pin pip and virtualenv in that case, and soon probably tox as well | 16:15 |
| scottda | Hi Infra team. Could I get a Core member to look at this review to enable logging of Cinder IRC: https://review.openstack.org/#/c/138965/ | 16:15 |
| jeblair | clarkb: http://paste.openstack.org/show/152785/ | 16:16 |
| *** marun has joined #openstack-infra | 16:16 | |
| clarkb | jeblair: looking | 16:16 |
| jeblair | clarkb: note i added logging to record the return value and current value of connection.input_buffer for the jenkins01 manager connection.... | 16:16 |
| anteaya | jeblair: thanks | 16:17 |
| jeblair | clarkb: what happens is that it triggers the edge poll two extra times, but, as far as i can tell from that logging, it simply returns eagain both times | 16:17 |
| jeblair | clarkb: so there's basically no signal that anything is wrong :/ | 16:17 |
| fungi | sdague: and also eradicate any use of pip install -U and even then i'm not sure wher ethat leaves us if one of our transitive requirements decides to start listing an install requirement on setuptools>=8 (pin those as they come up too i guess) | 16:17 |
| clarkb | fungi: there is no pip install -U, its an explicit update of setuptools | 16:18 |
| fungi | though admittedly, anything with install_requires on any version of setuptools is inherently broken | 16:18 |
| jeblair | clarkb: i have restarted it with an added log entry to record every byte read from that connection, in case i'm missing a way that it could exit that function after a partial read and reset the data | 16:18 |
| clarkb | devstack wants to be testing latest setuptools and honestly for this reason | 16:18 |
| *** andreaf has joined #openstack-infra | 16:18 | |
| clarkb | jeblair: ok | 16:18 |
| jeblair | clarkb: but none of the log statements about any kind of socket errors at all are being triggered | 16:18 |
| clarkb | jeblair: so as far as geard is concerned the connection to that worker is fine and not doing much | 16:18 |
| *** pcrews has joined #openstack-infra | 16:19 | |
| fungi | clarkb: i wasn't just talking about devstack, but i suppose we could pin setuptools<8 pip<6 virtualenv<12 indefinitely in devstack | 16:19 |
| jeblair | clarkb: yeah. the extra poll triggers are weird, but that's the only difference i see with anything else | 16:19 |
| clarkb | fungi: no I am trying to say that the code before our pin was correct :) it updates pip and setuptools to current | 16:19 |
| *** dannywilson has joined #openstack-infra | 16:19 | |
| *** dkranz has joined #openstack-infra | 16:19 | |
| clarkb | fungi: we want that so that when these things happen we know about them and can fix them prior to eg debian releasing | 16:19 |
| *** dannywil_ has joined #openstack-infra | 16:20 | |
| *** dannywilson has quit IRC | 16:20 | |
| jeblair | clarkb, fungi: btw, new process to safely restart geard: SIGSTOP zuul, restart geard, wait a minute or so for geard to get all the CAN_DO packets that register jobs, SIGCONT zuul | 16:21 |
| fungi | clarkb: sure, and i was trying to work through the end implications of sdague's suggestion to indefinitely pin setuptools to an earlier version | 16:21 |
| fungi | i assume he means "pin setuptools<8 until ubuntu 16.04 lts is released" and not "for ever and ever" | 16:21 |
| fungi | jeblair: that's slick! | 16:21 |
| clarkb | fungi: oh I missed that I don't think devstack should ever do that | 16:21 |
| clarkb | fungi: otherwise we deal with all this stuff when we try to update everything to next testbed distro | 16:22 |
| dstufft | or you know, you could just talk to the folks who make setuptools and make a compelling reason for why the end users shouldn't know that something in their environment might start acting wierd because they updated setuptools ;) | 16:22 |
| fungi | clarkb: yep, i'm assuming it would just come with an implicit "the openstack community has decided not to support newer setuptools/pip/virtualenv for a couple years" statement | 16:22 |
| sdague | dstufft: well, you didn't seem to be receptive to that when I brought it up earlier | 16:23 |
| anteaya | in case anyone wants to bless my reply to this person directing them to create new ci accounts with the self-serve process: https://etherpad.openstack.org/p/ptOB62hqhT | 16:23 |
| sdague | dstufft: what will act weird in the case that we have now | 16:23 |
| clarkb | anteaya: might be worth having them ping us when they don't need the old account anymore so we can deactivate it | 16:24 |
| anteaya | clarkb: I can put that in | 16:24 |
| dstufft | sdague: the case you have right now? Nothing because you don't use python-debian, but the same warning that is telling you about python-debian also told you about pytz which was super useful | 16:24 |
| dstufft | and you can't have one without the other really | 16:24 |
| clarkb | anteaya: lgtm | 16:25 |
| anteaya | clarkb: thanks | 16:25 |
| * anteaya waits a few more moments | 16:25 | |
| *** matel has joined #openstack-infra | 16:25 | |
| hashar | jeblair: clarkb: someone pinged at 11am about jenkins01 misbehaving (related to yesterday geard issue). But I guess you are already on it :] | 16:25 |
| sdague | dstufft: that's the part I don't understand honestly, because it seems like we care about pytz because we install it | 16:26 |
| sdague | we don't care about stuff that we never touch | 16:26 |
| dstufft | sdague: you didn't install pytz though, pytz was installed from the system | 16:26 |
| fungi | anteaya: added a suggestion | 16:26 |
| anteaya | fungi: ah and a good one it is | 16:26 |
| dstufft | in order for pkg_resources to know you cared about pytz and not python-debian it'd have to parse the ast of the things you're importing, figure out where they came from, and then warn | 16:26 |
| openstackgerrit | yolanda.robla proposed openstack-infra/storyboard-webclient: Retrieve task statuses from the backend https://review.openstack.org/138389 | 16:26 |
| *** Guest51388 has quit IRC | 16:27 | |
| anteaya | okay I will send that off then | 16:27 |
| *** MaxV has quit IRC | 16:28 | |
| *** bradjones has quit IRC | 16:28 | |
| sdague | dstufft: you are setting the legacy version to 0 for compare reasons, right? | 16:28 |
| sdague | so we actually don't need to care | 16:28 |
| sdague | because if we attempted to install or upgrade we'd get the pypi version | 16:28 |
| sdague | for the pytz case | 16:29 |
| dstufft | sdague: less than zero | 16:29 |
| sdague | ok, that's fine | 16:29 |
| *** erikwilson has joined #openstack-infra | 16:29 | |
| *** erikmwilson has quit IRC | 16:30 | |
| *** erikwilson is now known as erikmwilson_ | 16:30 | |
| *** erikmwilson has joined #openstack-infra | 16:30 | |
| sdague | dstufft: so, this user confusion thing is honestly a big deal. We've got tons of people out there using devstack to setup environments. And the most consistent way I can see of turning off a thing that none of them care about is just patching out that piece of setuptools after install. | 16:30 |
| *** nfedotov has quit IRC | 16:30 | |
| *** erikwilson has joined #openstack-infra | 16:30 | |
| openstackgerrit | Matthew Treinish proposed openstack-infra/project-config: Add nodepool script to preseed testrepository from subunit2sql https://review.openstack.org/136234 | 16:30 |
| *** erikwilson is now known as erikmwilson_ | 16:31 | |
| *** kumartin has joined #openstack-infra | 16:31 | |
| *** garyk1 has joined #openstack-infra | 16:31 | |
| *** erikmwilson_ has quit IRC | 16:31 | |
| mriedem | i'm looking at bug https://bugs.launchpad.net/os-apply-config/+bug/1403510 and it's impacting sqlalchemy-migrate (which i have a change in to get merged to unblock some CI testing), | 16:31 |
| uvirtbot | Launchpad bug 1403510 in zaqar "gate tests failed with "NameError: name 'StandardError' is not defined"" [High,In progress] | 16:32 |
| *** bradjones has joined #openstack-infra | 16:32 | |
| mriedem | i see some projects are just disabling sphinx in their requirements for now until the bug is fixed, | 16:32 |
| dstufft | sdague: the most consistent way not to run into situations like this is to use virtual environments instead of frankensystems that are half system packages and half pip packages ;) | 16:32 |
| *** garyk1 has quit IRC | 16:32 | |
| *** garyk1 has joined #openstack-infra | 16:32 | |
| mriedem | are there other efforts going on to fix https://bugs.launchpad.net/os-apply-config/+bug/1403510 globally? | 16:32 |
| *** erikwilson has joined #openstack-infra | 16:32 | |
| *** erikmwilson has quit IRC | 16:32 | |
| *** erikwilson is now known as erikmwilson | 16:32 | |
| mriedem | or is there going to be a pbr release with https://review.openstack.org/#/c/142561/ ? | 16:32 |
| *** erikwilson has joined #openstack-infra | 16:33 | |
| *** garyk has quit IRC | 16:33 | |
| anteaya | I'm wondering which wikipage I should link to for instructions for creating a self-service account for old ci account holders, this wikipage comes up well when searching for "third party" in the wiki search: https://wiki.openstack.org/wiki/ThirdPartySystems | 16:33 |
| openstackgerrit | Dmitry Teselkin proposed openstack-infra/system-config: Fix vhost configuration sections https://review.openstack.org/132170 | 16:33 |
| anteaya | I could add a link to the bottom | 16:33 |
| * anteaya waits for opposition | 16:33 | |
| sdague | dstufft: hence my previous comment that you didn't seem very receptive to addressing this | 16:33 |
| dstufft | sdague: my problem is, you have two ways for users to get confused here, one is that things don't work quite right when pkg_resources does things (like adding eggs to path and such) and for that you want a warning, and it's not just install time activities because of the way setuptools work | 16:33 |
| dstufft | the other way is the one you're concerned about | 16:33 |
| *** pcaruana has quit IRC | 16:33 | |
| sdague | yeh, because the other way around is going to be tons of confused users, most of whom know basically nothing about python, and all then sudden their system is emitting warnings on every python command | 16:34 |
| dstufft | like for instance, there was a pytz problem because of devstack, and it occured when attempting to import Babel because babel had dependency on >=0a | 16:34 |
| mriedem | dhellmann: mordred: do you know if there is going to be a pbr release for https://review.openstack.org/#/c/142561/ ? | 16:35 |
| dstufft | pytz>0a that is | 16:35 |
| dstufft | mriedem: there is one already | 16:35 |
| mriedem | dhellmann: mordred: if that fixes the busted py33 jobs? | 16:35 |
| dstufft | 0.10.4 IIRC | 16:35 |
| mriedem | dstufft: hasn't hit the mirrors yet? | 16:35 |
| mriedem | i didn't even see that commit in the logs on git.o.o for pbr | 16:35 |
| dstufft | sdague: setuptools 8.0 didn't start out with that warning, it got that warning because things broke here in openstack and people were like wat? | 16:36 |
| mriedem | http://git.openstack.org/cgit/openstack-dev/pbr/log/ | 16:36 |
| fungi | mriedem: it's in the feature/0.10 branch not master | 16:36 |
| mriedem | ah | 16:36 |
| dstufft | because the breakage may not be obvious | 16:36 |
| *** annegent_ has joined #openstack-infra | 16:36 | |
| fungi | mriedem: we weren't releasing off master because we didn't want to complicate things by releasing a version with all the semver changes yet | 16:36 |
| sdague | dstufft: so why isn't it something which is off by default? | 16:36 |
| *** Ala has quit IRC | 16:37 | |
| sdague | and the env can be set on for debugging | 16:37 |
| *** rkukura_ has joined #openstack-infra | 16:37 | |
| *** Ala has joined #openstack-infra | 16:37 | |
| fungi | sdague: i believe it was a reaction to the multiple statements from our community of "why did setuptools break this without a deprecation warning?" | 16:37 |
| fungi | where this is the closest thing to a warning setuptools can really provide | 16:38 |
| mriedem | fungi: hmm, ok, looks like it got into this py33 job but still failed https://jenkins07.openstack.org/job/gate-sqlalchemy-migrate-python33/3/console - i'm checking the log | 16:38 |
| dstufft | sdague: because it's sort of tautological, if you know about the issue you don't need the warning | 16:38 |
| *** koolhead17 has quit IRC | 16:38 | |
| *** dimtruck is now known as zz_dimtruck | 16:38 | |
| sdague | dstufft: but if we know something broke horribly, we can turn on debug to figure out why | 16:38 |
| mriedem | fungi: different errors, at least that's 'good' | 16:38 |
| sdague | so, sorry, I was not working for the setuptools 8 fire drill, so I don't know all the details | 16:38 |
| dstufft | sdague: well setuptools doesn't have an overall debug mode, maybe it should, and if it did maybe it'd be better to have this covered by that | 16:39 |
| *** rkukura has quit IRC | 16:40 | |
| *** rkukura_ is now known as rkukura | 16:40 | |
| sdague | so adhoc it, because this warning is adhoc (as is the way to turn it off) | 16:40 |
| openstackgerrit | yolanda.robla proposed openstack-infra/storyboard-webclient: Retrieve task statuses from the backend https://review.openstack.org/138389 | 16:40 |
| sdague | dstufft: do you understand my user confusion concerns here? | 16:41 |
| dstufft | the way to turn it off is a standard python feature that exists exactly for this thing | 16:41 |
| dstufft | this kind of thing* | 16:41 |
| sdague | except something like this should default off | 16:41 |
| dstufft | sdague: I understand yes, I'm just not sure that those user's confusion outweighs the confusion from people for whom this message helps | 16:42 |
| dstufft | the problem with default to off is that it requires you to know you need to turn it on | 16:42 |
| sdague | dstufft: that's because you don't have to field the issues | 16:42 |
| sdague | dstufft: and *all* the people that are in a position to fix things, are the people that will know how to turn it on | 16:42 |
| sdague | so it scopes the warning to the correct audience | 16:42 |
| mriedem | how easy is it to get a project's bugs tracked in launchpad? | 16:43 |
| dstufft | sdague: well the problem is that a lot of them don't know, at least right now, and the problems can be non obvious and end up effecting end users | 16:44 |
| fungi | mriedem: you create a new project in launchpad, and click on the "configure bug tracking" link there | 16:44 |
| mriedem | fungi: https://bugs.launchpad.net/sqlalchemy-migrate | 16:44 |
| mriedem | already exists | 16:44 |
| fungi | mriedem: then i'm guessing the question you asked isn't the question you meant to ask | 16:45 |
| *** tkelsey has joined #openstack-infra | 16:45 | |
| mriedem | well, how do i go about getting sqla-migrate bugs tracked in launchpad? i didn't create the project. | 16:45 |
| mriedem | mordred did i believe | 16:45 |
| fungi | mriedem: oh, whoever created that project will need to do it or will need to add you as an admin for that project | 16:45 |
| openstackgerrit | omri marcovitch proposed openstack-infra/system-config: Add Third Party FAQ paragraph https://review.openstack.org/141817 | 16:46 |
| openstackgerrit | Clark Boylan proposed openstack-dev/pbr: Write and read more complex git sha info https://review.openstack.org/141666 | 16:46 |
| openstackgerrit | Clark Boylan proposed openstack-dev/pbr: Only import sphinx during hook processing https://review.openstack.org/142840 | 16:46 |
| openstackgerrit | Clark Boylan proposed openstack-dev/pbr: Properly check for git before getting git dir https://review.openstack.org/142841 | 16:46 |
| fungi | mriedem: yep https://launchpad.net/sqlalchemy-migrate says mordred will need to do it | 16:46 |
| clarkb | fungi: dhellmann mordred dstufft ^ those are the house keeping changes to pbr from 0.10.4 to master. Some scrutiny would be good as the merge conflict resolution was not entirely trivial | 16:46 |
| mriedem | fungi: ok, thanks, i'm prepared to make an offering... | 16:47 |
| fungi | mriedem: or else mordred will need to delegate that to you so that you can do it | 16:47 |
| *** vigneshvar has joined #openstack-infra | 16:47 | |
| fungi | mriedem: i think he's on a plane, or three, so no idea when to expect him popping up | 16:47 |
| mriedem | does he have clones? | 16:47 |
| mriedem | or he just exists in multiple planes? | 16:48 |
| mriedem | like, metaphysical planes i mean, not airplane | 16:48 |
| fungi | mriedem: he exists in a state of quantum flux until you collapse his wave function | 16:48 |
| clarkb | I left out the postN stuff as the semver changes should cover that and I left out the dropping of gitshas because that will require changes to semver | 16:49 |
| *** yamamoto has quit IRC | 16:50 | |
| clarkb | now time for gear | 16:50 |
| *** sputnik13 has quit IRC | 16:50 | |
| mriedem | fungi: i think i'll just take the relatively easier route and get sqla-migrate to not blow up on py33 :) | 16:52 |
| anteaya | jeblair clarkb fungi pleia2 instructions for creating a new self serve CI account if you already have one: https://wiki.openstack.org/wiki/OldtoNewGerritCIAccount | 16:52 |
| mriedem | followed by the harder task of getting a migrate core to review it | 16:52 |
| matel | guys, only one chage is missing for citrix to use upstream project-config, could you take a look at this change: https://review.openstack.org/#/c/136700/ | 16:53 |
| *** armax has joined #openstack-infra | 16:54 | |
| jeblair | clarkb: okay i have logs of data received by gear from jenkins01 that did not result in a packet being returned. | 16:54 |
| clarkb | jeblair: I think I may see the bug in my gear change. It has to do with setting raw_bytes=b'' when segment is "false". We could be cutting a packet off in the middle we then assume the next bytes read are a packet header but they are not and things go crazy | 16:55 |
| jeblair | clarkb: i have to afk for a bit, but i think when i get back i can make progress | 16:55 |
| fungi | jeblair: jenkins-gearman plugin bug then you think? | 16:55 |
| clarkb | jeblair: that is good news | 16:55 |
| jeblair | fungi: it's a possibility, i have yet to reconstruct the packet data to see if it's valid | 16:55 |
| fungi | oh! received by gear _from_ jenkins01 | 16:55 |
| fungi | but still, yeah, maybe | 16:56 |
| jeblair | fungi: yep | 16:56 |
| *** sweston has joined #openstack-infra | 16:57 | |
| clarkb | jeblair: I think we may want to return None there but not change raw_bytes that way the next read will give us the data we need? or we would have to treat that as an error reconnect condition? I need to understand a bit more the situation where segment could be nil | 16:57 |
| clarkb | eg wht situation does recv return b'' and not raise an error with errno EAGAIN | 16:59 |
| *** dims_ has quit IRC | 16:59 | |
| *** dims__ has joined #openstack-infra | 16:59 | |
| *** andreaf has quit IRC | 17:00 | |
| openstackgerrit | Ivan Udovichenko proposed openstack-infra/system-config: Remove hardcoded variables from openstack_project module (Wiki) https://review.openstack.org/141374 | 17:00 |
| *** zz_gondoi is now known as gondoi | 17:00 | |
| *** andreaf has joined #openstack-infra | 17:01 | |
| *** Clint has joined #openstack-infra | 17:01 | |
| *** ivar-lazzaro has joined #openstack-infra | 17:01 | |
| *** mrmartin has quit IRC | 17:01 | |
| *** Sincler has quit IRC | 17:02 | |
| *** Sincler has joined #openstack-infra | 17:02 | |
| *** dims__ has quit IRC | 17:03 | |
| *** bdpayne has joined #openstack-infra | 17:04 | |
| *** jerryz_ has quit IRC | 17:04 | |
| *** gyee has joined #openstack-infra | 17:05 | |
| dstufft | sdague: ok, talked it over with Jason | 17:06 |
| dstufft | sdague: we're going to keep the warning, but silence it by default | 17:06 |
| sdague | dstufft: awesome, thank you | 17:06 |
| fungi | i will cease the workarounds i've been trying to hammer into devstack-gate in that case | 17:07 |
| mordred | mriedem: on it | 17:07 |
| fungi | thanks dstufft! | 17:07 |
| TheJulia | /win 2 | 17:07 |
| *** belmoreira has quit IRC | 17:07 | |
| mriedem | mordred: great, thanks | 17:07 |
| *** EmilienM has quit IRC | 17:08 | |
| openstackgerrit | Mate Lakat proposed openstack-infra/project-config: XenServer: Enable injection of XVA and ISO urls https://review.openstack.org/136700 | 17:08 |
| *** fandi has quit IRC | 17:08 | |
| mordred | mriedem: what is it I'm doing? | 17:08 |
| mriedem | mordred: hook up sqlalchemy-migrate to launchpad for bug tracking | 17:08 |
| mriedem | so i can open sqla-migrate bugs in LP | 17:08 |
| *** thedodd has quit IRC | 17:09 | |
| *** EmilienM has joined #openstack-infra | 17:09 | |
| mordred | cool. is there a launchpad team that should be driver/owner? | 17:09 |
| dstufft | https://bpaste.net/show/b2c4153327b8 | 17:10 |
| mordred | mriedem: I turned on the bug tracker and made you bug supervisor | 17:10 |
| *** avozza is now known as zz_avozza | 17:11 | |
| *** otter768 has joined #openstack-infra | 17:11 | |
| mriedem | mordred: thanks | 17:11 |
| mordred | clarkb: thanks for the forward port ... you didn't want to use the merge commit I already had as a base? | 17:11 |
| clarkb | mordred: I updated your existing change 141667 iirc | 17:11 |
| *** Masahiro has joined #openstack-infra | 17:13 | |
| fungi | dstufft: that looks ideal. maybe exposing an option (command line, config, whatever) in pip that can elicit those warnings would also be helpful | 17:13 |
| *** thedodd has joined #openstack-infra | 17:14 | |
| zaro | morning | 17:14 |
| anteaya | morning zaro | 17:14 |
| fungi | that gets the trigger up at the tool users are interacting with | 17:14 |
| *** jpich has quit IRC | 17:15 | |
| mordred | clarkb: k. I meant this one: https://review.openstack.org/#/c/142323/ | 17:15 |
| mordred | which is a merge commit and does the get rid of git sha part already as well | 17:15 |
| zaro | anteaya: want to double check with you regarding ls-members command. not sure if there was a real resolution there. did you still need that or can you live without it? | 17:16 |
| mordred | I was going to update it this morning - would you prefer a null merge just for tags and then cherry-picks? | 17:16 |
| *** otter768 has quit IRC | 17:16 | |
| *** fandi has joined #openstack-infra | 17:16 | |
| *** ayoung is now known as ayoung-afk | 17:16 | |
| clarkb | mordred: yes null merge for tags would be best then do cherry-picks | 17:16 |
| clarkb | mordred: I think that is easier to review | 17:17 |
| anteaya | zaro: it appears the group has decided I can live without it | 17:17 |
| *** zz_dimtruck is now known as dimtruck | 17:17 | |
| anteaya | zaro: so if I am requested to get this info and can't, it won't be a surprise to anyone | 17:17 |
| *** dims__ has joined #openstack-infra | 17:17 | |
| *** Masahiro has quit IRC | 17:17 | |
| dstufft | fungi: pip will probably just hardcode it to on | 17:18 |
| sdague | yeh, I think pip hardcoding on sounds completely reasonable | 17:18 |
| dstufft | fungi: pip is different than setuptools in that pip isn't invoked except in a "manage my packages scenario" | 17:19 |
| fungi | anteaya: zaro: yeah, the stated use case was to be able to find out a third-part ci's ssh username for them, which in theory they already know because they've configured it in their system | 17:19 |
| *** dtantsur is now known as dtantsur|afk | 17:19 | |
| dstufft | whereas pkg_resources is invoked for all sorts of random shit | 17:19 |
| *** thedodd has quit IRC | 17:19 | |
| *** thedodd has joined #openstack-infra | 17:19 | |
| fungi | dstufft: yeah, makes sense | 17:19 |
| *** shashankhegde has joined #openstack-infra | 17:19 | |
| anteaya | so if they can't find it, then we can't help them until they do | 17:19 |
| sdague | fungi: it's probably worth updating your patch to include a comment on how to turn on that warning if we want to | 17:20 |
| fungi | anteaya: i see that as a good thing. they can go away and learn about their system, then come back | 17:20 |
| sdague | it will be a good place to keep track of it, as it's right where devstack does the setuptools install | 17:20 |
| *** kumartin has quit IRC | 17:20 | |
| fungi | anteaya: keeps us from spoon-feeding them information. give someone a fish vs teach someone to fish ;) | 17:20 |
| *** hashar has quit IRC | 17:21 | |
| fungi | sdague: well, the setuptools pin revert is not the cleanest place to add that, but i'll add a separate patch for it | 17:22 |
| fungi | doing that now | 17:22 |
| sdague | fungi: ok, works for me | 17:22 |
| anteaya | well we have offloaded the screaming, so perhaps the teaching to fish will have room to happen | 17:22 |
| anteaya | I couldn't think with all the screaming | 17:22 |
| *** stevemar has quit IRC | 17:23 | |
| *** stevemar has joined #openstack-infra | 17:23 | |
| fungi | anteaya: for that matter, it's actually an issue specifically for people who have created their own gerrit account for their system, so there's a very specific url they can go to in gerrit to see their username as long as they log in | 17:23 |
| fungi | as well as their account id, all that | 17:24 |
| anteaya | right | 17:24 |
| *** zz_avozza is now known as avozza | 17:25 | |
| openstackgerrit | Matt Riedemann proposed openstack-infra/project-config: Revert "sqlalchemy-migrate: enable the py33 gate." https://review.openstack.org/142850 | 17:27 |
| *** achanda has joined #openstack-infra | 17:28 | |
| openstackgerrit | Matt Riedemann proposed openstack-infra/project-config: Revert "sqlalchemy-migrate: enable the py33 gate." https://review.openstack.org/142850 | 17:28 |
| zaro | fungi, anteaya : ok. i'll remove that item from the meeting agenda. I'll keep the ls-members registered-users fix on my todo list and will let you know when it appears on upstream. | 17:28 |
| anteaya | zaro: thanks, but as we identified, it might just be avaialbe to gerrit admins, which wouldn't allow me access anyway | 17:29 |
| anteaya | never hurts to be aware though | 17:30 |
| anteaya | so thanks | 17:30 |
| dstufft | sdague: https://github.com/jaraco/setuptools/pull/23 | 17:30 |
| *** ociuhandu has quit IRC | 17:30 | |
| *** ociuhandu has joined #openstack-infra | 17:30 | |
| zaro | anteaya: that's not necessarily true if i implement it :) doesn't really make sense to make that command exclusive to admins anyways because it's already available to all registered users from the UI. | 17:31 |
| zaro | anteaya: i mean the info is available. | 17:32 |
| BobBall | jeblair: Anita suggested she'd like your input on https://review.openstack.org/#/c/136700/ - Would you be able to have a look? Mate is leaving Citrix tomorrow so it'd be good to get this patch merged before he goes. | 17:32 |
| openstackgerrit | Matt Riedemann proposed openstack-infra/project-config: Revert "sqlalchemy-migrate: enable the py33 gate." https://review.openstack.org/142850 | 17:32 |
| *** markmcclain has joined #openstack-infra | 17:32 | |
| anteaya | zaro: okay I leave it with you | 17:34 |
| anteaya | sdague: if you could be so kind as to review: https://review.openstack.org/#/c/140409/ which creates a new stackforge repo from code split from neutron | 17:34 |
| anteaya | andreas is offline else I would include him too as he has been following the process as well | 17:35 |
| *** teran has quit IRC | 17:38 | |
| anteaya | BobBall: thanks, yes I like nodepool operators to know what is being changed in nodepool scripts | 17:38 |
| *** ssam2 has quit IRC | 17:40 | |
| BobBall | anteaya: of course :) | 17:41 |
| anteaya | :D | 17:41 |
| fungi | sdague: is https://review.openstack.org/142855 what you were wanting? | 17:42 |
| *** mrmartin has joined #openstack-infra | 17:42 | |
| sdague | fungi: yes, thanks | 17:43 |
| fungi | dstufft: ^ | 17:43 |
| *** cody-somerville has joined #openstack-infra | 17:43 | |
| *** reed has joined #openstack-infra | 17:44 | |
| fungi | i did also test that PYTHONWARNINGS=$PYTHONWARNINGS,whatever does work as expected even if PYTHONWARNINGS is initially unset | 17:44 |
| fungi | so this lets us append as needed | 17:44 |
| fungi | future-proofing | 17:45 |
| *** shashankhegde has quit IRC | 17:46 | |
| *** gyee has quit IRC | 17:48 | |
| *** matel has quit IRC | 17:48 | |
| clarkb | woo my cherry pick of dhellmans change was super wrong :) | 17:49 |
| clarkb | I am cleaning it up | 17:49 |
| fungi | i hadn't gotten a chance to look yet, but cherry-picking refactors are never trivial | 17:50 |
| *** yamamoto has joined #openstack-infra | 17:50 | |
| *** funzo has joined #openstack-infra | 17:51 | |
| *** garyk1 has quit IRC | 17:51 | |
| openstackgerrit | Merged openstack-dev/hacking: Remove Git commit message checks https://review.openstack.org/142585 | 17:53 |
| *** HeOS has quit IRC | 17:54 | |
| *** yamamoto has quit IRC | 17:56 | |
| openstackgerrit | Merged openstack-dev/hacking: Remove complex import rules https://review.openstack.org/142586 | 17:57 |
| *** amuller has quit IRC | 17:58 | |
| *** derekh has quit IRC | 17:58 | |
| clarkb | wow untangling this is a lot more fun on master | 18:00 |
| *** hdd has joined #openstack-infra | 18:00 | |
| openstackgerrit | Matthew Treinish proposed openstack-infra/project-config: Add nodepool script to preseed testrepository from subunit2sql https://review.openstack.org/136234 | 18:01 |
| clarkb | the docutiles things are more tightly coupled to git operations in packaging.py | 18:01 |
| mtreinish | fungi: so I pushed the subunit2sql release which fixes the performance issues and other bugs with sql2subunit, so I think ^^^ is good to go now | 18:01 |
| *** sabeen has quit IRC | 18:02 | |
| mtreinish | heh, although once we land that we'll have to update d-g so it pulls the right stream from .testrepository | 18:02 |
| *** thedodd has quit IRC | 18:03 | |
| *** yaguang has quit IRC | 18:04 | |
| *** thedodd has joined #openstack-infra | 18:04 | |
| *** thedodd has quit IRC | 18:04 | |
| *** thedodd has joined #openstack-infra | 18:04 | |
| *** tkelsey has quit IRC | 18:07 | |
| *** AJaeger has joined #openstack-infra | 18:07 | |
| AJaeger | mestery, I'll do a final review of the VMware-NSX patch an will approve it now unless you should loud STOP... | 18:08 |
| *** dims__ has quit IRC | 18:08 | |
| *** pblaho has quit IRC | 18:08 | |
| mestery | AJaeger: Works for me, thank you! I'm onboard with it merging. | 18:08 |
| *** rmcall has quit IRC | 18:08 | |
| *** thedodd has quit IRC | 18:09 | |
| *** e0ne_ has quit IRC | 18:09 | |
| *** rmcall has joined #openstack-infra | 18:09 | |
| *** dims__ has joined #openstack-infra | 18:10 | |
| dstufft | fungi: was there another PR that added the PYTHONWARNINGS that I missed? Or how is this passing -> https://review.openstack.org/#/c/142322/ | 18:11 |
| *** koolhead17 has joined #openstack-infra | 18:12 | |
| *** koolhead17 has joined #openstack-infra | 18:12 | |
| *** thedodd has joined #openstack-infra | 18:12 | |
| AJaeger | mestery: approved - thanks for your confirmation | 18:13 |
| mestery | AJaeger: yw, and thanks! | 18:14 |
| *** jaypipes is now known as jaypipes-afk | 18:14 | |
| fungi | dstufft: great question... it installed latest setuptools http://logs.openstack.org/22/142322/3/check/check-tempest-dsvm-full/fad5422/logs/devstacklog.txt.gz#_2014-12-18_17_23_34_192 | 18:14 |
| fungi | dstufft: though i see no "parsed as a legacy" in the logs | 18:15 |
| *** ivar-lazzaro has quit IRC | 18:15 | |
| dstufft | that is slightly confusing | 18:16 |
| dstufft | they wouldn't be run on a different machine that doesn't have python-debian installed on them would they? | 18:16 |
| *** dimtruck is now known as zz_dimtruck | 18:16 | |
| fungi | dstufft: good call... http://logs.openstack.org/22/142322/3/check/check-tempest-dsvm-full/fad5422/logs/pip-freeze.txt.gz | 18:17 |
| dstufft | fungi: tests pass, merge it quick! ;p | 18:17 |
| dstufft | at least it validates that setuptools 8 works once you remove the warnings | 18:18 |
| *** shashankhegde has joined #openstack-infra | 18:18 | |
| fungi | dstufft: so, as the fates would have it, the failures we saw for those jobs all ran in hpcloud, the successes of the same jobs all ran in rackspace | 18:19 |
| *** dims_ has joined #openstack-infra | 18:19 | |
| fungi | dstufft: hence, booted on snapshots derived from different base images | 18:19 |
| dstufft | fungi: obviously that's because I work at Rackspace | 18:19 |
| fungi | dstufft: and because hpcloud uses official ubuntu images rather than rackspace-modified images ;) | 18:20 |
| clarkb | I think I would like to not remove the pin until pbr master mostly works | 18:20 |
| *** dims__ has quit IRC | 18:20 | |
| dstufft | fungi: sure, you with your facts! | 18:21 |
| anteaya | AJaeger: you might get something from my soapbox speech in -neutron earlier: http://eavesdrop.openstack.org/irclogs/%23openstack-neutron/%23openstack-neutron.2014-12-18.log starting from 2014-12-18T15:52:08 <anteaya> so for the channel in general | 18:21 |
| fungi | clarkb: that's fair. but we're closer now... mostly waiting on a setuptools 8.2.1 or 8.3 or 9 or something | 18:22 |
| pc_m | hi: Is there a know issue with check-tempest-dsvm-neutron-full showing SSH timeout connecting to cirros VM? | 18:22 |
| clarkb | the more I change the more tests fail :) | 18:22 |
| clarkb | fungi: ya I am just finding that the more recent changes to master make dhellmanns change much more complicated there | 18:22 |
| clarkb | mordred: ^ any chance you want to stab at that as I am failing? | 18:22 |
| *** boris-42 has quit IRC | 18:23 | |
| *** harlowja_away is now known as harlowja | 18:23 | |
| dstufft | fungi: 2001: A setuptools Odyssey | 18:23 |
| clarkb | mordred: basicall the docutils classes in pbr/packaging.py have changed since 0.10.0 and we need to put current stuff in pbr/builddoc.py. But current stuff uses the write changelog method and generate authors and checks if things are a git dir | 18:23 |
| *** melwitt has joined #openstack-infra | 18:23 | |
| clarkb | mordred: so there is more coupling to packaging.py than before (not as easy to just split out get boolean option) | 18:23 |
| openstackgerrit | Merged openstack-infra/project-config: Add VMware-NSX to StackForge https://review.openstack.org/140409 | 18:24 |
| fungi | i'm sorry dstufft but you can't do that | 18:24 |
| * fungi is now known as fungi-9000 | 18:24 | |
| * anteaya pictures fungi with one red eye | 18:24 | |
| jeblair | clarkb: (btw, re your change, i was thinking that the admin command push also needs to happen in the try/finally) | 18:25 |
| * nibalizer would like fungi to open the bay doors | 18:25 | |
| *** dims_ has quit IRC | 18:26 | |
| * fungi starts to sing "daisy bell" | 18:26 | |
| *** dims__ has joined #openstack-infra | 18:28 | |
| mordred | clarkb: yah, I'll be back online soon | 18:29 |
| clarkb | mordred: let me link you the problematic lines | 18:30 |
| *** koolhead17 has quit IRC | 18:30 | |
| *** pblaho has joined #openstack-infra | 18:30 | |
| openstackgerrit | Matt Riedemann proposed openstack-infra/project-config: Make gate-sqlalchemy-migrate-python33 non-voting https://review.openstack.org/142850 | 18:30 |
| clarkb | mordred: https://git.openstack.org/cgit/openstack-dev/pbr/tree/pbr/packaging.py#n800 the three lines starting there | 18:30 |
| openstackgerrit | Travis Tripp proposed openstack/requirements: Angular xstatic for Version 1.3.7 https://review.openstack.org/142869 | 18:30 |
| *** koolhead17 has joined #openstack-infra | 18:31 | |
| clarkb | mordred: if I delete those it appears to be mostly ok | 18:31 |
| clarkb | mordred: except that of course that feature is removed | 18:31 |
| fungi | zaro: are you aware of any rate limiting in gerrit's http api which could cause https://storyboard.openstack.org/#!/story/2000084 ? | 18:32 |
| clarkb | mordred: if I pull out all the git related things into a pbr/git.py file and import that in builddoc.py and packaging.py I get test failure I don't understand | 18:32 |
| clarkb | (I updated the imports and paths in the tests too) | 18:33 |
| *** koolhead17 has quit IRC | 18:35 | |
| clarkb | I need to take a break then will come back to this | 18:36 |
| *** sputnik13 has joined #openstack-infra | 18:36 | |
| *** mpaolino has quit IRC | 18:36 | |
| Daviey | clarkb: can you pastebin the original errors you are seeing? | 18:38 |
| zaro | fungi: no rate limiting that i am aware of. | 18:38 |
| fungi | zaro: okay, might just be something with gertty | 18:39 |
| fungi | jaypipes-afk: is there any chance you have a proxy between your gertty and our gerrit? | 18:39 |
| clarkb | Daviey: http://logs.openstack.org/40/142840/1/check/gate-pbr-python27/6bffe66/console.html those are the original errors. They happen because there are now tests that check new functionality in the builddoc classes | 18:39 |
| * fungi found nothing in the gerrit error_log which related to that error | 18:39 | |
| clarkb | Daviey: adding that new functionality requires completely reorging the code to avoid circular imports | 18:39 |
| *** sputnik13 has quit IRC | 18:40 | |
| *** Ryan_Lane has joined #openstack-infra | 18:40 | |
| zaro | fungi: i think that might be due to client being idle too long? i think i see that in the web client as well, need to refresh the client then works again. | 18:40 |
| clarkb | jeblair: fwiw ps3 of my change seems to work relatively well under local testing but your ps4 runs a single job then gets stuck | 18:40 |
| Daviey | clarkb: Ah, not the error i was expecting. pbr is really sensitive to lack of entropy and has a non-obvious error.. I was expecting it to be that | 18:41 |
| *** jcoufal_ has quit IRC | 18:41 | |
| fungi | Daviey: those stem ultimately from the fact that docutils' answer to py3k is to run 2to3 at install time, and pbr is winding up with untranslated docutils loaded into the execution space. an ugly interaction indeed | 18:41 |
| clarkb | ya that happens too, we have haveged installed on the slaves | 18:41 |
| clarkb | and that seems to mitigate the entropy issues during testing | 18:41 |
| *** patrickeast has joined #openstack-infra | 18:42 | |
| Daviey | Yeah, i saw mention of workin in hpcloud but failing in rackspace which made me wonder if the different clouds have different entropy providers | 18:42 |
| fungi | Daviey: do you have a bug filed on the entropy issue? i'm not immediately coming up with reasons why pbr needs entropy | 18:42 |
| *** dizquierdo has quit IRC | 18:42 | |
| clarkb | fungi: it generates gpg keys when testing | 18:43 |
| Daviey | fungi: There is a bug that someone else raised... It was on my mental todo to make the error more obvious | 18:43 |
| fungi | clarkb: oh! its testsuite | 18:43 |
| fungi | got it | 18:43 |
| clarkb | fungi: its part of the semver stuff. I am honestly pretty close to wondering if we should put all the semver stuff on a feature branch and clean up master | 18:43 |
| clarkb | but that won't help current issues so meh | 18:43 |
| *** boris-42 has joined #openstack-infra | 18:44 | |
| fungi | yeah, it will want to test signed keys, but we _could_ just fake that out. no need to use actually random key material to test pbr's grokking signed tags | 18:44 |
| fungi | er, signed tags | 18:44 |
| clarkb | jeblair: I have confirmed that ps3 is better at moving large quantities of bytes than ps4. So I will probably look into that next as ps4 looks fine on the surface | 18:45 |
| AJaeger | thanks, anteaya ! I'm reading... | 18:46 |
| *** e0ne has joined #openstack-infra | 18:46 | |
| anteaya | AJaeger: it is kind of a summary of yesterday's conversation | 18:46 |
| fungi | clarkb: Daviey: i wonder if we can't just stuff some garbage into ~/.gnupg/random_seed | 18:47 |
| fungi | clarkb: Daviey: assuming the error stems from gpg --gen-key trying to populate that | 18:47 |
| dstufft | fungi: you could just commit some generated keys | 18:48 |
| Daviey | fungi: the LP bug is 1369768 | 18:48 |
| dstufft | pre-generated | 18:48 |
| fungi | dstufft: or that | 18:48 |
| *** e0ne is now known as e0ne_ | 18:48 | |
| clarkb | that was suggested and shot down in review iirc | 18:48 |
| Daviey | ISTR we have some pre-generated ssl keys in nova for similar reasons.. so that seems a reasonable fix | 18:49 |
| *** bradjones has quit IRC | 18:49 | |
| Daviey | I do remember reviewing a change and commenting on the key expiry date. | 18:50 |
| AJaeger | anteaya: do we need to explain more in the infra-manuals creator's file? | 18:50 |
| anteaya | AJaeger: I don't think we do | 18:50 |
| fungi | Daviey: clarkb: dstufft: looks like we could just `dd if=/dev/urandom of=~/.gnupg/random_seed bs=600 count=1` | 18:50 |
| anteaya | AJaeger: since this is not a workflow we want to encourage | 18:51 |
| *** bradjones has joined #openstack-infra | 18:51 | |
| anteaya | AJaeger: repo split is a necessary evil right now | 18:51 |
| AJaeger | but it happens... | 18:51 |
| anteaya | AJaeger: to redistribute load | 18:51 |
| anteaya | AJaeger: it does, but we would prefer not to encourage it | 18:51 |
| anteaya | as it is fraught with hazzards | 18:52 |
| clarkb | mordred: ping me when you get a chance to look at it. I am going to get food and come back to this with fresh eyes to see if I can get soemthing that works | 18:52 |
| anteaya | AJaeger: mostly I wanted you to be aware of the latest communication | 18:52 |
| anteaya | AJaeger: so you can take that information into your reviews | 18:52 |
| mordred | clarkb: ola | 18:52 |
| mordred | back online | 18:52 |
| AJaeger | understood. Let me rewrite the docs to see whether we need to explain that make upstream / import clearer. | 18:52 |
| anteaya | AJaeger: as you see fit | 18:53 |
| anteaya | AJaeger: I'm not in a hurry to have this in the docs though, but I will review your patch when you have one | 18:53 |
| AJaeger | anteaya: appreciated the pointer. | 18:53 |
| nelsnelson | fungi, It appears that the hypervisor-libvirt plugin for devstack/lib/nova_plugins currently makes no adjustments to the tempest configs. Do you recommend that I try to get some changes made to the hypervisor-libvirt devstack plugin directly? Or is there a way to augment the hypervisor-libvirt plugin from a customized set of external functions that I can just sort of drop in somewhere, such that devstack-gate would run my custom set of | 18:53 |
| nelsnelson | iniset instructions against the tempest config prior to invocation? | 18:53 |
| anteaya | :D | 18:53 |
| fungi | Daviey: clarkb: dstufft: oh! even better. the gnupg devs have our backs. there's a little-known --debug-quick-random command-line option | 18:54 |
| * AJaeger wanted to say "read again" instead of "rewrite" ;( | 18:54 | |
| clarkb | mordred: cool, did you see the code I linked? its those three lines that set up a circular dependency with packaging.py from builddoc.py | 18:54 |
| anteaya | AJaeger: ah | 18:54 |
| clarkb | mordred: so we need to split out all of that code from packaging.py | 18:54 |
| mordred | clarkb, fungi: btw - as I was in the car I had the same thought - we should put the semver stuff into a feature branch, make feature/0.10 master and then cherry pick the non-semver patches from current master on top | 18:54 |
| anteaya | AJaeger: by all means, currently reading stuff myself | 18:54 |
| mordred | clarkb: just about to look | 18:54 |
| anteaya | gertty to be honest | 18:54 |
| Daviey | fungi: Oh nice catch | 18:55 |
| fungi | Daviey: clarkb: dstufft: seems on gpg 1.x it's just --quick-random | 18:55 |
| AJaeger | anteaya: I'll look at it tomorrow... | 18:55 |
| mordred | clarkb: I still haven't fully read the original dhellmann patch | 18:55 |
| anteaya | AJaeger: sounds good | 18:55 |
| mordred | clarkb: so I might be a few in engaging with this | 18:55 |
| clarkb | mordred: sure no problem | 18:56 |
| mordred | however, I do think we should earnestly consider swapping master out | 18:56 |
| clarkb | mordred: ya I am leaning more and more towards that | 18:56 |
| mordred | clarkb: maybe I'll hack a little on making what that branch wants to look like, as well as the cherry-picks of the things that are on master already | 18:57 |
| *** AJaeger has quit IRC | 18:57 | |
| *** annegent_ has quit IRC | 18:58 | |
| jeblair | clarkb, fungi: i found the problem with jenkins01. and clarkb silently fixed it in his patch. ;) | 18:58 |
| clarkb | jeblair: wait what? does that make me a ninja? | 18:58 |
| fungi | jeblair: clearly he didn't want you to find it | 18:58 |
| fungi | clarkb: stealthy | 18:58 |
| clarkb | jeblair: I do think you should fix it properly and not rely on my patch yet as I still do not grok why ps3 is happy but ps4 is not? | 18:59 |
| clarkb | unless you grok ^ | 18:59 |
| *** e0ne_ has quit IRC | 19:00 | |
| openstackgerrit | James E. Blair proposed openstack-infra/gear: Fix reading headers of large packets https://review.openstack.org/142878 | 19:00 |
| jeblair | clarkb, fungi: ^ | 19:01 |
| *** atiwari has joined #openstack-infra | 19:01 | |
| jeblair | and yeah, i think we should merge that and run with it while we continue to hack on the other patch | 19:01 |
| clarkb | oh I had to do that because raw_bytes wasn't read one at a time | 19:01 |
| *** Masahiro has joined #openstack-infra | 19:02 | |
| clarkb | now I need to understand why the byte by byte process needs that | 19:02 |
| jeblair | clarkb: if we get EAGAIN in the middle of a packet | 19:02 |
| jeblair | clarkb: len(packet) will be something like, oh, say 1448. which is not equal to 12. | 19:02 |
| clarkb | oh because you could call it again | 19:03 |
| clarkb | ya | 19:03 |
| jeblair | an alternate approach would be to save code, ptype, and datalen across calls | 19:03 |
| *** yamamoto has joined #openstack-infra | 19:03 | |
| clarkb | this keeps the diff down at a little cpu time expense | 19:03 |
| clarkb | I am +2 should I approve too? | 19:03 |
| jeblair | actually, we should revisit that | 19:04 |
| fungi | jeblair: yep, so we were getting header batched with other data and then *boom* | 19:04 |
| jeblair | because we'd be unpacking it every time. | 19:04 |
| jeblair | so super inefficient | 19:04 |
| clarkb | jeblair: revisit in a new patch or change? | 19:04 |
| *** doude has quit IRC | 19:04 | |
| *** doude_ has joined #openstack-infra | 19:04 | |
| jeblair | clarkb: let me push a new patch | 19:04 |
| clarkb | jeblair: ok | 19:04 |
| * anteaya heads out for a walk | 19:05 | |
| *** Masahiro has quit IRC | 19:07 | |
| *** e0ne has joined #openstack-infra | 19:07 | |
| clarkb | jeblair: I also think _readRawBytes with bytes_to_read>1 is buggy when using ssl | 19:08 |
| *** yamamoto has quit IRC | 19:08 | |
| clarkb | as subsequent recv() calls could throw the ssl exceptions | 19:08 |
| openstackgerrit | Joe Gordon proposed openstack-infra/elastic-recheck: Add fingerprint fur bug 1403284 https://review.openstack.org/142310 | 19:09 |
| uvirtbot | Launchpad bug 1403284 in devstack "mysql (1040, 'Too many connections')" [Undecided,In progress] https://launchpad.net/bugs/1403284 | 19:09 |
| *** johnthetubaguy is now known as zz_johnthetubagu | 19:09 | |
| openstackgerrit | Joe Gordon proposed openstack-infra/elastic-recheck: Add fingerprint for bug 1403284 https://review.openstack.org/142310 | 19:10 |
| openstackgerrit | James E. Blair proposed openstack-infra/gear: Fix reading headers of large packets https://review.openstack.org/142878 | 19:10 |
| jeblair | clarkb, fungi: ^ | 19:10 |
| clarkb | jeblair: does that fix it? I don't think code is saved between EAGAIN exceptions? | 19:11 |
| fungi | ahh, i see. so this keeps us from doing a header read if we already have started to unpack | 19:11 |
| *** otter768 has joined #openstack-infra | 19:12 | |
| clarkb | oh I see its fixing it in the tight loop | 19:12 |
| clarkb | but we will unpack across EAGAINs wfm | 19:12 |
| fungi | right, this is all within one eagain | 19:12 |
| clarkb | +3 | 19:13 |
| fungi | previously we were just getting lucky because the remaining data could be exactly 12 bytes and then we'd have unpacked it | 19:14 |
| fungi | even if it wasn't a header | 19:14 |
| *** otter768 has quit IRC | 19:16 | |
| *** ivar-lazzaro has joined #openstack-infra | 19:19 | |
| *** ivar-lazzaro has quit IRC | 19:20 | |
| *** berendt has joined #openstack-infra | 19:20 | |
| *** berendt has quit IRC | 19:20 | |
| *** ivar-lazzaro has joined #openstack-infra | 19:21 | |
| *** ayoung-afk is now known as ayoung | 19:21 | |
| openstackgerrit | Jeremy Stanley proposed openstack-dev/pbr: Use unsafe OpenPGP keys for testing https://review.openstack.org/142884 | 19:22 |
| fungi | Daviey: ^ | 19:22 |
| fungi | it won't pass unit tests at the moment until we get master fixed | 19:22 |
| *** shashankhegde has quit IRC | 19:23 | |
| fungi | but i added . to the testenv deps list in tox.ini and it passes py27 that way | 19:23 |
| jeblair | clarkb: yeah, i chose this approach rather than making sure to save 3 vars on each iteration (and hoping i got the resets correct) | 19:24 |
| clarkb | jeblair: ya I think this works well enough for now | 19:24 |
| clarkb | jeblair: I think I understand why my change is broken in ps4 now. We are better able to read all the data off the wire in a readPacket call then we never get poll triggered to read the next packet out of the buffer | 19:25 |
| clarkb | jeblair: and since readPacket is only called when poll says there is data to read my code is broken | 19:25 |
| clarkb | not quite sure how I want to fix that yet | 19:26 |
| openstackgerrit | Matt Riedemann proposed openstack-infra/elastic-recheck: Add query for tempest bug 1403998 https://review.openstack.org/142885 | 19:26 |
| uvirtbot | Launchpad bug 1403998 in tempest "test_dhcp_ipv6 tests failing in the gate" [Undecided,New] https://launchpad.net/bugs/1403998 | 19:26 |
| *** hdd has quit IRC | 19:27 | |
| clarkb | I think maybe make read packet only read one packet at a time and not buffer anythibg. so get 12 bytes, read length, get remainder | 19:28 |
| *** annegent_ has joined #openstack-infra | 19:28 | |
| *** pradk has quit IRC | 19:30 | |
| *** shashankhegde has joined #openstack-infra | 19:30 | |
| jeblair | clarkb: that doesn't work as well for admin packets | 19:32 |
| *** ajo_ has quit IRC | 19:32 | |
| jeblair | clarkb: on the server side, readpacket is in a while loop | 19:32 |
| jeblair | clarkb: so the intent is that it should read to exhaustion every time | 19:32 |
| clarkb | jeblair: its in a while loop but it only reads from fd's with POLLIN set | 19:33 |
| jeblair | clarkb: once it is triggered, it should read until exhaustion | 19:33 |
| clarkb | jeblair: so we can read the first packet and the second into readPacket, return only the first packet, then never get triggeredagain | 19:33 |
| clarkb | jeblair: oh ok, maybe the fix then is to just call readPacket until it errors | 19:34 |
| jeblair | clarkb: see readFromConnection -- it will call it again | 19:34 |
| jeblair | clarkb: that's exactly what it does | 19:34 |
| *** HeOS has joined #openstack-infra | 19:34 | |
| clarkb | oh I see. so maybe this is a worker/client side issue | 19:34 |
| jeblair | clarkb: yeah, this may need rethinking for the client side | 19:34 |
| Daviey | fungi: I added a comment to your change | 19:35 |
| jeblair | clarkb: where i would expect that it would not get called again because there's no pending data (since it uses level-triggered select) | 19:35 |
| *** annegent_ has quit IRC | 19:35 | |
| *** dprince has quit IRC | 19:35 | |
| clarkb | jeblair: yup | 19:35 |
| openstackgerrit | Merged openstack-infra/elastic-recheck: Add query for neutron bug 1403291 https://review.openstack.org/142324 | 19:35 |
| uvirtbot | Launchpad bug 1403291 in neutron "test_server_connectivity_pause_unpause fails with "AssertionError: False is not true : Timed out waiting for 172.24.4.64 to become reachable"" [Undecided,New] https://launchpad.net/bugs/1403291 | 19:35 |
| openstackgerrit | Merged openstack-infra/gear: Fix reading headers of large packets https://review.openstack.org/142878 | 19:36 |
| lifeless | fungi: I thought we'd gotten haveged everywhere already ? | 19:36 |
| *** bradjones has quit IRC | 19:36 | |
| lifeless | did pbr get a release done? | 19:37 |
| lifeless | '07:56 <oopsiesdais_> Hi, I am running into some issues using `pip install fixtures` as of today. | 19:37 |
| lifeless | It appears pbr was updated in the past day, and I can install pbr just | 19:37 |
| lifeless | fine, but when fixtures tries to install it, it fails. Have you seen this? | 19:37 |
| lifeless | ' | 19:37 |
| mordred | lifeless: yes, we've made a release - 0.10.4 | 19:37 |
| mordred | lifeless: which contains fixes for all currently known problems | 19:37 |
| mriedem | getting this merged would unblock sqlalchemy-migrate patches https://review.openstack.org/#/c/142850/ | 19:37 |
| clarkb | but 0.10.3 was more likely to be the cause of ^ if that timestamp is utc | 19:37 |
| lifeless | clarkb: that would have been about 25 hours ago | 19:38 |
| clarkb | lifeless: ya 0.10.3 then which had a bug requiring git in some cases | 19:38 |
| clarkb | fixed in 0.10.4 | 19:38 |
| lifeless | kk | 19:38 |
| Ryan_Lane | http://docs.openstack.org/openstack-ops/content/nova-network-deprecation.html <-- is this still accurate anymore? | 19:38 |
| fungi | lifeless: we do have haveged installed everywhere, but that doesn't mean everyone who might run pbr's unit tests does | 19:39 |
| Ryan_Lane | (hi btw!) | 19:39 |
| clarkb | Ryan_Lane: iirc neutron is supposed to be testing nova net to neutron migrations this cycle | 19:40 |
| lifeless | fungi: then they can install it :) | 19:40 |
| *** bradjones has joined #openstack-infra | 19:40 | |
| Ryan_Lane | clarkb: does neutron now support flatdhcp? | 19:40 |
| clarkb | that and DVR allows you to have man routers like nova net but I think it still needs love | 19:40 |
| lifeless | fungi: I'm not arguing against the change btw :) | 19:40 |
| jogo | fungi: if you have a second, wondering if I could get your thoughts on https://review.openstack.org/#/c/141923/ (add timeout to er_safe_run) | 19:40 |
| Daviey | lifeless: Fine. But an error should say so.. :) | 19:40 |
| clarkb | Ryan_Lane: for that I am not sure | 19:40 |
| fungi | lifeless: it seems wasteful to suck away the kernel entropy pool needlessly | 19:40 |
| fungi | lifeless: but yes, they could | 19:40 |
| clarkb | Ryan_Lane: but feature parity has been a big thing lately (juno and kilo(master)) | 19:40 |
| lifeless | fungi: agreed. Note my *not arguing against it*. | 19:40 |
| * fungi nods | 19:41 | |
| jaypipes-afk | fungi: nope, no proxy | 19:41 |
| *** jaypipes-afk is now known as jaypipes | 19:41 | |
| lifeless | fungi: further, note that I tried to do it the first time around. | 19:41 |
| Ryan_Lane | so, it's still listed as deprecated, even though it's getting new features? :) | 19:41 |
| lifeless | fungi: see the config where the option should be | 19:41 |
| clarkb | Ryan_Lane: nova net should not get new features | 19:41 |
| clarkb | iirc its bug fixes only | 19:41 |
| lifeless | Ryan_Lane: what new features? | 19:41 |
| *** emagana has quit IRC | 19:41 | |
| *** jlk has joined #openstack-infra | 19:41 | |
| Ryan_Lane | so the new features were only for juno? | 19:41 |
| clarkb | Ryan_Lane: neutron is getting new features to match nova net | 19:42 |
| lifeless | Ryan_Lane: *what* new features? | 19:42 |
| clarkb | nova network is not getting new features | 19:42 |
| Ryan_Lane | I missed the last summit, so my info is likely out of date :( | 19:42 |
| fungi | jaypipes: zaro suggests it could be a session timeout from inactivity... does it maybe happen after suspend/resume or something? | 19:42 |
| jlk | Hey folks! Could somebody tell me what the status is on https://review.openstack.org/#/c/140893/ ? It looks like it's had all the appropriate checks, but has been sitting like that for near 24 hours. | 19:42 |
| *** teran has joined #openstack-infra | 19:42 | |
| Ryan_Lane | http://specs.openstack.org/openstack/nova-specs/specs/juno/implemented/better-support-for-multiple-networks.html | 19:42 |
| *** tonytan4ever has quit IRC | 19:43 | |
| clarkb | :/ apparently they are adding new features | 19:43 |
| Ryan_Lane | right. that's why I was asking about deprecation :D | 19:43 |
| jogo | clarkb: we unfroze nova networking because it was taking too long | 19:43 |
| Ryan_Lane | I remember in atlanta hearing that nova-network wasn't going to be deprecated anymore | 19:43 |
| clarkb | jogo: yes but for bug fixes only not features | 19:43 |
| *** atiwari has quit IRC | 19:44 | |
| clarkb | Ryan_Lane: ^ | 19:44 |
| clarkb | anyways my understanding was that neutron and nova are going to get migration testing working this cycle. I have a job up to do it but it doesn't work yet | 19:45 |
| jogo | clarkb: not exactly, AFAIK we tried to limit to features neutron already has | 19:45 |
| *** yfried has joined #openstack-infra | 19:45 | |
| jaypipes | fungi: no, no suspend or resume. I'll keep rooting around and see if I can get a pattern to it. | 19:45 |
| *** emagana has joined #openstack-infra | 19:45 | |
| lifeless | clarkb: https://review.openstack.org/#/c/93093/ FWIW | 19:45 |
| clarkb | and that kilo would be the release that allows you to migrate across. Along side this neutron is also adding feature parity. Things like DVR | 19:45 |
| jogo | clarkb: nova would like that very much, but AFAIK the ball is in neutron's court | 19:45 |
| Ryan_Lane | what's DVR? | 19:45 |
| jogo | mestery: ^ | 19:45 |
| clarkb | jogo: its in both of yours | 19:45 |
| clarkb | jogo: my understanding come out of that session is that both sides would have to work on this | 19:45 |
| Ryan_Lane | https://wiki.openstack.org/wiki/Neutron/DVR found it | 19:46 |
| clarkb | Ryan_Lane: distributed virtual router. Allows you to have multiple routers like nova multi host | 19:46 |
| mestery | DVR == Distributed Virtual Router | 19:46 |
| jogo | mestery: do you think Kilo will be the release where we have a nova-network -> neutron migration path working and tested? | 19:46 |
| mestery | jogo: Yes, that's the plan. | 19:47 |
| jogo | jlk: I ran a recheck on your patch hopefully that should get it moving again | 19:47 |
| mestery | In fact, I'm hopeful we mark it deprected in Kilo | 19:47 |
| mestery | nova-network | 19:47 |
| clarkb | and the test exists it jus needs work in grenade | 19:47 |
| jlk | jogo: alright, thanks. | 19:47 |
| jogo | mestery: excellent, and anything needed from the nova side | 19:47 |
| clarkb | and possibly in d-g to set up the scenario | 19:47 |
| mestery | jogo: I think the spec is filed over there from obondarev. | 19:47 |
| jogo | mestery: link? | 19:47 |
| mestery | jogo: Let me look | 19:47 |
| clarkb | iirc we actually want nova net and neutron running at the same time but current d-g says turn off nova net | 19:48 |
| jogo | mestery: thanks, as clarkb pointed out we should make sure this doesn't fall through the cracks | 19:48 |
| clarkb | jeblair: your latest gear change merged. Is that going to get applied soon? | 19:48 |
| mestery | jogo: Agreed, and I can't find the link, let me work with obondarev who is doing the work. | 19:48 |
| mestery | jogo: It's critical we make this migration work during Kilo so we can finally deprecate nova-network | 19:49 |
| jeblair | clarkb: yeah, i'll restart it again soon | 19:49 |
| jogo | mestery: thanks, tomorrow is the spec deadline but we have exceptions until Kilo-2 and this would get one if needed | 19:49 |
| jogo | but would prefer to not need | 19:50 |
| mestery | jogo: I've emailed obondarev on this, I'll do it again | 19:50 |
| mestery | IT's like herding cats at times ;) | 19:50 |
| jogo | mestery: yup, at least its not one of mordred's wet cats | 19:50 |
| mestery | rofl | 19:50 |
| *** ZZelle has joined #openstack-infra | 19:51 | |
| dhellmann | mordred: the only important differences in my patch were catching the right type of exception, and putting the classes in a place where if we decide we need them they can always be imported and the names registered in the command are never pointing to a value that can be None | 19:53 |
| dhellmann | mordred: I think your patch had an error with the function setting a name but not specifying that the variable was a global, but I'd gone pretty far with my patch before we figured that out | 19:53 |
| *** baoli has quit IRC | 19:59 | |
| mordred | kk | 20:02 |
| *** _nadya_ has joined #openstack-infra | 20:04 | |
| *** erikmwilson has quit IRC | 20:05 | |
| *** aysyd has quit IRC | 20:05 | |
| *** bswartz has quit IRC | 20:09 | |
| *** Longgeek has quit IRC | 20:10 | |
| *** shashankhegde has quit IRC | 20:11 | |
| *** zz_sabari is now known as zz_zz_sabari | 20:12 | |
| fungi | Daviey: lifeless: replied to your comments on 142884 along with some benchmarking | 20:17 |
| *** bradjones has quit IRC | 20:18 | |
| *** tonytan4ever has joined #openstack-infra | 20:18 | |
| *** garyk has joined #openstack-infra | 20:19 | |
| *** bdpayne has quit IRC | 20:19 | |
| garyk | anteaya: thanks for the review. any idea how long it takes till we see the group created on gerrit? | 20:20 |
| lifeless | fungi: so, just need to move the option and I'll +2 it | 20:21 |
| anteaya | garyk: it is created: https://review.openstack.org/#/admin/groups/545,members | 20:21 |
| fungi | lifeless: maybe i'm misunderstanding. it's implemented only as a command-line option from what i can find (for safety, so nobody sets it in their prefs) | 20:22 |
| anteaya | garyk: are you the first core and release group memeber? | 20:22 |
| garyk | anteaya: yes :) | 20:22 |
| *** andreykurilin has joined #openstack-infra | 20:22 | |
| lifeless | fungi: AIUI all the options can be either CLI or config, but if that ones different, ca la vie | 20:22 |
| anteaya | fungi: if you would be so kind to add garyk to vmware-nsx-core and -release | 20:22 |
| anteaya | garyk: then you can add others to -core as you see fit | 20:23 |
| lifeless | fungi: did you give it a try? | 20:23 |
| fungi | lifeless: i'm testing now to find out | 20:23 |
| garyk | anteaya: fungi:thanks! | 20:23 |
| *** Vek has quit IRC | 20:24 | |
| garyk | that would be great - we can start to kick the tires once this is done | 20:24 |
| anteaya | garyk: once this is done, well from our point of view, it is done | 20:24 |
| garyk | brilliant! | 20:24 |
| fungi | anteaya: garyk: done | 20:24 |
| anteaya | fungi: thank you | 20:24 |
| garyk | fungi: and anteaya: thanks! | 20:25 |
| anteaya | godspeed | 20:25 |
| fungi | lifeless: yeah, tried %quick-random and quick-random: 1 and a few other likely patterns, none seem to work in the preferences file | 20:26 |
| lifeless | fungi: well nuts. Ok. Can have a comment to that effect? | 20:27 |
| *** teran_ has joined #openstack-infra | 20:27 | |
| lifeless | fungi: (in the source file) | 20:27 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/project-config: Add nodepool script to preseed testrepository from subunit2sql https://review.openstack.org/136234 | 20:27 |
| fungi | lifeless: sure, will do. want me to also clean up the unimplemented entries in the preferences list while i'm at it? | 20:28 |
| lifeless | fungi: they are implemented, in 2.1 | 20:28 |
| fungi | lifeless: do we want to switch to testing with 2.1? | 20:28 |
| lifeless | fungi: http://superuser.com/questions/813793/unable-to-use-transient-key-when-generating-a-batch-gpg-key | 20:28 |
| fungi | lifeless: if so, this option is also named something different in gpg2 | 20:28 |
| lifeless | fungi: that wouldn't fix this bug; this bug is 'someone other than us running the tests may be surprised' | 20:29 |
| lifeless | fungi: another way to fix it would be a try:finally around the call that attaches a message saying 'check your entropy' | 20:29 |
| openstackgerrit | Elizabeth K. Joseph proposed openstack-infra/puppet-zanata: Add Rakefile, TODO & Wildfly configuration support https://review.openstack.org/142645 | 20:30 |
| *** teran has quit IRC | 20:30 | |
| lifeless | fungi: requiring 2.1 would also fix it I guess, in that <2.1 would error, and 2.1 wouldn't timeout | 20:30 |
| Daviey | lifeless / fungi: I started working on a branch that caught the exception, but disgarded it as fungi's fix seems smarter IMO | 20:31 |
| *** annegent_ has joined #openstack-infra | 20:31 | |
| openstackgerrit | Elizabeth K. Joseph proposed openstack-infra/puppet-zanata: Add Rakefile, TODO & Wildfly configuration support https://review.openstack.org/142645 | 20:32 |
| lifeless | OTOH requiring 2.1 would break testing with older distros | 20:32 |
| lifeless | since 2.0 was all that was present in e.g. utopic | 20:32 |
| * lifeless checks rmadison | 20:32 | |
| lifeless | indeed | 20:33 |
| lifeless | even vivid is still 2.0 | 20:33 |
| lifeless | fungi: whats the option called in 2.x? | 20:33 |
| openstackgerrit | Jeremy Stanley proposed openstack-dev/pbr: Use unsafe OpenPGP keys for testing https://review.openstack.org/142884 | 20:33 |
| fungi | lifeless: mentioned in the code comment now ^ | 20:34 |
| *** tkelsey has joined #openstack-infra | 20:34 | |
| *** erikmwilson has joined #openstack-infra | 20:35 | |
| *** erikmwilson has quit IRC | 20:35 | |
| anteaya | fungi mordred both of you proposed the exact change at the exact same minute to change the regex to accept rc instead of c | 20:35 |
| clarkb | mordred: ok I am about to look at pbr again. Have you had a chance to look yet? | 20:35 |
| anteaya | who wins? | 20:35 |
| anteaya | one gets merged and the other abandoned | 20:35 |
| fungi | lifeless: the --debug-quick-random documentation, as much as it is, does explicitly say it's cli-only https://www.gnupg.org/documentation/manuals/gnupg/Agent-Options.html | 20:35 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-infra/project-config: Support pre/post test hooks for nova-docker jobs https://review.openstack.org/142899 | 20:35 |
| anteaya | I'm not going to pick | 20:36 |
| jeblair | anteaya: the one with the lower number :) | 20:36 |
| anteaya | jeblair: very good | 20:36 |
| *** annegent_ has quit IRC | 20:36 | |
| fungi | anteaya: oh, whoops. i didn't notice | 20:37 |
| mordred | clarkb: I have not | 20:37 |
| *** shashankhegde has joined #openstack-infra | 20:37 | |
| *** erikmwilson has joined #openstack-infra | 20:37 | |
| anteaya | mordred one this one | 20:37 |
| anteaya | fungi: not a problem I just didn't want them to both sit | 20:37 |
| mordred | yay I win! | 20:38 |
| mordred | score one more commit point for monty | 20:38 |
| * fungi hands mordred a kewpie doll | 20:38 | |
| *** tkelsey has quit IRC | 20:39 | |
| jeblair | fastest git-review in the ...west? ...east? where are you anyway? ;) | 20:39 |
| jeblair | clarkb, fungi: restarting geard with fix now | 20:39 |
| clarkb | jeblair: ok | 20:39 |
| anteaya | somewhere spanish | 20:39 |
| anteaya | he greeted us with ola this morning/start of day | 20:39 |
| fungi | probably best not to waste time figuring out whose airspace he's in at any given moment | 20:40 |
| mordred | jeblair: that's a great question - and the answer has actually changed today at least once | 20:40 |
| mordred | today mordred is like the sun - rising in the east and setting in the west | 20:41 |
| jeblair | anteaya: if 'ola' is spanish he said 'wave' to us. | 20:41 |
| anteaya | did he | 20:42 |
| mordred | btw - from the twitters... | 20:42 |
| mordred | "There's no Dana. Only Zuul. #WorstDateIn5Words" | 20:42 |
| anteaya | is ola considered hello anywhere | 20:43 |
| anteaya | mordred: nice | 20:43 |
| jlk | mordred: lol | 20:43 |
| mordred | anteaya: it's considered hello when I say it :) | 20:43 |
| anteaya | mordred: I heard hello | 20:43 |
| anteaya | but wave is fine too | 20:43 |
| mordred | https://twitter.com/staypuft/status/545648920303390720 | 20:43 |
| mordred | just in case anyone cares to retweet | 20:43 |
| anteaya | see a lose nothing from not being on the twitters | 20:44 |
| fungi | anteaya: "hola" is spanish for hello | 20:44 |
| anteaya | better filter system | 20:44 |
| *** zz_zz_sabari is now known as zz_zz_zz_sabari | 20:44 | |
| anteaya | fungi: ah the h | 20:44 |
| anteaya | french canadians love the h | 20:44 |
| fungi | better yet, ¡Hola! | 20:44 |
| *** weshay has quit IRC | 20:44 | |
| anteaya | ah the upside down exclaimation then the h | 20:45 |
| *** harlowja has quit IRC | 20:45 | |
| *** harlowja_ has joined #openstack-infra | 20:45 | |
| mordred | fungi: that can be confused with misspelling "holla" as in slang for "holler" | 20:45 |
| jeblair | mordred: thus the exclamation! | 20:45 |
| fungi | everyone loves a good exclamation! | 20:46 |
| jeblair | or more than !!!one!111! | 20:46 |
| * jeblair used !!!one!111! in a sentence | 20:46 | |
| mordred | I should just start using the other one incorrectly¡ | 20:47 |
| fungi | mordred: why¿ | 20:47 |
| *** hashar has joined #openstack-infra | 20:47 | |
| mordred | fungi: ?why not¿ | 20:47 |
| fungi | ok° | 20:48 |
| mordred | it seems like the conversation has moved us into friday - maybe we're all in new zealand with lifeless and just don't know it | 20:48 |
| jeblair | mordred: so i want to replace a system call in a unit test. how should i go about doing that? | 20:48 |
| jedimike | feels like a friday | 20:48 |
| anteaya | I can take a friday | 20:49 |
| *** hashar has quit IRC | 20:49 | |
| mordred | jeblair: well ... | 20:49 |
| openstackgerrit | Merged openstack-infra/project-config: The ci-sandbox project should not require an ICLA https://review.openstack.org/142804 | 20:49 |
| mordred | jeblair: ksplice? | 20:49 |
| jeblair | mordred: sorry, standard library function | 20:49 |
| jeblair | mordred: in python | 20:49 |
| mordred | ah | 20:50 |
| mtreinish | jeblair: just mock it | 20:50 |
| mordred | well that you can just use monkeypatch in fixtures | 20:50 |
| jeblair | what's the mock lib du jour? | 20:50 |
| mtreinish | mock :) | 20:50 |
| *** Masahiro has joined #openstack-infra | 20:50 | |
| jeblair | okay, and when is appropriate to fixtures-monkeypatch vs mock? | 20:50 |
| mordred | if you're not already using a straight-up mocking library and there is a specific call you want to do, I'd just use monkeypatch | 20:51 |
| *** sarob has joined #openstack-infra | 20:51 | |
| mordred | if you want to do more than a passing amount of mocking, I'd use mock | 20:51 |
| jeblair | mtreinish, mordred: roger, thanks | 20:51 |
| *** annegent_ has joined #openstack-infra | 20:52 | |
| *** atiwari has joined #openstack-infra | 20:53 | |
| jedimike | i always forget which one i prefer, mock or mocker. Then I use one of them, and it's always the other one. | 20:53 |
| openstackgerrit | Merged openstack-infra/project-config: Go back to rc for pre-releases https://review.openstack.org/142791 | 20:54 |
| fungi | ttx: ^ | 20:55 |
| fungi | we're officially back to "rc" as of now | 20:55 |
| *** Masahiro has quit IRC | 20:55 | |
| mordred | woot! | 20:55 |
| *** hamzy_vacation has quit IRC | 20:55 | |
| fungi | (well, in about 5-10 minutes when that hits zuul's configuration anyway) | 20:55 |
| *** hamzy_vacation has joined #openstack-infra | 20:56 | |
| openstackgerrit | Merged openstack-infra/project-config: Make gate-sqlalchemy-migrate-python33 non-voting https://review.openstack.org/142850 | 20:56 |
| *** zz_johnthetubagu is now known as johnthetubaguy | 20:58 | |
| *** weshay has joined #openstack-infra | 20:59 | |
| *** ociuhandu has quit IRC | 21:00 | |
| fungi | Daviey: replied to your comment on 142884 but afaik no distros ship gnupg 2.x as "gpg" (they all use a separate "gpg2" mainly because 1.x is better suited to embedded/low-resource environments, headless/remote/automation and places where you don't want lots of dependent libs) | 21:00 |
| Daviey | fungi: thanks.. just read it.. +1 | 21:01 |
| mtreinish | fungi: arch symlinks /usr/bin/gpg to gpg2 | 21:01 |
| fungi | mtreinish: arch is insane | 21:01 |
| fungi | mtreinish: arch also ships a py3k /usr/bin/python | 21:01 |
| mtreinish | fungi: http://paste.openstack.org/show/152886/ | 21:01 |
| *** dprince has joined #openstack-infra | 21:02 | |
| *** garyk has quit IRC | 21:02 | |
| fungi | mtreinish: wow. even the gnupg authors recommend against that | 21:02 |
| mtreinish | they do and that's a constant frustration for me | 21:02 |
| mordred | clarkb: ^^ next time you suggest we use arch for something | 21:02 |
| viglesias | mordred: sounds like an over_arch_ing problem | 21:03 |
| * viglesias backs away slowly | 21:03 | |
| fungi | heh | 21:03 |
| *** _nadya_ has quit IRC | 21:03 | |
| *** kumartin has joined #openstack-infra | 21:03 | |
| *** baoli has joined #openstack-infra | 21:07 | |
| *** baoli has quit IRC | 21:07 | |
| *** baoli has joined #openstack-infra | 21:07 | |
| *** cdent has quit IRC | 21:08 | |
| *** pc_m has quit IRC | 21:08 | |
| fungi | mtreinish: even gnupg's official documentation refers to a "gpg" command for 1.x and "gpg2" for 2.x https://www.gnupg.org/documentation/manuals/gnupg/Invoking-GPG.html | 21:09 |
| jlk | but arch knows better! | 21:09 |
| *** ivar-laz_ has joined #openstack-infra | 21:10 | |
| dims__ | sdague: when you get a chance, can you please peek at https://review.openstack.org/#/c/142899/ ? anteaya requested a nova core to look see at pre/post hooks i am trying to add for nova-docker | 21:10 |
| anteaya | dims__: mostly I just want a nova core to say they are aware | 21:10 |
| *** baoli has quit IRC | 21:10 | |
| clarkb | mordred: so I startedfrom scratch to avoid any issues that might bite me and I still can't get it to work on master. Moving the iter_log stuff into pbg/git makes them explode. currently trying to debug that | 21:10 |
| anteaya | dims__: is sdague nova core? | 21:10 |
| mtreinish | fungi: heh, I'll file a bug with arch then. Although I expect it to fall on deaf ears | 21:10 |
| *** baoli has joined #openstack-infra | 21:11 | |
| *** signed8bit has quit IRC | 21:11 | |
| fungi | mtreinish: yeah, definitely seems like a distro bug (upstream expects users to be able to install and use both versions side-by-side after all) | 21:11 |
| anteaya | so he is | 21:11 |
| sdague | dims__: so... what do these things do? | 21:11 |
| sdague | and more importantly do the devstack plugins help with this? | 21:12 |
| *** zz_zz_zz_sabari is now known as zz_zz_zz_zz_saba | 21:12 | |
| *** ivar-lazzaro has quit IRC | 21:12 | |
| *** jedimike has quit IRC | 21:12 | |
| dims__ | sdague: in the post hook, i'd like to collect docker log at the very end of the job after all the tests are run | 21:12 |
| dims__ | sdague: https://review.openstack.org/#/c/142879/ is the nova-docker review where i add the shell scripts | 21:13 |
| *** otter768 has joined #openstack-infra | 21:13 | |
| dims__ | sdague: the "clean" phase of the devstack plugin gets run before the tempest jobs are run i believe | 21:13 |
| mtreinish | dims__: just copy the files to the log dir and they'll get copied automatically (that's what neutron did for their functional job) | 21:13 |
| *** baoli has quit IRC | 21:14 | |
| *** zz_dimtruck is now known as dimtruck | 21:14 | |
| mtreinish | dims__: oh, nm me I didn't look at your patches | 21:14 |
| dims__ | mtreinish: ./neutron/tests/functional/contrib/post_test_hook.sh | 21:15 |
| mtreinish | dims__: yep, which is exactly what you're adding. | 21:15 |
| dims__ | :) | 21:15 |
| *** baoli_ has joined #openstack-infra | 21:16 | |
| *** stevemar has quit IRC | 21:16 | |
| openstackgerrit | Doug Hellmann proposed openstack-infra/project-config: Generate specs.o.o root page from template https://review.openstack.org/141173 | 21:16 |
| sdague | dims__: the pre test is empty? | 21:16 |
| clarkb | somehow the iter_changelog yield type is changing? this is so weird | 21:17 |
| *** baoli_ has quit IRC | 21:17 | |
| dims__ | sdague: y for now, i may need to enable a few more tweaks in docker there just before the run | 21:17 |
| *** otter768 has quit IRC | 21:17 | |
| sdague | dims__: ok | 21:17 |
| dims__ | sdague: thanks | 21:18 |
| *** baoli_ has joined #openstack-infra | 21:18 | |
| dims__ | mtreinish: thanks as well | 21:18 |
| sdague | dims__: so this raises a good question of figuring out how to best have https://review.openstack.org/#/c/137054/ exposed via d-g | 21:19 |
| dims__ | sdague: looking | 21:21 |
| dims__ | sdague: y, that looks like devstack-only | 21:22 |
| sdague | yes it is | 21:23 |
| *** atiwari has quit IRC | 21:23 | |
| sdague | but part of the intent for creating the interface is making it easy to export up | 21:23 |
| *** signed8bit has joined #openstack-infra | 21:24 | |
| openstackgerrit | Clark Boylan proposed openstack-dev/pbr: Only import sphinx during hook processing https://review.openstack.org/142840 | 21:24 |
| openstackgerrit | Clark Boylan proposed openstack-dev/pbr: Properly check for git before getting git dir https://review.openstack.org/142841 | 21:24 |
| dims__ | sdague: right, there's a few projects that use the pre/post hooks via external shell script - http://paste.openstack.org/show/152905/ - there's a bunch that have the hook in their yaml files in project-config | 21:25 |
| *** sarob has quit IRC | 21:25 | |
| clarkb | dhellmann: mordred comments on https://review.openstack.org/#/c/142840/2 show the weirdness | 21:26 |
| clarkb | that should pass at least pep8 and py27 though | 21:26 |
| sdague | dims__: yeh | 21:26 |
| openstackgerrit | Adrian Turjak proposed openstack-infra/project-config: Add new Cloud-Pydashie project https://review.openstack.org/142240 | 21:27 |
| *** sweston_ has joined #openstack-infra | 21:27 | |
| openstackgerrit | Cameron Rowshanbin proposed openstack-infra/project-config: ec2-driver: Added gate checks https://review.openstack.org/142908 | 21:28 |
| notmyname | see the git sec issue? | 21:28 |
| *** sweston has quit IRC | 21:28 | |
| *** sweston_ is now known as sweston | 21:28 | |
| notmyname | https://github.com/blog/1938-vulnerability-announced-update-your-git-clients | 21:29 |
| *** ayoung has quit IRC | 21:29 | |
| dhellmann | clarkb: more context? | 21:29 |
| clarkb | dhellmann: porting your sphinx docutils change onto pbr master. Master docutils classes depend on git things in packaging.py so I had to rip that stuff out into pbr/git.py | 21:30 |
| dims__ | mtreinish: sdague: can one of you please leave a note that this is kosher pattern to follow for now? | 21:30 |
| clarkb | dhellmann: to avoid circular imports as you did with the options.get_boolean_option thing | 21:30 |
| *** mjturek has quit IRC | 21:30 | |
| dhellmann | clarkb: ok, that seems reasonable. I wish we'd taken a serious look at that refactoring/cleanup patch we had submitted. :-/ | 21:30 |
| jeblair | clarkb, fungi: i think we may need to look into whether we need a gerrit upgrade due to the issue notmyname raised ^ | 21:30 |
| clarkb | dhellmann: but doing that seems to have changed the behavior of _iter_changelog() I was getting "TypeError: must be unicode, not str" until I changed the string type yielded by _iter_changelog | 21:31 |
| clarkb | dhellmann: and I have no idea why moving those functions would do that | 21:31 |
| notmyname | jeblair: tl;dr affects clients on case-insensitive filesystems | 21:31 |
| *** sputnik13 has joined #openstack-infra | 21:31 | |
| *** ajo has joined #openstack-infra | 21:31 | |
| *** ihrachyshka has quit IRC | 21:32 | |
| clarkb | jeblair: hrm, gerrit is not a git client so I wouldn't expect it to need that, but jgit may do silly things so who knows | 21:32 |
| dhellmann | clarkb: does running tox reproduce the error? I can spend some time looking at it | 21:32 |
| clarkb | dhellmann: ya, if you remove the u from the strings in pbr/git.py:_iter_changelog() you should be able to reproduce | 21:32 |
| clarkb | dhellmann: eg just make those normal strings | 21:32 |
| jeblair | notmyname: yeah, so probably not critical for our systems; though perhaps the fixes add protection for vulnerable systems? clarkb: yeah, at any rate, jgit made a release due to it so ... | 21:32 |
| fungi | jeblair: clarkb: notmyname: i think the case-insensitive filesystem part also means the places where we're using git clients are probably not affected either | 21:33 |
| dhellmann | clarkb: from __future__ import unicode_literals | 21:33 |
| fungi | using git clients in our servers/automation i mean | 21:33 |
| dhellmann | packaging.py has that at the top | 21:33 |
| clarkb | dhellmann: oh does that implicitly make strings unicode? | 21:33 |
| notmyname | jeblair: fungi: right. it seems to mostly affect os x and windows. so openstack certainly has a lot of devs who need to update. still probably good to take care | 21:33 |
| dhellmann | clarkb: yes, it makes literal strings unicode by default instead of having to prefix with u | 21:34 |
| clarkb | dhellmann: I see. ok I can clean that up in a sec | 21:34 |
| dhellmann | clarkb: good for py2/3 compat | 21:34 |
| dhellmann | k | 21:34 |
| *** cameron__ has joined #openstack-infra | 21:34 | |
| mordred | notmyname: ++ | 21:34 |
| *** markmcclain has quit IRC | 21:34 | |
| clarkb | dhellmann: mordred fungi I think the other thing to review there is whether or not the code I moved makes sense in its new home. Maybe more code should be moved or less? | 21:35 |
| dhellmann | clarkb: I'm ok with moving things a little at a time, but I'll take a look | 21:36 |
| openstackgerrit | Clark Boylan proposed openstack-dev/pbr: Only import sphinx during hook processing https://review.openstack.org/142840 | 21:36 |
| openstackgerrit | Clark Boylan proposed openstack-dev/pbr: Properly check for git before getting git dir https://review.openstack.org/142841 | 21:36 |
| *** ayoung has joined #openstack-infra | 21:37 | |
| cameron__ | Hi, we've submitted a change to the project-config repo to add a job for our ec2-driver project here: https://review.openstack.org/#/c/142908/ | 21:37 |
| cameron__ | It would be great if it could get verified/reviewed | 21:38 |
| *** mrmartin has quit IRC | 21:40 | |
| fungi | dhellmann: for our security.openstack.org website where we're starting to publish ossa text rendered from rst via sphinx, we want to add an additional build step (which generates the rst from yaml). is it easy to have build_sphinx call arbitrary commands? | 21:41 |
| clarkb | fungi: aiui no, that is why the code in pbrthat does that stuff exists | 21:41 |
| anteaya | cameron__: the best way to get reviews it to give them, http://anteaya.info/blog/2013/03/21/reviewing-an-openstack-patch/ is a friendly introduction if you are just getting started | 21:41 |
| fungi | clarkb: aha, so we'd need a custom doc job to do that then, probably? | 21:42 |
| mordred | fungi: well, you can write a sphinx plugin | 21:42 |
| *** tkelsey has joined #openstack-infra | 21:42 | |
| mordred | fungi: and reference that sphinx plugin frmo your conf.py file | 21:42 |
| cameron__ | anteaya: Thanks! | 21:42 |
| clarkb | jeblair: so thinking a bit more abut this I think we should be ok, but updating to fixed jgit may be prudent just in case they snuck any other fixes in | 21:42 |
| anteaya | cameron__: you're welcome | 21:42 |
| mordred | fungi: that way you can just use build_sphinx and it'll do what you want | 21:42 |
| fungi | mordred: i'll double-check how complicated that is. hopefully not hard | 21:43 |
| mordred | fungi: looking at oslosphinx has an example of writing a sphinx plugin to do stuff | 21:43 |
| mordred | I dont' think it's too bad | 21:43 |
| fungi | mordred: oh! all the better. thanks | 21:43 |
| fungi | clarkb: does rebuilding our gerrit warfile update jgit automagically? because i don't see an updated gerrit release yet | 21:44 |
| clarkb | fungi: no, we have to bump the version in our pom.xml | 21:44 |
| *** amitgandhinz has quit IRC | 21:44 | |
| *** pblaho has quit IRC | 21:46 | |
| openstackgerrit | James E. Blair proposed openstack-infra/gear: Add readPacket unit test https://review.openstack.org/142910 | 21:46 |
| clarkb | dhellmann: fungi mordred ok the latest patchsets in that series are looking good according to jenkins. I think we should review them now | 21:46 |
| *** vigneshvar has quit IRC | 21:46 | |
| jeblair | clarkb: a test to help us out ^ (it catches the error we just fixed) | 21:46 |
| clarkb | jeblair: cool will review shortly | 21:46 |
| openstackgerrit | Adrian Turjak proposed openstack-infra/project-config: Add new Cloud-Pydashie project https://review.openstack.org/142240 | 21:46 |
| *** tkelsey has quit IRC | 21:47 | |
| jeblair | clarkb: i think with a few more of those we will be in a good position to unit test your change | 21:47 |
| dhellmann | fungi: there's an example of doing that in the governance repo, let me find the link | 21:47 |
| jeblair | clarkb: (it has a generalized method of simulating any network packet delivery oddities we want to do) | 21:48 |
| *** julim has quit IRC | 21:48 | |
| dhellmann | fungi: https://review.openstack.org/#/c/125788/ | 21:48 |
| clarkb | jeblair: nice now I am very interested in reviewing | 21:48 |
| anteaya | anyone know a timothy r chavez? he seems to be reviewing infra patches but I don't know his irc nick | 21:48 |
| anteaya | not that I have a question for him, I'm just curious | 21:48 |
| clarkb | anteaya: timrc | 21:49 |
| anteaya | ah wonderful thanks | 21:49 |
| mordred | anteaya: he works for cody-somerville | 21:50 |
| fungi | dhellmann: perfect--thanks again | 21:50 |
| anteaya | mordred: does he | 21:50 |
| anteaya | til | 21:50 |
| dhellmann | fungi: let me know if I can help at all | 21:50 |
| anteaya | or today I remembered, not sure if I had learned that before or not | 21:51 |
| *** cameron__ has quit IRC | 21:51 | |
| dhellmann | clarkb: do we want to merge those patches into separately if we know adding that egg-info writer causes issues? I guess we want them to all land together? | 21:52 |
| fungi | dhellmann: gmurphy is working on https://review.openstack.org/140848 and i'm sure he'd love feedback (though for now we're still tweaking the generated rst to match what we want to use in e-mails too, for convenience) | 21:52 |
| clarkb | dhellmann: I think that ince we have a release that works we should be fine merging things as individual changes as long as they are all present in the next release | 21:52 |
| clarkb | dhellmann: since we consume via releases and the tests which install themselves pass | 21:53 |
| fungi | clarkb: dhellmann: yeah i think it should be safe to just land them in series as they come | 21:53 |
| openstackgerrit | James E. Blair proposed openstack-infra/gear: Add readPacket unit test https://review.openstack.org/142910 | 21:53 |
| dhellmann | fungi, clarkb : ok, sounds good | 21:54 |
| fungi | clarkb: reviewing them now | 21:54 |
| *** ivar-laz_ has quit IRC | 21:54 | |
| *** shayneburgess has joined #openstack-infra | 21:55 | |
| *** ivar-lazzaro has joined #openstack-infra | 21:55 | |
| *** ivar-lazzaro has quit IRC | 21:55 | |
| openstackgerrit | Merged openstack-infra/project-config: Support pre/post test hooks for nova-docker jobs https://review.openstack.org/142899 | 21:56 |
| fungi | clarkb: don't we really need the . addition to tox.ini testenv deps as early as possible to make sure we're not testing 0.10.4 where we think we're testing the change? | 21:56 |
| *** ivar-lazzaro has joined #openstack-infra | 21:56 | |
| clarkb | fungi: good qusetion. I think the answer is yes | 21:58 |
| fungi | clarkb: or just accept that the second change is clean and not worry about someone trying to use the first possibly buggy one? | 21:58 |
| dhellmann | fungi: I left some comments on that ossa review | 21:58 |
| fungi | i'm less thrilled about the second option, but it could be pragmatic | 21:58 |
| fungi | dhellmann: awesome--thanks! | 21:59 |
| clarkb | fungi: hrm | 21:59 |
| clarkb | let me at least check the first commit works with . in its path (which would indicate it is clean) | 21:59 |
| *** wenlock has quit IRC | 22:00 | |
| openstackgerrit | James E. Blair proposed openstack-infra/gear: Add readPacket unit test https://review.openstack.org/142910 | 22:00 |
| *** salv-orlando has joined #openstack-infra | 22:01 | |
| *** dannywil_ has quit IRC | 22:01 | |
| *** crowshan has joined #openstack-infra | 22:01 | |
| *** dkranz has quit IRC | 22:01 | |
| clarkb | jeblair: in ^ why slice the first packet in the test? | 22:02 |
| *** patrickeast_ has joined #openstack-infra | 22:02 | |
| clarkb | does that trip the eagain? | 22:02 |
| clarkb | I think that must be what that does. ignore me | 22:02 |
| *** johnthetubaguy is now known as zz_johnthetubagu | 22:03 | |
| *** patrickeast has quit IRC | 22:04 | |
| *** patrickeast_ is now known as patrickeast | 22:04 | |
| fungi | that way we get multiple passes | 22:04 |
| *** dannywilson has joined #openstack-infra | 22:05 | |
| *** mbacchi has quit IRC | 22:05 | |
| timrc | clarkb, Hi, reading up | 22:05 |
| timrc | But I can guess "wtf is this tim guy!?" | 22:05 |
| *** yamamoto has joined #openstack-infra | 22:05 | |
| dhellmann | clarkb: +2 on that series | 22:05 |
| *** atiwari has joined #openstack-infra | 22:05 | |
| clarkb | dhellmann: cool | 22:06 |
| clarkb | fungi: so using . at the base of the series I only run into timeout errors for gpg things | 22:06 |
| dhellmann | clarkb: thanks for doing the work to cherry-pick those | 22:06 |
| openstackgerrit | Adrian Turjak proposed openstack-infra/project-config: Add new Cloud-Pydashie project https://review.openstack.org/142240 | 22:06 |
| dhellmann | clarkb: those errors are lack of entropy | 22:06 |
| clarkb | dhellmann: no problem. It needed to be done and I learned about future string things | 22:06 |
| clarkb | dhellmann: ya, I am not worried about them | 22:06 |
| fungi | clarkb: hah! you need https://review.openstack.org/142884 | 22:06 |
| *** atiwari has quit IRC | 22:07 | |
| *** annegent_ has quit IRC | 22:07 | |
| *** wenlock has joined #openstack-infra | 22:07 | |
| *** annegent_ has joined #openstack-infra | 22:08 | |
| *** dprince has quit IRC | 22:08 | |
| clarkb | hrm https://jenkins05.openstack.org/job/gate-tempest-dsvm-neutron-src-pbr/19/console failed so we may not be done | 22:08 |
| *** atiwari has joined #openstack-infra | 22:08 | |
| *** gilliard is now known as gilliard_afk | 22:08 | |
| clarkb | did we unpin setuptools? | 22:09 |
| clarkb | it looks like we did | 22:09 |
| dhellmann | fungi, clarkb, mordred : when you do releases of pbr, do you use the oslo_release.sh script in openstack-infra/release-tools to update launchpad, too? | 22:09 |
| *** yamamoto has quit IRC | 22:10 | |
| clarkb | arg I don't think we can update pbr without setuptools pin reverted... | 22:10 |
| crowshan | anteaya: Hey, I messaged you earlier about the patch I wanted reviewed here: https://review.openstack.org/#/c/142908/. I also read the article, and I was wondering if there was a quicker way to get our patch reviewed since it's pretty small and straightforward | 22:10 |
| clarkb | dhellmann: I haven't done a release of pbr so no | 22:10 |
| fungi | clarkb: i guess i should have -2'd or wip'd https://review.openstack.org/142322 after you said let's not merge it yet | 22:10 |
| anteaya | crowshan: I see you change your irc nick | 22:11 |
| anteaya | crowshan: do take a look at all the patches currently open in project-config | 22:11 |
| crowshan | anteaya: yeah, I changed it so that it matched my gerrit info better | 22:11 |
| anteaya | crowshan: then help me understand why to me your patch is more important than any of the others? | 22:11 |
| fungi | clarkb: i can revert my revert. just a sec | 22:11 |
| dtroyer | fungi is headed into inception-land | 22:12 |
| *** thedodd has quit IRC | 22:12 | |
| clarkb | fungi: or we put the remove sha1 suffixes at the base of that stack | 22:12 |
| clarkb | fungi: I think that may work | 22:12 |
| clarkb | mordred: ^ did you say you had that handy? | 22:12 |
| dhellmann | clarkb: are the pbr tests failing, then? | 22:12 |
| clarkb | dhellmann: ya the integration run is failing because pbr master still outputs versions with the git sha | 22:13 |
| clarkb | dhellmann: fungi so I think if we put the git sha1 info removal change at the bottom of the stack we can get through without revert reverts in devstack | 22:13 |
| clarkb | but I don't know if that change has been written yet | 22:13 |
| *** weshay has quit IRC | 22:13 | |
| jeblair | clarkb: yep (eagain) -- that's so that we simulate the tcp stack giving us half the first gearman-packet in one tcp-packet, then eagain, then the second half of the first gearman-pcaket along with the second gearman-packet both in the second tcp-packet | 22:13 |
| *** hichtakk has joined #openstack-infra | 22:13 | |
| clarkb | jeblair: also, any chance you want to write a test case like that for worker/client :) | 22:13 |
| jeblair | clarkb: in progress. i'm currently trying to figure out why it fails! :) | 22:14 |
| clarkb | jeblair: if not I willswitch gears (no pun intended) from pbr once pbr is settled and work on that | 22:14 |
| clarkb | jeblair: perfect | 22:14 |
| dhellmann | clarkb, fungi : I'm heading offline to go to our local meetup. I'll check back in tomorrow to see if you need more pbr reviews. | 22:14 |
| fungi | clarkb: https://review.openstack.org/142912 is an option now anyway | 22:14 |
| clarkb | dhellmann: thank you for all the help have fun at meetup | 22:14 |
| fungi | dhellmann: thanks for the help! have a good get-together | 22:14 |
| *** gondoi is now known as zz_gondoi | 22:15 | |
| dhellmann | clarkb, fungi : thanks, we're talking about the election tonight. Woo! | 22:15 |
| jeblair | clarkb: also todo is admin requests for both | 22:15 |
| fungi | dhellmann: excellent! remind everyone to vote | 22:15 |
| anteaya | asselin: help me understand why we need this patch reference by the other patch: https://review.openstack.org/#/c/140548/4 to me this is just noise | 22:16 |
| clarkb | fungi: I think mordred may have the cahnge for removing git shas from master somewhere ut it may have been in a tag merge which is why I didn't like it | 22:16 |
| *** pblaho has joined #openstack-infra | 22:16 | |
| clarkb | mordred: ^ that sound right? | 22:16 |
| anteaya | asselin: as any repo with upstream /your-repo-here/ will require a new patch to update anyway | 22:16 |
| fungi | clarkb: looking--maybe we can just extract it | 22:16 |
| asselin | anteaya, discussed with nibalizer and decided to not do https://review.openstack.org/#/c/140548/, but to leave it as a ref, so I left it WIP. But perhaps there's another way to refer to it | 22:17 |
| anteaya | I don't need to refer to it at all | 22:18 |
| *** zz_zz_zz_zz_saba is now known as zz_zz_zz_zz_zz_s | 22:18 | |
| anteaya | it serves no purpose for me | 22:18 |
| fungi | clarkb: seems so, yes https://review.openstack.org/142323 | 22:18 |
| anteaya | as I know that the content will need to be in place as a series of individual patches | 22:18 |
| asselin | anteaya, it's for those who will do the split, but we can find another place for it | 22:18 |
| anteaya | it is just noise I have to look at to ensure I am giving a through review | 22:18 |
| anteaya | asselin: that information is contained the the spec | 22:19 |
| anteaya | which is why I requested the patch to the spec last night to keep it updated | 22:19 |
| anteaya | lets reference the spec please, not another patch | 22:19 |
| mordred | clarkb: yah, it's in the tag merge | 22:19 |
| asselin | anteaya, I think we're misunderstanding. | 22:19 |
| anteaya | perhaps I am | 22:20 |
| anteaya | do explain | 22:20 |
| asselin | anteaya, but I will abandon that patch since it causes confusion and not necessary | 22:20 |
| clarkb | mordred: I am a bit confused how that ended up there because the code is drastically different in that area between the two branches right? | 22:20 |
| anteaya | asselin: do help me to understand if I am missing a piece | 22:20 |
| clarkb | mordred: so cehrry-picking/merging doesn't help | 22:20 |
| mordred | I did it all by hand | 22:21 |
| clarkb | mordred: any chance you want ot propose that as a normal change so that the tag merge is a noop change? | 22:21 |
| asselin | anteaya, you're misunderstanding the purpose of that patch, and that is because the purpose is of very limited value. Hence I agree to just get rid of it. | 22:22 |
| mordred | yah | 22:22 |
| clarkb | mordred: I think it makes it much easier to review as the merge commits are funky | 22:22 |
| anteaya | asselin: okay great thank you | 22:22 |
| mordred | sure. will do in an hour or so | 22:22 |
| crowshan | anteaya: Thanks, I definitely didn't think our patch was more important, just very quick to do, but anyways I'll look into reviewing the other patches | 22:22 |
| clarkb | mordred: awesome. then we can put my stackatop that and we should be fine iwthout a devstack revert revert | 22:22 |
| anteaya | asselin: can we also change the commit message in the important patch to remove the #1 portion as I do believe it is now redundant | 22:22 |
| asselin | anteaya, yes agree. makes sense | 22:23 |
| anteaya | crowshan: thank you, we appreciate all the help to get through the review backlog | 22:23 |
| fungi | clarkb: which would b e great, because otherwise we'll need a devstack revert revert revert once the other stuff merges | 22:23 |
| anteaya | asselin: thank you | 22:23 |
| anteaya | what is the current record for the longest chain of reverts? | 22:23 |
| *** sigmavirus24 is now known as sigmavirus24_awa | 22:24 | |
| *** bdpayne has joined #openstack-infra | 22:24 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 22:24 | |
| *** zz_zz_zz_zz_zz_s is now known as sabari | 22:24 | |
| fungi | anteaya: not sure, but i don't think i want to be the one holding the record whatever it is | 22:24 |
| anteaya | fair enough | 22:24 |
| anteaya | hot potato | 22:24 |
| *** timcline has quit IRC | 22:25 | |
| * anteaya checks her hot potatos in the oven | 22:25 | |
| *** amitgandhinz has joined #openstack-infra | 22:27 | |
| *** tonytan4ever has quit IRC | 22:27 | |
| openstackgerrit | Ramy Asselin proposed openstack-infra/project-config: Prepare project-config for puppet module split https://review.openstack.org/140523 | 22:27 |
| openstackgerrit | James E. Blair proposed openstack-infra/gear: Add unit test for readPacket (client mode) https://review.openstack.org/142914 | 22:28 |
| jeblair | clarkb: ^ client side. next admin. | 22:28 |
| asselin | anteaya, done ^^ | 22:28 |
| anteaya | asselin: thank you | 22:28 |
| anteaya | mtreinish: so the way I read https://review.openstack.org/#/c/136234/ you are still waiting on a bug fix, yeah? | 22:29 |
| mtreinish | anteaya: nope, the bugs have been fixed, and I pushed a release today for that | 22:30 |
| openstackgerrit | James E. Blair proposed openstack-infra/gear: Add unit test for readPacket (client mode) https://review.openstack.org/142914 | 22:30 |
| anteaya | mtreinish: sorry perhaps I didn't read the storyboard story correctly | 22:31 |
| openstackgerrit | K Jonathan Harker proposed openstack-infra/project-config: Split out lodgeit module https://review.openstack.org/142917 | 22:31 |
| * anteaya reads again | 22:31 | |
| mtreinish | anteaya: I guess I need to go and update storyboard, I'm just used to it being automatic | 22:31 |
| anteaya | mtreinish: ah perhaps that is it | 22:31 |
| *** funzo has quit IRC | 22:32 | |
| openstackgerrit | K Jonathan Harker proposed openstack-infra/system-config: Split out lodgeit module https://review.openstack.org/142918 | 22:32 |
| *** funzo has joined #openstack-infra | 22:33 | |
| mtreinish | anteaya: ok I updated https://storyboard.openstack.org/#!/story/2000023 | 22:33 |
| *** sweston_ has joined #openstack-infra | 22:33 | |
| fungi | BobBall: you seem to have been the instigator of the xen readme in devstack... the url here is no longer valid: https://git.openstack.org/cgit/openstack-dev/devstack/tree/tools/xen/README.md#n63 | 22:34 |
| *** sweston has quit IRC | 22:34 | |
| *** sweston_ is now known as sweston | 22:34 | |
| anteaya | mtreinish: lovely thank you | 22:34 |
| fungi | BobBall: any insights into how that needs to be changed? | 22:34 |
| * mtreinish misses the auto bug stuff... | 22:35 | |
| *** dims__ has quit IRC | 22:35 | |
| ashp | hmm, I just noticed that all our jobs are set to concurrent: false | 22:36 |
| ashp | despite our jenkins_job_builder being set for true | 22:36 |
| anteaya | asselin: can you co-ordinate with jesusaurus on the lodgeit module split? | 22:36 |
| ashp | Is there anywhere else in jenkins that this could be overridden that I'm unaware of that someone here knows? | 22:36 |
| anteaya | asselin: https://review.openstack.org/#/c/142918/ | 22:36 |
| *** avozza is now known as zz_avozza | 22:37 | |
| clarkb | ashp: is that a default flag that is being overridden in jjb? | 22:37 |
| asselin | anteaya, looking | 22:37 |
| *** mattfarina has quit IRC | 22:37 | |
| ashp | clarkb: I'm not sure, and I'm sort of confused, the only concurrent mention I have is in defaults.yaml | 22:37 |
| ashp | /var/jenkins/job_builder/defaults.yaml: concurrent: true | 22:37 |
| clarkb | ashp: if you edit that section in jobs that value wil be overridden | 22:38 |
| ashp | none of the jobs overwrite it | 22:38 |
| ashp | that was with a grep -ri | 22:38 |
| asselin | jesusaurus, you have time to coordinate on "split out lodgeit module" https://review.openstack.org/#/c/142918/ | 22:39 |
| *** Masahiro has joined #openstack-infra | 22:39 | |
| ashp | oh we're running 0.9, I wonder if this is just fixed | 22:39 |
| anteaya | mtreinish: so this adds the script is there a patch up that puts it in play? if yes can we get a link to it in the commit message? | 22:39 |
| *** pblaho has quit IRC | 22:40 | |
| mtreinish | anteaya: not yet, I wasn't exactly sure on how to do that. clarkb had indicated that it was automagic | 22:40 |
| anteaya | perhaps it is | 22:40 |
| clarkb | ya all nodepool scripts end up on the nodes during builds | 22:41 |
| anteaya | cool | 22:41 |
| anteaya | okay thanks | 22:41 |
| anteaya | mtreinish: I'm going to +1 because I can't see a reason to -1 put my knowledge of nodepool scripts is weak | 22:41 |
| *** zz_avozza is now known as avozza | 22:42 | |
| mtreinish | clarkb: and they're all executed too? That's the real question, because this will only work after the tempest repo is pulled in | 22:42 |
| mtreinish | anteaya: ok cool, thanks | 22:42 |
| clarkb | no you have to call them | 22:42 |
| *** unicell has quit IRC | 22:42 | |
| clarkb | via the enttry script listed in nodepool.yaml | 22:42 |
| openstackgerrit | Dave Walker proposed openstack-dev/pbr: Ignore non-release related tags https://review.openstack.org/142144 | 22:42 |
| anteaya | mtreinish: thank you | 22:42 |
| mtreinish | clarkb: ok that's what I was really asking before. Thanks, I'll push that too | 22:42 |
| openstackgerrit | K Jonathan Harker proposed openstack-infra/infra-specs: Use the correct project name https://review.openstack.org/142920 | 22:44 |
| *** Masahiro has quit IRC | 22:44 | |
| *** teran_ has quit IRC | 22:45 | |
| *** kgiusti has left #openstack-infra | 22:45 | |
| *** mfink__ has quit IRC | 22:45 | |
| *** rlandy has quit IRC | 22:47 | |
| clarkb | jeblair: you inverted the fake socket blocking value but don't toggle it for either the client or server | 22:47 |
| clarkb | jeblair: shouldn't one o fthem toggle the flag? I think server shoul dset blocking=0 | 22:48 |
| *** dannywilson has quit IRC | 22:49 | |
| *** dannywilson has joined #openstack-infra | 22:50 | |
| ashp | clarkb: 1.0 fixed it instantly, huh | 22:50 |
| *** btran has joined #openstack-infra | 22:51 | |
| *** mfink has joined #openstack-infra | 22:52 | |
| openstackgerrit | James E. Blair proposed openstack-infra/gear: Add admin readPacket unit tests https://review.openstack.org/142924 | 22:53 |
| jeblair | clarkb: where are you talking about? | 22:53 |
| *** atiwari has quit IRC | 22:53 | |
| btran | Hi, when setting up my devstack env, I keep getting this error: 'OSError: [Errno 11] Resource temporarily unavailable' at different service startups locations (c-api, n-api, etc.). Running on physical Ubuntu 12.04 server. Any ideas what my problem is? | 22:54 |
| fungi | clarkb: dtroyer: actually we do need to emergency merge and promote https://review.openstack.org/142912 until there's another setuptools release | 22:54 |
| openstackgerrit | Merged openstack-infra/elastic-recheck: Add fingerprint for bug 1403284 https://review.openstack.org/142310 | 22:54 |
| uvirtbot | Launchpad bug 1403284 in devstack "mysql (1040, 'Too many connections')" [Undecided,In progress] https://launchpad.net/bugs/1403284 | 22:54 |
| clarkb | jeblair: in the change that adds the client test, you go from if blocking to if not blocking | 22:54 |
| clarkb | jeblair: iirc the server is blocking and the client/workers are not | 22:54 |
| jeblair | clarkb: right. the logic there is the same because i also inverted the contents of the conditional | 22:55 |
| jeblair | clarkb: i just made both of the conditionals in that function match | 22:55 |
| clarkb | oh I see. Is something else setting the blocking flag? | 22:55 |
| jeblair | clarkb: the socket is set to blocking by ServerConnection | 22:55 |
| jeblair | clarkb: (so it's part of the tested code path) | 22:55 |
| clarkb | gotcha | 22:55 |
| *** mriedem has quit IRC | 22:56 | |
| anteaya | btran: you will probably get better receiption in #openstack-dev | 22:56 |
| btran | anteaya: ok thanks, will do that | 22:57 |
| anteaya | bradm: thanks and welcome to openstack | 22:57 |
| *** signed8b_ has joined #openstack-infra | 22:58 | |
| *** bdpayne_ has joined #openstack-infra | 22:58 | |
| *** david-ly_ has joined #openstack-infra | 22:58 | |
| *** packet has quit IRC | 22:58 | |
| *** rmcall has quit IRC | 23:00 | |
| jogo | clarkb: FYI http://lists.openstack.org/pipermail/openstack-operators/2014-December/005739.html | 23:00 |
| dtroyer | fungi: I'm +2 on 142912…sdague may have headed to family-land | 23:00 |
| jogo | thought you may want to comment on that thread | 23:00 |
| fungi | dtroyer: i went ahead and used gerrit admin powers to approve it | 23:00 |
| mordred | clarkb: ok. I'm back. sorry for the delay | 23:00 |
| dtroyer | fungi: np, was just going to ask about +A anyway | 23:01 |
| mordred | clarkb: on my plate now is remaking that merge commit as a flat commit on top of a null-merge for tags, right? | 23:01 |
| fungi | dtroyer: it's now been promoted ahead of the nova change in the gate it was in the process of killing | 23:01 |
| *** rmcall has joined #openstack-infra | 23:01 | |
| *** rmcall has quit IRC | 23:01 | |
| fungi | dtroyer: thanks for the speedy assist! | 23:02 |
| *** wenlock has quit IRC | 23:02 | |
| dtroyer | fungi: np, been watching this drama from the sidelines all week... | 23:03 |
| *** david-lyle has quit IRC | 23:03 | |
| *** bdpayne has quit IRC | 23:03 | |
| *** signed8bit has quit IRC | 23:03 | |
| *** SumitNaiksatam has quit IRC | 23:03 | |
| *** camunoz has quit IRC | 23:03 | |
| *** timrc has quit IRC | 23:03 | |
| *** ianw has quit IRC | 23:03 | |
| *** bcourt has quit IRC | 23:03 | |
| *** sileht has quit IRC | 23:03 | |
| *** andreykurilin has quit IRC | 23:03 | |
| *** rmcall has joined #openstack-infra | 23:03 | |
| fungi | clarkb: so anyway, the pep8 pin will be back in place in about an hour anyway, barring random gate-breaking bug killing the revert | 23:03 |
| fungi | er, s/pep8/setuptools 8/ | 23:04 |
| *** SumitNaiksatam has joined #openstack-infra | 23:04 | |
| *** jamielennox|away is now known as jamielennox | 23:05 | |
| mordred | fungi: we re-pinned? | 23:05 |
| *** ddieterly has quit IRC | 23:05 | |
| fungi | keeping an eye on it since it's my bad for not setting it wip after we realized it would actually pass jobs if they got lucky enough to run on the right provider | 23:05 |
| *** ddieterly has joined #openstack-infra | 23:06 | |
| fungi | mordred: yeah, we're still waiting for a setuptools release which silences the runtime warning by default, since we have tempest tests which fail parsing output otherwise | 23:06 |
| mordred | ah, yah. gotcha | 23:06 |
| *** camunoz has joined #openstack-infra | 23:06 | |
| *** timrc has joined #openstack-infra | 23:06 | |
| *** sileht has joined #openstack-infra | 23:07 | |
| *** bcourt has joined #openstack-infra | 23:07 | |
| *** ianw has joined #openstack-infra | 23:07 | |
| fungi | mordred: except we only have libs installed which trigger that on hpcloud. so it got lucky enough to run all affected tempest jobs on rackspace twice in a row | 23:07 |
| *** e0ne has quit IRC | 23:07 | |
| fungi | (and then demolished the gate for nearly an hour) | 23:09 |
| *** esker has quit IRC | 23:09 | |
| *** ddieterly has quit IRC | 23:11 | |
| *** kumartin has quit IRC | 23:11 | |
| openstackgerrit | Joe Gordon proposed openstack-infra/devstack-gate: rename subnode log directory https://review.openstack.org/140529 | 23:11 |
| *** e0ne has joined #openstack-infra | 23:13 | |
| clarkb | mordred: I think we just need the remove git sha change | 23:13 |
| mordred | sigh | 23:13 |
| clarkb | mordred: ? | 23:13 |
| mordred | I don't HAVE just that - I originally did this with git merge | 23:13 |
| *** otter768 has joined #openstack-infra | 23:14 | |
| mordred | I have the pbr.json and the git sha removal as one thing | 23:14 |
| clarkb | mordred: I think if ou put those tow together thats fine | 23:14 |
| mordred | phew. cool | 23:14 |
| clarkb | mordred: then I can drop the first change in my stack | 23:14 |
| *** dkliban is now known as dkliban_afk | 23:15 | |
| openstackgerrit | Ramy Asselin proposed openstack-infra/project-config: Add puppet-recheckwatch as split out module https://review.openstack.org/140430 | 23:15 |
| *** achanda has quit IRC | 23:15 | |
| clarkb | jeblair: my change fails atop your tests I am working on that now | 23:15 |
| openstackgerrit | Ramy Asselin proposed openstack-infra/infra-specs: Remove 2 modules that no longer exist https://review.openstack.org/142627 | 23:16 |
| openstackgerrit | Jamie Lennox proposed openstack/requirements: Update keystoneclient to 1.0.0 https://review.openstack.org/142929 | 23:17 |
| *** e0ne has quit IRC | 23:17 | |
| openstackgerrit | Monty Taylor proposed openstack-dev/pbr: Merge branch 'feature/0.10' https://review.openstack.org/142930 | 23:17 |
| openstackgerrit | Monty Taylor proposed openstack-dev/pbr: Port in git sha changes from 0.10 line https://review.openstack.org/142931 | 23:17 |
| mordred | clarkb: voila! | 23:17 |
| jeblair | clarkb: awesome! :) | 23:17 |
| *** otter768 has quit IRC | 23:18 | |
| clarkb | jeblair: looks like it gets the first packet ok but not the second because the fake socket raises eagain | 23:19 |
| clarkb | and client side hits eof. so I think I am being too greedy | 23:20 |
| clarkb | mordred: thank you. I will rebase in a bit | 23:20 |
| *** EmilienM is now known as EmilienM|pto | 23:21 | |
| *** pc_m has joined #openstack-infra | 23:21 | |
| clarkb | jeblair: woot think I figured it out | 23:23 |
| clarkb | jeblair: new patchset pushed, needed to reset need_bytes after getting bytes | 23:24 |
| openstackgerrit | Clark Boylan proposed openstack-infra/gear: Read more bytes per readPacket iteration https://review.openstack.org/140186 | 23:24 |
| clarkb | jeblair: also, when you have a chance I would like to talk about the situation where we set raw_bytes = b'' | 23:24 |
| jeblair | clarkb: k, reading your patch now | 23:25 |
| *** ryanpetrello has quit IRC | 23:25 | |
| *** amitgandhinz has quit IRC | 23:26 | |
| jeblair | clarkb: ok | 23:29 |
| clarkb | jeblair: so we read segment and if not segment then raw_bytes = b'' and return None | 23:29 |
| *** SumitNaiksatam has quit IRC | 23:29 | |
| *** wenlock has joined #openstack-infra | 23:29 | |
| *** jedimike has joined #openstack-infra | 23:29 | |
| *** SumitNaiksatam has joined #openstack-infra | 23:30 | |
| clarkb | jeblair: my concern with that is we may have valid packet data in raw_bytes we have just thrown out and the next read will assume the middle of a packet is the beginning of a packet | 23:30 |
| clarkb | jeblair: wouldn't it be better to keep raw_bytes as is and try again? or possibly close and restart that connection? | 23:30 |
| jeblair | clarkb: except there is no next read. that means it's disconnected. | 23:30 |
| clarkb | ah ok so that is the situation | 23:30 |
| jeblair | clarkb: so if we receive the empty string from recv and don't already have a packet, then we'll never get one | 23:31 |
| clarkb | I read recv(2) and at least in C land that may have been the cause but wasnt sure about python | 23:31 |
| jeblair | clarkb: that was mostly in there to make sure that on reconnect, we start with a good clean slate | 23:31 |
| jeblair | clarkb: however, i note the connect() method resets things properly | 23:31 |
| *** jedimike has quit IRC | 23:31 | |
| jeblair | clarkb: so if you want to drop the line, i think it will be okay. | 23:31 |
| clarkb | jeblair: I may just add a comment there. I think being defensive is fine | 23:31 |
| clarkb | ok now to rebase on mordeds change then clean up the admin detection | 23:32 |
| jeblair | clarkb: well, if we leave it there, then i think we also should clean up needs_bytes similarly... i'll leave a comment | 23:32 |
| clarkb | jeblair: thanks | 23:33 |
| clarkb | hrm rebase did not do what I expected there | 23:34 |
| *** ddieterly has joined #openstack-infra | 23:34 | |
| jeblair | clarkb: i think we still may have an issue on the client side in that readPacket will only be called once by the poll loop | 23:34 |
| jeblair | clarkb: (that's not covered by the new tests) | 23:35 |
| clarkb | oh wrong branch | 23:35 |
| clarkb | jeblair: yup | 23:35 |
| *** bswartz has joined #openstack-infra | 23:35 | |
| jeblair | clarkb: we could probably loop within the client poll loop to repeat the call to readPacket if Connection.input_buffer is not empty | 23:36 |
| openstackgerrit | Clark Boylan proposed openstack-dev/pbr: Only import sphinx during hook processing https://review.openstack.org/142840 | 23:36 |
| openstackgerrit | Clark Boylan proposed openstack-dev/pbr: Properly check for git before getting git dir https://review.openstack.org/142841 | 23:36 |
| clarkb | mordred: ^ is the other half | 23:37 |
| clarkb | mordred: I think you have failing tests that need updating | 23:38 |
| clarkb | silly tests | 23:39 |
| mordred | clarkb: /me stabs | 23:39 |
| *** rmcall has quit IRC | 23:40 | |
| *** dmsimard is now known as dmsimard_away | 23:41 | |
| NobodyCam | afternoon INFRA folk. would this be the correct place about a OpenStack Proposal Bot patch? | 23:42 |
| clarkb | NobodyCam: yup | 23:43 |
| NobodyCam | clarkb: TY its a little old but : https://review.openstack.org/#/c/128965 | 23:43 |
| *** crowshan has quit IRC | 23:44 | |
| *** r-daneel has quit IRC | 23:44 | |
| Daviey | Maybe it is too late now... but did anyone look to do an OpenStack all-in-one as for Qemu Advent Calendar? | 23:44 |
| *** sputnik13 has quit IRC | 23:46 | |
| fungi | Daviey: not to my knowledge. these days anyone worth their handlebar mustache and skinny jeans is making a docker image for that instead | 23:46 |
| Daviey | fungi: :) | 23:46 |
| *** sputnik13 has joined #openstack-infra | 23:46 | |
| clarkb | NobodyCam: and the question is? why is that test case removed? | 23:48 |
| *** pc_m_ has joined #openstack-infra | 23:48 | |
| NobodyCam | yep | 23:48 |
| clarkb | I love how people just go "I DON"T UNDERSTAND -1" | 23:48 |
| NobodyCam | gota love the comment ?!?!?!? | 23:48 |
| NobodyCam | I want to understand | 23:48 |
| NobodyCam | soo I ask | 23:48 |
| clarkb | so we merge tags back into master | 23:48 |
| clarkb | we do this so that when you are on master you always have that tag in your history | 23:49 |
| clarkb | so that logically on the git graph that version occurred in the psat prior to now | 23:49 |
| clarkb | we do this in a way that keeps master's code with a pointer to the old code. So its a special merge | 23:49 |
| clarkb | NobodyCam: the unfortunate bit is that gerrit doesn't render this well | 23:50 |
| *** SumitNaiksatam has quit IRC | 23:50 | |
| clarkb | NobodyCam: so the best way to review those is with git diff locally. you can do `git diff HEAD^` to see that there shouldn't be any diff against master | 23:50 |
| clarkb | and you can `git describe --always` to see that the tag is now in your history. But you have to git remote update before you do that to ensure you have the tag available | 23:51 |
| *** SumitNaiksatam has joined #openstack-infra | 23:51 | |
| *** pc_m has quit IRC | 23:51 | |
| *** gyee has joined #openstack-infra | 23:51 | |
| clarkb | and I can confirm running those that this is the case | 23:52 |
| *** pc_m_ has left #openstack-infra | 23:52 | |
| *** pc_m_ has joined #openstack-infra | 23:52 | |
| NobodyCam | ahh ... ok I'll test that here.... TY clarkb :) | 23:53 |
| clarkb | and if you grep for the test that is "removed" it is there | 23:53 |
| clarkb | I think that means that that test may have been removed on the tag side? | 23:53 |
| *** signed8b_ is now known as signed8bit | 23:54 | |
| *** fandi has quit IRC | 23:54 | |
| *** annegent_ has quit IRC | 23:55 | |
| *** baoli_ has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!