| *** denis_makogon has quit IRC | 00:01 | |
| *** johnthetubaguy has quit IRC | 00:01 | |
| *** michchap_ has joined #openstack-infra | 00:01 | |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/config: Convert subunit v1 log results to v2 format. https://review.openstack.org/57798 | 00:01 |
|---|---|---|
| clarkb | lifeless: that is basically what we had before, does it look sane to you? | 00:02 |
| clarkb | jeblair: I think we should get that change in asap | 00:02 |
| *** dangers is now known as danger_fo_away | 00:03 | |
| *** michchap has quit IRC | 00:03 | |
| *** mrodden1 has joined #openstack-infra | 00:04 | |
| *** mrodden has quit IRC | 00:06 | |
| jeblair | clarkb: lgtm | 00:07 |
| clarkb | it was working at one point but got lost in the shuffle of moving scripts around | 00:07 |
| openstackgerrit | Monty Taylor proposed a change to openstack-dev/pbr: Enable wheel processing in the tests https://review.openstack.org/56817 | 00:10 |
| openstackgerrit | Monty Taylor proposed a change to openstack-dev/pbr: Clean up integration script https://review.openstack.org/56816 | 00:10 |
| openstackgerrit | Monty Taylor proposed a change to openstack-dev/pbr: Serve local mirror using apache https://review.openstack.org/56752 | 00:10 |
| openstackgerrit | Monty Taylor proposed a change to openstack-dev/pbr: Use wheels for installation https://review.openstack.org/48803 | 00:10 |
| clarkb | jeblair: I am going to put that through, I will keep an eye on tests and revert it really quick if it causes problems, but I am pretty confident in it as we had it before | 00:10 |
| *** sarob has joined #openstack-infra | 00:10 | |
| *** loq_mac has joined #openstack-infra | 00:11 | |
| *** johnthetubaguy has joined #openstack-infra | 00:12 | |
| *** johnthetubaguy has quit IRC | 00:12 | |
| jog0 | so by my calculations the check jobs were failing at a rate of about 87% | 00:14 |
| clarkb | jog0: that change I just approved to run-unittests.sh should make the neutron unittests finish much quicker | 00:14 |
| jog0 | clarkb: woot | 00:15 |
| clarkb | so we probably don't need to worry about the timeout on that job once it gets in | 00:15 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/config: Remove jeepyb from the requirements gate https://review.openstack.org/56764 | 00:15 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/config: Create and upload wheels https://review.openstack.org/56760 | 00:15 |
| *** hogepodge has quit IRC | 00:16 | |
| *** hogepodge has joined #openstack-infra | 00:17 | |
| *** crank has quit IRC | 00:19 | |
| *** atiwari has quit IRC | 00:19 | |
| *** crank has joined #openstack-infra | 00:19 | |
| *** lchen has joined #openstack-infra | 00:20 | |
| openstackgerrit | A change was merged to openstack-infra/config: Convert subunit v1 log results to v2 format. https://review.openstack.org/57798 | 00:22 |
| *** mrodden1 has quit IRC | 00:23 | |
| *** mriedem1 has quit IRC | 00:24 | |
| *** loq_mac has quit IRC | 00:27 | |
| *** wenlock has quit IRC | 00:29 | |
| *** sarob has quit IRC | 00:29 | |
| *** sarob has joined #openstack-infra | 00:30 | |
| *** sarob has quit IRC | 00:35 | |
| jog0 | lifeless: ping | 00:35 |
| jog0 | tesr question | 00:35 |
| jog0 | testr | 00:35 |
| clarkb | I can attempt to answer maybe | 00:35 |
| lifeless | jog0: shoot | 00:36 |
| jog0 | so it came to our attention that testr runs the tests in a deterministic order the first time its run | 00:36 |
| jog0 | can we make that random | 00:36 |
| lifeless | you are misinformed. | 00:37 |
| jog0 | good | 00:37 |
| lifeless | testr doesn't run tests. | 00:37 |
| jog0 | it has a scheduler though right? | 00:37 |
| lifeless | It schedules tests; the first time schedule has no timing data so is entirely round robin. | 00:37 |
| lifeless | Thats set based, so will be wildly different on different Python versions and architectures. | 00:37 |
| lifeless | However | 00:38 |
| jog0 | so instead of roudn robin can we do random? | 00:38 |
| lifeless | testtools.run sorts tests always at OpenStacks request :) | 00:38 |
| *** MarkAtwood has quit IRC | 00:38 | |
| *** sarob has joined #openstack-infra | 00:38 | |
| jog0 | so we can undo that | 00:39 |
| lifeless | jog0: perhaps we can talk about the big picture | 00:39 |
| jog0 | or sort randomly or something | 00:39 |
| clarkb | you don't want to undo that, it makes test bisection harder | 00:39 |
| lifeless | jog0: rather than the proposed solution | 00:39 |
| sdague | so the issue is, we're doing post mortem on the grenade fail | 00:40 |
| sdague | and the problem is, the grenade job was completely stable when it merged | 00:40 |
| sdague | grenade job == using parallel tempest at the end of grenade | 00:40 |
| sdague | then 1 new tempest test landed | 00:41 |
| sdague | so the deterministic ordering of all the tests became a new deterministic ordering, which had the possibility of an out of memory race | 00:41 |
| lifeless | ok, so you want to find out about such pain earlier? | 00:42 |
| sdague | yes | 00:42 |
| sdague | and more often | 00:42 |
| sdague | so that we'd get more data from our multiple tempest runs on each job, instead of them all running the tests in the same order | 00:43 |
| *** yamahata_ has joined #openstack-infra | 00:43 | |
| lifeless | ok, so for clarity: there are two layers here: testr scheduling to backends, and backends running tests. | 00:44 |
| sdague | yep | 00:44 |
| lifeless | If we testresources in place as I want to, ordering w/in a backend will be fixed | 00:44 |
| lifeless | because it's an optimising ordering | 00:44 |
| sdague | lifeless: but how would that work in the gate, where it's fresh guest each time | 00:44 |
| lifeless | and the long term plan for exporting that data to testr to get optimising scheduling from testr too | 00:44 |
| *** mrodden has joined #openstack-infra | 00:45 | |
| sdague | we have nothing to optimize with | 00:45 |
| lifeless | testresources has metadata in the suite about the costs of bringing up fixtures | 00:45 |
| lifeless | it's how it works | 00:45 |
| lifeless | it builds a graph of all the resource transitions needed and generates a least cost approximate walk through it (using simulated annealing) | 00:45 |
| *** mrodden1 has joined #openstack-infra | 00:46 | |
| lifeless | I'm very surprised that you were seeing consistent partitioning from testr | 00:46 |
| sdague | lifeless: why? | 00:46 |
| lifeless | but I'd be delighted to add randomness in there | 00:47 |
| sdague | it's like random num generator with a fixed seed | 00:47 |
| *** changbl has quit IRC | 00:47 | |
| lifeless | sdague: sure | 00:48 |
| sdague | anyway, we're running off here for a bit. Just an idea. | 00:48 |
| *** mrodden has quit IRC | 00:49 | |
| lifeless | so, low hanging fruit - submit a bug to make testr perturbate the round robin allocation it does. | 00:49 |
| clarkb | oh thats right you guys are all together in berkely | 00:49 |
| lifeless | Happy to have that. | 00:49 |
| lifeless | testtools is more concerning - we filed and fixed it upstream in Python | 00:49 |
| lifeless | and I'm not sure that our medium term plan for tempest optimisation is compatible with randomness within a backend | 00:50 |
| lifeless | s/bug/patch/ | 00:50 |
| lifeless | I have to pop out for a little myself | 00:50 |
| clarkb | woot https://jenkins02.openstack.org/job/gate-cinder-python26/2901/console subunit format conversion seems fine | 00:55 |
| *** michchap_ has quit IRC | 00:56 | |
| *** xeyed4good has joined #openstack-infra | 00:58 | |
| *** michchap has joined #openstack-infra | 00:59 | |
| clarkb | now I need a guinea pig neutron change | 00:59 |
| *** ^d has quit IRC | 00:59 | |
| mordred | clarkb: https://review.openstack.org/#/c/55805/ | 00:59 |
| mordred | it's one of mine that's broken right now | 00:59 |
| mordred | you could recheck it | 01:00 |
| clarkb | cool thanks | 01:00 |
| mordred | without confusing anyone | 01:00 |
| *** nati_ueno has quit IRC | 01:01 | |
| *** nati_ueno has joined #openstack-infra | 01:01 | |
| *** colinmcnamara has joined #openstack-infra | 01:03 | |
| *** xeyed4good has left #openstack-infra | 01:05 | |
| clarkb | btw zuul is going to do a massive set of merges in a little bit \o/ | 01:05 |
| *** hogepodge has quit IRC | 01:06 | |
| *** colinmcnamara has quit IRC | 01:06 | |
| mordred | w00t! | 01:07 |
| clarkb | currently up to ~23 changes | 01:08 |
| clarkb | almost half the pipeline | 01:08 |
| *** reed has joined #openstack-infra | 01:16 | |
| *** rongze has joined #openstack-infra | 01:23 | |
| *** rongze has quit IRC | 01:27 | |
| *** nati_uen_ has joined #openstack-infra | 01:31 | |
| *** nati_uen_ has quit IRC | 01:31 | |
| *** nati_uen_ has joined #openstack-infra | 01:31 | |
| *** nati_ueno has quit IRC | 01:32 | |
| clarkb | :( the rsyslog stuff for the slaves conflicts with different rsyslog stuff in puppet | 01:35 |
| *** boris-42 has joined #openstack-infra | 01:36 | |
| *** markwash has quit IRC | 01:37 | |
| *** nosnos has joined #openstack-infra | 01:40 | |
| *** nosnos has quit IRC | 01:40 | |
| *** sarob_ has joined #openstack-infra | 01:40 | |
| *** nosnos has joined #openstack-infra | 01:41 | |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/config: Handle multiple rsyslog service defs. https://review.openstack.org/57807 | 01:42 |
| clarkb | that is a bit of a hackish workaround but should allow puppet to run on the slaves again | 01:42 |
| *** sarob has quit IRC | 01:44 | |
| clarkb | ran it on precise23 to double check it works and seems to be happy | 01:45 |
| *** sarob_ has quit IRC | 01:45 | |
| *** nati_uen_ has quit IRC | 01:46 | |
| *** boris-42 has quit IRC | 01:55 | |
| *** boris-42 has joined #openstack-infra | 01:56 | |
| *** boris-42 has quit IRC | 01:58 | |
| *** bingbu has joined #openstack-infra | 01:59 | |
| *** yamahata_ has quit IRC | 02:02 | |
| *** dcramer_ has joined #openstack-infra | 02:02 | |
| *** rongze has joined #openstack-infra | 02:03 | |
| *** ljjjustin has joined #openstack-infra | 02:10 | |
| *** jhesketh has joined #openstack-infra | 02:12 | |
| *** yaguang has joined #openstack-infra | 02:16 | |
| *** boris-42 has joined #openstack-infra | 02:17 | |
| *** mriedem has joined #openstack-infra | 02:18 | |
| *** dstanek has quit IRC | 02:20 | |
| *** jhesketh__ has joined #openstack-infra | 02:22 | |
| *** sjing has joined #openstack-infra | 02:36 | |
| *** dcramer_ has quit IRC | 02:37 | |
| openstackgerrit | A change was merged to openstack-infra/reviewstats: Fix disagreement percentage calculation https://review.openstack.org/57606 | 02:46 |
| *** sarob has joined #openstack-infra | 02:52 | |
| *** markwash has joined #openstack-infra | 02:53 | |
| openstackgerrit | A change was merged to openstack-infra/devstack-gate: Add stable/havana to devstack-gate https://review.openstack.org/52437 | 02:53 |
| *** nati_ueno has joined #openstack-infra | 02:55 | |
| *** boris-42 has quit IRC | 02:57 | |
| *** sarob has quit IRC | 02:58 | |
| *** sarob has joined #openstack-infra | 02:59 | |
| *** sarob has quit IRC | 02:59 | |
| *** sarob has joined #openstack-infra | 02:59 | |
| *** wenlock has joined #openstack-infra | 03:02 | |
| *** sarob has quit IRC | 03:06 | |
| *** sarob has joined #openstack-infra | 03:07 | |
| *** lchen has quit IRC | 03:07 | |
| *** sarob has quit IRC | 03:11 | |
| *** dstanek has joined #openstack-infra | 03:13 | |
| *** pcrews has quit IRC | 03:14 | |
| *** sarob has joined #openstack-infra | 03:16 | |
| *** jgrimm has joined #openstack-infra | 03:18 | |
| *** ljjjustin has quit IRC | 03:23 | |
| *** lchen has joined #openstack-infra | 03:23 | |
| *** reed has quit IRC | 03:25 | |
| *** CaptTofu_ has quit IRC | 03:26 | |
| *** krtaylor has joined #openstack-infra | 03:26 | |
| *** CaptTofu has joined #openstack-infra | 03:26 | |
| *** gyee has quit IRC | 03:45 | |
| *** nati_ueno has quit IRC | 03:47 | |
| *** sarob has quit IRC | 03:51 | |
| *** sarob_ has joined #openstack-infra | 03:51 | |
| *** nati_ueno has joined #openstack-infra | 04:07 | |
| *** dkliban_ has joined #openstack-infra | 04:12 | |
| mikal | jog0: you still around? | 04:12 |
| *** julim has joined #openstack-infra | 04:13 | |
| *** nati_ueno has quit IRC | 04:17 | |
| *** nati_ueno has joined #openstack-infra | 04:18 | |
| *** jgrimm has quit IRC | 04:18 | |
| *** dcramer_ has joined #openstack-infra | 04:20 | |
| *** julim has quit IRC | 04:23 | |
| *** mriedem has quit IRC | 04:24 | |
| *** markmcclain has joined #openstack-infra | 04:25 | |
| *** cody-somerville has quit IRC | 04:27 | |
| *** cody-somerville has joined #openstack-infra | 04:42 | |
| *** SergeyLukjanov has joined #openstack-infra | 04:43 | |
| *** nati_uen_ has joined #openstack-infra | 04:45 | |
| cyeoh | jog0: FWIW the v3 enable patch is now passing grenade. https://review.openstack.org/#/c/57589/ | 04:46 |
| *** nati_ueno has quit IRC | 04:48 | |
| *** colinmcn_ has joined #openstack-infra | 04:51 | |
| *** chandankumar has joined #openstack-infra | 04:53 | |
| *** rongze has quit IRC | 04:53 | |
| *** colinmcn_ has quit IRC | 04:58 | |
| *** sandywalsh has quit IRC | 04:58 | |
| *** changbl has joined #openstack-infra | 05:02 | |
| *** sarob_ is now known as sarob | 05:05 | |
| *** matsuhashi has joined #openstack-infra | 05:06 | |
| *** ljjjustin has joined #openstack-infra | 05:09 | |
| *** markmcclain has quit IRC | 05:13 | |
| *** markmcclain has joined #openstack-infra | 05:15 | |
| *** matsuhashi has quit IRC | 05:16 | |
| *** sarob has quit IRC | 05:17 | |
| *** sandywalsh has joined #openstack-infra | 05:18 | |
| *** SergeyLukjanov has quit IRC | 05:21 | |
| *** matsuhashi has joined #openstack-infra | 05:23 | |
| *** rongze has joined #openstack-infra | 05:24 | |
| *** sarob has joined #openstack-infra | 05:29 | |
| *** rpodolyaka1 has joined #openstack-infra | 05:31 | |
| *** rongze has quit IRC | 05:33 | |
| *** sarob has quit IRC | 05:34 | |
| *** sarob has joined #openstack-infra | 05:37 | |
| *** michchap has quit IRC | 05:40 | |
| *** michchap has joined #openstack-infra | 05:40 | |
| clarkb | the failures in http://logs.openstack.org/74/57774/2/gate/gate-tempest-devstack-vm-full/e592961/console.html appear to be the next round of bugs that will need sorting for sanity | 05:40 |
| clarkb | if they aren't in e-r tomorrow I will take a stab at adding them | 05:41 |
| clarkb | I have seen a handful of cases today though | 05:41 |
| *** sarob has quit IRC | 05:41 | |
| portante | me too | 05:41 |
| clarkb | cyeoh: were the API v3 tests failing due to a bug related to the API tests or just falling victim to all of the other bugs | 05:41 |
| portante | is there a bug opened for those SSH protocol errors? | 05:42 |
| clarkb | portante: I am not sure, I just noticed them now | 05:42 |
| clarkb | the volume thing was the one I had seen before | 05:42 |
| * clarkb looks at tempest bugs | 05:42 | |
| cyeoh | clarkb: the v3 api tests were causing 1251920 to occur a lot more than with them disabled | 05:43 |
| clarkb | cyeoh: interesting | 05:43 |
| clarkb | oh the teardownclass volume thing is related to the ssh test failing first, I should look at these more closely | 05:44 |
| cyeoh | clarkb: I'd guess it was just the higher probability of having the m1.small instances running at the same time as the v2 tests were | 05:45 |
| clarkb | portante: I don't see a bug yet, I will submit one really quickly | 05:46 |
| portante | k thx | 05:46 |
| *** jhesketh has quit IRC | 05:47 | |
| *** jhesketh has joined #openstack-infra | 05:47 | |
| *** nati_uen_ has quit IRC | 05:49 | |
| *** mihgen has joined #openstack-infra | 05:50 | |
| clarkb | portante: https://bugs.launchpad.net/tempest/+bug/1253896 with some quick analysis of the bug included | 05:53 |
| uvirtbot | Launchpad bug 1253896 in tempest "test_minimum_basic_scenario fails with SSHException: Error reading SSH protocol banner" [Undecided,New] | 05:53 |
| * portante looks | 05:53 | |
| portante | great, that will be helpful to match it up when we encounter those kinds of failures | 05:54 |
| *** markmcclain has quit IRC | 06:00 | |
| *** melwitt has quit IRC | 06:02 | |
| *** nati_ueno has joined #openstack-infra | 06:10 | |
| *** michchap has quit IRC | 06:19 | |
| *** bingbu has quit IRC | 06:20 | |
| clarkb | thinking about the various failures, would it help to try and run more tools like radon and provide that information in check tests? | 06:24 |
| clarkb | I am not sure how closely tests are examined if they pass and radon wouldn't really be able to give us a sane pass fail data set | 06:24 |
| *** michchap has joined #openstack-infra | 06:26 | |
| portante | for swift, it would be nice to have tools that detect 503s returned by the API, and highlight Tracebacks (though python-keystoneclient middleware emits Tracebacks all the time as part of its debug logging) | 06:27 |
| *** bingbu has joined #openstack-infra | 06:31 | |
| torgomatic | double bonus points for logging the txid from the headers in the 503 response; txids make *excellent* logstash queries | 06:32 |
| *** rongze has joined #openstack-infra | 06:37 | |
| *** matsuhashi has quit IRC | 06:44 | |
| *** nicedice has quit IRC | 06:48 | |
| *** matsuhas_ has joined #openstack-infra | 06:49 | |
| *** nsaje has joined #openstack-infra | 06:50 | |
| *** rpodolyaka1 has quit IRC | 06:51 | |
| *** yolanda has joined #openstack-infra | 06:54 | |
| *** matsuhas_ has quit IRC | 06:56 | |
| *** lchen has quit IRC | 06:57 | |
| *** yolanda has quit IRC | 06:57 | |
| *** lchen has joined #openstack-infra | 06:59 | |
| *** amotoki_zzz is now known as amotoki | 07:00 | |
| *** nati_ueno has quit IRC | 07:01 | |
| *** matsuhashi has joined #openstack-infra | 07:03 | |
| *** nati_ueno has joined #openstack-infra | 07:03 | |
| *** nati_ueno has quit IRC | 07:04 | |
| *** nati_uen_ has joined #openstack-infra | 07:04 | |
| *** sjing has quit IRC | 07:05 | |
| *** sjing has joined #openstack-infra | 07:07 | |
| openstackgerrit | Chmouel Boudjnah proposed a change to openstack-infra/config: Add minimal checks for swift-bench. https://review.openstack.org/57845 | 07:08 |
| *** sdake_ has quit IRC | 07:08 | |
| *** bingbu has quit IRC | 07:17 | |
| *** yolanda has joined #openstack-infra | 07:29 | |
| *** bingbu has joined #openstack-infra | 07:29 | |
| *** nicedice has joined #openstack-infra | 07:37 | |
| *** nati_uen_ has quit IRC | 07:38 | |
| *** mihgen has quit IRC | 07:41 | |
| *** jhesketh has quit IRC | 07:51 | |
| *** nati_ueno has joined #openstack-infra | 08:01 | |
| *** nati_ueno has quit IRC | 08:02 | |
| *** nati_ueno has joined #openstack-infra | 08:02 | |
| *** hdd has joined #openstack-infra | 08:02 | |
| *** amotoki has quit IRC | 08:10 | |
| *** dstanek has quit IRC | 08:13 | |
| jog0 | mikal: no, and cyeoh lets wait another day just to be safe | 08:19 |
| *** yamahata_ has joined #openstack-infra | 08:19 | |
| *** mihgen has joined #openstack-infra | 08:19 | |
| jog0 | portante: if you don't see a big feel free to file one and an elastic-search query here http://git.openstack.org/cgit/openstack-infra/elastic-recheck/tree/queries.yaml | 08:20 |
| jog0 | cyeoh: gate is still a little wobbly wnat to not rock the boat | 08:20 |
| jog0 | clarkb: if you are feeling sleepless I added some more things I want to mention in the email (https://etherpad.openstack.org/p/critical-patches-gatecrash-November-2013) | 08:21 |
| * jog0 goes to bed | 08:21 | |
| *** flaper87|afk is now known as flaper87 | 08:22 | |
| *** nosnos_ has joined #openstack-infra | 08:31 | |
| *** nosnos has quit IRC | 08:35 | |
| *** denis_makogon has joined #openstack-infra | 08:36 | |
| *** lchen has quit IRC | 08:37 | |
| *** lchen has joined #openstack-infra | 08:37 | |
| *** sjing has quit IRC | 08:37 | |
| *** cody-somerville has quit IRC | 08:44 | |
| *** cody-somerville has joined #openstack-infra | 08:45 | |
| *** osanchez has joined #openstack-infra | 08:47 | |
| *** Loquacity has quit IRC | 08:48 | |
| *** yamahata_ has quit IRC | 09:00 | |
| *** Loquacity has joined #openstack-infra | 09:01 | |
| *** jpich has joined #openstack-infra | 09:01 | |
| *** nosnos_ has quit IRC | 09:04 | |
| *** nosnos has joined #openstack-infra | 09:05 | |
| *** mkerrin has joined #openstack-infra | 09:05 | |
| *** fbo_away is now known as fbo | 09:07 | |
| *** nati_ueno has quit IRC | 09:07 | |
| *** dizquierdo has joined #openstack-infra | 09:08 | |
| *** nati_ueno has joined #openstack-infra | 09:09 | |
| *** bingbu has quit IRC | 09:10 | |
| *** nati_ueno has quit IRC | 09:10 | |
| *** derekh has joined #openstack-infra | 09:15 | |
| *** marun has joined #openstack-infra | 09:16 | |
| *** yassine has joined #openstack-infra | 09:16 | |
| openstackgerrit | Roman Prykhodchenko proposed a change to openstack-infra/config: Adds devstack-gate tests for Ironic https://review.openstack.org/53917 | 09:17 |
| *** johnthetubaguy has joined #openstack-infra | 09:18 | |
| *** marun has quit IRC | 09:22 | |
| *** marun has joined #openstack-infra | 09:23 | |
| *** johnthetubaguy1 has joined #openstack-infra | 09:26 | |
| *** johnthetubaguy has quit IRC | 09:26 | |
| *** ljjjustin has quit IRC | 09:29 | |
| *** boris-42 has joined #openstack-infra | 09:29 | |
| *** matsuhashi has quit IRC | 09:29 | |
| *** matsuhashi has joined #openstack-infra | 09:30 | |
| *** dizquierdo has quit IRC | 09:33 | |
| *** boris-42 has quit IRC | 09:34 | |
| *** dstanek has joined #openstack-infra | 09:41 | |
| *** Loquacity has quit IRC | 09:44 | |
| *** Loquacity has joined #openstack-infra | 09:45 | |
| *** dstanek has quit IRC | 09:46 | |
| *** SergeyLukjanov has joined #openstack-infra | 09:47 | |
| openstackgerrit | A change was merged to openstack-dev/pbr: Base python 2.7 skip list on parsed names https://review.openstack.org/54152 | 09:55 |
| *** denis_makogon has quit IRC | 09:59 | |
| openstackgerrit | A change was merged to openstack-infra/devstack-gate: Remove duplicate entries https://review.openstack.org/56920 | 10:03 |
| *** DinaBelova has joined #openstack-infra | 10:08 | |
| *** ace05__ has quit IRC | 10:09 | |
| *** pblaho has joined #openstack-infra | 10:11 | |
| ttx | fungi, mordred: please see https://review.openstack.org/#/c/57441/ -- kinda need those pages up asap | 10:14 |
| *** simonmcc has quit IRC | 10:40 | |
| *** simonmcc has joined #openstack-infra | 10:40 | |
| *** simonmcc has quit IRC | 10:41 | |
| *** simonmcc has joined #openstack-infra | 10:42 | |
| *** DinaBelova has quit IRC | 10:45 | |
| *** marun has quit IRC | 10:45 | |
| *** cody-somerville has quit IRC | 10:46 | |
| *** marun has joined #openstack-infra | 10:47 | |
| *** cody-somerville has joined #openstack-infra | 10:48 | |
| *** pblaho has quit IRC | 10:52 | |
| *** fbo is now known as fbo_away | 10:52 | |
| *** marun has quit IRC | 10:56 | |
| *** marun has joined #openstack-infra | 10:56 | |
| openstackgerrit | Liang Chen proposed a change to openstack-dev/pbr: Make sphinx builders configurable in LocalBuildDoc https://review.openstack.org/57892 | 10:57 |
| *** marun has quit IRC | 10:58 | |
| soren | How about some love for https://review.openstack.org/#/c/55519/ ? | 10:59 |
| *** yaguang has quit IRC | 10:59 | |
| openstackgerrit | Thierry Carrez proposed a change to openstack-infra/config: Add openstack/oslo.rootwrap project https://review.openstack.org/57896 | 11:01 |
| *** cody-somerville has quit IRC | 11:02 | |
| *** nati_ueno has joined #openstack-infra | 11:07 | |
| *** cody-somerville has joined #openstack-infra | 11:07 | |
| *** nati_uen_ has joined #openstack-infra | 11:10 | |
| *** pcm_ has joined #openstack-infra | 11:12 | |
| *** nati_ueno has quit IRC | 11:13 | |
| *** wenlock has quit IRC | 11:14 | |
| *** lcestari has joined #openstack-infra | 11:15 | |
| *** pcm_ has quit IRC | 11:19 | |
| *** pcm_ has joined #openstack-infra | 11:20 | |
| *** rfolco has joined #openstack-infra | 11:21 | |
| *** Loquacity has quit IRC | 11:29 | |
| *** dkliban_ has quit IRC | 11:30 | |
| *** Loquacity has joined #openstack-infra | 11:32 | |
| *** SergeyLukjanov has quit IRC | 11:33 | |
| *** DinaBelova has joined #openstack-infra | 11:36 | |
| *** fbo_away is now known as fbo | 11:42 | |
| *** dstanek has joined #openstack-infra | 11:42 | |
| *** SergeyLukjanov has joined #openstack-infra | 11:45 | |
| *** dstanek has quit IRC | 11:47 | |
| *** nsaje has quit IRC | 11:53 | |
| *** nsaje has joined #openstack-infra | 11:54 | |
| *** cody-somerville has quit IRC | 11:58 | |
| *** nsaje has quit IRC | 11:59 | |
| *** nati_uen_ has quit IRC | 12:05 | |
| *** matsuhashi has quit IRC | 12:09 | |
| *** matsuhashi has joined #openstack-infra | 12:10 | |
| *** DinaBelova has quit IRC | 12:13 | |
| *** matsuhashi has quit IRC | 12:15 | |
| *** xeyed4good has joined #openstack-infra | 12:20 | |
| *** michchap has quit IRC | 12:23 | |
| *** michchap has joined #openstack-infra | 12:23 | |
| *** CaptTofu has quit IRC | 12:28 | |
| *** CaptTofu has joined #openstack-infra | 12:28 | |
| portante | jog0: thanks | 12:29 |
| portante | notmyname: see above ^^^ link yaml thinky | 12:29 |
| *** SergeyLukjanov is now known as _SergeyLukjanov | 12:30 | |
| *** nsaje has joined #openstack-infra | 12:33 | |
| *** lchen has quit IRC | 12:36 | |
| *** nati_ueno has joined #openstack-infra | 12:36 | |
| *** nsaje has quit IRC | 12:38 | |
| *** lchen has joined #openstack-infra | 12:40 | |
| *** xeyed4good has quit IRC | 12:41 | |
| *** DinaBelova has joined #openstack-infra | 12:42 | |
| *** dolphm has joined #openstack-infra | 12:44 | |
| *** michchap has quit IRC | 12:45 | |
| *** michchap has joined #openstack-infra | 12:46 | |
| *** marun has joined #openstack-infra | 13:00 | |
| *** CaptTofu has quit IRC | 13:02 | |
| *** nati_ueno has quit IRC | 13:02 | |
| *** nati_ueno has joined #openstack-infra | 13:03 | |
| *** CaptTofu has joined #openstack-infra | 13:03 | |
| *** dkliban_ has joined #openstack-infra | 13:07 | |
| *** nati_ueno has quit IRC | 13:09 | |
| *** DinaBelova has quit IRC | 13:11 | |
| *** SergeyLukjanov has joined #openstack-infra | 13:12 | |
| *** DinaBelova has joined #openstack-infra | 13:14 | |
| *** dcramer_ has quit IRC | 13:17 | |
| *** dizquierdo has joined #openstack-infra | 13:21 | |
| dolphm | this is probably a dumb question- does pbr behave differently if you do setup.py develop vs setup.py install? | 13:24 |
| dolphm | or setuptools | 13:24 |
| dolphm | other than where things happen | 13:24 |
| *** zul has quit IRC | 13:27 | |
| *** oubiwann has joined #openstack-infra | 13:27 | |
| *** zul has joined #openstack-infra | 13:27 | |
| *** dkranz has joined #openstack-infra | 13:29 | |
| *** basic` has quit IRC | 13:29 | |
| *** pblaho has joined #openstack-infra | 13:33 | |
| *** basic` has joined #openstack-infra | 13:34 | |
| *** basic` has quit IRC | 13:35 | |
| *** basic` has joined #openstack-infra | 13:35 | |
| dolphm | i'm running into the same issue described here- https://bitbucket.org/pypa/setuptools/issue/73/ | 13:35 |
| *** nosnos has quit IRC | 13:38 | |
| *** DinaBelova has quit IRC | 13:40 | |
| *** xeyed4good has joined #openstack-infra | 13:42 | |
| *** thomasem has joined #openstack-infra | 13:42 | |
| dolphm | bah, i give up. pip install -e works | 13:44 |
| *** SergeyLukjanov has quit IRC | 13:45 | |
| *** julim has joined #openstack-infra | 13:47 | |
| openstackgerrit | Jaroslav Henner proposed a change to openstack-infra/jenkins-job-builder: Add properties testing. https://review.openstack.org/57654 | 13:49 |
| *** DinaBelova has joined #openstack-infra | 13:49 | |
| *** ArxCruz has quit IRC | 13:49 | |
| *** ArxCruz has joined #openstack-infra | 13:51 | |
| openstackgerrit | Jaroslav Henner proposed a change to openstack-infra/jenkins-job-builder: Add batch_tasks support. https://review.openstack.org/57469 | 13:51 |
| openstackgerrit | Jaroslav Henner proposed a change to openstack-infra/jenkins-job-builder: Add seealso to batch_tasks from promoted_build. https://review.openstack.org/57473 | 13:51 |
| *** CaptTofu has quit IRC | 13:53 | |
| *** CaptTofu has joined #openstack-infra | 13:53 | |
| *** bpokorny has joined #openstack-infra | 13:53 | |
| *** ArxCruz has quit IRC | 13:54 | |
| *** ArxCruz has joined #openstack-infra | 13:54 | |
| *** dprince has joined #openstack-infra | 13:55 | |
| anteaya | 1 in the gate - it appears we have caught up | 13:55 |
| anteaya | w00t! | 13:55 |
| anteaya | yay team | 13:55 |
| *** mihgen has quit IRC | 13:56 | |
| *** marun has quit IRC | 13:57 | |
| *** marun has joined #openstack-infra | 13:58 | |
| *** michchap has quit IRC | 13:58 | |
| *** michchap has joined #openstack-infra | 14:01 | |
| *** sandywalsh has quit IRC | 14:01 | |
| *** mihgen has joined #openstack-infra | 14:09 | |
| *** alcabrera has joined #openstack-infra | 14:09 | |
| *** ryanpetrello has joined #openstack-infra | 14:10 | |
| *** mfer has joined #openstack-infra | 14:11 | |
| *** SergeyLukjanov has joined #openstack-infra | 14:11 | |
| *** boris-42 has joined #openstack-infra | 14:13 | |
| *** julim has quit IRC | 14:13 | |
| *** michchap has quit IRC | 14:14 | |
| *** michchap has joined #openstack-infra | 14:14 | |
| *** rnirmal has joined #openstack-infra | 14:14 | |
| *** sandywalsh has joined #openstack-infra | 14:17 | |
| *** mriedem has joined #openstack-infra | 14:17 | |
| *** julim has joined #openstack-infra | 14:20 | |
| *** nicedice has quit IRC | 14:21 | |
| *** nicedice has joined #openstack-infra | 14:21 | |
| *** yaguang has joined #openstack-infra | 14:23 | |
| dkehn | clarkb: curious for nodes that run the testing (e.g. py27) how much memory to they have? just curious, for me on my servers to perform test what took 4GB now won't stay out of swap space, had to bump to 8GB | 14:28 |
| dkehn | clarkb: neutron related | 14:28 |
| *** ArxCruz has quit IRC | 14:30 | |
| *** herndon has joined #openstack-infra | 14:31 | |
| *** boris-42 has quit IRC | 14:32 | |
| *** boris-42_ has joined #openstack-infra | 14:32 | |
| *** eharney has joined #openstack-infra | 14:33 | |
| *** cody-somerville has joined #openstack-infra | 14:34 | |
| *** cody-somerville has quit IRC | 14:34 | |
| *** cody-somerville has joined #openstack-infra | 14:34 | |
| *** xeyed4good has left #openstack-infra | 14:34 | |
| fungi | ttx: looking | 14:34 |
| openstackgerrit | A change was merged to openstack-infra/config: Track icehouse development in releasestatus https://review.openstack.org/57441 | 14:37 |
| fungi | dkehn: we use 8gb flavors for those | 14:39 |
| dkehn | fungi: thx, yepper that'll do it | 14:39 |
| soren | Could someone throw some love at https://review.openstack.org/#/c/55519/ , please? | 14:39 |
| *** zoresvit has quit IRC | 14:41 | |
| anteaya | coming back again to my desire to have a conversation about testing resources | 14:44 |
| anteaya | so that developers know in advance what is required | 14:44 |
| anteaya | and we have a sense as a group what those resources are, so they don't keep ballooning | 14:44 |
| *** bnemec is now known as beekneemech | 14:53 | |
| *** dansmith is now known as damnsmith | 14:54 | |
| *** SergeyLukjanov has quit IRC | 14:55 | |
| *** dstanek has joined #openstack-infra | 14:55 | |
| *** boris-42_ has quit IRC | 14:58 | |
| *** russellb is now known as rustlebee | 14:59 | |
| *** dkliban_ is now known as dkliban | 14:59 | |
| *** dizquierdo has quit IRC | 15:00 | |
| *** ArxCruz has joined #openstack-infra | 15:00 | |
| *** dizquierdo has joined #openstack-infra | 15:01 | |
| *** alaski is now known as lascii | 15:02 | |
| notmyname | portante: I didn't see yaml | 15:03 |
| *** pblaho has quit IRC | 15:06 | |
| *** SergeyLukjanov has joined #openstack-infra | 15:07 | |
| *** markmcclain has joined #openstack-infra | 15:08 | |
| *** marun has quit IRC | 15:08 | |
| *** marun has joined #openstack-infra | 15:09 | |
| *** dcramer_ has joined #openstack-infra | 15:15 | |
| *** ruhe has joined #openstack-infra | 15:18 | |
| dstanek | i think my review got stuck; review.openstack.org/54210 says gate jobs started a few days ago | 15:21 |
| dstanek | is there anything i can do to kick it? | 15:21 |
| fungi | dstanek: it probably got dumped the first time we cleared the gate to try and push through fixes for the various nondeterministic bugs, and we didn't requeue those as far as i know. just leave a review comment on it which says "reverify no bug" and it should get added into the gate | 15:26 |
| dstanek | fungi: great thanks | 15:26 |
| *** marun has quit IRC | 15:29 | |
| dkranz | fungi: I followed a link in a recent build to http://logs.openstack.org/periodic/periodic-tempest-devstack-vm-stress/302442d/ but get Not Found | 15:30 |
| dkranz | fungi: Is this supposed to work? | 15:30 |
| openstackgerrit | Sergey Lukjanov proposed a change to openstack-infra/config: Enable #storyboard IRC channel logging https://review.openstack.org/57952 | 15:30 |
| dkranz | fungi: The link was from https://jenkins02.openstack.org/job/periodic-tempest-devstack-vm-stress/50/console | 15:30 |
| openstackgerrit | Alex Gaynor proposed a change to openstack-infra/config: Run the cinderclient tests under PyPy in gate and check https://review.openstack.org/57953 | 15:31 |
| *** rcleere has joined #openstack-infra | 15:31 | |
| fungi | dkranz: i suspect those links are being built incorrectly. may not have been updated since we started adding an extra directory level | 15:31 |
| * fungi tries something | 15:31 | |
| *** pblaho has joined #openstack-infra | 15:32 | |
| fungi | dkranz: seems it ended up in http://logs.openstack.org/periodic/periodic-tempest-devstack-vm-stress/devstack-precise-hpcloud-az2-663888/50/ | 15:35 |
| fungi | i suspect we didn't really mean to be using the slave name for that path component | 15:35 |
| *** mrodden has joined #openstack-infra | 15:35 | |
| fungi | i'll add to my queue trying to figure out why it's doing that | 15:36 |
| dkranz | fungi: ok thanks. Is there a general rule I can follow to do what you did to find it? | 15:36 |
| *** herndon has quit IRC | 15:36 | |
| dkranz | fungi: Until the issue is fixed? | 15:36 |
| *** wenlock has joined #openstack-infra | 15:36 | |
| fungi | dkranz: note the name of the slave where the job ran, mentioned in the top-right corner on the status page for that job... https://jenkins02.openstack.org/job/periodic-tempest-devstack-vm-stress/50/ | 15:37 |
| *** mrodden1 has quit IRC | 15:37 | |
| dkranz | fungi: ok, thanks. GOt it. | 15:37 |
| fungi | and then the /50 comes from the run number | 15:37 |
| *** datsun180b has joined #openstack-infra | 15:38 | |
| *** senk has joined #openstack-infra | 15:38 | |
| *** ArxCruz has quit IRC | 15:40 | |
| *** ArxCruz has joined #openstack-infra | 15:43 | |
| *** rongze has quit IRC | 15:45 | |
| openstackgerrit | Jeremy Stanley proposed a change to openstack-infra/config: Alphabetize projects.yaml again https://review.openstack.org/57959 | 15:48 |
| openstackgerrit | A change was merged to openstack-infra/os-loganalyze: Support Setting The Path Using an ENV Variable https://review.openstack.org/57783 | 15:57 |
| *** wenlock_ has joined #openstack-infra | 15:57 | |
| *** dstanek has quit IRC | 15:58 | |
| *** ruhe has quit IRC | 16:00 | |
| anteaya | dprince: this page http://smokestack.openstack.org/ 404s | 16:02 |
| dprince | anteaya: https://smokestack.openstack.org/ | 16:02 |
| dprince | ssl, albeit an unsigned one... | 16:03 |
| *** ericw has joined #openstack-infra | 16:04 | |
| *** ericw is now known as ewindisch | 16:04 | |
| *** rongze has joined #openstack-infra | 16:04 | |
| *** danger_fo_away is now known as dangers | 16:04 | |
| *** SergeyLukjanov has quit IRC | 16:05 | |
| *** UtahDave has joined #openstack-infra | 16:06 | |
| *** atiwari has joined #openstack-infra | 16:08 | |
| anteaya | dprince: what do you want to do about the link using http:// on http://ci.openstack.org/third_party.html ? | 16:10 |
| fungi | that can be updated in openstack-infra/config doc/source/third_party.rst i believe | 16:11 |
| anteaya | I ask because http://lists.openstack.org/pipermail/openstack-dev/2013-November/019219.html | 16:11 |
| anteaya | which requires all neutron plugins to have third party testing by i-2 | 16:11 |
| anteaya | so there are many new folks popping up asking how to imipliment | 16:12 |
| dprince | anteaya: I can fix the link I guess. I thought I had them all fixed... | 16:12 |
| anteaya | as the first and best example of third party testing, I am pointing them to your code dprince | 16:13 |
| dprince | anteaya: and they will likely hate me because I did things w/ Ruby. But that is fine. | 16:13 |
| *** DinaBelova has quit IRC | 16:13 | |
| anteaya | but the broken link is a trip hazzard | 16:14 |
| openstackgerrit | Dan Prince proposed a change to openstack-infra/config: Fix SmokeStack link to be the ssl version. https://review.openstack.org/57964 | 16:14 |
| dprince | anteaya: ^^ | 16:14 |
| *** ewindisch has quit IRC | 16:15 | |
| *** ewindisch has joined #openstack-infra | 16:15 | |
| mordred | morning all | 16:15 |
| krtaylor | morning mordred | 16:16 |
| dprince | mordred: late start for an east coaster now hey | 16:17 |
| mordred | dprince: heh. well, it's an early start when you're on the west coast | 16:17 |
| *** misterpink has joined #openstack-infra | 16:18 | |
| dprince | mordred: well then, I see you adjust nicely | 16:18 |
| *** senk has quit IRC | 16:29 | |
| *** wenlock has quit IRC | 16:29 | |
| *** misterpink has quit IRC | 16:30 | |
| *** datsun180b_ has joined #openstack-infra | 16:30 | |
| *** misterpink has joined #openstack-infra | 16:30 | |
| *** wenlock has joined #openstack-infra | 16:32 | |
| *** rongze_ has joined #openstack-infra | 16:32 | |
| *** rongze has quit IRC | 16:32 | |
| *** datsun180b_ has quit IRC | 16:33 | |
| *** datsun180b_ has joined #openstack-infra | 16:33 | |
| *** datsun180b has quit IRC | 16:34 | |
| *** datsun180b_ is now known as datsun180b | 16:34 | |
| * fungi disappears to get some lunch... bbiab | 16:39 | |
| zul | does pbr have anything to do with sphinx? | 16:40 |
| *** mihgen has quit IRC | 16:40 | |
| *** yassine has quit IRC | 16:41 | |
| *** fifieldt has quit IRC | 16:42 | |
| *** yaguang has quit IRC | 16:44 | |
| *** chandankumar has left #openstack-infra | 16:44 | |
| *** chandankumar has joined #openstack-infra | 16:44 | |
| chandankumar | Hello | 16:45 |
| chandankumar | i have to edit the wiki page of stackalytics | 16:46 |
| chandankumar | for that i am creating the account on openstack wiki page | 16:46 |
| *** reed has joined #openstack-infra | 16:46 | |
| chandankumar | for that i requires secret word, as word (secret) i am entering isnot acceptable while creating the account. | 16:47 |
| chandankumar | what is that secret word? | 16:47 |
| *** DinaBelova has joined #openstack-infra | 16:53 | |
| jeblair | zul: i think it does touch it a little; i think it has an option to auto-generate api docs | 16:53 |
| zul | jeblair: because im seeing newer versions of sphinx fail on kestone and noa | 16:54 |
| mordred | zul: there is a newer version of sphinx that is broken | 16:57 |
| mordred | zul: which is why we have it pinned, iirc | 16:57 |
| zul | mordred: yeah yeah ;) | 16:57 |
| mordred | zul: I do not believe the breakage is pbr related | 16:58 |
| mordred | however - it's entirely possible that something new has arisen | 16:58 |
| *** Hefeweizen has quit IRC | 16:58 | |
| *** jgriffith has quit IRC | 16:59 | |
| *** jgriffith has joined #openstack-infra | 16:59 | |
| *** Hefeweizen has joined #openstack-infra | 16:59 | |
| *** dkehn has quit IRC | 16:59 | |
| *** derekh has quit IRC | 17:00 | |
| *** dkehn has joined #openstack-infra | 17:00 | |
| openstackgerrit | Cyril Roelandt proposed a change to openstack/requirements: Add sphinxcontrib-docbookrestapi https://review.openstack.org/57978 | 17:00 |
| *** ewindisch has quit IRC | 17:02 | |
| *** julim has quit IRC | 17:05 | |
| openstackgerrit | A change was merged to openstack/requirements: Update python-troveclient version https://review.openstack.org/52137 | 17:05 |
| *** julim has joined #openstack-infra | 17:06 | |
| *** rfolco has quit IRC | 17:07 | |
| *** herndon has joined #openstack-infra | 17:07 | |
| *** med_ has quit IRC | 17:11 | |
| *** rpodolyaka has quit IRC | 17:12 | |
| *** ewindisch has joined #openstack-infra | 17:16 | |
| *** atiwari has quit IRC | 17:16 | |
| portante | notmyname: http://git.openstack.org/cgit/openstack-infra/elastic-recheck/tree/queries.yaml | 17:20 |
| *** jpich has quit IRC | 17:20 | |
| *** pblaho has quit IRC | 17:20 | |
| *** mrodden has quit IRC | 17:23 | |
| *** hogepodge has joined #openstack-infra | 17:25 | |
| openstackgerrit | A change was merged to openstack-infra/zuul: test: support voluptuous 7.0+ https://review.openstack.org/54696 | 17:29 |
| zaro | fungi: can we reverify on a change that has been approved but not merged? | 17:30 |
| mgagne | zaro: reverify no bug ? | 17:31 |
| mgagne | zaro: unless you have a bug associated with it | 17:32 |
| zaro | mgagne: yes, https://review.openstack.org/#/c/56397 | 17:32 |
| mgagne | zaro: make sure there is no dependencies | 17:32 |
| mgagne | zaro: cloverphp publisher is a dependency | 17:32 |
| mgagne | zaro: it needs to be approved for this one to get merged | 17:32 |
| zaro | mgagne: ahh forgot. i see why it didn't go thru now. | 17:33 |
| zaro | mgagne: would still like to get this one merged if you can take another look, https://review.openstack.org/#/c/56715 | 17:33 |
| openstackgerrit | Valeriy Ponomaryov proposed a change to openstack-infra/config: Add gate-manila-devstack job https://review.openstack.org/57987 | 17:34 |
| mgagne | zaro: I was able to build a python project on my windows slave. Jenkins looks to work fine on Windows 2012, just a PITA to setup... | 17:34 |
| zaro | mgagne: what type of connection? | 17:35 |
| jeblair | clarkb: bug in https://review.openstack.org/#/c/52689/ :( | 17:35 |
| mgagne | zaro: using javaws to start the slave. it's using JNLP. again, it's a PITA with a firewalled setup, you have to assign a static port for JNLP on the master. | 17:35 |
| *** mrodden has joined #openstack-infra | 17:36 | |
| pabelanger | Yay, voluptuous patch landed! | 17:36 |
| *** johnthetubaguy1 has quit IRC | 17:36 | |
| zaro | mgagne: yeah. don't like JNLP either. did you try ssh? | 17:37 |
| *** markwash has quit IRC | 17:37 | |
| *** sarob has joined #openstack-infra | 17:39 | |
| mgagne | zaro: no, this would require further more setup on the windows image and I wanted it to work asap. But it would probably worth it, our git repositories require ssh keys anyway. | 17:39 |
| mgagne | zaro: I just don't know yet how I'm gonna puppetize that thing. Installing packages/dependencies on Windows is such a pain... | 17:40 |
| zaro | mgagne: yeah. don't like JNLP either. did you try ssh? | 17:47 |
| mgagne | zaro: no, have any documentation I can read? | 17:48 |
| zaro | mgagne: opps, mean how about custom image with ssh? | 17:48 |
| *** ^d has joined #openstack-infra | 17:48 | |
| *** ^d has joined #openstack-infra | 17:48 | |
| *** misterpink has quit IRC | 17:48 | |
| *** dizquierdo has quit IRC | 17:51 | |
| openstackgerrit | Khai Do proposed a change to openstack-infra/jenkins-job-builder: update doc and add new JJB unit tests https://review.openstack.org/56715 | 17:51 |
| mgagne | zaro: here is the trick: the build I'm doing (on this windows slave) will support builds of Windows images I'm working on (which for various reasons shouldn't have ssh installed). I'm actually using a half-baked Windows image to setup my jenkins slave and then use it to build tools used to improve my windows images. Long story short: I'm dog fooding my windows images. | 17:51 |
| *** datsun180b_ has joined #openstack-infra | 17:52 | |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/zuul: Raise an exception on an invalid reporter https://review.openstack.org/57996 | 17:53 |
| mgagne | zaro: One thing I could do is add support for Windows in DIB so I could add tools to my base image and then use it for my slaves. | 17:53 |
| *** hogepodge has quit IRC | 17:54 | |
| *** datsun180b has quit IRC | 17:55 | |
| *** datsun180b_ is now known as datsun180b | 17:55 | |
| notmyname | jog0: any response to clayg on https://review.openstack.org/#/c/57334/ ? if that can't be merged, then we'll need to merge https://review.openstack.org/#/c/56712/ ins wift | 17:56 |
| *** che-arne has quit IRC | 17:57 | |
| openstackgerrit | A change was merged to openstack-infra/zuul: register reporters on test config https://review.openstack.org/55184 | 17:57 |
| *** rongze has joined #openstack-infra | 17:57 | |
| jeblair | notmyname: a quick glance at the comment suggests that clayg may be under the impression that "except Exception:" is not allowed by the rule, but in reality it is | 17:59 |
| jeblair | notmyname: dhellmann made some fairly compelling arguments that 'except:' is almost never what you want and 'except Exception:' is the best way to "catch 'em all" | 18:00 |
| *** hogepodge has joined #openstack-infra | 18:00 | |
| notmyname | jeblair: the point is that we specifically do not want a requirement for "except Exception". we want to be able to have bare excepts | 18:00 |
| jeblair | notmyname: (an approach i am also strongly in favor of) | 18:00 |
| notmyname | right, and I agree in general. but there are exceptions | 18:00 |
| *** sarob has quit IRC | 18:01 | |
| *** rongze_ has quit IRC | 18:01 | |
| jeblair | notmyname: okay -- so i fist just wanted to make sure that the discussion wasn't 'except something:' versus 'except:', so that's cool. | 18:01 |
| notmyname | jeblair: and I'd prefer to have a mechanism to say "no this is really what I mean" rather than turning the check off | 18:01 |
| jeblair | ugh, so we talked about this at the summit: https://etherpad.openstack.org/p/icehouse-oslo-hacking-updates | 18:02 |
| notmyname | jeblair: we currently have 2 patches that will fail flake 8 because they each have "except:" (as they should). similarly, we had a period of time during havana when we weren't really deployable for a while because a bare except got changed to "except Exception" | 18:02 |
| jeblair | and unfortunately, the only thing that got recorded in the etherpad was that everyone in the room seemed convinced that "except Exception" is the way to go, even though many of us wanted to remove the check | 18:03 |
| notmyname | jeblair: end result being that either the flake8 patch needs to land, or we'll turn off bare except checking in swift | 18:03 |
| *** ^d has quit IRC | 18:03 | |
| hub_cap | wasnt there a wiki article for setting up your own nodepool (so i can test the image cache changes) ? | 18:03 |
| *** ^d has joined #openstack-infra | 18:03 | |
| *** ^d has quit IRC | 18:03 | |
| jeblair | and while everyone thought that the docs should be updated to include more _reasoning_ about these things, the reasoning about "except Exception" stated at the time wasn't recorded | 18:03 |
| *** ^d has joined #openstack-infra | 18:03 | |
| *** SergeyLukjanov has joined #openstack-infra | 18:04 | |
| clarkb | jeblair: :( | 18:04 |
| clarkb | also good morning (really late start for me today :( ) | 18:04 |
| notmyname | jeblair: nobody is arguing that for the vast majority of the time you don't want to have bare excepts. this is similar to having "reverify no bug" ;-) | 18:04 |
| notmyname | bad idea most of the time, but sometimes needed | 18:04 |
| jeblair | and i only remember it had something to do with the fact that _very low level_ exceptions are the only ones that don't derive from Exception, and they are ones that you never really want to catch | 18:04 |
| notmyname | jeblair: eventlet timeouts do not derive from exception | 18:05 |
| jeblair | notmyname: !!? | 18:05 |
| notmyname | for good reason (and I'm curious how other projects handle that) | 18:05 |
| notmyname | eventlet.Timeout derives from BaseException | 18:05 |
| notmyname | so eg in Swift we have a lot of "except (Exception, Timeout):" | 18:06 |
| notmyname | which is good | 18:07 |
| jeblair | notmyname: i'm not in a position to argue the point, so i don't want to waste any more of your time (i am not opposed to the noqa flag certainly); you should probably talk to jog0 and dhellmann | 18:07 |
| notmyname | jeblair: ok :-) | 18:07 |
| notmyname | and I'm not trying to argue anything either. I'd prefer to see noqa for bare excepts supported, but there are (less optimal) ways to do the same thing | 18:08 |
| notmyname | jog0: dhellmann: you have been summoned | 18:09 |
| *** UtahDave has quit IRC | 18:09 | |
| jeblair | notmyname: oh sorry, just to set the tone of my last remark -- i meant it more in the sense of "i don't have all the facts at my disposal to have a productive conversation"; however, i am very interested in the conversation as I think i have some things to learn | 18:10 |
| notmyname | jeblair: no worries :-) | 18:10 |
| reed | fungi, thanks signing my key | 18:14 |
| clarkb | jeblair: for the bug in 52689, would caching the git config values in merger.Repo and checking if they have changed then updating them work? | 18:14 |
| clarkb | jeblair: if I understand correctly we just need to be able to update those values if they change in the config | 18:15 |
| reed | fungi, too bad your messages only confirmed the incredibly very sad state of encryption tools :( | 18:15 |
| jeblair | clarkb: yes, but they also need to update on a fresh start (if you shut down zuul, reconfig, then start), so you might need to handle that | 18:15 |
| jeblair | clarkb: basically, that's why there was an _initialized flag | 18:15 |
| *** sarob has joined #openstack-infra | 18:16 | |
| *** SergeyLukjanov has quit IRC | 18:17 | |
| *** melwitt has joined #openstack-infra | 18:18 | |
| *** markwash has joined #openstack-infra | 18:18 | |
| *** fungi has left #openstack-infra | 18:19 | |
| *** pafuent has joined #openstack-infra | 18:19 | |
| *** mrmartin has joined #openstack-infra | 18:21 | |
| *** sarob has quit IRC | 18:22 | |
| clarkb | jeblair: was this bug preexisting? I don't see how merger.email and merger._initialized could be updated after the initial clone with the old code | 18:22 |
| clarkb | jeblair: except for on a zuul restart | 18:22 |
| clarkb | is that the cause that you are talking about? | 18:22 |
| clarkb | * case | 18:23 |
| *** david-lyle is now known as david-lyle_afk | 18:24 | |
| jeblair | clarkb: right -- before zuul never expected to reclone after it started, so it's correct that case wasn't handled (it couldn't be). but it could handle a config change across a restart | 18:24 |
| clarkb | gotcha | 18:24 |
| *** DinaBelova has quit IRC | 18:25 | |
| *** sarob has joined #openstack-infra | 18:29 | |
| *** flaper87 is now known as flaper87|afk | 18:30 | |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/zuul: Add the possibility to pass the zuul url https://review.openstack.org/54196 | 18:31 |
| *** SergeyLukjanov has joined #openstack-infra | 18:31 | |
| *** chuck__ has joined #openstack-infra | 18:31 | |
| *** sarob_ has joined #openstack-infra | 18:32 | |
| clarkb | jeblair: I think I have a simple fix | 18:32 |
| clarkb | running tox over it now | 18:32 |
| *** sarob has quit IRC | 18:32 | |
| reed | jeblair, mordred: see my email to the i18n team | 18:35 |
| pafuent | Hi, I submitted my first patch to Openstack, and I'm not sure if Jenkins must run the merge with the actual status of the review (I got an Approved). Someone could check the status of the review and tell me if I need to do something else? (https://review.openstack.org/#/c/57001/) | 18:35 |
| reed | not urgent | 18:36 |
| reed | pafuent, welcome! :) | 18:36 |
| pafuent | reed: Thanks | 18:36 |
| mrodden | so this seems interesting http://logs.openstack.org/74/57774/2/gate/gate-tempest-devstack-vm-full/e592961/testr_results.html.gz | 18:37 |
| jeblair | reed: i'm very opposed to us paying for 'enterprise features' of transifex | 18:37 |
| jeblair | reed: i never thought that was what we decided at hk | 18:37 |
| mrodden | tempest.scenario.test_minimum_basic.TestMinimumBasicScenario failed with SSHException: Error reading SSH protocol banner | 18:37 |
| *** fungi has joined #openstack-infra | 18:37 | |
| *** sarob_ has quit IRC | 18:38 | |
| *** hogepodge has quit IRC | 18:39 | |
| *** osanchez has quit IRC | 18:39 | |
| *** sarob has joined #openstack-infra | 18:39 | |
| jeblair | reed: i even double checked https://etherpad.openstack.org/p/icehouse-summit-translation-tools-and-scripts | 18:39 |
| jeblair | reed: "ask for a quote for enterprise services" was not an action item | 18:39 |
| *** rfolco has joined #openstack-infra | 18:40 | |
| clarkb | pafuent: it has not merged, under status it says review in progress which will switch to merged when it does merge | 18:41 |
| *** sarob has quit IRC | 18:41 | |
| *** Ryan_Lane has quit IRC | 18:41 | |
| *** SergeyLukjanov has quit IRC | 18:41 | |
| clarkb | pafuent: I think that change probably got lost in the great gate upheaval of november 2013 | 18:41 |
| *** sarob has joined #openstack-infra | 18:42 | |
| clarkb | pafuent: the current fix to that problem is 'reverify no bug' but reverify may(did we make a decision on this yet) be going away in the near future | 18:42 |
| *** rongze has quit IRC | 18:42 | |
| mrodden | clarkb: all of reverify or just no bug? | 18:43 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/zuul: GC git.Repo objects in merger.Repo operations. https://review.openstack.org/55943 | 18:43 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/zuul: Clone repo if it is not on disk https://review.openstack.org/52689 | 18:43 |
| clarkb | mrodden: I would like all of reverify to go away, but I think I may lose that argument :) | 18:43 |
| clarkb | jeblair: ^ fixes for zuul | 18:43 |
| mrodden | what would be used instead | 18:43 |
| harlowja | jd__ yt | 18:43 |
| clarkb | mrodden: core reviewers can leave a new comment with +1 approved | 18:43 |
| pafuent | clarkb: So, I need to add a comment with 'reverify no bug'? | 18:44 |
| mrodden | ok | 18:44 |
| clarkb | pafuent: yes | 18:44 |
| pafuent | clarkb: Ok, thanks so much | 18:44 |
| fungi | clarkb: pafuent: yes, looks like zuul probably did not see that approval. it was likely timed while zuul was down or otherwise not watching the gerrit event stream while we were working through gate issues. i al;ready added a reverify comment on it a few minutes ago and it's in the gate now | 18:44 |
| clarkb | fungi: cool thanks | 18:44 |
| SpamapS | does this need to be re-queued somehow? https://review.openstack.org/#/c/51685/ ? | 18:44 |
| SpamapS | I don't see it listed on zuul.openstack.org | 18:44 |
| * jd__ stares at harlowja | 18:45 | |
| clarkb | SpamapS: I think you have the same problem as pafuent, give it a 'reverify no bug' | 18:45 |
| SpamapS | clarkb: ty | 18:45 |
| pafuent | clarkb: Thanks again | 18:45 |
| harlowja | jd__ i and a few others were wondering about https://etherpad.openstack.org/p/icehouse-oslo-service-synchronization (i think u did this session) anything we can help out with there? | 18:45 |
| fungi | clarkb: SpamapS: though that one does look like it had started gating back on wednesday, so it was probably actually in the gate when we dumped it | 18:46 |
| jd__ | harlowja: we need https://review.openstack.org/#/c/56927/ approved so Yassine can start sending patches | 18:46 |
| harlowja | interesting, tooz | 18:47 |
| harlowja | that will be the name, neat, haha | 18:47 |
| jd__ | yeah :-) | 18:47 |
| harlowja | well it has my +1 now to ;) | 18:47 |
| harlowja | wonder if someone from here can approve it ;) | 18:48 |
| jd__ | maybe we can stare at someone | 18:48 |
| fungi | reed: i assume you refer to the need to unfortunate need to import and publish the signature? note that uses a trick of cryptography... i can encrypt the signature for each uid and send them independently to their respective addresses, but only if you both have control of the decryption key and control of that e-mail address will it be possible for you to decrypt and publish the sig on a given uid | 18:49 |
| harlowja | jd__ what were your thoughts on having a locking api in that also, still service 'liveness' releated | 18:49 |
| clarkb | O_O standalone python libvirt on pypi for version 0.9.11 is possible | 18:49 |
| clarkb | now I really want to get cloud archive going again | 18:49 |
| harlowja | jd__ https://wiki.openstack.org/wiki/StructuredWorkflowLocks (stares at that, haha) | 18:49 |
| clarkb | jd__: what is the status on using ubuntu cloud archive on the test slaves? | 18:49 |
| jd__ | harlowja: I think it's a good idea, that could be another featureset of a driver -- in the case some driver don't support that part, and vice-versa | 18:50 |
| reed | fungi, not that, it's the fact that Thunderbird Enigmail fails to see the attachment... I still haven't found time to import them :( | 18:50 |
| fungi | reed: really? weird | 18:50 |
| harlowja | jd__ sure, taskflow itself would love such a library :) | 18:50 |
| jd__ | harlowja: I read -- I can't stare the wiki for too long, it always wins, it scares me. | 18:50 |
| clarkb | fungi: jeblair: mordred: until https://review.openstack.org/#/c/57807/ (or something like it) puppet won't run on the precise slaves | 18:50 |
| reed | jeblair, the Enterprise quote was implicit in the needs of the i18n team | 18:50 |
| jeblair | reed: no it wasn't. | 18:51 |
| clarkb | it runs ok in the d-g image build because those image builds don't use server.pp as a base | 18:51 |
| jeblair | reed: "Talk to Transifex, let them know how we feel about their change of course" | 18:51 |
| harlowja | jd__ :) scary wikis | 18:51 |
| jeblair | reed: that was the action item | 18:51 |
| jeblair | reed: nothing about quotes or extending our use of transifex | 18:51 |
| reed | jeblair, that was done | 18:51 |
| jog0 | notmyname: sorry haven't had a chance to review it | 18:51 |
| reed | there were also other questions for them in the discussions | 18:51 |
| clarkb | fungi: jeblair: an alternative to my proposed fix would be to rework where we make the rsyslog changes for d-g slaves | 18:51 |
| *** wenlock has quit IRC | 18:52 | |
| notmyname | jog0: ok. I know there's been a lot going on. just wanted to alert you on it again :-) | 18:52 |
| reed | jeblair, there are needs from the TX team stated | 18:52 |
| jd__ | clarkb: there's a patch to make them use cloud archive there https://review.openstack.org/#/c/48226/ | 18:52 |
| jeblair | reed: we decided to move from transifex to another platform, and to work with the communities of the other platforms on those needs | 18:52 |
| harlowja | jd__ well let me know if u need any help (taskflow peoples are in #openstack-state-management) - defintely interseted in helping this | 18:52 |
| jd__ | clarkb: I left a comment with the status (from what I know) today | 18:52 |
| *** Ryan_Lane has joined #openstack-infra | 18:53 | |
| reed | jeblair, the i18n team doesn't agree that that was the decision... the decision for them was that for the time being we stay on TX | 18:53 |
| jog0 | notmyname: the short of it, is I am fine with adding a noqa to almost anything | 18:53 |
| jeblair | reed: says who? | 18:53 |
| notmyname | jog0: great! merge it! ;-) | 18:53 |
| jog0 | I am blindly trusting you on there is a good reason for this | 18:53 |
| reed | jeblair, says Daisy, the team lead | 18:53 |
| jeblair | reed: where? | 18:53 |
| jog0 | notmyname: but I am a stickler for commit messages :) | 18:53 |
| clarkb | jd__: thanks, looking at that now | 18:53 |
| jd__ | harlowja: awesome, yassine's working on it right now with some guidance, and he'll push things as soon as repo is up -- I'll tell you when the show begins | 18:53 |
| reed | jeblair, not on a mlist :) | 18:53 |
| jeblair | reed: the decision is to stay on transifex while we work on moving to another platform carefully | 18:53 |
| jog0 | notmyname: oh and testing is magic docstring tests | 18:53 |
| jeblair | reed: then it didn't happen. :) | 18:53 |
| notmyname | jog0: well, it's not too blind there are references to current requirements for it :-) | 18:53 |
| reed | jeblair, there, you said that too | 18:54 |
| clarkb | jd__: lgtm hoping jeblair and fungi will take a look too https://review.openstack.org/#/c/48226/ | 18:54 |
| jog0 | notmyname: I didn't say everyone is blindly trusting you, just that I am. | 18:54 |
| reed | so, while we stay on TX, the i18n team needs to function | 18:54 |
| jeblair | reed: yeah, 'stay on platform while we work on the move' doesn't mean 'buy new enterprise features' | 18:54 |
| jog0 | as I have no reason to doubt you on this | 18:54 |
| jog0 | and I am *not* a python expert like dhellmann or others | 18:54 |
| jeblair | reed: if tx is already less functional than pootle, perhaps we should go ahead and move | 18:54 |
| reed | jeblair, there is no decision to buy anything, just to get a quote | 18:54 |
| reed | jeblair, it's not, unfortunately | 18:55 |
| harlowja | jd__ great | 18:55 |
| reed | pootle and the others have more limitations than TX, that was the result of the discussion in HK | 18:55 |
| reed | and of the evaluation done before | 18:55 |
| jog0 | notmyname: you hit a E901 in the test | 18:55 |
| jog0 | http://logs.openstack.org/34/57334/1/check/gate-hacking-python26/2d8b364/console.html#_2013-11-19_22_46_29_429 | 18:55 |
| jog0 | which is why its failing | 18:55 |
| jog0 | so fix that and cut and paste some of that comment into the commit message and +2 from me it is | 18:56 |
| jog0 | then we have to cut a hacking 0.8.1, which is trivial enough | 18:56 |
| jeblair | reed: sure, they all have problems; my point is that we decided to move, and therefore, we should spend effort there and not on transifex. | 18:57 |
| *** sarob has quit IRC | 18:57 | |
| reed | feel free to reply to my email to the i18n team so we get to the same page... | 18:57 |
| *** zul has quit IRC | 18:57 | |
| *** sarob has joined #openstack-infra | 18:58 | |
| jeblair | reed: (i'm not sure that others have more limitations than tx; all tools have some gaps) | 18:58 |
| reed | jeblair, indeed, they all do but we have here a social problem mostly, not a tech one | 18:58 |
| *** hogepodge has joined #openstack-infra | 18:58 | |
| clarkb | jog0: was there something specific on the etherpad you wanted me to look at? I am reading it and not seeing much of a diff | 18:58 |
| jog0 | clarkb: I added some notes at the bottom of things I want to have written down somewhere | 18:59 |
| reed | jeblair, the social issue is that the i18n team doesn't want to change tool, they are already familiar with TX which they perceive as the "best" | 18:59 |
| *** david-lyle_afk is now known as david-lyle | 18:59 | |
| jeblair | reed: then they should have said that in hk | 18:59 |
| reed | jeblair, whatever else we try to give them they will not like | 18:59 |
| jeblair | reed: but instead they agreed to change to a new tool | 18:59 |
| jog0 | clarkb: in part, explain the new flow for debugging gate | 18:59 |
| jog0 | as I have seen people try to help and not really sure how | 18:59 |
| reed | jeblair, one can argue they didn't agree to "changing", quite the contrary ... | 19:00 |
| *** hogepodge has quit IRC | 19:00 | |
| jog0 | clarkb: under items to cover | 19:00 |
| fungi | reed: should we consider privatizing the translations entirely since they're going to be done using proprietary tools now? | 19:00 |
| clarkb | jog0: thansk reading those now | 19:01 |
| jeblair | reed: wow, that's really disingenuous if that's what they're saying | 19:01 |
| notmyname | jog0: I'm not sure I know how to read that log error. (and it's clayg's patch) | 19:01 |
| reed | fungi, my opinion counts very little, I don't 'manage' them, I have little authority | 19:01 |
| fungi | reed: it doesn't sound like there's any intent for them to be part of openstack, the free software project any longer | 19:01 |
| jeblair | reed: you don't fly out to a summit, get together for a conversation, come to an agreement, and then say "oh, well i didn't technically agree to that" | 19:01 |
| jog0 | notmyname: 'incorrectly found E901' | 19:01 |
| *** wenlock has joined #openstack-infra | 19:02 | |
| jog0 | clarkb: I am thinking maybe we add a doc to e-r saying how we do things | 19:02 |
| jog0 | I started one | 19:02 |
| clarkb | jog0: ++ | 19:02 |
| *** mfer has quit IRC | 19:02 | |
| fungi | reed: this takes them into the realm of o'reilly press editing our manuals with proprietary tools. that effort isn't part of open stack, but rather a separate project which we happen to receive the end results from | 19:02 |
| jog0 | clarkb: https://review.openstack.org/#/q/status:open+project:openstack-infra/elastic-recheck+branch:master+topic:queries,n,z | 19:02 |
| jog0 | by start I mean add a few lines | 19:02 |
| jog0 | but its a start | 19:02 |
| reed | fungi, good point | 19:02 |
| jog0 | that what in the email we can say, if you want to help read x | 19:02 |
| openstackgerrit | A change was merged to openstack-infra/jenkins-job-builder: update doc and add new JJB unit tests https://review.openstack.org/56715 | 19:03 |
| *** sarob has quit IRC | 19:03 | |
| *** sarob has joined #openstack-infra | 19:03 | |
| *** sandywalsh has quit IRC | 19:04 | |
| fungi | jd__: clarkb: i'm -1 on that patch for the moment. i'd like to see nova weigh in on the efficacy of that new libvirt package, since they have an open task on the bug | 19:04 |
| *** sarob has quit IRC | 19:05 | |
| mordred | reed: I agree with jeblair | 19:05 |
| *** sarob has joined #openstack-infra | 19:05 | |
| clarkb | fungi: nova cores are proposing we stop supporting libvirt older than 0.9.11 (precise is 0.9.8) because we can have pip installable python bindings with 0.9.11 | 19:05 |
| mordred | I feel that we decided that we would investigate the options more fully so that we could understand what a migration plan would look like | 19:05 |
| clarkb | fungi: do we need to merge a discussion somehwere? | 19:06 |
| *** hogepodge has joined #openstack-infra | 19:06 | |
| mordred | reed: we under no circumstances did anything that would imply that we would pay tx money for anyting or move to an enterprise version of anything | 19:06 |
| fungi | clarkb: potentially, i think it's not reflected in either the proposed change or the bug. even just a hyperlink in the bug to the ml thread might help? | 19:06 |
| clarkb | fungi: sure, have a bug number? I will grab a link to the ml thread | 19:07 |
| fungi | clarkb: i'm more concerned with knowing that the 1.1.1 they've added to cloud archive is known to fix bug 1228977 | 19:07 |
| uvirtbot | Launchpad bug 1228977 in nova "n-cpu seems to crash when running with libvirt 1.0.6 from ubuntu cloud archive" [High,Confirmed] https://launchpad.net/bugs/1228977 | 19:07 |
| *** Ryan_Lane has quit IRC | 19:08 | |
| clarkb | fungi: ok, I will leave notes on the bug about the ml discussion | 19:08 |
| fungi | clarkb: the last comment, from the package maintainer, just said "here's a new version, please test" | 19:08 |
| mordred | in fact, I find the o'reilly thing pretty repulsive, but am willing to deal with it because annegentle is really awesome | 19:08 |
| *** Ryan_Lane has joined #openstack-infra | 19:08 | |
| fungi | wanting to make sure that the "test" isn't us adding cloud archive back to all our slaves | 19:08 |
| clarkb | jog0: left notes on the etherpad | 19:09 |
| clarkb | fungi: gotcha | 19:09 |
| jeblair | mordred: to be fair, they keep saying things like "we want to figure out how to work with your process". maybe monday's meeting will get us somewhere since we won't be on a time crunch. | 19:09 |
| *** sarob has quit IRC | 19:09 | |
| jeblair | mordred: and that's all i'm going to say on that, because, indeed, they have had a while to figure it out and haven't yet. :/ | 19:10 |
| *** sdake_ has joined #openstack-infra | 19:10 | |
| *** sdake_ has quit IRC | 19:10 | |
| *** sdake_ has joined #openstack-infra | 19:10 | |
| fungi | i take responsibility for a lot of the previous call outcome with o'reilly. they had something like 2 days to start editing and wanted a conference call to talk about process, but i was the only infra core who was available to make the call on such short notice | 19:10 |
| jeblair | fungi: well, it wasn't much different than the call before that. and probably the call before that too. but hey, maybe this time it'll be different. :) | 19:11 |
| * fungi hopes | 19:11 | |
| *** DinaBelova has joined #openstack-infra | 19:11 | |
| jeblair | reed, mordred, fungi: i have sent what i hope is a very clear message to the i18n list | 19:11 |
| mordred | so going | 19:11 |
| mordred | jeblair: awesome | 19:11 |
| jeblair | in particular, if people don't think what i said we agreed to is what we agreed to, then we have a problem. :) | 19:12 |
| mordred | I agree | 19:12 |
| *** oubiwann has quit IRC | 19:12 | |
| mordred | also, I'd like to just go on the record as saying that if hte problem is "transifex has gone closed-source" the solution to that problem is not "buy a commercial subscription" | 19:13 |
| mordred | I'd also like to say that, while I apprecite the feedback from our users on their opinions of the tool we're going to use, ultimately we have not made any tooling decisions based on popularity contests | 19:13 |
| mordred | but instead we have based them on actual technical needs | 19:13 |
| mordred | and we have pissed off many sets of people in the short term due to that | 19:14 |
| jeblair | mordred: that is quite true, and in turn have made a lot more people happy in the long run. | 19:14 |
| mordred | so I'd like to be clear that we're kind of going out of our way to work with people on a decision we're pretty sure will have impact on them here | 19:14 |
| mordred | jeblair: exactly | 19:14 |
| clarkb | fungi: bug updated, any chance you can read what I have and the ML threads and perhaps send a response to merge the threads? | 19:14 |
| jog0 | clarkb: thanks | 19:15 |
| mordred | openstack running key parts of its process on closed-source is not and never has been acceptable | 19:15 |
| fungi | clarkb: just saw the bug update--thanks! i'll loop back around on the ml | 19:15 |
| clarkb | fungi: I am trying to catch up on that bug too and may try to reproduce with 1.1.1 myself (though probably not today as I would need to spin up a proper tempest test bed, unless I hold a d-g node...) | 19:16 |
| clarkb | fungi: actually I think that isn't a half bad idea, we could hold an arbitrary d-g node, wait for tests to finish, add cloud archive and upgrade, then run tempest again | 19:16 |
| jeblair | reed: so, ah, thanks for the heads up. :) | 19:17 |
| fungi | clarkb: i spin up devstack-tempest slaves pretty regularly to run tests. i can just add cloud-archive in the same manner that puppet change would and try a tempest-full on it | 19:18 |
| fungi | i'll try it out here in a bit (i hope) | 19:18 |
| clarkb | fungi: woot that would be super helpful thanks | 19:18 |
| clarkb | (I forget that you are probably the person that know how to do that best) | 19:18 |
| clarkb | well you and ttx | 19:18 |
| *** sandywalsh has joined #openstack-infra | 19:19 | |
| *** sarob has joined #openstack-infra | 19:23 | |
| reed | jeblair, thank you for the reply. It's good to keep the conversation going | 19:24 |
| reed | I'll reach out to the translatewiki, zanata and pootle folks | 19:25 |
| *** pabelanger has quit IRC | 19:26 | |
| reed | and I learned about another tool yesterday, weblate | 19:27 |
| *** pabelanger has joined #openstack-infra | 19:27 | |
| *** hashar has joined #openstack-infra | 19:28 | |
| fungi | clarkb: maybe i'm misreading http://lists.openstack.org/pipermail/openstack-dev/2013-November/020117.html but does it sound to you like they're asking for us to run tests on machines with vt-d? i was mostly finished drafting a response to point them at the third-party testing doc, but figure a second set of eyes on their request would be good first | 19:29 |
| *** herndon has quit IRC | 19:33 | |
| openstackgerrit | Soren Hansen proposed a change to openstack-infra/config: Add BasicDB to stackforge https://review.openstack.org/55519 | 19:33 |
| portante | have folks seen this behavior before with grenade? https://bugs.launchpad.net/tempest/+bug/1254142 | 19:36 |
| clarkb | fungi: yes test nodes with vtd is how I read that | 19:36 |
| uvirtbot | Launchpad bug 1254142 in tempest "check grenade job failed with "fatal: The remote end hung up unexpectedly"" [Undecided,New] | 19:36 |
| portante | http://logstash.openstack.org/#eyJzZWFyY2giOiJtZXNzYWdlOlwiZmF0YWw6IFRoZSByZW1vdGUgZW5kIGh1bmcgdXAgdW5leHBlY3RlZGx5XCIgQU5EIGZpbGVuYW1lOmNvbnNvbGUuaHRtbCIsImZpZWxkcyI6W10sIm9mZnNldCI6MCwidGltZWZyYW1lIjoiYWxsIiwiZ3JhcGhtb2RlIjoiY291bnQiLCJ0aW1lIjp7InVzZXJfaW50ZXJ2YWwiOjB9LCJzdGFtcCI6MTM4NTE0ODcxODUzNH0= | 19:36 |
| clarkb | they will need third party testing because kvm passthrough doesn't seem to work well or be on our cloud provider radars | 19:36 |
| fungi | clarkb: okay, firing reply then. thanks@ | 19:36 |
| clarkb | fungi: lifeless may have opinions on using the tripleo cloud | 19:36 |
| fungi | clarkb: no doubt, once it's ready | 19:37 |
| *** rfolco has quit IRC | 19:37 | |
| soren | clarkb: Are you sure kvm passthrough would even suffice? They're specifically asking for not just VT-D (which I don't believe kvm passthrough offers) as well as some device that provides SR-IOV. | 19:37 |
| *** rpodolyaka has joined #openstack-infra | 19:38 | |
| *** SergeyLukjanov has joined #openstack-infra | 19:38 | |
| *** mrmartin has quit IRC | 19:39 | |
| *** SergeyLukjanov has quit IRC | 19:39 | |
| clarkb | soren: I actually don't | 19:40 |
| openstackgerrit | Matt Riedemann proposed a change to openstack-infra/elastic-recheck: Add query for bug 1177134 https://review.openstack.org/58014 | 19:40 |
| uvirtbot | Launchpad bug 1177134 in devstack "[ERROR] /opt/stack/new/devstack/exercises/bundle.sh:61 Image ami-00000001 not available within 15 seconds" [Medium,Triaged] https://launchpad.net/bugs/1177134 | 19:40 |
| clarkb | soren: does kvm passthrough only do vtx? | 19:40 |
| *** pafuent has left #openstack-infra | 19:41 | |
| soren | clarkb: Last I checked. | 19:41 |
| soren | clarkb: Which arguably was a couple of years ago. | 19:41 |
| *** ArxCruz has quit IRC | 19:42 | |
| clarkb | google doesn't return any "it works!" results so may still be the case | 19:43 |
| *** rongze has joined #openstack-infra | 19:43 | |
| *** SergeyLukjanov has joined #openstack-infra | 19:44 | |
| *** rongze has quit IRC | 19:48 | |
| *** dprince has quit IRC | 19:52 | |
| jog0 | clarkb: ping | 19:54 |
| *** rpodolyaka has quit IRC | 19:55 | |
| jog0 | so for the 18 hour number, that may have been true, but that doesn't reflect how long it would take to flush the queue | 19:55 |
| jog0 | if nothing merges it would take 145 hours for 145 patches to go through | 19:55 |
| jog0 | clarkb: for the graphite graph | 19:55 |
| jog0 | my html /js is weak | 19:55 |
| soren | clarkb: I'm not surprised. It's a lot of work with no reasonable real world use cases. | 19:56 |
| *** mrmartin has joined #openstack-infra | 19:58 | |
| *** herndon has joined #openstack-infra | 20:01 | |
| *** mrmartin has quit IRC | 20:03 | |
| clarkb | jog0: gotcha. my html/js is weak too :) | 20:05 |
| *** mrmartin has joined #openstack-infra | 20:05 | |
| clarkb | jog: I would copy pasta from the zuul status page | 20:05 |
| clarkb | it uses a grahpite js lib | 20:05 |
| jog0 | ohh thats what that was | 20:06 |
| *** reed has quit IRC | 20:06 | |
| jog0 | I just used an image | 20:06 |
| jog0 | http://paste.openstack.org/show/53834 | 20:06 |
| *** flaper87|afk is now known as flaper87 | 20:06 | |
| *** _david_ has joined #openstack-infra | 20:09 | |
| *** lcestari has quit IRC | 20:09 | |
| *** MarkAtwood has joined #openstack-infra | 20:09 | |
| *** mrmartin has quit IRC | 20:10 | |
| dkranz | clarkb: Can you give your thoughts on my last comment in https://bugs.launchpad.net/tempest/+bug/1253896 ? | 20:10 |
| uvirtbot | Launchpad bug 1253896 in tempest "test_minimum_basic_scenario fails with SSHException: Error reading SSH protocol banner" [Undecided,New] | 20:10 |
| dkranz | clarkb: This is a little outside my area. | 20:10 |
| clarkb | dkranz: interesting, I think you may be on to something | 20:12 |
| dkranz | clarkb: ok, good. But I'm not sure how to test the fix. | 20:13 |
| *** dolphm has quit IRC | 20:13 | |
| dkranz | clarkb: I can put up a patch and run it a bunch of times | 20:13 |
| clarkb | qemu is really slow. cirros is very lightweight but start 4 of them and we may bog down. I think relaxing timeouts if possible might be worth trying also catching exceptions where necessary | 20:13 |
| dkranz | clarkb: We have to catch for this one becuause the timeout is hard-wired in paramiko | 20:13 |
| clarkb | ya testing this stuff is hard because it may be related to the underlying hypervisors load too | 20:14 |
| dkranz | clarkb: But we could increase the console log wait time | 20:14 |
| dkranz | clarkb: I'll give it a try | 20:14 |
| clarkb | cool let me know how it goes | 20:14 |
| *** mfer has joined #openstack-infra | 20:15 | |
| jeblair | dkranz: i think since the "Banner:" debug line doesn't show up, it's probably the 15 second timeout for the first line it's hitting. | 20:18 |
| dkranz | jeblair: Makes sense. I am going to just catch all SSHException so it won't matter | 20:18 |
| *** nsaje has joined #openstack-infra | 20:19 | |
| *** nsaje has quit IRC | 20:19 | |
| *** chuck__ has quit IRC | 20:20 | |
| *** mrmartin has joined #openstack-infra | 20:22 | |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/config: Add graphite graph of gate failure rates to e-r page https://review.openstack.org/58019 | 20:23 |
| jog0 | clarkb jeblair ^ | 20:23 |
| jog0 | I'm sure I got osmething wrong on that | 20:23 |
| jog0 | being I wasn't sure how to locally test that out | 20:23 |
| clarkb | you should be able to just open the page locally. the other graphs will 404 but the graphite thing should render | 20:26 |
| *** dstanek has joined #openstack-infra | 20:26 | |
| jog0 | hmm it tried hittinh localhost instead of the proper graphite server | 20:27 |
| jog0 | digging | 20:27 |
| *** hogepodge has quit IRC | 20:27 | |
| clarkb | you might need to put it behind python simple http server but I doubt it | 20:27 |
| jog0 | clarkb: woot got it working | 20:28 |
| jog0 | was missing a line | 20:28 |
| jog0 | urL: | 20:28 |
| *** sarob has quit IRC | 20:29 | |
| jog0 | hmm the image is too small | 20:29 |
| clarkb | you can give dimensions to graphite | 20:29 |
| jog0 | clarkb: the image is big the page is rendering it small | 20:30 |
| openstackgerrit | A change was merged to openstack-infra/zuul: Doc: enhance gating reference with diagrams https://review.openstack.org/54648 | 20:30 |
| *** gyee has joined #openstack-infra | 20:31 | |
| *** dkliban has quit IRC | 20:31 | |
| *** thomasem has quit IRC | 20:31 | |
| *** hogepodge has joined #openstack-infra | 20:32 | |
| jog0 | its somet css stuff | 20:32 |
| chmouel | if you guys aren't too busy to review this (add gates tests to swift-bench) https://review.openstack.org/#/c/57845/ | 20:33 |
| chmouel | so we can have our first commit merging there | 20:33 |
| jog0 | clarkb: fixed | 20:38 |
| _david_ | clarkb, jeblair gerit-2.8-rc3 was released today. GA is going to be released in one or two weeks, ... | 20:38 |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/config: Add graphite graph of gate failure rates to e-r page https://review.openstack.org/58019 | 20:38 |
| clarkb | _david_ spearce -2'd zaros patch so it won't get in will it? | 20:39 |
| clarkb | zaro ^ | 20:39 |
| _david_ | recently we had a discussion to set up testing environment for it and prepare the migration? | 20:39 |
| _david_ | clarkb, zaro's patch is on master, we are talking about 2.8 final | 20:40 |
| clarkb | right we can't upgrade until we have that feature | 20:41 |
| _david_ | if and when zaro's patch is going to be approved, it can be only available in the next major, i. e. 2.9 | 20:41 |
| clarkb | or an equivalent thing | 20:41 |
| _david_ | clarkb, yeah, we need still core change anyway for WIP plugin to work, so, we would apply zaro's patch too | 20:42 |
| jeblair | wow, uh, the 2.8 ui has changed a lot. | 20:43 |
| _david_ | jeblair, the new change screen really rocks ;-) | 20:43 |
| jeblair | _david_: did you write it? | 20:43 |
| _david_ | even more cooler is the master now, with related changes tabs | 20:43 |
| _david_ | jeblair, Nope, wop wrote it, at least the first version, i contributed a lot to the new change screen | 20:44 |
| _david_ | s/wop/sop | 20:44 |
| jeblair | _david_: it doesn't fit on my 1280x1024 screen | 20:45 |
| openstackgerrit | A change was merged to openstack/requirements: Bump to using SQLAlchemy migrate 0.8.2. https://review.openstack.org/56662 | 20:45 |
| _david_ | jeblair, really? https://gerrit-review.googlesource.com/#/c/51996/ | 20:45 |
| *** alcabrera has quit IRC | 20:46 | |
| jeblair | _david_: that one is fine, zaro's has more tabs... screenshot coming | 20:46 |
| jeblair | https://gerrit-review.googlesource.com/#/c/48254/ | 20:46 |
| openstackgerrit | A change was merged to openstack-dev/pbr: Bump the development status classifier. https://review.openstack.org/57272 | 20:46 |
| _david_ | jeblair, that's true! | 20:46 |
| _david_ | jeblair, there are 4 tabs: 1/ dependencies, /2 same topic /3 conflicting changes with the curent change /4 cerry picked | 20:47 |
| _david_ | jeblair, those tabs are conditions, if they are empty, then they are hidden., ... | 20:47 |
| _david_ | on zaro's patch they are full, and shown | 20:47 |
| jeblair | http://imgur.com/WdrYFzB | 20:47 |
| *** dolphm has joined #openstack-infra | 20:47 | |
| *** chuck__ has joined #openstack-infra | 20:48 | |
| jeblair | _david_: also same topic says "400 Unsupported query:Owner" why is that? | 20:48 |
| _david_ | Is that on gerrit-review? | 20:49 |
| jeblair | _david_: yes | 20:49 |
| jeblair | _david_: on https://gerrit-review.googlesource.com/#/c/48254/28 | 20:49 |
| _david_ | jeblair, unbelievable ;-) | 20:49 |
| _david_ | jeblair, since Gerrit 2.8 there is secondary index = Lucene or Solr bilt in in Gerrit | 20:49 |
| _david_ | so that all new and shiny queries are working only if secondary index is enabled | 20:50 |
| _david_ | something was broken there ,-) | 20:51 |
| _david_ | jeblair, on my 15" wide screen laptop all is fine with the screen and all 4 tabs are visible | 20:51 |
| jeblair | _david_: yes, i can obviously make the window wider; the question is whether it's a good idea from a UI perspective to _require_ that people have widescreen displays in order to use gerrit | 20:52 |
| *** sarob has joined #openstack-infra | 20:53 | |
| *** miqui has quit IRC | 20:53 | |
| *** dkliban has joined #openstack-infra | 20:53 | |
| *** sarob has joined #openstack-infra | 20:53 | |
| _david_ | jeblair, yes they should ;-) Today each and every one has 15" wide screen laptop + 30" desktop display | 20:53 |
| jeblair | _david_: i think UIs that can reflow to accomodate various widths are best, but if you're going to assume you're on a desktop, assuming a width of 1024 is nice, or 1280 as a max. | 20:53 |
| openstackgerrit | A change was merged to openstack-infra/config: Switch gate-tempest-devstack-vm-neutron-large-ops to a voting job https://review.openstack.org/56629 | 20:54 |
| _david_ | jeblair, sop added only one tab: related changes | 20:54 |
| _david_ | jeblair, the other 3 was added recently ... | 20:54 |
| * fungi runs his browser on a 1080x1920 (portrait) display, shrinks the font and still has to scroll gerrit horizontally. horrible ui choice | 20:55 | |
| *** DinaBelova has quit IRC | 20:55 | |
| openstackgerrit | Armando Migliaccio proposed a change to openstack-infra/config: Avoid false positives for python27 UT gate job https://review.openstack.org/58022 | 20:55 |
| jeblair | _david_: i don't think it's a good assumption to make that everyone has a display like that. because it's clearly not true. | 20:55 |
| fungi | i also sometimes use it from a wearable with a 640x480 display and wow. scrolltastic | 20:55 |
| *** sarob_ has joined #openstack-infra | 20:55 | |
| openstackgerrit | Khai Do proposed a change to openstack-infra/config: Setup a private gerrit instance for security reviews https://review.openstack.org/47937 | 20:56 |
| *** sarob has quit IRC | 20:56 | |
| zaro | fungi: ^ | 20:56 |
| jeblair | _david_: and moreover, there's a huge amount of space wasted if you expand the window to hold all 4 tabs -- fully 25% of the space under them is wasted since the rest of the page isn't that wide. | 20:56 |
| openstackgerrit | Chmouel Boudjnah proposed a change to openstack-infra/config: Add minimal checks for swift-bench. https://review.openstack.org/57845 | 20:56 |
| portante | clarkb: grenade failed because pip could not be downloaded, should I file a bug, or just recheck the job? | 20:57 |
| jeblair | _david_: my display is actually 2560x1024 wide, but it spans two monitors. that means i can have gerrit in one and some terminals and an email client in another. | 20:58 |
| fungi | portante: i think we need a bug for that if there isn't one already. we shouldn't be trying to download pip directly from pypi.python.org during tests | 20:58 |
| jeblair | _david_: and even when i'm on a widescreen laptop, i only expand my browser to about 1024 wide so that i can fit some terminals beside it... | 20:58 |
| portante | fungi: k | 20:58 |
| jeblair | _david_: all of that is to say that people have different ways of working, with different display systems... | 20:58 |
| portante | filing ... | 20:58 |
| jeblair | _david_: and it's nice to try to accomodate them by not requiring that they run an app in full screen mode | 20:59 |
| _david_ | jeblair, can you remember as we have very vital discussion on #gerrit with mfick about third columnon old change screen? | 20:59 |
| fungi | as soon as i can get this displayport mst hub to work better under linux, i'll be able to run both my portrait displays in 1440x2560, and then maybe the browser one will be wide enough to not have to scroll gerrit | 21:00 |
| portante | fungi: it was trying to download from pypi.openstack.org, though | 21:00 |
| portante | Error <urlopen error timed out> while getting http://pypi.openstack.org/openstack/pip/pip-1.4.1.tar.gz (from http://pypi.openstack.org/openstack/pip/) | 21:00 |
| _david_ | jeblair, we wanted to put there verification status table, and he was against it, exactly for these reasons, ... | 21:00 |
| fungi | portante: oh, that's definitely weird | 21:00 |
| portante | that is a 404 | 21:00 |
| zaro | fungi: talked to clarkb and came up with the config in the rst file. | 21:00 |
| fungi | portante: definitely weird to get a 404 there... it certainly is a valid url and i can download from it just fine | 21:01 |
| fungi | zaro: awesome! | 21:01 |
| _david_ | jeblair, but beside screen question: are we going to setup a test gerrit installation for 2.8 in the near future? | 21:01 |
| portante | I get a 404 hitting it right now | 21:02 |
| portante | oopps | 21:02 |
| portante | no, it is ithere | 21:02 |
| jeblair | _david_: i never thought it had to make it wider; our gerrit 2.4 only uses a little less than 2/3 of the width at 1280 (a little more than 2/3 at 1024); plenty of room for a 3rd column, and of course it could wrap if needed | 21:02 |
| *** yolanda has quit IRC | 21:02 | |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/elastic-recheck: Swap in better query for 1239856 https://review.openstack.org/58024 | 21:02 |
| jeblair | _david_: and lots of room to the right of the reviewer table which takes < 1/2 the width | 21:02 |
| portante | must have been a simple service distruption | 21:02 |
| jeblair | _david_: right, anyway. :) | 21:03 |
| portante | is it really worth a bug for that, fungi? | 21:03 |
| jeblair | _david_, clarkb, zaro: i think we wanted to hold off on setting it up until the patch at least landed in master | 21:03 |
| portante | or perhaps file a bug so that we have a handle to recheck on? | 21:03 |
| jog0 | jeblair: glance-bugs is not open | 21:03 |
| jog0 | I got me a critical | 21:03 |
| jog0 | markwash: ^ | 21:03 |
| jog0 | https://bugs.launchpad.net/glance/+bug/1239856 | 21:03 |
| fungi | portante: i think so. i'd like to check the apache logs on static.o.o where that's hosted | 21:03 |
| uvirtbot | Launchpad bug 1239856 in glance "Registry client request GET /images/UUID raised ClientConnectionError" [Undecided,Confirmed] | 21:03 |
| portante | k, filing | 21:04 |
| jeblair | _david_, clarkb, zaro: and with a -2 on that patch (even though it looks like it's for fixable things, which might even be fixed) i'm still worried | 21:04 |
| portante | tempest? | 21:04 |
| zaro | getting a patch to land in upstream gerrit is an exercise in frustration! | 21:04 |
| _david_ | jeblair, anyway, because we are going to make a really big jump from 2.4 to 2.8 there is a lot to do with configuration upgrade, database migration, tests | 21:04 |
| fungi | portante: so having timestamps will help me there | 21:04 |
| fungi | portante: openstack-ci | 21:04 |
| portante | k | 21:04 |
| _david_ | zaro, if i can give you advice: take it easy, really, there is no difference at all, if the change is landed today or in two months | 21:05 |
| zaro | jeblair, _david_, clarkb : i have sent out another email to spearce to take another look. i think that's the third time. | 21:05 |
| _david_ | jeblair, i don't understand how installation setup depends on landing or not of zaro's change | 21:06 |
| jeblair | _david_: we can't even be remotely certain that the change will land. | 21:06 |
| _david_ | zaro, that was wrong, please don't send and don't ask, that makes the thing only worse, believe me | 21:06 |
| zaro | jeblair, _david_, clarkb : i also asked everybody that have reviewed it to re-review. | 21:06 |
| portante | fungi: https://bugs.launchpad.net/openstack-ci/+bug/1254167 | 21:06 |
| uvirtbot | Launchpad bug 1254167 in openstack-ci "pip fails to install, timeout fetching from pypi.openstack.org" [Undecided,New] | 21:06 |
| fungi | portante: thanks! | 21:07 |
| zaro | _david_: good to know. mfick told me that spearce doesn't mind being reminded | 21:08 |
| _david_ | jeblair, but for WIP plugin to work we still need to patch core, so who care, that we patch it twice, | 21:08 |
| _david_ | well the only difference between WIP core change and zaro's change that zaro's change touch database | 21:08 |
| jeblair | _david_: we don't want to run a fork, we want to run upstream. if these patches land, i think we're okay running a patched version because we know we won't have to work to carry the patch in the future | 21:09 |
| portante | fungi: i rechecked that job so we can see if it will repeat readily | 21:09 |
| jeblair | _david_: but until it's certan that they will land, we don't want to take the risk. | 21:09 |
| jeblair | _david_: that's how we ended up running the fork we're running today, which we want to stop doing. :) | 21:09 |
| _david_ | jeblair, may be i am missing something, but what about that guy: https://gerrit-review.googlesource.com/#/c/50250/ ? | 21:10 |
| jeblair | jog0: what launchpad team should have access? | 21:10 |
| pleia2 | jeblair: should all channels we interact with have jeblair set as founder? (I've been asking people to do it for gerritbot, but wondering about logging too) | 21:10 |
| _david_ | last time we discussed it with mordred we agreed upon to patch at least this for now. | 21:10 |
| jeblair | pleia2: yes, so that when we automate channel configuration, i can just grant access to the script | 21:12 |
| jog0 | jeblair: hrm not actually sure | 21:13 |
| jog0 | let me check | 21:13 |
| jog0 | sdague: ^ oh wait your on a plane | 21:13 |
| jeblair | _david_: i think we agreed that we would patch it against 2.8 if it went into master; but i don't want to speak for mordred. | 21:13 |
| jog0 | jeblair: two options come to mind | 21:14 |
| jog0 | https://launchpad.net/~tempest-bug-team | 21:14 |
| jog0 | https://launchpad.net/~openstack-qa-team | 21:14 |
| jog0 | but qa team has 157 members | 21:15 |
| _david_ | jeblair, it went not into master, it was rejected, now what? | 21:15 |
| jog0 | so jeblair I am not really sure | 21:15 |
| _david_ | jeblair, you guys are 4 major release behind master, and you say, you really need this feature, so it was migrated into a plugin and still depends on 10 lines of changes in core, that were rejected by mfick. And now you want to tell me, that it is a no go for you to patch 10 LoC in Gerrit core? | 21:17 |
| jeblair | _david_: i think monty's last comment on that patch is good -- it would be worth working on generic statuses so that we can run 2.8 (or, well, 2.9 now) without carrying local patches | 21:19 |
| jeblair | _david_: but again, i don't want to speak for mordred | 21:19 |
| jeblair | _david_: we should really get his input since he's been involved more on this | 21:19 |
| jog0 | clarkb: ping | 21:20 |
| jog0 | https://bugs.launchpad.net/tempest/+bug/1253896 | 21:20 |
| _david_ | jeblair, i am almost done with generic status change: https://gerrit-review.googlesource.com/#/c/50700 | 21:20 |
| uvirtbot | Launchpad bug 1253896 in tempest "test_minimum_basic_scenario fails with SSHException: Error reading SSH protocol banner" [Undecided,New] | 21:20 |
| jog0 | is paramiko I think | 21:20 |
| clarkb | jog0: pong | 21:21 |
| _david_ | jeblair, how to get his input? | 21:21 |
| jog0 | clarkb: at least paramiko doesn't work with eventlet | 21:21 |
| jeblair | _david_: cool, so it sounds like it's getting closer! | 21:21 |
| clarkb | jog0: I don't think tempest uses eventlet | 21:22 |
| jog0 | clarkb: yeah | 21:22 |
| jeblair | _david_: mordred is here most of the time, he's probably just in a meeting now or something. | 21:22 |
| jog0 | after I said the parallel I think it was wrong | 21:22 |
| jog0 | clarkb: could it be the VM taking too long to start up? | 21:23 |
| _david_ | jeblair, i already told you my opinion how to proceed in this situation: take these two patches zaro's (that /me voted +1) and mine and patch core | 21:23 |
| _david_ | jeblair, 2.8, and bcause it is almost stable now, i am here to ask you what are the next steps | 21:23 |
| clarkb | jeblair: https://review.openstack.org/#/c/57807/ any chance I can get you to review that really quick? need to get the subunit conversion on precise slaves to help neutron py27 failure false positives due to timeouts | 21:24 |
| clarkb | jog0: no the VM is up, we know this because the dropbear server is saying hello you are connected to dropbear version foo | 21:24 |
| jeblair | _david_: (not to distract again, but here's another bit of usability feedback; the comments section should be less wide; it's difficult for eyes to track back when reading long lines, so they should only be about 75 characters long at the most) | 21:24 |
| clarkb | jog0: I think it may be too slow to do the rsa stuff which is relatively expensive | 21:25 |
| jog0 | clarkb: right | 21:25 |
| jog0 | ohh you mean the entire env | 21:25 |
| jeblair | _david_: i personally want to wait until some of this stuff starts to merge. zaro and mordred may have different opinions which i'd like to hear from them. | 21:25 |
| jog0 | we can check the load at teh same time | 21:25 |
| _david_ | jeblair, that why it is so important that you upgrade to get your feedback and hear your voice in current and future development | 21:26 |
| jog0 | clarkb: loads no the devstack vm don't look bad | 21:27 |
| *** mrmartin has quit IRC | 21:27 | |
| jog0 | clarkb: maybe the metaserver didn't do its think in time? | 21:27 |
| clarkb | jog0: ohhhhh ya good point, the authorized key on the client side comes from the metadata server right? | 21:28 |
| clarkb | could be a problem there | 21:28 |
| *** dkliban has quit IRC | 21:28 | |
| *** ewindisch has quit IRC | 21:28 | |
| jeblair | clarkb: why do you need 2 service definitions? | 21:28 |
| clarkb | jeblair: because one is only for debian derivatives | 21:29 |
| *** vipul is now known as vipul-away | 21:29 | |
| clarkb | jeblair: and the other is for only slaves | 21:29 |
| *** vipul-away is now known as vipul | 21:29 | |
| clarkb | jeblair: and d-g image builds don't use server.pp | 21:29 |
| clarkb | (it is a bit of a mess) | 21:29 |
| jeblair | clarkb: i think they do? | 21:31 |
| jeblair | clarkb: slave_template -> slave -> server | 21:31 |
| clarkb | hmm why did the server builds work after I made the slave rsyslog changes? I will dig | 21:32 |
| jeblair | clarkb: cool. it _seems_ like defining the service in server and then notifying it from both places should work | 21:32 |
| fungi | i want to say we ran into something similar with the groups-dev server puppetry, and ended up being unable to have an object in one class notify a service in another class... but i could be misremembering | 21:34 |
| clarkb | jeblair: d-g slaves use slave_template -> jenkins::slave | 21:34 |
| clarkb | jeblair: there is no openstack_project::server | 21:34 |
| clarkb | jeblair: fungi: in this case I can't notify the rsyslog in server from slave.pp because d-g image builds don't build up from that point | 21:34 |
| *** _david_ has left #openstack-infra | 21:35 | |
| fungi | got it. openstack_server::slave not being in the chain there | 21:35 |
| fungi | er, openstack_project::slave | 21:35 |
| clarkb | right | 21:35 |
| jog0 | clarkb: metadata service has no hits on reboot | 21:35 |
| *** vipul is now known as vipul-away | 21:37 | |
| *** vipul-away is now known as vipul | 21:37 | |
| jeblair | clarkb: maybe the service definition should be in o_p::template then | 21:38 |
| mikal | So, who can approve https://review.openstack.org/#/c/56158/ ? (Not clarkb, he's already reviewed it) | 21:38 |
| clarkb | jeblair: I can do that | 21:38 |
| jeblair | clarkb: (since server includes template) | 21:38 |
| mikal | I hassle because it will make our tech writer's life much easier and is part of an evil plan to trick companies into hiring more tech writers | 21:38 |
| clarkb | jeblair: fungi: do we want all machines to have a larger rsyslog message buffer or just notify the service in openstack_p::template? | 21:38 |
| clarkb | jeblair: fungi: eg should I consolidate everything in template.pp or just the service resource | 21:39 |
| clarkb | I like consolidating because it makes the code less spaghetti | 21:39 |
| fungi | in this case, a minor increase in rsyslog buffer size across more machines as a trade off to easier-to-maintain puppet configuration sounds fine to me | 21:41 |
| devananda | hiya. I'm trying to sync latest db migration code and lockutils from oslo, and have run into this interesting thing: http://logs.openstack.org/16/56516/9/check/gate-ironic-python27/4d9b449/console.html | 21:41 |
| *** rnirmal has quit IRC | 21:41 | |
| devananda | chatting with mikal over in -ironic, but i thought ya'll might have some ideas about it, too | 21:41 |
| clarkb | devananda: looks like it is failing to create a file in the job workspace | 21:48 |
| *** dstanek has quit IRC | 21:48 | |
| clarkb | is it possible something is dleeting that dir before that test gets to it? | 21:48 |
| clarkb | in general shared paths like that are a bad thing. better to use tempdirs per test/process | 21:48 |
| *** dstanek has joined #openstack-infra | 21:51 | |
| *** dkliban has joined #openstack-infra | 21:51 | |
| *** gyee has quit IRC | 21:52 | |
| devananda | clarkb: everything in that changeset is coming from oslo ... | 21:53 |
| devananda | clarkb: it /used/ to use a per-test temp dir | 21:53 |
| clarkb | possible oslo bug then? | 21:54 |
| devananda | clarkb: compare L165 https://review.openstack.org/#/c/56516/6/ironic/tests/db/sqlalchemy/test_migrations.py | 21:54 |
| devananda | to L97 https://review.openstack.org/#/c/56516/9/ironic/openstack/common/db/sqlalchemy/test_migrations.py | 21:55 |
| markwash | jog0: hi. . can you help me figure out what's going on with https://bugs.launchpad.net/glance/+bug/1239856 ? | 21:55 |
| uvirtbot | Launchpad bug 1239856 in glance "Registry client request GET /images/UUID raised ClientConnectionError" [Undecided,Confirmed] | 21:55 |
| devananda | mikal: any idea why oslo would have stopped using a per-test tempdir in favor of a global / shared lock dir? | 21:55 |
| clarkb | wait what | 21:56 |
| clarkb | we are putting tests in oslo now? | 21:56 |
| clarkb | that seems wrong for some reason | 21:56 |
| devananda | clarkb: shared db test code | 21:57 |
| clarkb | oh is this locking across processes? | 21:57 |
| clarkb | we shouldn't do that | 21:57 |
| clarkb | instead each process should use a distinct schema in the DB, we haven't quite made that work for myslq because mysql permissiosn are funny, but we should just do that if this is the laternative | 21:58 |
| mikal | devananda: oh, I think that was because of a bug? | 21:58 |
| clarkb | (I was really hoping one of our mysql experts would make that change...) | 21:58 |
| mikal | devananda: the per lock dir was happening for non-test code? IIRC? | 21:58 |
| devananda | this http://git.openstack.org/cgit/openstack/oslo-incubator/tree/openstack/common/db/sqlalchemy/test_migrations.py#n169 | 21:59 |
| *** julim has quit IRC | 21:59 | |
| devananda | used to create a per-test temp dir and include that in the lock file prefix | 21:59 |
| devananda | it does not any more | 21:59 |
| devananda | afaik, this code was moved from $project/tests/db/sqlalchemy/test_migrations.py to $project/openstack/common/db/sqlalchemy/test_migrations.py | 22:00 |
| devananda | and also changed to no longer use a fixture // separate temp dirs per test | 22:00 |
| devananda | that second part doesn't make sense to me | 22:01 |
| *** denis_makogon has joined #openstack-infra | 22:01 | |
| devananda | i haven't followed any discussion that led to these changes, though. just ran into this when trying to apply the changes to ironic | 22:01 |
| *** mfer has quit IRC | 22:04 | |
| *** gyee has joined #openstack-infra | 22:09 | |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/config: Consolidate rsyslog configs. https://review.openstack.org/57807 | 22:09 |
| clarkb | jeblair: fungi ^ I think that addresses the concerns | 22:09 |
| clarkb | devananda: I have no idea either | 22:09 |
| devananda | clarkb: k, thanks. I'm testing preserving the fixtures.TempDir() behavior from the older code. I will propose it to oslo if it works | 22:10 |
| *** gyee_ has joined #openstack-infra | 22:10 | |
| *** gyee has quit IRC | 22:14 | |
| *** eharney has quit IRC | 22:14 | |
| fungi | clarkb: any thoughts on where we can go with further abstraction so we can differentiate things we want installed on devstack slaces and unit test slaves but not on trusted slaves? add a new jenkins::bare_slave class and rework o_p::bare_slave to just instantiate that? | 22:15 |
| fungi | s/slaces/slaves/ | 22:15 |
| clarkb | fungi: yeah, I think we need to take jenkins::slave.pp and split it into a couple manifests | 22:15 |
| clarkb | fungi: have one be "bare" then build the others on top of that | 22:15 |
| *** gyee_ has quit IRC | 22:16 | |
| clarkb | we sort of do that today with a conditional in slave.pp but it should be done at class level instead then you include jenkins::slave::base include jenkins::slave::all_the_stuff | 22:16 |
| fungi | because we sort of have that going on already in o_p but most of what i want to break out is over in the jenkins module unfortunately | 22:16 |
| *** gyee_ has joined #openstack-infra | 22:16 | |
| clarkb | yeah I think we can just start tearing that class apart and group related things in smaller classes | 22:16 |
| fungi | excellent. i'll head that direction with it | 22:17 |
| *** gyee_ has quit IRC | 22:17 | |
| clarkb | then compose with include instead of if | 22:17 |
| * fungi nods eagerly | 22:17 | |
| *** enarciso has joined #openstack-infra | 22:17 | |
| nibalizer | yea just a base class then some extending classes | 22:18 |
| nibalizer | this sounds like something i can help with, actually | 22:18 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/config: Remove +/-2 governance votes from tech committee https://review.openstack.org/58042 | 22:18 |
| jeblair | fungi: wha, huh? why are trusted slaves different than devstack/unit test slaves? | 22:19 |
| jeblair | (obviously, they get extra things, but why should they get fewer things) | 22:20 |
| jog0 | markwash: yeah | 22:21 |
| fungi | jeblair: we're installing things which various projects want to use to run tests, but are not available in our distributions (so we use a variety of thord-party package repositories) | 22:21 |
| fungi | er, third-party | 22:21 |
| jeblair | fungi: why are we doing that? | 22:21 |
| fungi | good question. should we stop? | 22:21 |
| jog0 | markwash: I left a comment with a logstash query URL that finds all the hits | 22:22 |
| *** pcm_ has quit IRC | 22:22 | |
| jeblair | fungi: probably? what projects are we talking about? | 22:22 |
| jog0 | and I see a DB error in the registry | 22:22 |
| * fungi will put together a list | 22:22 | |
| jog0 | but elasticsearch isn't collecting that | 22:22 |
| *** SergeyLukjanov has quit IRC | 22:22 | |
| fungi | jeblair: so, at the moment it's backports of py3k and pypy. we have a request from soren to add riak for tests he wants to run on a stackforge project | 22:24 |
| *** Ryan_Lane has quit IRC | 22:26 | |
| clarkb | we might get asked for node.js depending on how that horizon discussion goes | 22:26 |
| devananda | mikal, clarkb: fix for that locking in oslo - https://review.openstack.org/58044. this seems to have worked in ironic. | 22:26 |
| devananda | with that, i need to run! have a good weekend all! | 22:26 |
| clarkb | you too! | 22:26 |
| fungi | clarkb: yes, they do seem to be very close to deciding they want node.js as a test-time dependency | 22:26 |
| jeblair | so i don't think we've added an apt-repo for a non-openstack project. and we are pretty aggressive about not adding them for openstack projects too | 22:28 |
| jeblair | clarkb: where's this horizon discussion? | 22:28 |
| clarkb | jeblair: on the openstack-dev ml | 22:30 |
| jeblair | of course, under the 'BootstrapV3 and lessc' thread | 22:30 |
| *** ryanpetrello has quit IRC | 22:30 | |
| *** hashar has quit IRC | 22:30 | |
| pleia2 | jeblair: -ChanServ- jeblair has too many channels registered. | 22:32 |
| clarkb | nice jeblair wins | 22:32 |
| jeblair | ugh, that'll be a problem for the bot then | 22:33 |
| pleia2 | freenode can probably grant an acception, asking around | 22:33 |
| pleia2 | late on a friday is clearly not the time :) | 22:34 |
| jeblair | heh | 22:34 |
| cody-somerville | Why would you want the bot to be a founder anyhow? ie. why would it need that level of permission? | 22:35 |
| jeblair | cody-somerville: because no one other than the bot should. | 22:35 |
| openstackgerrit | A change was merged to openstack-infra/jeepyb: Allow automatic subscription to DocImpact bugs https://review.openstack.org/56158 | 22:36 |
| cody-somerville | Not to be difficult but why does it make a difference? | 22:36 |
| *** dangers is now known as danger_fo_away | 22:37 | |
| jeblair | cody-somerville: because people come and go all the time around here, but not projects. so asking people to grant other people founder status on irc is a losing proposition. | 22:37 |
| fungi | cody-somerville: in the past we've had random people register official channels, but then move on and the projects using those outlive them. getting freenode staff to hand over control of a registered channel to someone else has not been easy apaprently | 22:37 |
| jeblair | cody-somerville: we want to automate it like everything else. | 22:37 |
| fungi | er, what jeblair said | 22:38 |
| *** Ryan_Lane has joined #openstack-infra | 22:38 | |
| cody-somerville | small +f should be sufficient for that and works around the founder limit | 22:38 |
| *** Ryan_Lane has quit IRC | 22:38 | |
| *** Ryan_Lane has joined #openstack-infra | 22:38 | |
| *** rockyg has joined #openstack-infra | 22:39 | |
| *** mriedem has quit IRC | 22:39 | |
| *** enqae has joined #openstack-infra | 22:39 | |
| lifeless | clarkb: hi ? | 22:41 |
| *** ewindisch has joined #openstack-infra | 22:41 | |
| lifeless | clarkb: who needs 3rd party testing? | 22:41 |
| jeblair | cody-somerville: it would be nice for some account to have founder status | 22:41 |
| clarkb | lifeless: nova pci passthrough folks | 22:42 |
| jeblair | cody-somerville: and it would be nicer if that account were a bot | 22:42 |
| fungi | lifeless: the nova pci passthrough | 22:42 |
| fungi | lifeless: thread on the -dev ml | 22:42 |
| *** hogepodge has quit IRC | 22:43 | |
| mordred | cody-somerville: is there a reason to not transfer founder status to the bot? | 22:43 |
| *** mrodden has quit IRC | 22:45 | |
| cody-somerville | mordred: apparently we can't get the bot in #storyboard until we add jeblair as founder but ChanServ won't let it happen since jeblair has so many channels registered to his name. Was just suggesting founder status wasn't that important to stated goal and +f (ability to change the access list) would accomplish it w/o the limit. | 22:45 |
| mordred | cody-somerville: hah. fascinating | 22:45 |
| lifeless | fungi: clarkb: will look tomorrow | 22:45 |
| pleia2 | I found a freenode staffer, hopefully he can quickly toggle a bit ;) | 22:46 |
| lifeless | erm Monday | 22:46 |
| lifeless | clarkb: fungi: 3rd party for now, once the RH region comes online we should be able to do two regions etc etc | 22:46 |
| lifeless | clarkb: fungi: of course, we should also land the docs I sketched up about this ;) | 22:46 |
| fungi | lifeless: yeah, that basically parrots what i replied to the ml with. thanks! | 22:47 |
| *** ewindisch has quit IRC | 22:49 | |
| *** herndon has quit IRC | 22:50 | |
| *** UtahDave has joined #openstack-infra | 22:52 | |
| *** openstackinfra has joined #openstack-infra | 22:53 | |
| jeblair | openstackinfra: hi there! | 22:53 |
| jeblair | i registered openstackinfra so we can use it with the yet-to-be-written bot | 22:54 |
| jeblair | pleia2: if the op you found is responsive, maybe we can go ahead and bump the limit on openstackinfra, and maybe even start transferring founder status over | 22:54 |
| pleia2 | jeblair: ok, will do | 22:54 |
| *** hdd has quit IRC | 22:55 | |
| mordred | I am in favor of our new openstackinfra overlord | 22:56 |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/elastic-recheck: Add query for bug 1253896 https://review.openstack.org/58053 | 22:56 |
| uvirtbot | Launchpad bug 1253896 in tempest "test_minimum_basic_scenario fails with SSHException: Error reading SSH protocol banner" [Critical,Confirmed] https://launchpad.net/bugs/1253896 | 22:56 |
| *** bpokorny has quit IRC | 22:56 | |
| jeblair | clarkb, fungi, mordred: openstackinfra password added | 22:57 |
| fungi | awesome! | 22:57 |
| openstackgerrit | Elizabeth Krumbach Joseph proposed a change to openstack-infra/config: Add note about setting founder to IRC docs https://review.openstack.org/53925 | 22:58 |
| *** rcleere has quit IRC | 23:02 | |
| pleia2 | anyone know who the freenode GC (group contact) is for openstack? | 23:04 |
| *** sdake_ has quit IRC | 23:04 | |
| *** andreaf has quit IRC | 23:06 | |
| *** dcramer_ has quit IRC | 23:07 | |
| *** ^d has quit IRC | 23:07 | |
| *** datsun180b has quit IRC | 23:09 | |
| *** enqae has quit IRC | 23:09 | |
| jeblair | pleia2: there is none, they don't accept new groups | 23:10 |
| jeblair | pleia2: that's one of the difficulties with this | 23:10 |
| pleia2 | x_x | 23:11 |
| pleia2 | thanks | 23:11 |
| *** andreaf has joined #openstack-infra | 23:11 | |
| zaro | clarkb: about your comment in https://bugs.launchpad.net/openstack-ci/+bug/1203198 , did we have that discussion during summing? if so maybe updat this? | 23:11 |
| uvirtbot | Launchpad bug 1203198 in openstack-ci "Mirror jobs should not copy new packages if installation fails." [High,Triaged] | 23:11 |
| *** ewindisch has joined #openstack-infra | 23:12 | |
| *** sarob_ has quit IRC | 23:13 | |
| thingee | jog0: did reviews with stale gate starting jobs get started back up again? I saw a mention on the ML that they would once things were resolved. | 23:19 |
| *** dolphm_ has joined #openstack-infra | 23:20 | |
| *** dolphm has quit IRC | 23:20 | |
| *** ruhe has joined #openstack-infra | 23:20 | |
| *** dolphm_ has quit IRC | 23:20 | |
| *** denis_makogon has quit IRC | 23:20 | |
| *** rockyg has quit IRC | 23:21 | |
| *** gyee has joined #openstack-infra | 23:22 | |
| jeblair | thingee: if it hasn't merged and isn't on the zuul status page, you may want to leave a 'reverify no bug' comment | 23:23 |
| thingee | jeblair: thanks. Just noting in case there was something ran to catch these. | 23:24 |
| *** odyssey4me3 has joined #openstack-infra | 23:28 | |
| *** openstackinfra has quit IRC | 23:29 | |
| *** gyee has quit IRC | 23:34 | |
| *** ruhe has quit IRC | 23:35 | |
| *** gyee has joined #openstack-infra | 23:37 | |
| *** ewindisch has quit IRC | 23:44 | |
| *** flaper87 is now known as flaper87|afk | 23:46 | |
| *** ewindisch has joined #openstack-infra | 23:50 | |
| *** ewindisch has quit IRC | 23:51 | |
| *** odyssey4me3 has quit IRC | 23:54 | |
| *** UtahDave has quit IRC | 23:55 | |
| clarkb | mordred: fungi can you take a look at the bug zaro pointed out? I wasn't in that design session ( went to a different one ) | 23:56 |
| *** otherwiseguy has quit IRC | 23:57 | |
| *** Ryan_Lane has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!