Monday, 2014-08-25

*** harlowja_at_home has joined #openstack-infra00:00
clarkblifeless: oh and the worker allocation race00:00
clarkblifeless: that one should actually be simple to fix I would imagine? its where testr loses a worker bceause its gives two workers the same id number00:00
lifelessclarkb: I need more data on that one - bug 1316858133114000:02
lifelessclarkb: I need more data on that one - bug 131685800:02
uvirtbotLaunchpad bug 1316858 in testrepository "testr reports wrong work # on tests some times" [High,Triaged]
clarkblifeless: that is a case of that bug00:02
clarkblifeless: worker 1 there is the one that raced00:03
clarkbwe should have 4 workers but only 3 were recorded and the first one ran twice as many tests as it shoudl've so it was the one that got confused00:03
*** otherwiseguy has quit IRC00:05
lifelessclarkb: were there four pids?00:06
*** otherwiseguy has joined #openstack-infra00:06
clarkblifeless: yes, if you scroll up to where testr started there are four load lists00:07
*** ramashri_ has quit IRC00:08
*** otter768 has joined #openstack-infra00:08
*** annegentle has joined #openstack-infra00:10
openstackgerritBrant Knudson proposed a change to openstack-infra/devstack-gate: Stash Apache httpd logs
lifelessclarkb: so the thing that flummoxes me here00:15
lifelessclarkb: we use enumerate00:15
lifelessclarkb: so its kindof like saying 'when enumerate fails...'00:15
lifelessclarkb: and this is the tag logic00:16
lifeless                case = testtools.DecorateTestCaseResult(case,00:16
lifeless                    lambda result:testtools.StreamTagger(00:16
lifeless                        [result], add=['worker-%d' % pos]))00:16
lifelesswhere pos is from enumerate()00:16
lifelessI mean, obviously its happening00:19
lifelessbut wtf00:20
*** ramashri has joined #openstack-infra00:20
*** unicell has joined #openstack-infra00:20
lifelessclarkb: does it happen on every run ?00:20
clarkbit is very common yes00:21
*** otherwiseguy has quit IRC00:23
*** mmedvede has quit IRC00:23
*** otherwiseguy has joined #openstack-infra00:25
*** tsg has quit IRC00:28
*** tgohad has quit IRC00:29
*** otherwiseguy has quit IRC00:30
*** tgohad has joined #openstack-infra00:30
*** tsg has joined #openstack-infra00:30
*** annegentle has quit IRC00:31
r1chardj0n3shm, the pip-missing-reqs experimental job seems to have disappeared from nova... display for my experimental check job (114122) doesn't list it00:38
r1chardj0n3sgate-nova-pip-missing-reqs that was added in to be precise...00:40
clarkbr1chardj0n3s: its running in the silent queue now so runs on every patchset00:40
r1chardj0n3sI'm not sure what that means ;) but if it's "silent" how do I see the result?00:41
r1chardj0n3sah, I see the silent queue now, sorry00:41
r1chardj0n3s(I'm still not sure how to see the result of it though(00:41
clarkbr1chardj0n3s: the url there is
lifelessRun: 344 in 6.78047 sec.00:43
lifelessthats bogus - wall clock is 0.3s :)00:44
*** ominakov has quit IRC00:44
lifelessyou can't sum concurrent work!00:44
*** nosnos has joined #openstack-infra00:45
*** dkehn has quit IRC00:46
r1chardj0n3sthanks clarkb00:46
*** mmaglana has joined #openstack-infra00:47
*** harlowja_at_home has quit IRC00:47
*** annegentle has joined #openstack-infra00:50
*** tomoe_ has joined #openstack-infra00:53
*** annegentle has quit IRC00:54
openstackgerritDavid Pursehouse proposed a change to openstack-infra/jenkins-job-builder: SCM: rename `self` arguments to `parser`
*** yaguang has joined #openstack-infra00:56
*** asettle has joined #openstack-infra00:56
*** alexandra_ is now known as asettle00:57
ianwlifeless / clarkb: catching up on openstack-dev.  who knew calling setUp() twice was so common.  if you know of outstanding things i can help with it00:58
*** nosnos has quit IRC00:58
lifelessianw: I'm not aware of anything but thanks00:59
lifelessianw: I should have been more cautious I guess and warned rather than errored on day 101:00
*** nosnos has joined #openstack-infra01:01
*** annegentle has joined #openstack-infra01:02
clarkbr1chardj0n3s: I think we were hoping that once you were reasonably happy with it we could roll it into another test01:04
clarkbr1chardj0n3s: I don't know that it needs to be a dedicated test but may make sense as part of the existing requirements check or as part of the "pep8" job which runs static analysis01:04
r1chardj0n3sclarkb: one of the issues is it has to be very careful about virtualenv pollution from test tools01:06
clarkbr1chardj0n3s: all of the test tools should be part of the requirements though01:06
clarkbr1chardj0n3s: and if that is the case the existing requirements check may be best for it as that isn't in a test venv iirc01:07
r1chardj0n3sclarkb: but those requirements aren't installed as part of the application01:07
clarkboh I see what you mean. Basically the difference between requirements.txt and test-requirements.txt01:08
*** otter768 has quit IRC01:08
clarkbI think we can accomodate that in the existing requirements check job01:08
*** annegentle has quit IRC01:09
*** tiantian has joined #openstack-infra01:11
lifelesssdague: btw - the 'pythonlogging:'' name refers to the root logger - its nothing to do with subunit01:13
openstackgerritSteve Baker proposed a change to openstack-infra/config: check-heat-dsvm-functional pre/post hooks
lifelesssdague: you're throwing away actual information with your transform in subunit-trace.py01:13
*** tiantian has left #openstack-infra01:15
lifelessyayaya is green01:15
lifelessnow I just need two reviews from infra||oslo folk on it01:15
*** masayukig has quit IRC01:16
r1chardj0n3ssorry clarkb I was in a meeting there ;)01:18
*** nosnos has quit IRC01:19
*** signed8bit has joined #openstack-infra01:20
r1chardj0n3sso I guess the issue is that if we turn on pip-missing-reqs for everyone automagically, rather than project-by-project (and my preference, as the person doing it, is for the former ;) that does mean that some projects will just randomly start getting fails in the requirements check for random changes01:20
*** signed8bit has quit IRC01:20
r1chardj0n3s... unless nova is the only project that was actually failing ;)01:20
lifelessok thats interesting01:20
lifelessI can trigger01:21
lifeless - WARNING: missing Worker 0! Race in testr accounting.01:21
lifeless - WARNING: missing Worker 1! Race in testr accounting.01:21
lifeless - WARNING: missing Worker 2! Race in testr accounting.01:21
lifeless - Worker 3 (344 tests) => 0:00:01.939864s01:21
lifelessevery time by evaluating the iterator greedily01:21
clarkbr1chardj0n3s: yes, but we only run that check on requirements changes so it should make it relatively easy to clean up01:21
*** masayukig has joined #openstack-infra01:21
clarkbr1chardj0n3s: so everything won't turn red01:21
*** dimtruck is now known as zz_dimtruck01:21
r1chardj0n3sclarkb: ok, that sounds good. the actual mechanism is a deep dark scary secret from my perspective though, so how should we proceed?01:22
*** nosnos has joined #openstack-infra01:22
clarkbr1chardj0n3s: we update to run your thing01:24
clarkbr1chardj0n3s: then remove it from nova silent queue01:24
r1chardj0n3sclarkb: and the change pending against nova is replaced by one that just fixes nova's requirements.txt01:24
*** nosnos has quit IRC01:24
r1chardj0n3sclarkb: jhesketh (my tame infra person) is on leave at the moment, so it's going to be a couple off weeks before I can coordinate with him to get that all done01:26
clarkbr1chardj0n3s: I am happy to help though I have to run to dinner shortly01:27
clarkbso won't happen tonight :)01:27
clarkbr1chardj0n3s: but I figure since I am awake well into AU work hours why not01:27
r1chardj0n3sclarkb: ok, cool thanks. tomorrow is fine :)01:28
*** harlowja_at_home has joined #openstack-infra01:30
lifelessoh ffs half-ass lambdas in python01:30
lifelessof *course* thats what it is01:30
clarkblifeless: I am very curious to see the diff on the bugfix :)01:31
lifelessI haven't 100% verified01:33
lifelesslet me finish that before half-cocking it01:33
*** harlowja_at_home has quit IRC01:36
*** harlowja_at_home has joined #openstack-infra01:37
lifelessthe variable used to tag01:40
lifelessis being looked up in the containing context at execution time01:40
lifelessnot as at definition time01:40
lifelesse.g. its binding by label, not value01:40
*** armax has joined #openstack-infra01:44
*** yamahata has joined #openstack-infra01:45
lifelessclarkb: see trunk01:47
*** nosnos has joined #openstack-infra01:51
*** oomichi has joined #openstack-infra01:52
*** armax has quit IRC01:52
*** r1chardj0n3s is now known as r1chardj0n3s_afk01:54
*** nosnos has quit IRC01:58
*** nosnos has joined #openstack-infra01:58
*** Vega has joined #openstack-infra01:58
*** Vega is now known as Guest5690201:59
*** adalbas has quit IRC02:04
lifelessclarkb: do you have an example of bug 1331140 for me?02:06
uvirtbotLaunchpad bug 1331140 in testrepository "testr reports double failure counts" [Undecided,Incomplete]
*** esker has quit IRC02:09
*** annegentle has joined #openstack-infra02:10
*** rick_ has joined #openstack-infra02:12
*** annegentle has quit IRC02:15
*** xchu has joined #openstack-infra02:16
clarkbI couldnt find one02:16
*** annegentle has joined #openstack-infra02:18
rick_clarkb: HI02:20
*** xchu has quit IRC02:21
rick_clarkb: I have problem about how to debug jenkins publisher? I config the console log publisher, but I did not see any console log output.02:22
*** harlowja_at_home has quit IRC02:23
*** SumitNaiksatam has quit IRC02:28
*** SumitNaiksatam has joined #openstack-infra02:29
*** sdake_ has joined #openstack-infra02:30
*** r1chardj0n3s_afk is now known as r1chardj0n3s02:31
*** tgohad has quit IRC02:33
*** tsg has quit IRC02:33
*** Guest56902 has quit IRC02:36
*** otherwiseguy has joined #openstack-infra02:40
*** dims has quit IRC02:44
charzhi guys, I sent the request of third party service account one week ago. Is that ready?02:46
pleia2charz: what name/company? I can check to see if there were any responses02:47
charzpleia2: swiftstack-cluster-ci/SwiftStack02:47
*** dims has joined #openstack-infra02:47
charzpleia2: thanks :)02:47
*** dims has quit IRC02:47
pleia2charz: not seeing updates yet, anteaya was conferencing late last week so might just need catch up there, or SergeyLukjanov will get to it :)02:48
charzpleia2: So I need to talk with SergeyLukjanov? or send some info to him?02:50
*** KanagarajM has joined #openstack-infra02:51
pleia2charz: it'll be in a queue of accounts to set up, sorry it takes so long02:53
*** KanagarajM has quit IRC02:54
charzpleia2: It's ok, just make sure the status of account and I can schedule my tasks.02:56
charzplars: Thanks for your help. :)02:58
charzs/plars/pleia2 , sorry02:59
pleia2you're welcome03:01
*** baoli has quit IRC03:03
lifelessclarkb: if youhavetime03:04
*** hogepodge has quit IRC03:06
*** dkehn has joined #openstack-infra03:07
*** annegentle has quit IRC03:14
*** zz_dimtruck is now known as dimtruck03:18
*** Ryan_Lane has quit IRC03:25
*** sdake_ has quit IRC03:26
*** dimtruck is now known as zz_dimtruck03:28
*** otter768 has joined #openstack-infra03:28
*** Ryan_Lane has joined #openstack-infra03:34
*** _buzz has quit IRC03:40
*** ppai has joined #openstack-infra03:40
*** _buzz has joined #openstack-infra03:41
*** rick_ has quit IRC03:43
*** rick_ has joined #openstack-infra03:44
*** harlowja_at_home has joined #openstack-infra03:44
*** lttrl has quit IRC03:47
*** dims has joined #openstack-infra03:48
*** sdake_ has joined #openstack-infra03:52
*** gokrokve has joined #openstack-infra03:53
*** dims has quit IRC03:53
*** tsg has joined #openstack-infra03:59
*** tgohad has joined #openstack-infra03:59
*** Alexei_987 has joined #openstack-infra03:59
*** stevemar has quit IRC04:02
*** Ryan_Lane has quit IRC04:02
*** otter768 has quit IRC04:07
*** gokrokve has quit IRC04:10
*** sdake_ has quit IRC04:10
*** cnesa7 has quit IRC04:14
*** gokrokve has joined #openstack-infra04:15
*** Ryan_Lane has joined #openstack-infra04:21
*** harlowja_at_home has quit IRC04:22
*** loki184 has joined #openstack-infra04:26
*** gokrokve has quit IRC04:27
*** gokrokve has joined #openstack-infra04:28
*** daya_k has joined #openstack-infra04:28
*** amotoki has joined #openstack-infra04:29
*** yfried has quit IRC04:31
*** nosnos has quit IRC04:32
*** gokrokve has quit IRC04:32
*** nosnos has joined #openstack-infra04:33
*** ramashri has quit IRC04:35
*** thingee has quit IRC04:35
*** Ryan_Lane has quit IRC04:37
*** thingee has joined #openstack-infra04:37
*** torandu has quit IRC04:39
*** YorikSar has quit IRC04:41
*** YorikSar has joined #openstack-infra04:42
*** dims has joined #openstack-infra04:53
*** Ryan_Lane has joined #openstack-infra04:54
*** fabo has quit IRC04:56
*** YorikSar has quit IRC04:57
*** fabo has joined #openstack-infra04:57
*** dims has quit IRC04:59
*** otherwiseguy has quit IRC05:03
*** tgohad has quit IRC05:05
*** tsg has quit IRC05:05
*** flaper87|afk is now known as flaper8705:10
*** ociuhandu has quit IRC05:11
*** zz_dimtruck is now known as dimtruck05:19
*** Ryan_Lane1 has joined #openstack-infra05:25
*** dimtruck is now known as zz_dimtruck05:29
*** yfried has joined #openstack-infra05:35
*** Ryan_Lane has quit IRC05:38
*** _nadya_ has joined #openstack-infra05:45
*** HeOS has quit IRC05:46
*** _nadya_ has quit IRC05:49
*** miqui has quit IRC06:00
*** markmc has joined #openstack-infra06:03
*** k4n0 has joined #openstack-infra06:05
*** _nadya_ has joined #openstack-infra06:09
*** isaacb has joined #openstack-infra06:10
*** afazekas has joined #openstack-infra06:11
rick_Does anyone know how to convert Console Output of Jenkins to html file?06:13
*** MaxV has joined #openstack-infra06:18
*** cnesa7 has joined #openstack-infra06:20
*** trinaths has joined #openstack-infra06:22
*** dtantsur|afk is now known as dtantsur06:23
*** dizquierdo has joined #openstack-infra06:27
*** _nadya_ has quit IRC06:28
openstackgerritafazekas proposed a change to openstack-infra/devstack-gate: Fix ssh key specification in the README.rst
*** trinaths_afk has joined #openstack-infra06:34
*** YorikSar has joined #openstack-infra06:36
*** trinaths has quit IRC06:36
*** trinaths_afk is now known as trinaths06:36
*** daya_k has quit IRC06:38
*** dims has joined #openstack-infra06:42
*** dims has quit IRC06:47
jamielennoxi just pushed a tag to zuul for release, how long before i expect to see this in pypi? is there a log i can use to see the build?06:58
*** _nadya_ has joined #openstack-infra06:59
*** jlibosva has joined #openstack-infra06:59
*** _nadya_ has quit IRC07:04
davidlenwellfungi: earthquake was nothing07:06
*** jcoufal has joined #openstack-infra07:12
*** Longgeek has joined #openstack-infra07:16
*** _nadya_ has joined #openstack-infra07:17
*** HeOS has joined #openstack-infra07:17
*** zz_dimtruck is now known as dimtruck07:18
*** marun has quit IRC07:20
*** dizquierdo has quit IRC07:20
ttxjroll: they don't need to have code proposed to be marked started, just that the dev has started work on it07:23
flaper87SergeyLukjanov: hey, when you're around, mind +A ? :)07:23
ttxjroll: and Lucas said they gad people working on it07:23
ttxjroll: see
*** ihrachyshka has joined #openstack-infra07:24
*** flaper87 is now known as flaper87|afk07:26
*** ildikov has joined #openstack-infra07:27
*** dimtruck is now known as zz_dimtruck07:28
*** yolanda has joined #openstack-infra07:31
*** bookwar has joined #openstack-infra07:32
*** jamielennox is now known as jamielennox|away07:33
*** baohua has joined #openstack-infra07:33
*** praneshp has quit IRC07:34
*** luqas has joined #openstack-infra07:36
*** daya_k has joined #openstack-infra07:46
*** doude has joined #openstack-infra07:47
*** ihrachyshka has quit IRC07:48
*** Guest87807 has joined #openstack-infra07:48
*** amuller has joined #openstack-infra07:53
lifelessjamielennox|away: it triggers a job, visible in the zuul status page07:57
lifelessjamielennox|away: I don't think there is a review object to use to tie into that though :(07:57
*** jpich has joined #openstack-infra07:57
*** jistr has joined #openstack-infra07:59
*** IvanBerezovskiy has joined #openstack-infra07:59
*** dizquierdo has joined #openstack-infra07:59
*** pblaho has joined #openstack-infra08:05
*** loki184 has quit IRC08:07
*** derekh has joined #openstack-infra08:09
*** ociuhandu has joined #openstack-infra08:09
*** yfried_ has joined #openstack-infra08:14
*** tomoe_ has quit IRC08:14
*** yfried has quit IRC08:15
*** tomoe_ has joined #openstack-infra08:15
*** mflobo has joined #openstack-infra08:20
*** oanufriev has quit IRC08:24
*** enikanorov has quit IRC08:26
*** enikanorov has joined #openstack-infra08:26
*** _nadya_ has quit IRC08:27
yolandafungi, clarkb, jeblair, were you able to apply nodepool dib changes?08:27
*** yfried_ has quit IRC08:29
*** dims has joined #openstack-infra08:31
*** loki184 has joined #openstack-infra08:32
*** ihrachyshka has joined #openstack-infra08:33
*** yfried_ has joined #openstack-infra08:34
*** dims has quit IRC08:36
*** oanufriev has joined #openstack-infra08:36
*** k4n0 has quit IRC08:39
*** ihrachyshka has quit IRC08:48
*** ihrachyshka has joined #openstack-infra08:48
*** yfried_ has quit IRC08:50
*** k4n0 has joined #openstack-infra08:53
*** KanagarajM has joined #openstack-infra08:55
*** ihrachyshka has quit IRC08:56
*** rlandy has joined #openstack-infra08:59
*** yfried has joined #openstack-infra09:02
SergeyLukjanovflaper87|afk, approved09:05
*** mkollaro has joined #openstack-infra09:05
SergeyLukjanovcharz, pleia2, I hope I'll have some time today to create accounts09:06
*** mkollaro1 has joined #openstack-infra09:09
*** mkollaro has quit IRC09:09
*** Guest87807 has quit IRC09:09
*** rcarrillocruz has joined #openstack-infra09:10
*** Hal_ has joined #openstack-infra09:11
*** Hal_ has quit IRC09:13
*** Hal_ has joined #openstack-infra09:14
*** rcarrillocruz has quit IRC09:21
*** dmellado has quit IRC09:23
*** marun has joined #openstack-infra09:24
*** rcarrillocruz has joined #openstack-infra09:25
*** penguinRaider has quit IRC09:27
*** andreykurilin has joined #openstack-infra09:29
*** hashar has joined #openstack-infra09:29
*** ZZelle has joined #openstack-infra09:31
*** oanufriev has quit IRC09:38
*** oanufriev has joined #openstack-infra09:39
*** oanufriev has quit IRC09:41
*** sressot has quit IRC09:42
*** baohua has quit IRC09:42
*** dtantsur is now known as dtantsur|lunch09:43
*** penguinRaider has joined #openstack-infra09:43
*** yamahata has quit IRC09:43
*** ppai has quit IRC09:50
openstackgerritA change was merged to openstack-infra/release-tools: Use double-bracket style everywhere in rc*.sh
*** sressot has joined #openstack-infra09:50
*** rcarrillocruz has quit IRC09:51
*** pelix has joined #openstack-infra09:52
*** luqas has quit IRC09:52
*** luqas has joined #openstack-infra09:54
*** luqas has quit IRC09:58
*** rcarrillocruz has joined #openstack-infra09:58
*** pblaho is now known as pblaho|afk10:01
*** oanufriev has joined #openstack-infra10:01
openstackgerrityolanda.robla proposed a change to openstack-infra/zuul: Add check for ref being a string before applying regex
*** pcm_ has joined #openstack-infra10:05
*** e0ne has joined #openstack-infra10:05
*** jp_at_hp has joined #openstack-infra10:06
*** luqas has joined #openstack-infra10:06
*** ppai has joined #openstack-infra10:06
*** Hal_ has quit IRC10:08
*** e0ne has quit IRC10:08
*** pcm_ has quit IRC10:09
*** pcm_ has joined #openstack-infra10:10
*** marun has quit IRC10:11
*** e0ne has joined #openstack-infra10:12
*** marun has joined #openstack-infra10:12
*** tnurlygayanov has joined #openstack-infra10:15
*** e0ne has quit IRC10:15
openstackgerritA change was merged to openstack-infra/devstack-gate: Add to the list of projects
*** e0ne has joined #openstack-infra10:20
*** unicell has quit IRC10:21
*** e0ne has quit IRC10:21
*** p5ntangle has joined #openstack-infra10:22
*** e0ne has joined #openstack-infra10:26
*** e0ne has quit IRC10:28
*** hashar has quit IRC10:29
*** andreykurilin has quit IRC10:32
*** e0ne has joined #openstack-infra10:32
*** andreykurilin has joined #openstack-infra10:32
*** p5ntangl_ has joined #openstack-infra10:33
*** p5ntangle has quit IRC10:36
*** e0ne has quit IRC10:39
*** rick_ has quit IRC10:41
*** e0ne has joined #openstack-infra10:43
*** rcarrillocruz has quit IRC10:43
*** e0ne has quit IRC10:44
*** p5ntangl_ has quit IRC10:46
*** tomoe__ has joined #openstack-infra10:46
*** p5ntangle has joined #openstack-infra10:46
*** k4n0 has quit IRC10:47
*** rcarrillocruz has joined #openstack-infra10:47
*** flaper87|afk is now known as flaper8710:48
*** e0ne has joined #openstack-infra10:48
*** tomoe_ has quit IRC10:49
*** tomoe__ has quit IRC10:50
*** e0ne has quit IRC10:53
*** e0ne has joined #openstack-infra10:56
*** rcarrillocruz has left #openstack-infra10:56
*** rcarrillocruz has joined #openstack-infra10:57
*** mmaglana has quit IRC10:59
*** k4n0 has joined #openstack-infra10:59
*** pcm_ has quit IRC11:01
*** pcm_ has joined #openstack-infra11:03
*** e0ne has quit IRC11:03
*** pblaho|afk is now known as pblaho11:04
*** pcm_ has quit IRC11:04
*** tomoe_ has joined #openstack-infra11:05
*** pcm_ has joined #openstack-infra11:05
*** e0ne has joined #openstack-infra11:07
*** e0ne has quit IRC11:09
*** luqas has quit IRC11:12
*** penguinRaider has quit IRC11:12
*** e0ne has joined #openstack-infra11:12
*** e0ne has quit IRC11:15
*** daya_k has quit IRC11:15
*** yamahata has joined #openstack-infra11:16
*** e0ne has joined #openstack-infra11:19
*** _nadya_ has joined #openstack-infra11:19
*** mkollaro has joined #openstack-infra11:22
*** [HeOS] has joined #openstack-infra11:22
*** andreykurilin has quit IRC11:22
*** HeOS has quit IRC11:23
*** pblaho is now known as pblaho|afk11:24
*** dteselkin has quit IRC11:24
*** mkollaro1 has quit IRC11:24
*** dteselkin has joined #openstack-infra11:25
*** adalbas has joined #openstack-infra11:25
*** Hal_ has joined #openstack-infra11:27
*** e0ne has quit IRC11:27
*** dims has joined #openstack-infra11:29
*** e0ne has joined #openstack-infra11:32
*** omrim has joined #openstack-infra11:32
*** Hal_ has quit IRC11:32
*** ominakov has joined #openstack-infra11:32
*** rcarrillocruz has quit IRC11:33
*** Hal_ has joined #openstack-infra11:35
*** rcarrillocruz has joined #openstack-infra11:35
*** hashar has joined #openstack-infra11:36
*** omrim has quit IRC11:36
*** omr has joined #openstack-infra11:37
omrHi, I notice that from time to time I have "Merge Failed" on some commits in my CI setup.11:37
omrIn spite of the "Merge Failed" I saw the results of the tests.11:38
*** ppai has quit IRC11:39
omrIs someone know what is the root-cause for such a failed?11:39
*** pblaho|afk is now known as pblaho11:40
*** loki184 has quit IRC11:40
*** dizquierdo has quit IRC11:44
*** flaviof_zzz is now known as flaviof11:44
*** KanagarajM has quit IRC11:49
*** e0ne has quit IRC11:49
*** ppai has joined #openstack-infra11:51
joaomr: sorry, Didn't encounter that yet11:51
*** [HeOS] is now known as HeOS11:51
*** e0ne has joined #openstack-infra11:54
*** pblaho is now known as pblaho|afk11:57
*** rfolco has joined #openstack-infra11:58
*** e0ne has quit IRC11:58
*** dizquierdo has joined #openstack-infra11:58
*** e0ne has joined #openstack-infra12:02
*** MaxV has quit IRC12:03
*** flaper87 is now known as flaper87|afk12:04
*** baoli has joined #openstack-infra12:05
*** tomoe_ has quit IRC12:06
*** dkranz has joined #openstack-infra12:06
*** ihrachyshka has joined #openstack-infra12:07
*** flaper87|afk is now known as flaper8712:07
*** mbacchi has joined #openstack-infra12:09
*** e0ne has quit IRC12:09
*** ihrachyshka has quit IRC12:13
*** ihrachyshka has joined #openstack-infra12:14
*** mkollaro has quit IRC12:14
*** mkollaro1 has joined #openstack-infra12:14
omrjoa: Have you ever seen the Exception:  "ERROR zuul.Merger: Unable to reset repo <zuul.merger.merger.Repo object at>" from your merger.log12:15
joadon't think so, but lemme grep my meager logs12:15
*** aysyd has joined #openstack-infra12:16
omrjoa: Thanks12:16
joanope, none in my logs12:16
joawell, my CI doesn't run full-time, as it's not properly configured yet, missing some tweaks12:16
*** kgiusti has joined #openstack-infra12:20
*** ryanpetrello has quit IRC12:21
*** jaypipes has joined #openstack-infra12:22
*** e0ne has joined #openstack-infra12:24
*** ryanpetrello has joined #openstack-infra12:26
*** e0ne has quit IRC12:28
*** zz_dimtruck is now known as dimtruck12:29
sdaguemorning folks12:30
joamorning sdague12:30
*** weshay has joined #openstack-infra12:34
*** ominakov has quit IRC12:36
openstackgerritCedric Brandily proposed a change to openstack-infra/git-review: Prefer git-config over git-review config files
marunsdague: quick question, why is lockutils required to run tests? (seen in tox.ini)12:41
sdaguemarun: I don't know12:42
marunsdague: Hmmm, any thoughts on who would?12:42
sdaguelifeless? you are asking about the single test invocation, right?12:43
*** rcarrillocruz has quit IRC12:43
*** pblaho|afk is now known as pblaho12:45
*** pdmars has joined #openstack-infra12:46
marunsdague: I think it wraps all unit test invocations actually12:47
marunsdague: thanks for the pointer, I'll ping lifeless then.12:47
sdaguepossibly a thing that testtools should be importing that it doesn't?12:47
*** rcarrillocruz has joined #openstack-infra12:47
sdagueyeh, I don't know, that magic invocation came from somewhere else and I just carried it forward because it worked12:48
marunsdague: we do the same thing in neutron12:48
sdaguedoes it work without it?12:48
sdagueit's not in other projects12:49
marunsdague: ah, now I see12:49
fungiyolanda: nodepool now has the dib changes applied (plus some fixes for bugs they introduced) running in production now. should be completely in sync with the tip of nodepool's master branch at this point12:49
openstackgerritA change was merged to openstack/requirements: Rename bash8 to bashate (part 2)
marunsdague: It ensures a temp directory exists for locks so that any code that uses lockutils will work when invoked in a test.12:50
yolandafungi, that's great news12:51
*** dkliban_ has joined #openstack-infra12:51
sdagueoh, that's just to trigger the explode early if it doesnt?12:51
marunsdague: sorry for bugging you instead of just looking, but now you know something too. :)12:51
sdaguecool, learning something new is always good :)12:51
yolandafungi, are you using dib for building images now, or still using old approach?12:51
fungiyolanda: still using the traditional method for the moment12:52
yolandafungi, any blocker for not using dib?12:53
fungiyolanda: no idea. probably just someone's time to make sure it's working for our images in our various providers12:55
*** dkliban_ has quit IRC12:55
*** ppai has quit IRC12:56
*** sdake has quit IRC12:56
yolandafungi, ok, just let me know if there is something i can help with. I wrote also some changes for documentation:
*** mmedvede has joined #openstack-infra12:58
*** rcarrill` has joined #openstack-infra12:58
*** sdake has joined #openstack-infra12:58
*** _nadya_ has quit IRC13:00
*** ildikov_ has joined #openstack-infra13:01
*** rcarrillocruz has quit IRC13:01
*** ildikov has quit IRC13:01
*** dtantsur|lunch is now known as dtantsur13:03
*** hashar has quit IRC13:04
*** hashar has joined #openstack-infra13:04
*** _nadya_ has joined #openstack-infra13:05
*** dims has quit IRC13:06
*** dims has joined #openstack-infra13:06
*** miqui has joined #openstack-infra13:08
*** _nadya_ has quit IRC13:08
*** mmedvede has quit IRC13:09
*** mestery has quit IRC13:10
openstackgerritKurt Taylor proposed a change to openstack-infra/config: Change wording and typos in third_party.rst
*** ominakov has joined #openstack-infra13:12
*** mmedvede has joined #openstack-infra13:12
*** p5ntangle has quit IRC13:16
*** mestery has joined #openstack-infra13:17
*** dustins has joined #openstack-infra13:17
*** dizquierdo has quit IRC13:18
*** bknudson has quit IRC13:18
*** chandankumar has joined #openstack-infra13:20
*** mestery_ has joined #openstack-infra13:21
*** mestery has quit IRC13:22
*** mriedem has joined #openstack-infra13:24
*** trinaths has quit IRC13:26
*** Hal_ has quit IRC13:29
*** Hal_ has joined #openstack-infra13:29
*** dizquierdo has joined #openstack-infra13:31
*** stevemar has joined #openstack-infra13:31
*** rcarrill` has quit IRC13:33
*** rcarrillocruz has joined #openstack-infra13:33
*** dprince has joined #openstack-infra13:35
*** k4n0 has quit IRC13:35
mtreinishsdague, marun: heh, we did that manually in the tempest unit tests with:
marunmtreinish: ah, a great opportunity to remove some code then :)13:36
*** pcrews has joined #openstack-infra13:36
mtreinishmarun: yeah, probably13:37
*** bknudson has joined #openstack-infra13:37
*** nosnos has quit IRC13:38
*** wenlock has joined #openstack-infra13:38
*** nosnos has joined #openstack-infra13:38
*** sabeen has joined #openstack-infra13:40
*** jgrimm has quit IRC13:40
*** nosnos has quit IRC13:43
*** sballe has joined #openstack-infra13:43
*** chandankumar has quit IRC13:46
*** armax has joined #openstack-infra13:50
jrollttx: great, thanks :)13:51
*** jpeeler has joined #openstack-infra13:51
*** jpeeler has quit IRC13:51
*** jpeeler has joined #openstack-infra13:51
*** homeless has joined #openstack-infra13:52
* anteaya is home and very very happy to be so13:53
*** eharney has joined #openstack-infra13:53
*** dizquierdo has quit IRC13:54
sdagueanteaya: I'm with you on that :)13:54
flaper87lifeless: hey, could you take a look here whenever you've a chance?
anteayagood bye noises of elevators, ice machines and fire trucks, hello birds motorboats, hammers and cranes13:56
flaper87lifeless: not sure why that's happening but I remember seeing something w.r.t semver in the mailing list. It seems I can't replicate it locally13:56
Hal_Hi guys, do you know if there is any issue with git review ? I have the following message when I do a “git review” :
anteayacharz: your account has been created: 13052swiftstack-cluster-ciSwiftStack Cluster cinerama openstack-ci@swiftstack.com13:58
charzanteaya: Thanks, I got the mail.13:58
anteayacharz: great13:58
anteayahi Hal_13:58
Hal_anteaya: hi13:59
anteayaHal_: did you read the error message in the paste?13:59
fungiHal_: it looks like the account you're connecting with to upload your patch hasn't submitted contact info.
Hal_I submitted an email13:59
Hal_I am unable to submit more (error on submission)13:59
anteayaHal_: an email to where?13:59
anteayahey fungi14:00
Hal_anteaya: I do not understand14:00
openstackgerritCedric Brandily proposed a change to openstack-infra/git-review: Prefer git-config over git-review config files
Hal_I have full name and preferred email registered14:00
fungiHal_: that means it probably doesn't match your foundation profile. make sure you followed the instructions in order at
*** lucasagomes has joined #openstack-infra14:01
anteayafungi: what terribe things have I done or not done that I should attend to first?14:01
fungianteaya: nothing i know of14:01
anteayafungi: okay, thanks one down14:01
*** radez_g0n3 is now known as radez14:02
lucasagomeshi all, please some core in tempest can take a look at ? This is the last patch of a series and is blocking a blueprint in Ironic to be marked as Implemented. It's fairly small patch just adding some tests14:02
Hal_anteaya: I cannot find accurate agreement, I signed ICLA, the two others are not accurate (System CLA and US G CLA)14:02
Hal_I checked with someone else setup, he have the same agreement (ICLA) and only email in contact info and it works fine14:02
fungiHal_: did you join the openstack foundation as an individual member?14:03
fungiHal_: per the instructions i linked, you need to do that first, and make sure the primary e-mail address on your foundation member profile matches the preferred e-mail address you set in your gettit settings14:03
*** dmsimard_away is now known as dmsimard14:03
*** julim has joined #openstack-infra14:04
fungier, matches the preferred e-mail address you set in your gerrit settings14:04
Hal_fungi: is it a new requierement ?14:04
Hal_I was able to submit review for infra without all this administrative stuff14:04
fungiHal_: yeah, we've only been requiring that for anout 1.5 years, so fairly new depending on your definition of new14:04
fungiHal_: oh, some of the infra projects dfo not require a cla to submit changes14:04
Hal_fungi: ok14:05
*** dizquierdo has joined #openstack-infra14:06
*** chuckC has quit IRC14:07
krtaylorhi anteaya14:07
anteayahey krtaylor14:08
krtayloranteaya, for today's meeting, do I need to chair?14:08
anteayayou don't need to, you are welcome to if you would like, your choice14:08
Hal_fungi: I joined and set the preferred email address according to the one I submitted via the application form. Do I have to wait some minutes ?14:08
krtayloranteaya, I had an action to figure out how to change the cell background to red/yellow/green, but failed14:09
krtayloranteaya, for CI status, we have that hack for the broken type14:09
anteayakrtaylor: okay14:09
krtayloranteaya, but I cannot figure out how to pass color to the template14:09
fungiHal_: no, once you're able to successfully set address information at you ought to be able to push your patch for review14:09
anteayakrtaylor: what hack for what broken type? do you want to just explain in the meeting?14:10
Hal_fungi:  I still receive the same message :-(14:10
krtayloranteaya, it irritates me14:10
anteayakrtaylor: I can understand that14:10
fungiHal_: which same message?14:10
sdagueerw: were the docker tests all passing before it got upgraded from experimental -> silent?14:10
Hal_the same as
krtayloranteaya, yeah, I'll mention it there14:11
*** nelsnelson has joined #openstack-infra14:11
fungiHal_: did you successfully submit address information at yet?14:11
krtayloranteaya, I'll let you drive again this week, its nice to have the break :)14:11
Hal_Yes, the preferred email is set14:12
fungiHal_: street address information, not just e-mail address14:12
*** oomichi has quit IRC14:12
*** che-arne|2 is now known as che-arne14:12
sdagueerw: it looks like it's been 100% fail for 5 days14:13
*** signed8bit has joined #openstack-infra14:13
Hal_fungi: thanks. With street info it is working (why people I know do not need to fill this contact form ?)14:13
anteayakrtaylor: I understand, I'll drive this week, thanks for touching base14:13
fungiHal_: at you should see a line above the mailing address field like "Contact information last updated on August 25, 2014 at 2:10 PM." or something14:13
*** amotoki has quit IRC14:14
anteayakrtaylor: hey you are going to have to take some holidays sometime, once you have them scheduled let me know so I am sure I am able to cover meetings14:14
fungiHal_: explanation is at
openstackgerritDarragh Bailey proposed a change to openstack-infra/jenkins-job-builder: Create one parameters wrapper tag for all release params
Hal_fungi: Yes, I was able to update theses detailed contact info14:14
*** doug-fish has joined #openstack-infra14:14
Hal_fungi: thanks for the tips. Indeed, contact info are not shown once they have been provided ;)14:15
*** david-lyle has joined #openstack-infra14:17
Hal_fungi: anteaya thanks you very much for your help. I had to read the procedure more carefully (and as I was able once to submit a change, I thought I had already done everything already). Sorry14:17
krtayloranteaya, I have already taken the bulk of my vacation, the only exception is the week of Christmas I'll have off14:18
fungiHal_: no problem--sorry for the complications and thanks for the contribution!14:18
anteayaHal_: np, glad you are on your way14:18
*** ArxCruz has joined #openstack-infra14:18
*** otherwiseguy has joined #openstack-infra14:19
anteayakrtaylor: great, yes we will dicuss what the group wants to do over December holidays, we have some time for that14:19
*** wenlock has quit IRC14:19
*** yamahata has quit IRC14:22
*** mfer has joined #openstack-infra14:23
*** david-lyle has quit IRC14:23
*** david-lyle has joined #openstack-infra14:23
*** radez is now known as radez_g0n314:25
*** rcarrillocruz has quit IRC14:26
*** zz_gondoi is now known as gondoi14:27
*** enikanorov__ has joined #openstack-infra14:27
*** jcoufal_ has joined #openstack-infra14:28
*** Ng_ has joined #openstack-infra14:29
*** cnesa8 has joined #openstack-infra14:29
*** simonmcc___ has joined #openstack-infra14:29
*** funzo_ has joined #openstack-infra14:29
anteayaokay pleia2 when you are around I need to learn all the mailing list admin things14:29
*** jraim__ has joined #openstack-infra14:29
*** rcarrillocruz has joined #openstack-infra14:30
*** crinkle_ has joined #openstack-infra14:30
*** dguerri_ has joined #openstack-infra14:30
*** spiffxp- has joined #openstack-infra14:31
*** vipuls has joined #openstack-infra14:31
*** tristanC_ has joined #openstack-infra14:31
*** electrichead has joined #openstack-infra14:31
*** vladan_ has joined #openstack-infra14:31
*** dkliban1 has joined #openstack-infra14:31
erwsdague: I’ll take a look. It was passing consistently when I submitted the flip from experimental to silnet14:31
*** pleia2_ has joined #openstack-infra14:31
*** soren_ has joined #openstack-infra14:32
*** mgagne_ has joined #openstack-infra14:32
sdagueyeh, last pass was on Aug 20th14:32
*** cinerama` has joined #openstack-infra14:32
*** Clabbe has quit IRC14:32
*** sarob has quit IRC14:32
*** mgagne has quit IRC14:32
*** dimtruck has quit IRC14:32
*** dolphm has quit IRC14:32
*** dguerri has quit IRC14:32
*** vipul has quit IRC14:32
*** vladan has quit IRC14:32
*** freyes has quit IRC14:32
*** alaski has quit IRC14:32
*** dizquierdo has quit IRC14:32
*** jcoufal has quit IRC14:32
*** dkliban has quit IRC14:32
*** flaviof has quit IRC14:32
*** dtroyer has quit IRC14:32
*** cinerama has quit IRC14:32
*** lbragstad has quit IRC14:32
*** jraim has quit IRC14:32
*** Ng has quit IRC14:32
*** tristanC has quit IRC14:32
*** spiffxp has quit IRC14:32
*** cnesa7 has quit IRC14:32
*** funzo has quit IRC14:32
*** mtreinish has quit IRC14:32
*** simonmcc__ has quit IRC14:32
*** maurosr has quit IRC14:32
*** enikanorov_ has quit IRC14:32
*** jroll has quit IRC14:32
*** msuriar has quit IRC14:32
*** pleia2 has quit IRC14:32
*** JayF has quit IRC14:32
*** soren has quit IRC14:32
*** russellb has quit IRC14:32
*** therve has quit IRC14:32
*** redrobot has quit IRC14:32
*** jbryce has quit IRC14:32
*** crinkle has quit IRC14:32
*** Clabbe has joined #openstack-infra14:32
*** lbragstad has joined #openstack-infra14:32
*** dolphm has joined #openstack-infra14:32
*** mtreinish has joined #openstack-infra14:32
*** dimtruck has joined #openstack-infra14:32
*** spiffxp- is now known as spiffxp14:32
*** vipuls is now known as vipul14:32
sdagueso now it's just chewing up nodes and failing14:32
*** maurosr- has joined #openstack-infra14:32
*** dtroyer has joined #openstack-infra14:32
*** jbryce has joined #openstack-infra14:32
*** JayF has joined #openstack-infra14:32
*** dizquierdo has joined #openstack-infra14:32
*** maurosr- is now known as maurosr14:32
*** tristanC_ is now known as tristanC14:33
erwsdague: but I’ve had some other, more urgent issues (yes, truly) since the 20th. That’s lightening up a bit now.14:33
*** therve has joined #openstack-infra14:33
*** msuriar has joined #openstack-infra14:33
erwsdague: how do I view the silent checks?14:33
sdagueok, cool, no prob14:33
sdagueI was using logstash14:33
*** Ng_ is now known as Ng14:33
*** alaski has joined #openstack-infra14:33
*** jroll has joined #openstack-infra14:33
*** freyes has joined #openstack-infra14:33
*** russellb has joined #openstack-infra14:33
sdaguespecify build_name: AND message:"Finished"14:34
*** vladan_ is now known as vladan14:34
*** jgrimm has joined #openstack-infra14:34
sdagueI forget what the build name for the docker one is at the moment14:34
sdagueI was just wondering what was up with the long silent queue, then saw that docker is failing all the time14:34
*** simonmcc___ is now known as simonmcc__14:35
*** gargola has joined #openstack-infra14:36
*** tonytan4ever has joined #openstack-infra14:37
openstackgerritMatt Riedemann proposed a change to openstack/requirements: Bump minimum required version for simplejson to 2.2.0
mriedembknudson: ^14:39
mriedemdims: ^14:40
*** erlon has joined #openstack-infra14:40
*** erlon has quit IRC14:40
*** erlon has joined #openstack-infra14:40
*** hashar has quit IRC14:44
*** medieval1 has joined #openstack-infra14:44
*** ihrachyshka has quit IRC14:44
*** hashar has joined #openstack-infra14:44
*** datsun180b has joined #openstack-infra14:49
*** mmedvede has quit IRC14:49
*** bodepd_ has joined #openstack-infra14:49
*** mtreinish_ has joined #openstack-infra14:50
*** wenlock has joined #openstack-infra14:50
*** markmc_ has joined #openstack-infra14:50
*** Clabbe has quit IRC14:50
*** lbragstad has quit IRC14:50
*** bodepd has quit IRC14:50
*** jgrimm has quit IRC14:50
*** alaski has quit IRC14:50
*** dizquierdo has quit IRC14:50
*** JayF has quit IRC14:50
*** jbryce has quit IRC14:50
*** dtroyer has quit IRC14:50
*** dolphm has quit IRC14:50
*** dimtruck has quit IRC14:50
*** mtreinish has quit IRC14:50
*** markmc has quit IRC14:50
*** dimtruck has joined #openstack-infra14:50
*** Clabbe has joined #openstack-infra14:50
*** lbragstad has joined #openstack-infra14:50
*** dtroyer has joined #openstack-infra14:50
*** jbryce has joined #openstack-infra14:50
*** dolphm has joined #openstack-infra14:50
*** mmedvede_ has joined #openstack-infra14:50
*** alaski has joined #openstack-infra14:50
*** JasonF has joined #openstack-infra14:50
*** jgrimm has joined #openstack-infra14:50
*** mtreinish_ is now known as mtreinish14:51
*** dizquierdo has joined #openstack-infra14:51
*** HeOS has quit IRC14:51
*** rcarrillocruz has quit IRC14:52
sdaguemriedem: +A14:53
mriedemsdague: thanks, i wasn't sure about the gr-base.txt file14:54
sdagueit doesn't really matter14:54
sdagueit's just a test file, unless the particular lib is looked for in the tests, it won't really mater either way14:55
*** tsg has joined #openstack-infra14:55
*** tgohad has joined #openstack-infra14:55
*** ildikov has joined #openstack-infra14:55
*** rcarrillocruz has joined #openstack-infra14:56
*** crinkle_ is now known as crinkle14:56
*** hdd has joined #openstack-infra14:58
*** ildikov_ has quit IRC14:58
*** chuckC has joined #openstack-infra14:58
*** gokrokve has joined #openstack-infra14:59
*** pballand has joined #openstack-infra14:59
*** isaacb has quit IRC14:59
erwsdague: Nova arbitrarily changing driver interfaces, of course… I see the problem (or one problem), shouldn’t be hard to fix.15:00
*** annegentle has joined #openstack-infra15:00
ZZellewhois clarkb,15:01
sdagueerw: cool15:01
openstackgerritFrançois Magimel proposed a change to openstack-infra/config: Make py33 job non-voting for cloudkitty-dashboard
sdagueerw: it would be worth doing what the ironic folks did with their contract unit test in nova15:01
fungiZZelle: clarkb is clark boylan15:02
sdaguethen it would be clear which parts of the driver interface docker depends on15:02
fungiZZelle: or did you mean to /whois instead?15:02
ZZellefungi, hi, don't make fun of me :)15:02
fungiZZelle: wasn't making fun--sorry... thought it might be a legitimate question you had15:03
*** zns has joined #openstack-infra15:03
phschwartzMorning infra15:03
openstackgerrityolanda.robla proposed a change to openstack-infra/nodepool: Update documentation for using diskimage-builder
fungiphschwartz: morning--did your weekend investigations bear fruit?15:03
ZZellefungi, just checking if he was here to ask you and clarkb to look at gitreview changes15:03
*** emagana has joined #openstack-infra15:03
fungiZZelle: yeah, i don't do a very good job of visiting open changes for the git-review repo, thanks for the reminder. so many projects15:04
*** electrichead is now known as redrobot15:04
fungiZZelle: clarkb usually shows up in here around 1700 utc15:04
marunfungi: easy speedup of the neutron functional job here: :)15:05
ZZellefungi, i understand it's not really the most active gerrit project15:05
marunclarkb, jeblair: ^^15:05
*** pradk has joined #openstack-infra15:06
*** mestery_ is now known as mestery15:06
*** _buzz has quit IRC15:07
*** emagana has quit IRC15:08
erwsdague: contract unit test?15:08
*** emagana has joined #openstack-infra15:08
*** p5ntangle has joined #openstack-infra15:08
erwsdague: basically introduce a unit test into nova that screams if it breaks methods nova-docker depends on?15:09
erw(and then at least people know they’re breaking it?)15:09
*** mikedillion has joined #openstack-infra15:09
sdagueerw: yep, basically that15:09
*** atiwari has joined #openstack-infra15:10
erwsdague: interesting. thanks.15:10
*** markmcclain has joined #openstack-infra15:10
sdagueright, if you want to define a behavior, write a test for it. People can still change it if they need to, but at least they will know they are breaking yuo15:10
sdagueand it's a simple file to watch in review15:10
erwsdague: fwiw, this is the first time this actually happened since Havana15:10
erw(as far as I know)15:10
erwsdague: amazing, huh?15:11
sdaguethat surprises me15:11
*** lttrl has joined #openstack-infra15:11
*** unicell has joined #openstack-infra15:11
openstackgerritDarragh Bailey proposed a change to openstack-infra/jenkins-job-builder: Create one parameters wrapper tag for all release params
*** emagana has quit IRC15:13
erwsdague: I’ve had people run backports of the driver and they did have to make some changes, but I’ve been told they’ve been minimal — and usually for things like oslo.config, never (to my knowledge) the actual driver interface.15:13
*** radez_g0n3 is now known as radez15:13
*** yaguang has quit IRC15:14
sdagueerw: yeh, I don't know, but regardless expressing the interface bits you depend on as a contract is handy15:16
erwsdague: oh, agreed.15:16
*** Longgeek has quit IRC15:16
*** skolekonov has joined #openstack-infra15:17
*** markmcclain has quit IRC15:19
*** p5ntangle has quit IRC15:19
*** blitzrage is now known as leifmadsen15:19
*** markmcclain has joined #openstack-infra15:21
*** dtantsur is now known as dtantsur|afk15:23
*** unicell has quit IRC15:23
EmilienMhey infra, could someone from core have a look at - current puppet gate is broken and this patch would help us and fix it.15:23
jeblairEmilienM: and bundle will install the right ver of puppet from gems?15:26
fungiEmilienM: do the gate-{name}-puppet-unit-{puppet_version} jobs only run on ubuntu (no centos/fedora?)15:26
openstackgerrityolanda.robla proposed a change to openstack-infra/config: Add support for disk-image-builder in nodepool
openstackgerrityolanda.robla proposed a change to openstack-infra/config: Align with prepare-node in dib
openstackgerrityolanda.robla proposed a change to openstack-infra/config: Add pre-requisites needed for diskimage-builder
openstackgerrityolanda.robla proposed a change to openstack-infra/config: Add tools to hack on infra with Docker
sbadiaEmilienM: yep15:27
*** atiwari has quit IRC15:27
EmilienMfungi: only Ubuntu, yes.15:27
EmilienMjeblair: checking to be sure..15:28
*** p5ntangle has joined #openstack-infra15:28
*** coolsvap has quit IRC15:30
*** ildikov has quit IRC15:30
*** _buzz has joined #openstack-infra15:30
*** yfried_ has joined #openstack-infra15:31
*** mmaglana has joined #openstack-infra15:32
*** yfried has quit IRC15:33
*** trinaths has joined #openstack-infra15:33
*** amotoki has joined #openstack-infra15:33
*** linkid has joined #openstack-infra15:35
*** gokrokve has quit IRC15:35
EmilienMjeblair: just checked with sbadia, it's managed by puppet modules gemfile:
*** p5ntangle has quit IRC15:35
*** sballe_ has joined #openstack-infra15:35
*** mmaglana_ has joined #openstack-infra15:36
*** yfried_ has quit IRC15:36
sbadiaEmilienM jeblair yep (and
EmilienMgreat, thanks15:37
*** sballe has quit IRC15:37
*** mmaglana has quit IRC15:39
*** coolsvap has joined #openstack-infra15:42
EmilienMthanks fungi & jeblair btw15:43
sbadiayep, thanks!15:43
anteayaprotip: don't pick a fight with a bear, the bear will win15:44
*** atiwari has joined #openstack-infra15:44
anteayaneighbour just got bit, and he was completely surprised15:44
*** rcarrillocruz has quit IRC15:45
anteayaI'm just hoping they don't have to put the bear down, it is a good bear15:45
*** masayukig has quit IRC15:46
joaanteaya: wat ? :D are you living near a bear-housing forest ?15:46
*** markmcclain has quit IRC15:46
*** andreykurilin has joined #openstack-infra15:46
anteayajoa: we have had a great bear in the neighbourhood for about the last 7 years15:47
anteayawhen he sees me he runs away15:47
anteayathe neighbours have all been great about staying clean around their houses so there is nothing to attract the bear15:47
joayou aren't that scary though.15:48
anteayait is good to have a well trained bear around15:48
anteayaI'm fairly scary15:48
anteayaask other third party vendors15:48
joaahah :D15:48
*** rcarrillocruz has joined #openstack-infra15:49
*** hemna_ is now known as hemna15:50
*** flaper87 is now known as flaper87|afk15:52
*** masayukig has joined #openstack-infra15:52
stevemarbnemec, ping15:53
stevemarsometimes i really enjoy 'walking in' on these conversations, anteaya, you just don't hear enough bear stories nowadays15:54
anteayastevemar: I can give you some great bear stories, but this one is more a foolish neighbour story15:55
anteayahow he thought fighting a bear was a good choice is beyond me15:55
bnemecstevemar: pong15:55
stevemarbnemec, so about adding that file to oslotest (, i'm still not seeing how it'll be consumable by other projects?15:56
stevemari guess the other projects would update their tox.ini to point to oslotest?15:57
*** UtahDave has joined #openstack-infra15:57
dhellmannstevemar: the idea is if the script is in oslotest, it would be installed with oslotest and could be run like any other normal script15:58
dhellmannstevemar: instead of having different copies of it in every project, sourced from the incubator15:59
bnemecstevemar: Something like this:
* dhellmann nods15:59
bnemecThat way it will get installed in a reasonable location with oslotest.15:59
dhellmannwe probably need a more specific name for the script to avoid installation conflicts, but that's a relatively small thing15:59
openstackgerritGregory Haynes proposed a change to openstack-infra/config: Chown the repos for tripleo-ha job
*** Hal_ has quit IRC16:00
greghaynesclarkb: others ^16:00
greghaynesfixes a failing tripleo jerb16:00
*** gokrokve has joined #openstack-infra16:01
bnemecgreghaynes: So we can expect a 100% pass rate on the ha job after that? ;-)16:01
greghaynesbnemec: You know it!16:01
*** sballe_ has quit IRC16:01
*** sballe_ has joined #openstack-infra16:01
greghaynesderekh: psh, you always beat me to these things16:03
derekhgreghaynes: that change that requires that has been revert, but we should probably do it anyways for consistency16:03
derekhgreghaynes: :-)16:03
*** juzuluag has joined #openstack-infra16:03
phschwartzfungi: no, not any yet16:03
*** annegentle has quit IRC16:04
openstackgerritDarragh Bailey proposed a change to openstack-infra/jenkins-job-builder: Some tweaks to get closer to Python 3 compat
openstackgerritDarragh Bailey proposed a change to openstack-infra/jenkins-job-builder: Ensure dict orders are deterministic
*** hdd has quit IRC16:05
*** zns has quit IRC16:05
*** skolekonov has quit IRC16:05
stevemardhellmann, bnemec ahhh ok, so if i update setup.cfg with the script name, it'll be made available for other projects16:07
dhellmannstevemar: as long as they include oslotest in their test-requirementst.xt16:07
stevemarand in tox.ini, we can just reference it outright16:07
stevemardhellmann, well why didnt you just say that ! :P16:07
*** alkari has quit IRC16:07
dhellmannstevemar: sorry, I thought I was :-)16:07
greghaynesderekh: what is your goodsquishy url for tripleo jobs?16:08
*** otter768 has joined #openstack-infra16:08
dhellmannstevemar: keep in mind, the script may be installed "globally" (probably not, but could happen) so we need a name that's less generic for the script itself16:08
*** zns has joined #openstack-infra16:08
stevemardhellmann, any suggestions? ?16:09
dhellmannstevemar: maybe or or something?16:09
stevemardoude, sounds good to me, if something goes wrong, they'll go to mr. oslo :P16:09
stevemardhellmann, *16:09
dhellmannstevemar: makes sense16:10
derekhgreghaynes: that chown was needed for a bit   ( befor this merged
*** pblaho has quit IRC16:10
openstackgerritA change was merged to openstack-infra/config: Uninstall puppet before running puppet jobs
*** zns has quit IRC16:11
*** zns has joined #openstack-infra16:11
*** ramashri has joined #openstack-infra16:12
markvanfungi: Hi, that issue about rax nodes failing and hpcloud ones working is still happening.16:12
markvanI look at the,   did not see much to go on, but it seem it be cut off midway thru "L" packages.16:12
*** dane_leblanc has joined #openstack-infra16:13
markvanfungi: is there a reason we can't just get the autoconf tools installed on all nodes to allow native C extensions for ruby (and other languages) to compile properly?16:13
*** markmcclain has joined #openstack-infra16:15
*** pballand has quit IRC16:17
*** omr has quit IRC16:17
fungimarkvan: ahh, yes, it seems i'm discovering more and more that lodgeit on limits the content size16:18
fungimarkvan: it would be good to identify why your job is trying to run autoconf in one cloud provider and not another. if we just paper over this by installing it on them, then you're going to end up exercising different code paths depending on where it runs16:19
*** praneshp has joined #openstack-infra16:19
*** markmcclain has quit IRC16:20
markvanfungi: ok.  well, I really don't know if it's not running autoconf on the other nodes, the output is not verbose enough when it works.16:20
fungimarkvan: seems to work for this file size, so
markvanok, will take a look.16:21
fungimarkvan: i have to assume it's not running autoconf on the other nodes because i checked and it's really not installed and in the path even where the job is succeeding16:21
markvanI'll see about enabling more verbose or logging with bundler install, that should help some16:22
fungimarkvan: yes, i'll be happy to help try and interpret the debug info16:22
openstackgerritA change was merged to openstack-infra/jenkins-job-builder: SCM: rename `self` arguments to `parser`
*** dims_ has joined #openstack-infra16:23
markvanfungi: ok.  I'll review that list to see if anything stands out.  then try to add some debug, but I fear this might be a level or 2 lower then bundler itself and it just calls make to do the compiles.16:24
*** dims has quit IRC16:24
pleia2_good morning16:24
*** pleia2_ is now known as pleia216:24
*** _nadya_ has joined #openstack-infra16:26
*** radez is now known as radez_g0n316:26
*** amotoki has quit IRC16:27
fungimarkvan: right, the main reason i was focusing on the installed packages difference is that the build process arbitrarily deciding to build bindings to external libraries in some places and not others suggests that it's dependent on which libraries are installed16:28
openstackgerritMark Vanderwiel proposed a change to openstack-infra/config: Fix gate job timing issues with bundler
markvanfungi: good point.   Added verbose to this patch:  if you care to review it, that should help with tracking this down.16:30
*** emagana has joined #openstack-infra16:30
*** markmc_ has quit IRC16:30
*** loki184 has joined #openstack-infra16:30
*** Ryan_Lane has joined #openstack-infra16:32
openstackgerritDarragh Bailey proposed a change to openstack-infra/jenkins-job-builder: Added recursive option
*** pballand has joined #openstack-infra16:33
openstackgerritDarragh Bailey proposed a change to openstack-infra/jenkins-job-builder: Added recursive option
*** jistr has quit IRC16:33
*** mmaglana_ has quit IRC16:34
*** ildikov has joined #openstack-infra16:34
*** dangers_away is now known as dangers16:35
*** dims has joined #openstack-infra16:35
*** dims_ has quit IRC16:35
*** Ryan_Lane has quit IRC16:36
*** sabeen has quit IRC16:36
*** markmcclain has joined #openstack-infra16:37
*** juzuluag has quit IRC16:38
*** weshay has quit IRC16:41
*** loki184 has quit IRC16:45
*** weshay has joined #openstack-infra16:46
*** radez_g0n3 is now known as radez16:47
JasonFjeblair: I think jroll cleared up your concern with if you don't mind looking back over it16:47
*** hashar has quit IRC16:50
jeblairJasonF: ack16:50
*** alkari has joined #openstack-infra16:51
*** signed8bit is now known as signed8bit_ZZZzz16:51
*** dizquierdo has quit IRC16:51
*** dimtruck is now known as zz_dimtruck16:51
*** mmaglana has joined #openstack-infra16:52
*** sparkycollier has joined #openstack-infra16:53
*** afazekas has quit IRC16:53
*** bookwar has quit IRC16:53
*** sdake_ has joined #openstack-infra16:54
*** bcrochet has quit IRC16:55
openstackgerritJames E. Blair proposed a change to stackforge/gertty: Add a configurable keymap
openstackgerritJames E. Blair proposed a change to stackforge/gertty: Remove stray debug line
*** bhuvan has joined #openstack-infra16:56
*** marcoemorais has joined #openstack-infra16:56
*** signed8bit has joined #openstack-infra16:56
openstackgerritA change was merged to openstack-infra/jenkins-job-builder: Adding feature to specify the user id for jenkins user credentials
*** bcrochet has joined #openstack-infra16:57
jeblairJasonF: lgtm16:57
*** yamahata has joined #openstack-infra16:58
jeblairdhellmann: i added a 'light' palette to gertty (you can do '-p light', or set 'palette: light' in the config file). you can still add new palettes or alter individual colors in them in the config file, but i wanted to have an option for a good first experience for light background terminals16:58
*** baoli has quit IRC16:59
*** linkid has left #openstack-infra16:59
*** jlibosva has quit IRC16:59
jeblairdhellmann: if you have any local colors you'd like to incorporate into the builtin light palette, that would be quite welcome16:59
*** sarob has joined #openstack-infra16:59
*** sarob_ has joined #openstack-infra16:59
jeblairdhellmann: i sort of punted on a few things because i didn't really have good ideas; it all seemed so strange to me :)16:59
*** baoli has joined #openstack-infra17:00
jeblairdhellmann: (oh, these changes haven't merged yet; the tip of my branch is 116514)17:00
*** directxman12 has joined #openstack-infra17:00
clarkbo/ right at 170017:00
*** openstackgerrit has quit IRC17:00
jeblairclarkb: just in time for UTC beer17:01
*** marun is now known as marun_afk17:01
jrollJasonF: thanks for that poke17:01
JasonFthis is a fairly easy to review devstack-gate change, just needs at least one more core review :) http:////
* JasonF pokes jroll17:01
jrolljeblair: thanks for the review :)17:01
clarkbjeblair my secret is out17:01
*** harlowja has joined #openstack-infra17:01
jrollclarkb: mind looking at 112143? :)17:01
*** JasonF is now known as JayF17:02
directxman12hey, I have a question about the Nova blueprint use-libvirt-storage-pools (patch series starts here:  Namely, it requires some additional setup before tempest tests are run in order to pass (the initial storage pools have to be created).  How do I go about dealing with this?17:03
*** otter768 has quit IRC17:03
*** Hal_ has joined #openstack-infra17:03
clarkbjroll: see my comment17:06
*** zns has quit IRC17:06
clarkbdirectxman12: if tempest needs additional stuff devstack should probably set that up for it17:06
*** dane_leblanc has quit IRC17:07
*** derekh has quit IRC17:07
jrollclarkb: that gets set in lib/ironic, am I wrong to assume this should work?17:07
jrollclarkb: "agent_ssh" is not the default btw17:08
clarkbjroll: devstack gate runs before devstack so that variable will never be set if it comes from lib/ironic17:08
clarkbjroll: if you look in the script you will see that we default all of the DEVSTACK_GATE_* vars17:08
jrollactually, sec, I'm setting that somewhere else17:09
*** sparkycollier has quit IRC17:09
*** dane_leblanc has joined #openstack-infra17:09
jrollclarkb: so infra/config sets that17:10
*** rcarrillocruz has quit IRC17:10
jrollline 524 here
*** mgagne_ has quit IRC17:11
*** mgagne_ has joined #openstack-infra17:11
*** zns has joined #openstack-infra17:11
clarkbjroll: ok so you should just set the default then17:11
clarkbin the script17:12
*** rcarrillocruz has joined #openstack-infra17:12
jeblair#status log restarted gerritbot17:12
openstackstatusjeblair: finished logging17:12
*** jpich has quit IRC17:12
directxman12clarkb: would devstack accept a patch that did setup for a different patch that hasn't been merged yet?  It's kind of a chicken and egg problem, so I figured I'd ask here17:13
*** Ryan_Lane has joined #openstack-infra17:13
*** Ryan_Lane has quit IRC17:13
*** Ryan_Lane has joined #openstack-infra17:13
*** openstackgerrit has joined #openstack-infra17:13
clarkbdirectxman12: if the setup is necessary for a piece of functionality I don't see why not but I am not a devstack core17:13
jrollclarkb: ok, one sec17:13
jeblairi uploaded new versions of the zuul cloner changes17:13
directxman12clarkb: ok, thanks17:13
clarkbjeblair: ok I will try to rereview them17:13
clarkbfungi: yolanda I think we can use dib for the devstack images but the bare-* images need more work17:14
*** lucasagomes is now known as lucas-dinner17:14
clarkbfungi: yolanda so maybe starting with devstack images is a good next step17:14
markvanfungi: went thru the package list, don't see anything obvious in the lib diffs.  Also reviewed the logs again, but nothing obvious there either.  The other patch to add --verbose to bundler install is going in now, hopefully that will shed some light on this.17:15
ZZelleclarkb, jeblair, hi17:15
clarkbgreghaynes: you asked for review but there are two changes for that change. can we abandon one?17:15
*** mikedill_ has joined #openstack-infra17:15
clarkbZZelle: hellol17:15
ZZelleclarkb, jeblair, if you have some time, could you have a look to gitreview changes in gerrit?17:15
greghaynesclarkb: I did17:15
clarkbgreghaynes: oh derp I am blind. thanks17:15
greghaynesty if you +A the other :)17:16
openstackgerritJim Rollenhagen proposed a change to openstack-infra/devstack-gate: Add Ironic-python-agent support to devstack-gate
openstackgerritA change was merged to openstack-infra/jenkins-job-builder: Apply defaults to job-templates parameters
jrollclarkb, jeblair, one last try :) ^17:16
*** _nadya_ has quit IRC17:16
*** andreykurilin has quit IRC17:16
*** yfried_ has joined #openstack-infra17:16
*** mikedillion has quit IRC17:19
clarkbZZelle: I don't mind reviewing them but I disagree with a lot of them :) I figured it was better to let people actually interested in changes look17:21
*** yamahata has quit IRC17:21
*** tonytan4ever has quit IRC17:21
clarkbZZelle: in particular I don't understand why there should be a need for that. If you want code review (and thus submit) then submit immediately seems like the wrong thing to support in git review. As an alternative if review isn't required why not let people push directly?17:21
*** pballand has quit IRC17:22
clarkbif we are going to circumvent the tooling lets not use the tool to do the reviews17:22
dhellmannjeblair: thanks, I'll take a look. I may have special color requirements, since I had to eliminate the greens, too, because of a special sensitivity I have.17:22
*** flaviof_zzz has joined #openstack-infra17:22
*** Hal_ has quit IRC17:23
fungii agree on that one--it seems to be dragging further away from git-review as a tool to simplify contributing and veering into git-review as a convenience wrapper for project admins17:23
jeblairdhellmann: apparently there are 88 and 256 color terminals out there; i haven't added support in gertty for them yet, but with that, there may be other options, like greens with more blue or red in them17:23
jeblairdhellmann: urwid has a palette_test script if you want to see the colors17:24
clarkbjeblair: those 265 color terminals are amazing17:24
*** esmute has quit IRC17:24
jeblairurxvt supports 88 but not 256 apparently17:24
clarkbjeblair: there is a urxvt-256color :)17:24
*** markmcclain has quit IRC17:25
* fungi continues to get by with 8 colors at 2 intensities plus 8 background colors17:25
jeblairbut i limited myself to using 16 to make sure gertty had good universal support to start with17:25
jeblairclarkb: on neato17:25
clarkbjeblair: yup default to 16 is a great place to start17:25
dhellmannjeblair: this is what I'm using right now:
jeblairfungi: yeah, ==1617:25
fungiwhat's the standout option do?17:26
jeblairfungi: usually inverts fg/bg17:26
clarkbZZelle: the other one I need to wrap my head around is the config change. Not sure what the value is there, is it just simplification?17:26
*** otter768 has joined #openstack-infra17:26
*** bdpayne has joined #openstack-infra17:27
*** emagana has quit IRC17:28
*** emagana has joined #openstack-infra17:29
*** hogepodge has joined #openstack-infra17:29
*** yfried_ has quit IRC17:30
dhellmannjeblair: do you use a < 256 color terminal normally? or more? (my snark detector is borked this morning)17:31
clarkbjeblair: cloner stack lgtm at least as far as I got last week17:32
clarkbworking on the last change now17:32
*** a2hill has quit IRC17:32
*** emagana has quit IRC17:33
jeblairdhellmann: i normally run urxwt which supports 88, but i have not been using >16 with gertty because i wanted to design it to degrade gracefully (so not depend on anything > 16 by default)17:34
fungidhellmann: since i have the ansi color escapes memorized, i'd feel lost in a 256-color terminal17:34
dhellmannjeblair: ok, that makes sense17:35
*** yfried_ has joined #openstack-infra17:35
*** ramashri has quit IRC17:35
clarkbZZelle: or is the config functiaonlly wrong today?17:35
jeblairdhellmann: if you wanted to start playing with >16; the line to enable 88 is in, but just commented out17:35
*** ramashri has joined #openstack-infra17:35
dhellmannfungi: yeah, I have emacs set up with the solarized light theme, so I am on a 256 color terminal17:35
*** ramashri has quit IRC17:35
dhellmannjeblair: 16 colors should be enough for anyone, right? :-)17:36
*** ramashri has joined #openstack-infra17:36
* jeblair imagines his surroundings in 16 colors...17:36
* fungi still has frightening memories of swapping 4-color palettes on cga displays17:36
*** lttrl has quit IRC17:36
*** hogepodge has quit IRC17:37
*** esmute has joined #openstack-infra17:37
*** e0ne has joined #openstack-infra17:39
*** radez is now known as radez_g0n317:39
*** mgagne_ is now known as mgagne17:39
clarkbjeblair: just one small thing inline on the last change in the cloner stack17:40
*** tsg has quit IRC17:40
*** tgohad has quit IRC17:41
*** SumitNaiksatam has quit IRC17:41
clarkbfungi: jeblair: doesn't seem to have had any major opposition so I think I will go ahead and update that text today17:43
fungiclarkb: sounds good to me. i'm happy to +2 the community consensus there17:44
openstackgerritJames E. Blair proposed a change to openstack-infra/config: Switch to using single projects yaml job
openstackgerritJames E. Blair proposed a change to openstack-infra/config: Remove old projects yaml jobs
clarkbfungi: what do you think about dib'ing the devstack images?17:45
clarkbjeblair: ^ you too.17:46
fungiclarkb: um, i welcome our new glance-based overlords?17:46
jeblairclarkb: i think we have a small pile of changes to merge first, right?17:46
jeblairthe giant "use dib" patch among them, that i believe yolanda just updated this morning17:46
*** hogepodge has joined #openstack-infra17:46
clarkbjeblair: right, I think there is a bit of work that needs to go in for it. Mostly on the actually make dib work side17:47
ZZelleclarkb, it's not wrong, i would more say annoying17:47
fungijeblair: oh, btw i confirmed that unit tests since the image updates this weekend are indeed mentioning the image templates in the console logs now17:47
clarkbjeblair: mostly curious if you are ok with having two sets of image types for a while17:47
clarkbjeblair: or if we would prefer to do a bigger switch all at once17:47
*** ramashri has quit IRC17:48
*** gyee has joined #openstack-infra17:48
*** ryanpetrello has quit IRC17:48
*** dvorak has joined #openstack-infra17:48
ZZelleclarkb, ie, for users behind proxy you cannot overload scheme/port without updating every .gitreview17:48
*** sarob_ has quit IRC17:48
* anteaya adds her +1 to the 25 already on 11617617:48
*** Ajaeger has joined #openstack-infra17:48
anteayaI don't think I have ever seen so many +1's before17:49
jeblairclarkb: i think we should proceed, yes.  probably best to just do one type of image for now.17:49
ZZelleclarkb, is no more needed if we can change default gerrit remote name17:49
Ajaegeranteaya: agreed, this is impressive17:49
clarkbZZelle: I think mordred wanted git-review -d to just work without auth17:49
*** david-ly_ has joined #openstack-infra17:49
clarkbZZelle: and it would with that change iirc. So probably a couple use cases tied to it17:49
jrolljeblair, clarkb, mind looking at this again? added that variable to -wrap:
*** hdd has joined #openstack-infra17:50
ZZelleclarkb, nope, review.o.o only requires auth on push17:50
*** vhoward has left #openstack-infra17:50
*** Hal_ has joined #openstack-infra17:50
clarkbZZelle: it requires auth on both if the push and pull urls are both ssh urls17:50
dvorakanyone else tried mixing the use of the extended choice parameter plugin and normal parameters?  it looks like the extended choice plugin support was misplaced into the properties module17:51
clarkbZZelle: which is why treating them as different is important. You can do read only pull and ssh push that way17:51
dvorakand it's generating it's own XML block for parameters17:51
ZZelleclarkb, AH!, mordred is right, auth is required when using ssh (i am using https)17:51
ZZelleclarkb, right17:51
Ajaegerjeblair, clarkb: could you recheck , please? Clarkb you suggested it - and jeblair gives a -2. I'm happy to abandon but just like some extra confirmation...17:51
*** tonytan4ever has joined #openstack-infra17:51
clarkbAjaeger: jeblair: there was a reason we needed to sync iirc17:52
clarkbso while I agree with jeblair I think d-g may be a potential exception?17:52
dhellmannjeblair: if I wanted to write a thing to create an RSS feed of the eavesdrop meeting logs (not channel, just meeting), do you think it would be better to do that in the bot or in a cron job that processes the bot output daily?17:52
clarkbI am trying to remember what broke (if anything)17:52
clarkbdvorak: I don't think we have ever used it17:52
*** david-lyle has quit IRC17:53
jeblairdhellmann: i lean toward bot; i think it's amenable to that kind of thing17:53
jeblairclarkb, Ajaeger: i'm doing research17:53
dvorakclarkb: would a patch be welcome to fix it?  It'd probably involve moving all that code over to the parameters module, which would probably break existing extended choice users17:53
*** ryanpetrello has joined #openstack-infra17:53
dvorakI could duplicate the code and have the old code issue a deprecation warning if backwards compat is a concern17:53
*** bdpayne has quit IRC17:53
clarkbdvorak: yup patches very welcome and the deprecation warning method sounds good17:54
dhellmannjeblair: ok, that makes sense, I wasn't sure about concurrency issues but I suppose it deals with that already17:54
*** daya_k has joined #openstack-infra17:54
dvorakok, probably won't be too hard, I'll give it a shot.17:54
*** gokrokve_ has joined #openstack-infra17:54
*** Hal_ has quit IRC17:55
*** penguinRaider has joined #openstack-infra17:55
ZZelleclarkb, ok, the main advantage of preferring git-config over gitreview files is to allow to overload .gitreview without changing it (no sure we need to deprecate global/system gitreview config files)17:55
Ajaegerclarkb: what broke was: bashate was added to devstack-gate and that broke the requirments post job completely17:55
*** SumitNaiksatam has joined #openstack-infra17:55
AjaegerOnly once bashate was in requirements, it was fine again...17:55
clarkbAjaeger: thanks17:55
jeblairdhellmann: oh, one feed for all?  makes sense.  i don't have its concurrency model in my head right now, so i probably can't give you the advice you really need.  but i still think it's worth looking into the bot.17:55
*** bdpayne has joined #openstack-infra17:55
*** tsg has joined #openstack-infra17:55
*** tgohad has joined #openstack-infra17:55
dhellmannjeblair: yeah, or one per project, I guess, to eliminate the concurrency issue17:56
AjaegerSergeyLukjanov came up with
jeblairAjaeger, clarkb: here is where it was added:
dhellmannjeblair: I'll look at the bot code a bit when I have time; this is a background thing, obviously17:56
Ajaegerjeblair: thanks17:56
jeblairAjaeger, clarkb: i do not believe that devstack-gate itself having requirements was envisioned at the time17:57
Ajaegerjeblair: might be ;)17:57
jeblairAjaeger: also, we added bashate to requirements?17:57
*** tonytan4ever has quit IRC17:58
*** gokrokve has quit IRC17:58
Ajaegerjeblair: is the patch that broke it17:58
Ajaegerjeblair: yes, I cleaned up the bashate mess, it's now in requirements and bash8 was renamed everywhere to bashate17:58
*** gokrokve_ has quit IRC17:58
jeblairoh, bash8 was in requirements before, and you renamed it17:59
Ajaegerjeblair: bash8 was in requirements; bash8 got renamed to bashate; devstack-gate added bashate; everything broke - and then I renamed properly17:59
*** hogepodge has quit IRC17:59
Ajaeger(everything: requirements post job)17:59
*** esmute has quit IRC18:00
ZZelleclarkb, at first my idea was to allow overloading scheme/port using git-config but i realize it was already possible for rebase ... so i choose to allow to overload all .gitreview options18:00
jeblairAjaeger, clarkb: I think the solution is to remove devstack-gate from projects.txt in requirements18:00
clarkbZZelle: I see and this would be cases where setting http_proxy doesn't work/18:00
*** esmute has joined #openstack-infra18:00
clarkbjeblair: that works for me too18:01
jeblairAjaeger, clarkb: mordred added it so that select-mirror would dtrt; but we don't use select-mirror anymore, so it's not necessary18:01
*** hogepodge has joined #openstack-infra18:01
clarkbjeblair: agreed18:01
ZZelleclarkb, in my case, it's http_proxy not working with ssh :s18:01
AjaegerOk, I abandon my patch and propose a new one.18:01
*** markvan has quit IRC18:02
jeblairAjaeger: cool, thanks (for doing that, and finding the problem, and fixing it, and helping us understand what was going on... :)18:02
Ajaegerjeblair: speaking about cleanup up requirements: there's still and
Ajaegerjeblair: SergeyLukjanov was involved as well. This looked trivial initially...18:03
Ajaegerjeblair: could you check the docutils pin patches some time, please?18:03
*** markvan has joined #openstack-infra18:04
jeblairlooks like the horizon one is still outstanding18:04
Ajaegerjeblair: exactly, they want to have a spinx !=1.1.3 as part of the change...18:06
Ajaegerso, if that goes into requirements, they should be fine...18:06
openstackgerritAndreas Jaeger proposed a change to openstack/requirements: Remove devstack-gate from projects
jeblairwe used 1.1.3 for a long time18:07
*** _nadya_ has joined #openstack-infra18:08
*** amuller has quit IRC18:09
jeblairAjaeger: because sphinx 1.1.3 does not insist on the broken docutils.  sphinx 1.1.3 will work just fine now.18:11
clarkbjeblair: right I think there is a lot of confusion how our pins and things works18:11
Ajaegeryeah ;( So, some education for the horizon reviewers is needed...18:12
*** hogepodge has quit IRC18:12
*** _nadya_ has quit IRC18:13
*** gokrokve has joined #openstack-infra18:13
*** homeless has quit IRC18:14
* Ajaeger pings the blazar folks since they did not review the patch at all ...18:14
*** pballand has joined #openstack-infra18:14
jeblairAjaeger: would blazar block our ability to merge the removal?18:15
Ajaegerjeblair: blazar is in projects.txt of requirements repo18:15
*** emagana has joined #openstack-infra18:16
Ajaegerjeblair: So, I assume it blocks18:16
jeblairAjaeger: i hope the tests are configured so that only openstack projects block18:17
jeblairif not, we'll remove blazar if they're the last one standing18:17
jeblairdavid-ly_: can you lookt at ?  horizon is the last openstack project not to merge that change18:18
david-ly_jeblair: on it18:18
*** david-ly_ is now known as david-lyle18:18
Ajaegerjeblair: there's no special case for any kind of projects as far as I can see ;(18:19
*** e0ne has quit IRC18:19
*** _nadya_ has joined #openstack-infra18:19
*** e0ne has joined #openstack-infra18:20
openstackgerritJames E. Blair proposed a change to openstack-infra/zuul: Cloner: add project-specific branch overrides
jeblairyolanda, clarkb, fungi: i think my comment in 102583 is still relevant -- /opt/nodepool can't be the image build dir _and_ the vcsrepo dir at the same time18:22
*** vhoward has joined #openstack-infra18:24
*** hogepodge has joined #openstack-infra18:24
jeblairand the docker change still has unanswered questions; i'd love to have that in, but if we can't answer them, then we should probably move it out of the series18:24
clarkbjeblair: agreed, I think docker isn't an immediate thing we need. devstack image builds will work without it18:25
*** e0ne has quit IRC18:25
*** ociuhandu has quit IRC18:25
*** e0ne has joined #openstack-infra18:26
*** _nadya_ has quit IRC18:26
*** e0ne has quit IRC18:26
jeblairhrm.  the base of that series, 105931 has serious problems -- it looks like a rebase gone awry18:28
*** lttrl has joined #openstack-infra18:28
*** UtahDave has left #openstack-infra18:28
*** tonytan4ever has joined #openstack-infra18:29
*** emagana has quit IRC18:29
*** rloo has joined #openstack-infra18:30
clarkbjeblair: fungi: yesterday I was going to submit "Hudson Jenkins", "Gerrit Rietveld", and "Zuul Gozer" as the three most influential people in openstack. Then metacloud wanted irc nicks or twitter handles and only Gerrit has one :(18:30
*** _nadya_ has joined #openstack-infra18:30
rloohi, we've seen 2 gate-ironic-python26 failures with ConnectionError: HTTPConnectionPool.. errors (eg
*** dkliban1 is now known as dkliban18:31
rlooI'm not sure if that is related to this bug opened in Jan:
uvirtbotLaunchpad bug 1268725 in openstack-ci "ConnectionError: HTTPConnectionPool(host='', port=80): Max retries exceeded" [Medium,Triaged]18:32
*** hogepodge has quit IRC18:32
rlooor what the fix might be... Has anyone else noticed this?18:32
fungiclarkb: too bad. also, i'm thinking they may have a hard time coming up with 52 nominations when they only ask for three names18:32
*** emagana_ has joined #openstack-infra18:32
jeblairclarkb: or would "Zuul Gozerian" be more correct? :)  (they really don't accept people with a single name?)18:32
jeblairthat's just ...18:32
clarkbjeblair: you need first and last18:32
clarkband preferably a twitter handle18:33
*** _nadya_ has quit IRC18:33
jeblairgood point.  how could you be influential without a twitter handle?18:33
jeblairclarkb: also, i think we had an irc account for jenkins at some point.18:33
clarkbit was suggseted that zuul tweet merges18:33
fungirloo: we commonly see failures along those lines, because the internet is an unstable place and connections go astray with great frequency18:33
*** ominakov has quit IRC18:34
rloofungi: ahh. so 'recheck no bug'?18:34
jesusaurusyeah, but it looks like @TheGatekeeper is already taken18:34
fungirloo: or "recheck bug 1268725" or just "recheck"18:34
uvirtbotLaunchpad bug 1268725 in openstack-ci "ConnectionError: HTTPConnectionPool(host='', port=80): Max retries exceeded" [Medium,Triaged]
rloothx fungi.18:34
*** penguinRaider has quit IRC18:35
openstackgerritA change was merged to openstack-infra/config: Make jenkins own /opt/git in ha job
*** penguinRaider has joined #openstack-infra18:35
clarkbI am also happy to find out Asimov pronounced "robots" like Zoidberg18:35
clarkbso from nwo on I will call them rowbutts too18:36
jeblairclarkb: source?18:36
clarkbjeblair: its not the greatest source and I haven't been able to watch the tv show in question but
clarkbjeblair: its good enough for me beacuse this is awesome18:36
fungitruth is awesomer than cartoons18:37
*** hogepodge has joined #openstack-infra18:38
*** nelsnelson has quit IRC18:38
pleia2bug day tomorrow, sent reminder18:42
jeblairhrm.  we were supposed to have imported stuff into storyboard.  :/18:43
pleia2yeah, maybe one last LP bug day18:43
jeblairand here i thought the last one was the last one18:44
jeblairi'm pretty inclined to just copy the bugs we care about and shut it down.  or just declare bankruptcy and shut it down.18:44
*** ominakov has joined #openstack-infra18:44
*** marcusvrn has joined #openstack-infra18:44
pleia2could make tomorrow's bug day about doing the former18:44
fungiflaper87|afk: dims: dhellmann: lifeless: bug 1361251 sounds pretty severe?18:45
uvirtbotLaunchpad bug 1361251 in pbr "pbr is failing to parse versions" [Undecided,New]
dhellmannfungi, flaper87|afk: so far I think that's only affecting, isn't it?18:45
*** vhoward has left #openstack-infra18:46
* dhellmann scrolls down and sees otherwise18:46
fungidhellmann: it's the only one reported so far18:46
fungidhellmann: however we don't currently prevent building/uploading prereleases with those tags either18:47
fungidhellmann: so presumably we'll want to tighten the pre-release pipeline tag regex in our zul layout.yaml?18:47
*** doude has quit IRC18:47
fungier, zuul (sorry for the accidental ping, zul!)18:48
dhellmannfungi: I need to understand from lifeless why this tagging pattern is so special, but probably18:48
zulfungi:  its ok...i get a penny everytime someone pings me18:48
dhellmannlooks like that error failed a grenade test for swift, too18:48
*** doude has joined #openstack-infra18:48
fungizul: you must be a zullionaire by now18:48
zulfungi:  yes i just do the work for fun now18:48
yolandajeblair, just arrived, i'll take a look18:49
jogoI take it you guys are aware there doesn't seem to be enough/any f20 nodes18:50
*** hogepodge has quit IRC18:50
*** otherwiseguy has quit IRC18:50
jeblairjogo: yep, it's a known behavior in nodepool; it will get around to it eventually.18:51
jeblairjogo: there was an attempt to change that behavior, but it had some problems that haven't been identified/solved yet.18:52
jeblairjogo: see the nodepool change queue if you're interested18:52
*** hashar has joined #openstack-infra18:52
dhellmannfungi, lifeless: looks like that test went in with mordred's change:
*** lucas-dinner is now known as lucasagomes18:53
*** lucasagomes has left #openstack-infra18:53
jogojeblair: cool, thanks18:54
*** dtantsur|afk is now known as dtantsur18:54
*** vhoward has joined #openstack-infra18:55
*** otter768 has quit IRC18:57
*** e0ne has joined #openstack-infra18:57
*** zehicle_at_dell has joined #openstack-infra18:58
*** hogepodge has joined #openstack-infra18:58
clarkbfungi: knocking out the slave is a bit scary18:58
*** hogepodge has quit IRC18:59
*** pblaho has joined #openstack-infra19:02
openstackgerritJoshua Harlow proposed a change to openstack-infra/config: Switch taskflow to use the unittests slave script
*** pblaho has quit IRC19:02
Ajaegeris blazar dead? The channel is silent and no patches for a few weeks here:,n,z ? Did they take August off?19:03
anteayaAjaeger: best to chat with DinaBelova19:04
anteayamy understanding is that the blazer functionality has moved into other projects19:04
anteayabut DinaBelova would know more19:04
markvanfungi: on that autoconf issue, I see some successes with the bare-trusty-rax-dfw-1673292 node, and yet a failure on the bare-trusty-rax-iad-1673998, see and
markvanshouldnt thosre trusty rax nodes be acting the same?19:06
*** dane_leblanc has quit IRC19:06
Ajaegeranteaya: thanks, let's see whether DinaBelova is around...19:07
*** nelsnelson has joined #openstack-infra19:07
fungimarkvan: they *should* be acting the same, but it's possible they have local variations in base images or we had local issues with some nodepool image updates (also we know that there's something strange going on in dfw which seems to sometimes be booting different snapshots than we ask for)19:08
*** nelsnels_ has joined #openstack-infra19:08
Ajaegeranteaya: might be late for her...19:10
markvanfungi: interesting, ok...well that other patch to enable bundler install verbose is going thru, so hopefully that will allow us to better understand this.19:10
markvanfungi: btw, thx for all your help on this one.19:10
*** rcarrillocruz has quit IRC19:11
anteayaAjaeger: probably, yes19:11
fungimarkvan: you're welcome--i'm curious to get to the bottom of it too19:11
*** nelsnelson has quit IRC19:12
*** ramashri has joined #openstack-infra19:12
*** rcarrillocruz has joined #openstack-infra19:12
*** krtaylor has quit IRC19:13
*** ramashri has quit IRC19:15
*** ramashri has joined #openstack-infra19:15
*** Sukhdev has joined #openstack-infra19:16
*** changbl has quit IRC19:17
*** mestery has quit IRC19:20
openstackgerritAdam Gandelman proposed a change to openstack-infra/config: Consolidate and rename Ironic jobs
*** mestery has joined #openstack-infra19:20
openstackgerritClayton O'Neill proposed a change to openstack-infra/jenkins-job-builder: Move extended-choice support to parameters module
dvorak@clarkb: ^^ that's the JJB patch we discussed a little while ago19:21
*** Ryan_Lane has quit IRC19:22
*** marun_afk is now known as marun19:24
openstackgerritMatt Riedemann proposed a change to openstack-infra/elastic-recheck: Add query for keystone unit test xml hash bug 1361337
uvirtbotLaunchpad bug 1361337 in keystone "keystone.tests.test_serializer.XmlSerializerTestCase.test_collection_member random fails; lxml hashseed?" [Undecided,New]
*** e0ne has quit IRC19:26
*** ominakov has quit IRC19:26
*** sdake_ has quit IRC19:27
marunclarkb: well, that's embarrasing19:27
marunclarkb: I somehow managed to completely miss the fact that I couldn't override things by setting ENABLED_SERVICES19:27
*** sflanigan_ has joined #openstack-infra19:28
*** sflanigan_ has joined #openstack-infra19:28
*** camunoz_ has joined #openstack-infra19:28
*** ominakov has joined #openstack-infra19:29
*** hogepodge has joined #openstack-infra19:29
*** changbl has joined #openstack-infra19:29
openstackgerritMatt Riedemann proposed a change to openstack-infra/elastic-recheck: Add query for keystone unit test xml hash bug 1361337
uvirtbotLaunchpad bug 1361337 in keystone "keystone.tests.test_serializer.XmlSerializerTestCase.test_collection_member random fails; lxml hashseed?" [Undecided,New]
*** jamielennox|away has quit IRC19:30
*** camunoz has quit IRC19:31
*** sflanigan has quit IRC19:31
*** ianw has quit IRC19:32
*** jamielennox|away has joined #openstack-infra19:32
rcarrillocruzpleia2: hi there, hope you recovered from your gallbladder ok :-)19:33
rcarrillocruzi saw your post about bug day tomorrow19:33
*** ianw has joined #openstack-infra19:33
rcarrillocruzwill you host it in this channel or in #openstack-meeting?19:33
fungircarrillocruz: we coordinate them in this channel19:34
rcarrillocruzgood, i'll try to be around19:34
fungircarrillocruz: though there's not usually much going on in irc while we're going through bugs, other than just pinging one another quickly to figure out statuses that may not be accurately reflected before updating19:34
rcarrillocruzapart from flagging bugs no longer relevant and other stuff, do you also take assignment ?19:35
rcarrillocruznot sure about the format19:35
*** jp_at_hp has quit IRC19:35
fungircarrillocruz: sure, but it's really just an opportunity to spend some time focused on catch-up with bug triage and updates. we don't really do anything which couldn't be done any other day (if there are bugs you want to pick up or want to talk about, feel free to do so in here at any time)19:37
lifelessdhellmann: hi19:37
fungircarrillocruz: similarly, feel free to comment with any bugs you feel you have relevant information about19:37
fungier, within19:37
*** hogepodge has quit IRC19:38
*** kojack1 has joined #openstack-infra19:38
lifelessflaper87|afk: hi19:38
*** mmedvede_ has quit IRC19:38
openstackgerritSean Dague proposed a change to openstack/requirements: updated requirements repo readme
lifelessfungi: and yo19:38
fungiyo ho ho and a bottle of rum19:39
*** otherwiseguy has joined #openstack-infra19:39
lifelessflaper87|afk: ok, thats a bad version number in that we haven't got quite enough compat codefor in semver.19:40
lifelessflaper87|afk: I'll put up a patch for it today.19:40
*** mikedill_ has quit IRC19:41
*** hogepodge has joined #openstack-infra19:43
*** Ryan_Lane has joined #openstack-infra19:43
*** krtaylor has joined #openstack-infra19:45
*** jcoufal_ has quit IRC19:46
*** homeless has joined #openstack-infra19:47
sdaguedhellmann - as promissed, also I highlighted it with a mailing list thread19:47
dhellmannsdague: on my review list19:48
*** hogepodge has quit IRC19:49
dhellmannlifeless: mentions a logstash query that shows this is affecting other projects, too (swift, among others)19:49
uvirtbotLaunchpad bug 1361251 in pbr "pbr is failing to parse versions" [Undecided,New]19:49
*** pradk has quit IRC19:52
lifelessok, C just got up19:54
clarkbfungi fwiw I am +2 on the git review config logic cleanup and a var cleanup change19:54
*** ptoohill has joined #openstack-infra19:54
*** mmedvede_ has joined #openstack-infra19:54
clarkbfungi the second is an easy approval19:54
*** prad has joined #openstack-infra19:54
lifelessdhellmann: will help with this - but we have to get through 4 or so patches to use it19:54
lifelessdhellmann: I will do a sidebar patch to fix the parser to cope with this data input19:55
lifelessdhellmann: and and are all green19:56
lifelessbbiaf hours19:56
jeblairfungi: the graph suggests we're spending more time in delete; i haven't looked into it19:56
stevemardhellmann, the oslotest developer docs scare me :(19:57
stevemarstraight up copying the readme, that's a bold move19:57
dhellmannstevemar: maybe we should move this to #openstack-oslo19:58
*** rlandy has quit IRC19:58
fungijeblair: could be we're burning a lot straight from building to delete. we don't have a ton in error+deleting state (14 in rax and 13 in hpcloud)19:59
fungijeblair: i'm checking the nodepool logs now to see what's going on with them19:59
fungimight be quota problems or the floating ip leak20:00
*** baoli has quit IRC20:00
*** emagana_ has quit IRC20:00
*** otherwiseguy has quit IRC20:00
clarkbfloating ip leak would cause lots of node boot failures20:00
*** emagana has joined #openstack-infra20:00
*** marcoemorais has quit IRC20:01
*** pelix has quit IRC20:02
*** tonytan4ever has quit IRC20:02
*** zz_dimtruck is now known as dimtruck20:03
*** pballand has quit IRC20:03
*** fifieldt_ has joined #openstack-infra20:03
*** daya_k has quit IRC20:03
*** ociuhandu has joined #openstack-infra20:04
fungithe vast majority of LaunchStatusException occurrences in the log over the past 5 hours have been in hpcloud (by a couple orders of magnitude)20:05
*** emagana has quit IRC20:05
fungi~700 in hpcloud vs 8 in rax20:05
*** kgiusti has left #openstack-infra20:06
*** fifieldt has quit IRC20:06
*** Ryan_Lane has quit IRC20:06
*** dtantsur is now known as dtantsur|afk20:07
fungiall seem to be instances in error status20:07
clarkbfungi what does a floating ip list look like?20:07
fungidoesn't look like we're getting the exceptions we used to get creating/attaching floating ips, but checking now20:07
clarkbif it has ~600 rows and lost of unassociated IPs then we are likely leaking20:08
*** annegentle has joined #openstack-infra20:08
* clarkb is running an errand then finding lunch otherwise would look20:09
*** zehicle_at_dell has quit IRC20:09
fungi500 rows. getting an unassociated count now...20:09
clarkbat 500 we probably arent leaking too bad. we seem to notice it worst when we get up to our quota of 60020:10
fungiwell, actually 492 rows, 163 of which aren't associated with any instance20:10
*** Ryan_Lane has joined #openstack-infra20:10
*** otherwiseguy has joined #openstack-infra20:10
clarkbprobably worth cleaning up regardless20:10
fungibut again, usually we got different failure modes out of that20:10
fungithis is nodepool spinning while waiting for the instance to become available, and then seeing it transition into error20:11
fungitteggel: around?20:11
fungii'm also about to head out for an early dinner before the tourists descend on all the restaurants20:12
clarkbsilly tourists20:13
*** tonytan4ever has joined #openstack-infra20:13
*** pcrews has quit IRC20:14
*** marcoemorais has joined #openstack-infra20:18
*** pballand has joined #openstack-infra20:19
*** emagana has joined #openstack-infra20:19
*** rcarrillocruz has quit IRC20:21
*** ominakov has quit IRC20:21
*** marun has quit IRC20:21
jeblairhashar: do you know when you might be able to look at the cloner changes?  (it's mostly tests, only 2 small changes to the cloner itself)20:21
dhellmannlifeless: I'm not sure I see how addresses this parsing issue, except that it would have prevented the package from being created in the first place. Now that the package *does* exist, we need to keep allowing it to work, right?20:22
*** ominakov has joined #openstack-infra20:23
*** mikedillion has joined #openstack-infra20:23
*** weshay has quit IRC20:24
*** mjturek has joined #openstack-infra20:24
*** rcarrillocruz has joined #openstack-infra20:25
stevebakersdague: hey, do you feel strongly about project functional tests living in foo.tests.functional vs root package functionaltests?20:26
hasharjeblair: probably tomorrow. My Zuul cloner job has been blessed in production and is now voting :]20:27
hasharjeblair: gotta upgrade my Zuul install to pull some of your change and then I guess I will give a try at the cloner tests20:27
*** masayukig has quit IRC20:27
hasharjeblair: setting ZUUL_BRANCH as the default fallback branch sounds sane;  I am not sure why I havent done that originally20:27
*** lcheng has joined #openstack-infra20:28
sdaguestevebaker: can you explain what you mean by root package functional tests?20:28
jeblairhashar: cool, thanks20:28
sdagueso that I make sure I give a proper opinion20:28
stevebakersdague: like here
*** Ajaeger has quit IRC20:29
hasharjeblair: and thanks a ton for amending it :]20:29
*** gokrokve has quit IRC20:29
sdaguestevebaker: so I think part of the issue is that if these don't have namespacing, how does this work if things get pip installed?20:30
stevebakersdague: i think that means functionaltests wouldn't get packaged20:31
sdaguethis is one of the reasons why we had to do add nova hacking rules as nova.hacking, because otherwise things are likely to get weird20:31
sdaguestevebaker: but unit tests still would?20:31
stevebakersdague: yes, unit tests would. And functional tests would if they are in the heat namespace20:32
sdaguehonestly, I suspect that we're going to get bitten if we don't namespace these things correctly20:32
*** julim has quit IRC20:32
sdaguebut someone that understands python packaging better would be the person to best answer that20:32
*** miqui has quit IRC20:33
stevebakersdague: maybe heat.functionaltests would be better, the tox env matching pattern would be slightly less tortured20:33
sdaguestevebaker: what's the objection to putting it in heat.tests?20:33
ianwjeblair: i'm having quite a lot of trouble coming up with a more extensive nodepool test, to try and debug some of the capacity issues20:34
ianwjeblair: i may be taking the wrong approach20:34
stevebakersdague: comments line 59
ianwjeblair: have you seen anything like OperationalError: (OperationalError) (1213, 'Deadlock found when trying to get lock; try restarting transaction') 'UPDATE node SET ip=%s WHERE = %s' ('fake', 21L) before?20:34
*** bhuvan_ has joined #openstack-infra20:34
*** masayukig has joined #openstack-infra20:34
ianwjeblair: this is when I call deleteNode() to try to simulate nodes coming and going more fully20:35
*** otherwiseguy has quit IRC20:35
*** terryw has joined #openstack-infra20:35
sdagueso I would agree you probably want to separate this by directory and not by tags20:35
stevebakeroh yeah, not by tags. if the spec says that then ignore it20:36
marcusvrnasselin: hi! Now I definitely started the CI installation using your script. One thing that I have a question is if I have to deploy 2 machines using your script (one for jenkins master and another one for nodepool) or just one that will run jenkins master and nodepool.20:36
jeblairianw: nodepool isn't really meant to be used like a library, so i wouldn't expect you to be able to call deletenode from outside of nodepool itself and have it work20:36
sdagueheat.functionaltests would probably be fine. But for an example both glance and neutron do foo.tests.unit and foo.tests.functional20:37
stevebakerjeblair, clarkb: could I trouble you for a review?
*** bhuvan has quit IRC20:37
jeblairianw: so the functional tests take the approach of setting up the preconditions and letting it just run20:37
sdagueso that's a common pattern that projects doing this are using, and replicating it would be cool unless there is a huge reason not to20:37
asselinmarcusvrn, you can do it all in one machine20:37
sdagueswift does the same, except s/tests/test/20:38
stevebakersdague: OK, I'll talk to zaneb and come up with something20:38
asselinmarcusvrn, it's possible you hit some recent bugs that would prevent you from doing that. They got merged last week. Update to my latest and you should be ok.20:39
*** marun has joined #openstack-infra20:40
sdaguestevebaker: I put a 0 comment in about it20:40
clarkbsdague stevebaker I know mordred thinks all tests should go in the package20:40
ianwjeblair: well the problem seems to be at capacity, when a lot of requests are coming and going.  how would you approach the testing20:40
clarkbso that you can import them20:40
ianwjeblair: also, i'm more familiar with now & where it is logging, is it possible to get the logs for the time when it went crazy with the allocator patch20:40
marcusvrnasselin: nice! thks20:41
jeblairianw: i would reproduce it by running nodepool locally with a config that uses fake providers to get the exact conditions down, then i would see if i could do that in a test20:42
jeblairianw: yeah, i think we can try to extract it; i'm a bit busy now though, ask again if i don't get to it soon.20:43
ianwjeblair: ok, thanks20:44
openstackgerritJames E. Blair proposed a change to openstack-infra/nodepool: Correct ssh key instructions in README
jogovery random question, a month or so back I heard mordred was looking into switching nodepool to use the rebuild command did anything come of that?20:47
*** sdake_ has joined #openstack-infra20:49
anteayapleia2: if I am an admin for an email list am I automatically subscribed?20:50
*** pcrews has joined #openstack-infra20:53
*** bhuvan_ has quit IRC20:56
fungiheading out for a bit... back soon20:58
pleia2anteaya: I don't think so20:58
pleia2if someone knows how admin password for ml is is handled, please tell anteaya :)20:59
anteayapleia2: okay thanks20:59
*** dane_leblanc has joined #openstack-infra20:59
pleia2once you're logged in as admin, I can help from there21:00
anteayaah not sure how to log in without a password21:01
* anteaya checks my email alerts again21:01
clarkbthe admin passwords for lists should be emailed to you when it is created21:01
clarkbthen you change it21:01
*** zns has quit IRC21:01
pleia2thanks clarkb21:02
*** laca has quit IRC21:02
clarkbthe initial password is basically a very poorly kept shared secret21:02
*** ociuhandu has quit IRC21:03
*** sparkycollier has joined #openstack-infra21:03
anteayaclarkb: thanks, changed21:04
*** rloo has left #openstack-infra21:05
*** pballand has quit IRC21:06
*** sparkycollier has quit IRC21:06
*** david-lyle has quit IRC21:06
*** lttrl has quit IRC21:08
*** pdmars has quit IRC21:09
nibalizerjesusaurus: im working on some verbage for 99990, want me to msg it to you or just upload a new patchset to that reveiw?21:10
*** pdmars has joined #openstack-infra21:10
jeblairclarkb, hashar: our git prep scripts (gerrit-git-prep and devstack-gate) both have the option to clone from /opt/git/$PROJECT, and then update the remote to git.o.o before doing work21:10
jesusaurusnibalizer: feel free to just upload a new patch21:11
jeblairclarkb, hashar: (where /opt/git are cached versions of the repos on our images, which may be days old)21:11
stevebakersdague: could you pop into #heat to talk functional test packaging?21:11
lyxusis the vArmour CImaintainer here ?21:11
jeblairclarkb, hashar: but zuul-cloner doesn't support that21:11
jeblairclarkb, hashar: we could use it to work directly on /opt/git (it would correctly update from git.o.o in that case)21:12
hasharjeblair: clarkb: on my setup i have a gerrit replica to the slave instances and  git clone --reference   from there21:12
jeblairhashar: how do you git clone --reference with zuul-cloner?21:12
hasharjeblair: nope. I am going to dish out the --reference system21:12
*** mfer has quit IRC21:12
hashar(on my setup)21:12
clarkbjeblair: the tripleo tests were broken recently because mordred wrote a change for them that switched them to using /opt/git directly with no updates21:13
*** linkid has joined #openstack-infra21:13
clarkbjeblair: so I think there is already some assumption that those repos can be consumed directly. I would not be opposed to doing it this way21:13
jeblairclarkb: i don't think that's the assumption at all21:13
clarkbjeblair: it was mordreds assumption when he broke tripleo21:13
jeblairclarkb: just because mordred got a change wrong doesn't mean that it's universal21:13
clarkbI agree21:13
jeblairhashar: i don't understand21:14
clarkbI was thoroughly confused that anyone would use the cache directly :)21:14
hasharsorry late21:14
*** pdmars has quit IRC21:14
*** mbacchi has quit IRC21:14
jeblairhashar: so you're going to stop using --reference?21:14
hasharjeblair: zuul-cloner has two git url:   --zuul-url which points to a the merger having the Zuul merged commit21:14
hasharjeblair: and a git_base_url  which is some reference repository (i.e. gerrit)21:14
*** linkid has left #openstack-infra21:15
hasharjeblair: on my setup I do:  zuul-cloner --zuul-url git://merger01/
hasharit clones from gerrit, then fetch from git://merger0121:15
nibalizerjesusaurus: kk will do21:15
jeblairclarkb: the problem with using /opt/git directly is that it may make it hard to put repos in certain places (eg devstack; this is a minor problem), and extra hard to put repos in two places (grenade; this is more important)21:16
*** gokrokve has joined #openstack-infra21:16
clarkbjeblair: yup, I think the abstraction of treating them as read only caches is a good on21:16
*** Sukhdev has quit IRC21:16
hasharso git would get something reasonably up to date by pulling from gerrit (or in your case /opt/git ), then get the diff from the merger git repo21:16
jeblairhashar: right, our setup has three: local cache (/opt/git), authoritative repo (git.o.o), zuul (zm0X.o.o)21:16
clarkbjeblair: hashar: maybe a git_cache_url would be a good addition?21:17
jeblairhashar: except that zuul-cloner would not consult the zuul merger when updating a branch tip, only when setting up a change21:17
clarkbjeblair: hashar: beacuse I think having a cache of some sort is a feature that zuul cloner should support especailly for bigger repos21:17
*** Sukhdev has joined #openstack-infra21:18
*** zns has joined #openstack-infra21:19
*** dkranz has quit IRC21:19
jeblairclarkb: yeah, i think to support our 3-tier system with the cloner as-is (say for the grenade job) we would have to run cloner, then copy those repos to /opt/stack/old, then run cloner again and copy the repos to /opt/stack/new; all that copying of repos around seems like something cloner should be able to do21:19
hasharI have copy pasted a bunch of things from devstack-gate21:19
hasharbut iirc the idea is to clone from  git_base_url argument21:19
hasharthen change the remote with git remote --set-url  to point to the merger21:20
hasharand fetch the Zuul ref21:20
jeblairhashar: i believe it changes the remote to the authoritative server (git.o.o in our case); not to the merger21:20
*** mikedillion has quit IRC21:21
jeblairhashar: devstack-gate does it slightly differently than gerrit-git-prep...21:21
*** daneleblanc has joined #openstack-infra21:22
jeblairdevstack-gate does: rsync /opt/git/$PROJECT /opt/stack/new (and /opt/stack/old if running grenade), resets the origin remote to, git remote update, fetch zuul ref from merger21:23
jeblairgerrit-git-prep does: clone /opt/git/$PROJECT to $WORKSPACE/$PROJECT, resets the origin remote to, git remote update, fetch zuul ref from merger21:23
jeblairbasically the same though, with an rsync vs a clone21:24
jeblairhashar: i kind of like your idea of using --reference though...21:24
hasharthe git_base_url argument is Cloner.git_url .   In prepareRepo, the first action is to invoke cloneUpstream() that uses that git_base_url21:24
jesusauruswhats the reason for d-g doing an rsync and g-g-p doing a clone?21:24
jesusauruswhy not do it the same in both?21:24
hasharit updates from it, then prune stale remoes21:24
jeblairjesusaurus: written at different times?  i think they're effectively the same.21:25
clarkbjesusaurus: d-g rsyncs everything, ggp clones one repo21:25
jesusaurusclarkb: ah21:25
hasharjeblair: so I guess you want to clone from /opt/git  then remote update with the authoritative repository.   Would need another parameter added ( maybe --authoritative ?)21:25
jeblairhashar: so i was thinking i like your use of --reference21:25
jeblairhashar: maybe we could have zuul-cloner use --reference when it does it's clone, then it wouldn't need to mess with remotes (which seems kind of hacky)21:26
jeblairhashar: (and it makes the operation of zuul-cloner much more like git itself)21:26
jeblairso easier to understand21:27
jeblair(and of course, it would be optional)21:27
jeblairso we would do 'zuul-cloner --reference /opt/git/ $PROJECTS'21:29
hasharjeblair: the reason I came with --reference is that we wiped the job workspace and recloned big repositories from Gerrit.  for mediawiki/core that takes roughly 2:30 minutes :D21:29
jeblairand it would do 'git clone --reference /opt/git/$PROJECT$PROJECT /path/to/repo'21:29
hasharinstead we moved to use git clean -xqdf  ( option "clear workspace" in Jenkins git plugin)21:29
openstackgerritChristian Fetzer proposed a change to openstack-infra/jenkins-job-builder: Add support for the Jenkins Valgrind publisher plugin.
hasharjeblair: yeah that will speed up the initial clone, granted /opt/git and /path/to/repo are on the same device21:30
*** maurosr has quit IRC21:30
*** markmcclain has joined #openstack-infra21:30
jeblairjesusaurus: that's why it rsyncs ^ :)21:30
clarkbjeblair: hashar so --reference wouldn't be a git ref but a pointer to a cache of repos?21:30
jeblairjesusaurus: they are on different devices21:30
hasharclarkb: when you do  git clone --reference /srv/cache/repo  /srv/workspace/repo  (granted /srv is a partition), git is smart enough to create hardlinks from /srv/cache/repo to /srv/workspace/repo for the .git objects and pack files21:31
*** maurosr has joined #openstack-infra21:31
jesusaurushashar: oh cool21:32
hasharclarkb: though whenever you repack  the original repo  /srv/cache/repo ,  you end up screwing the workspace :D21:32
hasharso instead of cloning / copying across network disk, you save a huuuuge chunk of I/O21:32
*** jgrimm has quit IRC21:32
*** jgrimm has joined #openstack-infra21:32
clarkbhashar: oh I didn't know about that feature21:32
hasharif your /opt/git caches are packed, git just hardlink the few packs21:32
openstackgerritBrant Knudson proposed a change to openstack/requirements: Add a tox env to check requirements overlap
clarkbthis is a final call to protest the proposed code review -1 default text21:33
*** dprince has quit IRC21:33
hasharclarkb: I learn a git feature every week since we switched from svn :]  (last one:  git rebase --preserve-merges ,  let you rebase your branch keeping the merge commit instead of fast forwarding them).21:33
hasharclarkb: I pointed to a launchpad bug, could not assign it to you though.21:34
clarkbjeblair: did you want to weigh in before I change the text?21:34
clarkbhashar: ya, I will update the change but wanted to do that after we reached "consensus"21:34
clarkbsince changing the commit message will wipe out votes21:35
hasharclarkb: I assumed that.  You probably dont want to clear out the 50+  +1 present on the patchset hehe21:35
hasharah people were probably not aware / paying attention that all votes will be phased out21:35
*** alkari has quit IRC21:35
*** dimtruck is now known as zz_dimtruck21:35
hasharmight need some kind of "sorry folks" announcement21:35
clarkbnah the actual change is manual so I think thats ok21:36
*** dizquierdo has joined #openstack-infra21:36
hasharjeblair: so for zuul cloner, some tweak can be made in zuul/lib/  prepareRepo()21:37
jeblairclarkb, hashar: so --reference sounds like it may be a good idea if we can assume we're on the same device; but now that we're discussing that, i'm not certain that will be the case21:37
jeblair(or that we should make that assumption)21:38
hasharjeblair: another possibility would be to check whether a Zuul ref exist and if so use a shallow clone (only copying the ref, not the whole repo)21:38
clarkbjeblair: right especailly for the non d-g jobs21:38
hasharjeblair: and else, fallback to do a shallow clone of the tip of the branch from the authoritative repo21:38
hasharthis way you dont clone the whole repo which might be just fine21:38
jeblairhashar: we won't be able to use that (we have all kinds of things that don't work with shallow clones)21:38
hasharon my prod boxes, I have a SSD which has the Gerrit replica and the workspace21:40
jeblairhashar, clarkb: so maybe we're back to adding an argument to do the 'clone from cached repo if present then reset remote and update' thing we're doing now?21:41
clarkbjeblair: I think so21:41
clarkbjeblair: its simple and should work just about everywhere so I am a fan21:41
hasharsounds fine21:41
hasharprobably want to rename a few function in the Cloner21:41
jeblairokay, i'll double check some things and write up a change if it all looks good21:42
hashargit_base_url is used by cloneUpstream()21:42
hasharbase, upstream, authoritative etc...21:42
hasharthe semantic is a bit confusing (sorry)21:42
*** mriedem has quit IRC21:42
clarkbI am going to poke at nodepool errors a bit mroe21:44
clarkbjeblair: your feedback on the code review -1 text change would be great21:45
*** mmedvede_ has quit IRC21:45
jeblairclarkb: are you sure?  i'm increasingly a fan of notmyname's old idea of changing it to "-1".  :)21:45
clarkbjeblair: just -1?21:46
clarkbjeblair: I personlly don;t care a whole lot either way. I have always read the existing text as please go update21:46
clarkbbut that may be a bad native english speaker assumption21:46
jeblairclarkb: yes, based on the idea that we should not be speaking for the reviewers who are capable of speaking for themselves...21:46
jeblairclarkb: i think the whole conversation is based on the flawed assumption that the -1 text is the only context21:46
jeblairclarkb: no one should ever leave a -1 review without text indicating why, and what needs to happen for it to be resolved21:47
*** MaxV has joined #openstack-infra21:47
hasharthat has been a recurring complain in our community :(21:47
*** jgrimm has quit IRC21:47
hasharaka Gerrit speaking for ourselves.21:47
hasharanyway, late there.  Will do some review of the zuul cloner patches tomorrow21:47
jeblairhashar: cool, thanks.  good night!21:47
hasharI looked at them on Sunday, I love all the new patches21:47
hashargotta test them out a bit :]21:48
*** zz_dimtruck is now known as dimtruck21:48
hasharand if something is not compatible, I can always adjust my jenkins jobs21:48
*** radez_g0n3 is now known as radez21:48
*** reed has joined #openstack-infra21:49
*** dims has quit IRC21:50
jeblairclarkb: so i think in both cases (bare and devstack), the workspace and cached git repos are actually on the same device21:50
jeblairclarkb: in the case of bare, /opt/ and $WORKSPACE are both on /21:50
clarkbwe do have quite a few rax nodes that must be in the normal can't delete mode21:50
*** dims has joined #openstack-infra21:50
jeblairclarkb: in the case of devstack, /opt is the ephemeral disk, but the contents have been moved to it, and all the actual work happens in /opt/stack/{old,new}21:51
clarkband rax-dfw seems to be failing almost 100% for us21:51
clarkbjeblair: oh right we don't remount on the bare nodes21:51
*** annegentle has quit IRC21:52
jeblairso we _could_ use --reference, but we'd be stuck with it (we couldn't ever change to having the cached repos on / and doing work on the ephemeral disk21:52
jeblairand i'm not sure i want to add that complication)21:52
*** annegentle has joined #openstack-infra21:52
clarkbyeah and the repacks could potentially bite us though chances of that are low21:52
*** laca has joined #openstack-infra21:52
*** mmedvede has joined #openstack-infra21:53
hasharI am not sure what happens when the ref repo is repacked21:53
*** stevemar has quit IRC21:53
hasharbut I guess git will copy the new pack as a result21:53
hasharor bails out with the hardlink being moot / disappearing21:53
jeblairyeah, just another reason not to go down that road21:54
clarkbhashar: well if its all hard links it should be mostly transparent21:54
clarkbhashar: but that assumes that everything is a hard link which it can't be if they have different work dirs21:54
*** radez is now known as radez_g0n321:54
hasharonly the .git/objects are hardlinked though21:54
hasharthe workspace is a copy21:54
hashar(hopefully , dont quote me)21:54
*** david-lyle has joined #openstack-infra21:55
*** dims has quit IRC21:55
hasharall of that is a bit scary to me, so I never really looked deep into it.21:55
hasharreally sleeping now :-D21:56
clarkbhashar: go to sleep :)21:56
clarkbgood night21:56
*** tonytan4ever has quit IRC21:56
clarkbjeblair: would you prefer I left the -1 text as is now and bring up "-1" as an alternative21:56
clarkbjeblair: I do prefer -1 more now that I think of it21:57
*** hashar has quit IRC21:57
*** tonytan4ever has joined #openstack-infra21:57
*** enikanorov_ has joined #openstack-infra21:57
*** ildikov_ has joined #openstack-infra21:57
*** MaxV has quit IRC21:58
*** datsun181b has joined #openstack-infra21:58
*** enikanorov has quit IRC21:58
*** dims has joined #openstack-infra21:59
*** dims_ has joined #openstack-infra21:59
*** armax_ has joined #openstack-infra21:59
clarkb{"message": "No valid host was found. ", "code": 500, "created": "2014-08-25T21:58:02Z"} is the error we are getting from hpcloud22:00
clarkbI think that means the scheduler is unable to find anywhere to run 30GB nodes22:00
*** sthillma has joined #openstack-infra22:00
*** ildikov has quit IRC22:00
clarkbjogo: sdague ^ is that a proper interpretation?22:00
*** directxman12 has quit IRC22:00
*** datsun180b has quit IRC22:01
*** armax has quit IRC22:01
*** serverascode has quit IRC22:01
*** armax_ is now known as armax22:01
*** jamielennox|away has quit IRC22:01
*** datsun181b has quit IRC22:01
*** datsun180b has joined #openstack-infra22:02
*** serverascode has joined #openstack-infra22:02
*** MaxV has joined #openstack-infra22:02
*** dmsimard is now known as dmsimard_away22:02
*** emagana has quit IRC22:02
openstackgerritSpencer Krum proposed a change to openstack-infra/infra-specs: Put the puppet modules in their own projects/repos
clarkbsdague: also the sideways ironic job is in the experimental queue now22:02
clarkbsdague: and we should probably keep reviewing the multinode d-g chagne /me does that now22:03
nibalizerjesusaurus: ^^ new verbage22:03
*** dims has quit IRC22:03
nibalizeralso i dont know how to restructure test22:03
nibalizerer text22:03
clarkbhrm it needs a rebase I guess I will go ahead and do that22:03
*** jamielennox|away has joined #openstack-infra22:03
clarkbnibalizer: its like markdown but with slightly more syntax22:03
jesusaurusnibalizer: yeah, i keep wanting to write markdown in my rst, but that mostly doesnt work22:04
*** armax has quit IRC22:04
openstackgerritJames E. Blair proposed a change to openstack-infra/zuul: Add cache-dir argument to cloner
*** reed has quit IRC22:05
*** dims has joined #openstack-infra22:05
*** dims_ has quit IRC22:06
jeblairclarkb: not really.  i'm +0 on the change, i don't think it hurts and i'm glad to see people care :)22:06
*** yolanda_ has joined #openstack-infra22:06
*** zz_johnthetubagu has quit IRC22:06
*** pcm_ has quit IRC22:06
openstackgerritClark Boylan proposed a change to openstack-infra/devstack-gate: Simplify multinode setup
jeblairclarkb: if you wait a few more days, i'll probably +2 it when i get to it in my normal review cycle :)22:06
openstackgerritClark Boylan proposed a change to openstack-infra/devstack-gate: multi-node dg setup
clarkbjeblair: ok, I can wait22:07
jeblairclarkb: if you want to push it in faster, i don't object22:07
*** gondoi is now known as zz_gondoi22:07
clarkbI just rebased the mutlinode d-g changes so that we can see if they still run properly22:07
clarkbit was a mostly trivial rebase that conflicted with sideways grenade22:07
jeblairi pushed up a cache-dir change to cloner; it's untested and undocumented, but i think that's a way we could go:
jeblair(undocumented -> poorly documented)22:07
*** yolanda has quit IRC22:08
clarkbjeblair: in that case maybe I will go ahead and do that now and we can at least see what people think then bring up the textless -122:08
*** dims has quit IRC22:08
*** dangers` has joined #openstack-infra22:08
jeblairclarkb: if people don't like this one, textless is probably the way to go :)22:09
*** dims has joined #openstack-infra22:09
*** jgrimm has joined #openstack-infra22:09
jeblairclarkb: i believe at the time, we figured we needed the text to be a literal "-1" because otherwise you would not be able to see voting history in the comments22:09
anteayanibalizer: whitespace 9999022:09
jeblairclarkb: but with new gerrit, perhaps it could be a literal "".22:09
*** david-ly_ has joined #openstack-infra22:09
jeblairclarkb: not worth worrying about until we get there though22:09
*** kojack1 has left #openstack-infra22:10
*** dims has quit IRC22:10
*** dims_ has joined #openstack-infra22:10
*** zz_johnthetubagu has joined #openstack-infra22:10
*** zz_johnthetubagu is now known as johnthetubaguy22:10
*** dangers has quit IRC22:10
*** mjturek has quit IRC22:10
*** david-lyle has quit IRC22:10
*** ramashri has quit IRC22:10
*** dguerri_ has quit IRC22:10
clarkbjeblair: ya doing whitespace or nil strings in ini files is hard22:10
clarkbI pushed the change22:10
openstackgerritSpencer Krum proposed a change to openstack-infra/infra-specs: Put the puppet modules in their own projects/repos
nibalizeranteaya: fixed ^22:11
clarkband it shows up22:11
nibalizeri clearly dont have my whitespace showing vimrc on this machine22:11
openstackgerritClark Boylan proposed a change to openstack-infra/config: Propose new Code Review -1 text
anteayanibalizer: thanks22:12
*** dguerri has joined #openstack-infra22:12
pleia2nibalizer: line 73?22:12
clarkbnibalizer: want the vimrc chunk for it?22:12
jesusaurusnibalizer: awesome thanks for that. and while you have it open for editing, i apparently had trouble spelling "openstack_project" under Implementation22:12
*** homeless_ has joined #openstack-infra22:12
pleia2clarkb: ooh, I'd like that too22:12
*** HenryG_ has joined #openstack-infra22:12
nibalizerclarkb: i have one but i'd like to see yours22:13
clarkbpleia2: nibalizer: set list then set listchars=tab:>-,trail:-,extends:#,nbsp:-22:13
clarkbthat is what I use22:13
nibalizerbut i just stole hunners22:13
*** marun_ has joined #openstack-infra22:13
nibalizeri just dont thave the 'dev env' set up on the new 'dev' machine22:13
nibalizermaybe i should play with puppeting that again... no wait, tahts madness22:13
*** doug-fish1 has joined #openstack-infra22:13
*** ramashri has joined #openstack-infra22:13
openstackgerritSpencer Krum proposed a change to openstack-infra/infra-specs: Put the puppet modules in their own projects/repos
*** doug-fish1 has left #openstack-infra22:14
*** achanda has joined #openstack-infra22:14
*** bodepd_ has quit IRC22:14
*** doug-fish has quit IRC22:14
openstackgerritSpencer Krum proposed a change to openstack-infra/infra-specs: Put the puppet modules in their own projects/repos
jesusauruspleia2: clarkb: i use these two lines:
nibalizerpleia2: jesusaurus fixed i believe22:14
*** rfolco has quit IRC22:14
pleia2nibalizer: looks good :)22:15
clarkbjesusaurus: I like listchars because it differentiates tabs and spaces22:15
*** marun has quit IRC22:15
*** homeless has quit IRC22:15
*** sflanigan_ has quit IRC22:15
*** SumitNaiksatam has quit IRC22:15
*** HenryG has quit IRC22:15
*** Hefeweizen has quit IRC22:15
*** bodepd has joined #openstack-infra22:16
*** sballe__ has joined #openstack-infra22:16
*** zns has quit IRC22:16
jogoclarkb: yes, that is correct22:17
jogoclarkb: hpcloud is telling you it can't find space for your instance22:17
clarkbjogo: thanks22:17
jogoclarkb: so its 'full'22:17
jogoclarkb: are we seeing that for every call?22:18
*** Hefeweizen has joined #openstack-infra22:18
jeblairnibalizer: thanks for adding the sections about adding module boilerplate; but i think we should do that after we import the new projects so the changes are reviewed22:18
clarkbjogo: no, but for a chunk of them so I think we are floating around the limit22:18
jogoclarkb: do you know what happend to the investigation of using nova rebuild for nodepool22:18
*** david-ly_ is now known as david-lyle22:18
clarkbjogo: I think mordred never pushed code for it22:19
jeblairi don't see any pending changes for it in gerrit22:19
jogoclarkb: so that doesn't mean the idea was scratched, just somewhere 'in progresss'22:19
*** dizquierdo has quit IRC22:19
clarkbjogo: because I looked recently and couldn't find it. The tricky bit is knowing when to delete vs rebuild when scaling back22:19
*** terryw has quit IRC22:19
jeblairyeah, it's not trivial, because it needs  yeah that ^22:19
*** datsun180b has quit IRC22:19
*** sballe_ has quit IRC22:19
jogoclarkb: oh yeah, that makes it fun22:19
nibalizerjeblair: okay22:19
jogoclarkb jeblair: thanks I mainly wanted to make sure there wasn't any nova side issue with rebuild that you guys have hit yet22:20
jeblairjogo: no, however i think tteggel had some nova-side improvements in mind related to that...22:20
openstackgerritSpencer Krum proposed a change to openstack-infra/infra-specs: Put the puppet modules in their own projects/repos
nibalizerjeblair: how is that ^^?22:21
jeblairjogo: er, sorry, i meant phschwartz22:21
*** hdd_ has joined #openstack-infra22:21
*** markmcclain has quit IRC22:21
*** hdd_ has quit IRC22:21
jeblairnibalizer: is the order of 5,6 correct?22:22
*** zns has joined #openstack-infra22:22
lifelessdhellmann-phone: 114403 means that pbr wouldn't be erroring22:22
lifelessdhellmann-phone: I'm prepping another patch to recognise the 0.0.1{a/b/rc}{\d} format22:22
nibalizerjeblair: sortof depends on what we want22:22
*** hdd has quit IRC22:23
jeblairnibalizer: if it will work in that order, fine with me :)22:23
nibalizeri think we want that order22:23
nibalizeralso there is nothing in here about publishing to the forge22:23
nibalizerbut 'meh'22:23
jesusaurusnibalizer: jeblair: we also want to add the implied step of removing the module from config22:24
*** gokrokve has quit IRC22:24
*** annegentle has quit IRC22:24
nibalizerjesusaurus: yes this22:24
openstackgerritSpencer Krum proposed a change to openstack-infra/infra-specs: Put the puppet modules in their own projects/repos
*** tonytan4ever has quit IRC22:26
jogois there any way to sort out how many *pep8 jobs are run a day?22:26
*** dustins has quit IRC22:26
clarkbjogo: yes, graphite captures that22:27
*** ArxCruz has quit IRC22:27
nibalizerjesusaurus: done22:28
*** sflanigan_ has joined #openstack-infra22:28
*** sflanigan_ has joined #openstack-infra22:28
jogoclarkb: oh right22:28
*** rfolco has joined #openstack-infra22:28
*** andreykurilin has joined #openstack-infra22:29
*** pballand has joined #openstack-infra22:30
*** gokrokve has joined #openstack-infra22:31
*** ominakov has quit IRC22:31
*** aysyd has quit IRC22:32
*** dangers` is now known as dangers_away22:32
lifelessfungi: dhellmann-phone:the swift error is odd22:32
lifelessgit describe22:33
*** otherwiseguy has joined #openstack-infra22:34
*** hdd has joined #openstack-infra22:34
lifelessso the tag in swift that should be picked up is 2.0.022:35
*** prad has quit IRC22:35
*** julim has joined #openstack-infra22:36
*** pballand has quit IRC22:38
*** mmedvede has quit IRC22:39
*** julim has quit IRC22:41
*** chuckC has quit IRC22:41
*** signed8bit is now known as signed8bit_ZZZzz22:42
*** chuckC has joined #openstack-infra22:43
fungilifeless: i missed the swift tag/pbr-related error. is there a bug report describing it?22:43
*** chuckC has quit IRC22:43
*** SumitNaiksatam has joined #openstack-infra22:45
clarkbcool mutlinode d-g reviewed. It is coming together22:46
clarkbneeds to accmodate newer grenade var semantics though22:46
openstackgerritJames E. Blair proposed a change to openstack-infra/zuul: Add cache-dir argument to cloner
jeblairokay, that's tested and documented ^22:46
*** che-arne has quit IRC22:46
*** otherwiseguy has quit IRC22:48
*** weshay has joined #openstack-infra22:48
*** krtaylor has quit IRC22:48
jeblairperhaps we should "grep -v attic" when caching the git repos in the nodepool scripts22:50
openstackgerritA change was merged to openstack-infra/elastic-recheck: Add query for keystone unit test xml hash bug 1361337
uvirtbotLaunchpad bug 1361337 in keystone "keystone.tests.test_serializer.XmlSerializerTestCase.test_collection_member random fails; lxml hashseed?" [Undecided,New]
clarkbjeblair: ++22:51
*** zns has quit IRC22:51
fungii could get onboard with that. we shouldn't expect any project to actually make use of a retired project in tests22:51
*** annegentle has joined #openstack-infra22:51
clarkbjeblair: cloner change lgtm except for a doc tyop22:52
clarkbfungi: I woudl argue testing anything in the attic is a bug22:52
clarkbfungi: did you see that I got the error message from hpcloud nodes? and that rax does have a quite a few nodes in the undeletable delete state?22:53
clarkbfungi: also I chagned the -1 string. so can be approved whenever22:53
fungiclarkb: yeah, there are something like 13 or 14 in error+deleting state in rax, though i don't doubt that there are more stuck in other states and undeletable22:53
*** jgrimm is now known as zz_jgrimm22:54
fungioh yikes22:54
*** flaviof_zzz is now known as flaviof22:54
fungidfw has 116 in error+deleting suddenly22:54
*** rcarrillocruz has quit IRC22:54
clarkbfungi: yeah I think they happened right after you looked22:54
clarkbthey were ~1 hour old when I looked22:55
*** mmedvede has joined #openstack-infra22:55
fungiyep, it was definitely *not* that way before i went to dinner22:55
lifelessfungi: so where is this regex I need to fix?22:55
*** reed has joined #openstack-infra22:55
fungilifeless: i was merely suggesting the regex via the link in that pbr bug needed comparing to what we consider a reasonable pre-release pattern22:56
*** bhuvan has joined #openstack-infra22:57
fungilifeless: that determines what patterns we look for in tag names to determine whether we should fire corresponding jobs in zuul'22:57
fungis pre-release pipeline22:57
*** zul has quit IRC22:57
* fungi is designing his next keyboard with the carriage return at least a furlong away from all normal keys22:58
clarkbjogo: can you review 110907?22:58
clarkbjogo: it is related to the largeops job22:58
*** rcarrillocruz has joined #openstack-infra22:59
jogoclarkb: looking22:59
clarkbfungi: kinesis22:59
lifelessfungi: righto so22:59
clarkbjogo: and see my comment about the logs22:59
lifelessfungi: it doesn't stop the tags being pushed ?22:59
jogoI was just about to get ready to cut a new version of hacking with the new flak8 which has multiprocessing on by default, but alas the flake8 release notes lie22:59
clarkbjogo: if they all mutliplex on a single file that may be easiest for d-g but harder for python logging22:59
*** dimtruck is now known as zz_dimtruck23:00
fungilifeless: it doesn't stop the tags being pushed *but* it does stop us building tarballs and uploading releases with them23:00
lifelessfungi: so thats going to want to match the semver rules23:00
lifelessfungi: in the spec // doc23:00
lifelessfungi: which is way more restrictive23:01
lifelessfungi: why is 'g' in that regex today?23:01
openstackgerritSpencer Krum proposed a change to openstack-infra/config: Fix depercation warnings in templates for unatttended upgrades
fungilifeless: i think it was used for milestones23:01
*** sabeen has joined #openstack-infra23:02
lifelessfungi: any idea why we're not seeing the 2.0.0 tag in the swift jobs?23:02
fungialso possible it was only used for grizzly milestones23:02
*** zns has joined #openstack-infra23:03
*** bknudson has quit IRC23:03
fungilifeless: no, i was asking if you had a link to some backstory on that--first i'm hearing of it23:03
lifelessfungi: the error for new/swift with rc1 unrecognized implies that we're seeing one of the rc1 tags in the swift repo23:03
lifeless2014-08-25 14:35:34.140 | error in setup command: Error parsing /opt/stack/new/swift/setup.cfg: ValueError: Unknown remainder [u'rc1']23:03
fungiahh, investigating now23:03
lifelessbut git describe in swift master for me shows23:04
lifelessso its possible I guess that there is a bug in the current pbr23:04
openstackgerritSpencer Krum proposed a change to openstack-infra/config: Fix deprecation warnings in exim module
lifelessbut current pbr doesn't do its own walking yet23:04
*** markmcclain1 has joined #openstack-infra23:04
*** annegentle has quit IRC23:06
fungilifeless: git describe on tip of swift master tells me 2.1.0.rc123:06
lifelessfungi: ah23:08
lifelessfungi: I must not have pulled recently enough23:08
lifelessso that should be
fungihuh. browsing i'm getting conflicting states. i wonder if one of the cluster members behind that load balancer isn't in sync23:09
jogoclarkb: reviewed, your comments were correct we are not collecting all 10 n-cpu logs23:09
*** chuckC has joined #openstack-infra23:09
clarkbjogo: thanks23:09
*** sdake_ has quit IRC23:09
*** zul has joined #openstack-infra23:09
*** dane_leblanc has quit IRC23:09
*** erlon has quit IRC23:10
*** daneleblanc has quit IRC23:10
funginope, all 5 servers in the cluster have 8d02147 as the tip of master23:11
fungiwhich is what's tagged as 2.1.0.rc123:11
fungiweird... shows a different commit as the branch tip compared to
fungiforce refreshing my browser doesn't seem to help either23:13
clarkbfungi: there is a 5 minute cache in cgit23:13
clarkbis the ref > 5 minutes old?23:14
fungishouldn't be that... the newer ref is 2 days old23:14
fungiyeah, gerrit says it merged 08-23 18:0323:14
clarkbjeblair: any chance you want to comment on 114840 (we discussed how you would prefer to always use public IPs for those jobs)23:15
*** oomichi has joined #openstack-infra23:16
*** oomichi is now known as oomichi_away23:17
*** jamielennox|away is now known as jamielennox23:17
*** Sukhdev has quit IRC23:17
*** sabeen has quit IRC23:17
*** sabeen has joined #openstack-infra23:18
*** pdmars has joined #openstack-infra23:19
*** andreykurilin has quit IRC23:19
fungistranger still... haproxy says it balanced both requests to the same git server23:20
clarkbrussellb: comments on 11334923:27
openstackgerritlifeless proposed a change to openstack-dev/pbr: Raise an error if preversion versions are too low
openstackgerritlifeless proposed a change to openstack-dev/pbr: Look for and process sem-ver pseudo headers in git
openstackgerritlifeless proposed a change to openstack-dev/pbr: Refactor oneline changelog walking.
openstackgerritlifeless proposed a change to openstack-dev/pbr: Only consider tags that look like versions.
*** pdmars has quit IRC23:28
openstackgerritA change was merged to openstack-infra/config: Enable parallelism on DVR job
*** pdmars has joined #openstack-infra23:28
fungilifeless: it looks like the only two changes to hit that error in the past 48 hours (according to logstash) were 115265 and 11661823:29
clarkbfungi: want to give that a go? ( you commented onthe original 14 days as well so giving you a chance to read it)23:29
*** nelsnels_ has quit IRC23:30
fungiclarkb: thanks--approved23:30
*** reed_ has joined #openstack-infra23:31
clarkbfungi: thank you!23:31
*** marcusvrn has quit IRC23:31
clarkbnibalizer: I think I asked you about this but if you say package {'foo': ensure => absent} in puppet is it an error if that package doesnt' exist at all?23:31
clarkbnibalizer: take my foo example literally. would that break?23:31
clarkbnibalizer: context for this is 11388423:32
*** reed_ has quit IRC23:33
*** gokrokve_ has joined #openstack-infra23:33
*** pdmars has quit IRC23:33
* clarkb tests23:33
*** krtaylor has joined #openstack-infra23:33
*** reed has quit IRC23:34
*** MaxV has quit IRC23:34
lifelessfungi: so I think it will be very common for swift going forward with the 2.0.0.rc1 tag23:34
lifelessfungi: thus i'm on it now23:34
clarkbnibalizer: it looks like it may not be an error23:34
openstackgerritSpencer Krum proposed a change to openstack-infra/config: Fixing deprecation warnings in logrotate module
clarkbnibalizer: puppet apply -e "package {'foo': ensure => absent}" was my test23:35
fungilifeless: i just find it odd that none of the other swift changes have received updates today in the ~16 hours since that tag was pushed, for that one change to be the only hit23:35
lacahi. is anyone here doing 3rd party testing on anything other than linux?23:36
*** hemna is now known as hemna_23:36
lifelessfungi: sure23:36
*** gokrokve has quit IRC23:36
clarkblaca: I think alexpilotti is doing it for windows23:36
*** annegentle has joined #openstack-infra23:36
alexpilotticlarkb laca: hi there23:37
jeblairclarkb: istr nibalizer said that was okay23:37
alexpilottilaca: hi23:38
*** baoli has joined #openstack-infra23:38
lacaI work for Oracle and we would be keen on doing 3rd party testing on Solaris23:38
lacabut devstack seems to be a stumbling block, since it's linux specific23:38
lacawondering how to get around that23:38
alexpilottilaca: on Hyper-V we are using Devstack on linux for the controller and Hyper-V compute nodes23:39
alexpilottilaca: since what we need to test are the compute nodes23:39
clarkbjeblair: awesome. I went ahead and +2'd it23:40
alexpilottilaca: we’ew also porting the rest of the OpenStack components to Windows23:40
jeblairlaca: not everyone tests using devstack; you can set up your environment however you want23:40
alexpilottilaca: at that point we’ll have to abandon Devstack and move to a Windows equivalent23:40
nibalizerclarkb: ya in my testing it is A-OK with doing that23:40
lacaalexpilotti: hmm... ok23:40
nibalizerthat being said, you're probably doing it wrong23:40
clarkbnibalizer: most definitely23:41
jeblairthat should be our new motto23:41
nibalizerclarkb: so why not if statement?23:41
clarkbnibalizer: beacuse it doesn't matter23:41
alexpilottilaca: so if you need to test the entire stack on Solaris, you have to come up with an alternative unfortunately23:41
lacajeblair, alexpilotti, I was looking at the wiki and reading the code and it seemed that devstack-gate testing is a must23:41
clarkbnibalizer: and will work on all of the other node types23:41
nibalizerfirewalld exists on what? EL?23:41
clarkbnibalizer: centos, trusty, f21, f2023:41
*** atiwari has quit IRC23:41
alexpilottilaca: not really23:41
nibalizeroh ew23:41
clarkbnibalizer: on everything but precise apparently23:41
alexpilottilaca: it’s just a common approach, but not mandatory23:42
jeblairclarkb, nibalizer: new +1 text "lgtm.  that being said, you're probably doing it wrong"23:42
clarkbjeblair: :P23:42
nibalizerso do an if !lsbdistcodename != 'precise'23:42
nibalizeror you know, with the right logic operators and stuff23:42
*** armax has joined #openstack-infra23:42
clarkbnibalizer: or don't worry about it23:42
clarkbnibalizer: this is like doing my DNS edits in sysconfig files23:42
clarkbnibalizer: it will "fail" on ubuntu but thats fine bceause meh23:42
alexpilottilaca: anyway, bringing up your own equivalent solution shouldn’t be too hard23:43
jeblairlaca: ah, if this is to meet a requirement from a specific project, probably best to talk to the ptl of that project and see what they'll accept23:43
nibalizerclarkb: :/23:43
*** annegentle has quit IRC23:43
clarkblaca: out of curiousity where does devstack break on solaris? cinder maybe?23:44
*** zns has quit IRC23:44
jeblairclarkb: it does have lots of apt/yum stuff23:44
clarkbjeblair: oh right that stuff23:45
clarkbI was thinking that the config services and run them in screen bits should be mostly friendly23:45
clarkbcompetely forgot about package installations23:45
jrollwe should just compile everything instead23:46
lacaclarkb: from a brief look, lots of differences in the userland.  Even simple tings like options of grep, and more tricky stuff like figuring out the ip address of the system, etc23:46
nibalizerjeblair: so im circling around to 'why puppetmaster not go' from ages ago23:46
clarkblaca: but you can isntall gnu utils on solaris23:46
nibalizerand its a super derpy problem from what I can tell23:46
clarkblaca: or you could when I sued solaris way back when23:46
jrollthat's a great typo23:47
jeblairnibalizer: we look for things that make us go23:47
lacaclarkb: you can, in fact it's installed by default, but the scripts would have to run ggrep instead of grep and stuff.  But that's the simple stuff.23:47
*** dims_ has quit IRC23:48
*** eharney has quit IRC23:48
lacait could be ported to solaris but it would be quite a bit of work23:48
*** dims has joined #openstack-infra23:48
lacaI might have a go at it23:48
*** otherwiseguy has joined #openstack-infra23:49
*** bdpayne has quit IRC23:49
clarkbsdague: any chance you can weigh in on 115004? I think it is similar to some of the other stuff you were talking about today with heat??23:49
clarkblaca: ya I realize now especially with the package stuff that it would be non trivial23:49
fungilaca: there are people who have used other openstack installation frameworks to build a third-party testing system too, so it's not necessarily required that you do integration testing specifically with devstack... however porting (parts of) that may be easier than starting over23:50
ianwclarkb: (Not use firewalld on the jenkins slaves) <- we should probably just abandon this with the devstack work-around merged.  from my discussions firewall daemon installation seems like something that should be done in image prep in d-i-b23:50
clarkbianw: oh we got the devstack thing to go in?23:50
clarkbianw: I wasn't sure about that and was happy to kill it on our side too for the reasosn in that review23:50
clarkbianw: maybe we double tap it?23:50
jeblairlaca: just so you know, i'm not certain the devstack maintainers would accept patches upstream to support solaris.  i'm not meaning to discourage you -- just want to make sure we're on the same page (and if you are interested in that, you should ask around)23:51
ianwclarkb / jeblair: on that, as the two reviewers on , what's the status of d-i-b ?23:51
lacajeblair: sure, thanks, will do before I anything.  currently just sizing the effort.23:51
*** dims has quit IRC23:52
*** dims has joined #openstack-infra23:52
jeblairianw: i looked at the patch series earlier today, and it has some problems; i think yolanda was going to give it another pass23:52
ianwclarkb: i'm pretty happy to use firewalld to maintain the status quo on centos7 & f21 (when it is out & I get nodes going) if it works23:52
jeblairianw: mostly in the patches that depends on23:52
ianwjeblair: ok, yeah there is a lot there.  my issue is that hp has no centos7 images, so i can't get full testing going23:53
lacaif I managed to figure it out correctly, I can use most of the ci infrastructure, but I would need my own jenkins-job-building kind of thing that uses a different method for deploying the code from git instead of devstack23:53
ianwjeblair: if you know of hp cloud contacts, there's plenty of people willing to help get the images ready for them.  all they need to do is add them :)23:54
jeblairianw: it's definitely next in the process, and i think clarkb, fungi and i are all ready to give it a go when it's ready.  the dependent changes shouldn't be hard to fix23:54
jeblairianw: sometimes tteggel is around; he might know who to ping23:54
ianwjeblair: which would you choose of days / weeks / months? :)23:55
openstackgerritA change was merged to openstack-infra/config: Add non-voting checklinks job for manuals
jeblairianw: for what?23:56
jeblairianw: are you asking what i'd prefer or what i expect? :)23:56
*** ominakov has joined #openstack-infra23:57
ianwjeblair: having nodepool switched to using d-i-b images.  what you expect :)  i think we'd probably all prefer "yesterday" :)23:57
jeblairyeah, cause i mean, i _wanted_ to merge the change today...23:57
openstackgerritA change was merged to openstack-infra/config: Remove an unused node type
*** unicell has joined #openstack-infra23:57
*** unicell has quit IRC23:58
*** unicell has joined #openstack-infra23:58
*** tgohad has quit IRC23:58
*** tsg has quit IRC23:58
jeblairianw: best totally non-binding guess is we'll have some images working within 2 weeks.23:58
*** reed_ has joined #openstack-infra23:58
*** ildikov_ has quit IRC23:58
*** ominakov has quit IRC23:59
*** wenlock has quit IRC23:59

Generated by 2.14.0 by Marius Gedminas - find it at!