Monday, 2018-02-12

« Sunday, 2018-02-11 Index Tuesday, 2018-02-13 »
*** pengdake_ has joined #openstack-mistral00:00
*** pengdake_ has quit IRC00:06
openstackgerritVitalii Solodilov proposed openstack/mistral master: Correction of comments for the #539039 review  https://review.openstack.org/54326700:10
openstackgerritVitalii Solodilov proposed openstack/mistral master: Minor improvements of logging  https://review.openstack.org/53860200:14
*** zhurong has joined #openstack-mistral00:39
openstackgerritVitalii Solodilov proposed openstack/mistral master: Add the 'error_data' parameter to the FailAction  https://review.openstack.org/52854301:09
*** yangyapeng has quit IRC01:25
*** yangyapeng has joined #openstack-mistral01:26
*** yangyapeng has quit IRC01:31
*** pengdake_ has joined #openstack-mistral01:31
*** yangyapeng has joined #openstack-mistral01:32
*** yangyapeng has quit IRC01:36
*** zhurong has quit IRC01:57
*** pengdake_ has quit IRC02:00
*** pengdake_ has joined #openstack-mistral02:00
*** yangyapeng has joined #openstack-mistral02:02
*** d0ugal_ has quit IRC02:03
*** yangyapeng has quit IRC02:05
*** yangyapeng has joined #openstack-mistral02:06
*** d0ugal_ has joined #openstack-mistral02:14
openstackgerritVitalii Solodilov proposed openstack/mistral master: Fixed false positive retrying task if action is already completed  https://review.openstack.org/53974502:37
openstackgerritVitalii Solodilov proposed openstack/mistral master: Fixed the length of a task name  https://review.openstack.org/52849402:57
*** d0ugal_ has quit IRC03:02
*** mcdoker181818 has joined #openstack-mistral03:09
*** itlinux has quit IRC03:10
*** d0ugal_ has joined #openstack-mistral03:13
openstackgerritVitalii Solodilov proposed openstack/mistral master: Fixed the length of a task name  https://review.openstack.org/52849403:36
*** gkadam has joined #openstack-mistral03:40
*** itlinux has joined #openstack-mistral04:02
*** d0ugal_ has quit IRC04:06
*** gkadam is now known as gkadam-afk04:09
*** d0ugal_ has joined #openstack-mistral04:15
*** itlinux has quit IRC04:24
*** itlinux has joined #openstack-mistral04:30
*** pengdake_ has quit IRC04:30
openstackgerritRenat Akhmerov proposed openstack/mistral master: Add the 'error_data' parameter to the FailAction  https://review.openstack.org/52854304:37
*** gkadam-afk is now known as gkadam04:38
*** akazakov has quit IRC04:44
rakhmerovthrash|g0ne: hi, can you please investigate https://review.openstack.org/#/c/539553/ ?04:45
*** kushalAgrawal has quit IRC04:46
*** pengdake_ has joined #openstack-mistral04:53
*** mcdoker181818 has quit IRC05:10
*** mcdoker181818 has joined #openstack-mistral05:12
*** pengdake_ has quit IRC05:22
*** hardikjasani has joined #openstack-mistral05:30
*** pengdake_ has joined #openstack-mistral05:33
*** itlinux has quit IRC06:10
*** jtomasek has joined #openstack-mistral07:43
*** AlexeyAbashkin has joined #openstack-mistral07:51
*** jpich has joined #openstack-mistral08:54
*** d0ugal_ has quit IRC09:03
*** d0ugal has joined #openstack-mistral09:03
*** d0ugal has quit IRC09:03
*** d0ugal has joined #openstack-mistral09:03
openstackgerritQi Peng proposed openstack/mistral master: Add a step to install tox  https://review.openstack.org/54110909:13
rakhmerovd0ugal: hi09:24
d0ugalrakhmerov: Hey09:25
rakhmerovthere's one useful commit in the client that we didn't merge to stable/queens and it didn't become a part of the latest client release09:25
rakhmerovhttps://github.com/openstack/python-mistralclient/commit/e2ff3df5e89ba8ff271bf6db6d2dc57aab2a9bb409:25
rakhmerovthe corresponding server side functionality is merged and is part of the release09:26
rakhmerovI wonder if we are allowed to backport such a patch to stable/queens and release another client09:26
rakhmerovsay 3.2.109:26
rakhmerovor patch versions are only for bug fixes?09:26
rakhmerovif so maybe 3.3.0 would work09:27
d0ugalyeah, we would have to do 3.3.009:27
rakhmerovok09:27
rakhmerovthen would you please take care of the release once it's backported?09:28
rakhmerovI'll do a backport now09:28
d0ugalrakhmerov: I have done the backport :)09:28
rakhmerovooh09:28
d0ugalsure, I can do that09:28
rakhmerovwhen?09:28
d0ugalNow09:28
* d0ugal waits for gerrit to catch up09:28
d0ugalhttps://review.openstack.org/#/c/543393/09:28
d0ugalNot sure why openstackgerrit didn't post it09:29
rakhmerovooh, I see09:29
rakhmerovthanks!09:29
rakhmerovawesome09:29
rakhmerovI just want to demo this new feature internally and realized that it's not released on the client yet09:29
d0ugalyeah, whoops :)09:29
rakhmerov:)09:30
rakhmerovyeah, I missed it somehow09:30
d0ugalyeah, because we didn't do the release09:30
rakhmerovyes09:30
d0ugalwe would have noticed then09:30
rakhmerovit was merged on the client after we made a client release out of stable/queens09:30
d0ugalyeah09:30
d0ugalbut if we had done the final release I guess we would have looked for missing backports.09:31
rakhmerovyes09:31
d0ugalanyway, fairly easy to resolve.09:31
rakhmerovd0ugal: another topic.. I still didn't get may replies on the office hours thing )09:31
d0ugalIndeed :/09:32
d0ugalrakhmerov: I guess we can just start it and see what happens.09:32
d0ugalrakhmerov: we need to do some PTG planning.09:32
*** shardy has joined #openstack-mistral09:33
rakhmerovyeah, honestly today I'm not ready yet (other urgent things). How about Wed 3.00 UTC ?09:33
rakhmerovooh, no!09:33
rakhmerovit's too early for you..09:33
rakhmerov Fri 8.00 UTC09:34
d0ugal3am is a little early09:34
rakhmerovyeah, I figured )09:34
d0ugalI can do Friday - but I guess we wont get many others. Hopefully apetrich09:34
rakhmerovthat's ok09:34
rakhmerov3 people is ok, I'll try to get Andras too09:34
d0ugalgreat09:34
rakhmerovhe'll be at the PTG too and he wants to bring some stuff09:34
rakhmerovgood stuff09:35
rakhmerovok, deal09:35
rakhmerovI'll update everyone in the ML09:36
d0ugalrakhmerov: sounds good.09:37
d0ugalrakhmerov: I wonder if we could get an IRC bot to announce the start and end of office ours for each session here09:37
* apetrich is reading back09:37
rakhmerovd0ugal: hah09:37
rakhmerovgood idea09:37
d0ugalrakhmerov: and maybe it could ping all the cores or something too.09:37
rakhmerovwe definitely can09:38
rakhmerovyeah-yeah09:38
rakhmerovyou know how to do that? Is there some exsiting bot that could be configured?09:38
rakhmerovor a template, I don't know..09:38
d0ugalI don't - but I assume there must be plenty of examples around09:38
rakhmerovok09:38
d0ugalwe could look at openstackgerrit to start with09:38
d0ugalI'm not sure where we would host it09:38
rakhmerovright09:38
d0ugalMaybe something we could as infra09:39
rakhmerovmaybe we could discuss it on Friday too on our first office hours session )09:39
d0ugalask infra*09:39
d0ugalsounds good09:39
apetrichFriday is doable for me09:39
*** shardy has quit IRC09:43
*** shardy has joined #openstack-mistral09:43
rakhmerovd0ugal: https://review.openstack.org/#/c/543393 is checked09:46
rakhmerovbtw, it's not really good that the devstack job is disabled there now..09:46
rakhmerovI asked thrash|g0ne to look at why his patch fails09:46
d0ugalrakhmerov: indeed09:46
rakhmerovthat restores it09:46
rakhmerovd0ugal: so please approve the patch09:46
d0ugalright, I can see if he hast time later today09:46
d0ugaldone :)09:46
rakhmerovyep, thanks09:47
*** pengdake_ has quit IRC10:04
rakhmerovapetrich, d0ugal: when comfortable please review https://review.openstack.org/#/c/542065/, https://review.openstack.org/#/c/541199/, https://review.openstack.org/#/c/541161/, https://review.openstack.org/#/c/543267/10:25
rakhmerovand one more: https://review.openstack.org/#/c/528543/10:25
*** because has joined #openstack-mistral10:34
*** because is now known as Guest532910:35
d0ugalSure10:37
apetrichlooking10:44
openstackgerritMerged openstack/mistral master: Update mysql connection in doc  https://review.openstack.org/54119911:03
openstackgerritMerged openstack/mistral master: Fix error sql about privileges in doc  https://review.openstack.org/54116111:03
*** jaosorior has joined #openstack-mistral11:10
rakhmerovd0ugal, apetrich: when you really feel brave and that this is your day: https://review.openstack.org/#/c/467799/11:23
rakhmerov:)11:23
* apetrich doesn't ;)11:23
* apetrich needs more sleep11:24
rakhmerovthis one had a long way, very much expected it will become a part of rocky11:24
rakhmerov:)11:24
rakhmerovdon't be so scared, it's huge, yes, but mostly because of the tests11:24
apetrichanyway looking at it :)11:25
d0ugalrakhmerov: hah, I was worried you'd ask me to look at that one11:25
rakhmerovI've only half way reviewing it11:25
rakhmerovd0ugal: you were worried right :)11:25
d0ugalI guess I will try at some point11:25
d0ugalIt needs some docs at least11:25
rakhmerovbut no rush really, we'll just need to do it one day )11:25
d0ugalI have no idea how you even use it11:25
d0ugalsure11:25
d0ugalI guess it is something that we might even need to discuss at the PTG11:25
rakhmerovyes11:26
rakhmerovI'll do my best to review it this week11:26
rakhmerovBob (who was very much interested in it) already did11:26
rakhmerovbut he's not a core (yet)11:26
d0ugalI noticed11:26
d0ugalHas Bob been testing it out and trying it too?11:27
*** akazakov has joined #openstack-mistral11:44
*** katkapilatova has joined #openstack-mistral11:49
openstackgerritSagi Shnaidman proposed openstack/python-mistralclient master: Revert "Clean up keystone authentication"  https://review.openstack.org/54343812:02
d0ugalrakhmerov: ^ looks like we merged a regression12:11
*** gkadam has quit IRC12:14
*** yangyapeng has quit IRC12:14
*** Guest5329 has quit IRC12:17
*** yangyapeng has joined #openstack-mistral12:30
*** yangyapeng has quit IRC12:35
*** yangyapeng has joined #openstack-mistral12:47
openstackgerritSagi Shnaidman proposed openstack/python-mistralclient master: Add TripleO jobs for mistral client  https://review.openstack.org/54344712:57
*** pengdake_ has joined #openstack-mistral13:03
apetrichd0ugal, when did that land that I didn't see it happening13:03
d0ugalapetrich: ages ago :)13:04
apetrich26th of january?13:04
apetrichouch. how come?13:04
d0ugalapetrich: yeah, 26th jan - I thought longer actually13:05
d0ugalOnly an issue now because tripleo uses released clients only13:05
d0ugaland we have no tripleo CI on mistralclient13:05
d0ugalbut we should soon! https://review.openstack.org/#/c/543447/13:05
d0ugalOh, you reviewed already :)13:06
*** pengdake_ has quit IRC13:06
apetrichwhat that why we had some issues on our patches that the rest were not seeing?13:06
apetrichbut13:06
apetrichI've installed mistral master more recently than 26th13:07
* apetrich is confused13:07
d0ugalYeah, I don't fully understand13:07
d0ugalwhen was mistralclient last released13:07
apetrichoh mistralclient13:07
d0ugaland then I guess the version needs to be bumped in instack-undercloud?13:07
apetrichI've not installed python-mistralclient master13:08
apetrichoh ok13:08
openstackgerritMerged openstack/mistral master: Correction of comments for the #539039 review  https://review.openstack.org/54326713:08
apetrichthat makes more sense13:08
d0ugala little bit more :)13:08
apetrichthat keystone crap needs to be unified cof* cof* mistral-extra* cof13:08
d0ugallol13:09
d0ugalapetrich: do it!13:09
apetrichI need to have something for my throat :)13:09
apetrichI will push for it for next cycle13:09
*** hardikjasani has quit IRC13:09
*** weshay|ruck is now known as weshay13:10
*** thrash|g0ne is now known as thrash13:18
apetrichd0ugal, and now it is abandoned the patch13:18
*** dprince has joined #openstack-mistral13:24
therveapetrich, Yeah looks like another patch is the issue13:35
apetrichtherve, please don't tell me it is not an even older patch :)13:36
therveapetrich, https://review.openstack.org/#/c/455174/ apparently13:36
therveWhich is not a big surprise given how much of a mess that was13:36
apetrichyeah I remember that13:37
thervehttps://github.com/openstack/python-mistralclient/compare/3.1.4...3.2.0 is depressing13:40
openstackgerritBrad P. Crochet proposed openstack/python-mistralclient master: Add script for unit test coverage job  https://review.openstack.org/44846513:49
openstackgerritThomas Herve proposed openstack/python-mistralclient master: Don't override session during auth  https://review.openstack.org/54346013:51
therved0ugal, apetrich ^^^13:51
apetrichtherve, good catch13:52
*** kong has quit IRC13:55
*** toure|gone is now known as toure13:59
openstackgerritMike Fedosin proposed openstack/mistral master: Cache action definitions  https://review.openstack.org/52021814:49
apetrichd0ugal, lol I think the crontrigger fix might be causing the failure on the therve's patch. Look around here and grep delete http://logs.openstack.org/60/543460/1/check/python-mistralclient-devstack-dsvm/8961435/job-output.txt.gz#_2018-02-12_14_27_22_50539614:55
d0ugaloh14:55
apetrichd0ugal, it is deleting the execution we expect to see and thus failing the test. the thing that deletes I think is a crontrigger from another test14:56
d0ugalapetrich: I don't follow - I see executions being deleted?14:56
d0ugalwhelp14:56
apetrichd0ugal, no I'm wrong14:57
d0ugallol14:57
* d0ugal waits14:57
apetrichsorry false alarm14:57
apetrichbut something is deleting the executions14:57
apetrichnot sure what14:57
d0ugalapetrich: I think that may be tempest cleaning up between tests14:57
d0ugalor at least the base test class cleaning up14:57
d0ugalI think tempest may do something automatically but there is code for it too14:58
apetrichnot during the tests it fails here https://github.com/openstack/python-mistralclient/blob/master/mistralclient/tests/functional/cli/v2/cli_tests_v2.py#L82114:59
d0ugalapetrich: oh, yeah, that isn't tempest14:59
d0ugalhrm14:59
d0ugalwhy are these tests so long14:59
apetrichbut the previous cli call has a "--marker %s" that is on that highlighted line on the log15:00
apetrichI have no idea where those deletes are coming from15:00
thrashd0ugal: apetrich I *think* (and this is from an old look at the code) that the tempest cleanup is a bit too aggressive.15:02
thrashbut I could be thinking of the tests in mistral itself.15:03
d0ugalI didn't even know we had tempest tests in mistralclient.15:03
thrashd0ugal: well, now you know. And knowing is half the battle.15:04
d0ugal:)15:04
thrash:D15:04
d0ugalso they say15:04
thrashthat's what G.I. Joe says...15:04
thrashhttp://images.huffingtonpost.com/2016-03-07-1457387813-3802365-PHOTO2.jpeg15:04
d0ugalapetrich, thrash - the deletes come from the execution_create function :)15:05
d0ugalof course15:05
d0ugalhttps://github.com/openstack/python-mistralclient/blob/master/mistralclient/tests/functional/cli/v2/base_v2.py#L244-L24915:06
d0ugalbut that should happen after the test has finished :/15:06
thrashd0ugal: race condition?15:06
d0ugaltempest can run tests in parallel, so I suspect it is something like that.15:07
d0ugalbut I don't fully understand how that would happen here15:07
d0ugalbecause the deletion is by UUID, so should be safe15:08
d0ugalapetrich: what does --marker do?15:17
openstackgerritMike Fedosin proposed openstack/mistral master: Cache action definitions  https://review.openstack.org/52021815:18
therved0ugal, Marker makes your filter start at that id15:22
d0ugalah15:22
d0ugalJust found that15:22
d0ugalI wonder if that is buggy15:23
*** dprince has quit IRC15:23
therved0ugal, Isn't the issue https://review.openstack.org/#/c/529078/ not being there yet?15:23
d0ugaltherve: We might need something like that. These are mistralclient tempest tests tho' - I am not sure they have the same issue15:24
d0ugalThey might15:24
d0ugalbut the deletions etc. I've seen are by UUID, so I think they should be safe15:25
therveOh, different tests, great15:25
d0ugal:)15:25
d0ugalIndeed, I only learned today we had tempest tests in mistralclient15:25
d0ugalI rechecked your patch, it is almost finished.15:25
d0ugalI am only going to keep digging deeper if this fails in the same way15:25
therveOh that failure15:27
therveYeah it's totally those tests being wrong15:27
thrashd0ugal: the --marker thing is not a bug... --marker is supposed to be 'everything *after* this uuid'15:28
thervethrash, Yeah, but the test doesn't provide stable order15:29
thrashtherve: nice15:29
therveIt creates 2 executions, and expects them to be ordered that way15:29
thrashtherve: doh15:29
d0ugalif the api sorted by creation by default I guess that'd work15:30
d0ugaland I thought it did15:30
therved0ugal, Right but the precision is probably a second15:30
d0ugalTrue15:30
therveSo if you create them too fast, it's not reliable15:30
d0ugalso... sleep(1)!15:30
thrashwhen in doubt, sleep!15:31
* therve goes get the gasoline15:31
d0ugallol15:31
thrashtherve: planning to go out by fire?15:31
thervethrash, Please bring the matches!15:31
* thrash yields a flamethrower.15:31
therveAlso every time you put sleep in a test, a baby polar bear dies15:32
thrash*wields... *sigh*15:32
d0ugalI don't want to add a sleep15:33
d0ugalbut I also don't want to rewrite this test now15:33
d0ugaland I do want the fix to land15:33
therveConundrum!15:34
d0ugalI have a plan15:34
therveDoes it involve a ticket to the Bahamas?15:35
d0ugalThat is plan b15:35
therveTHAT is a good plan15:35
d0ugalCan we assume that the order of "execution-list" will be consistent?15:36
therveI *think* so15:36
d0ugalI *hope* so15:37
therveYeah otherwise it's another bug to fix15:38
therveBut it sounds like you want to slightly rewrite the test :)15:39
d0ugalSlightly15:39
d0ugaland technically I still don't want to15:39
*** toure is now known as toure|biab15:41
openstackgerritDougal Matthews proposed openstack/python-mistralclient master: Fix a race condition with execution creation  https://review.openstack.org/54350115:44
d0ugalThe --sort_keys part of this test is laughable15:45
d0ugalhttps://github.com/openstack/python-mistralclient/blob/master/mistralclient/tests/functional/cli/v2/cli_tests_v2.py#L823-L83015:45
d0ugalThey are both there, clearly must be sorted corrctly!15:45
*** dprince has joined #openstack-mistral15:48
openstackgerritDougal Matthews proposed openstack/python-mistralclient master: Fix a race condition with execution creation  https://review.openstack.org/54350115:55
openstackgerritChen Eilat proposed openstack/mistral master: Change 403 exception message to something more informative and accurate  https://review.openstack.org/54351215:57
*** thrash is now known as thrash|biab16:00
apetrichd0ugal, I like the pragmatic fix16:04
d0ugalHopefully it works.16:04
*** itlinux has joined #openstack-mistral16:08
*** itlinux has quit IRC16:10
*** itlinux has joined #openstack-mistral16:22
*** itlinux has quit IRC16:23
*** thrash|biab is now known as thrash16:27
*** bobh has joined #openstack-mistral16:30
*** itlinux has joined #openstack-mistral16:35
*** toure|biab is now known as toure16:57
rakhmerovd0ugal: I support the idea with sorting. That is reliable. Other 2 options: add a auto-increment field to DB schema so that it reliably orders by creation time or don't expect any order in the test17:00
d0ugalrakhmerov: it seems like overkill to add that to the db just for a few tests :)17:04
rakhmerovwell, it's not for that test )17:04
rakhmerovit's a general problem17:04
rakhmerovwe can't reliably sort things when fetching from API17:05
rakhmerovdepends on how DB works (which may be different)17:05
d0ugalYeah, truee17:05
d0ugalI'm just not sure if that is an issue otherwise17:05
*** jpich has quit IRC17:07
rakhmerovd0ugal: not a big issue probably, but once in a while someone who writes client code working with Mistral API stumbles upon it17:27
rakhmerovwe've seen a number of people17:27
rakhmerovand every time I had to explain the root of the problem17:28
rakhmerovso that they don't rely on order17:28
*** mcdoker181818 has quit IRC17:34
*** AlexeyAbashkin has quit IRC17:36
apetrichno implicit is pretty standard on places like mongo and that has been there now for a long time also on HA databases so if they don't ask for a sort key I feel taht the order that it comes one is up to the underlaying infra and should be considered random17:39
apetrich#my2cents :)17:39
apetrichno implicit order*17:39
*** mcdoker181818 has joined #openstack-mistral17:42
openstackgerritSean McGinnis proposed openstack/mistral-extra stable/pike: Fix tox.ini to use new constraints list management style  https://review.openstack.org/54356317:50
*** harlowja has joined #openstack-mistral18:14
*** akazakov has quit IRC18:45
*** itlinux has quit IRC18:49
*** itlinux has joined #openstack-mistral18:51
*** itlinux has quit IRC18:53
*** itlinux has joined #openstack-mistral18:55
*** dprince has quit IRC18:58
*** dprince has joined #openstack-mistral19:11
*** jtomasek has quit IRC19:14
*** dprince has quit IRC19:15
*** dprince has joined #openstack-mistral19:15
*** mcdoker181818 has quit IRC19:23
*** mcdoker181818 has joined #openstack-mistral19:26
*** bobh has quit IRC19:27
openstackgerritDougal Matthews proposed openstack/python-mistralclient master: Don't override session during auth  https://review.openstack.org/54346019:44
openstackgerritDougal Matthews proposed openstack/python-mistralclient master: Add TripleO jobs for mistral client  https://review.openstack.org/54344719:44
*** bobh has joined #openstack-mistral19:57
*** mcdoker181818 has quit IRC20:19
*** dprince has quit IRC20:19
*** kong has joined #openstack-mistral20:28
*** livelace-link has joined #openstack-mistral20:44
*** itlinux has quit IRC20:51
*** bobh has quit IRC21:00
openstackgerritMerged openstack/python-mistralclient master: Remove broken tox-cover job  https://review.openstack.org/54206521:05
openstackgerritMerged openstack/mistral-extra stable/pike: Fix tox.ini to use new constraints list management style  https://review.openstack.org/54356321:19
*** apetrich has quit IRC21:32
*** apetrich has joined #openstack-mistral21:33
*** mcdoker181818 has joined #openstack-mistral21:56
*** bobh has joined #openstack-mistral22:00
*** bobh has quit IRC22:04
*** bobh has joined #openstack-mistral22:06
*** bobh has quit IRC22:07
*** bobh has joined #openstack-mistral22:08
*** threestrands has joined #openstack-mistral22:12
*** katkapilatova has quit IRC22:55
*** AlexeyAbashkin has joined #openstack-mistral23:21
*** AlexeyAbashkin has quit IRC23:26
*** bobh has quit IRC23:32
« Sunday, 2018-02-11 Index Tuesday, 2018-02-13 »

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!