Thursday, 2016-11-17

*** bobh has joined #openstack-mistral00:02
*** bobh has quit IRC00:09
*** bobh has joined #openstack-mistral00:12
*** bobh has quit IRC00:18
*** bobh has joined #openstack-mistral00:34
*** bobh has quit IRC00:36
*** bobh has joined #openstack-mistral00:36
*** bobh has quit IRC00:41
*** rrecio_ has quit IRC00:51
*** bobh has joined #openstack-mistral01:58
*** bobh has quit IRC02:29
*** bobh has joined #openstack-mistral03:13
*** bobh has quit IRC03:17
*** bobh has joined #openstack-mistral03:22
*** bobh has quit IRC03:41
*** bobh has joined #openstack-mistral03:46
*** bobh has quit IRC04:28
*** thrash|g0ne has quit IRC04:51
*** thrash has joined #openstack-mistral04:51
*** thrash has joined #openstack-mistral04:51
*** hparekh has joined #openstack-mistral04:52
*** jaosorior has joined #openstack-mistral06:14
rakhmerovd0ugal: woow!06:23
rakhmerovwhat was that regression?06:23
rakhmerovbefore merging it let's please discuss it06:23
rakhmerov^Gal^: you here?06:32
rakhmerovhi06:32
openstackgerritMerged openstack/mistral-dashboard: Remove mox in test-requirement.txt  https://review.openstack.org/39777006:33
openstackgerritMerged openstack/mistral-dashboard: Adding files to .gitignore  https://review.openstack.org/39765606:38
openstackgerritMerged openstack/mistral-dashboard: Cleanup unused files  https://review.openstack.org/39764806:38
*** jaosorior has quit IRC06:41
*** jaosorior has joined #openstack-mistral06:42
openstackgerritMerged openstack/mistral-dashboard: mistral-dashboard: entities name change for system consistency  https://review.openstack.org/39763706:46
rakhmerovd0ugal, ddeja, hparekh: guys, could you please review a bunch of small patches?06:53
rakhmerovwhere I gave +206:53
openstackgerritMerged openstack/python-mistralclient: Add cancelled state to action executions  https://review.openstack.org/38470007:10
openstackgerritMerged openstack/python-mistralclient: Updated from global requirements  https://review.openstack.org/39861807:10
openstackgerritMerged openstack/python-mistralclient: Remove unused pylint  https://review.openstack.org/39764207:10
* ddeja grabs cup of tea and start reviewing07:19
ddejarakhmerov ^07:19
rakhmerovok!07:19
rakhmerovthank you07:19
ddejarakhmerov: btw, small question: do you have some list of companies that uses Mistral on production which you can share?07:25
rakhmerovsharatss: hi, can you please read my comment on https://review.openstack.org/#/c/396763/07:25
rakhmerovddeja: honestly, I'm afraid that I don't know well at this point. I have some in my head but this list is not full obviously. This question was recently raised by some of my managers and I thought that we need to find out07:27
rakhmerovI'm thinking about creating something like a survey or at least take stats from PyPi (but we need to figure out how to filter out OpenStack CI downloads from it)07:29
ddejarakhmerov: OK, thanks07:30
rakhmerovyeah, sorry07:30
ddejano, no problem07:30
rakhmerovI need to know this better but I need to figure out how to gather this info07:30
ddejajust this crazy bug from last week07:31
rakhmerovyes07:31
ddejawith hanigng transactions on OVH production07:31
rakhmerovbut you know, surprisingly we didn't come across this bug at Nokia07:31
ddejaand I started to wonder who else started to use mistral, but if there is no info, that's ok07:31
rakhmerovnobody noticed it, maybe because we don't make too many API calls07:32
ddejathat may be the case07:32
rakhmerovit's rather our workflows a big, and in the engine we didn't have this issue07:32
rakhmerovonly when using REST API07:32
ddejaOK07:33
rakhmerovddeja: you can count Nokia, StackStorm, TripleO, OVH (I don't know much about OVH though)07:33
rakhmerovmaybe Huawei too, but I'm not 100% sure07:34
rakhmerovthey wanted to use it07:34
ddejaoh, cool07:34
rakhmerovbut through StackStorm there's a huge number of Mistral users07:34
rakhmerovI just don't have a full list07:34
rakhmerovbut I know it includes companies like Symantec, Cisco etc.07:35
rakhmerovas far as that bug, this is actually my fault because I saw strange behavior in some of our unit tests before but didn't drill down to the root cause. In some of the tests if we made DB API call lazy loading still worked even outside of open transaction07:36
ddejarakhmerov: and technical question: do you think failur of dsvm gate in this patch https://review.openstack.org/#/c/397125/1 is related to the patch itself?07:37
rakhmerovI should have found a reason a while ago but postponed it07:37
rakhmerovlet me see..07:37
ddejabecouse in other recent patch we got all success07:37
rakhmerovno, I'm sure it's not related07:37
rakhmerovI know the reason of this failure07:37
rakhmerovgonna fix it07:38
ddejaoh, ok07:38
ddeja+2 +W then07:38
rakhmerovok07:38
openstackgerritMerged openstack/mistral-dashboard: Fixed action screen "run" button CSS issue  https://review.openstack.org/39836107:42
openstackgerritMerged openstack/mistral-lib: Added the Options for Internationalization output in conf.py  https://review.openstack.org/39767507:42
rakhmerovddeja: that failure has already been fixed actually. This patch was just checked earlier than we fixed this07:43
ddejaoh, ok07:45
ddejarakhmerov: just found this https://github.com/Tesora/tesora-mistral It looks like we can add another complany to the list...07:57
rakhmerovwhat the..07:58
rakhmerov:)07:58
rakhmerovooh, it's a fork07:58
rakhmerovI see07:58
sharatssrakhmerov: yes i read it07:58
rakhmerovddeja: yeah, I knew they were going to look at it but didn't know about any uses07:59
sharatssrakhmerov: i can take up the job of replacing MagicMock07:59
rakhmerovsharatss: sure, go ahead :)07:59
sharatssrakhmerov: thanks :)07:59
rakhmerovsharatss: are you ok to merge Winson's patch as is?07:59
sharatssrakhmerov: let it merge. As a part of my change i will replace it again08:00
rakhmerovddeja: I had a couple of meetings with them about a year ago08:00
rakhmerovsharatss: ok, good08:00
rakhmerovsharatss: please remove your -1 then08:01
sharatssrakhmerov: done :)08:01
ddejarakhmerov: ok, I see08:01
openstackgerritMerged openstack/mistral: Replace retrying with tenacity  https://review.openstack.org/39840808:08
openstackgerritMerged openstack/mistral: Added releasenote for Senlin Action Pack  https://review.openstack.org/39759208:08
*** shardy_afk is now known as shardy08:15
openstackgerritSharat Sharma proposed openstack/python-mistralclient: Removed the extra space from tox.ini  https://review.openstack.org/39768308:19
openstackgerritMerged openstack/python-mistralclient: Added the reno for stable/mitaka stable/newton and stable/liberty  https://review.openstack.org/39773608:21
*** mgershen has joined #openstack-mistral08:27
*** mgershen has quit IRC08:28
d0ugalrakhmerov: hey08:43
d0ugalrakhmerov: The transaction patches broke mistral briefly. sharatss noticed it because devstack *is* voting on python-tripleoclient08:43
rakhmerovyes, I'm here08:44
rakhmerovjust a sec08:44
d0ugalrakhmerov: there was an open patch to fix it, but we didn't make sure they merged at the same time08:44
rakhmerovyou mean that test_action_execution[sanity] test?08:44
rakhmerovor something else?08:44
rakhmerovyes, I saw that too08:45
d0ugalI forget which one exactly08:45
rakhmerovit was fixed by another patch that xavierhardy sent08:45
d0ugalbut simple things like "mistral workflow-list" failed08:45
d0ugalYeah08:45
rakhmerovooh, really?08:45
rakhmerovaaah, yes08:45
rakhmerovafter only the first patch, yes08:45
rakhmerovindeed08:45
rakhmerovso, now it's ok, right?08:45
d0ugalYeah, as far as I know08:46
*** mgershen has joined #openstack-mistral08:46
d0ugalbut I think we should have the job voting, I don't see any reason why we wouldn't :)08:46
xavierhardy\o/08:47
xavierhardysorry, wrong window08:47
rakhmerovd0ugal: well, let me think.. There was a reason actually to make it non-voting08:47
rakhmerove.g. some tests that were not stable enough (because of how they are designed)08:48
rakhmerovlike the test for ssh_proxied action08:48
rakhmerovhm..08:48
rakhmerovakuznetsova: hi Nastya, do you remember why we didn't want to make our devstack gate voting?08:49
rakhmerovd0ugal: actually, I want we to make it voting but I just wanna make sure it's ok08:49
*** jpich has joined #openstack-mistral08:49
rakhmerovd0ugal: I think I'm ok to make it voting. In case some of the unstable tests (those that fail due to their design itself) give us some consistent pain we can disable them for now till we fix them08:51
rakhmerovI don't want them to slow down our dev process08:51
d0ugalrakhmerov: we should try and make it more stable :)08:51
rakhmerovyes, I agree08:51
d0ugalrakhmerov: it is already voting on python-mistralclient - I don't see it being a problem htere08:52
d0ugalthere08:52
d0ugal(except when we broke it recently)08:52
rakhmerovddeja, hparekh, kong, akuznetsova: guys, maybe you have any objections?08:52
rakhmerovit's still kinda worrisome for me08:52
d0ugalhaha, okay08:52
d0ugalwe can always make it non-voting again if it is a problem and we are unable to fix it quickly08:52
rakhmerovon the client they are different08:52
rakhmerovyeah08:53
rakhmerovok, if there's not any other objections let's do it08:53
rakhmerovcourage is what will help us move forward :)08:53
sharatssrakhmerov: d0ugal i also insist on fixing py35 gate jobs on client08:53
rakhmerovit's high time we did it I guess08:53
rakhmerovsharatss: ok, we will08:54
d0ugalsharatss: makes sense08:54
sharatssrakhmerov: d0ugal i have a doubt08:54
rakhmerovabout what?08:54
sharatsswhy was the openstack/common removed from all clients? any particular reasons for it?08:55
rakhmerovwasn't it you who did this? :)08:55
rakhmerovI thought it was you :)08:55
sharatssrakhmerov: definitely not :P08:55
hparekhd0ugal: rakhmerov Yeah tests are not stable and we shavr to make it stable but as of now IMHO it should be non-voting08:55
rakhmerovhm... ok08:56
rakhmerovsharatss: I guess it was because we just don't use openstack/common08:56
rakhmerovwhich is ok08:56
hparekhsharatss: May be pawnesh has done it08:56
hparekhI guess you know him08:57
rakhmerovhparekh: yeah, this is my concern too but as far as I remember there's only one really painful test (for action std.ssh_proxied) that fails pretty often08:57
sharatsshparekh: yes i do08:57
rakhmerovhparekh: recently I didn't see a lot of other failures actually08:57
rakhmerovsharatss: so what is your concern about openstack/common?08:58
sharatsshparekh: rakhmerov: the fix for py35 was within the common directory08:58
rakhmerovooh08:58
sharatssrakhmerov: in cliutils.py08:58
hparekhrakhmerov: Yes right. I will look into this and submit a patch if I can find the reason08:58
rakhmerovpawnesh: ^08:58
rakhmerovpawnesh: can you comment?08:58
pawneshhparekh: i just removed client.py from openstack/common08:59
rakhmerovhparekh: thanks man, Nikolay actually looked at it but didn't figure out why it fails sometimes, maybe it's something with the test env itself (it's pretty hairy)08:59
rakhmerovhparekh: so, what's your opinion? Is it ok to make the gate voting and if that test turns to be too painful for us just disable it for now?09:00
rakhmerovand then fix it asap and enable back09:00
rakhmerovpawnesh, sharatss: guys, so what do we need to do? Revert cliutils.py back or what?09:01
rakhmerovsorry, I'm not too strong at that client side stuff09:01
sharatssrakhmerov: hparekh: if we want py35 as voting as of now its better to revert it09:01
sharatssrakhmerov: if we have time to fix it maybe we can fix it and make it to voting09:02
hparekhrakhmerov: How about all core reviewers check the tests failed in the gate and ignore if it is ssh_proxied one and if test is failing then we can investigate. Meanwhile  I am looking into this test and try to make it more stable09:02
rakhmerovsharatss: yes, sure. My understanding is that we don't have too much urgency in making it voting. Do whatever you think is needed. Be sure to discuss with pawnesh since he made this change09:04
sharatssrakhmerov: sure i might need hparekh help09:05
rakhmerovhparekh: hm.. we can do it. I usually look at the tests, but you know, for that exact patch I forgot :) And everybody else did. Which led to a serious regression09:05
rakhmerovd0ugal: what do you think?09:05
* d0ugal reads up09:05
d0ugalsorry, I was getting coffee09:05
rakhmerovhparekh, d0ugal: how about the following? For now we keep it non-voting and hparekh is looking into how to make that test more stable?09:06
hparekhsharatss: Sure I will help you. Let me know09:06
sharatsshparekh: thanks :)09:06
rakhmerovwith the promise that we'll be investigating every failure09:06
d0ugalrakhmerov: sure, or I can look into it09:06
d0ugalI do see one failure that looks invalid at the moment09:06
rakhmerovwe can keep that patch in infra for not (w/o merging it)09:07
rakhmerovonce we fix the test we can immediately merge the patch into infra09:07
rakhmerovd0ugal: which one?09:07
*** jaosorior is now known as jaosorior_lunch09:08
d0ugalrakhmerov: https://review.openstack.org/#/c/397761/09:08
d0ugalhttp://logs.openstack.org/61/397761/2/check/gate-mistral-devstack-dsvm/68f012f/console.html#_2016-11-15_16_08_18_34217609:08
d0ugalrakhmerov: acutally, ignore me09:08
rakhmerovd0ugal: I already looked at it, it's fixed09:08
d0ugalrakhmerov: that is the error from when we broken it recently09:08
d0ugalYeah09:08
rakhmerovyes :)09:08
* d0ugal is still waking up09:08
rakhmerovhaha :)09:08
rakhmerovhave your coffee man! :))09:09
d0ugalI am used to working with americans, you are all testing my brain about 4 hours earlier than normal09:09
rakhmerovnp09:09
rakhmerovhaha :))09:09
sharatssrakhmerov: and you doubt me so much :(09:09
rakhmerovd0ugal: so, are you ok to hold on with making it voting? Hopefully, for just a few days09:10
rakhmerovsharatss: what do you mean?09:10
sharatssrakhmerov: u thought that i had removed the directory :P09:10
rakhmerovsorry, my fault09:10
sharatssrakhmerov: haha :D i was kidding09:11
rakhmerovI don't always remember who has done what09:11
rakhmerov:))09:11
d0ugalrakhmerov: sure, I abandonded the change to be safe. I can restore it later.09:11
rakhmerovok09:11
rakhmerovd0ugal, hparekh: I would prefer if we had some period within we need to make that test stable09:12
rakhmerovhow about a week?09:12
rakhmerovsay, if it's not done by the end of the next week we disable the test for now make the gate voting09:12
hparekhyeah sounds good09:13
rakhmerovI think it would be fair, otherwise it may take too long09:13
d0ugalsure, sounds good09:14
*** DaveTurner has quit IRC09:14
d0ugalI have never used tempest before, I'll have to look at how it works :)09:14
* ddeja also thinks we can wait a week09:17
rakhmerovd0ugal: not a rocket sience )09:19
rakhmerovddeja: ok!09:19
d0ugalsharatss: I was looking into the openstack/common being deleted.09:29
d0ugalsharatss: https://github.com/openstack/python-mistralclient/commit/b1614333eb8c3b14aa84b185e8d02aebe8da2a0d09:29
d0ugalsharatss: and https://github.com/openstack/python-mistralclient/commit/b3805b6d082fe59910c64f6a0852de53353b050409:29
sharatssd0ugal: same here09:29
d0ugalsharatss: oh, cool :)09:30
d0ugalsharatss: so it seems that we just moved away from it and everything left was deleted09:30
d0ugalsharatss: so I don't think it is an issue?09:30
sharatssd0ugal: i am trying a small change and running tox. let me see if py35 passes09:30
*** testac3 has joined #openstack-mistral09:31
*** testac3 has left #openstack-mistral09:31
d0ugalsharatss: cool :)09:31
sharatssd0ugal: maybe we can solve this with the existing code without  reverting back anything09:31
sharatss:)09:31
*** jaosorior_lunch is now known as jaosorior09:43
*** openstackgerrit has quit IRC09:48
*** openstackgerrit has joined #openstack-mistral09:49
d0ugalsharatss: it looks like python35 is failing due to deprecation warnings09:49
d0ugalsharatss: but they don't come from anything we use09:50
d0ugalsharatss: I mean, they are caused by libraries we use, not mistral itself09:50
d0ugalso I think they need to be fixed in those packages?09:50
openstackgerritMerged openstack/python-mistralclient: Removed the extra space from tox.ini  https://review.openstack.org/39768310:05
*** shardy is now known as shardy_mtg10:16
openstackgerritRenat Akhmerov proposed openstack/mistral: Add timestamp at the bottom of every page  https://review.openstack.org/39776110:16
openstackgerritMerged openstack/mistral: Migrated to the new oslo.db enginefacade  https://review.openstack.org/39712510:22
openstackgerritMerged openstack/mistral: Use mock for a bad HTTP call in unit tests  https://review.openstack.org/39676310:22
sharatssd0ugal: yes. i dont think adding python warnings broke py35 right?10:43
d0ugalsharatss: You don't think they did?10:46
d0ugalI think they did10:47
d0ugalhttp://logs.openstack.org/83/397683/3/check/gate-python-mistralclient-python35-nv/547640c/console.html#_2016-11-17_08_23_04_60118510:47
d0ugalFor example ^10:47
sharatssd0ugal: now i get a new error10:47
d0ugalThat test checks that stderr is empty, but it isn't10:47
d0ugaland it isn't empty because of deprecation warnings, but they come from the stdlib in areas we don't use10:47
d0ugalbut they are used by cliff10:47
sharatssd0ugal: shall i test by removing the deprecation warnings?10:48
d0ugalsharatss: If you like.10:49
d0ugalsharatss: I would guess that somebody is probably already working on fixing it in cliff10:49
d0ugalso maybe we should wait for that?10:49
sharatssd0ugal: but the tests are passing in other projects10:50
sharatssd0ugal: take a look at this now http://paste.openstack.org/show/589560/10:50
d0ugalsharatss: oh, I didn't know that bit. Interesting.10:50
sharatssd0ugal: this is what i get after running tox now10:51
d0ugalsharatss: that is the same error that I linked in the logs ^10:51
sharatssd0ugal: yes. AssertionError. err...10:52
d0ugalYou run your tests as root! :-o10:53
d0ugalsharatss: a new version of cliff was tagged two days ago10:54
sharatssd0ugal: hmm.. so we will wait for the fix from cliff side?10:56
d0ugalsharatss: https://github.com/openstack/cliff/blob/master/cliff/commandmanager.py#L9110:57
d0ugalThat is where one deprecation warning comes from10:57
d0ugaland the other comes from here: https://github.com/openstack/cliff/blob/master/cliff/app.py#L38710:58
d0ugalso I don't think it has been fixed in cliff.10:58
d0ugalMaybe other projects just ignore the deprecation warnings? or maybe they don't check for them?10:58
*** shardy_mtg is now known as shardy11:00
sharatssd0ugal: are you planning to propose it in cliff?11:04
d0ugalsharatss: no, no plans at the moment11:04
* d0ugal is already wearing too many hats11:04
sharatssd0ugal: haha :) ok i will see if i can11:05
d0ugalsharatss: I wonder if the new method is in python211:05
d0ugalI guess that might be why they have not changed it yet11:05
sharatssd0ugal: so what do you say? we will wait for them?11:06
d0ugalsharatss: Yeah, it is Python 3 only11:06
d0ugalsharatss: I guess there isn't an easy fix :)11:06
d0ugalsharatss: so either ignore the deprecation warnings, or update the test11:06
*** khamtamtun has joined #openstack-mistral11:06
d0ugalChecking for stderr is not very reliable because of possible warnings like this11:07
sharatssd0ugal: you are right. I checked in projects that are passing 3511:08
sharatssd0ugal: they are ignoring deprecation warnings11:08
d0ugalsharatss: cool, then we should do that :)11:08
sharatssd0ugal: checking..11:10
openstackgerritSharat Sharma proposed openstack/python-mistralclient: Removing deprecation warnings to pass py35  https://review.openstack.org/39893111:11
sharatssd0ugal: ^11:11
*** khamtamtun has quit IRC11:12
d0ugalsharatss: cool, looks good - we shall see if that helps with py35 :)11:13
sharatssd0ugal: i think it should11:13
d0ugalIt seems most projects just ignore them all, I thought they might just ignore specific warnings11:13
d0ugalYeah, I am running it locally to see.11:13
d0ugalI guess I am impatient11:14
sharatssd0ugal: please check it11:14
*** kamtamtun has joined #openstack-mistral11:14
d0ugalsharatss: passed for me11:20
sharatssd0ugal: what's taking jenkins so long? :))11:21
d0ugalsharatss: tripleo11:21
d0ugal:)11:21
d0ugalactually, no, tripleo doesn't CI against mistralclient11:21
sharatssd0ugal: ok. will wait for it to pass11:22
d0ugalsharatss: it is passing11:23
d0ugalsharatss: just waiting for devstack, the others passed11:23
sharatssd0ugal: hmm...11:24
d0ugaldevstack is the slowest :)11:24
sharatssd0ugal: it wont fail anyway. so we are safe :)11:26
*** hparekh has quit IRC11:27
*** sharatss has quit IRC11:31
*** sharatss has joined #openstack-mistral11:32
*** kamtamtun has quit IRC11:45
*** kamtamtun has joined #openstack-mistral11:52
*** kamtamtun has quit IRC11:53
*** kamtamtun has joined #openstack-mistral12:04
*** kamtamtun has quit IRC12:05
*** catintheroof has joined #openstack-mistral12:46
*** sharatss has quit IRC12:57
*** sharatss has joined #openstack-mistral12:57
*** chlong has joined #openstack-mistral13:00
*** bobh has joined #openstack-mistral13:01
*** bobh has quit IRC13:01
*** bobh has joined #openstack-mistral13:01
*** thrash is now known as thrash|g0ne13:06
*** sharatss has quit IRC14:18
*** sharatss has joined #openstack-mistral14:19
*** bobh has quit IRC14:20
*** jaosorior has quit IRC14:41
*** mgershen1 has joined #openstack-mistral14:45
*** mgershen has quit IRC14:47
*** mgershen1 has quit IRC15:00
openstackgerritOpenStack Proposal Bot proposed openstack/mistral: Updated from global requirements  https://review.openstack.org/39905215:01
*** mgershen has joined #openstack-mistral15:02
*** janki has joined #openstack-mistral15:22
*** bobh has joined #openstack-mistral15:55
openstackgerritMichal Gershenzon proposed openstack/mistral-specs: Yaql Tasks Function  https://review.openstack.org/39419416:24
openstackgerritMichal Gershenzon proposed openstack/mistral-specs: Yaql Tasks Function  https://review.openstack.org/39419416:28
*** rrecio has joined #openstack-mistral16:32
*** rrecio_ has joined #openstack-mistral16:36
*** rrecio has quit IRC16:36
*** janki has quit IRC17:04
*** DaveTurner has joined #openstack-mistral17:47
*** jpich has quit IRC18:47
*** shardy has quit IRC18:57
*** rbrady is now known as rbrady-afk19:28
*** gyee has joined #openstack-mistral19:38
*** openstackgerrit has quit IRC19:48
*** openstackgerrit has joined #openstack-mistral19:49
*** rbrady-afk is now known as rbrady20:12
*** bobh has quit IRC20:44
*** bobh has joined #openstack-mistral20:46
*** bobh has quit IRC20:51
*** rrecio_ has quit IRC21:31
*** jamielennox is now known as jamielennox|away21:40
*** rrecio has joined #openstack-mistral21:56
*** jamielennox|away is now known as jamielennox22:24
*** pawnesh has quit IRC23:17
*** catintheroof has quit IRC23:29
*** catintheroof has joined #openstack-mistral23:30
*** catintheroof has quit IRC23:35

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!