09:03:14 <gmann> #startmeeting qa
09:03:15 <openstack> Meeting started Thu Nov 22 09:03:14 2018 UTC and is due to finish in 60 minutes.  The chair is gmann. Information about MeetBot at http://wiki.debian.org/MeetBot.
09:03:16 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
09:03:18 <openstack> The meeting name has been set to 'qa'
09:03:23 <gmann> who all here today
09:03:27 <mbindlish> Hi
09:04:19 <gmann> mbindlish: hi
09:05:20 <gmann> let's start
09:05:23 <gmann> #link https://wiki.openstack.org/wiki/Meetings/QATeamMeeting
09:05:28 <gmann> agenda ^^
09:05:50 <gmann> but if it two of us then we can directly discuss the required topic
09:06:00 <gmann> mbindlish: did you start the bug triage things ?
09:06:35 <mbindlish> i have started but this time doesn't have updated one
09:06:37 <mbindlish> https://etherpad.openstack.org/p/QA-Stein-Bug-Triage
09:07:02 <mbindlish> by next week will update with proper things
09:07:37 <gmann> we reached to 30 new bugs :(
09:07:55 <gmann> mbindlish: ok any bug you investigated or want to disucss ?
09:08:01 <mbindlish> yes, its outdated. In next week. I will share the updated one
09:08:13 <mbindlish> yes I need to discuss one of issue
09:08:27 <mbindlish> https://review.openstack.org/#/c/577730/
09:09:11 <mbindlish> Test is always failing, didn't get the proper reason. Sometimes got failed with py35, sometimes with py36
09:09:18 <gmann> that is patch we can discuss.
09:09:23 <mbindlish> yes
09:09:25 <gmann> i eman from bug
09:09:28 <gmann> mean
09:09:32 <mbindlish> ok, not yet
09:10:24 <gmann> ok, and can you also add the comparison number from previous week so that we can get to know the progress of that week. like new 30 ->28
09:11:03 <mbindlish> @gmann I was thinking to append in previous week, with a seprator line. Is it ok?
09:11:30 <gmann> did not get
09:12:15 <mbindlish> I mean i will share the report for first week, on above of that will share the report for 2nd week and so on...
09:12:22 <mbindlish> if 1st week has 30 new bugs
09:12:32 <mbindlish> 2nd will have 28 new bugs in total
09:12:41 <mbindlish> 3rd will have 25 new bugs
09:13:00 <gmann> that is good, but in current week you can add like x->y which will be very clear
09:13:09 <mbindlish> sure
09:13:13 <gmann> thanks
09:13:26 <gmann> back to your patch
09:13:34 <mbindlish> yeah
09:15:33 <gmann> what is issue
09:16:06 <mbindlish> zuul is failing as test case is failing. If test pass for py27 then same test fail for py35
09:16:08 <gmann> did you check the felipe comment ?
09:16:29 <mbindlish> @gmann yes I checked but not able to conclude the solution
09:16:58 <gmann> humm, he commented on nov 3 seems but did not see your comment on that.
09:17:44 <mbindlish> actually I understood to call the function from entry point which
09:17:48 <gmann> mbindlish: one good practice, do not wait for such discussion for office hour. you can discuss on gerrit also or catch the person on same day/night if possible. otherwise gerrit is best async communication
09:17:52 <mbindlish> i will do
09:18:31 <mbindlish> but that fix seems not the conclusion, wich will still give the same error
09:19:24 <gmann> that is ok. i mean you can always start discussion on patch also for valid/invalid or question etc
09:19:45 <mbindlish> @gmann sure
09:20:12 <gmann> mbindlish: you mean by setting the non null value will give same error ?
09:20:48 <mbindlish> there are 2 test cases: 1 with some value, 2nd with empty data
09:21:03 <mbindlish> empty data test case is failing for random env
09:21:20 <mbindlish> like if it got pass in py27 it will fail in py35
09:21:33 <mbindlish> or if it got pass in py35 then it may fail in py27
09:21:56 <mbindlish> it is consistent
09:22:03 <mbindlish> inconsistent*
09:22:05 <gmann> i see, did you find the reason
09:22:31 <mbindlish> no :( so i am asking
09:23:25 <gmann> so what i understood form comment is other parallel test changing the CONF value and test expecting the empty value will fail
09:23:57 <mbindlish> yes
09:27:14 <gmann> mbindlish: in both test, i can see that issue
09:27:40 <mbindlish> ohh
09:29:00 <gmann> as CONF is global path and you can see the value can change from other tests
09:29:37 <gmann> that was one of the reason i asked you to test it via more inner check instead of checking CONF._path
09:30:13 <mbindlish> what if I am reassigning blank value, shouldn't it will update?
09:30:26 <gmann> and that you can do by setting some config value and then checking those value in innner most function
09:30:47 <gmann> mbindlish: how you will reassign and control not other test run in between
09:31:11 <mbindlish> hmm..
09:31:30 <gmann> let's check with config options value instead of global var check
09:32:06 <mbindlish> I think there was no feasiblity of setting config value, but again I will check
09:33:47 <gmann> mbindlish: commented on patch
09:34:38 <gmann> keep updating on gerrit about your finding, do not hold till irc disucssion.
09:34:47 <mbindlish> @gmann ok
09:35:03 <gmann> i can check that test later
09:35:14 <gmann> anything else you have to discuss?
09:35:14 <mbindlish> One more thing I want to ask other than this
09:35:20 <gmann> sure
09:35:45 <mbindlish> Can we add debug env in tox like below:
09:35:45 <mbindlish> [testenv:debug]
09:35:45 <mbindlish> basepython = python3
09:35:45 <mbindlish> commands = oslo_debug_helper -t tempest/tests {posargs}
09:36:32 <mbindlish> to run the tox with debug, I have checked in congress for such thing
09:37:17 <gmann> does not existing env put debug logs ?
09:38:23 <mbindlish> I think it does for failures only
09:40:52 <gmann> or with --debug you can run
09:41:04 <gmann> i am not against of, having that is useful i think.
09:41:47 <gmann> mbindlish: anything else ?
09:41:51 <mbindlish> I think i tried with --debug, but it doesn't work
09:42:05 <gmann> ok, you can add that is ok.
09:42:39 <mbindlish> @gmann ok, thanks
09:42:57 <mbindlish> @gmann : nothing else from my side
09:43:13 <gmann> let's close then.
09:43:14 <gmann> thanks
09:43:20 <mbindlish> thanks
09:43:21 <gmann> #endmeeting