13:01:35 <chaconpiza> #startmeeting monasca
13:01:35 <opendevmeet> Meeting started Tue Jul  6 13:01:35 2021 UTC and is due to finish in 60 minutes.  The chair is chaconpiza. Information about MeetBot at http://wiki.debian.org/MeetBot.
13:01:35 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
13:01:35 <opendevmeet> The meeting name has been set to 'monasca'
13:01:51 <adriancz> Hi
13:01:59 <chaconpiza> Hi adriancz
13:02:10 <chaconpiza> The agenda as usual https://etherpad.opendev.org/p/monasca-team-meeting-agenda
13:03:35 <chaconpiza> Let's start
13:03:53 <chaconpiza> #topic [Ceilometer Monasca Client] Lots of warnings and stack traces during test runs
13:04:02 <chaconpiza> https://bugs.launchpad.net/ceilometer/+bug/1934220
13:04:24 <chaconpiza> created by Matthias Runge
13:05:22 <chaconpiza> I investigated the issue
13:05:32 <chaconpiza> and was able to reproduce it
13:06:33 <chaconpiza> In Ceilometer there are unittests related to Monasca, basically monasca-publisher, monasca-data-filter and monasca-client
13:07:01 <chaconpiza> The issue reported in Launchpad is from monasca-client
13:07:34 <chaconpiza> There are 8 unittest of Ceilometer's monasca-client
13:09:18 <chaconpiza> some of these tests check scenarios with a mocked wrong response
13:09:38 <chaconpiza> for example, unauthenticated user
13:10:28 <chaconpiza> all those tests work ok. But of course the stacktracke of the exception is displayed in the output
13:10:47 <chaconpiza> This is what Matthias don't like
13:11:21 <chaconpiza> What do you think Adrian?
13:12:29 <adriancz> also  I don't like this
13:13:12 <chaconpiza> The question, is it ok to display in the output of the tests stack traces of the mocked exceptions?
13:14:30 <adriancz> i don't like this exception in the output
13:16:21 <adriancz> in my opinion this look ugly and sometimes is hard to read test result because of that.
13:16:57 <adriancz> we should see exception only when test fails
13:17:08 <chaconpiza> Check this unittest from Nova:
13:17:08 <chaconpiza> https://zuul.opendev.org/t/openstack/build/85f0fc646ca949a3a07649cc26dcb5df/log/job-output.txt
13:17:36 <chaconpiza> Line No. 51901
13:17:47 <chaconpiza> (It takes a little bit to load)
13:18:47 <chaconpiza> Can you see that?
13:19:14 <adriancz> yes, I can
13:19:57 <adriancz> i don't like this output
13:20:01 <chaconpiza> So, it is a similar situation: A mocked exception or mocked class that leads to the exception, as expected
13:20:18 <adriancz> imagine this output  when you try to run tests in your local machine
13:20:29 <adriancz> and one test fail
13:20:32 <chaconpiza> and then the test finishes with the  `{4} nova.tests.unit.network.test_network_info.NetworkInfoTests.test_hydrate [0.031209s] ... ok`
13:20:44 <adriancz> and you try to search for exception
13:21:03 <chaconpiza> Yes, I think this is the point of Matthias.
13:21:31 <chaconpiza> in the other hand, all unittests in OpenStack present this behavior
13:23:06 <chaconpiza> I tried inside of the test to Try&Catch the exception and then Log some text about it. To replace the Traceback with the a Log file
13:23:21 <chaconpiza> But, unfortunately without success.
13:23:50 <chaconpiza> even Try&Catch with general Exception and BaseException doesn't work.
13:24:21 <chaconpiza> Adrian do you have any idea?
13:24:42 <adriancz> at this pond I have no idea
13:25:06 <adriancz> i would need check the code
13:25:17 <adriancz> and play with this unit tests
13:26:01 <chaconpiza> I will take a look on the monasca-client unittests from the monasca-pythonclient repo. And compare them with the monasca-client unittests from the Ceilometer.
13:26:30 <adriancz> +1
13:26:55 <chaconpiza> alright, anyways I will treat it with low priority, since nothing is really broken.
13:27:44 <chaconpiza> let's move forward
13:28:32 <chaconpiza> #topic [Monasca-Agent] Revert "Replace fnmatch with oslo.utils.fnmatch"
13:28:38 <chaconpiza> https://review.opendev.org/c/openstack/monasca-agent/+/799585
13:29:43 <adriancz> +2 from me
13:29:59 <chaconpiza> Because of some issues in PY2 with the standard fnmatch library, the oslo.utils.fnmach was used
13:30:00 <adriancz> i think we should merge this
13:30:10 <chaconpiza> Yes, I am agreed.
13:31:25 <chaconpiza> Same change is on Ceilometer, Barbican, etc.
13:32:06 <adriancz> do we need to cherry pick this to stable branches ?
13:32:50 <chaconpiza> I was thinking on that, because Wallaby, Victoria and Ussuri are PY3 only
13:33:34 <chaconpiza> I will wait a little bit before to merge it into master, to see if somebody from the other projects complains on that change.
13:35:03 <adriancz> Good idea
13:35:24 <chaconpiza> But I am almost sure that the change will be accepted.
13:35:25 <chaconpiza> Ok
13:35:41 <chaconpiza> #topic Monasca Images Build (Stable Branches)
13:36:28 <chaconpiza> Here as mentioned before, let's stop building everyday Train
13:37:12 <chaconpiza> Train is already in EM stage
13:37:13 <adriancz> I agree
13:37:15 <chaconpiza> https://review.opendev.org/c/openstack/releases/+/790758
13:37:27 <chaconpiza> No need to build it everyday
13:38:13 <chaconpiza> Beside of the Train branch, we have issue in the other stable branches
13:38:29 <chaconpiza> Wallaby ,Victoria and Ussuri
13:39:30 <chaconpiza> The only one that is ok (with exception of Train) is Monasca-Notification
13:40:06 <chaconpiza> Let's start to fix them, Adrian do you have any preference?
13:40:19 <chaconpiza> Persister, Api, Agent, Common, etc...
13:41:11 <adriancz> i think i can start with api or persister
13:41:59 <chaconpiza> Alright, take please persister I will take Api
13:42:09 <adriancz> I think w should first check common
13:42:27 <adriancz> because all image depend on this
13:42:42 <chaconpiza> Monasca-Common is building ok from master branch
13:42:52 <adriancz> perfect
13:43:10 <adriancz> Ok, I will take persister
13:43:36 <chaconpiza> persister from master use as a base image monasca-common from master
13:43:56 <chaconpiza> persister from Wallaby use as a base image monasca-common from master    too
13:44:12 <chaconpiza> persister from Victoria use as a base image monasca-common from master    too
13:44:14 <chaconpiza> etc...
13:44:43 <chaconpiza> this is a second change we need to do after fixing the images:
13:45:01 <chaconpiza> persister from Wallaby SHOULD use as a base image monasca-common from Wallaby
13:45:04 <chaconpiza> and so on
13:46:00 <chaconpiza> ok, those are all topic from the agenda
13:46:15 <chaconpiza> Do you have any other topic to discuss today?
13:48:23 <adriancz> No
13:48:37 <adriancz> i have no topic
13:48:46 <chaconpiza> Then let's close the meeting
13:48:54 <chaconpiza> Thanks see you next week
13:48:57 <adriancz> Thanks Martin
13:49:01 <adriancz> bye
13:49:32 <chaconpiza> #endmeeting