13:00:03 <chaconpiza> #startmeeting monasca
13:00:04 <openstack> Meeting started Tue Jan 12 13:00:03 2021 UTC and is due to finish in 60 minutes.  The chair is chaconpiza. Information about MeetBot at http://wiki.debian.org/MeetBot.
13:00:05 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
13:00:07 <openstack> The meeting name has been set to 'monasca'
13:00:30 <chaconpiza> Hi
13:00:47 <bandorf> Hi, everybody
13:00:48 <chaconpiza> The agenda as usual https://etherpad.opendev.org/p/monasca-team-meeting-agenda
13:00:55 <chaconpiza> Hello Matthias
13:01:36 <chaconpiza> Let's start with the topics
13:01:42 <chaconpiza> #topic Status: RED - pip resolver issue with publish-openstack-releasenotes-python3
13:02:08 <dougsz> hello
13:02:16 <chaconpiza> Hi Doug
13:02:25 <bandorf> Hi, Doug
13:03:04 <chaconpiza> Hervé Beraud sent already the code changes to accomplish this global change
13:03:19 <chaconpiza> https://review.opendev.org/q/(projects:openstack/monasca+OR+project:openstack/python-monascaclient)+AND+topic:fix-relmgt-pip-doc
13:03:50 <chaconpiza> I have added myself as a reviewer, please feel free to take a look on the changes
13:05:00 <chaconpiza> Here they wrote the list of affected repositories http://paste.openstack.org/show/801396/
13:06:34 <chaconpiza> #topic Tempest test Docker container, failing tests with timerange
13:07:52 <chaconpiza> I am still investigating why these 3 tests fail when using the docker containers, but in devstack they pass ok
13:08:41 <chaconpiza> dougsz, do you remember whether these test, with timerange, require specific conditions?
13:09:00 <chaconpiza> like specific influx version, or something like that?
13:09:57 <dougsz> Hmm, I don't recall anything like that
13:10:12 <dougsz> which tests specifically?
13:10:23 <dougsz> ah I see on the etherpad..
13:10:25 <chaconpiza> monasca_tempest_tests.tests.api.test_dimensions.TestDimensions.test_list_dimension_names_with_metric_name_with_timerange
13:10:35 <chaconpiza> yes, those in the agenda
13:11:10 <dougsz> These are the recently added tests from Bharat? ..just going to look
13:11:14 <chaconpiza> failing with:     testtools.matchers._impl.MismatchError: ['value_1', 'value_2', 'value_3'] != ['value_2', 'value_3']
13:11:31 <chaconpiza> yes, those from Bharat.
13:11:59 <chaconpiza> Anyways, when I stack the devstack, and execute the Tempest test they are ok
13:12:36 <dougsz> I have a bad feeling about this - the timerange for dimensions  was only approximately respected by Influx
13:13:42 <bandorf> What about influx versions used in both cases? I think, this is a good direction. One metric more or less - could easily be caused by an influx change
13:13:50 <chaconpiza> I set the same configuration in the influx container than in devstack
13:14:00 <bandorf> Ah, ok
13:14:46 <dougsz> Interesting
13:14:47 <chaconpiza> Same version, in devstack 1.7.6 and Official docker image 1.7.6-alpine
13:15:44 <chaconpiza> Ok, I will continue and report here the findings.
13:15:56 <chaconpiza> #topic AOB
13:15:58 <dougsz> I think the main thing to know, is that the timerange isn't strictly respected by InfluxDB
13:16:09 <chaconpiza> good point
13:16:46 <dougsz> You may need to widen the window, if it still seems approximately sensible in the test
13:17:13 <dougsz> or narrow it, by the look of your results :\
13:17:25 <chaconpiza> yes, it sounds like a good idea
13:17:26 <dougsz> something like that - it is far from ideal really
13:18:04 <bandorf> The time stamps used are generated by monasca? Or by influxdb?
13:18:43 <dougsz> I think they come from Monasca, and are attached to the metric
13:18:54 <dougsz> and then you say to influx, give me all the dimensions (like hostname) over this time window
13:19:07 <dougsz> and it only loosely respects the time window
13:19:17 <bandorf> OK, but in this case we could easily verify if they are correct, right?
13:19:28 <bandorf> Just query the influxdb
13:19:45 <dougsz> that true - run a raw query
13:20:02 <chaconpiza> I will
13:20:15 <bandorf> I understand your concern, regarding "loosely". But then, it could happen sometimes, sometimes not...
13:20:27 <bandorf> Martin, I assume this is 100% reproducible
13:20:36 <bandorf> ? was missing
13:21:27 <chaconpiza> yes, everytime it fails when deployed in containers
13:21:44 <chaconpiza> and everytime it's ok in devstack
13:22:53 <chaconpiza> in github.com/monasca/monasca-docker we still using the Tempesttests docker image generated from the time when tempest test was a subdirectory in monasca-api repository. I want to use the new image generated from Tempesttest own repository
13:23:35 <chaconpiza> One more issue, as you know since November Docker Hub limits the amount of pulls
13:23:44 <bandorf> Martin, if you want, we can have a call after the weekly.
13:23:44 <bandorf> I have some ideas, but takes too long to write down
13:24:00 <dougsz> Cool, along those lines I've made some progress on the Kolla Monasca CI job, so it now passes reliably using the latest Tempest tests, should hopefully be merged soon
13:24:03 <chaconpiza> ^^ sure
13:24:53 <chaconpiza> Anonymous users can pull 100 images in a period of 6 hours, it is based on external IP
13:25:46 <chaconpiza> So, some of the Travis VMs share the external IP and sometimes it is not possible to pull the images for the deploy.
13:26:03 <dougsz> :(
13:26:16 <chaconpiza> I am adding he code to check the amount of pulls that left for that external IP
13:27:09 <chaconpiza> With a registerd docker user the amount is 200 pulls and it is independant of the IP
13:28:47 <chaconpiza> Ok, those are all topics from agenda. Thanks for the hints.
13:29:00 <dougsz> thanks both
13:29:17 <bandorf> Thanks, bye
13:29:43 <chaconpiza> If there arn't other topics let's see you next week. Don't forget to add you topics in the agenda.
13:29:50 <chaconpiza> Bye
13:31:54 <chaconpiza> #endmeeting