17:00:00 <hogepodge> #startmeeting refstack
17:00:01 <openstack> Meeting started Tue Apr 17 17:00:00 2018 UTC and is due to finish in 60 minutes.  The chair is hogepodge. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:00:02 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:00:05 <openstack> The meeting name has been set to 'refstack'
17:00:09 <tosky> o/
17:00:15 <hogepodge> #link https://etherpad.openstack.org/p/refstack-meeting-18-04-17 agenda
17:01:11 <tosky> I was confused by the date :)
17:01:41 <mguiney> o/
17:02:27 <hogepodge> #chair tosky
17:02:28 <openstack> Current chairs: hogepodge tosky
17:02:38 <hogepodge> #topic python-tempestclient
17:02:55 <hogepodge> take it away! Yeah, the every other week is a bit confusing. :-)
17:03:17 <tosky> sooo
17:03:22 <hogepodge> looks like a lot of changes merged in the last couple of weeks
17:03:59 <tosky> yes, especially the big breaking ones
17:04:28 <tosky> the big refactor landed yesterday, so now it should be easier to add support for new service
17:04:31 <tosky> services*
17:04:46 <tosky> there is still space for improvement of course
17:05:19 <tosky> that means that few reviews needs to be updated (merge conflict), namely
17:05:31 <tosky> #link https://review.openstack.org/#/c/528380/ Make tempest setup image verification aware
17:05:51 <tosky> #link https://review.openstack.org/#/c/519714/ WIP: Add EC2-API support
17:06:00 <tosky> (this ^^ one was not in the agenda)
17:06:44 <tosky> #link https://review.openstack.org/#/c/546115/ Make main function more modular
17:06:53 <hogepodge> so what's the timeline for refstack integration?
17:07:21 <tosky> so, about that, mkopec is working on
17:07:25 <tosky> #link https://review.openstack.org/#/c/554480/ Use --non-admin for non admin users in zuul jobs
17:07:37 <tosky> which should ensure that we work properly with non admin users
17:07:56 <hogepodge> excellent
17:08:06 <tosky> the proper integration has not received updates in a while, but I think that it was waiting for that change
17:08:08 <tosky> namely
17:08:12 <tosky> #link https://review.openstack.org/#/c/541273/ Generate tempest.conf in refstack client (WIP)
17:08:24 <tosky> Chandan also had other things to fix around
17:08:54 <tosky> so while I don't know exactly the status, I know for sure that it's not forgotten :)
17:09:32 <tosky> all the work above should cover the 3 open stories, at least partially
17:09:48 <tosky> or at least
17:10:03 <tosky> #link https://storyboard.openstack.org/#!/story/2001695 Fix tempestconf job having demo user
17:10:04 <tosky> and
17:10:21 <tosky> #link https://storyboard.openstack.org/#!/story/2001696 Refactor tempestconf in order to integrate with refstack_client tool
17:11:04 <tosky> there is no work in progress yet (afaik) on
17:11:13 <tosky> #link https://storyboard.openstack.org/#!/story/2001693 Generate accounts.yaml when --os-cloud is used with python-tempestconf
17:11:26 <hogepodge> ok
17:11:27 <tosky> but it's probably better to tackle that after the refactor stabilizes
17:11:32 <hogepodge> yeah
17:11:38 <tosky> and I think it's everything from my side for today
17:12:18 <hogepodge> thank you!
17:12:23 <hogepodge> questions from anyone?
17:13:01 <hogepodge> #topic refstack
17:13:02 <mguiney> not from me,, i'll take a look at those and review later today
17:13:27 <hogepodge> There's a new review up to modify refstack-client, to help with nfv testing
17:13:35 <hogepodge> Also to account for newer tempest release
17:13:36 <hogepodge> https://review.openstack.org/#/c/561777/ Conform with tempest 18.0.0 when getting test IDs
17:14:05 <hogepodge> Once the unit testing is sorted out we'll want to do some manual testing just to make sure it's not breaking anything for the users.
17:14:32 * mguiney nods
17:14:54 <hogepodge> Subunit upload is still in progress. Waiting on a few things, including subunit2sql patches being finalized
17:15:02 * mguiney nods
17:15:04 <hogepodge> #link https://review.openstack.org/#/c/543836/ Add ability to grab stream processing variables from kwargs
17:15:12 <hogepodge> #link https://review.openstack.org/#/c/537775/ Add delete by uuid functions
17:15:13 <mguiney> it looks as though there have been some movement on those
17:15:42 <hogepodge> I think we're still waiting on mtreinish to give his stamp of approval
17:15:51 <mguiney> locally, i'm working on getting the oslo conf issues sorted, though that hasnt been at the top of my stack, so i haven't finished getting that patch ready
17:16:11 <hogepodge> ok
17:16:39 <mguiney> i can ping him today, unless there have been more changes suggested, in which case i'll amend the patch first
17:16:40 <hogepodge> Did the advice from the oslo folks make sense?
17:16:53 <mguiney> (saw that there had been reviews, haven't seen them yet
17:16:54 <mguiney> )
17:17:08 <mguiney> and yes, it really pointed me at some factors i hadnt been considering
17:17:10 <hogepodge> I've pinged him a few times (including here) ;-) but yeah, we should stay on top of moving that forward so we're unblocked. You can ask in the qa room also
17:17:33 <mguiney> i think i'll do that, more eyes on a patch are always better
17:18:37 <hogepodge> New capability sources patch
17:18:49 <mguiney> ah, yes
17:19:08 <mguiney> This one is *so close* to being ready
17:19:18 <hogepodge> #link https://review.openstack.org/#/c/547246/ allow for the addition of new capability sources
17:19:35 <mguiney> after much fighting with how javascript does things, I finally got it rendering the page again
17:19:54 <hogepodge> ok, good
17:19:56 <mguiney> its currently working on the assumption that the selected guideline type is powered/compute/storage
17:20:00 <hogepodge> looks like unit test and lint problems?
17:20:25 <mguiney> so i have to change that so that if one of the addons is selected, itll use that instead
17:20:54 <hogepodge> ok
17:20:54 <mguiney> that said, given that the mechanism for grabbing the proper source is now working, that should be relatively minor
17:21:00 * mguiney knocks on wood
17:21:00 <hogepodge> good
17:21:12 <mguiney> yes, it appears that it is (re npm tests)
17:21:56 <mguiney> I'm rather confused by that, as nothing npm-lint related should have been touched here, and i'm wondering if I may have included the yarn patch as a base by accident
17:22:37 <hogepodge> ah, maybe
17:22:58 <hogepodge> Yarn is being included in this patch
17:23:25 <hogepodge> oh wait, i was looking at the wrong one
17:23:26 <mguiney> ah, i'll fix that as soon as meeting is out
17:23:47 <hogepodge> Speaking of that
17:23:50 <mguiney> oh whoops. either way, i'll investigate post meeting.
17:23:57 <hogepodge> #link https://review.openstack.org/#/c/559459/3 Convert from bower usage to yarn
17:24:18 <hogepodge> I'm not sure why we're using bower away, we should just do that once and make the patch completely dependent on yarn, unless I'm missing something.
17:25:07 <mguiney> I agree, my latest version of the patch (which is not quite finished) doesnt include bower-away as a dep
17:25:45 <hogepodge> ok
17:25:56 <hogepodge> Anything else on the refstack front?
17:25:59 <tosky> anyway, ".psst! Your project can stop working at any moment because its dependencies can change. " is definitely scary a lot
17:26:08 <mguiney> the thing i'm currently running into, however, is that when i change the package statements so that npm pulls from https endpoints rather than git ones, the package retrievals are failing
17:26:17 <mguiney> tosky: yeah no kidding
17:26:19 <tosky> "let's change completely the packaging system", nice move
17:27:33 <mguiney> I plan on spinning up a fresh env today and trying to work pas the current roadblock, have been really surprised by the amount of fight that yarn has been putting up in the pursuit of making it work
17:27:42 <mguiney> *past, not pas
17:27:45 <hogepodge> we can look at it later
17:27:48 * mguiney nods
17:27:49 <hogepodge> today, when we meet
17:27:54 <mguiney> i may be missing something silly
17:27:59 <mguiney> otherwise, that's all i have
17:28:08 <hogepodge> Is there anything else?
17:29:31 <hogepodge> ok, have a great day everybody!
17:29:34 <hogepodge> #endmeeting