19:00:37 <hogepodge> #startmeeting refstack
19:00:38 <openstack> Meeting started Tue Sep 19 19:00:37 2017 UTC and is due to finish in 60 minutes.  The chair is hogepodge. Information about MeetBot at http://wiki.debian.org/MeetBot.
19:00:39 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
19:00:42 <openstack> The meeting name has been set to 'refstack'
19:00:57 <hogepodge> #topic Agenda
19:01:16 <mguiney> o/
19:01:43 <catherineD> o/
19:02:03 <hogepodge> #link https://etherpad.openstack.org/p/refstack-meeting-17-09-19 Agenda
19:02:03 <luzC> o/
19:02:20 <hogepodge> Please check over the agenda and add any additional items
19:02:46 <pvaneck> o/
19:03:41 <hogepodge> #topic PTG Followup
19:04:05 <hogepodge> Thanks to everyone for attending the PTG, either in person or remotely. I feel like we were pretty productive. It was great to see everyone.
19:04:48 <luzC> +1
19:05:06 <hogepodge> #link https://www.dropbox.com/sh/bqd2i7uy988qehk/AAB-GRH37novuE-vFlE7zsvZa?dl=0 PTG Team Photos
19:05:22 <mguiney> (yay!)
19:06:15 <hogepodge> The two major work items we came out with from the PTG were subunit upload and Tempest Autoconfig. I think we're going to make great progress on both.
19:06:24 <catherineD> everyone looks good in the picture :-)
19:06:48 <mguiney> yeah, great team photos :)
19:06:58 <hogepodge> catherineD: we need to photoshop you and pvaneck in :-D
19:07:09 <catherineD> :-)
19:07:21 <mguiney> its too bad we couldnt bring a laptop down with catherineD on it, at the time ;P
19:07:27 <pvaneck> lol, paint crop in
19:07:33 <hogepodge> haha yes
19:08:34 <hogepodge> API docs was also an area we identified for work (right now the docs are all specs)
19:08:35 <catherineD> I found that it was not very productive to attend remotely ... but still better to have a remote session than nothing
19:08:56 <catherineD> hogepodge: ++ on API doc
19:08:57 <hogepodge> catherineD: were you able to hear us?
19:09:19 <catherineD> maybe 30% of the time ...
19:09:39 <mguiney> :(
19:10:00 <hogepodge> Ah, I thought our conference phone was working better than it was. :-( Maybe microphones next time? ;-)
19:10:14 <catherineD> especially the SIG session ... could not hear much because of multiple speakers in different positions
19:10:57 <hogepodge> Good to know. We'll be sure to pass around the microphone as much as we can next time.
19:11:25 <hogepodge> Many thanks to gema for bringing the speakerphone
19:11:35 <mguiney> gema++
19:11:47 <hogepodge> Any other comments on the content or work items?
19:11:57 <hogepodge> Or feedback on how I can run it better?
19:13:24 <hogepodge> Ok, with the PTG behind us.
19:13:35 <hogepodge> #topic OpenStack Summit Sydney
19:13:53 <hogepodge> I wanted to get a sense of attendance and accepted talks
19:14:11 <hogepodge> Who is planning on attending? mguiney and I will be there
19:14:14 <catherineD> hogepodge: nope you did a great job to have a good number of attendees at the RefStack session
19:14:34 <hogepodge> catherineD: thank you
19:17:13 <catherineD> I  do not have any speaker session this time .. .but there is a forum session that Rocky put my name as one of the panelists.. As usual, we will need travel approval ...
19:18:03 <hogepodge> catherineD: ok, let me know if you need help with anything
19:18:18 <luzC> I don't have speaker sessions either, and I'm not attending the summit this time
19:18:35 <hogepodge> luzC: :-(
19:20:33 <hogepodge> On to the next topic
19:20:47 <hogepodge> #topic RefStack Server Cleanup
19:20:58 <mguiney> ahhh yes
19:21:13 <hogepodge> One of the outcomes from last week were a number of patches that clean up the code base, mostly around documentation
19:21:48 <hogepodge> mguiney: has two, one that was just merged, and another that isn't passing the gate yet but is an API fix
19:22:28 <hogepodge> catherineD: pvaneck: luzC: can you take a look at https://review.openstack.org/#/c/504535/2 ?
19:23:44 <hogepodge> To me it looks like it's failing because the mock is expecting a different function signature. I just want to make sure that the change is accessing the database and api correctly
19:24:01 <mguiney> yep. It's just a matter of changing the thing to a MagicMock
19:24:24 <mguiney> but i need to figure out first *which* mock needs to be changed
19:25:33 <mguiney> so that should be a quick fix
19:25:44 * mguiney knocks on wood
19:26:16 <luzC> the patch is only changing the unit test case for the api...
19:26:36 <luzC> is that the intention?
19:26:45 <luzC> I'll take a look later today
19:27:06 <hogepodge> thanks luzC
19:27:16 <hogepodge> I have a couple of patches that are wip
19:27:54 <luzC> probably patch need a depends-on tag
19:27:55 <hogepodge> I mean, one is, the other is changing the endpoint for run-in-docker
19:28:15 <hogepodge> #link https://review.openstack.org/#/c/503178/
19:28:15 <pvaneck> so right now 'product_version_id' is not returned in the test response based on https://github.com/openstack/refstack/blob/master/refstack/db/sqlalchemy/models.py#L50
19:28:46 <pvaneck> based on default_allowed_keys there
19:29:08 <hogepodge> pvaneck: hmm
19:29:40 <mguiney> ohhhh ok
19:29:45 <mguiney> that actually explains a lot
19:29:47 <pvaneck> product_version is based on the defined relationship which wil give you keys ('id', 'version', 'cpid', 'product_info')
19:30:14 <mguiney> out of curiosity, why are we inserting it as "product_version_id"
19:30:18 <pvaneck> that's why ['product_version']['id'] is checked as opposed to ['product_version_id']
19:30:34 <pvaneck> inserting where?
19:31:31 <mguiney> line 118-119 of the same file
19:31:44 <mguiney> er hold on lemme find a link
19:32:32 <mguiney> https://github.com/openstack/refstack/blob/master/refstack/tests/api/test_results.py#L118-L119
19:32:57 <pvaneck> i believe that is just the key that the api requests for put on a test object
19:33:05 <mguiney> ahhhhh ok, thank you
19:33:23 <mguiney> that clarifies a lot, actually. I'll fix it asap
19:34:20 <pvaneck> I think the reason for not returning it in the response was that a product_version dict containing (id, version, cpid, product_info) is returned and product_version_id would be duplicate of product_version.id
19:34:45 <luzC> I have to leave earlier, I'll check the pending reviews later today :-)
19:34:56 <hogepodge> luzC: ok
19:35:16 <mguiney> luzC: bye! have a good day!
19:35:40 <hogepodge> luzC: thanks for dropping in
19:36:18 <hogepodge> pvaneck: mguiney: do you feel like there's enough information for a review and action on that patch?
19:36:30 <mguiney> absolutely
19:37:39 <pvaneck> not sure exactly what the patch is doing. is it only supposed to be a one line change?
19:37:44 <mguiney> actually... given that that is the only  change in that patch, it may be better to discard it
19:37:55 <mguiney> sorry, took another look, reconsidered
19:37:58 <catherineD> mguiney: ++
19:38:16 <mguiney> i had forgotten that i'd split that and a few other changes, my bad
19:38:37 <pvaneck> yea, i see that patch set one had additional changes
19:39:14 <hogepodge> pvaneck: there were many changes that addressed a few different problems, so we broke the patch up
19:39:27 <pvaneck> okay
19:39:42 <hogepodge> looks like we merged a lot of the other patches
19:39:57 <hogepodge> take a look at the rest on the cleanup list and see if there's anything amiss with them.
19:40:06 <pvaneck> slowly going through
19:40:22 <hogepodge> #topic RefStack Verification Field Update
19:40:28 <hogepodge> mguiney: what's the status on this?
19:40:41 <mguiney> ready for review (and hopefully merge)
19:40:56 <hogepodge> Looks like three active patches
19:41:07 <hogepodge> #link https://review.openstack.org/#/c/504358/1/ Documentation update
19:41:24 <hogepodge> #link https://review.openstack.org/#/c/499956/ Script bug fix
19:42:01 <hogepodge> #link https://review.openstack.org/#/c/503495/1 Token auth (is that part of this patchset)
19:42:13 <mguiney> nope!
19:42:29 <hogepodge> Ok, so just the first two?
19:42:31 <mguiney> the third patch is just an improvement
19:43:11 <mguiney> yes. So the documentation patch isnt even quite part of the bugfix, just an assumption of knowledge that i noticed I hadn't written down while fixing the initial bug
19:43:17 <hogepodge> Ok. So focus on the first two?
19:43:20 <mguiney> so really, more of a cleanup patch
19:43:44 <hogepodge> Can we set a target for running the update and schedule it with infra once the patches are merged?
19:44:07 <mguiney> target date?  absolutely
19:45:15 <hogepodge> mguiney: will you coordinate it with clarkb or fungi for the database backup and script run?
19:45:48 <mguiney> absolutely. I also have a test sheet generator, in case anyone has any desire to do more testing
19:46:05 <mguiney> just to save anyone interested the carpal tunnel
19:46:15 <fungi> yeah, happy to help on that when things with the gerrit upgrade calm down a little
19:47:13 <hogepodge> ok, moving on to the next topic
19:47:29 <hogepodge> #topic RefStack Subunit Upload
19:47:31 <mguiney> excellent, thank you :)
19:48:22 <hogepodge> The first step is to prepare the database to handle subunit. We have a patch up for review
19:48:28 <hogepodge> #link https://review.openstack.org/#/c/498735/ Alembic table
19:49:11 <mguiney> That's been functional a while, the most recent patch is just cleaning up some unneccesary complexity in the code flow
19:49:55 <hogepodge> mguiney: ok. pvaneck catherineD it looks like we should add a unit test too to make the code coverage pass?
19:50:28 <pvaneck> yea, some test(s) that touch the new code
19:51:03 <hogepodge> If we can keep that test in the green (even though it's non-voting) I'd prefer that
19:52:20 <hogepodge> We had a long discussion at the PTG about next steps.
19:53:20 <hogepodge> mguiney: has an action to write the API spec for the subunit upload
19:53:35 <mguiney> i've gotten a good ways into that
19:53:43 <hogepodge> luzC: has an action to determine a method for validating the subunit input
19:54:06 <hogepodge> hogepodge: has an action to write general API documentation (probably with swagger)
19:54:24 <hogepodge> I'll add those to the meeting etherpad so we can track them going forward
19:55:21 <hogepodge> Running low on time so let's see if we can finish off the next topic
19:55:28 <chandankumar> hogepodge: can we move refstack meeting one hour early?
19:57:01 <hogepodge> how would that work for everyone else? I don't think we have time to discuss it today, but add it to the open discussion section and we can make it our first agenda item for next week
19:57:25 <hogepodge> #topic RefStack Client
19:57:25 <pvaneck> works for me
19:57:44 <chandankumar> hogepodge: sure, that will work.
19:57:49 <hogepodge> no movement on either of the active patches, will check in with David to see what the progress is.
19:57:59 <hogepodge> chandankumar: thanks
19:58:11 <hogepodge> #topic Open Discussion
19:58:45 <mguiney> (re timeslot move, that would actually be *better* for me)
19:58:48 <hogepodge> chandankumar: in general earlier works for me too. We would need to find open meeting space if we moved it.
19:59:15 <chandankumar> mguiney: yup, we need to check with catherineD also is it works for her or not?
19:59:19 <hogepodge> It's a bigger topic anyway, I think we should be tick-tocking for more international coverage
19:59:51 <hogepodge> We have contributors from the Asia-Pacific region that I'd like to give a better chance at attending to.
19:59:59 * mguiney nods
20:00:06 <hogepodge> Almost out of time, but good starter topic for next week. Thanks chandankumar!
20:00:11 <catherineD> yea
20:00:15 <hogepodge> Thanks everyone! It went fast!
20:00:20 <hogepodge> #endmeeting