14:00:27 <mattmceuen> #startmeeting airship 14:00:28 <openstack> Meeting started Tue Jan 29 14:00:27 2019 UTC and is due to finish in 60 minutes. The chair is mattmceuen. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:30 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:32 <openstack> The meeting name has been set to 'airship' 14:00:33 <mattmceuen> #topic Rollcall 14:00:39 <mattmceuen> good morning / evening everyone! 14:00:39 <sthussey> here 14:00:48 <dwalt> o/ 14:00:54 <evgenyl> o/ 14:00:55 <mattmceuen> Agenda du jour: https://etherpad.openstack.org/p/airship-meeting-2019-01-29 14:00:55 <b-str> o/ 14:00:56 <michael-beaver> good morning 14:01:00 <levmorgan> Hiya 14:01:02 <seaneagan> o/ 14:01:04 <nishant_> good morning 14:01:08 <georgk> hello 14:01:14 <mattmceuen> We'll give it 2mins for folks to join and add any discussion topics to the etherpad 14:02:53 <mattmceuen> #topic Airship-in-a-Bottle issues 14:02:59 <levmorgan> So, I've been to a few meetings, but I realized I haven't introduced myself yet. I'm Lev, I've been doing work on Pegleg. Thanks to everyone who's been helping out with reviews. 14:03:20 <b-str> Hi Lev 14:03:25 <mattmceuen> Hey Lev, welcome and thanks for the Pegleg work, was looking at some of that last week and looking really good 14:03:34 <levmorgan> Thanks! 14:03:55 <mattmceuen> So for AIAB - my new years resolution 14:04:01 <mattmceuen> Is not to make this a standing item on our agenda 14:04:37 <mattmceuen> There were a couple issues reported on the ML, and then v1k0d3n reported one over the weekend which I believe matched one of the ML issues 14:04:49 <mattmceuen> dwalt can you please summarize issues? 14:05:08 <dwalt> Certainly. 14:05:51 <dwalt> A lot of folks on the mailing list have been reporting issues with AIAB behind the proxies. Roman has compiled a list of sites that are compatible with proxy usage 14:06:39 <dwalt> There is also a patch that adds support to the dev-minimal site, but it hasn't received a lot of attention lately 14:06:50 <mattmceuen> "sites that are compatible with proxy usage" - what does that refer to? 14:07:19 <dwalt> Not all of the sites work behind proxy servers 14:07:32 <dwalt> I'm not sure if this is listed in the docs anywhere, I couldn't find it 14:07:54 <dwalt> #link https://review.openstack.org/#/c/615387/ 14:08:38 <dwalt> e.g. sites refers to dev-minimal, multinode 14:08:38 <mattmceuen> Oh I see, "airship site definition sites", my head was stuck on "web sites" :) 14:09:08 <dwalt> I think it would be good to get some attention on this patch again and maybe update the docs to reflect this 14:09:52 <mattmceuen> So dev-minimal does not work behind proxy yet -- as far as we know, is that the only site that's been reported as having issues proxy-wise recently? 14:10:27 <sthussey> I'd prefer to just totally get rid of AIAB 14:10:33 <smithd> multi-gate has given me fits 14:10:45 <dwalt> This is the most information I can find on that mattmceuen 14:10:46 <mattmceuen> v1k0d3n is not here but hopefully will see this later - please let us know if you were behind a proxy when you hit that maas deployment issue 14:10:47 <dwalt> # link http://lists.airshipit.org/pipermail/airship-discuss/2019-January/000278.html 14:11:27 <nishant_> I was facing issues on AIAB dev minimal with MaaS import jobs failing 14:11:33 <nishant_> last week 14:13:12 <mattmceuen> nishant_: that was not behind a proxy, correct? 14:13:29 <nishant_> It was behind proxy 14:13:32 <mattmceuen> ok 14:14:14 <dwalt> I can give it a run outside of a proxy and see if the maas issues persist 14:14:27 <mattmceuen> I know romang has a job running in a lab that tests AIAB not behind a proxy (and I think it would be valuable to get this reporting out somewhere) but: 14:14:40 <mattmceuen> I wonder if it would be more valuable to have an AIAB job that runs behind a proxy 14:14:48 <mattmceuen> as that seems to be a very common use case for folks 14:14:55 <mattmceuen> and seems to slip by us more often 14:14:59 <dwalt> mattmceuen: do you know which site that job runs? 14:15:05 <mattmceuen> nope 14:15:16 <mattmceuen> we should follow up with romang later 14:15:35 <mattmceuen> I added the two PS from Roman's email to the "please review" list for the day 14:15:57 <mattmceuen> Any other discussion on the AIAB topic? 14:16:38 <dwalt> nothing from me 14:17:06 <mattmceuen> ok cool. Thanks for bringing this up dwalt. 14:17:10 <mattmceuen> #topic Airship Community Feedback session 14:17:46 <mattmceuen> FYI all -- in lieu of our normally scheduled design call this thursday, we'll be hosting an Airship Community Feedback call on the same bridge 14:17:58 <mattmceuen> Chris Hoge will be facilitating 14:18:14 <mattmceuen> That's: This Thursday, 10am US Central Time 14:18:20 <mattmceuen> https://attcorp.webex.com/meet/rp2723 14:19:20 <mattmceuen> Would appreciate folks joining and offering any and all feedback so we can continue to improve our community, be inclusive, and make it fun for all us devs 14:19:59 <mattmceuen> That's all I have on that - see you then 14:20:06 <mattmceuen> #topic Requests for Review 14:20:42 <mattmceuen> So these were from romang's email earlier, re: AIAB proxy work. Let's get some eyes on them please: 14:20:42 <mattmceuen> https://review.openstack.org/#/c/615387/ (AIAB Proxy) 14:20:42 <mattmceuen> https://review.openstack.org/#/c/615387/ (AIAB Proxy) 14:20:50 <mattmceuen> in fact they're the same one 14:20:53 <mattmceuen> so review it twice 14:21:04 <roman_g> :D 14:21:09 <mattmceuen> Hey roman_g! 14:21:10 <roman_g> o/ 14:21:40 <mattmceuen> Any other patchsets that could really use some reviews today, folks? 14:23:16 <mattmceuen> #topic Roundtable 14:23:28 <mattmceuen> Open floor, anything else we'd like to bring up 14:24:41 <georgk> I have a question 14:24:46 <georgk> if nobody else has anything 14:25:03 <b-str> go for it! 14:25:31 <georgk> what is your recommended way to debug documents rendered by deckhand 14:26:16 <b-str> one thought... Shipyard can export rendered documents after they are ingested into the site 14:26:17 <georgk> has anybody worked on a CLI client? 14:26:47 <mattmceuen> shipyard has a CLI client :) lives in the API container, right b-str? 14:27:00 <georgk> well, shipyard yes 14:27:08 <georgk> ok, didnt consider this 14:27:16 <b-str> it does, but it's probably best to use the script and remote docker, but however you want to run it. 14:27:20 <b-str> https://airshipit.readthedocs.io/projects/shipyard/en/latest/CLI.html#get-configdocs 14:27:29 <b-str> https://airshipit.readthedocs.io/projects/shipyard/en/latest/CLI.html#get-renderedconfigdocs 14:27:45 <b-str> also, there's the concept of validation of the docs 14:27:56 <b-str> you can do a validation without committing them using: 14:28:18 <b-str> https://airshipit.readthedocs.io/projects/shipyard/en/latest/CLI.html#commit-configdocs with the --dryrun option 14:28:18 <georgk> right, thats what I am looking for 14:28:26 <georgk> ok, cool 14:28:45 <roman_g> Deckhand docs builds are still failing. Could someone look at it, please? I as not able to get it working. Thank you. 14:29:04 <roman_g> May be call me in w/ screen-sharing, and I will learn from you. 14:29:04 <georgk> ok, I´ll check that out. Thanks 14:29:11 <roman_g> Would greatly appreciate. 14:29:26 <mattmceuen> thanks georgk! 14:29:58 <b-str> Roman, I looked into it a bit 14:30:04 <b-str> but gave up. 14:30:12 <roman_g> Deckhand docs build error: sphinx.errors.ExtensionError: Could not import extension sphinxcontrib.apidoc (exception: No module named 'sphinxcontrib.apidoc') 14:30:26 <b-str> it looks like something changed in the RTD platform more than something that changed in the codebase 14:30:50 <roman_g> Latest successfull build Completed Oct. 29, 2018. 5:50 p.m. 14:31:10 <roman_g> 3 months passed as it is outdated 14:31:19 <roman_g> b-str: anything we can do together? 14:31:26 <b-str> right, and if you look at the code on either side of those commits, it doesn't seem to have any breaking changes - but... I could also be very wrong. 14:31:44 <roman_g> b-str: that was first thing I checked :( 14:32:04 <b-str> I think we could probably turn off the apidoc usage and see if it goes through 14:32:12 <b-str> but that eliminates a bunch of docs 14:32:59 <roman_g> Well, as long as it works on localhost, can we tell people to build their own api docs? 14:33:32 <b-str> I mean, that's not a great long term solution 14:34:12 <b-str> for now, we can ask that of course... 14:35:08 <roman_g> ...ok, will verify and submit change to disable apidocs 14:38:26 <mattmceuen> Let's not lose track of that fmonteiro did some good stuff there that we should get working again 14:39:34 <b-str> indeed 14:39:55 <mattmceuen> anything else guys? 14:40:08 <roman_g> No. 14:41:20 <mattmceuen> alright - thanks all, have a good week! 14:41:23 <mattmceuen> #endmeeting