14:01:29 #startmeeting sahara 14:01:30 Meeting started Thu Dec 13 14:01:29 2018 UTC and is due to finish in 60 minutes. The chair is tellesnobrega. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:01:31 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:01:33 The meeting name has been set to 'sahara' 14:01:59 hi 14:03:11 hi 14:03:21 is jeremy around? 14:03:52 maybe not 14:03:57 lets start 14:04:05 #topic News/Updates 14:04:38 o/ 14:04:43 hey jeremyfreudberg 14:05:03 I'm mainly working on APIv2 stuff, documentation and 500 issue 14:05:12 we can talk more about them later 14:06:52 tosky, jeremyfreudberg news from you? 14:07:07 I reviewed the patch which fixes ambari image generation on ubuntu (still some issues) 14:07:30 sent the email about the split of the repositories to openstack-discuss@ (sorry for messing up the subject) 14:08:11 worked for a (unrelated) presentation on RBAC in openstack, but that's the reason for that WIP review which adds some RBAC tests to Sahara 14:09:18 and in general I should work on the code for https://review.openstack.org/#/c/622492/ , if accepted 14:09:27 and that's it 14:10:58 jeremyfreudberg? 14:12:02 nothing new really, but the semester is over after mid-next-week 14:12:11 nice 14:12:55 will you be able to take some time to work on some apiv2 them? 14:13:07 yes, that's the idea 14:13:13 that's my priority for sure 14:13:16 awesome 14:14:25 moving on 14:14:53 #topic APIv2 14:15:07 I'm currently working on 2 issues 14:15:25 the 500 case and the documentation 14:15:33 lets start with the 500 case 14:16:12 I talked with tosky and we decided to make the change happen only on APIv2 14:17:03 apiv2 only makes sense to me too 14:17:36 what I'm trying to make it work right now is a decorator that will get the command and check in a dictionary if the params passed are on the whitelist 14:18:35 does that seems ok to you? 14:19:59 well, i think this was something that i at least partially suggested on the review, so, yes, it sounds good :) 14:20:03 tosky? 14:20:34 jeremyfreudberg, yes it is :) 14:20:58 ehm, that's one the reviews that I really didn't check 14:22:43 it would be good if both could take a look, I'm hitting some weird errors on tempest (the change is only on APIv2 and tempest runs against api v1 and it is failing) 14:22:57 also scenarios I couldn't find the errors, if anyone can give me a hand with that 14:23:24 the first question is: are you sure that the change is v2 only? :) 14:23:37 in any case, the solution is, in theory, elegant: it's the most straightforward way to encapsulate that different API view functions have different allowed query string values 14:23:39 iirc some of the v2 code used v1, or the other way, or some common code 14:23:48 tosky, good question 14:24:30 tosky, it should be at this point, but I will check 14:25:30 tellesnobrega: check the logs of sahara-api 14:25:32 tellesnobrega: did you find out where these things are coming from: https://review.openstack.org/#/c/623300/9/sahara/utils/api.py 14:25:40 it's definitely going through apiv2 code 14:25:51 (my question is unrelated to tosky's question) 14:26:00 http://logs.openstack.org/00/623300/9/check/sahara-tests-tempest/94d7ad0/controller/logs/apache/sahara-api_log.txt.gz#_2018-12-12_21_41_49_395831 14:26:41 jeremyfreudberg, I cleaned up it a lot now, I'll test and see if it is better 14:27:11 jeremyfreudberg, it looks more like this now https://review.openstack.org/#/c/623300/9/sahara/service/validation.py 14:27:20 a lot less params 14:27:38 tellesnobrega: great, cool 14:27:47 the ones I kept are search_opts passed on the client that it uses to create the query 14:27:54 tosky, thanks, I will fix that 14:30:22 that is all I have on apiv2 14:30:40 hopefully jeremyfreudberg will have time to work on microversioning 14:31:42 any topics you want to cover? 14:32:13 an update on plugin split 14:32:39 sure 14:32:43 #topic Split Plugin 14:33:15 I sent the email to openstack-discuss@; we had one reaction so far with some feedback which should be implemented (description for plugins) 14:33:58 apart from that, the codebase seems to be fine, according my previous tests, so I'd be leaning towards merging the pending patches, rebasing, and try to push the split as soon as possible 14:34:13 and fix the breakages when they shows 14:34:34 at least a puppet core wrote that he can help when I nagged him earlier today 14:34:46 the openstack-ansible part is something that I can help fixing too 14:35:10 if anyone else has any comment on the way the split is done, this is the right time :) 14:35:31 it looks like the right path 14:37:36 nothing else from my side about the splitting 14:37:51 jeremyfreudberg? 14:39:49 not right now 14:40:15 anything else? 14:40:59 not on this topic, just a plea to recheck few pending reviews with +2 (I think that tellesnobrega already ninja-approved everything that he could, so it's on jeremyfreudberg now :) 14:41:17 yes, i'll give some things a look 14:41:31 I did but didn't want to force approve important stuff 14:42:05 jeremyfreudberg, if you have time today or tomorrow to review would be great so we can merge most of the stuff and we can move on with the split sequece 14:42:08 sequence 14:43:02 yup 14:43:15 thanks 14:43:55 I guess we can close it early today 14:45:55 thanks jeremyfreudberg and tosky 14:46:06 lets keep up the good work 14:46:11 see you all next week 14:46:34 see you! 14:46:36 #endmeeting