17:00:10 #startmeeting Solum Team Meeting 17:00:10 Meeting started Tue Oct 6 17:00:10 2015 UTC and is due to finish in 60 minutes. The chair is devkulkarni. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:00:11 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:00:14 The meeting name has been set to 'solum_team_meeting' 17:00:21 #topic Roll Call 17:00:24 ed cranford 17:00:25 Devdatta Kulkarni 17:00:56 today's agenda #link https://wiki.openstack.org/wiki/Meetings/Solum#Agenda_for_2015-10-06_1700_UTC 17:01:30 datsun180b: hopefully folks saw yesterday's email 17:01:50 well i know i did 17:02:00 :) 17:02:14 hi vijendar 17:02:22 Hi devananda 17:02:36 Hi devkulkarni :) 17:03:06 vijendar: :) tab complete fails sometimes 17:03:22 devkulkarni: yes 17:03:36 hi muralia1 17:03:37 o/ 17:03:45 we are just getting started 17:03:59 we have datsun180b vijendar as well 17:04:04 cool. didnt realize the meeting was in a new room. 17:04:21 muralia1: yeah, our regular room is occupied at this hour 17:05:31 I am guessing that others will trickle in.. we can move forward as we have a quorum 17:05:47 #topic Announcements 17:05:53 Solum Fishbowl session details 17:05:58 Thursday, October 29, 2.40 pm - 3.20 pm 17:06:06 #link https://docs.google.com/spreadsheets/d/1tpLN5emWhcMmSmkn8z_HuclcjnEPevP77BhdnFN9KCs/pubhtml?gid=5&single=true 17:06:52 I am guessing that the fishbowl session is essentially a place to present the current capabilities of the project and engage with the community through Q and A 17:07:12 We have above listed FB slot 17:07:20 I will be going to Tokyo 17:07:54 In the open discussion topic, we can discuss thoughts/ideas about how to make most of this fb session 17:08:25 are there any other announcements from the team members? 17:08:38 none from me 17:08:59 alright 17:09:01 devkulkarni: have you prepared any slides to present in flishbowl? 17:09:10 vijendar: no, not yet. 17:09:41 I was thinking of pulling in various talks that the team had submitted and consolidating that in a presentation 17:10:16 will be reaching out to muralia, datsun180b, james_li, dimtruck, you, randallburt regarding this 17:10:33 cool 17:10:42 ok 17:11:02 yep .. muralia1, datsun180b, keep an eye out for my email/pings on irc 17:11:16 will do 17:11:26 awesome 17:12:07 lets move to next topic .. we can discuss more about tokyo summit in the open discussion 17:12:20 #topic Review Action Items 17:12:27 (DONE) Changing meeting time to 1700 UTC 17:12:42 Glad that you all were able to make it today 17:13:17 For folks in central time zone, the current time of noon is going to be for only few meetings 17:13:35 till day light saving starts. after that, it will be 11.00am 17:13:58 DST ends on the first Sunday in November 17:14:35 so starting November 3, we will be at 11.00am 17:14:37 right? 17:14:46 sounds right to me 17:15:09 datsun180b: thanks for correcting.. our current meeting time will be till DST ends, not starts 17:15:22 cool 17:15:22 "Spring forward, Fall back" 17:15:54 he he.. nice way to remember 17:15:56 haha 17:16:26 so yes, starting November 3, we will be at 11.00am CST for folks in CST 17:16:46 ok, second action item was 17:16:52 (DONE) Solum bug list update 17:17:09 I spent most of last Friday going through Solum's bug list 17:17:20 #link https://bugs.launchpad.net/solum/ 17:17:27 we have 131 bugs 17:17:32 is that what happened 17:17:44 there were 20/30 which were marked Incomplete 17:18:01 I went through all of them.. some of them were obsolete 17:18:08 some were actually not incomplete 17:18:20 so now, there should not be any bug which is marked incomplete 17:18:35 still, 131 is lot 17:18:49 I hope we can get to triaging them soon 17:19:04 it is possible that many of them are obsolete 17:19:33 I plan to do similar cleanup for bugs for python-solumclient 17:19:40 in next couple of days 17:20:11 ok, so that completes me giving updates on the action items 17:20:27 #topic Blueprint/Bug Review and Discussion 17:20:39 Updating documentation https://review.openstack.org/#/c/230628/ 17:20:56 If you can take a moment to review above, that would be great 17:22:40 looking 17:22:50 datsun180b: ok 17:24:03 added a comment. small nit 17:24:56 muralia1: sure, will take care of it 17:26:16 muralia1, datsun180b: will address your comments today 17:26:21 should be very quick 17:26:48 if you can keep an eye out for patch on solum channel that will be great 17:27:08 ok, the second item today is: 17:27:17 Fixing a bug in wsme pecan exception wrapper https://review.openstack.org/#/c/230115/ 17:27:26 some background on this one 17:28:10 recently I had started noticing that in solum-api logs, there was a exception line getting printed for every api call 17:28:35 turns out that that was happening because of the wsme_pecan exception wrapper had a bug 17:28:40 this patch fixes that bug 17:29:11 basically, the code was checking for 'pecan.response.status' whereas it should have been checking for 'pecan.response.status_code' 17:30:30 the patch also does not affect the original purpose of the code, which is to hide exception messages when something fails in the server (cause for status code to be set to 500) 17:31:24 datsun180b, muralia1: here is the patch link again https://review.openstack.org/#/c/230115/ 17:32:22 +2 17:32:32 thanks muralia1 17:32:54 +2+A 17:33:04 awesome datsun180b 17:33:37 next one is Updating devstack setup https://review.openstack.org/#/c/230222/ 17:33:44 so, some background on this one 17:34:18 recently we have got a new contributor who is trying to get a working devstack environment 17:34:47 while helping her out, I realized that our devstack setup script was not up-to-date 17:34:59 above patch makes it so 17:35:16 what has changed? 17:35:46 basically the devstack script did not include steps to clone and setup nova-docker 17:35:55 and also 17:36:20 nova's configuration file seems to be not coming pre-populated with the virt_driver option 17:36:39 so I have included that line in solum's devstack/lib/solum 17:37:27 ok 17:37:37 makes sense 17:38:50 a related point to this patch, about whether to use nova-docker in devstack setup or not 17:39:19 in last two days I have spent several hours trying to get the combination of glance, nova-docker, heat working together 17:39:30 with no success 17:39:56 so, this morning I have started testing the combination of swift, nova, heat 17:40:12 which we have not really exercised a whole lot in the devstack environment 17:40:43 this path is taking me all the way till 'DEPLOYING APP' state 17:41:08 but heat is returning with stack-create failed due to swift signal time out 17:41:15 so currently debugging that 17:42:54 ok, the next one is 17:43:02 Updating Glance interactions https://review.openstack.org/#/c/230588/ 17:43:14 here is background on this one 17:43:34 recently glance's interactions from within solum started failing 17:43:48 we submitted a bug to glance, which is now merged 17:44:16 but until that releases to pypi we won't be able to use it 17:44:39 also, I realized that we should update our glance commands to reflect the changes 17:45:00 this patch makes those changes https://review.openstack.org/#/c/230588/ 17:45:20 muralia1: I wanted to check with you on something 17:45:29 sure 17:45:55 I remember seeing glance image-tag method in our code 17:46:06 do you know if that is supported in glance version 1? 17:46:20 hmm, not sure actually 17:46:44 but i dont think we are using glnace tags anbymore 17:46:46 I found that that line in our code had added —os-image-api-version flag with value set to 2 17:46:55 oh okay 17:47:05 I thought for tagging languagepacks we were using that 17:47:08 let me look 17:47:14 we were originally storing languagepack matadata there. 17:47:32 i think we removed it so we can store LP metadata in the solum db 17:48:00 muralia1: https://github.com/openstack/solum/blob/master/contrib/lp-cedarish/docker/build-lp#L114 17:48:29 because this command is already using —os-image-api-version 2 17:48:44 I thought of changing everywhere to include that flag 17:48:59 it might be a v2 only thing 17:49:14 hmm.. thats what I am thinking 17:49:19 I need to double check it 17:49:54 also, for what purpose were we tagging lp images in glance? 17:50:18 the lp create command takes metadata 17:50:25 such as compiler version, etc 17:50:39 hmmm 17:50:41 we were originally storing that metadata in glance labels 17:50:49 but that doesnt make it easy for searching 17:51:05 we should change that to store metadata in the solum db 17:51:19 aren't we already storing the metadata in solum's db? 17:52:00 checking 17:52:59 cause if we are already storing this in solum db, then there is no need to tag the images in glance 17:53:00 we might be storing it in the image table tags 17:53:25 ok, I will verify in my testing. 17:53:31 we need to test it by creating an lp and passing in some metadata. then check the image table tags 17:53:45 sure, I can test that 17:54:03 going back to earlier point.. 17:54:34 lets say we are storing the metadata in solum db, then we can probably just work with glance version 1 17:55:11 as except for the image-tag-update method, no other method is v2 only 17:55:21 yes 17:56:00 ok.. and in that case, we can modify glance methods to use version 1 17:56:09 so thats why I have -1 on this patch 17:56:30 as I want to test these various scenarios 17:56:43 datsun180b, vijendar: thoughts? 17:57:45 we are almost at the end our meeting time 17:58:03 devkulkarni: I missed part of the conversation. will sync up with you later (offline) 17:58:12 vijendar: sounds good 17:58:15 we can pick it up in #solum 17:58:23 sounds good datsun180b 17:58:56 I will end the meeting then 17:59:04 yup 17:59:07 you all have good lunch break 17:59:15 and see you in solum in a bit 17:59:35 thanks muralia1, datsun180b, vijendar for joining today. we had great discussion 17:59:46 see ya. 17:59:47 see you next week at the same time 17:59:55 #endmeeting