17:01:43 #startmeeting community 17:01:43 Meeting started Mon Mar 30 17:01:43 2015 UTC and is due to finish in 60 minutes. The chair is reed. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:01:45 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:01:46 hi 17:01:47 The meeting name has been set to 'community' 17:01:53 hi guys 17:01:58 hi 17:02:59 so let's review the action items of last meeting 17:03:00 #link https://wiki.openstack.org/wiki/Community/MeetingAgenda#Agenda_for_March_30.2C_2015 17:03:38 mrmartin, let's start from askbot 17:03:53 but first 17:04:07 #topic check previous action items http://eavesdrop.openstack.org/meetings/community/2015/community.2015-03-23-17.01.html 17:04:23 reed to prepare a visual diagram based on dizquierdo's rst file: done 17:04:52 #link https://wiki.openstack.org/wiki/File:Grimoire-arch.png 17:05:05 dizquierdo to look at other openstack-infra/puppet-* modules and refactor mlstats puppet 17:05:12 how is that going, dizquierdo_? 17:05:20 reed, we decided to delay this, till we get a proper view of the architecture 17:05:24 although dpose is working on that 17:05:26 ok 17:05:29 or at least he's aware of that 17:06:27 #info dizquierdo_ delayed the refactoring mlstats puppet module until architecture is more clear 17:06:30 how about the next action item? dizquierdo to add to the spec the description of database management 17:06:44 reed, I added to the spec file info about the databases 17:06:57 and also information related to Automator, the tool in charge of automating the whole process 17:06:58 automator? 17:07:00 ok 17:07:08 retrieval + running all of the tools 17:07:16 https://etherpad.openstack.org/p/activity-infratization-spec 17:07:34 lines 54-65 and 75-136 are the new ones 17:07:34 ok, if we have time, we'll discuss those later 17:07:41 ok 17:08:10 and the following topics are for mrmartin, part of the already scheduled topics 17:08:16 well reed 17:08:18 just a comment 17:08:25 there were pending issues in GitHub 17:08:29 we advanced in some of them 17:08:36 at least closed one, and there are comments in the others 17:08:40 right, dizquierdo_, we have those as later topics 17:08:45 ah ok, sorry 17:08:52 #topic Infratization of Askbot 17:09:23 #info reed has done more testing reported results to the community list... 17:09:47 so askbot 17:09:54 #info Chinese search on the new server works as in the production server, only with roman characters 17:10:04 the Chinese patch was accepted, and we have this roman char issue 17:10:09 so the two servers are similarly broken, right? 17:10:20 yeah, but I can debug that 17:10:53 ok, we need to decide today if we move as it is and fix it later 17:10:59 solr communicates through a REST like api, so it is possible to identify whether solr config is bad, or askbot sending wrong indexes or search queries 17:11:00 or fix it now before moving 17:11:27 I can allocate time for that tomorrow, let's check it, if it means a quick fix, apply that 17:11:34 did you find out about the pip package? 17:11:43 is it a good source to install? 17:12:14 when I deployed the pip first, it was a very recent release 17:12:41 ok 17:12:51 i say let's go for pip then 17:12:58 but basically I support the pip or some packaged form 17:13:13 yes, we can work with Evgeny on that I guess 17:13:15 because it won't broke a system 17:13:24 if a new github patch arrives 17:13:44 so we should wait until tomorrow to set a deadline for the switch? 17:13:58 yes I suggest that 17:14:29 #info check back with mrmartin on March 31st to decide a deadline for switching the server 17:14:45 mrmartin, how about the backup scripts? what was the decision there? 17:15:15 so there was a little confusion, and we needed to backup pgsql what was a non-existent puppet module for infra 17:15:30 https://review.openstack.org/167631 17:15:35 https://review.openstack.org/168117 17:15:42 this two patch just waiting for final approval 17:16:10 ok 17:16:22 and if it is there, we I just need to add a bup resource, a third patch to manage the backup app at instance side. 17:16:23 do those need to merge before we can cut-off the old server? 17:16:55 anyway this third patch will require a little time to pass infra, because it is requires manual intervention at backup server side (create credentials, keys, etc.) 17:17:05 anyway 17:17:10 I found something :D 17:17:30 I just realized that we have an ask-staging.o.o instance, and actual ask.o.o is creating daily backups there 17:17:41 #info the backup of the new askbot server is depending on https://review.openstack.org/167631 and https://review.openstack.org/168117, and possibly a third patch 17:18:07 good :) 17:18:33 we need something like that for bup resource entry: https://github.com/openstack-infra/system-config/blob/master/modules/openstack_project/manifests/groups.pp#L86 17:19:05 so that's the askbot status 17:20:13 ok, let's go on with groups 17:20:47 cool 17:21:15 #topic groups status 17:21:25 #info post-pone the decision to move askbot to the new server until March 31st, pending sync-up with reed and mrmartin after infra meeting 17:21:38 #topic groups status 17:21:50 so this event location integration with o.o is done 17:21:57 looks nice finally. 17:22:32 The security fix and Commons 3.15 -> 3.23 upgrade done 17:22:40 and a little notice here 17:22:41 #info integration of events with openstack.org is done 17:22:42 :) 17:22:53 #info The security fix and Commons 3.15 -> 3.23 upgrade done 17:23:16 this upgrade broke only one thing, new users who had an o.o profile and don't have an account at groups.o.o cannot login 17:23:49 i saw a patch for that: is that issue fixed? 17:24:03 because some code in token handling eat up all of the php process memory and either segfaulted or timed out. 17:24:23 lovely 17:24:29 I realized that on Friday, was a bit hard to find what the heck happend 17:25:11 because independently everything was working, and integration of three modules commons_activity_streams, tokens and groups_ouath2 caused this bug 17:25:24 so the patch is out now 17:25:32 both on dev, and production 17:25:32 is the site working correctly? 17:25:40 needs to work now 17:25:55 what do you mean? 17:26:10 I can test it quickly 17:26:52 ok, thansk 17:26:59 ok, it works well now 17:27:11 anything else on the groups portal? 17:27:36 ok, so originally after the sec upgrade, I started to work on this notification 17:28:00 when somebody register a new group, we are getting an email about that (or sends an email to ambassador@ ML) 17:28:19 cool 17:28:28 tom wants the email to go to ambassadors list 17:28:40 it is in progress, because I run into that upgrade issue 17:28:47 no problem 17:28:49 ok 17:28:53 the portal have a very nice Messaging system concept 17:28:54 next topic 17:29:10 so that's to status, I'll go with that after the ask solr task. 17:29:22 #topic activity board progress on github issues 17:29:26 #link https://github.com/VizGrimoire/VizGrimoireJS/issues 17:29:36 mrmartin, yep, askbot is priority 17:30:02 reed, Luis was in charge of filtering that list of issues 17:30:07 and one of them was closed 17:30:22 and I'd say that in the rest of them, he's waiting for some extra info 17:30:42 ok, I'll go through them 17:30:46 the one closed was this one: https://github.com/VizGrimoire/VizGrimoireJS/issues/55 17:30:57 and this should be deployed to the activity board in the following days 17:31:12 cook 17:31:17 cool 17:31:56 if there is nothing special to discuss, let's continue on the tracker 17:31:56 this is all the progress related to the GH issues 17:32:01 sure 17:32:25 sounds good... 17:32:29 anything else to discuss? 17:32:50 reed, not sure if mrmartin had the chance to have a look at the vagrant machine 17:32:54 although this is related to the first topic 17:33:14 I'll check it on this week, this sec fix was priority. 17:33:14 I think it's better to save time to focus on the items with shorter deadlines, like the Report 17:33:27 yeah, agreed with that priority, mrmartin 17:33:58 oh reed, is this the place to talk about the quarterly report progress? 17:34:04 I have a bunch of reading to do and things to discuss with dizquierdo_, so we should probably close the joined meeting and move to ad-hoc channel 17:34:11 sure 17:34:26 dizquierdo_, we can close the meeting, send mrmartin to work and you and I can focus on the quarterly report 17:34:27 ok. 17:34:35 ok 17:34:40 send mrmartin to sleep :D 17:34:43 :) 17:34:44 indeed :) 17:34:47 that's what I meant :) 17:34:53 #endmeeting