16:01:27 <kozhukalov> #startmeeting Fuel 16:01:28 <openstack> Meeting started Thu May 21 16:01:27 2015 UTC and is due to finish in 60 minutes. The chair is kozhukalov. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:01:29 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:01:32 <openstack> The meeting name has been set to 'fuel' 16:01:35 <kozhukalov> #chair kozhukalov 16:01:36 <openstack> Current chairs: kozhukalov 16:01:54 <kozhukalov> agenda for today is quite small 16:02:12 <kozhukalov> #link https://etherpad.openstack.org/p/fuel-weekly-meeting-agenda 16:02:31 <kozhukalov> #topic HCF status 16:02:56 <kozhukalov> we still have more than 5 bugs 16:03:11 <kozhukalov> so we can not introduce hcf 16:03:20 <kozhukalov> but most of them are in progress 16:03:48 <mihgen> kozhukalov: any success with weird MCollective issues? 16:04:23 <kozhukalov> we've fixed 3 bugs related to mcollective this week 16:04:35 <kozhukalov> but i believe you are asking about this one 16:04:46 <kozhukalov> #link https://bugs.launchpad.net/fuel/+bug/1452153/ 16:04:46 <openstack> Launchpad bug 1452153 in Fuel for OpenStack "IBP does not start due to mcollective problem" [High,Confirmed] - Assigned to Fuel provisioning team (fuel-provisioning) 16:05:08 <mihgen> yes 16:05:24 <kozhukalov> first, last comments to this bug was from env based on 377 build 16:05:30 <kozhukalov> quite old one 16:05:41 <kozhukalov> second there is no access to this env 16:06:07 <kozhukalov> but I've asked them and hope they'll give access tonight 16:06:29 <mwhahaha> I think by chance i ran in to that problem earlier this week. mcollective was running and i ended up having to start another instance and the deploy finished 16:06:29 <kozhukalov> third, from logs it became clear 16:07:11 <mihgen> any idea how we could repeatedly reproduce it? 16:07:13 <kozhukalov> that mcollective stoped and didn't send any log messages right after it was restarted by nailgun-agent 16:07:42 <mwhahaha> no idea for mine, there were actually 2 mcollective agents running at the time. when i did a service restart it started up a 3rd 16:07:56 <mwhahaha> it hasn't been consistant 16:08:08 <kozhukalov> this bug is about failed mcollective on bootstrap, not on target OS 16:08:28 <mwhahaha> yea it was stuck durring the bootstrap 16:09:01 <kozhukalov> anyway, access is needed 16:09:12 <kozhukalov> to figure out what is going on 16:09:50 <kozhukalov> in a nutshell, this bug is in progress 16:10:29 <kozhukalov> network is ok, iptables is ok, it is 100% mcollective related problem 16:11:08 <kozhukalov> are there any other questions/comments about hcf 16:11:12 <kozhukalov> ? 16:11:53 <kozhukalov> ok, open discussion then 16:12:00 <kozhukalov> #topic Open Discussion 16:12:01 <mihgen> we are close to it.. let's nail down the remaining issues.. 16:12:35 <kozhukalov> mihgen: we even have nailgun for that 16:12:44 <kozhukalov> -) 16:13:13 <kozhukalov> is there anything to discuss? 16:13:36 <kozhukalov> many people are treveling to Openstack Summit 16:13:45 <angdraug> anyone heard anything interesting from Vancouver? 16:13:58 <kozhukalov> yes, i've heard 16:14:14 <kozhukalov> people werer very excited about app catalog 16:14:18 <kozhukalov> great success 16:14:37 <angdraug> I'd like to plug the user survey results: 16:14:38 <angdraug> #link http://superuser.openstack.org/articles/openstack-users-share-how-their-deployments-stack-up 16:14:51 <angdraug> Fuel did very well among the deployment tools :D 16:15:15 <angdraug> #3 overall, #2 in dev/test use cases 16:15:21 <ogelbukh> I wonder why the grand total is >100% tho :) 16:15:30 <kozhukalov> interesting 16:15:35 <angdraug> multichoice question 16:15:39 <ogelbukh> does fuel counts as puppet as well? 16:15:46 <angdraug> I suspect that yes 16:16:16 <angdraug> survey didn't originally have Fuel on the list, it was added after overwhelming number of people entered "Fuel" as the "other" option 16:16:25 <kozhukalov> fuel is third place 16:16:59 <angdraug> so it's likely that some fuel users picked puppet 16:17:23 <angdraug> next time, it will be more objective since it will be in the list from the start 16:17:25 <kozhukalov> yes, we hope so 16:17:56 <mihgen> no tripleO in there? 16:18:44 <angdraug> what's that? :p 16:18:57 <kozhukalov> by the way I'm a little bit disappointed by the fact that many important features were thrown away from 7.0 release 16:19:13 <kozhukalov> mihgen: any comments on the 7.0 scope? 16:19:57 <mihgen> we'd need to update blueprints. With the current team in place we have to be realistic if we want to fit into 7.0 schedule (let me share link ) 16:20:12 <mihgen> #link https://wiki.openstack.org/wiki/Fuel/7.0_Release_Schedule 16:20:35 <mihgen> many things can be done as plugins 16:20:54 <mihgen> but for core work, main drivers would be granular deployment and flexible networking 16:21:09 <IvanKliuk> mihgen: is there features' list for 7.0? 16:21:10 <kozhukalov> mihgen: does that mean that all team members are going to be busy on 7.0 features? 16:21:21 <mihgen> this should become a baseground for further changes 16:22:00 <mihgen> we'd need to reflect plannings in the list of blueprints for the next release 16:22:14 <kozhukalov> i mean there are some quite complicated features which probably need more time than just one month, and they have a chance to be thrown away from 7.1 as well 16:22:24 <ogelbukh> I'm a little concerned with advanced networking becoming kinda SDN solutoin of it's own.. 16:23:03 <mihgen> ogelbukh: what do you mean by that? You've got to do underlay anyway and Fuel should be flexible here 16:23:05 <ogelbukh> but the most disappointing thing is postponing volume manager works 16:23:07 <mihgen> currently it's not 16:23:26 <ogelbukh> yes, I mean flexibility of underlay 16:23:41 <mihgen> ogelbukh: we've not found enough python folks to work on volume manager 16:24:01 <mihgen> so if there are any from other than Fuel Core teams, then it might be possible 16:24:17 <ogelbukh> mihgen: well, as it's refactoring work, I guess even 1 developer can make a difference 16:24:45 <mihgen> ogelbukh: current estimate I'm aware of is 2.5 folks for the whole cycle 16:25:01 <kozhukalov> ogelbukh: i'd say it rather total re-working than re-factoring 16:25:02 <ogelbukh> mihgen: for the full scope of the work? 16:25:03 <mihgen> the work can be started of course, especially design 16:25:07 <ogelbukh> kozhukalov: oh 16:25:11 <ogelbukh> alright then :) 16:25:16 <ogelbukh> yeah 16:25:31 <ogelbukh> I'd say let's create a spec by the 7.0 16:25:38 <mihgen> ogelbukh: yeah here the issue is it needs total rework finally.. 16:25:51 <mihgen> with many workarounds applied over time 16:25:59 <angdraug> ironic how the term "refactor" was introduced as an opposite to "rewrite", and now it became a synonym ... 16:26:09 <kozhukalov> yes, I think, at least the design must be in place before 7.0 ff 16:26:14 <ogelbukh> because I'd really like to understand what it should look like before the first patch :) 16:26:45 <ogelbukh> (and what it shouldn't look like fwiw) 16:26:54 <ogelbukh> kozhukalov: +1 16:27:55 <kozhukalov> in fact, i've started working on design of new volume manager but 16:28:06 <kozhukalov> haven't finished yet 16:28:33 <kozhukalov> still in my head 16:28:47 <ogelbukh> kozhukalov: OK :) 16:29:36 <kozhukalov> it is important not to make it pain in my a.. 16:29:41 <kozhukalov> -) 16:29:56 <kozhukalov> ok, any other updates, comments, suggestions? 16:30:39 <IvanKliuk> is there features' list for 7.0? 16:30:51 <kozhukalov> is someone is going to make a talk about Fuel during summit? 16:31:03 <kozhukalov> IvanKliuk: yes, there is 16:31:12 <kozhukalov> but it is in jira 16:31:33 <IvanKliuk> kozhukalov: OK 16:31:35 <kozhukalov> IvanKliuk: please ask akislitsky for the link 16:31:35 <angdraug> it should all be reflected in blueprints in LP, eventually 16:32:06 <kozhukalov> angdraug: right, but the entire discussion is in jira 16:32:35 <angdraug> yes :( 16:32:35 <kozhukalov> about which features are going to be a part of 7.0 and which aren't 16:33:23 <kozhukalov> ok, looks like we are done 16:33:33 <kozhukalov> closing? 16:33:46 <kozhukalov> thanks everyone for attending 16:34:00 <kozhukalov> #endmeeting