19:01:21 #startmeeting Fedora Cloud SIG 19:01:21 Meeting started Wed Mar 25 19:01:21 2015 UTC. The chair is kushal. Information about MeetBot at http://wiki.debian.org/MeetBot. 19:01:21 Useful Commands: #action #agreed #halp #info #idea #link #topic. 19:01:26 #topic Roll Call 19:01:36 .hellomynameis kushal 19:01:37 kushal: kushal 'Kushal Das' 19:01:37 .hellomynameis slimshady 19:01:40 dustymabe: Sorry, but you don't exist 19:01:41 .hellomynameis rtnpro 19:01:42 rtnpro: rtnpro 'Ratnadeep Debnath' 19:01:48 .hellomynameis dustymabe 19:01:49 dustymabe: dustymabe 'Dusty Mabe' 19:01:49 dustymabe, :) 19:02:19 .hello oddshocks 19:02:22 oddshocks: oddshocks 'David Gay' 19:02:31 .hello roshi 19:02:34 roshi: roshi 'Mike Ruckman' 19:03:15 Anyone else? 19:03:46 * jsmith is here 19:04:29 don't think so 19:04:35 #topic Last meeting's action items 19:04:48 * jzb help find scollier additional maintainer for dockerfiles 19:04:49 * jzb (and others) update dockerfiles for dnf 19:04:49 * kushal wil ask Lennart for more specific use case and details for #93 19:04:49 * create proposal page on wiki for Atomic spin for f23 forward, call for 19:04:49 vote on mailing list. 19:04:49 * jzb create proposal page on wiki for Atomic spin for f23 forward, call 19:04:51 for vote on mailing list. 19:05:34 jzb, Any updates? 19:05:44 .fas corey84 19:05:45 Corey84: corey84 'Corey84' 19:05:49 (a bit late) 19:06:06 I am yet to ask Lennart was he was traveling to Indonesia and on leave till 29th. 19:06:41 Rest of them are on jzb 19:07:25 I think we can move to the tickets 19:07:35 #topic Maintaining Fedora docker images for f22 https://fedorahosted.org/cloud/ticket/97 19:07:44 Anyone else has update for this? 19:07:54 I'm going to do some helping of scollier with the docker files 19:08:04 haven't gotten to it yet - but it's on the radar at least 19:08:27 we talked a bit about it after the last meeting - and checks jzbs first item from last meeting 19:08:35 roshi, can you please add an action item then? 19:08:57 kushal I think for this ticket someone was going to go to server WG and try to see if "cloud" could own Docker image? 19:09:06 what do you mean kushal ? 19:09:22 kushal: from the ticket looks like it was JZB 19:09:26 roshi, meeting action item :) 19:09:31 dustymabe, yup. 19:09:37 dustymabe, and he is not here today. 19:09:45 :( 19:10:02 Maybe post in the ticket to ask for status update? 19:10:11 dustymabe, can you please do that? 19:10:17 I guess I mean what for? For me helping with it? 19:10:25 roshi, yeah :) 19:10:50 did you give char to anyone yet? 19:10:57 I think you have to do it since you're chair 19:11:36 roshi, ah 19:11:42 roshi, this should work anyway 19:11:57 #chair roshi dustymabe Corey84 rtnpro oddshocks 19:11:57 Current chairs: Corey84 dustymabe kushal oddshocks roshi rtnpro 19:12:07 roshi, sorry, I missed that 19:12:31 no worries :) 19:13:33 Moving to next ticket then. 19:13:34 #action roshi to help with docker files 19:13:44 kushal: I updated the ticket 19:13:48 dustymabe, thanks :) 19:14:03 #topic Care and Feeding, Fedora Dockerfiles https://fedorahosted.org/cloud/ticket/84 19:14:19 hi kushal, i must say, I'm very, very distracted in meetings ATM. 19:14:21 sorry. 19:14:31 scollier, no problem. 19:15:10 Moving to next. 19:15:25 #topic Getting sha256sum published for the cloud images https://fedorahosted.org/cloud/ticket/93 19:15:32 No updates for this too. 19:15:42 Lennart is on holiday till 29th of this month. 19:16:33 I think we will have a lot more time for open floor today :) 19:16:43 #topic Atomic as separate spin (or, going the other way, main cloud edition)? https://fedorahosted.org/cloud/ticket/96 19:17:25 jzb, is supposed to create the proposal page for the same. 19:17:34 For Fedora 23 onwards. 19:17:43 he sent an email to the list with some proposal 19:17:45 was that it? 19:18:17 dustymabe, ah, just now saw that email 19:18:22 https://fedoraproject.org/wiki/Cloud/Atomic-2-Week-Proposal 19:18:58 dustymabe, thanks :) 19:19:02 np 19:20:33 Moving to next. 19:20:34 #topic Producing Updated Cloud/Atomic Images https://fedorahosted.org/cloud/ticket/94 19:20:41 oddshocks, any updates on this one? 19:21:15 Haven't heard from dgilmore or releng on this 19:21:39 looks like this is probably not going to happen til at least F22? 19:21:45 what's on the ticket is all I know 19:21:48 dustymabe: probably 19:23:08 rel-eng created new images. 19:23:13 I did the initial testing. 19:23:25 There was the issue of missing eth0 sometime. 19:23:33 I think I dropped a mail to the list. 19:23:46 I have more to say during open floor. 19:24:26 Yeah, I know there have been updates. But I don't think there's been a whole system set up for how it'll happen between us/them/qa yet 19:25:52 oddshocks, Okay. 19:26:24 I think we can move to Open Floor now. 19:26:56 oddshocks, btw, can you poke rel-eng once? 19:27:06 I mean over email / track 19:28:32 kushal: yeah 19:28:37 thanks 19:28:40 I have before, sent out an email. No replies. 19:28:43 I'll send another 19:28:53 #topic Open Floor 19:28:54 Just to ask about the status of the whole collaborative process? 19:29:01 oddshocks, yup. 19:29:03 cool 19:29:17 I have few points for open floor. 19:29:51 I was trying out the kickstart with the hack for languages related to https://bugzilla.redhat.com/show_bug.cgi?id=156477 19:29:54 https://bugzilla.redhat.com/show_bug.cgi?id=156477 19:30:28 x86_64 rawhide is 264 MB in my local box without the hack, and 240MB with the hack. 19:30:36 Means there is difference. 19:31:08 But as my imagefactory setup happily fails to reboot the image, it actually declares it a complete fail. 19:31:19 I will have to mount the image manually, and test it. 19:31:23 Images rather. 19:31:53 On the other hand, yesterday I wrote a blog post on Tunir. 19:32:02 http://kushaldas.in/posts/tunir-a-simple-ci-with-less-pain.html 19:32:23 I also ported the cloud tests using python unittest for tunir. 19:32:33 https://github.com/kushaldas/tunirtests 19:32:38 kushal, +1 19:32:47 kushal: sounds great 19:32:50 Feel free to test the system and add new tests. 19:33:08 It already found few existing issues in the current images. 19:33:38 I am planning to add extra tests for atomic image. 19:33:54 Things like checking if docker actually works or not. 19:34:32 roshi, Do you have anything on mind for new testcases? 19:35:03 For the normal cloud image, I am already testing installation of a package (pss in this case). 19:35:11 not really, but we should really get stuff integrated into taskotron - so we're running tests in one place and not multiple 19:35:41 taskotron is going to be the place for tests to be ran - and then we won't have to maintain the infro for it or ask someone else to do it 19:35:44 roshi, taskotron can just reuse my unittests, they are simple python unittests. 19:35:57 we'd just have to maintain my tasks 19:36:15 you just have to get them into the task format that taskotron can consume 19:36:21 Okay. 19:36:31 I'm working on bits for the disposable clients for them to be able to run the tests on 19:36:48 so if you could help with that effort (cloud tests in taskotron), it'd be great 19:37:00 and tunir already has some bits in it that can be used, I'd imagine 19:37:07 roshi, Okay, I will look into that. 19:37:24 I'm not totally familiar with the taskotron internals - but that's been a goal for a long time 19:37:40 perhaps get with tflink and find out what he has to say about what's needed to get some more movement on that 19:37:40 can I share updates from my end? 19:38:03 but that's all I had - because then we can get new testcases into that system once the system is working 19:38:16 sure thing rtnpro - that's EOL for me 19:39:10 I started working on creating a framework on top of datagrepper to empower widgets and dashboards needed by various fedora apps 19:39:12 https://github.com/rtnpro/fedwidgets 19:39:37 I got my initial prototype validated by threebean 19:39:41 * roshi is excited about this bit 19:40:05 the concept is pretty simple 19:40:29 have a denormalized data layer on top of datagrepper for faster data retrieval 19:41:16 people write widgets in Python to listen to fedmsg and save denormalized data in proper format 19:41:38 and widgets also render data (JSON) through an API 19:42:08 people can use JS to render this data as needed 19:42:52 I could not work on this idea last week and this week as I was in between finalizing job change 19:43:03 it's almost done now 19:43:14 so, I will resume work on it ASAP 19:43:25 I will keep you folks posted :) 19:43:41 rtnpro: thanks 19:44:19 EOL 19:44:33 Any other update for open floor? 19:45:37 nothing here 19:45:39 Yeah 19:45:42 First, AMI lifetimes 19:46:06 I sent out a summary to the list of how people seemed to feel about them https://lists.fedoraproject.org/pipermail/cloud/2015-March/005148.html 19:46:41 https://lists.fedoraproject.org/pipermail/cloud/2015-March/005148.html 19:46:54 I don't want to take up too much time here, but I propose a vote, and also to immediately (without a vote, besides now), delete all alpha, beta, TC, and RC AMIs from before the F21 release 19:47:09 sorry, that first part was confusing 19:47:28 the first vote i mentioned was for the lifetime of final release AMIs, or if we keep them forever 19:47:51 and I'm just asking for +1s now to get rid of the non-final F21 amis people tested with 19:47:59 I already got one +1 on the list 19:48:03 oddshocks, +1 from me on that. 19:48:34 oddshocks: +1 on getting rid of non-final AMIs 19:48:45 Cool, if no objections, will do today 19:48:59 I think we should send a follow up email to the list 19:49:13 dustymabe, +1 19:49:14 dustymabe: as in, "we're about to do this because we have +1s, any objections?" 19:49:18 and state what we have decided along with a time and date that the action will be performed 19:49:21 oddshocks, yup. 19:49:37 this gives people a chance to come out of the woodwork to enlighten us on possible issues. 19:49:50 but also gives us a course of action 19:50:13 cool 19:50:14 will do 19:50:34 I have one other thing for open floor. Did anyone see that thread about enhanced networking in our AMIs? 19:51:00 We're running out of time here and I don't want to leave anyone else without time, but if people wanna check that out and give any input, that'd be cool 19:51:38 oddshocks, I saw, but have to read it again to understand it properly. 19:52:06 oddshocks: I am waiting on an okay that things are right, or a no they are not we have the fixes in and please do a new tree 19:52:10 Basically, "are there any problems with doing this" or whatever. The guy wants us to toggle the flag on to enable enhanced networking (kernel mod I think) for our AMIs 19:52:12 oddshocks: I have gotten neither 19:52:22 thanks, dgilmore :) 19:52:38 oddshocks: at this point everything is in the CloudWG's hands 19:52:48 dgilmore, Okay, understood. 19:52:55 as it has been for a few weeks now 19:53:59 okay, I must have misunderstood 19:54:03 anyway, that's all from me 19:54:36 ok all of that was a bit confusing 19:54:41 was dg talking about sriov? 19:54:53 dgilmore, we will test out the networking changes as suggested by in the bugzilla, and then will ask you to create a new tree. 19:55:15 Okay, that line is confusing due to missing words :( 19:56:08 dennis was talking about cloud image updates with releng 19:56:26 I was talking about enhanced networking kernel mod in our AMIs 19:56:32 all good now 19:57:16 We have 3 minutes left. 19:58:41 I am close this meeting then. 19:58:48 +1 19:58:50 If no one has anything to say. 19:58:55 3 19:58:56 2 19:58:58 1 19:59:01 #endmeeting