16:00:20 <jgriffith> #startmeeting cinder 16:00:21 <openstack> Meeting started Wed Apr 10 16:00:20 2013 UTC. The chair is jgriffith. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:22 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:00:24 <openstack> The meeting name has been set to 'cinder' 16:00:30 <jgriffith> hey folks! 16:00:35 <kmartin> hey 16:00:37 <xyang_> hi 16:00:41 <avishay> hi 16:00:45 <DuncanT> Hey 16:00:51 <hemna> morning 16:00:53 <dricco> hi 16:00:55 <eharney> hi 16:01:03 <jgriffith> nice turn out :) 16:01:20 <jgriffith> So I don't have much in the way of formalities today 16:01:26 <jgallard> hi 16:01:31 <winston-d> hi 16:01:32 <jgriffith> Trying to do too much before next week 16:01:38 <jgriffith> winston-d: do you EVER sleep! 16:01:54 <winston-d> jgriffith : :) 16:01:57 <jgriffith> Ok... let's roll 16:02:04 <jgriffith> #topic summit updates/plans 16:02:05 <vincent_hou_> hi 16:02:11 <jgriffith> vincent_hou_: hey there! 16:02:13 <avishay> vincent_hou_: hi 16:02:25 <smulcahy> lo 16:02:28 <jgriffith> I pushed session out to sched.org incase you didn't notice 16:02:37 <vincent_hou_> nice to see you. 16:02:44 <jgriffith> anything anybody sees as concerning so far? 16:03:02 <jgriffith> anybody think of a session topic that we need and don't have? 16:03:18 <avishay> jgriffith: didn't check the sched but did you get the notice that winston-d and i are merging our similar sessions? 16:03:26 <hemna> jgriffith, state machine? 16:03:41 <winston-d> avishay : yeah, i told jgriffith about that. 16:03:42 <jgriffith> avishay: well... I got it yesterday from winston-d 16:03:43 <hemna> are we touching that for Havana? 16:03:52 <jgriffith> avishay: but I had already pushed 16:04:01 <jgriffith> and I think we can fill two sessions on those topics 16:04:09 <jgriffith> so I scheduled them back to back and left them seperate 16:04:14 <avishay> jgriffith: yes we definitely can, ok 16:04:15 <DuncanT> I meant to ping Avishy about mentioning state machines duing smart shutdown but forgot 16:04:19 <jgriffith> we can meld the discussion across the two 16:04:25 <jgriffith> Oh snap! 16:04:27 <jgriffith> states! 16:04:32 <hemna> :P 16:04:51 <jgriffith> Ok... I have one reserved place-holder, that may be where that goes 16:05:03 <jgriffith> do we need an entire session for states? 16:05:18 <hemna> kinda felt like a big one? 16:05:21 <jgriffith> ie do we have enough thought around it to actually be productive in 40 minutes? 16:05:49 <hemna> not I 16:05:56 <winston-d> maybe not 16:05:59 <avishay> not sure 16:06:14 <jgriffith> I think we're going to miss it again 16:06:20 <jgriffith> Ok.. I proose this 16:06:20 <DuncanT> I can talk for 40 minutes on it, but I don't think it needs 40 minutes 16:06:29 <jgriffith> DuncanT: fair 16:06:36 <jgriffith> alright.. plan: 16:06:54 <jgriffith> We'll talk informally between now and summit sessions 16:07:05 <jgriffith> and we'll see where it might tie in to the existing sessions 16:07:20 <jgriffith> ie as was mentioned during the shutdown session 16:07:21 <kmartin> jgriffith: want to mention the etherpad naming convention? https://etherpad.openstack.org/Summit-Havana-Cinder-<session name> 16:07:39 <jgriffith> kmartin: Oh.. yes, thanks! 16:07:47 <jgriffith> so for those that don't know 16:07:51 <avishay> with capital letters? 16:08:10 <jgriffith> If you're doing a session create an etherpad as described. 16:08:25 <jgriffith> oops.. I'd prefer no caps, but if we've already started down that path, so beit 16:08:52 <avishay> any way to rename etherpads? i already made one. or just copy-paste into a new one? 16:09:04 <kmartin> some have been start already but easy to change 16:09:14 <hemna> soren, I've never "run" a session before 16:09:15 <jgriffith> avishay: I don't think you can rename, just copy/paste 16:09:22 <hemna> is there some preso required? 16:09:23 <avishay> jgriffith: ok no problem 16:09:24 <xyang_> jgriffith: can you do a summit session with slides? 16:09:32 <hemna> or just show up with an etherpad and start talking? 16:09:42 <jgriffith> xyang_: etherpad and start talking typically 16:09:43 <xyang_> jgriffith: is etherpad required? 16:09:46 <DuncanT> Copy/ paste an put a link in the old one to the new one 16:09:47 <jgriffith> xyang_: yes 16:09:50 <hemna> ok 16:09:55 <avishay> DuncanT: OK 16:10:05 <jgriffith> ehterpad allows us to have a record of the conversation 16:10:06 <hemna> at this point for the 2 sessions, I don't have much other than the ideas for the topics 16:10:12 <kmartin> solved: summit-havana-cinder-<session-name> 16:10:14 <xyang_> ok thanks 16:10:18 <jgriffith> it also allows people to participate remotely 16:10:29 <jgriffith> kmartin: thank you sir! 16:10:34 <winston-d> xyang_ : i think you were there in last summit? 16:10:39 <jgriffith> hemna: that's fine 16:10:44 <jgriffith> hemna: not uncommon 16:10:47 <xyang_> winston-d: yes 16:10:50 <hemna> ok 16:10:55 <hemna> I just feel unprepared :P 16:11:06 <hemna> It's a learning process at this point for me. 16:11:09 <jgriffith> hemna: Well fix that dan it :) 16:11:12 <jgriffith> hemna: just kidding 16:11:19 <jgriffith> hemna: So remember the idea is to collaborate 16:11:28 <jgriffith> hemna: not to just say "here's what I'm doing" 16:11:40 <avishay> jgriffith: where do we put the etherpads? i saw some wiki link but it didn't have the cinder sessions on it 16:11:56 <winston-d> hemna : :) it's more like discussion than presentation. so if you can get others to talk you don't even have to prepare anything. :p 16:11:56 <hemna> :) 16:12:06 <kmartin> avishay: https://etherpad.openstack.org 16:12:06 <hemna> ok perfect 16:12:11 <jgriffith> avishay: That's what kmartin was pointing out 16:12:12 <hemna> I'm looking forward to it 16:12:29 <jgriffith> so you just create it with the convention he gave above ^^ 16:12:34 <jgriffith> is that what you meant? 16:12:42 <winston-d> there should be a wiki to keep track of all etherpads 16:12:44 <avishay> i mean, how do people find the etherpad for a session 16:12:45 <DuncanT> Is there an index of pads somewhere? 16:12:51 <hemna> winston-d, +1 16:12:59 <jgriffith> avishay: and if I *find* the master key (which we have for the entire summit) I'll add them 16:13:08 <hemna> you can't search on the etherpad site :( 16:13:10 <vincent_hou> summit-havana-cinder-<session-name>. do we keep the <>? 16:13:18 <avishay> vincent_hou: no 16:13:22 <jgriffith> DuncanT: typically we'll have a key when the sessions get going 16:13:23 <DuncanT> vincent_hou: No 16:13:29 <jgriffith> DuncanT: also that's why the naming convention 16:13:32 <vincent_hou> ok 16:13:46 <jgriffith> But I'll find out where we're putting the key 16:13:46 <avishay> so meanwhile i'll make an etherpad and hope i don't lose the URL :P 16:14:00 <jgriffith> I should've updated the session info with the etherpad link 16:14:06 <avishay> https://wiki.openstack.org/wiki/Summit/Grizzly/Etherpads 16:14:06 <jgriffith> I meant to do that after last summit 16:14:07 <DuncanT> jgriffith: I know I screwed up all mine last year and had to do some rapid fixup, but the gods of wifi coverage were kind and I managed to sort it before the session started :-) 16:14:22 <kmartin> DuncanT: Grizzly etherpads: https://wiki.openstack.org/wiki/Summit/Grizzly/Etherpads 16:14:23 <avishay> oops that's the old one 16:14:28 <jgriffith> haha... Yes, you and I both had some issues as well as some others 16:14:33 <avishay> https://wiki.openstack.org/wiki/Summit/Havana/Etherpads 16:14:44 <jgriffith> avishay: perfect! thanks 16:15:19 <hemna> ok who's creating the Cinder section? 16:15:22 <hemna> so we don't all do it :) 16:15:34 <jgriffith> I'm doing it now 16:15:44 <jgriffith> Once it's up you can put your own session etherpad link in 16:15:52 <avishay> just an off-topic announcement - the patch to fix jenkins tests just went through - i will add 'recheck's 16:15:52 <hemna> thanks 16:17:12 <smulcahy> avishay: thanks - you putting rechecks on all? 16:17:27 <jgriffith> avishay: coolness! 16:17:35 <avishay> smulcahy: well those for master ... we'll need to deal with grizzly backports now.... 16:17:38 <avishay> :( 16:17:57 <jgriffith> Ok... the cinder section is up 16:17:59 <smulcahy> yeah, they need to pull your patch too I think 16:18:30 <jgriffith> alright... 16:18:37 <jgriffith> So on sessions, there's one outlier here 16:19:04 <jgriffith> http://summit.openstack.org/cfp/edit/45 16:19:23 <jgriffith> So this one is an interesting topic and I think it's important 16:19:26 <avishay> jgriffith: try again without the 'edit' 16:19:31 <avishay> jgriffith: 'forbidden' 16:19:33 <jgriffith> however, I don't feel there's enough info 16:19:35 <jgriffith> DOH 16:19:38 <jgriffith> every time!!! 16:19:46 <avishay> :) 16:19:48 <winston-d> s/edit/details/ 16:20:31 <jgriffith> http://summit.openstack.org/cfp/details/45 16:20:35 <DuncanT> jgriffith: Unconference it ? 16:20:36 <avishay> jgriffith: did you talk to lifeless ? 16:20:38 <jgriffith> winston-d: thanks 16:20:40 <jgriffith> DuncanT: yes 16:20:49 <jgriffith> avishay: yes extensively 16:21:05 <jgriffith> My concern is that the problem definition isn't very clear 16:21:13 <jgriffith> we'll never be productive in 40 minutes 16:21:21 <avishay> jgriffith: i see 16:21:24 <jgriffith> It took me that long to get all of the pieces to this put together 16:21:29 <jgriffith> let alone think about solving it :) 16:21:39 <jgriffith> So I've prposed to him we unconf it 16:21:40 <DuncanT> It does seem like an interesting problem space though 16:21:46 <jgriffith> DuncanT: indeed 16:22:02 <jgriffith> but it's a bit vague on what it is 16:22:12 <jgriffith> it doesn't fit nicely into any existing model's IMO 16:22:18 <jgriffith> anyway... 16:22:22 <jgriffith> wanted to make everyone aware 16:22:27 <jgriffith> we'll be talking about it 16:22:37 <jgriffith> but may not have a formal session 16:23:02 <winston-d> thanks for the heads-up. that's interesting topic 16:24:16 <jgriffith> cool cool 16:24:29 <jgriffith> #topic bugs 16:24:33 <DuncanT> Are we still planning on a testing unconf too? 16:24:42 <jgriffith> DuncanT: indeed 16:24:51 <j_king> yay testing. :) 16:24:56 <jgriffith> j_king: :) 16:24:57 <avishay> what's an unconference? 16:25:07 <jgriffith> Unconference is the same as a session 16:25:10 <jgriffith> but we hijack a room 16:25:19 <jgriffith> and just get together even less formally 16:25:24 <avishay> gotcha 16:25:54 <jgriffith> So there's a number of annoying little things showing up 16:26:07 <jgriffith> I just wanted to put the call out, PLEASE make sure you're testing this week 16:26:12 <hemna> Is there a list of these unconfs ? 16:26:19 <jgriffith> and of course triaging/filing and fixing 16:26:24 <jgriffith> and don't forget about reviews 16:26:26 <DuncanT> hemna: They get posted at the event 16:26:28 <jgriffith> hemna: no 16:26:29 <hemna> k 16:26:55 <jgriffith> That's all I got... anybody else? 16:26:59 <jgriffith> #topic open discussion 16:27:25 <hemna> we have a stress test tool that I wrote here that we are using to stress test cinder and our drivers 16:27:35 <jgriffith> hemna: nice! 16:27:52 <avishay> anyone know when does the next grizzly version come out with backports? 16:27:58 <avishay> usually 2 months, right? 16:28:01 <hemna> it's not really ready for public consumption, but would we be interested in making it available for public consumption or as part of the project somehow? 16:28:14 <jgriffith> avishay: thorw it on your github :) 16:28:23 <jgriffith> s/thorw/throw/ 16:28:31 <jgriffith> GRRRR 16:28:32 <jgriffith> hemna: 16:28:39 <avishay> jgriffith: right, that'll fly :) 16:28:45 <DuncanT> hemna: Sounds good - we've got a similar tool. There's a /really/ old version on github but another release would be a plan 16:28:53 <hemna> ok 16:29:00 <hemna> I'll see what I can do 16:29:13 <jgriffith> hemna: so if it can be made generic for the other drivers, it would be really cool to contribute! 16:29:23 <hemna> It's very generic right now 16:29:23 <avishay> indeed 16:29:25 <jgriffith> ie tempest augment etc 16:29:28 <hemna> it's a threaded python app 16:29:36 <jgriffith> hemna: excellent! 16:29:42 <hemna> that basically uses the cinder client apis 16:29:51 <jgriffith> hemna: we seem to think alike ;) 16:29:53 <hemna> as well as the nova apis 16:29:55 <j_king> shareable block devices among several instances? 16:29:58 <hemna> to create N vms 16:30:07 <hemna> Y volumes, then attaches, detaches, etc 16:30:11 <jgriffith> I think it would be interesting to get something like that built up and added to CI 16:30:13 <avishay> j_king: that will be discussed at the summit 16:30:23 <vincent_hou> hemna: will it be an independent project? 16:30:26 <j_king> avishay: ah cool.. I'd like to follow that one closely 16:30:30 <hemna> We've found bugs in our drivers using the tool 16:30:41 * j_king interested in distributed coordination 16:30:44 <hemna> let me see if I can approval to post the code on my github 16:30:50 <hemna> then we can decide what to do with it....if anything 16:30:53 <jgriffith> hemna: that would be awesome 16:30:58 <avishay> hemna: +1 16:31:24 <winston-d> hemna : tempest has already some stress testing cases, maybe that's the right place to contribute 16:31:31 <hemna> ok 16:31:37 <hemna> I haven't mucked w/ tempest yet 16:31:47 <jgallard> winston-d, +1 16:32:27 <avishay> hmmm end meeting? 16:32:31 <jgallard> just for information, we have started some thoughts about: a way to place instances nearly to their volumes with ceph backend, for those who are interested an etherpad is available here: https://etherpad.openstack.org/instance-volume-collocation 16:32:32 <hemna> we have another script that is a continuous test script...it runs indefinitely to do various cinder actions. 16:32:45 <xyang_> kmartin: have you found out where the new block storage admin guide is located 16:33:05 <kmartin> xyang_: yep 16:33:09 <avishay> jgallard: IBM has similar thoughts for GPFS 16:33:21 <xyang_> kmartin: link? 16:33:32 <kmartin> xyang_: http://docs.openstack.org/trunk/openstack-block-storage/admin/content/index.html 16:33:37 <jgallard> avishay, oh great! do you have some pointers? 16:33:39 <xyang_> kmartin: thanks 16:33:55 <avishay> jgallard: https://blueprints.launchpad.net/cinder/+spec/gpfs-vm-placement 16:34:23 <jgallard> avishay, ok, I didn't saw it, thanks 16:34:26 <avishay> jgallard: http://summit.openstack.org/cfp/details/228 16:34:35 <winston-d> jgallard : avishay will there be session for these two? 16:35:06 <avishay> winston-d: jgallard: the second link is a session 16:35:34 <jgallard> winston-d, probably it will be: http://openstacksummitapril2013.sched.org/event/4dec95bea6e9d109f32448dc66cd6275#.UWVfsBeBrgU ? 16:36:02 <avishay> jgallard: don't think so, let me check 16:36:26 <kmartin> jgriffith: https://etherpad.openstack.org/summit-havana-cinder-multi-attach-and-ro-volumes was added to https://wiki.openstack.org/wiki/Summit/Havana/Etherpads#Cinder 16:36:44 <jgallard> avishay, ok, thanks 16:36:58 <jgriffith> kmartin: :) 16:37:04 <hemna> added my 2 sessions to the wiki 16:37:06 <jgriffith> cool.. we're rolling now 16:37:24 <avishay> jgallard: kind of a pain to map proposals to sessions 16:37:36 <jgallard> avishay, YES!!! 16:37:37 <jgallard> :) 16:38:34 <rushiagr> hi, sry m late 16:39:18 <jgriffith> rushiagr: no worries 16:39:33 <jgriffith> rushiagr: there seems to be something interesting going on with the netapp tests 16:39:39 <jgriffith> rushiagr: ie they're failing :( 16:39:50 <jgriffith> rushiagr: not sure if folks already reched out to you on that or not? 16:40:10 <rushiagr> jgriffith: yeah, was tracking that 16:40:15 <jgriffith> rushiagr: cool 16:40:25 <jgriffith> alright.. I really don't have anything 16:40:34 <jgriffith> and we're all around in #cinder anyway 16:40:38 <rushiagr> made some progress on that, but yet to decide how to fix it off 16:40:49 <jgriffith> shall we call it a meeting? 16:40:58 <hemna> sounds good 16:41:02 <jgriffith> alright 16:41:06 <jgriffith> thanks everyone! 16:41:28 <jgriffith> For those that will be in Portland, looking forward to meeting you all in person 16:41:33 <rushiagr> thanks 16:41:35 <jgallard> thanks! 16:41:38 <xyang_> thanks 16:41:43 <winston-d> see u next week! 16:41:49 <vincent_hou> c u 16:41:49 <jgriffith> For those that can't make it, I try to get on IRC and maintain some communications on what's going on 16:41:51 <xyang_> see u 16:41:59 <jgriffith> #endmeeting