16:00:15 <smcginnis> #startmeeting Cinder
16:00:16 <openstack> Meeting started Wed Aug 24 16:00:15 2016 UTC and is due to finish in 60 minutes.  The chair is smcginnis. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:00:17 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:00:19 <openstack> The meeting name has been set to 'cinder'
16:00:20 <smcginnis> Agenda: https://wiki.openstack.org/wiki/CinderMeetings#Next_Cinder_Team_meeting
16:00:21 <Swanson> Hello
16:00:29 <jgregor> Hello!
16:00:31 <scottda> hola
16:00:31 <smcginnis> ping - dulek duncant eharney geguileo winston-d e0ne jungleboyj jgriffith thingee smcginnis hemna xyang1 tbarron scottda erlon rhedlind jbernard _alastor_ bluex patrickeast dongwenjuan JaniceLee cFouts Thelo vivekd adrianofr mtanino yuriy_n17 karlamrhein diablo_rojo jay.xu jgregor baumann rajinir wilson-l reduxio wanghao thrawn01 chris_morrell stevemar watanabe.isao,tommylike.hu
16:00:32 <ntpttr__> o/
16:00:33 <baumann> Hiya
16:00:34 <tommylikehu> hello
16:00:38 <geguileo> o/
16:00:44 <bswartz> .o/
16:00:46 <xyang1> hi
16:00:47 <jungleboyj> o/
16:00:51 <DuncanT> Hey
16:00:53 <smcginnis> Short agenda today.
16:01:00 <smcginnis> Probably means will run out of time. :)
16:01:12 <smcginnis> #topic Announcements
16:01:16 <tommylikehu> sounds cool
16:01:24 <smcginnis> It's non-client library freeze.
16:01:40 <smcginnis> If we get hemna's patch through, I'll be requesting a release later today.
16:01:46 <smcginnis> Of os-brick that is.
16:01:59 <patrickeast> Did that FC fix go in yet?
16:02:08 * patrickeast goes to look
16:02:20 <geguileo> smcginnis: Any chance of this getting in?  https://review.openstack.org/#/c/231135/
16:02:24 <smcginnis> patrickeast: I thought it did yesterday, but I woudl wait on that too.
16:02:25 <geguileo> patrickeast: No, not yet
16:02:35 <erlon> hey
16:02:37 <geguileo> patrickeast: If you are talking about mine
16:02:42 <tbarron> hi
16:02:42 <smcginnis> Oh yeah, wanted hemna to take a peek at that one.
16:02:58 <smcginnis> But it can probably go through.
16:02:59 <patrickeast> geguileo: yep that's the one
16:03:12 <smcginnis> patrickeast: You said you tested that and it improved things, right?
16:03:27 <patrickeast> smcginnis: yea it needs to be in
16:03:56 <geguileo> We have the same issue in iSCSI multipath cleanup: https://review.openstack.org/#/c/231134/
16:04:01 <hemna> geguileo, I'll look
16:04:07 <geguileo> hemna: Thanks!
16:04:23 <hemna> geguileo, xtremio fc is failing on that one
16:04:24 <hemna> fwiw
16:04:30 <hemna> have to make sure that's not an issue
16:04:45 <smcginnis> Next week is N-3, feature freeze, client library freeze, string freeze, and requirements freeze.
16:04:52 <smcginnis> So please keep that in mind.
16:05:15 <smcginnis> Soft string freeze, actually, but still - we should try to avoid making any translatable string changes if possible.
16:05:31 <smcginnis> That will really help the translation team get everything done in time.
16:06:00 <xyang1> hemna: where?
16:06:15 <smcginnis> #link https://review.openstack.org/#/c/357214/ Proposed Ocata schedule
16:06:26 <erlon> smcginnis: friday?
16:06:38 <geguileo> hemna: That was failing due to the volume type issue
16:06:44 <smcginnis> erlon: I think Thursday at the latest.
16:06:54 <erlon> smcginnis: ok
16:07:11 <smcginnis> Just so everyone is aware, Ocata is going to be shorter as we transition over to the new release cadence.
16:07:19 <smcginnis> Or rather Summit cadence.
16:07:27 <smcginnis> So Ocata will be a bit different.
16:07:36 <smcginnis> You can see the proposed schedule at that link.
16:07:37 <xyang2> geguileo: wasn't that one passed already?
16:07:44 <smcginnis> A lot more compressed.
16:07:59 <xyang2> geguileo: I didn't see XtremIO CI failure on your patch
16:08:12 <smcginnis> Another reason I would like O focused more on bug fixes and stabilization, if we can.
16:08:20 <Swanson> Need more and faster releases with less in them and buggier. Should go well.
16:08:31 <geguileo> xyang2: Apparently I didn't request the recheck, how do I trigger the recheck only for XtremIO_FC os-brick?
16:08:32 <hemna> xyang2, https://review.openstack.org/#/c/231135/
16:08:37 <hemna> FC failed
16:09:07 <smcginnis> Should be "run-EMC XIO"
16:09:07 <geguileo> Yeah, and I have just checked that the error is in the scheduler:  InvalidVolumeType: Invalid volume type: volume_type cannot be None
16:09:07 <bswartz> Swanson: I think you're being sarcastic but faster releases with less in them would be awesome IMO
16:09:27 <smcginnis> geguileo: Ah, doesn't have the fix. Rebase?
16:09:30 <geguileo> xyang2: Thanks!
16:09:44 <xyang2> geguileo: np
16:09:45 <smcginnis> bswartz: Sometimes it's hard to tell when he's being sarcastic or not. ;)
16:09:47 <jungleboyj> Swanson: So pessimistic!
16:10:07 <Swanson> jungleboyj, I'm a naturally sunny and positive person.
16:10:13 <smcginnis> Hah!
16:10:20 <geguileo> smcginnis: I don't think it needs a rebase, but let's see the results first  :-)
16:10:53 <smcginnis> geguileo: If it's still hitting the "volume type can't be null" thing, that fix should be in there.
16:11:07 <jungleboyj> Swanson: So I have heard.  ;-)
16:11:07 <smcginnis> But yeah, we'll see.
16:11:11 <hemna> ok if it pukes over that I'll just shove it through
16:11:18 <geguileo> lol
16:11:27 <geguileo> If it fails I'll rebase and resubmit
16:12:19 <smcginnis> The thing is gate will rebase on it's checks, but most CI's don't, so external issues aren't avoided (or hit) in third party CI without rebasing the patch.
16:12:40 <smcginnis> #topic Open Discussion
16:12:41 <geguileo> smcginnis: Oooooh, I didn't know that
16:13:02 <smcginnis> geguileo: Yeah, it's a pretty subtle thing.
16:13:10 <smcginnis> Nothing else on the agenda.
16:13:22 <smcginnis> The driver deprecations/unsupported thing seems to have some fans.
16:13:34 <smcginnis> I saw Ironic is going to adopt that approach too.
16:13:38 <DuncanT> I have just noticed on the post on mailing list re NFS clone
16:13:38 <hemna> that patch is on it's way in no ?
16:13:51 <jungleboyj> smcginnis: I have lost track of that.  Are we moving towards the approach we discussed?
16:13:58 <smcginnis> hemna: I gave +2. I think we need to bribe someone else. :)
16:14:04 <smcginnis> jungleboyj: Looks like it.
16:14:15 <hemna> smcginnis, so Ironic is going to do the same thing ?
16:14:21 <smcginnis> It makes the tag enforcers happy, it's probably better on our end users.
16:14:36 <DuncanT> That's on the minimum feature list
16:14:39 <jungleboyj> smcginnis: +2
16:14:39 <smcginnis> hemna: Looks like it. They had a thread on doing that and I didn't see any dissenters.
16:14:47 <hemna> ah cool.
16:14:55 <hemna> so Cinder is actually leading something for a change :)
16:14:55 <smcginnis> DuncanT: What's the NFS clone thread?
16:15:00 <smcginnis> hemna: ;)
16:15:12 <smcginnis> hemna: We're the troublemakers.
16:15:20 <DuncanT> smcginnis: Apparently the NFS driver doesn't support clone?
16:15:34 <bswartz> DuncanT: you need snapshots before you can do clone
16:15:43 <hemna> yes we are :)
16:15:48 <smcginnis> DuncanT: jungleboyj is working on that now, I believe.
16:15:55 <jungleboyj> DuncanT: I just opened a bug for that and am going to work on fixing that as well with the snapshot support.
16:16:03 <DuncanT> bswartz: Oh, that's on the minimum feature list too :-)
16:16:14 <jungleboyj> DuncanT: Yep.
16:16:22 <bswartz> DuncanT: it's been "almost done" for more than 2 years now
16:16:31 <tommylikehu> lol
16:16:42 <smcginnis> bswartz: True
16:16:43 <jungleboyj> bswartz: I am here to save the day!?!
16:16:45 <DuncanT> bswartz: I might put the deprecation patch up and force the issue
16:16:48 <tbarron> there were nasty nova dependencies
16:16:50 <smcginnis> But now jungleboyj has a vested interest in it. ;)
16:17:08 <erlon> tbarron: +1 yep
16:17:11 <tbarron> DuncanT: that might not have motivated nova ..
16:17:29 <bswartz> I will shed no tears if the NFS driver is ripped out
16:17:43 <DuncanT> tbarron: Vendors and distros can put pressure on nova if needed....
16:18:26 <erlon> smcginnis: Im doing some investigation as well, actually theres something beyond nova
16:18:32 <smcginnis> erlon: Oh?
16:18:48 <erlon> smcginnis: the NFS snapshots right?
16:19:35 <smcginnis> erlon: What's the issue beyond Nova?
16:20:31 <erlon> smcginnis: firstly Nova would need to consider the backing file chain for the snapshtos, currently it only takes the basefile
16:21:20 <erlon> smcginnis: them, libvirt need to attach the snapshot file, I couldn't get that working
16:21:38 <erlon> smcginnis: even using only virsh commands
16:21:56 <smcginnis> erlon: Hmm, not good.
16:22:36 <eharney> erlon: this is all stuff that should be working already... at least it did the last time i ran this code
16:22:42 <cFouts> o/
16:22:53 <erlon> smcginnis: yep, if you try to attach the basefile, which has the same permissions it works, if you try the snapshot it returns a permission error
16:23:26 <jungleboyj> eharney: I am going to play around with it some more later this week too.
16:23:32 <erlon> eharney: where you able to atach a snapshot to an instance?
16:24:05 <eharney> i'll give the code a quick smoke test shortly and see what i get
16:24:20 <smcginnis> OK, anything else today?
16:24:32 <jungleboyj> I was able to create and delete snapshots.  Hadn't tried attaching them.
16:25:13 <smcginnis> Going once...
16:25:26 <smcginnis> Going twice...
16:25:36 <smcginnis> OK, thanks everyone.
16:25:42 <jungleboyj> Thanks smcginnis !
16:25:46 <erlon> jungleboyj: what I observed is that after you create the snapshos, when you detach the volume, and re-attach it, nova will attach the base file
16:25:46 <smcginnis> #endmeeting