15:00:39 #startmeeting cinder_bs 15:00:39 Meeting started Wed Sep 22 15:00:39 2021 UTC and is due to finish in 60 minutes. The chair is enriquetaso. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:39 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:39 The meeting name has been set to 'cinder_bs' 15:01:04 #topic os-brick bug 15:01:13 simondodsley :) 15:01:22 Thanks 15:01:26 Bug https://bugs.launchpad.net/os-brick/+bug/1944474 was raised yesterday and the customer (a large global telco) are very keen for it to be looked at. 15:01:39 It appears that when an error 6 is retuned from iscsiadm there are no reties made even after the latest patch https://review.opendev.org/c/openstack/os-brick/+/796671 which is mmetrged into os-brick 3.0.7 15:01:39 <_pewp_> hemna ( ・_・)ノ 15:01:52 The customer would like these retries otherwise they have to keep checking the hypervisor and manually rescanning to get all the paths back. 15:02:02 Could someone who is more knowlegeable of the iscsi connector code in os-brick take a look at this? TBH I don't want to mess with os-brick code... 15:02:42 #link https://bugs.launchpad.net/os-brick/+bug/1944474 15:02:56 Fix proposed: 15:02:56 #link https://review.opendev.org/c/openstack/os-brick/+/810331 15:03:46 Ah - so this is already in the works - great to know 15:04:18 can we get this bug added to the commit message for the fix? 15:05:07 I love how on the ball you all are... 15:05:09 looks like that was sofia's -1 15:05:17 yep - just saw that 15:05:42 I'll let the customer know about this patch and they can try it out 15:05:53 simondodsley: if you could keep an eye on that patch, and propose backports (you can do it from gerrit) that would be helpful 15:06:07 will do 15:06:33 we will probably be releasing from ussuri soon after the xena release 15:06:51 because i think ussuri goes into EM mode like 2 weeks afterwards 15:07:06 so we'll be doing a final ussuri release of everything 15:07:46 so simondodsley your other task is to make sure we don't propose a ussuri os-brick release before that fix is included 15:07:58 on it 15:08:07 I will also keep a note 15:08:16 cool, ty whoami-rajat 15:08:19 thanks 15:08:26 thanks everyone for this rapid response 15:08:57 cinder has 2 modes: rapid and glacial 15:09:02 :) 15:09:17 Cool, can I move on with the rest of the bugs? 15:09:54 #topic [RBD] Volume can't be removed if other volumes were created from a snapshot of this volume and were not removed 15:10:01 #link https://review.opendev.org/c/openstack/cinder/+/810402 15:10:07 -I'm confused about this rosmaita. Do I need to write up a release note as a known issue ? Because the target is RC2 instead of RC1. 15:10:25 looking 15:11:15 no, your current release note is fine 15:11:24 i think we can target this one to rc-2 15:12:05 i just did ... are there any objections to considering this a release-critical bug? 15:13:01 ok, hearing no objections, i will add it to the etherpad 15:13:08 thanks! 15:13:23 #topic Announcements 15:13:27 Cinder has six new bugs reported this period. 15:13:33 Three of them are driver related so I won't mention them today. 15:13:43 One it's the  cinder-manage db version fails with: AttributeError: 'Engine' object has no attribute 'get_main_option' that we've already mentioned in the main meeting today.#link http://lists.openstack.org/pipermail/openstack-discuss/2021-September/024994.html 15:13:53 oops 15:13:54 #link http://lists.openstack.org/pipermail/openstack-discuss/2021-September/024994.html 15:14:42 The next one is from last week, but I think I should mentioned it here 15:14:49 #topic bug_1 Create a separate [quota] section for quota parameters 15:14:56 #link https://bugs.launchpad.net/cinder/+bug/1934025 15:15:09 Summary: Takashi Kajinami proposed to migrate all quota parameters to a new [quota] section so that users and operators can easily understand these options are for quota features. 15:15:10 Manila and Nova have all quota parameters in the [quota] section. 15:15:25 Patch proposed to master: 15:15:35 #link https://review.opendev.org/c/openstack/cinder/+/772181 15:15:44 should mark as Wishlist importance 15:16:07 moved to Wishlist 15:16:19 we should put this on the PTG etherpad 15:16:31 would be good to distinguish quotas from limits 15:16:42 and get ready for the unified-quotas stuff 15:16:49 sounds good 15:17:27 OK, that's all I have for today's meeting 15:17:40 i wrote https://bugs.launchpad.net/cinder/+bug/1944577 for the manage+encryption stuff 15:17:40 #topic Open Discussion 15:17:49 definitely needs some work, i'll go chase fixing things up there 15:17:57 #link https://bugs.launchpad.net/cinder/+bug/1944577 15:18:06 we block unmanaging encrypted volumes, we probably need to block managing them too 15:18:15 er, managing to encrypted types, that is 15:18:27 ok, that makes sense 15:18:35 because doing that would require a user being able to associate a key w/ the volume 15:18:38 and we don't have a way to do that 15:18:46 so you just end up with a volume that isn't put together right 15:18:48 good point! 15:19:04 could make that work eventually, but for now we need to at least make it not dangerous 15:19:31 ok, i will write up a release note saying don't do this and ping you for reveiw 15:19:39 Merged openstack/cinder stable/ussuri: Allow removing NFS snapshots in error status https://review.opendev.org/c/openstack/cinder/+/809669 15:19:41 ok 15:20:17 Sofia Enriquez proposed openstack/cinder stable/train: Allow removing NFS snapshots in error status https://review.opendev.org/c/openstack/cinder/+/810406 15:20:50 eharney: is it easy block managing an encrypted volume? 15:20:57 thanks eharney for the new bug, it's definitely more detailed than the original 15:21:01 might be better to just do that now and include it in RC-2 15:21:15 rosmaita: should be, i just need to, as usual, determine which place in the code to do it 15:21:47 can just do it up front in the API on the request, i guess 15:21:57 that's what i was thinking 15:23:25 ok, let's touch base on monday and we can see where we are on this 15:23:55 sound good 15:24:14 sounds* 15:24:44 any other bug for the open discussion? 15:26:04 OK 15:26:15 ending meeting, thanks !! 15:26:16 #endmeeting