15:03:46 <enriquetaso> #startmeeting cinder_bs 15:03:46 <opendevmeet> Meeting started Wed Aug 24 15:03:46 2022 UTC and is due to finish in 60 minutes. The chair is enriquetaso. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:03:46 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:03:46 <opendevmeet> The meeting name has been set to 'cinder_bs' 15:03:56 <enriquetaso> Full list of bugs 15:03:56 <enriquetaso> #link https://lists.openstack.org/pipermail/openstack-discuss/2022-August/030084.html 15:04:07 <enriquetaso> We have three bugs to discuss today 15:04:18 <rosmaita> o/ 15:04:40 <enriquetaso> #topic I can't attach a nfs+luksv1 volume because _attach_encryptor only supports iscsi protocol 15:04:42 <enriquetaso> hi rosmaita 15:04:48 <enriquetaso> #link https://bugs.launchpad.net/nova/+bug/1987311 15:04:53 <enriquetaso> I've opened a bug against Nova regarding NFS encryption. I'll try to reach the nova team this week. 15:05:21 <rosmaita> nice bug report there 15:05:59 <enriquetaso> ha, yep, thanks eharney for debugging this with me 15:07:09 <enriquetaso> moving on 15:07:16 <enriquetaso> #topic [docs] Cannot start VM from disk with burst QoS. 15:07:23 <enriquetaso> #link https://bugs.launchpad.net/cinder/+bug/1982945 15:07:47 <enriquetaso> This bug was originally a nova bug but was moved this week i think. I marked this as a documentation bug but I could be wrong. 15:09:49 <enriquetaso> Last bug: 15:09:51 <enriquetaso> #topic RBD support extends for in-use volume 15:09:59 <enriquetaso> #link https://bugs.launchpad.net/cinder/+bug/1900406 15:10:09 <enriquetaso> This is a last-month bug that I forgot to mention. 15:10:09 <enriquetaso> The configuration rbd_volume_local_attach = False but looks like Cinder ignores it and allows rbd to extend the volume anyway. 15:10:30 <enriquetaso> I marked it as low importance because i think it may be an easy fix 15:10:57 <eharney> rbd_volume_local_attach is a nova option... i guess what this is saying is that when using that, extend should not be allowed, but it is? 15:11:17 <opendevreview> Gorka Eguileor proposed openstack/cinder master: Expose group in Volume OVO https://review.opendev.org/c/openstack/cinder/+/779578 15:11:56 <rosmaita> enriquetaso: btw, i agree with your idea to start https://bugs.launchpad.net/cinder/+bug/1982945 as a doc bug 15:12:06 <eharney> what would the fix be? 15:13:31 <enriquetaso> eharney, yes, that's what I understood. I haven't check in the code but maybe it's a simple CONF check while extending 15:13:57 <eharney> well, i think this bug at least needs a better description of what's happening and a proposal, for it to be tagged as low-hanging-fruit 15:14:06 <geguileo> that would be a nova issue, right? 15:14:09 <eharney> it's barely clear to me what is happening there, much less a new-comer 15:14:10 <geguileo> os-brick doesn't know about that 15:14:12 <eharney> geguileo: probably 15:14:27 <geguileo> if you tell os-brick to extend, then it will try to do so 15:14:28 <enriquetaso> i'll remove the low-hanging-fruit :P my bad 15:15:09 <geguileo> enriquetaso: oh, it should be easy 15:15:13 <eharney> the bug also doesn't state what happens 15:15:17 <eharney> or what is supposed to happen 15:15:19 <eharney> well, ok 15:15:44 <enriquetaso> maybe i should link nova team to the bug as well since rbd_volume_local_attach is a nova option 15:16:58 <enriquetaso> so, we need more description on the bug, sadly it's almost one year-old but, hope the reporter reply 15:17:29 <geguileo> I would just ask in the bug what they mean 15:17:40 <geguileo> because I have read it now, and I don't even understand what they are talking about 15:17:56 <geguileo> Nova seems to do the right thing on the driver side and doesn't call os-brick 15:18:20 <geguileo> so it just calls cinder to extend, and the librbd on the hypervisor will detect the new size and it should just work 15:18:54 <geguileo> this sentence is weird: rbd and libvirt will extend rbd image. rbd should not extend rbd image. 15:21:42 <opendevreview> Alexander Deiter proposed openstack/cinder master: Fix Infinidat driver to backup attached volume https://review.opendev.org/c/openstack/cinder/+/851640 15:21:55 <enriquetaso> OK 15:22:11 <enriquetaso> so the bug is less clear than I thought 15:22:44 <enriquetaso> I'll ask for more information and what they mean 15:23:10 <eharney> it might be more productive to not worry about the report too much, and just go assess what kind of test coverage we have for rbd extend, both local attach and not 15:23:20 <eharney> if we are interested in chasing this 15:23:51 <enriquetaso> i can do that as well 15:23:59 <eharney> we should be able to find out whether it works or not without the reporter's involvement 15:25:03 <sfv880> Hello! if possible, I would like to ask you to review the fixes for the Infinidat driver: 15:25:08 <sfv880> Fix Infinidat driver to return all iSCSI portals | https://review.opendev.org/c/openstack/cinder/+/849022 15:25:09 <sfv880> Fix Infinidat driver to use TLS/SSL communication | https://review.opendev.org/c/openstack/cinder/+/849846 15:25:09 <sfv880> Fix Infinidat driver generic volume migration | https://review.opendev.org/c/openstack/cinder/+/850320 15:25:10 <sfv880> Fix Infinidat driver multi-attach feature | https://review.opendev.org/c/openstack/cinder/+/850321 15:25:13 <sfv880> Fix Infinidat driver to backup attached volume | https://review.opendev.org/c/openstack/cinder/+/851640 15:25:17 <sfv880> Thank you very much! 15:25:41 <enriquetaso> lol 15:26:23 <enriquetaso> eharney, makes sense, i'll check that and add some comments on the bug report with this information 15:26:34 <enriquetaso> OK, last 4 minutes 15:26:40 <enriquetaso> #topic open discussion 15:26:52 <enriquetaso> do we have any bug to discuss ? 15:29:21 <rosmaita> nothing from me 15:29:23 <enriquetaso> eharney, {0} tempest.api.volume.test_volumes_extend.VolumesExtendAttachedTest.test_extend_attached_volume [28.431093s] ... ok looks like it's fine 15:29:48 <enriquetaso> anyway i need to see more carefully 15:29:53 <enriquetaso> rosmaita, OK :) 15:30:42 <eharney> that doesn't cover local attach vs not 15:30:53 <eharney> also need to see how much behavior that actually checks 15:31:06 <eharney> is it supposed to extend inside the instance, etc 15:31:32 <opendevreview> Takashi Natsume proposed openstack/cinder master: Fix a deprecation warning about regex https://review.opendev.org/c/openstack/cinder/+/854428 15:31:41 <enriquetaso> eharney, thanks 15:31:50 <enriquetaso> OK, thanks everyone! 15:31:53 <enriquetaso> #endmeeting