14:01:00 <rosmaita> #startmeeting cinder 14:01:00 <opendevmeet> Meeting started Wed Mar 9 14:01:00 2022 UTC and is due to finish in 60 minutes. The chair is rosmaita. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:01:00 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:01:00 <opendevmeet> The meeting name has been set to 'cinder' 14:01:07 <rosmaita> #topic roll call 14:01:15 <jungleboyj> o/ 14:01:21 <whoami-rajat> Hi 14:01:21 <eharney> howdy 14:01:21 <simondodsley> hi 14:01:30 <tosky> hi 14:01:40 <fabiooliveira> o/ 14:01:45 <bingyanh> o/ 14:01:58 <enriquetaso> hi 14:02:07 <geguileo> hi! o/ 14:02:08 <e0ne> hi 14:02:29 <kkaras_intel> hello 14:02:43 <rosmaita> hello everyone 14:02:59 <rosmaita> #link https://etherpad.opendev.org/p/cinder-yoga-meetings 14:03:11 <rosmaita> let's get started! 14:03:17 <yuval> hey 14:03:38 <rosmaita> daylight savings time in USA starts 13 March, so this meeting may be 1 hour later for you next week depending on where you are 14:03:52 <rosmaita> so, if you show up and no one is here, come back in an hour 14:04:00 <rosmaita> you will not have missed anything 14:04:06 <jungleboyj> :-) 14:04:19 <rosmaita> the first Yoga Release Candidate (RC-1) must be made this week 14:04:31 <rosmaita> at that time, the stable/yoga branch will be cut 14:04:38 <rosmaita> master becomes the Zed development branch 14:04:58 <rosmaita> any changes must be made in master and then backported to stable/yoga to be in the release 14:06:23 <rosmaita> so, let's prioritize 14:06:32 <rosmaita> https://etherpad.opendev.org/p/cinder-yoga-release-critical-changes 14:07:06 <rosmaita> first, for new drivers coming out in Yoga, if you have a bugfix you discovered, it would be good to get those into the Yoga release 14:07:35 <rosmaita> btw, these don't require a release note because the driver doesn't exist yet 14:08:09 <rosmaita> the "transition-to-alembic" patches continue to be important 14:09:14 <rosmaita> and there aren't too many left 14:09:37 <whoami-rajat> are the merged alembic changes stable enough that we can move the rest for Zed? or they need to be in yoga only 14:10:20 <rosmaita> i need to talk to stephenfin about that 14:11:11 <rosmaita> the other things we should take a look at are the "known issues" in the xena release 14:11:24 <rosmaita> https://docs.openstack.org/releasenotes/cinder/xena.html#known-issues 14:13:45 <geguileo> rosmaita: there are a couple of issues on the nvmet target 14:14:07 <geguileo> known by me, I mean 14:14:08 <rosmaita> you mean left over from xena, or new ones for yoga 14:14:25 <enriquetaso> I'm not currently working on the encryption problem but it's on my TODO list. If anyone would like to work with me on that to speed things up, let me know.. :P 14:15:06 <geguileo> rosmaita: as far as I know we've had them since the begining of the nvmet target 14:15:10 <rosmaita> enriquetaso: that's Bug #1888680 ? https://bugs.launchpad.net/cinder/+bug/1888680 14:15:12 <geguileo> we just didn't know 14:15:52 <enriquetaso> yep, the ones mentioned on Known Issues 14:16:34 <rosmaita> geguileo: ok, let's talk later and work out the text about nvmet known issues 14:17:06 <rosmaita> i believe jon bernard is workign on the cgroups v1 issue 14:18:13 <rosmaita> Bug #1937084 (race condition between delete attachment and delete volume) ... pretty sure geguileo fixed that and it's been backported 14:18:18 <rosmaita> \o/ 14:18:26 <geguileo> yup 14:19:15 <rosmaita> https://bugs.launchpad.net/cinder/+bug/1895035 ("Ceph backup driver could not restore a volume other than ceph") doesn't have a patch associated with it in launchpad 14:19:22 <rosmaita> anyone know the status of that one? 14:20:57 <enriquetaso> it's ready for upstream reviews 14:21:13 <rosmaita> cool, do you have a review link? 14:21:21 <enriquetaso> #link https://review.opendev.org/c/openstack/cinder/+/750782/ 14:22:22 <rosmaita> ty 14:22:26 <enriquetaso> \o/ 14:23:26 <rosmaita> ok, last thing was a warning to operators not to manage really old qcow2 images into cinder from an nfs backend 14:23:50 <eharney> bug 1937084 is mentioned as something we are working on in the release notes -- but it's been fixed 14:24:25 <rosmaita> yes, thanks to geguileo i think 14:25:40 <rosmaita> as far as the operator warning goes (about managing v2 qcow2s into cinder from nfs backend), we can repeat in Y and think about blocking this in code in Z 14:27:05 <whoami-rajat> sounds good, operators need to make this transition eventually 14:28:06 <rosmaita> any questions or observations before we move on? 14:29:20 <rosmaita> ok, only one agenda item, so maybe this will be a quick meeting 14:29:34 <rosmaita> #topic question about translation pattern 14:30:04 <rosmaita> well, you can look here to see what I'm talking about: 14:30:13 <rosmaita> #link https://review.opendev.org/c/openstack/cinder/+/819778/8/cinder/volume/drivers/huawei/huawei_conf.py#436 14:30:16 <jungleboyj> rosmaita: You are going to jinx it. :-) 14:30:44 <rosmaita> yeah, somehow a discussion of logging is always contentious 14:31:50 <eharney> i feel like i'm missing something about how this is supposed to work 14:32:10 <jungleboyj> :-) 14:32:12 <eharney> if you want the log messages to be in english, would you not just have the service running with an en_US locale or so? 14:32:37 <rosmaita> i think if you do that, the exception messages that show up to the user are english too 14:32:55 <eharney> are they? really not sure 14:33:03 <AndroUser> Yuval is still here 14:33:11 <rosmaita> the idea was that operators want the logs in english to maximize the chance they'll get a hit when googling the message 14:33:20 <rosmaita> well, tbh, i am not sure either 14:33:34 <geguileo> I think everything uses the same language 14:33:38 <geguileo> (but could be wrong) 14:34:03 <eharney> i guess i was hoping that API requests were requesting a language via http headers 14:34:50 <eharney> at any rate, it would seem like a good idea to understand the requirements before we ask everyone to duplicate all of their error messages in the code, because that's just... uhy :/ 14:35:04 <rosmaita> i figured that since they made a big deal about it in the oslo.log docs, this must be what you have to do 14:35:12 <eharney> maybe so 14:35:19 <rosmaita> but also maybe not 14:35:32 <eharney> which, as was pointed out in the review, surely means that most of them in cinder aren't done right 14:36:05 <rosmaita> yep 14:36:53 <rosmaita> ok, seems like the thing to do short-term is continue our current practice 14:36:58 <rosmaita> don't translate log messages 14:37:08 <rosmaita> do translate messages for exceptions 14:37:12 <eharney> https://docs.openstack.org/oslo.i18n/latest/user/guidelines.html 14:37:15 <eharney> says "_() is preferred for any user facing message, even if it is also going to a log file." 14:37:21 <jungleboyj> rosmaita: That was my latest understanding. 14:37:41 <jungleboyj> But then we have the line that eharney highlights. :-) 14:38:10 <rosmaita> keep reading, though: "However, now that log messages are no longer translated, it is unfortunately necessary to use two separate strings: a plain one for the log message, and a translatable one for the exception." 14:38:32 <eharney> dramatic twist! yeah... 14:38:58 <jungleboyj> *duh duh ... dah!* 14:39:11 <jungleboyj> So, that would appear to be the answer then. 14:39:24 <rosmaita> i wonder if anyone at all is actually doing this (where 'this'==duplicate strings) 14:39:41 <rosmaita> i bet everyone stops reading at the line eharney pointed out 14:39:47 <jungleboyj> Honestly, I don't think I have seen that in the code. 14:39:55 <jungleboyj> Yep. 14:40:02 <rosmaita> i don't think i've seen it anywhere in cinder 14:40:04 <eharney> well, most people look for info, and find it, and don't assume that the next paragraph is going to tell you the opposite thing :) 14:40:06 <geguileo> but is someone translating exceptions? 14:40:18 <geguileo> if nobody is translating anything, what's the point? 14:40:24 <eharney> also true 14:40:34 <rosmaita> well, we do get occasional .po file updates 14:41:57 <rosmaita> ok, let's take eharney's sentence "_() is preferred for any user facing message, even if it is also going to a log file. This ensures that the translated version of the message will be available to the user." as the current cinder approach 14:42:38 <rosmaita> we can ask the TC about this at the PTG 14:43:00 <bingyanh> ok 14:43:04 <eharney> it looks like we maybe just fully removed a few languages this year from cinder's translations. wonder why 14:43:38 <eharney> https://review.opendev.org/c/openstack/cinder/+/809944 14:44:53 <jungleboyj> Interesting. 14:44:54 <geguileo> Spanish survived!!! 14:45:09 <jungleboyj> :-) 14:45:10 <enriquetaso> \o/ 14:45:14 <enriquetaso> ha 14:45:21 <jungleboyj> Survivors. 14:45:33 <jungleboyj> French and Italian ... gone. 14:45:40 <AndroUser> Hola 14:45:47 <geguileo> lol 14:45:53 <jungleboyj> :-) 14:46:02 <geguileo> _('Hello') 14:46:12 * fabiooliveira hahaha 14:46:21 <rosmaita> i have no idea, the instructions to PTLs are to merge those patches right away as long as they don't break any tests 14:46:31 <jungleboyj> Exception: Hola 14:46:44 <jungleboyj> Hola Mundo! 14:46:49 <enriquetaso> :P 14:47:12 * jungleboyj stops before he embarrasses himself 14:47:32 <tosky> ups 14:47:49 <tosky> maybe a threshold issue? 14:48:20 <tosky> I guess I will have to take a look 14:48:58 <rosmaita> tosky: how do you mean? they didn't translast at least x% of the messages, and thus were dropped? somehting like that? 14:49:31 <jungleboyj> Or lack of translation volunteers. 14:50:21 <rosmaita> ok, so we have discussed this and reached a conclusion ... namely, continue to do what we have been doing 14:50:23 <rosmaita> thanks! 14:50:29 <rosmaita> #topic open discussion 14:50:47 <jungleboyj> ++ 14:53:22 <eharney> https://review.opendev.org/c/openstack/cinder/+/800704 is a patch that just removes a bunch of dead code if someone wants an easy review 14:53:29 <eharney> it was approved a while ago but got stuck in a merge conflict or something 14:53:44 <jungleboyj> where is Sean? He loves dead code. 14:55:05 <eharney> ok 14:55:31 <enriquetaso> do we have a mypy patch for this week? 14:56:05 <eharney> well https://review.opendev.org/c/openstack/cinder/+/774288 was the mypy patch from a few weeks ago, it was approved but didn't merge because nobody has reviewed its parent patch yet 14:56:49 <enriquetaso> oops 14:57:03 <rosmaita> just realized my comment on the parent is still in draft 14:57:15 <rosmaita> (it was a positive comment) 14:57:18 <enriquetaso> #link https://review.opendev.org/c/openstack/cinder/+/818830 14:57:18 <eharney> there are more of them out there under the "mypy" topic 14:57:25 <enriquetaso> I'll review the parent 15:00:49 <rosmaita> ok, that's all we have time for ... 15:00:52 <rosmaita> #endmeeting