| *** dviroel|out is now known as dviroel | 00:20 | |
| *** dviroel is now known as dviroel|out | 00:32 | |
| *** dviroel|out is now known as dviroel | 11:23 | |
| whoami-rajat | #startmeeting cinder | 14:00 |
|---|---|---|
| opendevmeet | Meeting started Wed Apr 20 14:00:22 2022 UTC and is due to finish in 60 minutes. The chair is whoami-rajat. Information about MeetBot at http://wiki.debian.org/MeetBot. | 14:00 |
| opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 14:00 |
| opendevmeet | The meeting name has been set to 'cinder' | 14:00 |
| whoami-rajat | #topic roll call | 14:00 |
| simondodsley | o/ | 14:00 |
| tosky | o/ | 14:00 |
| felipe_rodrigues | o/ | 14:00 |
| eharney | hi | 14:01 |
| jbernard | o/ | 14:01 |
| jungleboyj | o/ | 14:01 |
| nahimsouza[m] | o/ | 14:01 |
| caiquemello[m] | o/ | 14:01 |
| whoami-rajat | #link https://etherpad.openstack.org/p/cinder-zed-meetings | 14:01 |
| rosmaita | o/ | 14:02 |
| lucasmoliveira059 | o/ | 14:02 |
| whoami-rajat | we've a good turnout | 14:03 |
| whoami-rajat | so let's start | 14:03 |
| whoami-rajat | #topic announcements | 14:03 |
| *** dasm|off is now known as dasm | 14:03 | |
| whoami-rajat | first, Cinder mid cycle dates | 14:03 |
| enriquetaso | hi | 14:03 |
| whoami-rajat | #link https://de46dad8d0940c6cc81e-27618e467ab9b983fdfb7a89cc8b48ca.ssl.cf2.rackcdn.com/837494/3/check/openstack-tox-docs/e70ce68/docs/zed/schedule.html | 14:03 |
| e0ne_ | hi | 14:03 |
| whoami-rajat | I've reproposed the Zed schedule for cinder to include midcycle 1 (at R-18) and midcycle 2 (at R-9) | 14:04 |
| whoami-rajat | as a placeholder, I've marked the dates, 1st June (Wednesday) and 3rd August (Wednesday) 1400-1600 UTC as the proposed dates | 14:04 |
| fabiooliveira | hi | 14:04 |
| whoami-rajat | would like to know if there are any conflicts of the above dates with other events or country specific holiday | 14:04 |
| whoami-rajat | also would like to know if the date/time is not comfortable and I should do a poll instead | 14:05 |
| rosmaita | i won't be available for Aug 3, but that's just me | 14:05 |
| whoami-rajat | I've proposed the timing overlapping with the cinder meeting (+ one extra hour) as it works for everyone | 14:05 |
| whoami-rajat | oh, the idea is to have the midcycle at R-9 so we can schedule any day of the week | 14:06 |
| rosmaita | i will just watch the video when i get back | 14:07 |
| simondodsley | same for me | 14:07 |
| whoami-rajat | sure, or we can do a poll as well, if others also feel more comfortable on another day | 14:08 |
| whoami-rajat | ok, i guess we're going to do a poll for R-9 then | 14:08 |
| whoami-rajat | any conflicts for R-18? | 14:08 |
| rosmaita | no conflict for me | 14:09 |
| simondodsley | no | 14:10 |
| sfernand | both dates works fine for the netapp folks | 14:10 |
| whoami-rajat | cool, I will change the release schedule to exclude R-9 for now and prepare a poll | 14:10 |
| whoami-rajat | #action whoami-rajat to prepare a poll for R-9 midcycle | 14:10 |
| whoami-rajat | great | 14:10 |
| whoami-rajat | moving on | 14:10 |
| whoami-rajat | second, Monkeypatching FIPS | 14:11 |
| whoami-rajat | #link http://lists.openstack.org/pipermail/openstack-discuss/2022-April/028250.html | 14:11 |
| whoami-rajat | Ade lee sent out a mail regarding monkeypatching paramiko as paramiko uses md5 which makes it FIPS non complaint | 14:11 |
| whoami-rajat | (the mail also mentions cinder) | 14:11 |
| whoami-rajat | I still haven't looked at the cinder code for any occurrence but just for awareness, we need to wait for the fix to land in paramiko or adopt any other lib, until then projects are planning to monkeypatch it | 14:11 |
| e0ne_ | #link https://github.com/openstack/cinder/search?q=paramiko | 14:11 |
| eharney | we have some drivers that use paramiko | 14:12 |
| whoami-rajat | thanks e0ne_ , so we are using paramiko | 14:12 |
| e0ne_ | whoami-rajat: np | 14:12 |
| whoami-rajat | so someone (probably me) will reply to the mail about cinder using paramiko | 14:13 |
| eharney | AFAIK it is only used for a few drivers to talk to the backend, so one option is that those drivers won't work in FIPS mode until it's fixed | 14:13 |
| whoami-rajat | ok | 14:13 |
| simondodsley | some of those backends may not be FIPS compliant anyway so wouldn't be used in a FIPS environment | 14:13 |
| whoami-rajat | i guess then it shouldn't be much of an issue since the FIPS jobs we will be running are mostly lvm, ceph, nfs related | 14:14 |
| e0ne_ | should we add t FIPS support requirement for new drivers? | 14:14 |
| whoami-rajat | and third party CI still doesn't run on centOS | 14:14 |
| whoami-rajat | so we need to wait for FIPS support in ubuntu to make it a requirement for drivers | 14:15 |
| eharney | right | 14:15 |
| rosmaita | we can strongly suggest that drivers keep it in mind | 14:15 |
| whoami-rajat | ok, that's a relief | 14:15 |
| simondodsley | something in the new drivers documentation? | 14:16 |
| rosmaita | or the review checklist | 14:16 |
| rosmaita | or both | 14:16 |
| rosmaita | :) | 14:16 |
| rosmaita | basically, don't use md5 and only use well-respected crypto libraries | 14:16 |
| whoami-rajat | both sounds good, one cinder wide and one driver specific | 14:17 |
| rosmaita | and don't implement your own crypto under any circumstances | 14:17 |
| eharney | yeah, a review checklist for such things seems like a good idea | 14:17 |
| e0ne_ | rosmaita: :) | 14:17 |
| rosmaita | whoami-rajat: you can give me an action item on this | 14:17 |
| whoami-rajat | rosmaita, regarding documentation or replying to the email thread? | 14:18 |
| whoami-rajat | or both :) | 14:18 |
| rosmaita | docs | 14:18 |
| whoami-rajat | ok | 14:18 |
| whoami-rajat | i will reply to the email with our discussion | 14:18 |
| whoami-rajat | #action rosmaita to include FIPS related documentation for new drivers | 14:18 |
| whoami-rajat | #action whoami-rajat to respond to the mail thread with our discussion | 14:19 |
| rosmaita | my personal opinion is that we should use something other than paramiko in ssh_utils | 14:19 |
| whoami-rajat | maybe Ade has some idea about good alternatives, can discuss with him | 14:20 |
| whoami-rajat | that's all for my announcements, does anyone have any other announcement? | 14:20 |
| whoami-rajat | guess not, so let's move to topics | 14:22 |
| whoami-rajat | #topic vote on NetApp proposal to backport ONTAP REST API support | 14:22 |
| whoami-rajat | Fernando suggested that we should do this in the video + IRC meeting (that is next week) so everyone can express their opinions about it and then we can vote | 14:22 |
| whoami-rajat | e0ne_, will you be available in next week's video + IRC meeting? | 14:22 |
| e0ne_ | whoami-rajat: I really hope so | 14:23 |
| jungleboyj | :-) | 14:23 |
| e0ne_ | if I won't join, my current vote is to don't allow such backports | 14:23 |
| geguileo | ok | 14:23 |
| sfernand | e0ne_: ok | 14:24 |
| whoami-rajat | ok, so my personal opinion is it would make sense to have a discussion on it before voting else we will end up upvoting or downvoting it without any concrete conclusion | 14:24 |
| eharney | yes, i think more discussion is needed | 14:24 |
| simondodsley | I can't make the next meeting either and my vote would be to not allow these backports - specify supported ONTAP versions that drivers support instead | 14:24 |
| rosmaita | right now, i'm with e0ne_ on this as well | 14:25 |
| whoami-rajat | ok, so let's discuss it next week and maybe we can drag the discussion till mid cycle to properly finalize it, till then the netapp team can work on their patch for master | 14:26 |
| sfernand | whoami-rajat: ++ | 14:27 |
| whoami-rajat | cool, anything else on this topic? | 14:27 |
| whoami-rajat | guess not, moving on then | 14:28 |
| whoami-rajat | #topic release notes with driver patches | 14:28 |
| whoami-rajat | rosmaita, that's you | 14:28 |
| rosmaita | basically, just what i said on the agenda | 14:28 |
| geguileo | rosmaita: copy/paste then };-) | 14:29 |
| rosmaita | (which i will paste in here as soon as i find it) | 14:29 |
| rosmaita | note to reviewers and driver maintainers: any driver fix that will be backported should have a release note | 14:29 |
| whoami-rajat | note to reviewers and driver maintainers: any driver fix that will be backported should have a release note | 14:29 |
| rosmaita | ok, we agree on this :) | 14:29 |
| whoami-rajat | +1 | 14:30 |
| geguileo | rosmaita: in general ANY driver fix should have a release note | 14:30 |
| geguileo | regardless of the backporting aspect | 14:30 |
| jungleboyj | ++ | 14:30 |
| simondodsley | +1 | 14:30 |
| whoami-rajat | usually allowed backports are mostly bugfixes so it's rare I've seen one without a releasenote | 14:30 |
| rosmaita | well, i was reviewing the other day, and this came up | 14:31 |
| geguileo | rosmaita: and we should also say: driver bug fixes MUST have a launchpad bug explaining what's fixing | 14:31 |
| rosmaita | so maybe this is more for reviewers to keep in mind | 14:31 |
| simondodsley | the huawei patches referenced later in this meeting are missing them | 14:31 |
| e0ne_ | geguileo: +1 | 14:31 |
| rosmaita | geguileo: and the commit message should make sense | 14:31 |
| rosmaita | and the bug report should also make sense | 14:32 |
| geguileo | rosmaita: oooooh, that's a good one ;-) | 14:32 |
| rosmaita | and good engineering practices in the code | 14:32 |
| whoami-rajat | so we already have an example | 14:32 |
| e0ne_ | rosmaita: and not include 'proprietary crypto algorithm for a driver' | 14:32 |
| geguileo | rosmaita: that's sometimes subjective ;-P | 14:32 |
| geguileo | e0ne_: +1 | 14:32 |
| geguileo | whoami-rajat: you were working on a reviewer doc patch, these points can be added there | 14:33 |
| e0ne_ | oops... I missed 'implementation' word in my previous message:( | 14:33 |
| whoami-rajat | geguileo, sure, these are some good ideas | 14:34 |
| rosmaita | ok, for now, reviewers, please keep the above in mind | 14:34 |
| e0ne_ | rosmaita: +1 | 14:34 |
| rosmaita | and if you see a driver patch with no release note, that's an easy -1 !!! | 14:35 |
| whoami-rajat | I'm happy rosmaita is providing lots of points for my review doc, probably rosmaita can co-author it as well! | 14:35 |
| rosmaita | whoami-rajat: you can have those points for free :) | 14:36 |
| whoami-rajat | :D | 14:36 |
| whoami-rajat | ok, so i think the point is pretty straightforward here | 14:36 |
| whoami-rajat | we can move on to the next topic then | 14:36 |
| whoami-rajat | #topic fixing some DB API function signature inconsistencies | 14:37 |
| whoami-rajat | rosmaita, that's you again | 14:37 |
| rosmaita | i will be real quick (for a change) | 14:37 |
| rosmaita | #link https://review.opendev.org/c/openstack/cinder/+/837542/3 | 14:37 |
| rosmaita | the question is, how far do we want to go in fixing this on this patch | 14:37 |
| rosmaita | please take a look, i have comments and stephen has responses | 14:38 |
| rosmaita | i have an opinion, but we should probably get some other people in on the action | 14:38 |
| rosmaita | so if you are interested in consistency between the db api interface and the sqlalchemy implementation, take a look | 14:38 |
| rosmaita | and if you don't care, something will happen | 14:38 |
| rosmaita | that's it | 14:38 |
| whoami-rajat | i can see 10 review comments, is there any one in specific or all are pointing to a general idea ? | 14:39 |
| rosmaita | there are maybe 3 general issues | 14:40 |
| geguileo | am I the only one that thinks we should completely drop that cinder/db/api.py file? | 14:41 |
| whoami-rajat | ok, so we can take a look at rosmaita's comments | 14:41 |
| eharney | geguileo: it's a reasonable idea | 14:41 |
| geguileo | or is there somebody that thinks we'll ever support something different than SQLA? | 14:41 |
| rosmaita | i suspect not | 14:42 |
| eharney | geguileo: Nova did it here: https://review.opendev.org/c/openstack/nova/+/799524 | 14:43 |
| whoami-rajat | that sounds like a good idea for cinder midcycle | 14:43 |
| geguileo | a pain to review, but makes sense to me | 14:43 |
| whoami-rajat | that's a long list of file changes... | 14:44 |
| rosmaita | then i guess it doesn't matter what happens in that patch | 14:44 |
| rosmaita | because we will be making changes later anyway | 14:44 |
| rosmaita | let's get rid of the legacy enginefacade first though | 14:44 |
| rosmaita | i.e., do not stop reviewing the current patches!!! | 14:45 |
| whoami-rajat | yep, that's a good reminder for a priority item ^^ | 14:45 |
| rosmaita | #link http://tiny.cc/cinder-legacyfacade | 14:45 |
| whoami-rajat | anyway, we can discuss the merging of db/api and db/sqlalchemy/api later | 14:46 |
| rosmaita | ^^ that's a handy review dashboard | 14:46 |
| geguileo | it kills me a bit when somebody reorders methods in a file... | 14:46 |
| rosmaita | yeah, but it is kind of nice to have the related ones together | 14:46 |
| rosmaita | saves my control key from getting worn out in emacs | 14:47 |
| whoami-rajat | so let's move on as we've other topics to cover as well (and less time) | 14:47 |
| whoami-rajat | #topic EM release for Victoria is April 27 | 14:47 |
| whoami-rajat | jbernard, that's you | 14:47 |
| jbernard | heya, last victoria release is next week | 14:48 |
| jbernard | we have 6 outstanding patches: https://review.opendev.org/q/branch:stable/victoria+project:openstack/cinder+status:open | 14:48 |
| whoami-rajat | so i forgot this in my announcement, jbernard has stepped up to take up stable release maintenance! | 14:48 |
| enriquetaso | \o/ | 14:48 |
| rosmaita | jbernard: ++ | 14:48 |
| simondodsley | congrats | 14:49 |
| eharney | excellent | 14:49 |
| rosmaita | i guess i should drop my -2 on https://review.opendev.org/c/openstack/cinder/+/822939 ? | 14:49 |
| whoami-rajat | jbernard++ thanks! | 14:49 |
| jungleboyj | \o/ | 14:49 |
| jungleboyj | Thank you jbernard ! | 14:49 |
| rosmaita | not sure the issue has been settled yet, but i will leave it up to other reviewers | 14:49 |
| jbernard | if im understanding the process correctly, I need some time for the release team to process the request, so we need the patches we care about merged by Monday (Apr 25) at the latest (i estimate) | 14:49 |
| whoami-rajat | so stable cores can take a look at the open patches as we're nearing the final victoria release (mostly have votes as of now) | 14:50 |
| simondodsley | rosmaita that one is abandoned anyway | 14:50 |
| rosmaita | simondodsley: ty! | 14:50 |
| eharney | rosmaita: i'll go fix up that whole thing after i fix master | 14:50 |
| rosmaita | ok, cool | 14:50 |
| whoami-rajat | rosmaita, eharney has abandoned it, so you can remove it when he restores? | 14:50 |
| rosmaita | sure | 14:51 |
| jbernard | ok, we're already down to 5, making progress ;) | 14:51 |
| rosmaita | jbernard: you are already killing it as release manager! | 14:51 |
| eharney | https://review.opendev.org/c/openstack/cinder/+/831188 is ineligible currently, so easy to skip that one | 14:51 |
| whoami-rajat | jbernard, that's right, we should do our release before 27th so release team can get time to make the transition easier for victoria to EM | 14:51 |
| whoami-rajat | eharney, yep, most probably it won't make to victoria in time | 14:52 |
| whoami-rajat | so we can skip patches that are currently open in master | 14:53 |
| whoami-rajat | or are far away from making it to victoria | 14:53 |
| whoami-rajat | unless people want that change in | 14:53 |
| whoami-rajat | and working towards getting it in | 14:53 |
| whoami-rajat | so, stable cores, take a look at the open patches in victoria | 14:54 |
| whoami-rajat | moving on | 14:54 |
| whoami-rajat | #topic Replace distutils in drivers | 14:54 |
| whoami-rajat | simondodsley, that's you | 14:54 |
| simondodsley | as mentioned in the agenda I have a patch up to replace driver usage of distutils (https://review.opendev.org/c/openstack/cinder/+/832130) to stop any future bugs, but some of the affected drivers don't have CIs - can we just merge this patch without those CI, or does anyone know who owns these CI systems? | 14:54 |
| eharney | unit tests should be sufficient for this change | 14:56 |
| whoami-rajat | i can see there are 6 drivers using it | 14:56 |
| eharney | CI results are not really interesting | 14:56 |
| whoami-rajat | if those driver maintainers can take a look, would be good | 14:56 |
| jbernard | the nature of the changes do not impact the driver logic itself, i would vote to merge personally | 14:56 |
| simondodsley | finmding the CI maintainers is the biggest issue | 14:56 |
| simondodsley | we need a list of current maintainers | 14:57 |
| rosmaita | i think a message to the ML and then we mark all those as unsupported (if they aren't already) | 14:57 |
| whoami-rajat | rosmaita, that sounds like a good idea | 14:57 |
| rosmaita | well, the 3rd party CIs are supposed to keep their info in the wiki up to date | 14:57 |
| simondodsley | but if there isn't even a CI... | 14:58 |
| simondodsley | eg I cannot find an IBM DS8k CI | 14:58 |
| rosmaita | i actually thought that driver had been removed | 14:58 |
| simondodsley | or a powerflex | 14:58 |
| whoami-rajat | this seems to be a concern with re-occurring changes that modify driver code | 14:58 |
| eharney | i'm not sure why it's a concern with this change | 14:59 |
| simondodsley | DS8k is still listed as supported as is powerflex | 14:59 |
| whoami-rajat | eharney, not specifically this change but in general driver maintainers not being active | 15:00 |
| whoami-rajat | i will send out a mail to the ML mentioning the drivers in question here, in the meantime if the team thinks this change is good with unit tests only then we can merge as well | 15:00 |
| simondodsley | this change just highlights the lack of visibility on 3rd party CIs | 15:00 |
| rosmaita | eharney: it's just a general erosion of standards | 15:00 |
| rosmaita | but eharney has a good point, we shouldn't let it hold up this patch, i guess | 15:01 |
| whoami-rajat | rosmaita, yep, as stated above, if people think the change is good with UTs then we can go ahead and merge | 15:01 |
| whoami-rajat | but i will send out the mail anyway | 15:01 |
| whoami-rajat | and we're out of time | 15:01 |
| whoami-rajat | last topic | 15:02 |
| whoami-rajat | #topic request for review of backports -- i think we already got reviews on them | 15:02 |
| whoami-rajat | to add releasenote | 15:02 |
| simondodsley | OK - I guess I will have to work out how to do a UT for this - BTW I hate writing in mock :) | 15:02 |
| whoami-rajat | so ganso you can make those changes and team can review again | 15:02 |
| whoami-rajat | simondodsley, i think eharney means the current UTs should validate your change (or maybe i understood it wrong) | 15:03 |
| simondodsley | and add a statement that this has been tested on the stable branch | 15:03 |
| whoami-rajat | let's continue discussion in cinder channel (after bugsquad meeting) | 15:03 |
| eharney | i don't know if the current ones do or not, they might, or coverage might need to be added | 15:03 |
| simondodsley | whoami-rajat - ah, ok | 15:03 |
| whoami-rajat | thanks everyone! | 15:03 |
| whoami-rajat | #endmeeting | 15:03 |
| opendevmeet | Meeting ended Wed Apr 20 15:03:52 2022 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 15:03 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/cinder/2022/cinder.2022-04-20-14.00.html | 15:03 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/cinder/2022/cinder.2022-04-20-14.00.txt | 15:03 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/cinder/2022/cinder.2022-04-20-14.00.log.html | 15:03 |
| jungleboyj | Thank you! | 15:04 |
| *** dviroel is now known as dviroel|lunch | 15:31 | |
| *** dviroel|lunch is now known as dviroel | 16:31 | |
| *** dasm is now known as dasm|off | 20:53 | |
| *** dviroel is now known as dviroel|out | 22:27 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!