21:00:56 <timburke> #startmeeting swift
21:00:56 <opendevmeet> Meeting started Wed May 25 21:00:56 2022 UTC and is due to finish in 60 minutes.  The chair is timburke. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:00:56 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:00:56 <opendevmeet> The meeting name has been set to 'swift'
21:01:04 <timburke> who's here for the swift meeting?
21:01:10 <ade_lee> o/
21:01:13 <mattoliver> o/
21:02:49 <timburke> as usual, the agenda's at https://wiki.openstack.org/wiki/Meetings/Swift
21:03:02 <timburke> and as usual, i haven't updated it like i should have ;-)
21:03:11 <timburke> first up
21:03:16 <timburke> #topic gate failures
21:03:30 <timburke> our gate's busted!
21:03:38 <timburke> but https://review.opendev.org/c/openstack/swift/+/842817 should fix it
21:04:45 <timburke> tl;dr is that with openstack dropping py36 and py37 support for zed, we've got to maintain our own constraints file for the old versions again
21:04:51 <mattoliver> oh thanks for dinning into hat
21:04:55 <mattoliver> *that
21:05:17 <timburke> see also https://governance.openstack.org/tc/reference/runtimes/zed.html
21:05:18 <mattoliver> *digging
21:05:27 <mattoliver> wow, you can tell it's early here, I can't even type :P
21:06:43 <timburke> and http://lists.openstack.org/pipermail/openstack-discuss/2022-May/028611.html
21:06:57 <mattoliver> ok and centos stream 9 for zed, interesting
21:07:06 <timburke> now i just need to get the fix merged ;-)
21:08:02 <timburke> yeah, i started looking at getting a centos stream 9 job running somewhere... pyeclib or liberasurecode, maybe? but i don't think i actually got it passing
21:08:54 <timburke> ah! sure enough, libec: https://review.opendev.org/c/openstack/liberasurecode/+/820969
21:09:33 <timburke> last run was a few months ago, though, and i don't remember what the failure mode was
21:09:57 <ade_lee> timburke, its definitely worth re-running
21:09:58 <mattoliver> yeah, just attempted to click on the output.
21:10:27 <timburke> anyway, that's all i had about the gate: if your patches keep failing, don't freak out just yet ;-)
21:10:31 <timburke> next up
21:10:40 <timburke> #topic sha1 deprecation
21:11:32 <mattoliver> cschwede-: has +2'ed the first in the chain
21:11:39 <timburke> we've got a bunch of patches we ought to land! ade_lee even managed to get cschwede to do some reviews, so it *must* be important ;-)
21:11:48 <mattoliver> lol
21:12:03 <ade_lee> and you got me to join this meeting too :)
21:12:15 <timburke> it's like whoa
21:12:50 <ade_lee> so yeah - pretty please, this is the patch thats holding up a bunch of fips jobs across projects
21:12:55 <ade_lee> (patches)
21:13:07 <timburke> do we know what the status is on the tempest dependency? https://review.opendev.org/c/openstack/tempest/+/832771
21:13:13 <mattoliver> 525771 also has a +2 in the second last patchset from pete.. so maybe we just land
21:14:03 <ade_lee> huh - I actually didn't realize we had a tempest dependency ..
21:15:07 <timburke> the tempurl remove-sha1-from-defaults can't pass with tempest master
21:15:29 <ade_lee> so -- we need to fix a few tests ..
21:15:41 <ade_lee> or a few more tests ..
21:16:21 <ade_lee> although I guess that things pass right now with the tests that were fixed?
21:17:34 <ade_lee> timburke, I haven't been following closely - but is there an issue in fixing the formpost tests?
21:19:00 <timburke> in tempest? yeah, we've got a bit of a chicken-and-egg problem since swift currently *only* supports sha1 for formpost. can't move tempest to using sha256 for those until swift actually supports it
21:19:59 <ade_lee> but mattoliver subsequent patch adds support for sha256 , right?
21:20:20 <ade_lee> https://review.opendev.org/c/openstack/swift/+/838434
21:20:56 <timburke> yup. and sha1 is still an allowed-by-default digest, which is as it should be
21:22:31 <timburke> there could absolutely be another patch written against tempest to switch the formpost tests to sha256. can't be part of https://review.opendev.org/c/openstack/tempest/+/832771, though, since it'll depend on mattoliver's patch, which eventually depends on 832771
21:22:43 <ade_lee> timburke, ok -- so if I understand this -- to get the tempest patch in -- all we need to do is fix test_object_temp_url_negative.py as well.
21:22:57 <ade_lee> and yeah we defer the formpost stuff
21:23:04 <ade_lee> to a different patch
21:23:46 <timburke> i think so? seems right to me. unfortunately, i don't think i have much leverage on getting tempest code merged, though
21:23:49 <ade_lee> then we need a second patch which fixes the tempest tests for formpost , which depends on mattoliver formpost patch?
21:24:08 <ade_lee> ack - I can talk to martin
21:24:33 <ade_lee> I just didn't realize there was still a tempest issue
21:24:42 <timburke> cool, thanks! and i'll work on getting my gate fixed and patches +A'ed
21:25:41 <timburke> #topic ring v2
21:25:41 <ade_lee> timburke, for the formpost patches, presumably those can be merged without changes to tempest
21:25:52 <timburke> #undo
21:25:52 <opendevmeet> Removing item from minutes: #topic ring v2
21:26:08 <ade_lee> and then we can do the tempest change to actually make sure its tested ..
21:26:16 <timburke> ade_lee, well, except they depend on the tempurl patch (at least, as things currently stand)
21:26:37 <ade_lee> well yeah - we're asuming that goes in
21:26:41 <ade_lee> assuming
21:27:18 <ade_lee> for the formtest changes -- I see there is something about optionally specifying the digest
21:27:18 <timburke> and that going in is dependent upon the tempest change to tempurl tests going in
21:28:07 <ade_lee> does that mean that when we set up the tempest tests for formpost +  sha256, we'll have to add some config?
21:28:57 <timburke> don't think so, but i haven't looked closely at mattoliver's patch since it got broken up. it's on my list for this week
21:29:35 <timburke> should be able to just swap sha1 to sha256, similar to the tempurl tempest patch
21:29:47 <mattoliver> if you use sha256 it'll detect that
21:29:52 <mattoliver> no extra config
21:30:00 <ade_lee> cool - so plan is -- I'll fix up that addtional test, and talk to martin about merging tempest patch without formpost changes
21:30:10 <timburke> the follow-up should just give clients an option for another way to spell it
21:30:14 <ade_lee> then we can merge all the swift patches
21:30:33 <ade_lee> and then we can add extra patch to tempest to switch formpost tests
21:30:47 <timburke> yup! sounds right to me
21:30:52 <mattoliver> +1
21:30:58 <ade_lee> cool thanks!
21:31:39 <timburke> #topic ring v2
21:32:09 <timburke> sorry, mattoliver, i still haven't gotten to following up on your review :-(
21:32:40 <mattoliver> nps, i mean it'll be good to get too, but you are saving our gate bacon :P
21:33:07 <mattoliver> I tried to address most the todos in the follow up.
21:33:37 <timburke> thanks for that -- though i think there's one part i won't be taking as-is
21:33:57 <timburke> (around unrecognized digest algos)
21:34:58 <timburke> let's see... the memcache patches are mostly resolved now...
21:35:15 <timburke> #topic eventlet, logging, and locking
21:35:37 <timburke> mattoliver, am i remembering right that you were going to try this out in our staging environment?
21:35:57 <timburke> i *know* we still haven't tried it out in prod yet
21:36:52 <mattoliver> I was suppose to. But onboarding new devs got in the way. I know I said it last week, but I'll try generating a load and giving it whirl this week.
21:37:22 <mattoliver> Seems we have Als backend ratelimint in there 2.
21:37:27 <timburke> sounds good -- just wanted to make sure i wasn't the one who'd committed to it ;-)
21:37:36 <mattoliver> so load generation we might be able to do 2 birds one stone
21:37:41 <timburke> yup!
21:37:42 <mattoliver> lol
21:38:20 <timburke> #topic backend ratelimiting
21:38:48 <timburke> we're up to 2 +2s on https://review.opendev.org/c/openstack/swift/+/836046/ ! should we wait on the results of your testing before merging?
21:39:16 <timburke> or should i just blindly +A once we've got the gate passing again? ;-)
21:40:40 <mattoliver> good question, in theory it can be turned off. But maybe work on the gate and hopefully by then we'll have some staging results
21:40:54 <timburke> sounds good
21:41:28 <timburke> #topic ptg
21:41:55 <timburke> i almost forgot! they've set the dates for the next (in person!) PTG
21:41:57 <timburke> #link http://lists.openstack.org/pipermail/openstack-discuss/2022-May/028601.html
21:42:06 <mattoliver> \o/
21:42:21 <timburke> October 17-20th, in Columbus, Ohio
21:43:01 <mattoliver> It'll be so awesome to meet up face to face with everyone again!
21:43:14 <timburke> for sure
21:43:22 <timburke> i can hardly wait
21:43:27 <mattoliver> I should confirm work will send me, then get my esta organised so I can enter the US :P
21:43:39 <timburke> that's all i've got
21:43:41 <mattoliver> got a few months :P
21:43:47 <timburke> #topic open discussion
21:43:56 <timburke> anything else we should bring up this week?
21:45:04 <mattoliver> Been playing with the next steps for not moving to Cleaved until we're ready. Hopefully have something up to review today or tomorrow
21:45:29 <timburke> nice!
21:45:37 <mattoliver> code just doesn't look right yet. OCD :P
21:46:13 <timburke> oh yeah, and i should go re-review https://review.opendev.org/c/openstack/swift/+/841143
21:46:36 <mattoliver> yeah, then we can land it
21:46:51 <mattoliver> it's better then what we have now. And saved our bacon in prod.
21:46:57 <mattoliver> so think it's a win
21:47:05 <timburke> 👍
21:47:22 <timburke> all right, i'll let you get on with your morning, then
21:47:31 <timburke> thanks for coming, and thanks for working on swift!
21:47:35 <timburke> #endmeeting