Wednesday, 2026-01-14

opendevreviewASHWIN A NAIR proposed openstack/swift master: s3api: fix s3api mpu complete log for correct c/o  https://review.opendev.org/c/openstack/swift/+/96691701:45
opendevreviewASHWIN A NAIR proposed openstack/swift master: sq: adding test to validate same concept for multipart object copy  https://review.opendev.org/c/openstack/swift/+/97333702:36
mattoliverIt's been a while since we had a swift meeting. 07:01
mattoliverI don't have much to talk about (still catching up) but thought it might be worth while anyway. 07:01
mattoliverAt least so there are minutes. 07:01
mattoliverEven if I talk to myself :P 07:02
mattoliver#startmeeting swift07:02
opendevmeetMeeting started Wed Jan 14 07:02:08 2026 UTC and is due to finish in 60 minutes.  The chair is mattoliver. Information about MeetBot at http://wiki.debian.org/MeetBot.07:02
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.07:02
opendevmeetThe meeting name has been set to 'swift'07:02
mattoliveranyone where (other then me) for the swift meeting?07:02
mattoliverAs always agenda is at: 07:02
mattoliver#link https://wiki.openstack.org/wiki/Meetings/Swift07:02
mattoliverNot many topics, and most of them are place holders, so they should go quickly :) 07:03
mattoliver#topic eventlet removal07:03
mattoliverThis is still a thing, but I know at nvidia we've been distracted with other things and holidays. 07:03
mattoliverWe were going to have a video meeting with cschwede about it. We should organise that soon. Advartise it here so other can also attend. We'd probably use meetpad (like we do for PTGs).07:05
mattoliverBut other then that not sure there has been too much progress, this yeah for sure!07:05
mattoliver#link https://review.opendev.org/q/project:openstack/swift+branch:feature/threaded+is:open07:05
mattoliver^ To track things on the feature branch07:05
mattoliverMoving on..07:05
mattoliver#topic Timestamp collisions07:06
mattoliverStill a thing.07:06
mattoliverTurns out getting confidence on a major change to core swift code (even though the Timestamp class is plumbed thoughout) is still a hard concept have confidence in. There are many things that expect certain timestamps in a particular format. 07:07
mattoliverBut we've landed ALOT of precursor patches07:08
mattoliverLots of test cleanup to use timestamp classes so we can see what breaks etc.07:08
mattoliverI've been working on the nano patch some, rebasing on all these changes to see how it fairs. But the plan is still the jitter patch07:09
mattoliverStill want to fix it properly, but version 2 will just be a random int to stop the bleeding.07:09
opendevreviewMerged openstack/swift master: probe tests: fix timestamp handling  https://review.opendev.org/c/openstack/swift/+/97246307:10
mattoliverjianjian has come up with a great idea to first just target obj PUT's. As that's were we see the collisions happen, AND means we don't have to worry about offsets07:10
opendevreviewMerged openstack/swift master: test_backend.py: refactor TestUpdateNewItemFromExisting  https://review.opendev.org/c/openstack/swift/+/97253307:10
mattoliverSo think that might be where we go07:10
mattoliverThere has also been work downstream to try and better understand from a client POV how this is happening.07:10
mattoliverSo learning alot, and will make swift better!07:11
mattoliverNext..07:11
mattoliver#topic Fixing dockerhub saio image07:11
mattoliverWe have an saio docker image that's beening built via zuul.. but it's broken. timburke and I have been trying to get it working again and re-learn how it was set up (thanks Thiago, not only for the initial work but helping us re-learn). 07:13
mattoliver#link https://review.opendev.org/c/openstack/swift/+/97319407:13
patch-botpatch 973194 - swift - Update dockerhub secret - 2 patch sets07:13
mattoliverStill a ways to go, but I have saved the account creds somewhere safe (and shared with timburke ). 07:13
mattoliverI'll make some notes and put them somewhere useful for the project once we've got it working.07:14
mattoliver#topic open discussion07:14
mattoliverthat's all I really have07:14
mattoliverThere are other things I'd like to loop back to and continue to work on and land when we have cycles. My vacuum patch, SO_TIMESTAMPING, tracing. And talking with csmart some way to better manage disabled disks in core swift. So plenty of work to do.07:16
mattoliverI know timburke has been playing around with liberasurecode so lots of cool stuff happening there too! If anyone wants something to work on, we have way more work then we can possibly do, so good time to work on Swift! 07:17
mattoliverAnyway, that's all I got.07:17
mattoliverThanks for coming and thanks for working on swift!07:17
mattoliver#endmeeting07:17
opendevmeetMeeting ended Wed Jan 14 07:17:39 2026 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)07:17
opendevmeetMinutes:        https://meetings.opendev.org/meetings/swift/2026/swift.2026-01-14-07.02.html07:17
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/swift/2026/swift.2026-01-14-07.02.txt07:17
opendevmeetLog:            https://meetings.opendev.org/meetings/swift/2026/swift.2026-01-14-07.02.log.html07:17
mattoliverhey kota , was just me, so talked to myself :P you don't usually come to this one :) 07:18
kotathx mattoliver for organizing the meeting staff. sorry for recent my absence. 07:19
kotabut happy to see to confirm this time still working.07:20
kotaah, ok. https://wiki.openstack.org/wiki/Meetings/Swift was updated. I looked at the wiki last year, I was not sure when the meeting starts in the new year.07:21
kotathanks again for the house keeping of the community.07:22
cschwedemattoliver: sorry, missed the updated time - will update my calendar entry. thx for the summary, I just read the backlog07:25
cschwedemattoliver: re eventlet removal - i'm working on a summary doc as well as some key patches. ack regarding another virtual call, we should schedule that one soon07:26
mattolivercschwede: nps, I should've reminded people. Kk sounds good! 07:40
opendevreviewAlistair Coles proposed openstack/swift master: s3api: do not set x-timestamp header  https://review.opendev.org/c/openstack/swift/+/97327214:51
opendevreviewAlistair Coles proposed openstack/swift master: timestamps: add SimpleTimestamp for ShardRanges  https://review.opendev.org/c/openstack/swift/+/96874014:51
opendevreviewAlistair Coles proposed openstack/swift master: WIP Timestamp: add random jitter to offset field  https://review.opendev.org/c/openstack/swift/+/96773814:51
opendevreviewAlistair Coles proposed openstack/swift master: object_server: return 400 if POST timestamp has jitter  https://review.opendev.org/c/openstack/swift/+/97253514:51
opendevreviewAlistair Coles proposed openstack/swift master: tolerate but round meta timestamp jitter in backend  https://review.opendev.org/c/openstack/swift/+/97253614:51
opendevreviewAlistair Coles proposed openstack/swift master: sq? Timestamp: default with_jitter=False  https://review.opendev.org/c/openstack/swift/+/97221714:51
timburkethanks for running this week's meeting, mattoliver! i had wanted to do one last week, but had an orthodontist appointment for my son17:36
opendevreviewShreeya Deshpande proposed openstack/swift master: proxy-logging: Add 'api' labels  https://review.opendev.org/c/openstack/swift/+/93788423:14

Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!