Wednesday, 2025-09-24

opendevreviewTim Burke proposed openstack/liberasurecode master: Run helgrind (and rename memcheck target)  https://review.opendev.org/c/openstack/liberasurecode/+/96214001:26
opendevreviewTim Burke proposed openstack/liberasurecode master: Split unittest job  https://review.opendev.org/c/openstack/liberasurecode/+/96214901:26
opendevreviewTim Burke proposed openstack/liberasurecode master: Split unittest job  https://review.opendev.org/c/openstack/liberasurecode/+/96214901:29
opendevreviewTim Burke proposed openstack/liberasurecode master: Run helgrind (and rename memcheck target)  https://review.opendev.org/c/openstack/liberasurecode/+/96214001:29
opendevreviewTim Burke proposed openstack/liberasurecode master: Split unittest job  https://review.opendev.org/c/openstack/liberasurecode/+/96214901:38
opendevreviewTim Burke proposed openstack/liberasurecode master: Run helgrind (and rename memcheck target)  https://review.opendev.org/c/openstack/liberasurecode/+/96214001:38
opendevreviewMatthew Oliver proposed openstack/swift master: reconstructor: Use last_part_node when building SYNC jobs  https://review.opendev.org/c/openstack/swift/+/83500106:39
mattoliverHey everyone, whose ready for a swift meeting?07:00
mattoliver#startmeeting swift07:01
opendevmeetMeeting started Wed Sep 24 07:01:05 2025 UTC and is due to finish in 60 minutes.  The chair is mattoliver. Information about MeetBot at http://wiki.debian.org/MeetBot.07:01
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.07:01
opendevmeetThe meeting name has been set to 'swift'07:01
mattoliverWho is here for the swift meeting?07:01
cschwedeo/07:01
mattolivernice, hey cschwede !07:02
mattoliverat least i wont be talking to myself :P 07:02
mattoliverAs usual the agenda is at:07:02
mattoliver#link https://wiki.openstack.org/wiki/Meetings/Swift07:02
mattoliverI'm going to skip the bug topic, I forgot to remove it. Instead we'll move it to a topic at the PTG :)07:03
mattoliverso 07:03
mattoliver#topic eventlet removal POC update07:03
mattoliver#link https://review.opendev.org/q/status:open+project:openstack/swift+branch:feature/threaded07:03
mattoliverWe have a feature branch, so that might be helpful ^07:03
mattoliverHow goes things with the POC cschwede ?07:04
cschwedeNo big updates this time, been on PTO last week. That said, pretty good progress before PTO - unit+functional tests passing for the account-server :)07:04
cschwedeCurrently working on more fixes to eventually pass all tests without eventlet too07:05
mattolivernice! and fair enough with the PTO :) 07:06
mattoliverI hope the time off was awesome!07:06
cschwedeThanks, it was - been out in the mountains :D07:07
mattoliverI'm jealous! 07:07
mattoliverI have been meaning to give the acount-server sans eventlet a whirl, just haven't seem to find the time yet. Work is always crazy busy.07:07
mattoliverLet's move on then07:07
mattoliver#topic October vPTG07:08
mattoliverThe vPTG is coming up07:08
mattoliverOct 27-3107:08
cschwedeWow, only one month left until PTG - time flies by!07:08
mattoliverAnd as in PTGs of very old, I'll be jetlagged, because the week b4 I'll be in the US and would have only just got home. So seems fitting :P07:08
mattoliver#link https://ptg.openinfra.org/07:09
mattoliver#link https://etherpad.opendev.org/p/swift-ptg-gazpacho07:09
mattoliverWe now have the etherpad to fill out. I just dded the bug triage topic :P 07:09
mattoliverStill need to add others, will do that soon. 07:09
cschwedeYes, let's discuss some plans to get better with the triage - and make more progress. Thx!07:10
mattoliverBut please if there is anything you want to talk about, and cool ideas, add them so we can discuss and make swift even better!07:10
mattoliverYeah, like just remove old bugs :P 07:10
mattoliverAlso don't forget to register for the PTG07:11
mattoliverThats more informational, so let's move on07:11
mattoliver#topic ringv2 next steps07:11
mattoliverChain starts:..07:11
mattoliver#link https://review.opendev.org/c/openstack/swift/+/95526307:11
mattoliverI bring this up because it's something I'm digging into.07:12
mattoliverAnd the chain has grown and about to get even longer :P 07:12
mattoliverThe first few patches are just some cleaning things up. And a ring/ringdata refactor to make the later patches better07:12
mattolivers/better/cleaner/07:13
mattoliverThe tail of the chain is making the reconstructor smarter by looking for hashes on an old primary, and if there is don't build a sync job. Because it still has frags it hasn't moved (ie don't bother rebuilding if the old primary still has it).07:14
cschwedeAh, that sounds like a great improvement!07:15
mattoliverCurrently I'm working on the proxy being smarter with old proxies, ie GETs to look at old primaries (kinda as a first handoff) 07:15
cschwedesmarter with old _primaries_?07:15
mattoliveryeah, sorry. Using the rings history tables, so on "reads" and if there is an old primary, check it before other handoffs.07:16
cschwedethx, these improvements sound pretty great.07:17
mattoliverBut I've moved the last_parts_moved to the ring too, so it can also know how long ago the part was moved and should it bother checking07:17
mattoliverthanks. They're still in flux, esp the later patches. But we can talk all about them at the PTG :) 07:17
mattoliverSpeaking of other interesting patches I'm working on.. 07:18
mattoliver#topic unique proxy offsets07:18
mattoliverSo clayg and acoles are working on improvments to timestamp collisions for EC in diskfile. 07:19
mattoliverAnd they're making it much more robust and much better.07:19
mattoliverAt the same time, I07:19
mattoliveram working trying to stop timestamp collisions from happening at all by using a modified and simlified version of the lamport clock algorithm07:20
mattoliver#link https://review.opendev.org/c/openstack/swift/+/95900907:20
mattoliverBasically giving every proxy a unique offset to add to x-timestamps, so even if we have a timestamp collision, the offset automatically breaks the tie07:21
mattoliverSo far, it seems to work great. We already have Timestamp class throughout swift which supports offsets AND it's already plumbed through to diskfile07:21
mattoliverSo far all the probetests that we have to simulate the EC timestamp collisions are fixed by it too.07:22
cschwedeJust read the commit msg - really like the details in the msg, kudos07:22
mattoliverJust need to eyes on it because the generating of a unique proxy number (automatically) is always hard07:23
mattoliverthanks :) 07:23
mattoliverMoving on to the last topic I have07:23
mattoliver#topic Relinker - allow clobber-hardlink-collision07:24
mattoliverSo we've been doing another part power increase on our EC policy. And had relinker's blocked and not finishing07:24
cschwedeuhh, what happened there?07:25
mattoliverTurns out, out cluster is really big and we've had some more timestamp collisions.. and we've to got a place where the old and new object are 2 different inodes and not hardlinked.07:25
mattoliverie one writes the first and then hardlinks it to the 2nd location. As the first one somehow writes the first one and then fails to harklink the second because it exists.07:26
mattoliverThen when the relinker comes along it also fails because the file is already there with a different inode.07:26
cschwedeouch!07:27
mattoliverThere wasn't many so the SRE could manually unlink the new location and let the relinker do the hardlink again.07:27
mattoliverBut it was still manual07:27
mattoliverAgain, our timestamp collision stuff will also solve this eventually07:27
cschwedeso even at your very large cluster size this seems to be a very rare event, given that it can be fixed manually in a timely manner?07:28
mattoliverbut in the mean time, clayg has a patch to give SRE the option to clobber in the relinker07:28
mattoliveryeah, that and we have dedicated SREs :P 07:28
cschwede#link https://review.opendev.org/c/openstack/swift/+/96176107:29
mattoliverbut yeah. enough to slow things down as to figure out why a bunch of disks never end07:29
mattoliveroh yeah, thanks cschwede :)07:29
mattoliverSo that patch allows ops to set an option to globber the file when hardlinking. Just to make is easer for them if there is a bunch on a node that they want ot run manually07:30
mattoliverAnyway, thats all I have07:30
mattoliver#topic open discussion07:30
mattoliverI've already talked about all my fun stuff :) 07:30
mattoliverAnything else to bring up?07:30
cschwedeNothing from my side, thx07:31
cschwedeAs always, thx mattoliver for all the updates and your time!07:31
mattoliverWell then let's call it early.07:31
mattoliverThanks for coming and thanks for working on swift!07:31
mattoliverThanks cschwede 07:31
mattoliver#endmeeting07:31
opendevmeetMeeting ended Wed Sep 24 07:31:45 2025 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)07:31
opendevmeetMinutes:        https://meetings.opendev.org/meetings/swift/2025/swift.2025-09-24-07.01.html07:31
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/swift/2025/swift.2025-09-24-07.01.txt07:31
opendevmeetLog:            https://meetings.opendev.org/meetings/swift/2025/swift.2025-09-24-07.01.log.html07:31
opendevreviewAlistair Coles proposed openstack/swift master: WIP sq: make FragZipper configurable  https://review.opendev.org/c/openstack/swift/+/96200310:04
*** diablo_rojo_phone_ is now known as diablo_rojo_phone14:19
*** diablo_rojo_phone is now known as Guest2727514:20
opendevreviewTim Burke proposed openstack/pyeclib master: Switch to black for style checks  https://review.opendev.org/c/openstack/pyeclib/+/83669416:13
opendevreviewAlistair Coles proposed openstack/swift master: WIP: include explicit uid in .data file name  https://review.opendev.org/c/openstack/swift/+/96220517:16
opendevreviewChristian Ohanaja proposed openstack/swift master: Adds cycle time tracking to sharder recon logs  https://review.opendev.org/c/openstack/swift/+/96220917:52
opendevreviewChristian Ohanaja proposed openstack/swift master: Adds cycle time tracking to sharder recon logs  https://review.opendev.org/c/openstack/swift/+/96220917:59
opendevreviewClay Gerrard proposed openstack/swift master: add minimal trace to get_hashes  https://review.opendev.org/c/openstack/swift/+/95437918:18
opendevreviewASHWIN A NAIR proposed openstack/swift master: common.db_replicator: include container path in logs, not just db path  https://review.opendev.org/c/openstack/swift/+/93337921:27
opendevreviewASHWIN A NAIR proposed openstack/swift master: account-broker: add path property with lazy, resilient cache  https://review.opendev.org/c/openstack/swift/+/96028923:53
opendevreviewASHWIN A NAIR proposed openstack/swift master: brokers: add resilient path property with lazy cache  https://review.opendev.org/c/openstack/swift/+/96028923:55

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