21:05:47 <mattoliver> #startmeeting swift 21:05:47 <opendevmeet> Meeting started Wed Dec 18 21:05:47 2024 UTC and is due to finish in 60 minutes. The chair is mattoliver. Information about MeetBot at http://wiki.debian.org/MeetBot. 21:05:47 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 21:05:47 <opendevmeet> The meeting name has been set to 'swift' 21:06:23 <mattoliver> Agenda which hasn't been updated is at: 21:06:28 <mattoliver> #link https://wiki.openstack.org/wiki/Meetings/Swift 21:07:17 <mattoliver> Tim needed to take his daughter to the Orthodonist today, so basically skipping the meeting. I'll just run through the only item so it's written in the meeting logs 21:07:30 <mattoliver> #topic christmas and new years 21:07:51 <mattoliver> The swift meeting will be skipped for the next 2 weeks due to the upcoming holidays. 21:08:26 <mattoliver> #topic open discussion 21:09:03 <mattoliver> Tim and I will be meeting with Seongsoo Cho and his team to talk sharding tomorrow. I'm looking forward to it. 21:09:56 <jianjian> nice, any particular sharding feature Seongsoo is going to work on? 21:10:37 <mattoliver> it more giving them confidence to run sharding themselves in production. 21:10:52 <mattoliver> More people using sharding means more work on it I hope ;) 21:10:59 <mattoliver> And I've been working on some changes to sharding replication and in particular: p 901095 21:10:59 <patch-bot> https://review.opendev.org/c/openstack/swift/+/901095 - swift - db_replication: big containers that haven't sunk s... - 3 patch sets 21:12:24 <mattoliver> We've also noticed that 404s on EC can be quite expensive/slow because if they walk 2 * replicas that's alot to check. So I've got a patch to allow request_node_count be overridden in policy overrides.: p 937667 21:12:25 <patch-bot> https://review.opendev.org/c/openstack/swift/+/937667 - swift - Allow request_node_count to be overridden per policy - 2 patch sets 21:12:55 <mattoliver> So we can set it to something like your ndata or something more meaningful, exp for heads. 21:13:17 <mattoliver> heck even replica count would be better :P 21:13:19 <jianjian> yes, we are going to run it in production and see how does it go 21:13:28 <mattoliver> \o/ 21:14:09 <mattoliver> Other then that, I'm trying to get this chain reviewed and landed: p 909802 21:14:10 <patch-bot> https://review.opendev.org/c/openstack/swift/+/909802 - swift - s3api: Additional checksums for MPUs - 39 patch sets 21:14:23 <mattoliver> because it's been around a long time and tim and ash have been patient and asking nicely :P 21:14:48 <jianjian> 💚 21:15:25 <mattoliver> But I'm going on leave for a few weeks at the end of the week, so hopefully push it as far as I can over the next 2 days. I've +2ed the first 2 patches. 21:15:51 <mattoliver> Anyway, anything else to bring up, otherwise we'll end the meeting :) 21:15:54 <mattoliver> ? 21:16:28 <jianjian> I don't have anything to add, thanks matt 21:16:47 <mattoliver> cool, then thanks for coming/reading and thanks for working on swift! 21:16:51 <mattoliver> #endmeeting