| *** gtema has quit IRC | 00:04 | |
| *** zaitcev has quit IRC | 01:49 | |
| *** gtema has joined #openstack-swift | 02:00 | |
| *** zaitcev has joined #openstack-swift | 02:03 | |
| *** ChanServ sets mode: +v zaitcev | 02:03 | |
| *** gtema has quit IRC | 02:04 | |
| *** zamba has quit IRC | 03:03 | |
| *** zamba has joined #openstack-swift | 03:24 | |
| *** psachin has joined #openstack-swift | 03:31 | |
| *** djhankb has quit IRC | 03:32 | |
| *** m75abrams has joined #openstack-swift | 03:34 | |
| *** djhankb has joined #openstack-swift | 03:35 | |
| *** gyee has quit IRC | 03:50 | |
| *** gtema has joined #openstack-swift | 04:01 | |
| *** rcernin has quit IRC | 04:03 | |
| *** gtema has quit IRC | 04:06 | |
| *** rcernin has joined #openstack-swift | 04:24 | |
| *** evrardjp has quit IRC | 04:33 | |
| *** evrardjp has joined #openstack-swift | 04:33 | |
| *** gtema has joined #openstack-swift | 06:02 | |
| *** gtema has quit IRC | 06:07 | |
| *** gtema has joined #openstack-swift | 06:34 | |
| *** gtema has quit IRC | 06:49 | |
| *** gtema has joined #openstack-swift | 06:50 | |
| *** rcernin has quit IRC | 06:54 | |
| *** takamatsu has quit IRC | 08:02 | |
| *** dsariel has quit IRC | 08:36 | |
| *** takamatsu has joined #openstack-swift | 08:58 | |
| *** takamatsu has quit IRC | 09:07 | |
| *** takamatsu has joined #openstack-swift | 09:07 | |
| *** dsariel has joined #openstack-swift | 09:10 | |
| *** mikecmpbll has joined #openstack-swift | 09:29 | |
| *** takamatsu has quit IRC | 09:53 | |
| *** tkajinam has quit IRC | 10:33 | |
| *** takamatsu has joined #openstack-swift | 11:06 | |
| *** rcernin has joined #openstack-swift | 12:08 | |
| *** mikecmpbll has quit IRC | 12:31 | |
| *** mikecmpbll has joined #openstack-swift | 12:34 | |
| *** m75abrams has quit IRC | 13:00 | |
| *** rcernin has quit IRC | 13:40 | |
| *** gtema has quit IRC | 14:42 | |
| *** psachin has quit IRC | 14:43 | |
| *** gtema has joined #openstack-swift | 15:13 | |
| *** gtema has quit IRC | 15:18 | |
| *** gtema has joined #openstack-swift | 15:40 | |
| *** gtema_ has joined #openstack-swift | 15:52 | |
| *** gtema has quit IRC | 15:55 | |
| *** dsariel has quit IRC | 16:17 | |
| *** dsariel has joined #openstack-swift | 16:17 | |
| *** patchbot has joined #openstack-swift | 16:34 | |
| *** gyee has joined #openstack-swift | 16:51 | |
| DHE | does this sound like a good idea? a cron job that, hourly, runs swift-object-replicator -o /etc/swift/object-server-handoffsonly.conf # where said file obviously specifies handoffs only mode, and kept separate from the normal replicator job | 16:55 |
|---|---|---|
| clayg | maybe! do you have a lot of handoffs sometimes? | 17:17 |
| *** gtema_ has quit IRC | 17:23 | |
| clayg | hey if anyone else is sharding containers watch your memcache misses: https://bugs.launchpad.net/swift/+bug/1890643 | 17:28 |
| openstack | Launchpad bug 1890643 in OpenStack Object Storage (swift) "shard range cache can get too big for memcache" [Undecided,New] | 17:28 |
| clayg | I was trolling some code the other day and this smelled for me: https://github.com/openstack/swift/blob/master/swift/proxy/controllers/base.py#L2187 | 17:31 |
| clayg | Should we be consuming response.body for 503 on GET to container listing? | 17:31 |
| clayg | When my container servers were getting overwhelmed I noticed a bunch of connections in CLOSE_WAIT; but don't necessarily know if that's because the proxy wasn't reading/closing error connections 🤷♂️ | 17:32 |
| *** dsariel has quit IRC | 17:33 | |
| *** mikecmpbll has quit IRC | 17:36 | |
| *** mikecmpbll has joined #openstack-swift | 17:40 | |
| clayg | I kinda feel like a healthy connection when closed ends up in TIME_WAIT 🤔 | 17:46 |
| DHE | clayg: I've been put in the unpleasant position of having a lot of data to move around, some big honking machines with nice hardware, but an inadequate network connection to support them. I'm raising the weights of these new machines in small increments and watching them peg their 1gig ports as they fill over time. | 17:54 |
| DHE | honestly I'm pegging them by manually running such an object-replicator command manually in order to get the data moved as fast as possible. it sucks. | 17:55 |
| *** patchbot has quit IRC | 18:01 | |
| *** patchbot has joined #openstack-swift | 18:01 | |
| clayg | we often kick things into handoff mode during a rebalance; sorry it's not easy going for you 😥 | 18:04 |
| *** timburke has quit IRC | 18:30 | |
| *** timburke has joined #openstack-swift | 18:30 | |
| *** ChanServ sets mode: +v timburke | 18:30 | |
| *** gmann is now known as gmann_afk | 18:35 | |
| DHE | well it's not your fault. I blame the switches for only having 1gig ports. | 18:42 |
| DHE | and even the ones that do have 10gig+ speeds I get messages asking if it's expected that we hit 80% utilization on it | 18:43 |
| DHE | :/ | 18:43 |
| *** gtema has joined #openstack-swift | 19:23 | |
| *** gtema has quit IRC | 19:28 | |
| *** dsariel has joined #openstack-swift | 20:18 | |
| *** mikecmpbll has quit IRC | 20:41 | |
| *** mikecmpbll has joined #openstack-swift | 20:47 | |
| *** dsariel has quit IRC | 21:01 | |
| *** rcernin has joined #openstack-swift | 21:23 | |
| *** rcernin has quit IRC | 21:27 | |
| *** rcernin has joined #openstack-swift | 21:34 | |
| *** rcernin has quit IRC | 21:42 | |
| *** rcernin has joined #openstack-swift | 22:12 | |
| *** hoonetorg has quit IRC | 22:15 | |
| *** rcernin has quit IRC | 22:17 | |
| *** gmann_afk is now known as gmann | 22:35 | |
| *** rcernin has joined #openstack-swift | 22:47 | |
| *** rcernin has quit IRC | 22:52 | |
| *** tkajinam has joined #openstack-swift | 22:55 | |
| *** rcernin has joined #openstack-swift | 23:20 | |
| mattoliverau | morning | 23:33 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!