| *** gyee has quit IRC | 02:04 | |
| *** tdasilva_ has joined #openstack-swift | 02:24 | |
| *** ChanServ sets mode: +v tdasilva_ | 02:24 | |
| *** tdasilva has quit IRC | 02:25 | |
| *** tdasilva_ is now known as tdasilva | 02:25 | |
| *** coreycb has quit IRC | 02:36 | |
| *** coreycb has joined #openstack-swift | 02:38 | |
| timburke | zaitcev, looks like it's willing to start even without the section (!) -- but setting something like `shard_container_threshold = asdf` breaks it pretty good | 02:41 |
|---|---|---|
| zaitcev | timburke, that does not sound like an officially approved way unless we document it, right? | 02:42 |
| timburke | oh, maybe that's because i had the section also defined in another file in the directory... | 02:42 |
| zaitcev | Yeah | 02:42 |
| zaitcev | I commented it out on mine and it failed. | 02:42 |
| timburke | our .conf.d support is cool ... but sometimes it hides what i wanted to do a bit | 02:43 |
| *** psachin has joined #openstack-swift | 03:21 | |
| *** rcernin has quit IRC | 03:38 | |
| *** rcernin has joined #openstack-swift | 03:46 | |
| *** m75abrams has joined #openstack-swift | 04:07 | |
| *** evrardjp has quit IRC | 04:33 | |
| *** evrardjp has joined #openstack-swift | 04:33 | |
| *** rcernin has quit IRC | 07:07 | |
| *** rcernin has joined #openstack-swift | 07:09 | |
| *** rcernin has quit IRC | 07:23 | |
| *** mattia has joined #openstack-swift | 08:41 | |
| *** tkajinam has quit IRC | 08:55 | |
| *** mugsie has quit IRC | 10:07 | |
| *** mugsie has joined #openstack-swift | 10:10 | |
| *** manuvakery has quit IRC | 11:53 | |
| *** gregwork has quit IRC | 12:00 | |
| *** gtema has joined #openstack-swift | 12:57 | |
| gtema | with the default account_autocreate=False, what is the proper way to create account? | 13:45 |
| *** m75abrams has left #openstack-swift | 14:03 | |
| *** psachin has quit IRC | 15:09 | |
| *** gtema_ has joined #openstack-swift | 15:37 | |
| *** gyee has joined #openstack-swift | 15:38 | |
| *** gtema has quit IRC | 15:40 | |
| *** gtema_ has quit IRC | 16:40 | |
| ormandj | timburke: fwiw, we've got ussuri + your patch live, that user is not experiencing issues | 17:45 |
| timburke | \o/ | 17:45 |
| ormandj | i'll ping back tomorrow after it's been cooking for a day, but so far so good, to the tune of lots of requests per second | 17:46 |
| *** TViernion has quit IRC | 17:58 | |
| *** TViernion has joined #openstack-swift | 18:07 | |
| *** TViernion has quit IRC | 18:21 | |
| -openstackstatus- NOTICE: Due to a flood of connections from random prefixes, we have temporarily blocked all AS4837 (China Unicom) source addresses from access to the Git service at opendev.org while we investigate further options. | 18:23 | |
| *** TViernion has joined #openstack-swift | 18:27 | |
| zaitcev | Not sure if I should put it here... Knowing timburke's thorough and inquisitive nature, I'm afraid he's going to write a mountain of tests in the next hour, but... | 18:37 |
| zaitcev | Did anyone notice that our object-server's parent process spins with 10 ms timeouts? | 18:38 |
| zaitcev | There's a "with Timeout(loop_timeout, exception=False)" that seems like controlling it, and the loop_timeout is 0.5. | 18:39 |
| zaitcev | But strace shows select(0, NULL, NULL, NULL, {tv_sec=0, tv_usec=9964}) = 0 (Timeout) | 18:39 |
| zaitcev | Basically, I installed SAIO on my laptop, and with just swift-init main start the CPU consumption is a couple of percent. | 18:40 |
| zaitcev | And all that's happening is this looping with 10ms timeouts. | 18:40 |
| DHE | I believe most of the prefork supervisor processes do that, not just object-server | 18:41 |
| *** gtema has joined #openstack-swift | 18:41 | |
| zaitcev | No daemons are started, you know. "Main" only. | 18:41 |
| zaitcev | So, it's in swift/common/wsgi.py | 18:41 |
| *** gtema has quit IRC | 18:46 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: proxy-logging: Use swift.backend_path if available https://review.opendev.org/728298 | 19:09 |
| *** viks____ has quit IRC | 19:22 | |
| *** noonedeadpunk has quit IRC | 20:02 | |
| *** noonedeadpunk has joined #openstack-swift | 20:03 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: swift-container-info: better shard range summary https://review.opendev.org/738692 | 20:34 |
| clayg | maybe some good low hanging fruit for anyone wanting to get into sharding -> p 737056 (cc mattoliverau) | 20:35 |
| patchbot | https://review.opendev.org/#/c/737056/ - swift - swift-container-info: Don't show all shard ranges ... - 2 patch sets | 20:35 |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: proxy-logging: Use swift.backend_path if available https://review.opendev.org/728298 | 21:23 |
| clayg | timburke: this one I think is maybe ready? https://review.opendev.org/#/c/735738/ | 21:47 |
| patchbot | patch 735738 - swift - s3api: Don't do naive HEAD request for auth - 4 patch sets | 21:47 |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: WIP: s3api: Make multi-deletes async https://review.opendev.org/648263 | 22:00 |
| *** rcernin has joined #openstack-swift | 22:38 | |
| *** tkajinam has joined #openstack-swift | 22:48 | |
| *** rcernin has quit IRC | 22:48 | |
| *** rcernin has joined #openstack-swift | 22:49 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!