| *** viks____ has joined #openstack-swift | 01:49 | |
| *** gyee has quit IRC | 03:18 | |
| *** psachin has joined #openstack-swift | 03:37 | |
| *** renich has quit IRC | 03:45 | |
| *** evrardjp has quit IRC | 04:36 | |
| *** evrardjp has joined #openstack-swift | 04:36 | |
| *** psachin has quit IRC | 04:56 | |
| openstackgerrit | Matthew Oliver proposed openstack/python-swiftclient master: Ussuri contrib docs community goal https://review.opendev.org/727633 | 05:39 |
|---|---|---|
| *** manuvakery has joined #openstack-swift | 05:52 | |
| *** ccamacho has joined #openstack-swift | 06:00 | |
| *** onovy has quit IRC | 06:53 | |
| *** onovy has joined #openstack-swift | 06:59 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: Fix pep8 job https://review.opendev.org/727644 | 07:24 |
| *** rpittau|afk is now known as rpittau | 07:29 | |
| openstackgerrit | Merged openstack/python-swiftclient master: Support v3 application credentials auth. https://review.opendev.org/699457 | 08:36 |
| *** dtantsur|afk is now known as dtantsur | 09:13 | |
| *** manuvakery has quit IRC | 10:12 | |
| *** yuxin_ has quit IRC | 10:49 | |
| *** yuxin_ has joined #openstack-swift | 10:54 | |
| *** rpittau is now known as rpittau|bbl | 11:04 | |
| rledisez | morning | 12:06 |
| rledisez | I was reading https://unixism.net/loti/index.html and one sentence in the intro caught my curiosity: "If you are going to be building something that deals with thousands or even hundreds of thousands of requests per hour, you need not bother with asynchronous I/O." | 12:06 |
| rledisez | So I looked at my numbers and my most solicited service (container) handles about 220 req/s per core, and the service with the most concurrent connection (proxy) seems to only handle about 100 concurrent connections. | 12:07 |
| rledisez | So I took 1h to do a simple test: bench container server with one worker pinned on one core with two implementations: the normal/eventlet wsgi server and a threadpool-based wsgi server (pool size=500). Actually, the performance difference is not huge (between 3 and 4%) and as soon as I start to remove some eventlet code from common.utils, it improves | 12:08 |
| rledisez | So I'm now wondering if we really need asynchronicity. Maybe the Linux kernel is better than eventlet at managing concurrency (and how much concurrent connection do we expect on a service?) | 12:08 |
| *** rpittau|bbl is now known as rpittau | 12:19 | |
| *** tkajinam has quit IRC | 12:24 | |
| *** ianychoi has quit IRC | 12:55 | |
| openstackgerrit | Merged openstack/swift master: Fix pep8 job https://review.opendev.org/727644 | 14:59 |
| *** cwright has quit IRC | 15:35 | |
| *** dtantsur is now known as dtantsur|afk | 15:36 | |
| *** cwright has joined #openstack-swift | 15:36 | |
| *** gyee has joined #openstack-swift | 15:43 | |
| *** manuvakery has joined #openstack-swift | 15:56 | |
| *** rpittau is now known as rpittau|afk | 16:21 | |
| *** camelCaser has quit IRC | 16:22 | |
| *** ccamel has joined #openstack-swift | 16:22 | |
| *** evrardjp has quit IRC | 16:36 | |
| *** evrardjp has joined #openstack-swift | 16:36 | |
| openstackgerrit | Tim Burke proposed openstack/python-swiftclient master: Application credential support follow-up https://review.opendev.org/727845 | 17:30 |
| *** viks____ has quit IRC | 17:37 | |
| openstackgerrit | Merged openstack/swift master: py27: Suppress UnicodeWarnings in ShardRange setters https://review.opendev.org/725458 | 17:44 |
| *** manuvakery has quit IRC | 19:36 | |
| *** ccamacho has quit IRC | 20:54 | |
| kota_ | morning | 20:58 |
| seongsoocho | kota_: morning ~ | 20:58 |
| kota_ | seongsoocho: o/ | 20:58 |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: Breakup reclaim into batches https://review.opendev.org/727876 | 20:59 |
| timburke | almost meeting time! | 20:59 |
| zaitcev | "<timburke> .... unfortunately i've been a bit distracted, so i'm not entirely sure myself." ... if I launched 6 replicator instances or only 5. So, do you feel lucky, punk? | 21:19 |
| tdasilva | mattoliverau: if need be, we could just use something like zoom too, no? | 21:57 |
| mattoliverau | yeah true! | 21:57 |
| timburke | we want to keep things to | 21:57 |
| timburke | generally-accessible means where possible, but i'm not opposed to moving to zoom or something if we're running out of slotted time for an in-progress discussion | 21:58 |
| timburke | the nice thing about the centralized PTG rooms (i expect) will be that people know where to go to find swift folks, ask questions, etc | 21:59 |
| mattoliverau | +1 | 22:11 |
| timburke | added a new "PTG Patches" section to https://wiki.openstack.org/wiki/Swift/PriorityReviews | 22:19 |
| timburke | (of course, listing a patch there doesn't prevent us from looking at, thinking about, reviewing, or merging the patch *before* the PTG ;-) | 22:27 |
| timburke | think of it as an opportunity to list things where you'd be disappointed if *no progress* was made on it that week -- if we can get that progress before then, so much the better | 22:31 |
| *** tkajinam has joined #openstack-swift | 22:46 | |
| *** threestrands has joined #openstack-swift | 23:29 | |
| *** rcernin has quit IRC | 23:31 | |
| *** rcernin has joined #openstack-swift | 23:32 | |
| *** early has quit IRC | 23:38 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!