| *** nadeem has joined #openstack-swauth | 03:45 | |
| *** nadeem has quit IRC | 03:46 | |
| *** openstackgerrit has quit IRC | 09:47 | |
| *** openstackgerrit has joined #openstack-swauth | 09:48 | |
| -openstackstatus- NOTICE: Zuul is moving in very slow motion since roughly 13:30 UTC; the Infra team is investigating. | 16:43 | |
| *** klamath has joined #openstack-swauth | 19:56 | |
| klamath | howdy, i was wondering if swauth can work swift multi-region | 19:57 |
|---|---|---|
| onovy | we have swauth in multi-region swift | 20:02 |
| klamath | no issues with memcache? | 20:03 |
| onovy | BUT :) we have one memcached cluster per region | 20:03 |
| klamath | ahh ok | 20:03 |
| onovy | but that's classic in single-region swift too | 20:04 |
| onovy | every proxy/store have memcached daemon, and memcache.conf is same for whole region | 20:04 |
| klamath | awesome, trying to figure some options out while keystone federation comes out | 20:04 |
| onovy | so no problem if you try to communicate with other proxy with same tokens inside region | 20:04 |
| onovy | but when you communicate with different region, token is not in memcache, so swauth will contact backend swift | 20:05 |
| onovy | load it to local memcached | 20:05 |
| onovy | this traffic to swift store is acceptable - low | 20:05 |
| onovy | "federation" is done by swift itself here | 20:06 |
| *** nadeem has joined #openstack-swauth | 20:15 | |
| klamath | onovy: on your two sites, do you use the same swift hash prefix and suffix? | 20:45 |
| onovy | yes | 20:45 |
| onovy | i think this is mandatory | 20:46 |
| klamath | and you can read and write to both of them without issue | 20:46 |
| onovy | yes | 20:46 |
| klamath | thank you | 20:46 |
| onovy | swift hash prefix and suffix must be same for whole cluster | 20:46 |
| onovy | even in multiregion | 20:46 |
| *** nadeem has quit IRC | 22:56 | |
| *** klamath has quit IRC | 23:16 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!