| mattoliverau | A container is actaully just a way to break down the namespace, so really is a metadata layer. objects are stored throughout the cluster, they're the main bytes written to disk. Containers are sqlite databases replicated across the cluster like objects. So when a new object is added, they are placed on a storage server and then the container is updated. If the cluster is really busy and the container update fails the | 00:19 |
|---|---|---|
| mattoliverau | update will happen asynchronously. So the object listings inside containers can lag a bit sometimes, but you can always access the objects even if the container is yet to be updated. This is the eventual consistency nature of Swift, but is what allows it to scale massively even over higher latency links. | 00:19 |
| tdasilva | mattoliverau, abelur: just fyi...from the links abelur shared earlier, i believe his swift is really ceph rgw | 00:24 |
| mattoliverau | oh.. well then that's different. You don't get eventual consistency, and not quite all the Swift api is covered. And it wont scale over higher latency links. | 00:26 |
| *** gyee has quit IRC | 00:28 | |
| mattoliverau | Hopefully there are multiple rgw gateways otherwise it's a bottleneck in a ceph cluster. But it looks like it's vexxhost so I'd assume they'd know what there doing. Ceph is awesome at block, and ok as object so long as you don't want to scale over larger distance regions. | 00:28 |
| abelur | what is the better alternative to Ceph, on the cloud provider? | 00:31 |
| abelur | I'll have to bring this up with the provider since scalibility/consistantcy is going to be super imporant considering we'd have ~5k-10k created/cleaned everyday :| | 00:33 |
| *** rcernin has quit IRC | 00:34 | |
| *** rcernin has joined #openstack-swift | 00:35 | |
| *** irclogbot_2 has joined #openstack-swift | 00:56 | |
| *** irclogbot_2 has quit IRC | 01:00 | |
| *** tdasilva has quit IRC | 01:15 | |
| *** irclogbot_1 has joined #openstack-swift | 01:25 | |
| *** irclogbot_1 has quit IRC | 01:34 | |
| *** irclogbot_0 has joined #openstack-swift | 02:26 | |
| *** irclogbot_0 has quit IRC | 02:30 | |
| *** altlogbot_1 has joined #openstack-swift | 02:47 | |
| *** altlogbot_1 has quit IRC | 02:52 | |
| *** BjoernT has joined #openstack-swift | 03:21 | |
| *** irclogbot_0 has joined #openstack-swift | 03:22 | |
| *** irclogbot_0 has quit IRC | 03:26 | |
| *** BjoernT has quit IRC | 03:26 | |
| *** BjoernT has joined #openstack-swift | 03:27 | |
| *** psachin has joined #openstack-swift | 03:44 | |
| *** irclogbot_2 has joined #openstack-swift | 03:52 | |
| *** irclogbot_2 has quit IRC | 03:56 | |
| *** BjoernT has quit IRC | 04:02 | |
| *** baojg has joined #openstack-swift | 04:02 | |
| openstackgerrit | Merged openstack/python-swiftclient master: Authors/changelog for 3.8.0 https://review.opendev.org/670215 | 04:14 |
| *** rcernin has quit IRC | 04:54 | |
| *** irclogbot_2 has joined #openstack-swift | 05:00 | |
| *** new_student1411 has joined #openstack-swift | 05:07 | |
| *** pcaruana has joined #openstack-swift | 05:13 | |
| *** jistr has quit IRC | 05:15 | |
| *** new_student14119 has joined #openstack-swift | 05:15 | |
| *** irclogbot_2 has quit IRC | 05:16 | |
| *** jistr has joined #openstack-swift | 05:18 | |
| *** new_student1411 has quit IRC | 05:18 | |
| *** altlogbot_2 has joined #openstack-swift | 06:10 | |
| *** altlogbot_2 has quit IRC | 06:14 | |
| *** irclogbot_3 has joined #openstack-swift | 06:36 | |
| *** irclogbot_3 has quit IRC | 06:40 | |
| *** rdejoux has joined #openstack-swift | 07:32 | |
| *** ccamacho has joined #openstack-swift | 07:59 | |
| *** altlogbot_3 has joined #openstack-swift | 08:02 | |
| *** altlogbot_3 has quit IRC | 08:04 | |
| *** altlogbot_1 has joined #openstack-swift | 08:12 | |
| *** gkadam has joined #openstack-swift | 08:15 | |
| *** altlogbot_1 has quit IRC | 08:16 | |
| *** gkadam has quit IRC | 08:17 | |
| *** altlogbot_1 has joined #openstack-swift | 08:18 | |
| *** tkajinam has quit IRC | 08:19 | |
| *** altlogbot_1 has quit IRC | 08:22 | |
| *** altlogbot_3 has joined #openstack-swift | 08:24 | |
| *** altlogbot_3 has quit IRC | 08:29 | |
| *** Fidde has joined #openstack-swift | 08:31 | |
| *** dosaboy has quit IRC | 08:47 | |
| *** aluria has quit IRC | 08:48 | |
| *** altlogbot_1 has joined #openstack-swift | 08:54 | |
| *** altlogbot_1 has quit IRC | 08:58 | |
| *** dosaboy has joined #openstack-swift | 09:00 | |
| *** irclogbot_0 has joined #openstack-swift | 09:00 | |
| *** dosaboy has quit IRC | 09:03 | |
| *** aluria has joined #openstack-swift | 09:03 | |
| *** dosaboy has joined #openstack-swift | 09:04 | |
| *** irclogbot_0 has quit IRC | 09:04 | |
| *** irclogbot_1 has joined #openstack-swift | 10:32 | |
| *** irclogbot_1 has quit IRC | 10:38 | |
| *** irclogbot_1 has joined #openstack-swift | 10:43 | |
| *** irclogbot_1 has quit IRC | 10:44 | |
| *** dosaboy has quit IRC | 10:45 | |
| *** aluria has quit IRC | 10:56 | |
| *** dosaboy has joined #openstack-swift | 11:02 | |
| *** altlogbot_3 has joined #openstack-swift | 11:04 | |
| *** altlogbot_3 has quit IRC | 11:08 | |
| *** tesseract has joined #openstack-swift | 11:08 | |
| *** test_user has joined #openstack-swift | 11:10 | |
| *** aluria has joined #openstack-swift | 11:11 | |
| *** altlogbot_3 has joined #openstack-swift | 11:13 | |
| *** altlogbot_3 has quit IRC | 11:16 | |
| *** dosaboy has quit IRC | 11:27 | |
| *** dosaboy has joined #openstack-swift | 11:28 | |
| *** dosaboy has quit IRC | 11:28 | |
| *** rdejoux has quit IRC | 12:00 | |
| *** rdejoux has joined #openstack-swift | 12:01 | |
| *** viks___ has quit IRC | 12:46 | |
| *** viks___ has joined #openstack-swift | 12:48 | |
| *** baojg has quit IRC | 12:49 | |
| *** baojg has joined #openstack-swift | 12:50 | |
| *** baojg has quit IRC | 12:50 | |
| *** baojg has joined #openstack-swift | 12:50 | |
| *** baojg has quit IRC | 12:51 | |
| *** baojg has joined #openstack-swift | 12:51 | |
| *** baojg has quit IRC | 12:52 | |
| *** baojg has joined #openstack-swift | 12:52 | |
| *** baojg has quit IRC | 12:52 | |
| *** baojg has joined #openstack-swift | 12:53 | |
| *** baojg has quit IRC | 12:53 | |
| *** baojg has joined #openstack-swift | 12:54 | |
| *** baojg has quit IRC | 12:54 | |
| *** baojg has joined #openstack-swift | 12:54 | |
| *** baojg has quit IRC | 12:55 | |
| *** baojg has joined #openstack-swift | 12:55 | |
| *** baojg has quit IRC | 12:56 | |
| *** baojg has joined #openstack-swift | 12:57 | |
| *** baojg has quit IRC | 12:57 | |
| *** baojg has joined #openstack-swift | 12:57 | |
| *** baojg has quit IRC | 12:58 | |
| *** baojg has joined #openstack-swift | 12:58 | |
| *** baojg has quit IRC | 12:59 | |
| *** baojg has joined #openstack-swift | 12:59 | |
| *** baojg has quit IRC | 13:00 | |
| *** baojg has joined #openstack-swift | 13:00 | |
| *** baojg has quit IRC | 13:00 | |
| *** baojg has joined #openstack-swift | 13:01 | |
| *** baojg has quit IRC | 13:01 | |
| *** baojg has joined #openstack-swift | 13:02 | |
| *** baojg has quit IRC | 13:02 | |
| *** baojg has joined #openstack-swift | 13:02 | |
| *** baojg has quit IRC | 13:03 | |
| *** baojg has joined #openstack-swift | 13:03 | |
| *** baojg has quit IRC | 13:04 | |
| *** baojg has joined #openstack-swift | 13:04 | |
| *** baojg has quit IRC | 13:04 | |
| *** baojg has joined #openstack-swift | 13:05 | |
| *** baojg has quit IRC | 13:05 | |
| *** baojg has joined #openstack-swift | 13:05 | |
| *** baojg has quit IRC | 13:06 | |
| *** baojg has joined #openstack-swift | 13:06 | |
| *** baojg has quit IRC | 13:07 | |
| *** baojg has joined #openstack-swift | 13:12 | |
| *** baojg has quit IRC | 13:13 | |
| *** baojg has joined #openstack-swift | 13:14 | |
| *** baojg has quit IRC | 13:15 | |
| *** baojg has joined #openstack-swift | 13:16 | |
| *** baojg has quit IRC | 13:16 | |
| *** baojg has joined #openstack-swift | 13:17 | |
| *** baojg has quit IRC | 13:17 | |
| *** baojg has joined #openstack-swift | 13:18 | |
| *** baojg has quit IRC | 13:18 | |
| *** baojg has joined #openstack-swift | 13:18 | |
| *** baojg has quit IRC | 13:19 | |
| *** baojg has joined #openstack-swift | 13:19 | |
| *** baojg has quit IRC | 13:20 | |
| *** baojg has joined #openstack-swift | 13:20 | |
| *** baojg has quit IRC | 13:20 | |
| *** baojg has joined #openstack-swift | 13:21 | |
| *** baojg has quit IRC | 13:21 | |
| *** baojg has joined #openstack-swift | 13:22 | |
| *** baojg has quit IRC | 13:22 | |
| *** baojg has joined #openstack-swift | 13:22 | |
| *** baojg has quit IRC | 13:23 | |
| *** baojg has joined #openstack-swift | 13:23 | |
| *** baojg has quit IRC | 13:24 | |
| *** baojg has joined #openstack-swift | 13:24 | |
| *** baojg has quit IRC | 13:24 | |
| *** baojg has joined #openstack-swift | 13:25 | |
| *** baojg has quit IRC | 13:25 | |
| *** baojg has joined #openstack-swift | 13:26 | |
| *** baojg has quit IRC | 13:27 | |
| *** baojg has joined #openstack-swift | 13:27 | |
| *** baojg has quit IRC | 13:28 | |
| *** baojg has joined #openstack-swift | 13:28 | |
| *** baojg has quit IRC | 13:28 | |
| *** baojg has joined #openstack-swift | 13:29 | |
| *** baojg has quit IRC | 13:29 | |
| *** baojg has joined #openstack-swift | 13:29 | |
| *** baojg has quit IRC | 13:30 | |
| *** BjoernT has joined #openstack-swift | 13:32 | |
| *** irclogbot_0 has joined #openstack-swift | 13:37 | |
| *** irclogbot_0 has quit IRC | 13:38 | |
| *** irclogbot_0 has joined #openstack-swift | 14:10 | |
| *** FlorianFa has quit IRC | 14:11 | |
| *** altlogbot_2 has joined #openstack-swift | 14:14 | |
| *** hoonetorg has quit IRC | 14:40 | |
| *** hoonetorg has joined #openstack-swift | 14:56 | |
| *** zaitcev has joined #openstack-swift | 15:07 | |
| *** ChanServ sets mode: +v zaitcev | 15:07 | |
| *** sasregulus has joined #openstack-swift | 15:25 | |
| *** Fidde has quit IRC | 15:26 | |
| *** baojg has joined #openstack-swift | 15:31 | |
| *** baojg has quit IRC | 15:36 | |
| *** tdasilva has joined #openstack-swift | 15:36 | |
| *** ChanServ sets mode: +v tdasilva | 15:36 | |
| *** gyee has joined #openstack-swift | 16:01 | |
| *** aluria has quit IRC | 16:26 | |
| sasregulus | Can I remove the quarantine directory after the corrupt data has replicated back? | 16:40 |
| tdasilva | sasregulus: yes, the quarantine directory needs to be manually cleaned up by operators | 16:51 |
| *** psachin has quit IRC | 17:03 | |
| clayg | when you have auto_shard = true in your container-sharder config... do you still get to use the manage-shard-ranges cli tool to enable containers to shard - or it'll just start going on doing it's own thing? | 17:09 |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: WIP: symlink-backed versioned_writes https://review.opendev.org/633857 | 17:11 |
| *** baojg has joined #openstack-swift | 17:32 | |
| *** baojg has quit IRC | 17:36 | |
| *** zaitcev has quit IRC | 17:53 | |
| *** zaitcev has joined #openstack-swift | 18:04 | |
| *** ChanServ sets mode: +v zaitcev | 18:04 | |
| *** new_student14119 has quit IRC | 18:19 | |
| *** rdejoux has quit IRC | 18:22 | |
| *** zaitcev_ has joined #openstack-swift | 18:31 | |
| *** ChanServ sets mode: +v zaitcev_ | 18:31 | |
| *** zaitcev has quit IRC | 18:35 | |
| *** ccamacho has quit IRC | 18:48 | |
| *** irclogbot_0 has quit IRC | 18:49 | |
| timburke | clayg, it'll just go, as i recall. don't see anything about manage-shard-ranges in the probe test, either | 18:52 |
| *** irclogbot_3 has joined #openstack-swift | 18:53 | |
| clayg | wild, so I don't guess I really understand who does the find_and_replace --enable when auto is turned on 🤷♂️ | 18:53 |
| timburke | replica 0 | 18:54 |
| timburke | which gets dicey when there's a rebalance | 18:54 |
| *** zaitcev_ has quit IRC | 19:32 | |
| *** tesseract has quit IRC | 19:33 | |
| *** zaitcev_ has joined #openstack-swift | 19:44 | |
| *** ChanServ sets mode: +v zaitcev_ | 19:44 | |
| clayg | I think we're falling into the tombstone branch tim -> https://github.com/openstack/swift/blob/2.21.0/swift/container/reconciler.py#L515 | 20:03 |
| clayg | that would match my observations anyway... that I got tombstones for the bs object names | 20:03 |
| timburke | clayg, right -- it's easier to repro the tombstone case, and it certainly seemed like i was seeing the tombstone case in the cluster that prompted me to write up the bug... but i don't think there's anything about this that makes it *exclusive* to the tombstone case | 20:05 |
| clayg | oh i gotcha 👍 | 20:09 |
| *** henriqueof has joined #openstack-swift | 20:38 | |
| timburke | oh yeah, swiftclient release requested: https://review.opendev.org/#/c/670439/ | 20:43 |
| patchbot | patch 670439 - releases - swiftclient 3.8.0 release - 1 patch set | 20:43 |
| *** pcaruana has quit IRC | 20:59 | |
| *** baojg has joined #openstack-swift | 21:33 | |
| *** baojg has quit IRC | 21:38 | |
| *** BjoernT has quit IRC | 21:47 | |
| *** ianychoi has quit IRC | 21:54 | |
| *** guimaluf has quit IRC | 21:57 | |
| openstackgerrit | Merged openstack/swift master: Move calls to self.app outside of error handling https://review.opendev.org/665488 | 22:12 |
| openstackgerrit | Tim Burke proposed openstack/swift master: ring: Track more properties of the ring https://review.opendev.org/670673 | 22:48 |
| openstackgerrit | Tim Burke proposed openstack/swift master: WIP: Allow ring lookups via a service https://review.opendev.org/670674 | 23:10 |
| openstackgerrit | Merged openstack/swift master: py3: Fix header_to_environ_key https://review.opendev.org/664438 | 23:13 |
| openstackgerrit | Merged openstack/swift master: reconciler: Enqueue right work for shard containers https://review.opendev.org/670239 | 23:13 |
| openstackgerrit | Tim Burke proposed openstack/swift master: ring: Track more properties of the ring https://review.opendev.org/670673 | 23:21 |
| openstackgerrit | Tim Burke proposed openstack/swift master: WIP: Allow ring lookups via a service https://review.opendev.org/670674 | 23:21 |
| openstackgerrit | Tim Burke proposed openstack/swift feature/losf: Add tests for vfile.py https://review.opendev.org/666378 | 23:32 |
| *** baojg has joined #openstack-swift | 23:34 | |
| *** baojg has quit IRC | 23:39 | |
| *** zaitcev__ has joined #openstack-swift | 23:42 | |
| *** ChanServ sets mode: +v zaitcev__ | 23:42 | |
| *** zaitcev_ has quit IRC | 23:46 | |
| *** baojg has joined #openstack-swift | 23:52 | |
| openstackgerrit | Tim Burke proposed openstack/swift feature/losf: Add tests for vfile.py https://review.opendev.org/666378 | 23:54 |
| *** baojg has quit IRC | 23:57 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!