| openstackgerrit | zhufl proposed openstack/swift master: Use assertIn to check whether substring is in str https://review.openstack.org/652583 | 00:58 |
|---|---|---|
| *** baojg has joined #openstack-swift | 01:35 | |
| *** threestrands has joined #openstack-swift | 01:47 | |
| *** baojg has quit IRC | 02:53 | |
| *** zaitcev_ has quit IRC | 03:49 | |
| *** zaitcev_ has joined #openstack-swift | 04:00 | |
| *** ChanServ sets mode: +v zaitcev_ | 04:00 | |
| *** zaitcev__ has joined #openstack-swift | 04:17 | |
| *** ChanServ sets mode: +v zaitcev__ | 04:17 | |
| *** zaitcev_ has quit IRC | 04:21 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: py3: port obj/reconstructor https://review.openstack.org/652819 | 04:44 |
| *** e0ne has joined #openstack-swift | 04:53 | |
| *** e0ne has quit IRC | 04:56 | |
| *** gyee has quit IRC | 05:08 | |
| *** e0ne has joined #openstack-swift | 05:21 | |
| *** e0ne has quit IRC | 05:25 | |
| *** baojg has joined #openstack-swift | 05:37 | |
| *** pcaruana has joined #openstack-swift | 06:26 | |
| *** zaitcev_ has joined #openstack-swift | 06:26 | |
| *** ChanServ sets mode: +v zaitcev_ | 06:26 | |
| *** zaitcev__ has quit IRC | 06:30 | |
| *** hogepodge has quit IRC | 06:32 | |
| *** nottrobin has quit IRC | 06:33 | |
| *** zaitcev_ has quit IRC | 06:39 | |
| *** nottrobin has joined #openstack-swift | 06:41 | |
| *** nottrobin has quit IRC | 06:45 | |
| *** zaitcev_ has joined #openstack-swift | 06:52 | |
| *** ChanServ sets mode: +v zaitcev_ | 06:52 | |
| *** nottrobin has joined #openstack-swift | 06:53 | |
| *** hogepodge has joined #openstack-swift | 06:56 | |
| openstackgerrit | Matthew Oliver proposed openstack/swift master: py3: Port the tempurl middleware https://review.openstack.org/652928 | 07:01 |
| *** e0ne has joined #openstack-swift | 07:06 | |
| *** e0ne has quit IRC | 07:09 | |
| *** rdejoux has joined #openstack-swift | 07:12 | |
| *** e0ne has joined #openstack-swift | 07:16 | |
| *** rcernin has quit IRC | 07:20 | |
| *** ccamacho has joined #openstack-swift | 07:48 | |
| *** e0ne has quit IRC | 07:52 | |
| *** threestrands has quit IRC | 07:58 | |
| *** e0ne has joined #openstack-swift | 07:58 | |
| *** hseipp has joined #openstack-swift | 07:59 | |
| *** e0ne has quit IRC | 07:59 | |
| *** gkadam has joined #openstack-swift | 08:09 | |
| *** gkadam has quit IRC | 08:10 | |
| *** e0ne has joined #openstack-swift | 08:10 | |
| *** tkajinam has quit IRC | 08:24 | |
| openstackgerrit | Merged openstack/swift master: Use assertIn to check whether substring is in str https://review.openstack.org/652583 | 08:25 |
| *** zaitcev__ has joined #openstack-swift | 08:34 | |
| *** ChanServ sets mode: +v zaitcev__ | 08:34 | |
| *** zaitcev_ has quit IRC | 08:37 | |
| *** rcernin has joined #openstack-swift | 10:02 | |
| *** baojg has quit IRC | 10:56 | |
| *** baojg has joined #openstack-swift | 10:57 | |
| *** baojg has quit IRC | 11:01 | |
| *** pcaruana has quit IRC | 11:17 | |
| *** baojg has joined #openstack-swift | 11:23 | |
| *** ccamacho has quit IRC | 11:34 | |
| *** zaitcev_ has joined #openstack-swift | 11:37 | |
| *** ChanServ sets mode: +v zaitcev_ | 11:37 | |
| *** zaitcev__ has quit IRC | 11:40 | |
| *** nguyenhai has joined #openstack-swift | 11:47 | |
| *** nguyenhai has quit IRC | 11:52 | |
| *** pcaruana has joined #openstack-swift | 12:05 | |
| *** baojg has quit IRC | 12:05 | |
| *** ybunker has joined #openstack-swift | 12:39 | |
| ybunker | hi all, question... how can i recreate the ring files of an already running cluster? | 12:42 |
| rledisez | hi ybunker. what do you have now? *.builder or *.ring.gz? | 12:49 |
| ybunker | rledisez: hi :-), i have both files {account|container|object}.builder and {account|container|object}.ring.gz | 12:50 |
| rledisez | ybunker: if you have both builder and ring.gz, you have everything you need. what are you thinking you need to recreate something? | 12:52 |
| rledisez | s/what/why/ | 12:52 |
| ybunker | rledisez: because i need to manually delete some replica partitions on a 100% fully node drives :-(, so first im going to look for spare partitions, and then delete them with rm -rf on the fully node, and i read that it then needs to 'recreate' the ring so the files started to move around again, since the reference of the replicas are going to be missed | 12:54 |
| *** hseipp has quit IRC | 13:02 | |
| *** zaitcev__ has joined #openstack-swift | 13:07 | |
| *** ChanServ sets mode: +v zaitcev__ | 13:07 | |
| *** pcaruana has quit IRC | 13:07 | |
| *** zaitcev_ has quit IRC | 13:11 | |
| rledisez | ybunker: i don't think you really want to do that. I may not have the whole context. Do you have freespace in your cluster? If so, you can just reduce the weight of the full devices and let swift re-organize (rebalance) the data | 13:15 |
| ybunker | rledisez: the problem is that we actually have a lot of free space on other nodes, but since fully drives are at 100% of used space, I have understood that 100% full drives are unable to move data out of themselves, because at a high level, they must have storage space to commit transactions and such before they can move data away from themselves. So we are like stuck at this point :-( | 13:17 |
| *** ccamacho has joined #openstack-swift | 13:20 | |
| rledisez | I would go for: | 13:20 |
| rledisez | 1/ stop the object-server on the full devices (so that new data won't come) | 13:20 |
| rledisez | 2/ set handoffs_first = True in the replicator configuration | 13:20 |
| rledisez | 3/ make some free space (few MB will be ok) by removing "useless" data (eg: quarantine). otherwise, mv some data to another drive. as soon as you have freespace enought, you can restore | 13:20 |
| rledisez | 4/ rebalance | 13:21 |
| ybunker | rledisez: i've already set the handsoff_first and handsoff_delete on the fully nodes, and in fact there is 10G of free space on the drives.. but, it don't actually move anything from those drives and they are still at 100% :-( | 13:27 |
| *** pcaruana has joined #openstack-swift | 13:33 | |
| *** thyrst has quit IRC | 13:42 | |
| ybunker | rledisez: also, i can't leave the object-replicator daemons running 24h because general latency of the cluster goes to the roof :-(, so I've setup to run those daemons on an 4h windows at night | 13:43 |
| *** zaitcev_ has joined #openstack-swift | 13:44 | |
| *** ChanServ sets mode: +v zaitcev_ | 13:44 | |
| *** thyrst has joined #openstack-swift | 13:44 | |
| *** zaitcev__ has quit IRC | 13:48 | |
| *** zaitcev_ has quit IRC | 13:56 | |
| *** zaitcev_ has joined #openstack-swift | 14:10 | |
| *** ChanServ sets mode: +v zaitcev_ | 14:10 | |
| *** mattia has joined #openstack-swift | 14:22 | |
| *** rcernin has quit IRC | 14:27 | |
| *** baojg has joined #openstack-swift | 14:35 | |
| *** nottrobin has quit IRC | 15:03 | |
| *** nottrobin has joined #openstack-swift | 15:03 | |
| *** zaitcev__ has joined #openstack-swift | 15:20 | |
| *** ChanServ sets mode: +v zaitcev__ | 15:20 | |
| *** zaitcev_ has quit IRC | 15:24 | |
| *** pcaruana has quit IRC | 15:27 | |
| *** pcaruana has joined #openstack-swift | 15:30 | |
| *** e0ne has quit IRC | 15:31 | |
| *** pcaruana has quit IRC | 15:40 | |
| *** pcaruana has joined #openstack-swift | 15:46 | |
| *** hseipp has joined #openstack-swift | 15:49 | |
| *** gyee has joined #openstack-swift | 15:50 | |
| *** pcaruana has quit IRC | 15:55 | |
| *** pcaruana has joined #openstack-swift | 16:02 | |
| *** hseipp has quit IRC | 16:06 | |
| *** ybunker has quit IRC | 16:07 | |
| *** pcaruana has quit IRC | 16:08 | |
| *** zaitcev__ has quit IRC | 16:09 | |
| *** hseipp has joined #openstack-swift | 16:10 | |
| *** zaitcev__ has joined #openstack-swift | 16:23 | |
| *** ChanServ sets mode: +v zaitcev__ | 16:23 | |
| *** hseipp has quit IRC | 16:24 | |
| *** pcaruana has joined #openstack-swift | 16:36 | |
| *** ybunker has joined #openstack-swift | 16:39 | |
| *** ccamacho has quit IRC | 16:53 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/python-swiftclient master: Dropping the py35 testing https://review.openstack.org/653046 | 16:58 |
| *** itlinux has joined #openstack-swift | 16:59 | |
| *** zaitcev__ is now known as zaitcev | 17:23 | |
| *** baojg has quit IRC | 17:24 | |
| *** e0ne has joined #openstack-swift | 17:32 | |
| *** e0ne has quit IRC | 17:49 | |
| *** zaitcev has quit IRC | 17:56 | |
| *** zaitcev has joined #openstack-swift | 18:08 | |
| *** ChanServ sets mode: +v zaitcev | 18:08 | |
| *** rdejoux has quit IRC | 18:21 | |
| *** irclogbot_1 has quit IRC | 18:39 | |
| *** irclogbot_2 has joined #openstack-swift | 18:41 | |
| *** baojg has joined #openstack-swift | 19:25 | |
| *** itlinux has quit IRC | 19:30 | |
| *** baojg has quit IRC | 19:30 | |
| *** itlinux has joined #openstack-swift | 19:34 | |
| *** zaitcev has quit IRC | 19:45 | |
| clayg | do we really not have a helper to turn a swob.Response into a swob.HTTPException? | 19:50 |
| *** ybunker has quit IRC | 19:53 | |
| *** zaitcev has joined #openstack-swift | 19:58 | |
| *** ChanServ sets mode: +v zaitcev | 19:58 | |
| *** pcaruana has quit IRC | 20:19 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: Move from py35 to py37 for gating py3 jobs https://review.openstack.org/649485 | 20:54 |
| *** zaitcev_ has joined #openstack-swift | 21:11 | |
| *** ChanServ sets mode: +v zaitcev_ | 21:11 | |
| *** zaitcev has quit IRC | 21:14 | |
| *** baojg has joined #openstack-swift | 21:26 | |
| *** baojg has quit IRC | 21:30 | |
| *** itlinux has quit IRC | 21:53 | |
| *** itlinux has joined #openstack-swift | 22:07 | |
| *** zaitcev_ has quit IRC | 22:38 | |
| *** rcernin has joined #openstack-swift | 22:38 | |
| *** zaitcev_ has joined #openstack-swift | 22:52 | |
| *** ChanServ sets mode: +v zaitcev_ | 22:52 | |
| *** tkajinam has joined #openstack-swift | 22:54 | |
| *** guimaluf has quit IRC | 23:21 | |
| *** baojg has joined #openstack-swift | 23:27 | |
| *** baojg has quit IRC | 23:31 | |
| *** itlinux has quit IRC | 23:35 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!