*** diablo_rojo has quit IRC | 01:16 | |
*** BjoernT has joined #openstack-swift | 02:20 | |
*** BjoernT_ has joined #openstack-swift | 02:29 | |
*** BjoernT has quit IRC | 02:31 | |
*** e0ne has joined #openstack-swift | 03:15 | |
*** e0ne has quit IRC | 03:19 | |
*** psachin has joined #openstack-swift | 03:26 | |
*** BjoernT_ has quit IRC | 03:26 | |
*** BjoernT has joined #openstack-swift | 03:27 | |
*** BjoernT has quit IRC | 03:28 | |
*** BjoernT_ has joined #openstack-swift | 03:29 | |
*** BjoernT_ has quit IRC | 03:34 | |
*** BjoernT has joined #openstack-swift | 03:34 | |
*** BjoernT has quit IRC | 04:10 | |
*** BjoernT has joined #openstack-swift | 04:12 | |
*** BjoernT has quit IRC | 04:26 | |
*** pcaruana has joined #openstack-swift | 06:22 | |
*** ccamacho has joined #openstack-swift | 07:09 | |
*** rdejoux has joined #openstack-swift | 07:36 | |
*** mvkr has quit IRC | 07:46 | |
*** mvkr has joined #openstack-swift | 08:00 | |
baffle | Is there a way to change the zone of a device, similar to how you can change ports/IPs with swift-ring-builder set_info? Or should one instead remove/rebalance/readd device? The drives are in in the wrong zone but correct server, and have a weight of 0 and no data. | 08:12 |
---|---|---|
*** pcaruana has quit IRC | 08:13 | |
*** pcaruana has joined #openstack-swift | 08:18 | |
*** tkajinam has quit IRC | 08:21 | |
*** e0ne has joined #openstack-swift | 08:55 | |
*** pcaruana has quit IRC | 11:11 | |
*** pcaruana has joined #openstack-swift | 11:42 | |
*** NM has joined #openstack-swift | 12:37 | |
tdasilva | baffle: currently you have to remove/readd device. Coincidently there's a patch about to be merged to add the options to change zone and region: patch 291990 | 12:49 |
patchbot | https://review.opendev.org/#/c/291990/ - swift - Add commands to ring-builder to change region / zone - 5 patch sets | 12:49 |
*** BjoernT has joined #openstack-swift | 13:18 | |
clayg | so I think i'll be playing with trying to make swift-get-nodes and other tools that take paths unquote command line args by default and support --raw for when needed | 13:34 |
*** NM has quit IRC | 13:40 | |
clayg | or maybe I'll look at the query stuff some more, timburke maybe dropped some knowledge p 682138 | 13:46 |
patchbot | https://review.opendev.org/#/c/682138/ - swift - Allow internal clients to use null namespace - 7 patch sets | 13:46 |
*** NM has joined #openstack-swift | 13:47 | |
*** jrosser has quit IRC | 14:14 | |
*** jrosser has joined #openstack-swift | 14:15 | |
*** baffle has quit IRC | 14:20 | |
*** csmart has quit IRC | 14:20 | |
*** diablo_rojo has joined #openstack-swift | 14:54 | |
*** rpittau has joined #openstack-swift | 14:59 | |
*** NM has quit IRC | 15:10 | |
openstackgerrit | Merged openstack/swift master: Add commands to ring-builder to change region / zone https://review.opendev.org/291990 | 15:16 |
*** zaitcev has joined #openstack-swift | 16:00 | |
*** ChanServ sets mode: +v zaitcev | 16:00 | |
*** rpittau is now known as rpittau|adk | 16:02 | |
*** rpittau|adk is now known as rpittau|afk | 16:02 | |
*** rdejoux has quit IRC | 16:13 | |
*** e0ne has quit IRC | 16:17 | |
*** gyee has joined #openstack-swift | 16:22 | |
*** gyee_ has joined #openstack-swift | 16:22 | |
*** gyee_ has quit IRC | 16:23 | |
*** gyee has quit IRC | 16:23 | |
*** gyee has joined #openstack-swift | 16:23 | |
*** csmart has joined #openstack-swift | 16:34 | |
*** NM has joined #openstack-swift | 16:39 | |
*** NM has quit IRC | 17:20 | |
*** pcaruana has quit IRC | 17:21 | |
*** psachin has quit IRC | 17:22 | |
*** baffle has joined #openstack-swift | 17:25 | |
*** pcaruana has joined #openstack-swift | 18:00 | |
*** e0ne has joined #openstack-swift | 18:28 | |
*** gyee has quit IRC | 18:33 | |
*** spartakos has joined #openstack-swift | 19:16 | |
*** gyee has joined #openstack-swift | 19:18 | |
*** NM has joined #openstack-swift | 19:19 | |
*** pcaruana has quit IRC | 19:32 | |
*** spartakos has quit IRC | 19:50 | |
*** NM has quit IRC | 19:54 | |
openstackgerrit | Clay Gerrard proposed openstack/swift master: Allow internal clients to use null namespace https://review.opendev.org/682138 | 19:58 |
clayg | ^ timburke i don't get it - it *really* looks like tests are confirming that all the ways I can cook up to use %00 in the existing query params (prefix, marker, delimiter) it's working | 19:59 |
*** spartakos has joined #openstack-swift | 19:59 | |
clayg | it seems like the sqlite query_args ? substitution *has* to be interpolating the null byte correctly - i'm really confused | 19:59 |
clayg | I'll keep writing more tests | 19:59 |
*** spartakos has quit IRC | 20:02 | |
*** NM has joined #openstack-swift | 20:48 | |
*** tonyb has joined #openstack-swift | 20:53 | |
*** diablo_rojo has quit IRC | 20:56 | |
*** NM has quit IRC | 21:27 | |
*** BjoernT has quit IRC | 21:35 | |
openstackgerrit | Merged openstack/swift master: Start building py3 docker images https://review.opendev.org/686419 | 21:57 |
mattoliverau | morning | 22:16 |
*** e0ne has quit IRC | 22:26 | |
*** BjoernT has joined #openstack-swift | 22:37 | |
*** BjoernT has quit IRC | 22:42 | |
*** rcernin has joined #openstack-swift | 23:08 | |
donnyd | timburke: How could I measure the object-exipre cycle times? | 23:39 |
*** NM has joined #openstack-swift | 23:41 | |
timburke | donnyd, there's an object.recon file with a bunch of interesting stats in JSON. default location is /var/cache/swift/object.recon | 23:46 |
timburke | cycle time should be in the object_expiration_pass key (which isn't the most obvious name... :-/) | 23:46 |
donnyd | So do I need to periodically parse this file? | 23:50 |
donnyd | also I blew up my first disk but am not quite sure how to replace it | 23:50 |
donnyd | like I know how to replace a disk in a linux machine.. but I am unsure what I need to do with swift when I replace the busted disk | 23:51 |
*** hoonetorg has quit IRC | 23:53 | |
timburke | yeah, periodic parsing; once every five minutes or so is probably plenty? | 23:58 |
timburke | as far as replacing a disk -- if you've got the replacement on hand, swap it out, format it as XFS, mount it where the busted disk used to be mounted. shouldn't even need a ring change | 23:59 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!