| *** renich has joined #openstack-swift | 03:33 | |
| *** psachin has joined #openstack-swift | 03:35 | |
| *** evrardjp has quit IRC | 04:33 | |
| *** evrardjp has joined #openstack-swift | 04:33 | |
| *** renich has quit IRC | 05:18 | |
| openstackgerrit | zhangboye proposed openstack/swift master: Add py38 package metadata https://review.opendev.org/730264 | 07:31 |
|---|---|---|
| *** tkajinam_ has quit IRC | 08:05 | |
| *** rcernin has quit IRC | 12:39 | |
| *** onovy has quit IRC | 12:43 | |
| *** psachin has quit IRC | 14:26 | |
| *** psachin has joined #openstack-swift | 14:35 | |
| *** psachin has quit IRC | 14:42 | |
| *** gyee has joined #openstack-swift | 15:17 | |
| *** psachin has joined #openstack-swift | 16:10 | |
| *** psachin has quit IRC | 16:17 | |
| *** psachin has joined #openstack-swift | 16:20 | |
| clayg | timburke: the updated test with that patches the batch size looks brilliant - i don't know how to easily apply it form the paste - can you push over, i think the change is more or less ready to go | 16:36 |
| *** psachin has quit IRC | 16:38 | |
| *** psachin has joined #openstack-swift | 16:41 | |
| timburke | clayg, 👍 -- i'll also do something similar for the container test. what do you think of the updated batch size? | 16:48 |
| *** renich has joined #openstack-swift | 16:50 | |
| *** psachin has quit IRC | 16:52 | |
| clayg | 30K seems big, let's just use a listing? I see almost no good reason for reclaim to be fast? | 16:53 |
| clayg | If you have lots of small databases or not very many deleted items the batch size doesn't matter | 16:54 |
| clayg | if you have a big database laying on the disk with a lot of tombstone rows - let one gt spin on it for a minute or two, NBD | 16:54 |
| openstackgerrit | Merged openstack/swift master: Bump up timeouts for DSVM and probe test jobs https://review.opendev.org/729778 | 16:57 |
| *** psachin has joined #openstack-swift | 16:59 | |
| *** psachin has quit IRC | 17:18 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: Breakup reclaim into batches https://review.opendev.org/727876 | 18:38 |
| openstackgerrit | Tim Burke proposed openstack/swift master: Breakup reclaim into batches https://review.opendev.org/727876 | 18:44 |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: Breakup reclaim into batches https://review.opendev.org/727876 | 19:38 |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: Breakup reclaim into batches https://review.opendev.org/727876 | 19:39 |
| openstackgerrit | Tim Burke proposed openstack/swift master: Breakup reclaim into batches https://review.opendev.org/727876 | 19:53 |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: WIP: Fix Versioned Object If-Match https://review.opendev.org/730414 | 20:10 |
| clayg | nah, this can't be right lp bug #1880239 | 20:17 |
| openstack | Launchpad bug 1880239 in OpenStack Object Storage (swift) "x-versions-enabled objects do not support if-matchc conditional GET" [Undecided,New] https://launchpad.net/bugs/1880239 | 20:17 |
| timburke | clayg, did you maybe forget to restart your proxy after hopping off p 724393 ? | 20:19 |
| patchbot | https://review.opendev.org/#/c/724393/ - swift - versioning: Have versioning symlinks make pre-auth... - 3 patch sets | 20:19 |
| clayg | @timburke for some reason I thought the select query would return the last row regardless | 20:19 |
| timburke | clayg, nope -- if there are fewer than OFFSET rows, it won't return anything | 20:20 |
| clayg | yes, that forgot to reset proxy - happy Friday! | 20:20 |
| timburke | the test will be great, though! we should probably add a similar one for Range headers | 20:21 |
| clayg | oh that makes sense - the symlink change only effects reserved namespace objects - so that's why only versoined was effected anyway | 20:21 |
| timburke | clayg, so on the header pass-through -- is there anything that we're actively *worried* about? like, "i know this header would be super bad if it got through"? i felt like since we're limited to GETs and HEADs, the danger was pretty low | 20:25 |
| clayg | 100% agreed - it wasn't even overly cautious - I was positive some test would fail - i just wanted to started with the simplest thing that could possibly work before I started to think about it | 20:26 |
| clayg | I mean, I get a *little* paranoid anytime I'm useing a preauth request but I think the helpers are there to help so passing in orig_req.headers makes sense to me? I think? | 20:28 |
| zaitcev | Awww | 20:37 |
| zaitcev | Decided that today was the day to review patch 727876 | 20:38 |
| patchbot | https://review.opendev.org/#/c/727876/ - swift - Breakup reclaim into batches - 9 patch sets | 20:38 |
| zaitcev | ... it's already approved | 20:38 |
| zaitcev | Why was I born into this cruel world. | 20:38 |
| zaitcev | Good job though | 20:39 |
| *** renich has quit IRC | 20:56 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: swift-dsvm: Enable s3api https://review.opendev.org/722550 | 21:08 |
| openstackgerrit | Tim Burke proposed openstack/swift master: Breakup reclaim into batches https://review.opendev.org/727876 | 21:16 |
| openstackgerrit | Tim Burke proposed openstack/swift master: swift-dsvm: Enable s3api https://review.opendev.org/722550 | 21:20 |
| *** renich has joined #openstack-swift | 21:28 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: swift-dsvm: Enable s3api https://review.opendev.org/722550 | 21:29 |
| openstackgerrit | Tim Burke proposed openstack/swift master: swift-dsvm: Enable s3api https://review.opendev.org/722550 | 21:44 |
| openstackgerrit | Merged openstack/swift master: Add py38 package metadata https://review.opendev.org/730264 | 22:15 |
| *** djhankb has quit IRC | 22:23 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: swift-dsvm: Enable s3api https://review.opendev.org/722550 | 22:23 |
| *** djhankb has joined #openstack-swift | 22:23 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: swift-dsvm: Enable s3api https://review.opendev.org/722550 | 22:44 |
| timburke | :-( i guess my unicode users/secrets were a little *too* crazy | 22:45 |
| openstackgerrit | Tim Burke proposed openstack/swift master: swift-dsvm: Enable s3api https://review.opendev.org/722550 | 23:05 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!