| mattoliverau | timburke: sure. if where is a boring session, I'll use the chance to review it ;) | 00:30 |
|---|---|---|
| mattoliverau | *there | 00:30 |
| mattoliverau | sigh, I can't le type | 00:30 |
| timburke | yay! only, wait, you guys keep raving about how great lca is... you won't get to this until you get back to the hotel, will you!? | 00:30 |
| mattoliverau | potentailly that is true, and the penguin dinner is tonight.. so might be evevn later ;) | 00:31 |
| mattoliverau | but will get to it when I can | 00:31 |
| timburke | thanks mattoliverau! | 00:32 |
| zaitcev | dest_brokers = {} # map shard range -> broker | 02:54 |
| zaitcev | dest_brokers[dest_shard_range] = destination | 02:54 |
| zaitcev | TypeError: unhashable type: 'ShardRange' | 02:54 |
| zaitcev | Thanks, Matt | 02:54 |
| zaitcev | (ironically this time) | 02:54 |
| zaitcev | how about I just use id() and move on | 02:55 |
| openstackgerrit | zhouxinyong proposed openstack/slogging master: Change openstack-dev to openstack-discuss in setup.cfg https://review.openstack.org/632890 | 03:06 |
| *** psachin has joined #openstack-swift | 03:09 | |
| openstackgerrit | zhouxinyong proposed openstack/swift-specs master: Change openstack-dev to openstack-discuss in setup.cfg https://review.openstack.org/632896 | 03:14 |
| *** baojg has quit IRC | 03:24 | |
| *** baojg has joined #openstack-swift | 03:43 | |
| *** spsurya has joined #openstack-swift | 04:33 | |
| *** gyee has quit IRC | 05:01 | |
| *** baojg has quit IRC | 05:02 | |
| *** tkajinam has quit IRC | 05:02 | |
| *** e0ne has joined #openstack-swift | 05:03 | |
| *** e0ne has quit IRC | 05:03 | |
| *** baojg has joined #openstack-swift | 05:04 | |
| *** baojg has quit IRC | 05:04 | |
| *** baojg has joined #openstack-swift | 05:05 | |
| openstackgerrit | Pete Zaitcev proposed openstack/swift master: py3: port the container sharder https://review.openstack.org/570320 | 05:07 |
| *** tkajinam has joined #openstack-swift | 05:07 | |
| zaitcev | What a nightmare. It better be ending soon. | 05:10 |
| zaitcev | timburke, mattoliverau: I'm going to look at some of the reviews you guys were talking about recently. You want anything, add me to reviewer list, so the website shows the patch in the list. | 05:15 |
| *** tkajinam has quit IRC | 05:18 | |
| *** tkajinam has joined #openstack-swift | 05:18 | |
| *** zaitcev has quit IRC | 06:10 | |
| *** ccamacho has quit IRC | 06:26 | |
| *** ccamacho has joined #openstack-swift | 07:12 | |
| *** baojg has quit IRC | 07:15 | |
| *** baojg has joined #openstack-swift | 07:16 | |
| *** baojg has quit IRC | 07:16 | |
| *** baojg has joined #openstack-swift | 07:17 | |
| *** baojg has quit IRC | 07:25 | |
| *** baojg has joined #openstack-swift | 07:25 | |
| *** e0ne has joined #openstack-swift | 07:35 | |
| *** tkajinam_ has joined #openstack-swift | 08:04 | |
| *** tkajinam has quit IRC | 08:06 | |
| *** rcernin has quit IRC | 08:29 | |
| *** tkajinam_ has quit IRC | 08:31 | |
| *** baojg has quit IRC | 08:43 | |
| *** baojg has joined #openstack-swift | 08:45 | |
| *** ianychoi has quit IRC | 08:55 | |
| *** hseipp has joined #openstack-swift | 09:14 | |
| *** e0ne has quit IRC | 09:41 | |
| *** e0ne has joined #openstack-swift | 09:56 | |
| *** mikecmpbll has joined #openstack-swift | 10:08 | |
| *** mikecmpb_ has joined #openstack-swift | 10:33 | |
| *** mikecmpbll has quit IRC | 10:34 | |
| *** mikecmpb_ is now known as mikecmpbll | 11:41 | |
| *** ybunker has joined #openstack-swift | 12:28 | |
| ybunker | what's the recommended kernel version for ubuntu 16.04.5 LTS where rsync has no problems or oom at all :) ? | 12:30 |
| *** takamatsu has joined #openstack-swift | 12:31 | |
| *** psachin has quit IRC | 12:32 | |
| *** pcaruana has quit IRC | 12:40 | |
| *** e0ne has quit IRC | 12:59 | |
| *** hseipp has quit IRC | 13:02 | |
| *** pcaruana has joined #openstack-swift | 13:20 | |
| *** e0ne has joined #openstack-swift | 13:20 | |
| *** baojg has quit IRC | 13:28 | |
| admin6 | Hi team, I can’t reach to make swift-dispersion-report works with keystone auth v3. Authentication seems to work but the proxy log en error : proxy: tenant mismatch: AUTH_Dev != Prod (txn: txf960300182c04233abc2e-005c49c0ef) (client_ip: 127.0.0.1). Do you have any suggestion ? | 13:59 |
| *** e0ne has quit IRC | 14:14 | |
| *** baojg has joined #openstack-swift | 14:31 | |
| *** ccamacho has quit IRC | 14:32 | |
| *** e0ne has joined #openstack-swift | 14:34 | |
| *** ccamacho has joined #openstack-swift | 14:35 | |
| *** takamatsu has quit IRC | 15:00 | |
| *** kukacz_ has quit IRC | 15:20 | |
| *** kukacz has joined #openstack-swift | 15:31 | |
| *** pcaruana has quit IRC | 15:35 | |
| *** kukacz has quit IRC | 15:42 | |
| *** ianychoi has joined #openstack-swift | 15:44 | |
| *** kukacz has joined #openstack-swift | 15:46 | |
| *** pcaruana has joined #openstack-swift | 15:55 | |
| *** kukacz has quit IRC | 15:58 | |
| *** kukacz has joined #openstack-swift | 16:00 | |
| *** gyee has joined #openstack-swift | 16:06 | |
| *** e0ne has quit IRC | 16:09 | |
| *** takamatsu has joined #openstack-swift | 16:12 | |
| *** gyee has quit IRC | 16:14 | |
| *** ccamacho has quit IRC | 16:23 | |
| *** kukacz has quit IRC | 16:24 | |
| *** gyee has joined #openstack-swift | 16:25 | |
| *** kukacz has joined #openstack-swift | 16:26 | |
| *** admin6 has left #openstack-swift | 16:31 | |
| *** pcaruana has quit IRC | 16:48 | |
| ybunker | is there a way to throttle the obj-replicator process so I can leave it running all day long with just a few latency impact?, I try setting the rsync_bwlimit to 10 for example on the object-server.conf file, but it seems that is not working at all | 16:49 |
| *** takamatsu has quit IRC | 16:54 | |
| *** zaitcev has joined #openstack-swift | 17:07 | |
| *** ChanServ sets mode: +v zaitcev | 17:07 | |
| *** mikecmpbll has quit IRC | 17:25 | |
| *** ccamacho has joined #openstack-swift | 17:28 | |
| *** e0ne has joined #openstack-swift | 17:59 | |
| *** e0ne has quit IRC | 18:13 | |
| *** takamatsu has joined #openstack-swift | 18:15 | |
| *** e0ne has joined #openstack-swift | 18:27 | |
| DHE | that's a bandwidth usage setting though. a filesystem crawl will still impact disk performance even though rsync isn't really sending anything. | 18:32 |
| *** ccamacho has quit IRC | 18:38 | |
| *** mikecmpbll has joined #openstack-swift | 18:45 | |
| ybunker | got it, and is there a way to handle that? to put some limitation | 18:48 |
| DHE | not to rsync directly, but if you have replicator jobs set up with ionice settings they will be inherited down to rsync on the local machine | 18:58 |
| DHE | on the side receiving the incoming connection you might need to have xinetd run a wrapper to set ionice before launching rsync or such | 18:59 |
| DHE | I don't know if there's a better way | 18:59 |
| *** e0ne has quit IRC | 19:02 | |
| zigo | We have a patch for the encryption issue we found out. | 19:04 |
| zigo | we will open bug and patch soonish. | 19:05 |
| DHE | XFS needs a metadata disk feature... | 19:09 |
| *** takamatsu has quit IRC | 19:09 | |
| *** baojg has quit IRC | 19:22 | |
| zigo | Dhe: that is not the issue. | 19:25 |
| *** e0ne has joined #openstack-swift | 19:26 | |
| DHE | sorry no that's in relation to what ybunker is asking | 19:38 |
| timburke | :-( https://bugs.launchpad.net/swift/+bug/1784753 is still a thing... | 19:40 |
| openstack | Launchpad bug 1784753 in OpenStack Object Storage (swift) "Add ratelimiting to more background daemons" [Undecided,New] | 19:40 |
| timburke | we got some daemons to do more ratelimiting, but still need object-reconstructor and object-replicator ... and idk that we'll be able to throttle rsync real well in the replicator if your cluster's full enough that there tends to be a bunch of objects per suffix... | 19:41 |
| *** mahatic has quit IRC | 19:42 | |
| zigo | DHE: Your thoughts about this one? https://salsa.debian.org/openstack-team/services/swift/blob/debian/rocky/debian/patches/fix-decryption-of-object-if-metadata-is-missing.patch | 19:51 |
| zigo | Please let us test this before we open the bug and review... | 19:51 |
| DHE | oh I don't work for the project. sorry. | 19:52 |
| *** e0ne has quit IRC | 19:54 | |
| zigo | :) | 19:56 |
| timburke | zigo: out of curiosity, what were the two different paths? how did the object come to be at the *wrong* path? | 19:58 |
| timburke | at first glance, patch seems sane enough -- i'd probably want to log a warning or something about the path mismatch. makes me very glad we stuck path in the key_id! | 19:59 |
| *** ybunker has quit IRC | 20:06 | |
| *** rcernin has joined #openstack-swift | 21:02 | |
| *** itlinux has joined #openstack-swift | 21:19 | |
| *** baojg has joined #openstack-swift | 21:23 | |
| *** baojg has quit IRC | 21:27 | |
| zigo | timburke: Somehow, the specs tells that what should be in use for the decryption, is the object name. However, it's not what's in use when encrypting, it's a truncated version of the object name (or path...). | 22:05 |
| zigo | timburke: When the object name is large enough, then they don't match. | 22:06 |
| zigo | That's the first problem. | 22:06 |
| timburke | eep! | 22:06 |
| zigo | The 2nd one, is that in some cases, we have no metadata recorded in the object itself. | 22:06 |
| zigo | It appears it's only saved in the container. | 22:06 |
| zigo | We haven't figured why exactly. | 22:07 |
| zigo | But it doesn't really mater, because even if we figure out where the issue is, we still need to have the (soon) proposed patch in order to recover wrongly written objects. | 22:08 |
| timburke | no metadata at all? or it's missing some of the encryption metadata? if you run swift-object-info on one of the affected .data files, what does it say? | 22:11 |
| *** rcernin has quit IRC | 22:11 | |
| *** rcernin has joined #openstack-swift | 22:12 | |
| timburke | makes me happy we had the outcome we did on the review: https://review.openstack.org/#/c/328208/11/swift/common/middleware/keymaster.py@85 | 22:20 |
| patchbot | patch 328208 - swift (feature/crypto-review) - Enable object body and metadata encryption (MERGED) - 16 patch sets | 22:20 |
| *** itlinux has quit IRC | 22:38 | |
| *** baojg has joined #openstack-swift | 22:53 | |
| *** tkajinam has joined #openstack-swift | 22:57 | |
| *** spsurya has quit IRC | 23:04 | |
| *** baojg has quit IRC | 23:55 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!