| *** nellysmitt has quit IRC | 00:00 | |
| *** ho has joined #openstack-swift | 00:04 | |
| *** IRTermite has joined #openstack-swift | 00:04 | |
| ho | good morining guys! | 00:04 |
|---|---|---|
| *** Nadeem has quit IRC | 00:06 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 00:08 |
| *** reed has quit IRC | 00:08 | |
| *** abhirc has joined #openstack-swift | 00:16 | |
| notmyname | hondak: hi! | 00:21 |
| notmyname | ho: hi! | 00:21 |
| notmyname | tab complete fail | 00:21 |
| *** annegent_ has joined #openstack-swift | 00:22 | |
| *** annegent_ has quit IRC | 00:24 | |
| *** lcurtis has quit IRC | 00:24 | |
| *** lcurtis has joined #openstack-swift | 00:25 | |
| mattoliverau | ho: morning | 00:32 |
| ho | notmyname: mattoliverau: hi! | 00:38 |
| ho | notmyname: How about the weather in SFO? | 00:38 |
| notmyname | ho: mattoliverau: when do you leave? I'm guessing pretty soon | 00:39 |
| notmyname | ho: http://forecast.io/#/f/37.7771,-122.4196 | 00:39 |
| ho | notmyname: I will leave tomorrow. | 00:39 |
| mattoliverau | notmyname: I'm on a 11:45 flight out tomorrow (Saturday) and arrive at 10:55 in SFO on saturday :) | 00:39 |
| notmyname | looks like rain this weekend and then pretty normal otherwise. cool | 00:39 |
| notmyname | mattoliverau: what will you do with the extra hour?! | 00:40 |
| mattoliverau | well it's 36 degrees today, then down to 17ish in SFO, so that will be a shock | 00:40 |
| mattoliverau | I don't know, 16 hours of flying (18 hours of transit) and still get an extra hour :P | 00:41 |
| ho | notmyname: thanks for the info. need to have an umbrella | 00:43 |
| *** imkarrer has quit IRC | 00:43 | |
| *** lcurtis has quit IRC | 00:43 | |
| *** dmorita has joined #openstack-swift | 00:45 | |
| notmyname | ho: it normally doesn't rain very hard here. | 00:45 |
| ho | ho: people don't use umbrellas? | 00:46 |
| occupant | yeah, this is supposed to be another superstorm like the one we got in december | 00:57 |
| occupant | I'd bought rain boots just before that happened. Perfect timing. | 00:57 |
| *** zul has quit IRC | 01:03 | |
| *** rmcall has quit IRC | 01:09 | |
| *** abhirc has quit IRC | 01:13 | |
| *** tellesnobrega_ has joined #openstack-swift | 01:25 | |
| *** openstack has joined #openstack-swift | 01:34 | |
| *** abhirc has joined #openstack-swift | 01:44 | |
| *** nellysmitt has joined #openstack-swift | 01:56 | |
| *** nellysmitt has quit IRC | 02:00 | |
| *** fandi has quit IRC | 02:11 | |
| *** lcurtis has joined #openstack-swift | 02:48 | |
| *** nellysmitt has joined #openstack-swift | 03:02 | |
| *** nellysmitt has quit IRC | 03:06 | |
| *** panbalag has quit IRC | 03:13 | |
| *** bill_az_ has quit IRC | 03:34 | |
| *** redbo has quit IRC | 03:55 | |
| *** early has quit IRC | 04:04 | |
| *** tellesnobrega_ has quit IRC | 04:07 | |
| *** ppai has joined #openstack-swift | 04:09 | |
| *** redbo has joined #openstack-swift | 04:14 | |
| *** ChanServ sets mode: +v redbo | 04:14 | |
| *** ppai has quit IRC | 04:19 | |
| *** kota_ has joined #openstack-swift | 04:22 | |
| *** early has joined #openstack-swift | 04:25 | |
| *** SkyRocknRoll has joined #openstack-swift | 04:26 | |
| *** early has quit IRC | 04:43 | |
| *** EmilienM|afk has quit IRC | 04:45 | |
| *** EmilienM has joined #openstack-swift | 04:46 | |
| *** nellysmitt has joined #openstack-swift | 05:06 | |
| *** nellysmitt has quit IRC | 05:11 | |
| openstackgerrit | Daisuke Morita proposed openstack/swift: Improvement of logging for PolicyError https://review.openstack.org/141238 | 05:34 |
| *** echevemaster has quit IRC | 05:42 | |
| *** jasondotstar has quit IRC | 05:43 | |
| *** early has joined #openstack-swift | 05:47 | |
| *** early has quit IRC | 05:50 | |
| *** lcurtis has quit IRC | 05:51 | |
| *** anticw has quit IRC | 06:01 | |
| *** anticw has joined #openstack-swift | 06:02 | |
| *** early has joined #openstack-swift | 06:06 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Imported Translations from Transifex https://review.openstack.org/153477 | 06:10 |
| *** early has quit IRC | 06:12 | |
| *** early has joined #openstack-swift | 06:23 | |
| *** zaitcev has quit IRC | 06:31 | |
| *** openstackgerrit has quit IRC | 06:35 | |
| *** openstackgerrit has joined #openstack-swift | 06:35 | |
| *** early has quit IRC | 06:43 | |
| *** early has joined #openstack-swift | 06:49 | |
| *** nellysmitt has joined #openstack-swift | 07:07 | |
| *** chlong has quit IRC | 07:12 | |
| *** nellysmitt has quit IRC | 07:12 | |
| *** chlong has joined #openstack-swift | 07:13 | |
| *** kota_ has quit IRC | 07:16 | |
| *** nshaikh has joined #openstack-swift | 07:39 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift: Efficient Replication for Distributed Regions https://review.openstack.org/99824 | 07:52 |
| *** TobiasE has quit IRC | 07:58 | |
| *** TobiasE has joined #openstack-swift | 08:00 | |
| *** rledisez has joined #openstack-swift | 08:02 | |
| *** nellysmitt has joined #openstack-swift | 08:15 | |
| *** tdasilva has joined #openstack-swift | 08:39 | |
| openstackgerrit | Thiago da Silva proposed openstack/swift: WIP: single-process https://review.openstack.org/149329 | 08:40 |
| tdasilva | eikke: hi, just uploaded your changes ^^^ | 08:41 |
| *** tdasilva has quit IRC | 08:43 | |
| *** foexle has joined #openstack-swift | 08:44 | |
| *** mmcardle has joined #openstack-swift | 08:45 | |
| *** acoles_away is now known as acoles | 08:46 | |
| *** chlong has quit IRC | 08:53 | |
| acoles | cschwede: morning! this may be a quick review if you have chance https://review.openstack.org/#/c/152903/ | 08:56 |
| *** jordanP has joined #openstack-swift | 08:58 | |
| *** TobiasE has quit IRC | 09:00 | |
| acoles | notmyname: good move adding that comment around delay_auth_decision. frustrates me that we are maintaining that sample config when the defaults don't even work, and cannot link to a keystonemiddleware doc for the other options, last time i looked it was out of date :( | 09:01 |
| acoles | notmyname: but its friday so i'm not grumpy | 09:01 |
| *** ho has quit IRC | 09:22 | |
| *** silor has joined #openstack-swift | 09:31 | |
| *** joeljwright has joined #openstack-swift | 09:37 | |
| openstackgerrit | Merged openstack/swift: Imported Translations from Transifex https://review.openstack.org/153477 | 09:39 |
| *** early has quit IRC | 09:51 | |
| *** bus-104 has joined #openstack-swift | 10:05 | |
| *** chlong has joined #openstack-swift | 10:08 | |
| *** aix has joined #openstack-swift | 10:16 | |
| *** tellesnobrega_ has joined #openstack-swift | 10:18 | |
| *** tellesnobrega_ has quit IRC | 10:28 | |
| *** tellesnobrega_ has joined #openstack-swift | 10:37 | |
| *** dmorita has quit IRC | 10:39 | |
| *** tellesnobrega_ has quit IRC | 10:45 | |
| *** silor1 has joined #openstack-swift | 10:48 | |
| *** silor has quit IRC | 10:50 | |
| *** silor has joined #openstack-swift | 10:53 | |
| *** silor1 has quit IRC | 10:54 | |
| *** silor has quit IRC | 11:07 | |
| *** reed has joined #openstack-swift | 11:21 | |
| *** zigo has quit IRC | 11:42 | |
| *** TobiasE has joined #openstack-swift | 11:54 | |
| *** torgomatic has quit IRC | 11:59 | |
| *** zul has joined #openstack-swift | 11:59 | |
| *** zul has quit IRC | 12:01 | |
| *** zul has joined #openstack-swift | 12:01 | |
| *** torgomatic has joined #openstack-swift | 12:01 | |
| *** ChanServ sets mode: +v torgomatic | 12:01 | |
| *** nellysmi_ has joined #openstack-swift | 12:04 | |
| *** TobiasE has left #openstack-swift | 12:05 | |
| *** nellysmitt has quit IRC | 12:06 | |
| *** jroll has quit IRC | 12:06 | |
| *** sileht_ has joined #openstack-swift | 12:10 | |
| *** panbalag has joined #openstack-swift | 12:17 | |
| *** geaaru has joined #openstack-swift | 12:18 | |
| *** zul has quit IRC | 12:19 | |
| *** esmute has quit IRC | 12:19 | |
| *** yuanz has quit IRC | 12:19 | |
| *** ipolyzos has quit IRC | 12:19 | |
| *** dmsimard_away has quit IRC | 12:19 | |
| *** anderstj has quit IRC | 12:19 | |
| *** CrackerJackMack has quit IRC | 12:19 | |
| *** glange has quit IRC | 12:19 | |
| *** sileht has quit IRC | 12:19 | |
| *** chrisnelson has quit IRC | 12:19 | |
| *** cppforlife_ has quit IRC | 12:19 | |
| *** sileht_ is now known as sileht | 12:19 | |
| *** jroll has joined #openstack-swift | 12:21 | |
| *** zul has joined #openstack-swift | 12:21 | |
| *** esmute has joined #openstack-swift | 12:21 | |
| *** yuanz has joined #openstack-swift | 12:21 | |
| *** ipolyzos has joined #openstack-swift | 12:21 | |
| *** dmsimard_away has joined #openstack-swift | 12:21 | |
| *** anderstj has joined #openstack-swift | 12:21 | |
| *** CrackerJackMack has joined #openstack-swift | 12:21 | |
| *** glange has joined #openstack-swift | 12:21 | |
| *** chrisnelson has joined #openstack-swift | 12:21 | |
| *** cppforlife_ has joined #openstack-swift | 12:21 | |
| *** sendak.freenode.net sets mode: +v glange | 12:21 | |
| *** nexusz99 has joined #openstack-swift | 12:21 | |
| *** nexusz99 has quit IRC | 12:22 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 12:24 |
| *** diurchenko has joined #openstack-swift | 12:24 | |
| diurchenko | Hey. http://docs.openstack.org/api/openstack-object-storage/1.0/content/serialized-response-formats.html Says I can use one way or another. But what if I supply both format query arg and Accept http header. Which is primary? | 12:26 |
| *** sileht has quit IRC | 12:28 | |
| *** sileht has joined #openstack-swift | 12:29 | |
| *** early has joined #openstack-swift | 13:01 | |
| *** nshaikh has left #openstack-swift | 13:08 | |
| *** early has quit IRC | 13:08 | |
| *** joeljwright1 has joined #openstack-swift | 13:30 | |
| *** joeljwright has quit IRC | 13:32 | |
| *** silor has joined #openstack-swift | 13:32 | |
| *** silor has quit IRC | 13:37 | |
| *** mahatic has joined #openstack-swift | 13:46 | |
| *** silor has joined #openstack-swift | 13:59 | |
| *** jroll has quit IRC | 14:03 | |
| *** jroll has joined #openstack-swift | 14:03 | |
| *** silor has quit IRC | 14:06 | |
| *** ppai has joined #openstack-swift | 14:08 | |
| *** ppai has quit IRC | 14:08 | |
| *** bill_az_ has joined #openstack-swift | 14:10 | |
| *** mahatic has quit IRC | 14:16 | |
| *** annegent_ has joined #openstack-swift | 14:24 | |
| *** mahatic has joined #openstack-swift | 14:29 | |
| *** geaaru_ has joined #openstack-swift | 14:35 | |
| *** annegent_ has quit IRC | 14:55 | |
| *** jrichli has joined #openstack-swift | 14:56 | |
| *** annegent_ has joined #openstack-swift | 15:09 | |
| *** dmsimard_away is now known as dmsimard | 15:25 | |
| *** rdaly2 has joined #openstack-swift | 15:28 | |
| *** geaaru_ has quit IRC | 15:30 | |
| *** lnxnut has joined #openstack-swift | 15:54 | |
| *** haomaiwang has joined #openstack-swift | 15:55 | |
| hurricanerix_ | Anybody have any objections to the idea of modifying TempURL to allow keys on a per container basis? If not, I may try to throw something together for/at the hackathon. | 16:04 |
| cschwede | acoles: thx, review done :) | 16:05 |
| cschwede | hurricanerix_: i like the idea, and i already wrote a small POC for this to try different ideas. i think this could be extended to formpost as well: https://github.com/cschwede/swift-formpost2 | 16:06 |
| hurricanerix_ | cschwede: oh nice! I'll take a look. | 16:07 |
| *** david-lyle_afk is now known as david-lyle | 16:07 | |
| acoles | cschwede: great! joeljwright1 (the author) will be pleased to see that fix land | 16:08 |
| openstackgerrit | Mahati proposed openstack/swift: Ring checker in swift-recon https://review.openstack.org/153617 | 16:09 |
| *** heroanxiaobo has joined #openstack-swift | 16:09 | |
| *** heroanxiaobo has left #openstack-swift | 16:10 | |
| *** heroanxiaobo has joined #openstack-swift | 16:21 | |
| *** openstackgerrit has quit IRC | 16:21 | |
| *** openstackgerrit has joined #openstack-swift | 16:22 | |
| hurricanerix_ | cschwede: So do you think it would be acceptable to move this to the TempURL & FormPOST middleware? If so, I would be happy to work on it. | 16:22 |
| *** SkyRocknRoll has quit IRC | 16:23 | |
| cschwede | hurricanerix_: i think we should discuss this next week (you’re in SFO, right?), but yes, from my pov it would be great to see a container-based temp-url-key upstream | 16:23 |
| *** heroanxiaobo has left #openstack-swift | 16:24 | |
| hurricanerix_ | cschwede: Yup, I will be there. Sounds good to me, looking forward to discussing it. | 16:24 |
| *** annegent_ has quit IRC | 16:26 | |
| *** rmcall has joined #openstack-swift | 16:26 | |
| joeljwright1 | acoles: cschwede: Thanks so much for reviewing that patch so quickly | 16:30 |
| cschwede | joeljwright1: you’re welcome! | 16:30 |
| *** rmcall has quit IRC | 16:33 | |
| *** rmcall has joined #openstack-swift | 16:34 | |
| *** silor has joined #openstack-swift | 16:36 | |
| *** dencaval has joined #openstack-swift | 16:40 | |
| torgomatic | hurricanerix_: if you mean the ability to set tempurl keys on a container and use them to GET/PUT/etc. *objects* in that container, then that sounds good | 16:44 |
| torgomatic | if you mean the ability to use tempurls to *manipulate* containers, that's not so good | 16:44 |
| torgomatic | IMO | 16:44 |
| egon | torgomatic: A tempurl to manipulate a container might be useful, for some sort of "one-time permissions" or "remote support" use cases... | 16:51 |
| *** IRTermite has quit IRC | 16:51 | |
| hurricanerix_ | torgomatic: yes, I was thinking GET/PUT/... objects in the container. | 16:52 |
| torgomatic | egon: yeah, but given that containers have ACLs and stuff, the security implications would need lots of careful thought | 16:52 |
| torgomatic | otherwise your "temporary" access might end up being permanent :| | 16:52 |
| torgomatic | hurricanerix_: cool. ship it! | 16:52 |
| hurricanerix_ | hlol | 16:53 |
| hurricanerix_ | *lol | 16:53 |
| *** annegent_ has joined #openstack-swift | 16:57 | |
| *** BAKfr has quit IRC | 16:58 | |
| notmyname | good morning | 16:58 |
| mahatic | good morning! | 16:58 |
| notmyname | hurricanerix_: yup, I like tempurls/container (with the caveat torgomatic said) | 16:58 |
| mahatic | notmyname, how does that look? -> https://review.openstack.org/153617 , Is that right? | 17:00 |
| *** BAKfr has joined #openstack-swift | 17:01 | |
| notmyname | cschwede: got a sec to talk about the summit proposal? | 17:01 |
| *** diurchenko has quit IRC | 17:01 | |
| notmyname | mahatic: in general, yes. | 17:03 |
| notmyname | mahatic: but... | 17:03 |
| *** annegent_ has quit IRC | 17:04 | |
| *** IRTermite has joined #openstack-swift | 17:04 | |
| notmyname | mahatic: swift doesn't have a requirement on requests right now. I'd prefer not to add it. there's plenty of examples elsewhere in the codebase to pull from | 17:05 |
| mahatic | issue with requests being backward compatible? | 17:05 |
| notmyname | also,t here should probably be some eror handling (eg what if the header doesn't exist) | 17:06 |
| *** AllMessedUp has joined #openstack-swift | 17:06 | |
| notmyname | but that one is minor | 17:06 |
| notmyname | mahatic: for the time being, if you wanted to work with requests so that you have something and can move on to the reporting part, that's ok | 17:06 |
| notmyname | mahatic: but know that it won't land with the dependency on requests | 17:06 |
| *** annegent_ has joined #openstack-swift | 17:07 | |
| notmyname | does that make sense? | 17:07 |
| mahatic | yup, but I think I'd make that change and then move to the reporting part | 17:08 |
| notmyname | ok | 17:08 |
| AllMessedUp | Question guys: I have a grizzly cluster that's doing a TON of replication. I see the "REPLICATE" messages in the originating machine's syslog but there is no activity in the rsync log and the partition isn't copied to the destination storage node. Does the "REPLICATE" message indicate an attempted replication via rsync? | 17:08 |
| mahatic | notmyname, but a general question - requests is not used so far because of the compatibility issue? | 17:08 |
| *** swiftishard has joined #openstack-swift | 17:09 | |
| notmyname | mahatic: more because if we don't need the dependency, we don't add it. and most of those parts of swift were written before requests was a serious option | 17:09 |
| mahatic | notmyname, oh okay. I was guessing the latter | 17:10 |
| notmyname | AllMessedUp: (I hope your nick isn't related to your question ;-) if you look in the sample configs (https://github.com/openstack/swift/blob/master/etc/object-server.conf-sample#L149) you can slow down some of the replication with run_pause and concurrency | 17:11 |
| *** annegent_ has quit IRC | 17:12 | |
| notmyname | hmm..rereading your question, maybe there's something else | 17:12 |
| notmyname | AllMessedUp: the REPLICATE verb is used by the storage servers to check that the data is properly in sync | 17:12 |
| notmyname | AllMessedUp: if something is not in sync, then rsync is used to transfer data | 17:12 |
| cschwede | notmyname: sure | 17:12 |
| notmyname | AllMessedUp: but you're seeing a partition that isn't being copied? | 17:12 |
| notmyname | cschwede: do you have to have all of the speakers on the submission? | 17:13 |
| AllMessedUp | notmyname: correct - the partition doesn't exist on the "destination" storage node | 17:13 |
| cschwede | notmyname: from my pov not, but it might raise more interest if there are more people? | 17:13 |
| AllMessedUp | we have a ton (~200k) of time_waits and I was thinking replication may be causing it based on all the REPLICATE entries in the log | 17:14 |
| notmyname | cschwede: I think you should add clayg and kota and mattoliverau and acoles. and we can validate next week if they can't/won't do it | 17:14 |
| notmyname | cschwede: in addition to you me and peluse | 17:14 |
| AllMessedUp | notmyname: seeing virtually no activity in the rsync log | 17:14 |
| notmyname | AllMessedUp: are you using a separate replicaton network (I don't recall if that's in grizzly or not) | 17:15 |
| cschwede | notmyname: ok, good idea. i already talked to acoles earlier, and i hope we can edit it until monday evening | 17:15 |
| AllMessedUp | notmyname: no replication network - single vlan | 17:16 |
| AllMessedUp | notmyname: looking at samples | 17:16 |
| notmyname | ok | 17:17 |
| *** abhirc has quit IRC | 17:17 | |
| AllMessedUp | TY | 17:17 |
| cschwede | notmyname: done | 17:19 |
| AllMessedUp | notmyname: reduced concurrency to 2 | 17:22 |
| AllMessedUp | notmyname: just to confirm, this log entry means the partition was checked and not replicated? | 17:22 |
| AllMessedUp | Feb 6 11:21:45 xxxxxxxx object-server x.x.x.x - - [06/Feb/2015:17:21:45 +0000] "REPLICATE /sds/94051" 200 1506 "-" "-" "-" 0.0006 | 17:22 |
| *** early has joined #openstack-swift | 17:23 | |
| *** dc has joined #openstack-swift | 17:23 | |
| panbalag | Hi, I'm trying to run the functional tests against a devstack setup I have. I created a test.conf on the server where I have the functional tests. Still not all parameters (like auth_version) are getting populated and tests are not running. Has anyone executed the functional tests recently? | 17:25 |
| *** mahatic has quit IRC | 17:25 | |
| *** mahatic has joined #openstack-swift | 17:26 | |
| *** silor has quit IRC | 17:27 | |
| *** rledisez has quit IRC | 17:29 | |
| notmyname | AllMessedUp: sorry, got pulled away here at the office (such is life) | 17:30 |
| notmyname | cschwede: thanks | 17:31 |
| notmyname | panbalag: you can run functional tests against any routable swift endpoint (ie everything from your saio to public cloud providers) | 17:31 |
| notmyname | AllMessedUp: let me look about that log line. ie what does a 200 response to a REPLICATE mean | 17:32 |
| openstackgerrit | Donagh McCabe proposed openstack/swift: Add multiple reseller prefixes and composite tokens https://review.openstack.org/137086 | 17:32 |
| AllMessedUp | notmyname: so I reduced replicator concurrency from 20 to 2 and I've got about 100k less time_waits -- sound expected/feasible? | 17:34 |
| notmyname | AllMessedUp: so that log line doesn't tell you that kind of info. it does tell you that the mount point and partition are there. otherwise it would have returned 507. but it returns in the body the serialized hashes that are then compared to see if anything needs to be transferred | 17:35 |
| notmyname | AllMessedUp: yes. I'd completely expect less requests to happen with lower concurrency | 17:35 |
| notmyname | AllMessedUp: so what you're seeing is no rsync connections, but the partition should be on both servers. right? | 17:39 |
| AllMessedUp | notmyname: yes, think so -- I'm going to use swift-get-nodes to check the existence of all the objects | 17:41 |
| AllMessedUp | notmyname: not seeing any connections to rsync(873) - check via "netstat -nat | grep :873" | 17:43 |
| notmyname | AllMessedUp: have you checked that the rings are the same as you expect on all your nodes? you can do that for your whole cluster with`swift-recon --md5`. or you can check by hand | 17:46 |
| *** annegent_ has joined #openstack-swift | 17:47 | |
| AllMessedUp | notmyname: yes sir - got a external test that checks the ring files md5sums | 17:47 |
| notmyname | great | 17:47 |
| notmyname | AllMessedUp: one thing you can do is stop replication on one node and then run it in once mode to see if things are wokring are expected | 17:48 |
| *** jordanP has quit IRC | 17:51 | |
| notmyname | cschwede: please let me know asap the link for that proposal so we can add it to our "list of stuff to promote" | 17:54 |
| notmyname | AllMessedUp: you can also tell the replicator to specifically replicate certain partitions or drives. so that may also help in your testing | 17:55 |
| cschwede | notmyname: there is no preview link available yet | 18:00 |
| *** early has quit IRC | 18:00 | |
| *** gyee has joined #openstack-swift | 18:03 | |
| *** early has joined #openstack-swift | 18:06 | |
| *** geaaru has quit IRC | 18:13 | |
| openstackgerrit | Mahati proposed openstack/swift: Ring checker in swift-recon https://review.openstack.org/153617 | 18:15 |
| *** tab___ has joined #openstack-swift | 18:17 | |
| *** early has quit IRC | 18:20 | |
| mahatic | notmyname, I modified it. Regarding the report, I can probably show for a request like this "swift-recon object --<command>" , if they are all object servers and if not, show what they are (container/account) | 18:21 |
| *** acoles is now known as acoles_away | 18:27 | |
| notmyname | mahatic: yup, exactly. any ideas on what the command should be? | 18:27 |
| mahatic | notmyname, I don't know, validateringservers? | 18:28 |
| notmyname | mahatic: --validate-ring might be more clear and easy to read | 18:29 |
| notmyname | or --validate-servers | 18:29 |
| notmyname | ya, I like that one better | 18:29 |
| mahatic | notmyname, okay, cool | 18:29 |
| mahatic | notmyname, how about the current modification of OPTIONS support? | 18:29 |
| mahatic | can you review that? | 18:30 |
| notmyname | not now. engineering meeting time now. at a glance, it looks better (no requests dependency), but maybe not quite right. still, shouldn't be a blocker for doing the other stuff for now | 18:30 |
| notmyname | mahatic: thanks for working on it | 18:31 |
| mahatic | notmyname, okay. thanks for the review. Please do post any further comments on gerrit when you're free | 18:31 |
| *** Nadeem has joined #openstack-swift | 18:32 | |
| *** mmcardle has quit IRC | 18:36 | |
| *** early has joined #openstack-swift | 18:38 | |
| *** silor has joined #openstack-swift | 18:46 | |
| *** abhirc has joined #openstack-swift | 18:52 | |
| *** early has quit IRC | 18:53 | |
| *** rdaly2 has quit IRC | 18:58 | |
| *** dc has quit IRC | 19:08 | |
| AllMessedUp | notmyname: can you give me a ballpark on the number of connections a object-replicator would make in any given replication run? | 19:13 |
| *** annegent_ has quit IRC | 19:15 | |
| *** silor1 has joined #openstack-swift | 19:22 | |
| AllMessedUp | notmyname: I've dropped storage server swift services to concurrency = 2. wondering why I still have >100k TIME_WAITs | 19:22 |
| *** silor has quit IRC | 19:24 | |
| *** mariusleu has quit IRC | 19:36 | |
| AllMessedUp | notmyname: from one of the storage node logs: object-replicator 11344998 suffixes checked -- does that mean than many network connections were done in the cluster to check replicas? | 19:46 |
| *** early has joined #openstack-swift | 19:50 | |
| *** mariusleu has joined #openstack-swift | 19:53 | |
| *** early has quit IRC | 19:53 | |
| *** mahatic has quit IRC | 19:55 | |
| *** early has joined #openstack-swift | 19:58 | |
| *** joeljwright has joined #openstack-swift | 20:00 | |
| *** annegent_ has joined #openstack-swift | 20:01 | |
| *** joeljwright1 has quit IRC | 20:02 | |
| *** early has quit IRC | 20:03 | |
| *** early has joined #openstack-swift | 20:07 | |
| *** bill_az_ has quit IRC | 20:09 | |
| *** dencaval has quit IRC | 20:09 | |
| *** mahatic has joined #openstack-swift | 20:11 | |
| *** early has quit IRC | 20:14 | |
| *** AllMessedUp has quit IRC | 20:15 | |
| *** mariusleu has quit IRC | 20:18 | |
| *** early has joined #openstack-swift | 20:24 | |
| *** nellysmi_ has quit IRC | 20:27 | |
| *** annegentle has quit IRC | 20:29 | |
| openstackgerrit | Merged openstack/python-swiftclient: Fix deleting SLO segments on overwrite https://review.openstack.org/152903 | 20:34 |
| *** silor1 has quit IRC | 20:38 | |
| *** bill_az_ has joined #openstack-swift | 20:38 | |
| *** jordanP has joined #openstack-swift | 20:47 | |
| *** early has quit IRC | 20:49 | |
| *** haomaiw__ has joined #openstack-swift | 20:58 | |
| *** jordanP has quit IRC | 21:01 | |
| *** haomaiwang has quit IRC | 21:01 | |
| *** tab___ has quit IRC | 21:01 | |
| *** imkarrer has joined #openstack-swift | 21:04 | |
| *** fbo has quit IRC | 21:14 | |
| *** fbo_ has joined #openstack-swift | 21:14 | |
| *** zaitcev has joined #openstack-swift | 21:22 | |
| *** ChanServ sets mode: +v zaitcev | 21:22 | |
| imkarrer | Good afternoon. I understand that the container reconciler will move objects in the wrong storage policy and that the container replicator detects these misplaced objects, but how do objects get assigned to the wrong storage policy in the first place? Is it a naming collision between the hash rings? | 21:23 |
| notmyname | imkarrer: through a combination of concurrent requests, failures, and network partitions | 21:24 |
| notmyname | maybe add in some rebalancing in there too for good measure | 21:24 |
| notmyname | basically, anything that would result in all replicas of a container not being available (but still durably in the cluster) at the time of the subsequent request | 21:25 |
| notmyname | which is really hard to make happen | 21:26 |
| imkarrer | notmyname: thank you. | 21:27 |
| *** BaconIsK_ has joined #openstack-swift | 21:27 | |
| zaitcev | notmyname: If yinyin never comes back -- and hopefuly you forgive this unpleasant hypothesis -- at what point is it appropriate for me to take over and re-submit his patch with some small touch-ups? | 21:30 |
| notmyname | zaitcev: which patch? | 21:31 |
| zaitcev | https://review.openstack.org/134855 | 21:31 |
| *** rdaly2 has joined #openstack-swift | 21:32 | |
| *** BaconIsK_ has quit IRC | 21:32 | |
| zaitcev | I am concerned that he may feel robbed | 21:32 |
| zaitcev | And it's his 1st patch. | 21:32 |
| zaitcev | If you know his identity and can contact him, it would be helpful. | 21:33 |
| notmyname | zaitcev: you could resubmit the content with you as the author. you could keep him as the author and you as the commiter. you could resubmit as your patch and then add a "co-authored-by" in the commit message | 21:33 |
| zaitcev | My investigation established that his patch actually fixed a problem and right away I cleared 10,000 tombstones on one node. | 21:33 |
| notmyname | nope, I don't know him/her or have any contact except for the email on the patch | 21:34 |
| zaitcev | Oh, right. Git magic. | 21:34 |
| notmyname | zaitcev: but it's been around a _long_ tiime (the patch). I think you should feel free to take it over | 21:34 |
| zaitcev | oh, it was? okay | 21:34 |
| notmyname | the last comment by the author was in november | 21:35 |
| zaitcev | I'll get back to it when I'm done with SO_REUSEPORT | 21:36 |
| notmyname | cool, thanks | 21:36 |
| *** dmsimard is now known as dmsimard_away | 21:47 | |
| mattoliverau | Leaving for airport now.. The flying and waiting begins! Cya all soon! | 21:58 |
| *** annegent_ has quit IRC | 22:00 | |
| *** zacksh has quit IRC | 22:04 | |
| *** zacksh has joined #openstack-swift | 22:04 | |
| *** zacksh has quit IRC | 22:04 | |
| *** abhirc has quit IRC | 22:05 | |
| *** zacksh has joined #openstack-swift | 22:05 | |
| notmyname | mattoliverau: safe travels :-) | 22:07 |
| *** zigo has joined #openstack-swift | 22:11 | |
| *** mahatic has quit IRC | 22:12 | |
| *** jrichli has quit IRC | 22:13 | |
| *** erlon has quit IRC | 22:26 | |
| *** nellysmitt has joined #openstack-swift | 22:27 | |
| *** annegentle has joined #openstack-swift | 22:30 | |
| *** nellysmitt has quit IRC | 22:32 | |
| *** lnxnut has quit IRC | 22:35 | |
| *** abhirc has joined #openstack-swift | 22:39 | |
| *** swiftishard has quit IRC | 22:46 | |
| *** rdaly2 has quit IRC | 22:52 | |
| *** tellesnobrega_ has joined #openstack-swift | 22:54 | |
| *** annegentle has quit IRC | 22:57 | |
| *** annegentle has joined #openstack-swift | 22:57 | |
| *** BaconIsKing has joined #openstack-swift | 23:08 | |
| *** reed has quit IRC | 23:11 | |
| *** BaconIsKing has quit IRC | 23:15 | |
| *** abhirc has quit IRC | 23:15 | |
| *** annegentle has quit IRC | 23:18 | |
| *** IRTermite has quit IRC | 23:21 | |
| *** dmsimard_away is now known as dmsimard | 23:24 | |
| *** dmsimard is now known as dmsimard_away | 23:25 | |
| *** IRTermite has joined #openstack-swift | 23:37 | |
| *** abhirc has joined #openstack-swift | 23:37 | |
| *** Nadeem has quit IRC | 23:38 | |
| *** imkarrer has quit IRC | 23:41 | |
| *** annegentle has joined #openstack-swift | 23:49 | |
| *** abhirc has quit IRC | 23:56 | |
| *** annegentle has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!