| *** csd has joined #openstack-swift | 00:00 | |
| peluse | torgomatic: FYI with /info on policies the output for when un-named policies are used is not all that great. Exmaples: (a) nothing defined you will see one un-named policy (implied 0) (b) >1 unnamed you will only see one blank policy as there's nothing unique when the dict item is added. Think its worth making these cases prettier? | 00:16 |
|---|---|---|
| *** Midnightmyth has joined #openstack-swift | 00:22 | |
| peluse | I could call them <unnamed-1>, <unnamed-2>, etc.... | 00:22 |
| *** matsuhashi has joined #openstack-swift | 00:26 | |
| portante | notmyname, *all*: I have been away from swift work for the last few days, mostly, is the wiki page still good for review priorities | 00:27 |
| portante | ? | 00:27 |
| gholt | torgomatic: glange: Thanks guys on the ssync review; hoping to do a release within a week or so and didn't want to lag another month on ssync testing. :) | 00:27 |
| *** _bluev has quit IRC | 00:29 | |
| peluse | portante: you know my input :) | 00:30 |
| gholt | Damn, I figured I'd do a trade since Sam approved my change I'd get one of his. Apparently he has a lot pending... let's see then. | 00:30 |
| peluse | gholt: any interest in looking at the patch to ssync to support policies? https://review.openstack.org/#/c/65347/ | 00:31 |
| gholt | I can look that over, unfortunately my spin up time on ec is pretty high, but that's okay, I can at least see if anything pops out at me. | 00:32 |
| peluse | gholt: thanks. there's no EC in this one, its just support for multiple object rings via storage policies | 00:32 |
| torgomatic | gholt: the DLO patch is probably the most pressing, but also the largest, which is probably why it's been sitting so long :) | 00:37 |
| gholt | peluse: Do you expect an upgrade to policies to essentialy be: shutdown all background jobs (replicators, updaters, etc.) then upgrade main servers (proxy, object, container, account) then turn back on the jobs? Or can a default policy be assumed when an server is talking to a new server during an upgrade? | 00:39 |
| gholt | torgomatic: Yeah, that one is begging for completion. | 00:39 |
| peluse | gholt: if no new policies are defined in between when the proxy and the other services are upgraded everything will work (which is the likely case). I guess you could screw yourself up if for some reason you updated the proxy, defined policy 1 before upgrading the rest and then tried to use it | 00:42 |
| peluse | gholt: I actually just added a todo item today to write up the end user docs for this and we'll be sure to cover the proper upgrade steps... | 00:43 |
| gholt | Well on this, and I may be reading it wrong, won't the ssync_receiver blow up if it receives a request from the sender that doesn't have a policy? | 00:44 |
| peluse | gholt: so if the proxy is done first it will always add the proper default before anything is sent to the backend. So another posible screw up in the upgrade would be to do the services first... | 00:45 |
| gholt | With hundreds of hosts, everything will be done in every order you can imagine. :) | 00:46 |
| peluse | gholt: good point! | 00:46 |
| gholt | Which is fine if it blows up, obviously not fine if it could interpret something incorrectly, best if it can not blow up and interpret a sane default. Heh | 00:47 |
| peluse | gholt: thanks, I'll add a note to the TODO item on the docs to brainstorm the upgrade proceedures which may result in some other general code changes as well as specific instructions | 00:48 |
| gholt | I'm making the assumption there's a concept of a default policy that corresponds what was running prior to the upgrade for all the existing data. | 00:48 |
| swifterdarrell | gholt: (cc pandemicsyn) portante: you guys seen this bug? https://bugs.launchpad.net/swift/+bug/1274312 | 00:48 |
| swifterdarrell | gholt: (cc pandemicsyn) portante: early-quorum-related | 00:48 |
| gholt | swifterdarrell: Hadn't seen it, but I could imagine it's an issue | 00:50 |
| peluse | gholt: correct. there's 2 concepts here, a 'policy 0' which is used for access to containers that exist already and have no policy and a 'default' which is used to create new containers when no policy is specified (allowing for something other than repl for new containers) | 00:50 |
| swifterdarrell | gholt: cool; that 2nd suggested solution seem okay (I like it better than the first one I thought of) | 00:50 |
| swifterdarrell | gholt: (sorry that was supposed to end in a ?) ;) | 00:51 |
| peluse | gholt: BTW 'policy 0' and the default can indeed be one and the same but they don't have to be... | 00:52 |
| gholt | swifterdarrell: Yeah, that or just make a subclass that everybody uses which overrides waitall. Same same | 00:52 |
| *** Midnightmyth has quit IRC | 00:52 | |
| peluse | gotta run... catch you guys later | 00:53 |
| gholt | peluse: Gotcha, 0 and default are distinct concepts but may be identical in reality. Cool | 00:53 |
| swifterdarrell | gholt: *nod* I haven't looked yet, but my hunch is that every user actually wants the new functionality (i.e. every caller of waitall() really wants wait-some-time-then-give-up-with-appropriate-draining) | 00:55 |
| swifterdarrell | gholt: obvs, if that's not true, I'd leave some way to get ahold of the old behavior | 00:55 |
| swifterdarrell | gholt: that's even assuming I get some time to implement the fix! heh | 00:55 |
| gholt | What's the side effect of the bug? Timeouts raised in weird places causing odd things to explode? | 00:57 |
| gholt | Also, I think GreenAsyncPile was redbo's brainchild. :) | 00:58 |
| swifterdarrell | gholt: it doesn't actually hurt anything, best I can tell; the symptom I observed at least once was 3rd backend request ("laggard") on an obj PUT finished up after 1.5s with flying colors; The client got the 201 after just over 0.5s (the post_quorum_timeout). | 00:58 |
| swifterdarrell | gholt: then... TWO MINUTES LATER, the log has a line for that same (uniquely-named) object with Timeout(10s) | 00:59 |
| swifterdarrell | gholt: lol? | 00:59 |
| gholt | Oh gotcha. | 00:59 |
| swifterdarrell | gholt: to be fair, the GreenAsyncPile is sweet, it's just not getting drained in some cases now | 00:59 |
| swifterdarrell | gholt: if the laggard really did take > 10s the timeout would be legit and should get logged | 00:59 |
| swifterdarrell | gholt: as for the 2m delay in logging, I got nothin' | 00:59 |
| * swifterdarrell shrugs | 00:59 | |
| gholt | Leaving stuff sitting for all that extra time == bad (for memory, general resources) and the error report it makes is wrong. | 01:00 |
| swifterdarrell | yup | 01:00 |
| swifterdarrell | but it's not causing any new req failures that weren't already gonna fail | 01:00 |
| swifterdarrell | no data loss, etc. | 01:00 |
| swifterdarrell | no lying to the client | 01:00 |
| swifterdarrell | just confusion for the operator ;) | 01:00 |
| gholt | Good, nobody likes emergency fixes. :) | 01:00 |
| swifterdarrell | gholt: ya, I consider this one as a "sometime in the next release" | 01:01 |
| portante | swifterdarrell: I have not seen that bug, but did read your report | 01:04 |
| *** csd has quit IRC | 01:08 | |
| *** krtaylor has joined #openstack-swift | 01:19 | |
| *** zackf has quit IRC | 01:24 | |
| *** krtaylor has quit IRC | 01:26 | |
| *** nosnos has joined #openstack-swift | 01:38 | |
| *** zaitcev has quit IRC | 01:48 | |
| *** zaitcev has joined #openstack-swift | 01:48 | |
| *** alpha_ori has quit IRC | 01:55 | |
| *** alpha_ori has joined #openstack-swift | 01:58 | |
| *** otherjon_ has quit IRC | 02:13 | |
| openstackgerrit | Samuel Merritt proposed a change to openstack/swift: Move all DLO functionality to middleware https://review.openstack.org/63326 | 02:40 |
| *** pheadron1 has joined #openstack-swift | 03:10 | |
| *** krtaylor has joined #openstack-swift | 03:12 | |
| *** pheadron has quit IRC | 03:13 | |
| openstackgerrit | A change was merged to openstack/swift: Config option to lower the timeout for recoverable object GETs. https://review.openstack.org/69509 | 03:27 |
| *** matsuhashi has quit IRC | 03:29 | |
| *** nshaikh has joined #openstack-swift | 03:34 | |
| *** chandankumar_ has joined #openstack-swift | 03:36 | |
| *** chandankumar_ has quit IRC | 03:40 | |
| *** chandankumar_ has joined #openstack-swift | 03:41 | |
| *** leizhang has joined #openstack-swift | 03:44 | |
| *** chandankumar_ has quit IRC | 03:46 | |
| *** chandankumar_ has joined #openstack-swift | 03:47 | |
| *** chandankumar_ has quit IRC | 03:47 | |
| *** chandankumar__ has joined #openstack-swift | 03:47 | |
| *** chandankumar__ has left #openstack-swift | 03:49 | |
| *** leizhang has quit IRC | 04:01 | |
| *** matsuhashi has joined #openstack-swift | 04:02 | |
| *** saurabh_ has joined #openstack-swift | 04:23 | |
| *** pheadron has joined #openstack-swift | 04:25 | |
| *** pheadron1 has quit IRC | 04:28 | |
| *** pheadron1 has joined #openstack-swift | 04:30 | |
| madhuri | notmyname: ping?? | 04:33 |
| *** pheadron has quit IRC | 04:33 | |
| *** gvernik has quit IRC | 04:42 | |
| openstackgerrit | A change was merged to openstack/swift: Some functional tests for object versioning https://review.openstack.org/62954 | 04:44 |
| *** matsuhashi has quit IRC | 05:14 | |
| *** matsuhas_ has joined #openstack-swift | 05:17 | |
| *** ppai has joined #openstack-swift | 05:26 | |
| *** pheadron has joined #openstack-swift | 05:27 | |
| *** pheadron1 has quit IRC | 05:30 | |
| *** hurricanerix has quit IRC | 05:37 | |
| *** gyee has quit IRC | 05:43 | |
| *** zaitcev has quit IRC | 05:45 | |
| *** zackf has joined #openstack-swift | 05:51 | |
| *** zackf has quit IRC | 05:56 | |
| *** psharma has joined #openstack-swift | 06:00 | |
| *** madhuri has quit IRC | 06:03 | |
| *** pheadron1 has joined #openstack-swift | 06:14 | |
| *** pheadron has quit IRC | 06:17 | |
| Anju | hii all, | 06:34 |
| Anju | Is swift s3 compatible ? | 06:34 |
| koolhead17 | Anju: did you check the swift doc by any chance :) | 06:41 |
| Anju | koolhead17: hii. yes :) :) | 06:41 |
| koolhead17 | you would not have asked then :P | 06:42 |
| koolhead17 | Anju: http://docs.openstack.org/grizzly/openstack-object-storage/admin/content//configuring-openstack-object-storage-with-s3_api.html << See if it helps cheers | 06:42 |
| Anju | koolhead17: so by adding the middlware only | 06:42 |
| Anju | we can interact from s3 api to swift / | 06:42 |
| koolhead17 | ya | 06:42 |
| koolhead17 | see the doc :P | 06:43 |
| koolhead17 | good luck | 06:43 |
| Anju | koolhead17: thanks | 06:43 |
| *** acoles has quit IRC | 08:14 | |
| *** odyssey4me has joined #openstack-swift | 08:17 | |
| *** acoles has joined #openstack-swift | 08:20 | |
| *** fbo_away is now known as fbo | 08:22 | |
| *** mlipchuk has joined #openstack-swift | 08:34 | |
| *** nosnos has quit IRC | 08:39 | |
| *** nosnos has joined #openstack-swift | 08:39 | |
| *** nacim has joined #openstack-swift | 08:46 | |
| *** madhuri has joined #openstack-swift | 08:46 | |
| madhuri | Hello could anyone tell me what does fast POST method means? | 08:47 |
| *** _bluev has joined #openstack-swift | 08:53 | |
| *** matsuhas_ has quit IRC | 09:33 | |
| *** ppai has quit IRC | 09:37 | |
| *** matsuhashi has joined #openstack-swift | 09:39 | |
| *** xga has joined #openstack-swift | 09:42 | |
| *** nosnos has quit IRC | 09:46 | |
| *** nosnos has joined #openstack-swift | 09:46 | |
| *** nosnos has quit IRC | 09:47 | |
| *** Manish_ has joined #openstack-swift | 09:48 | |
| Manish_ | Hi.. | 09:48 |
| *** matsuhashi has quit IRC | 09:50 | |
| *** ppai has joined #openstack-swift | 09:54 | |
| *** Manish_ has quit IRC | 09:55 | |
| *** acoles_ has joined #openstack-swift | 10:26 | |
| *** acoles_ has quit IRC | 10:44 | |
| *** Midnightmyth has joined #openstack-swift | 10:48 | |
| *** xga has quit IRC | 10:51 | |
| *** ppai has quit IRC | 11:07 | |
| *** xga has joined #openstack-swift | 11:08 | |
| openstackgerrit | A change was merged to openstack/swift: Skip delete_at_update for replication requests https://review.openstack.org/68529 | 11:11 |
| *** ppai has joined #openstack-swift | 11:20 | |
| *** pheadron1 has quit IRC | 11:39 | |
| *** xga has quit IRC | 11:40 | |
| *** Midnightmyth has quit IRC | 11:43 | |
| *** pheadron has joined #openstack-swift | 11:54 | |
| *** xga has joined #openstack-swift | 11:56 | |
| *** ppai has quit IRC | 12:06 | |
| *** nacim has quit IRC | 12:10 | |
| *** ppai has joined #openstack-swift | 12:24 | |
| *** ppai has quit IRC | 12:42 | |
| *** nacim has joined #openstack-swift | 12:43 | |
| *** Midnightmyth has joined #openstack-swift | 12:55 | |
| *** foexle has joined #openstack-swift | 13:00 | |
| *** haomaiwa_ has quit IRC | 13:23 | |
| *** judd7_ has joined #openstack-swift | 13:28 | |
| *** haomaiwang has joined #openstack-swift | 13:30 | |
| *** judd7 has quit IRC | 13:31 | |
| *** haomaiwa_ has joined #openstack-swift | 13:33 | |
| *** haomaiwang has quit IRC | 13:35 | |
| *** xga_ has joined #openstack-swift | 14:01 | |
| *** xga has quit IRC | 14:01 | |
| *** psharma has quit IRC | 14:02 | |
| *** mjseger has quit IRC | 14:26 | |
| *** Bamberrouse has joined #openstack-swift | 14:33 | |
| *** Bamberrouse has left #openstack-swift | 14:33 | |
| *** mlipchuk has quit IRC | 14:51 | |
| *** nshaikh has left #openstack-swift | 14:58 | |
| *** byeager has joined #openstack-swift | 15:00 | |
| *** bada has joined #openstack-swift | 15:06 | |
| *** xga has joined #openstack-swift | 15:25 | |
| *** xga_ has quit IRC | 15:26 | |
| *** xga has quit IRC | 15:37 | |
| *** pberis has joined #openstack-swift | 15:40 | |
| *** hurricanerix has joined #openstack-swift | 15:41 | |
| *** pberis has quit IRC | 15:41 | |
| *** xga has joined #openstack-swift | 15:41 | |
| *** pberis has joined #openstack-swift | 15:42 | |
| *** hurricanerix has quit IRC | 15:42 | |
| *** hurricanerix has joined #openstack-swift | 15:43 | |
| *** zackf has joined #openstack-swift | 15:45 | |
| *** pheadron1 has joined #openstack-swift | 15:45 | |
| *** pheadron has quit IRC | 15:48 | |
| *** tongli has joined #openstack-swift | 15:50 | |
| *** xga has quit IRC | 15:56 | |
| *** xga has joined #openstack-swift | 15:58 | |
| *** pheadron has joined #openstack-swift | 16:10 | |
| *** pheadron1 has quit IRC | 16:13 | |
| *** xga has quit IRC | 16:18 | |
| *** xga has joined #openstack-swift | 16:26 | |
| *** k00lkidz has joined #openstack-swift | 16:28 | |
| *** zul has quit IRC | 16:29 | |
| *** xga has quit IRC | 16:30 | |
| k00lkidz | Swift is try to put objects into an account that doesn't exist. The auth that shows up in the logs doesn't match up with the auth on all the current account when doing a swift list. | 16:31 |
| *** zul has joined #openstack-swift | 16:32 | |
| *** xga has joined #openstack-swift | 16:32 | |
| k00lkidz | Anyone got a remedy? | 16:32 |
| *** rook][ has joined #openstack-swift | 16:33 | |
| _bluev | k00lkidz: do you have account_autocreate set in your proxy-server.conf ? | 16:36 |
| k00lkidz | yes | 16:38 |
| _bluev | k00lkidz: tempauth or keystone ? | 16:44 |
| k00lkidz | swauth | 16:44 |
| _bluev | if you run "swauth-list -A http://127.0.0.1/auth/ -K <<supersecret>> <<account_name>>" run the proxy server, does it give the account URL you expect ? | 16:45 |
| k00lkidz | the problem is none of the account url when I run the command you listed matches the one in the logs | 16:47 |
| *** xga has quit IRC | 16:49 | |
| *** xga has joined #openstack-swift | 16:49 | |
| *** gyee has joined #openstack-swift | 16:51 | |
| *** byeager has quit IRC | 17:01 | |
| *** byeager has joined #openstack-swift | 17:01 | |
| k00lkidz | _bluev: do you have a clue on what happen? | 17:03 |
| *** byeager_ has joined #openstack-swift | 17:04 | |
| *** byeager has quit IRC | 17:05 | |
| k00lkidz | gholt: I am experiencing PUT errors in Swift using Swauth. Swift is trying to PUT objects into an account that doesn't exist. When I did a swauth-list on all current accounts, none of the account urls match up with the errored account URL in the logs. | 17:06 |
| k00lkidz | gholt: have you heard of such error? solution? | 17:07 |
| _bluev | k00lkidz: my guess is that the user is forcing the storageURL and us not using the one returned from authentication with swauth | 17:15 |
| *** byeager_ has quit IRC | 17:18 | |
| *** byeager has joined #openstack-swift | 17:19 | |
| k00lkidz | _bluev: do you have a good way to trace back to the source? I didn't see any connection info. | 17:21 |
| *** byeager has quit IRC | 17:23 | |
| *** _bluev has quit IRC | 17:26 | |
| *** zackf1 has joined #openstack-swift | 17:34 | |
| k00lkidz | _bluev: the source is from another storage server according to the logs. How/why does the storage server use/know of the account? Is there a way to tell the storage server to not use the non-existent account? | 17:34 |
| *** zaitcev has joined #openstack-swift | 17:35 | |
| *** xga has quit IRC | 17:36 | |
| *** xga has joined #openstack-swift | 17:36 | |
| *** zackf has quit IRC | 17:37 | |
| *** csd has joined #openstack-swift | 17:44 | |
| *** gvernik_ has joined #openstack-swift | 17:45 | |
| *** gvernik_ has quit IRC | 17:50 | |
| *** gvernik has joined #openstack-swift | 17:50 | |
| *** xga has quit IRC | 17:50 | |
| *** xga has joined #openstack-swift | 17:51 | |
| *** shri has joined #openstack-swift | 17:55 | |
| *** krtaylor has quit IRC | 17:56 | |
| *** byeager has joined #openstack-swift | 18:08 | |
| *** byeager has quit IRC | 18:09 | |
| *** byeager has joined #openstack-swift | 18:09 | |
| *** k00lkidz has quit IRC | 18:10 | |
| *** Midnightmyth has quit IRC | 18:10 | |
| *** Midnightmyth has joined #openstack-swift | 18:11 | |
| *** byeager has quit IRC | 18:12 | |
| *** byeager has joined #openstack-swift | 18:12 | |
| *** byeager has quit IRC | 18:17 | |
| *** xga has quit IRC | 18:17 | |
| *** zackf has joined #openstack-swift | 18:21 | |
| *** zackf1 has quit IRC | 18:21 | |
| shri | notmyname, dfg: Just wanted to bring this to your attention: https://bugs.launchpad.net/swift/+bug/1274634 | 18:25 |
| shri | would be great if we fix this is 1.12 | 18:25 |
| notmyname | hello world. just opened my IRC client for the first time in 2 days | 18:26 |
| notmyname | shri: 1.12 has already been released | 18:27 |
| shri | oh.. :-) | 18:27 |
| notmyname | shri: it's something that could be pretty easily added. it's always 1.0 right now | 18:27 |
| notmyname | shri: you should propose a patch to add it :-) | 18:28 |
| shri | you guys will be way faster in doing it | 18:28 |
| *** nacim has quit IRC | 18:30 | |
| *** judd7_ has quit IRC | 18:32 | |
| *** judd7 has joined #openstack-swift | 18:32 | |
| *** fbo is now known as fbo_away | 18:33 | |
| *** gyee has quit IRC | 18:34 | |
| *** byeager has joined #openstack-swift | 18:36 | |
| *** odyssey4me has quit IRC | 18:53 | |
| *** _bluev has joined #openstack-swift | 19:10 | |
| *** markd has joined #openstack-swift | 19:11 | |
| *** byeager_ has joined #openstack-swift | 19:25 | |
| *** byeager has quit IRC | 19:26 | |
| *** gvernik has quit IRC | 19:34 | |
| *** k00lkidz has joined #openstack-swift | 19:37 | |
| *** csd has quit IRC | 19:38 | |
| *** gvernik has joined #openstack-swift | 19:40 | |
| *** shri1 has joined #openstack-swift | 19:41 | |
| *** fbo_away is now known as fbo | 19:42 | |
| *** shri has quit IRC | 19:44 | |
| *** csd has joined #openstack-swift | 19:45 | |
| *** gyee has joined #openstack-swift | 19:47 | |
| k00lkidz | gholt: We currently have a Swift cluster using Swauth. The storage servers, for some reason, are using an nonexistent account url to PUT objects for replication. Why causes this and how do we stop them? | 19:49 |
| *** finite has joined #openstack-swift | 19:50 | |
| *** fbo is now known as fbo_away | 19:52 | |
| *** pberis has quit IRC | 19:57 | |
| rook][ | hello all, anyone have a sec to answer an ACL question? I have two users in an account and want one of the read only to a container. Is that possible? | 20:13 |
| rook][ | swift stat thecontainer Account: AUTH_838dec975ad948e1aab29c203f5dfb48 Container: thecontainer Objects: 0 Bytes: 0 Read ACL: cosbenchro Write ACL: cosbench Sync To: Sync Key: Accept-Ranges: bytes X-Timestamp: 1391112176.91371 Content-Type: text/plain; charset=utf-8 | 20:15 |
| rook][ | BUT, the ro account can write in the container..... | 20:15 |
| rook][ | TIA | 20:15 |
| *** _bluev has quit IRC | 20:29 | |
| rook][ | also tried to put '-' in the ACL: Write ACL: -cosbenchro | 20:30 |
| gholt | k00lkidz: No idea what all that means exactly. Any pastes of actual log lines? | 20:32 |
| gholt | I mean, they can have lots of [redacted] in them, but knowing as close to the actual error text helps. | 20:32 |
| k00lkidz | gholt: do you prefer pastebin or directly in irc? | 20:33 |
| k00lkidz | gholt: the logs | 20:34 |
| k00lkidz | gholt: logs: https://gist.github.com/anonymous/8718150 | 20:35 |
| k00lkidz | gholt: here is a better log example: https://gist.github.com/anonymous/8718203 | 20:39 |
| k00lkidz | gholt: I put some comments before each logs | 20:40 |
| rook][ | new ACLs same behavior | 20:40 |
| rook][ | Read ACL: cosbench:cosbenchro Write ACL: cosbench:cosbench | 20:40 |
| k00lkidz | gholt: the storage server is trying to replicate using an invalid account url causing PUT errors from what I can tell | 20:40 |
| gholt | koolhead17: I see, I don't think that has anything to do with Swauth. | 20:41 |
| gholt | koolhead17: Just looks like maybe you had an account at one time and now no longer do? And now there are containers that are trying to report to their account and cannot. | 20:41 |
| gholt | koolhead17: This looks strinkingly similar to a report some one else sent my way the other day about missing accounts. The container names were similar to the one you posted of 113600819885433552455 | 20:42 |
| koolhead17 | gholt: k00lkidz ^^ | 20:42 |
| gholt | Oh, sorry, too many people that are kool in here. ;) | 20:43 |
| k00lkidz | gholt: that also person was me | 20:43 |
| k00lkidz | gholt: :) | 20:43 |
| gholt | Okay, so now why do you think it's Swauth? | 20:43 |
| k00lkidz | gholt: I am not sure this is right, but I thought the storage server reference their account url from swauth since it seems to be the account management component. | 20:45 |
| k00lkidz | gholt: I can't locate the containers for the auth url since the partition doesn't exist. | 20:47 |
| gholt | k00lkidz: None of what you posted has auth involved, tbh. It's all backend chatter and they don't use auth to talk to one another. | 20:49 |
| gholt | Last time we talked we were pretty sure the cluster lost the account dbs somehow, though we couldn't determine how, and that the best course of action was to recreate the account hash db, let things run, and see what turns up in that account. | 20:50 |
| k00lkidz | gholt: since I don't know the name of the account, how do I recreate the account hash db? | 20:51 |
| *** _bluev has joined #openstack-swift | 20:51 | |
| gholt | You can do a PUT directly to the account server with the AUTH_xxx is talking about. You'll probably also have to include an x-timestamp header, though x-timestamp: 1 will probably work fine. | 20:52 |
| *** s00perk00lkidz has joined #openstack-swift | 20:55 | |
| k00lkidz | gholt: can you give me an example? I have never try putting an object directly | 20:59 |
| gholt | k00lkidz: Oh sure, if I look at your first gist https://gist.github.com/anonymous/8718150 it would be something like: curl -XPUT -Hx-timestamp:1 http://<ip_of_server_that_logged_that_line>:<account_server_port>/sde/367796/AUTH_ad4a8b7b-fdfd-4401-aad5-082907b6bf3a | 21:09 |
| k00lkidz | gholt: Just tried, didn't work. | 21:17 |
| k00lkidz | gholt: still getting the same logs. | 21:17 |
| k00lkidz | gholt: clarification, the curl went through, but it didn't fixed the issue. | 21:17 |
| gholt | Okay, well I'm not sure how much I can help from here. | 21:21 |
| k00lkidz | gholt: do you know of anyone else that can help? | 21:22 |
| *** bsdkurt has quit IRC | 21:22 | |
| gholt | Well, I mean, if I was there running that cluster I could help. :) But it's very hard for me to try to help over irc through someone else's hands and eyes. Sorry. :/ I'm not sure if somebody in here else is better at this than me. | 21:23 |
| *** bsdkurt has joined #openstack-swift | 21:24 | |
| *** hurricanerix has quit IRC | 21:28 | |
| *** ozone227 has joined #openstack-swift | 21:42 | |
| *** foexle has quit IRC | 21:52 | |
| *** krtaylor has joined #openstack-swift | 21:56 | |
| openstackgerrit | Jon Snitow proposed a change to openstack/swift: Privileged acct ACL header, new ACL syntax, TempAuth impl. https://review.openstack.org/63227 | 22:07 |
| openstackgerrit | Jon Snitow proposed a change to openstack/swift: Non-blocking but useful improvements to Account ACLs https://review.openstack.org/69321 | 22:07 |
| *** pheadron1 has joined #openstack-swift | 22:08 | |
| *** pheadron has quit IRC | 22:10 | |
| *** Trixboxer has quit IRC | 22:11 | |
| *** pheadron1 is now known as pheadron | 22:11 | |
| *** finite has quit IRC | 22:12 | |
| *** s00perk00lkidz has quit IRC | 22:16 | |
| openstackgerrit | Samuel Merritt proposed a change to openstack/swift: Add Storage Policy Support to the Updater https://review.openstack.org/64737 | 22:21 |
| *** ozone227 has quit IRC | 22:31 | |
| *** k00lkidz has quit IRC | 22:33 | |
| *** rook][ has quit IRC | 22:33 | |
| openstackgerrit | Peter Portante proposed a change to openstack/swift: In-process swift server for functional tests https://review.openstack.org/66108 | 22:43 |
| *** zackf is now known as zackf1 | 22:48 | |
| *** zackf has joined #openstack-swift | 23:11 | |
| *** zackf1 has quit IRC | 23:13 | |
| *** zackf is now known as zakcf | 23:29 | |
| *** zakcf is now known as zackf | 23:29 | |
| *** byeager_ has quit IRC | 23:30 | |
| *** zackf1 has joined #openstack-swift | 23:52 | |
| *** shri has joined #openstack-swift | 23:52 | |
| *** zackf has quit IRC | 23:53 | |
| *** shri1 has quit IRC | 23:54 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!