| *** germs has joined #openstack-swift | 00:01 | |
| *** germs has quit IRC | 00:01 | |
| *** germs has joined #openstack-swift | 00:01 | |
| openstackgerrit | Merged openstack/swift master: Support -d <devs> and -p <partitions> in DB replicators. https://review.openstack.org/548628 | 00:04 |
|---|---|---|
| openstackgerrit | Merged openstack/swift feature/deep: container replicator _in_sync should consider deleted shard ranges https://review.openstack.org/553875 | 00:04 |
| openstackgerrit | Merged openstack/swift feature/deep: Pass existing ranges to ContainerBroker._find_shard_ranges https://review.openstack.org/554310 | 00:04 |
| openstackgerrit | Merged openstack/swift master: Use state when you have it https://review.openstack.org/553927 | 00:05 |
| *** germs has quit IRC | 00:05 | |
| *** itlinux has joined #openstack-swift | 00:23 | |
| *** tovin07_ has joined #openstack-swift | 00:45 | |
| *** lifeless_ has joined #openstack-swift | 00:46 | |
| *** lifeless has quit IRC | 00:46 | |
| *** tovin07_ has quit IRC | 00:48 | |
| *** tovin07_ has joined #openstack-swift | 00:49 | |
| *** itlinux has quit IRC | 00:51 | |
| *** cshastri has joined #openstack-swift | 00:59 | |
| *** tovin07_ has quit IRC | 01:29 | |
| *** tovin07_ has joined #openstack-swift | 01:29 | |
| hugokuo | Swift does need some ways to throttle the bandwidth and diskIO for all consistency workers(auditor, replicator and reconstructor). There's a serious problem from operator's angle of view. While there's more and more data in the cluster, consistency worker does its job pretty hard on disks. That brings the disk utilization up to 100%. | 01:40 |
| hugokuo | Then the new incoming requests could be affected. | 01:41 |
| mattoliverau | acoles: lol, long live the composter :P | 01:42 |
| mattoliverau | hugokuo: there is the ionice options etc | 01:42 |
| mattoliverau | so you can give more precidence to user traffic | 01:43 |
| hugokuo | I remember that now. It's worth to give it a try. I didn't use that before. Not sure of efficiency with ionice. | 01:44 |
| DHE | besides the existing io priority system, you could use control groups to limit total disk IO... | 01:48 |
| openstackgerrit | Merged openstack/swift feature/deep: Add auto_shard option to enable or disable leader behaviour https://review.openstack.org/553781 | 01:59 |
| *** germs has joined #openstack-swift | 02:02 | |
| *** germs has quit IRC | 02:06 | |
| *** wes_dillingham has quit IRC | 02:19 | |
| *** links has joined #openstack-swift | 02:40 | |
| *** itlinux has joined #openstack-swift | 02:42 | |
| *** psachin has joined #openstack-swift | 03:02 | |
| *** bkopilov has quit IRC | 03:08 | |
| *** cshastri has quit IRC | 03:16 | |
| *** cshastri has joined #openstack-swift | 03:31 | |
| *** kei_yama has quit IRC | 03:44 | |
| *** itlinux has quit IRC | 03:49 | |
| *** kei_yama has joined #openstack-swift | 04:01 | |
| *** germs has joined #openstack-swift | 04:03 | |
| *** germs has quit IRC | 04:03 | |
| *** germs has joined #openstack-swift | 04:03 | |
| *** germs has quit IRC | 04:07 | |
| *** m_kazuhiro has joined #openstack-swift | 04:15 | |
| m_kazuhiro | good morning | 04:16 |
| *** armaan has quit IRC | 04:17 | |
| *** armaan has joined #openstack-swift | 04:18 | |
| *** armaan has quit IRC | 04:25 | |
| *** armaan has joined #openstack-swift | 04:26 | |
| *** bkopilov has joined #openstack-swift | 04:41 | |
| openstackgerrit | Kazuhiro MIYAHARA proposed openstack/swift master: Refactor tempurl functional test's try-finally https://review.openstack.org/554412 | 04:56 |
| *** armaan has quit IRC | 05:31 | |
| *** armaan has joined #openstack-swift | 05:31 | |
| *** armaan has quit IRC | 05:53 | |
| *** armaan has joined #openstack-swift | 05:54 | |
| *** armaan has quit IRC | 06:02 | |
| *** germs has joined #openstack-swift | 06:04 | |
| *** germs has quit IRC | 06:08 | |
| *** psachin has quit IRC | 06:27 | |
| openstackgerrit | Merged openstack/swift master: Encapsulate some general utility in utils https://review.openstack.org/553928 | 06:45 |
| *** psachin has joined #openstack-swift | 06:47 | |
| *** neonpastor has joined #openstack-swift | 07:01 | |
| *** Supun has joined #openstack-swift | 07:08 | |
| *** d0ugal has joined #openstack-swift | 07:12 | |
| *** d0ugal has quit IRC | 07:12 | |
| *** d0ugal has joined #openstack-swift | 07:12 | |
| *** d0ugal has quit IRC | 07:21 | |
| *** rcernin has quit IRC | 07:23 | |
| *** rcernin has joined #openstack-swift | 07:24 | |
| *** rcernin has quit IRC | 07:24 | |
| *** armaan has joined #openstack-swift | 07:25 | |
| *** tovin07 has joined #openstack-swift | 07:31 | |
| *** pcaruana has joined #openstack-swift | 07:34 | |
| *** Supun has quit IRC | 07:45 | |
| *** m_kazuhiro has quit IRC | 07:48 | |
| *** hseipp has joined #openstack-swift | 07:48 | |
| *** d0ugal has joined #openstack-swift | 07:49 | |
| *** tesseract has joined #openstack-swift | 08:00 | |
| *** armaan has quit IRC | 08:02 | |
| *** germs has joined #openstack-swift | 08:05 | |
| *** germs has quit IRC | 08:05 | |
| *** germs has joined #openstack-swift | 08:05 | |
| *** Supun has joined #openstack-swift | 08:05 | |
| *** m_kazuhiro has joined #openstack-swift | 08:05 | |
| *** germs has quit IRC | 08:08 | |
| *** armaan has joined #openstack-swift | 08:13 | |
| *** gkadam has joined #openstack-swift | 08:22 | |
| *** _alastor_ has quit IRC | 08:30 | |
| *** _alastor_ has joined #openstack-swift | 08:31 | |
| acoles | good morning | 08:31 |
| m_kazuhiro | acoles: good morning | 08:32 |
| openstackgerrit | Kazuhiro MIYAHARA proposed openstack/swift master: Add object-expirer new mode to execute tasks from general task queue https://review.openstack.org/517389 | 08:32 |
| *** geaaru has joined #openstack-swift | 08:36 | |
| *** armaan has quit IRC | 08:43 | |
| *** armaan has joined #openstack-swift | 08:43 | |
| acoles | m_kazuhiro: o/ | 08:44 |
| *** d0ugal has quit IRC | 09:02 | |
| *** jiaopengju_ is now known as jiaopengju | 09:02 | |
| openstackgerrit | Kazuhiro MIYAHARA proposed openstack/swift master: Refactor tempurl functional test's try-finally https://review.openstack.org/554412 | 09:03 |
| *** cbartz has joined #openstack-swift | 09:07 | |
| *** _glb has quit IRC | 09:30 | |
| *** glb has joined #openstack-swift | 09:31 | |
| *** armaan has quit IRC | 09:38 | |
| *** armaan has joined #openstack-swift | 09:39 | |
| olc | i understand that swift has not mechanisms that can guarantee that the ring is consistent, but what happens in case of inconsistency? | 09:43 |
| *** Supun has quit IRC | 09:44 | |
| *** Supun has joined #openstack-swift | 09:45 | |
| *** glb has quit IRC | 09:50 | |
| *** glb has joined #openstack-swift | 09:51 | |
| *** Supun has quit IRC | 09:59 | |
| *** tovin07_ has quit IRC | 10:01 | |
| *** germs has joined #openstack-swift | 10:05 | |
| *** germs has quit IRC | 10:05 | |
| *** germs has joined #openstack-swift | 10:05 | |
| kei-ichi | I'd like to add one thing as agenda for next swift meting. "Question about slogging unittest" | 10:07 |
| kei-ichi | I would like to decide policy for running unittest. | 10:07 |
| kei-ichi | Inside from slogging unittest, many of swift libray is used like following. | 10:07 |
| kei-ichi | ``` | 10:07 |
| kei-ichi | from swift.common.utils import readconf, TRUE_VALUES | 10:07 |
| kei-ichi | ... | 10:07 |
| kei-ichi | from swift.common.exceptions import ChunkReadTimeout, LockTimeout | 10:07 |
| kei-ichi | ``` | 10:07 |
| kei-ichi | Sometimes import just a tool. --> pattern-1 | 10:08 |
| kei-ichi | But sometimes using exception class directly. --> pattern-2 | 10:08 |
| kei-ichi | I'm now wondering should I completly copy those kind of code into slogging ? | 10:08 |
| kei-ichi | Or keep using this by importing from swift ? | 10:08 |
| kei-ichi | Let me discuss about above point in next meeting o/ | 10:08 |
| *** germs has quit IRC | 10:10 | |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep: Use state rather than subtlety to make shards of shards authoritative https://review.openstack.org/551356 | 10:14 |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep: Root leader decides when shard containers shard https://review.openstack.org/552120 | 10:14 |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep: Store epoch and root container in shard ranges table https://review.openstack.org/552668 | 10:14 |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep: Remove unnecessary timestamp shenanigans when creating shard container https://review.openstack.org/552133 | 10:26 |
| *** mvk has quit IRC | 10:31 | |
| *** armaan has quit IRC | 10:46 | |
| *** mvk has joined #openstack-swift | 10:46 | |
| *** cshastri has quit IRC | 10:56 | |
| *** m_kazuhiro has quit IRC | 10:59 | |
| *** rcernin has joined #openstack-swift | 11:02 | |
| *** Supun has joined #openstack-swift | 11:04 | |
| *** rcernin has quit IRC | 11:06 | |
| *** Supun has quit IRC | 11:18 | |
| *** Supun has joined #openstack-swift | 11:21 | |
| *** pcaruana has quit IRC | 11:23 | |
| *** psachin has quit IRC | 11:28 | |
| *** psachin has joined #openstack-swift | 11:34 | |
| *** armaan has joined #openstack-swift | 11:48 | |
| *** Supun has quit IRC | 11:53 | |
| *** pcaruana has joined #openstack-swift | 11:55 | |
| *** germs has joined #openstack-swift | 12:06 | |
| *** germs has quit IRC | 12:06 | |
| *** germs has joined #openstack-swift | 12:06 | |
| *** germs has quit IRC | 12:10 | |
| *** wes_dillingham has joined #openstack-swift | 12:14 | |
| *** Supun has joined #openstack-swift | 12:16 | |
| *** bkopilov has quit IRC | 12:32 | |
| *** openstackgerrit has quit IRC | 12:33 | |
| *** NM has joined #openstack-swift | 12:36 | |
| *** Fr33m1nd has joined #openstack-swift | 12:37 | |
| Fr33m1nd | Hi | 12:37 |
| *** d0ugal has joined #openstack-swift | 12:41 | |
| *** d0ugal has quit IRC | 12:41 | |
| *** d0ugal has joined #openstack-swift | 12:41 | |
| *** Supun has quit IRC | 13:06 | |
| *** openstackgerrit has joined #openstack-swift | 13:08 | |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep: Handle exceptions in _send_shard_ranges https://review.openstack.org/554531 | 13:08 |
| *** kei_yama has quit IRC | 13:13 | |
| *** Supun has joined #openstack-swift | 13:18 | |
| *** psachin has quit IRC | 13:21 | |
| *** armaan has quit IRC | 13:24 | |
| *** armaan has joined #openstack-swift | 13:25 | |
| *** OrionX_ has joined #openstack-swift | 13:26 | |
| *** OrionX has quit IRC | 13:30 | |
| *** bkopilov has joined #openstack-swift | 13:51 | |
| *** links has quit IRC | 13:51 | |
| *** bob_cheesey has joined #openstack-swift | 13:51 | |
| bob_cheesey | afternoon all, wonder if anyone could help me figure out where an object would be on disk? i've used swift-get-nodes and tracked it down to a specific dir on a device, however that dir only contains a file called 1504771716.00350.data | 13:52 |
| bob_cheesey | i've checked the other locations it could be and they're the same | 13:53 |
| acoles | bob_cheesey: the .data file is the object file | 13:56 |
| acoles | the format of the filename is <timestamp>.data | 13:57 |
| bob_cheesey | ah ok, i was under the impression that the file should reside on disk there | 13:57 |
| bob_cheesey | as in, the filename you expect | 13:57 |
| bob_cheesey | however i guess that doesn't work if it's in a bunch of subdirs | 13:57 |
| acoles | swift-object-info will show you details | 13:58 |
| acoles | bob_cheesey: object files don't have the name of the object, if that's what you mean | 13:59 |
| bob_cheesey | acoles: yeah that's what i was expecting but i think i mis-understood what i was reading | 13:59 |
| acoles | but if you run swift-object info on a .data file it will you show you the object name | 14:00 |
| bob_cheesey | yup and it's what i expect | 14:02 |
| bob_cheesey | thanks! | 14:02 |
| *** germs has joined #openstack-swift | 14:07 | |
| *** germs has quit IRC | 14:07 | |
| *** germs has joined #openstack-swift | 14:07 | |
| bob_cheesey | ok, so following on from that question, a customer of ours somehow 'accidentally' (their words) set an expiration date on some objects | 14:09 |
| bob_cheesey | fortunately they're still on disk but they can't be retrieved (obviously). is it possible to remove that metadata? | 14:09 |
| *** sai_ is now known as sai-mtg | 14:09 | |
| *** OrionX_ has quit IRC | 14:11 | |
| *** germs has quit IRC | 14:11 | |
| *** saint_ has joined #openstack-swift | 14:11 | |
| acoles | bob_cheesey: first, if the expiry time has passed but there is still a .data file then the object-expirer hasn't got round to deleting the object, but it will unless you stop it | 14:15 |
| bob_cheesey | yup, it hasn't be deleted yet | 14:16 |
| bob_cheesey | the expirer is currently disabled so it won't blitz it | 14:16 |
| acoles | bob_cheesey: ok that's good. there's no API means to undelete it however. I'm just thinking about how you might do it under-the-hood. | 14:20 |
| *** pcaruana has quit IRC | 14:21 | |
| *** armaan has quit IRC | 14:25 | |
| *** armaan has joined #openstack-swift | 14:25 | |
| *** pcaruana has joined #openstack-swift | 14:28 | |
| bob_cheesey | ok, that would be handy if it were possible. i've just gotten the list back from the customer and it's quite substantial so i'm not recovering them by hand | 14:28 |
| acoles | bob_cheesy: one approach (the easiest I have thought of so far) is to use swift's internal client to post new (empty) object metadata, thus clearing the x-delete-at metadata. However, that would also clear any other metadata your customer may have included with their objects (e.g. x-object-meta-<whatever>). | 14:34 |
| bob_cheesey | acoles: would any metadata they've set show up with swift-object-info? | 14:34 |
| acoles | the internal client can include a special header that causes the expiry status of the object to be ignored | 14:34 |
| acoles | bob_cheesey: yes | 14:35 |
| bob_cheesey | in which case that should work as the only thing in the user metadata dict is the delete-at header | 14:35 |
| acoles | in swift-object-info under section 'User Metadata:' | 14:35 |
| acoles | and there may be stuff in Other Metadata: section that I haven't remembered, other than X-Delete-At | 14:36 |
| *** d0ugal has quit IRC | 14:38 | |
| bob_cheesey | ok, i think they're more interested in recovering the data than anything else at the mo | 14:39 |
| bob_cheesey | thanks for your help! | 14:39 |
| *** OrionX has joined #openstack-swift | 14:40 | |
| acoles | bob_cheesey: I'm just trying something out... | 14:40 |
| bob_cheesey | ok! | 14:40 |
| *** cshastri has joined #openstack-swift | 14:45 | |
| *** OrionX has quit IRC | 14:46 | |
| acoles | bob_cheesey: what version of swift? | 14:49 |
| *** itlinux has joined #openstack-swift | 14:50 | |
| acoles | going to need 2.15.0 or later | 14:50 |
| bob_cheesey | ah no older than that | 14:50 |
| bob_cheesey | :( | 14:50 |
| acoles | :( | 14:50 |
| bob_cheesey | i can hack together something with bash to retrieve each file and preserve the directory structure so that may be the only way of doing this | 14:51 |
| acoles | bob_cheesey: I'm on a call, will get back to you | 14:54 |
| bob_cheesey | no worries - i'm heading off soon | 14:54 |
| *** armaan has quit IRC | 14:57 | |
| *** armaan has joined #openstack-swift | 14:57 | |
| *** jistr is now known as jistr|mtg | 15:01 | |
| openstackgerrit | Alexandre Lécuyer proposed openstack/swift master: Add round_robin_iter function to common/utils https://review.openstack.org/554582 | 15:05 |
| openstackgerrit | Alexandre Lécuyer proposed openstack/swift master: Change object_audit_location_generator() to yield for a single policy. https://review.openstack.org/548589 | 15:06 |
| *** OrionX has joined #openstack-swift | 15:07 | |
| *** d0ugal has joined #openstack-swift | 15:10 | |
| *** Supun has quit IRC | 15:14 | |
| *** armaan has quit IRC | 15:19 | |
| *** armaan has joined #openstack-swift | 15:19 | |
| *** jistr|mtg is now known as jistr | 15:36 | |
| *** Supun has joined #openstack-swift | 15:37 | |
| olc | is `swift-get-nodes -a account.ring.gz demo` well supposed to give the path where the account database of user 'demo' is located in? | 15:44 |
| *** OrionX has quit IRC | 15:45 | |
| olc | cccccchkrlkdtvenlfnlfdbfhuvejjkdtjntjhenrrej | 15:54 |
| acoles | olc: yes | 15:55 |
| timburke | note that many auth systems add an "auth prefix" so a user demo might have an account like AUTH_demo (and it's the *account* name that you'd want to pass to swift-get-nodes) | 15:57 |
| acoles | bob_cheesey: sorry, had meeting. FWIW this is what I came up with but it requires swift 2.15.0 https://gist.github.com/alistairncoles/722e3e8a6d7107c9fd93c1db7ff7d73c | 16:01 |
| olc | thanks guys, i have tested with demo, AUTH_demo and AUTH_5e55447707cd47ebb433dc24320b571f and none returns back a correct path | 16:01 |
| olc | i'll investigate further | 16:02 |
| acoles | olc: by 'correct' do you mean there is not db file at the path? that may happen if the user has not yet created any containers | 16:02 |
| olc | by correct I mean that directory does not exist | 16:03 |
| olc | i have created (uploaded) an object with that acound | 16:04 |
| olc | *account | 16:04 |
| olc | anyway I'll look deeper in it | 16:05 |
| olc | thanks | 16:05 |
| *** Supun has quit IRC | 16:11 | |
| *** Supun has joined #openstack-swift | 16:12 | |
| *** Fr33m1nd has quit IRC | 16:20 | |
| *** gyee has joined #openstack-swift | 16:26 | |
| notmyname | good morning | 16:28 |
| *** mikecmpbll has joined #openstack-swift | 16:30 | |
| notmyname | mikecmpbll: oh hi. bob_cheesey was the one asking about expiry | 16:30 |
| *** cshastri has quit IRC | 16:31 | |
| mikecmpbll | 👍🏼 | 16:35 |
| *** armaan has quit IRC | 16:35 | |
| acoles | bob_cheesey: https://gist.github.com/alistairncoles/27c086457a1251e9f7f4b85f57a3b553 use it at your own risk of course. You might alternatively zip up all the .data files and send them back to your customer for them to upload again, without x-delete- headers :) depends on size. | 16:36 |
| *** armaan has joined #openstack-swift | 16:36 | |
| *** joeljwright has quit IRC | 16:41 | |
| *** Supun has quit IRC | 16:44 | |
| *** gkadam has quit IRC | 16:47 | |
| *** mvk has quit IRC | 16:49 | |
| *** cschwede_ has quit IRC | 17:00 | |
| *** mvk has joined #openstack-swift | 17:03 | |
| *** cbartz has quit IRC | 17:06 | |
| *** mikecmpbll has left #openstack-swift | 17:07 | |
| *** Supun has joined #openstack-swift | 17:11 | |
| *** hseipp has quit IRC | 17:22 | |
| openstackgerrit | Tim Burke proposed openstack/swift feature/deep: Add swift-manage-shard-ranges entry point; drop find_shard_ranges script https://review.openstack.org/554639 | 17:23 |
| acoles | timburke: thanks ^ | 17:32 |
| *** OrionX has joined #openstack-swift | 17:46 | |
| *** OrionX has quit IRC | 17:50 | |
| *** Supun has quit IRC | 17:53 | |
| *** Supun has joined #openstack-swift | 17:54 | |
| *** armaan has quit IRC | 18:15 | |
| *** armaan has joined #openstack-swift | 18:16 | |
| *** geaaru has quit IRC | 18:22 | |
| *** mikecmpbll has joined #openstack-swift | 18:24 | |
| *** gyee has quit IRC | 18:30 | |
| zaitcev | mother of go | 18:32 |
| zaitcev | or god, in this case | 18:32 |
| zaitcev | I added +class TestECObjectControllerPP(BaseTestECObjectController, unittest.TestCase): | 18:33 |
| zaitcev | Basically it re-runs all the exact tests we have for Proxy, only with PUT+POST forced on | 18:34 |
| zaitcev | Surely PUT+POST is fully compatible, right? Well, no. Tests interfere with each other because we examine what the object server wrote into _datadir, and our object server exists across a bunch of tests. | 18:35 |
| zaitcev | And since all the files use the name "pimento" | 18:36 |
| zaitcev | there is more | 18:40 |
| zaitcev | AssertionError: '<html><h1>Ac' != 'HTTP/1.1 201' | 18:40 |
| clayg | WHOA, no that does not sound good : | 18:44 |
| zaitcev | someone forgot to read the response body | 18:49 |
| tdasilva | notmyname: is the LOSF meeting confirmed for tomorrow? | 18:51 |
| notmyname | tdasilva: yes | 18:51 |
| tdasilva | notmyname: ack, thanks! | 18:52 |
| clayg | FYI, if the gate will allow it to merge patch 554582 was pretty obvious | 18:58 |
| patchbot | https://review.openstack.org/#/c/554582/ - swift - Add round_robin_iter function to common/utils | 18:58 |
| acoles | timburke: p 554639 needs test_find_shard_ranges.py deleting but we ought to preserve some of the tests | 19:07 |
| patchbot | https://review.openstack.org/#/c/554639/ - swift (feature/deep) - Add swift-manage-shard-ranges entry point; drop fi... | 19:07 |
| *** Supun has quit IRC | 19:27 | |
| *** psachin has joined #openstack-swift | 19:28 | |
| *** d34dh0r53 has quit IRC | 19:37 | |
| *** d34dh0r53 has joined #openstack-swift | 19:37 | |
| *** psachin has quit IRC | 19:44 | |
| *** wes_dillingham has quit IRC | 19:46 | |
| openstackgerrit | Merged openstack/swift master: Refactor tempurl functional test's try-finally https://review.openstack.org/554412 | 19:47 |
| *** psachin has joined #openstack-swift | 19:48 | |
| openstackgerrit | Merged openstack/swift feature/deep: Add script tool to manage container db shard ranges https://review.openstack.org/553882 | 19:54 |
| *** tesseract has quit IRC | 19:56 | |
| *** armaan has quit IRC | 19:57 | |
| *** armaan has joined #openstack-swift | 20:05 | |
| *** psachin has quit IRC | 20:08 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift feature/deep: Imperceptible performance improvement https://review.openstack.org/554686 | 20:08 |
| *** psachin has joined #openstack-swift | 20:09 | |
| *** germs has joined #openstack-swift | 20:10 | |
| *** germs has quit IRC | 20:10 | |
| *** germs has joined #openstack-swift | 20:10 | |
| *** germs has quit IRC | 20:13 | |
| *** psachin has quit IRC | 20:14 | |
| *** openstackgerrit has quit IRC | 20:33 | |
| *** openstackgerrit has joined #openstack-swift | 20:46 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift feature/deep: Add a bit more context to errors https://review.openstack.org/554692 | 20:46 |
| openstackgerrit | Clay Gerrard proposed openstack/swift feature/deep: Imperceptible performance improvement https://review.openstack.org/554686 | 20:46 |
| *** Renich has joined #openstack-swift | 21:01 | |
| Renich | Good day. How can I get the usage data of a user? | 21:01 |
| Renich | I want to know how much storage is he/she using on swift | 21:01 |
| Renich | For now, I have a script that parses over all all the accounts and gets theh list of containers to, subsecuaently, get the "bytes_used" for every container. This takes a long time. | 21:03 |
| clayg | Renich: that's basically it - someone is trying to make https://github.com/notmyname/slogging an openstack managed program | 21:04 |
| clayg | or whatever the term is for doing reviews on gerrit and hosted in git.o.o and mirrored into github.com/openstack | 21:04 |
| clayg | anyway - it probably does about the same as your doing - but maybe with some concurrency? | 21:04 |
| Renich | http://dpaste.com/382FG6S | 21:05 |
| Renich | clayg: OK. I'll take a look. Thanks! | 21:05 |
| clayg | oh... what you're doing is using the api - the slogging stuff trolls the actual on-disk databases | 21:05 |
| Renich | clayg: yes, I could make a parser for this using crystal ;) | 21:06 |
| Renich | ah, OK. So one can query the DBs? | 21:06 |
| Renich | cool | 21:06 |
| Renich | I'll have to look at that code to figure it out | 21:07 |
| Renich | clayg: thanks a lot for the pointers! | 21:07 |
| clayg | Renich: you know what - try what you're doing but with `object store account show` | 21:07 |
| clayg | i think it'll give you per account aggreate bytes used in a single api request per credentials | 21:08 |
| clayg | might be good enough for a while | 21:08 |
| openstackgerrit | Tim Burke proposed openstack/swift feature/deep: Add swift-manage-shard-ranges entry point; drop find_shard_ranges script https://review.openstack.org/554639 | 21:14 |
| Renich | clayg: awesome, will do! Was trying to build the deb packages for slogging but it doesn't work. Also, issues seem too... old... Maybe it's abandoned? | 21:16 |
| clayg | yeah... i might call it "finished" - notmyname ??? | 21:17 |
| Renich | clayg: awesome! It shows Bytes for the account! Cool! Useful! | 21:17 |
| timburke | it'll probably be a little out of date... like the container stats, only more so | 21:18 |
| timburke | most likely not enough to be a problem | 21:18 |
| notmyname | I'm not working on slogging, but IIRC there was some `python setup.py build_deb` thing with it. talk to kei-ichi about what the current status of slogging is. he's wanting to continue work on it | 21:18 |
| *** wes_dillingham has joined #openstack-swift | 21:22 | |
| *** itlinux has quit IRC | 21:31 | |
| *** NM has quit IRC | 21:31 | |
| Renich | Thanks a lot clayg, for the help! | 21:44 |
| *** germs has joined #openstack-swift | 22:10 | |
| *** germs has quit IRC | 22:10 | |
| *** germs has joined #openstack-swift | 22:10 | |
| *** rcernin has joined #openstack-swift | 22:13 | |
| mattoliverau | Morning | 22:14 |
| *** germs has quit IRC | 22:14 | |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep: Clean up stats dumped to recon cache and logged https://review.openstack.org/554358 | 22:21 |
| timburke | acoles: go to sleep! | 22:26 |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep: Clean up stats dumped to recon cache and logged https://review.openstack.org/554358 | 22:28 |
| timburke | no! no! bad acoles! sleep! | 22:29 |
| *** ianychoi__ is now known as ianychoi | 22:32 | |
| openstackgerrit | Pete Zaitcev proposed openstack/swift master: PUT+POST and its development test https://review.openstack.org/427911 | 22:37 |
| mattoliverau | lol | 22:40 |
| zaitcev | sorry, I had to kill pimento | 22:40 |
| zaitcev | https://review.openstack.org/#/c/427911/26/test/unit/proxy/test_server.py | 22:40 |
| patchbot | patch 427911 - swift - PUT+POST and its development test | 22:40 |
| openstackgerrit | Merged openstack/swift feature/deep: Handle exceptions in _send_shard_ranges https://review.openstack.org/554531 | 22:47 |
| openstackgerrit | Merged openstack/swift master: Add round_robin_iter function to common/utils https://review.openstack.org/554582 | 22:47 |
| *** armaan has quit IRC | 22:48 | |
| *** zaitcev has quit IRC | 22:56 | |
| *** gerardgine has joined #openstack-swift | 22:58 | |
| clayg | what does "kill pimeto" mean!? | 23:05 |
| clayg | notmyname: were you trying to get legacy-tempest-dsvm-neutron-full out of our gate tests? Is there a *not* "legacy-*" version of that test? | 23:13 |
| notmyname | no, I wasn't trying to get anything out of our gate | 23:13 |
| notmyname | the legacy-* jobs are the ones that are still in -infra repos after the zuul v3 merge | 23:13 |
| clayg | oh, that's helpful - thank you for that! | 23:14 |
| openstackgerrit | Merged openstack/swift feature/deep: Use state rather than subtlety to make shards of shards authoritative https://review.openstack.org/551356 | 23:17 |
| *** Anticime1 is now known as Anticimex | 23:29 | |
| openstackgerrit | Tim Burke proposed openstack/swift feature/deep: Defend against sharding an expanding root container https://review.openstack.org/554733 | 23:40 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!