| *** RockKuo has joined #openstack-swift | 00:11 | |
| *** matsuhashi has joined #openstack-swift | 00:18 | |
| *** zackf has joined #openstack-swift | 00:58 | |
| *** fifieldt has joined #openstack-swift | 01:19 | |
| *** nosnos has joined #openstack-swift | 01:30 | |
| *** zackf has quit IRC | 02:13 | |
| *** mlipchuk has quit IRC | 02:24 | |
| *** chandan_kumar has joined #openstack-swift | 02:29 | |
| *** chandan_kumar has quit IRC | 02:30 | |
| *** matsuhashi has quit IRC | 02:47 | |
| *** nosnos has quit IRC | 03:02 | |
| *** matsuhashi has joined #openstack-swift | 03:04 | |
| *** nosnos has joined #openstack-swift | 03:20 | |
| *** matsuhashi has quit IRC | 03:29 | |
| *** chandan_kumar has joined #openstack-swift | 03:30 | |
| *** haomai___ has quit IRC | 03:30 | |
| *** chandan_kumar has quit IRC | 03:31 | |
| *** chandan_kumar has joined #openstack-swift | 03:31 | |
| *** cheri has joined #openstack-swift | 03:39 | |
| *** chandan_kumar has quit IRC | 03:48 | |
| *** zhiyan_ is now known as zhiyan | 03:50 | |
| *** nosnos has quit IRC | 03:51 | |
| *** sfineberg has quit IRC | 04:02 | |
| *** madhuri has quit IRC | 04:03 | |
| *** zackf has joined #openstack-swift | 04:03 | |
| *** madhuri has joined #openstack-swift | 04:03 | |
| *** sfineberg has joined #openstack-swift | 04:04 | |
| *** zackf has quit IRC | 04:07 | |
| *** chandan_kumar has joined #openstack-swift | 04:08 | |
| *** zackf has joined #openstack-swift | 04:33 | |
| *** zhiyan is now known as zhiyan_ | 04:39 | |
| *** chandan_kumar has quit IRC | 05:02 | |
| *** Midnightmyth has joined #openstack-swift | 05:03 | |
| *** ppai has joined #openstack-swift | 05:08 | |
| *** saju_m has joined #openstack-swift | 05:13 | |
| *** chandan_kumar has joined #openstack-swift | 05:27 | |
| *** StevenK has quit IRC | 05:29 | |
| *** wkelly has quit IRC | 05:29 | |
| *** StevenK has joined #openstack-swift | 05:30 | |
| *** wkelly has joined #openstack-swift | 05:30 | |
| *** nshaikh has joined #openstack-swift | 05:33 | |
| *** scohen has joined #openstack-swift | 05:40 | |
| *** scohen has quit IRC | 05:43 | |
| *** Midnightmyth has quit IRC | 05:51 | |
| *** cheri has quit IRC | 06:18 | |
| *** cheri has joined #openstack-swift | 06:25 | |
| *** cheri has quit IRC | 06:29 | |
| *** scohen has joined #openstack-swift | 06:36 | |
| *** saju_m has quit IRC | 06:41 | |
| *** psharma has joined #openstack-swift | 06:50 | |
| *** RockKuo has quit IRC | 06:51 | |
| *** saju_m has joined #openstack-swift | 06:54 | |
| openstackgerrit | Gil Vernik proposed a change to openstack/swift: Support for content-length in the upload object method for internal client. https://review.openstack.org/85650 | 06:59 |
|---|---|---|
| *** chandan_kumar has quit IRC | 07:02 | |
| *** h6w has left #openstack-swift | 07:03 | |
| *** chandan_kumar has joined #openstack-swift | 07:07 | |
| *** mlipchuk has joined #openstack-swift | 07:27 | |
| *** Longgeek_ has joined #openstack-swift | 07:29 | |
| *** cheri has joined #openstack-swift | 07:30 | |
| *** Longgeek_ has quit IRC | 07:34 | |
| *** Longgeek_ has joined #openstack-swift | 07:38 | |
| *** mmcardle has joined #openstack-swift | 08:00 | |
| *** nacim has joined #openstack-swift | 08:00 | |
| *** mlipchuk has quit IRC | 08:02 | |
| *** foexle has joined #openstack-swift | 08:03 | |
| *** gvernik has joined #openstack-swift | 08:09 | |
| gvernik | I uploaded a patch that failed with some error in Jenkis, i never saw before. Can someone explain me what is wrong with it? https://review.openstack.org/#/c/85650/ | 08:10 |
| *** tanee is now known as tanee-away | 08:13 | |
| *** tanee-away is now known as tanee | 08:14 | |
| *** mkerrin has joined #openstack-swift | 08:14 | |
| *** RockKuo has joined #openstack-swift | 08:15 | |
| *** [1]gvernik has joined #openstack-swift | 08:16 | |
| *** gvernik has quit IRC | 08:16 | |
| *** [1]gvernik is now known as gvernik | 08:16 | |
| *** mlipchuk has joined #openstack-swift | 08:20 | |
| openstackgerrit | Victor Stinner proposed a change to openstack/python-swiftclient: Fix swiftclient.client.quote() for Python 3 https://review.openstack.org/84102 | 08:21 |
| *** mlipchuk has quit IRC | 08:25 | |
| *** sungju_ has quit IRC | 08:29 | |
| openstackgerrit | Victor Stinner proposed a change to openstack/python-swiftclient: Fix swiftclient.client.quote() for Python 3 https://review.openstack.org/84102 | 08:30 |
| *** mlipchuk has joined #openstack-swift | 08:34 | |
| *** gadb has joined #openstack-swift | 08:35 | |
| *** jamie_h has joined #openstack-swift | 08:35 | |
| *** matsuhashi has joined #openstack-swift | 08:35 | |
| *** nosnos has joined #openstack-swift | 08:40 | |
| gadb | hugokuo : hi | 08:47 |
| hugokuo | gadb: hi | 08:47 |
| gadb | how are u today? | 08:48 |
| hugokuo | gadb: good thx/ | 08:48 |
| gadb | i still test about delete .ts file ;( | 08:48 |
| gadb | i have some question. | 08:49 |
| gadb | do you have some time for me? | 08:49 |
| hugokuo | gadb: yes | 08:51 |
| gadb | when i delete hashes.pki file all object dir, delete ts file and remake hashes.pki consisit "<80>^B}q^A." text | 08:51 |
| hugokuo | gadb: https://bugs.launchpad.net/swift/+bug/1301728 | 08:51 |
| gadb | in my think, that text means "empty" partition | 08:51 |
| gadb | right? | 08:51 |
| gadb | thx. that was my problam | 08:52 |
| hugokuo | gadb: you can load the pickle for human readable format | 08:53 |
| gadb | i make scirpt for delete all object hashes.pki file. | 08:53 |
| gadb | so, swift delete all .ts file. | 08:53 |
| gadb | swift can not find hashes.pki file. then try reconstitution hashes.pki file. | 08:54 |
| hugokuo | sure... if the .ts already over the reclaim_age and the hash_suffix is None. | 08:54 |
| gadb | yeah. | 08:54 |
| *** d89 has joined #openstack-swift | 08:54 | |
| hugokuo | Almost right. | 08:54 |
| gadb | but, i wonder when delete empty partition | 08:54 |
| gadb | partitions are never delete? swift only create? | 08:55 |
| gadb | my goal is clear all swift needless i-node | 08:56 |
| gadb | hugokuo : your bug report so cool :) | 09:00 |
| gadb | that is my problum. exactly | 09:01 |
| hugokuo | gadb: Perhaps someone will review it later. | 09:04 |
| hugokuo | gadb: anyhow, I think the partition directory will not be removed. But it doesn't matter from my point of view. | 09:05 |
| gadb | hugokuo : thx. i will review too. in my think. partitions are reuse. but i wondoer how many partition can create by swfit. and what logic | 09:09 |
| hugokuo | gadb: easy.... that's the basic of Swift. You can have a look at the ring mechanism. | 09:10 |
| gadb | okey. i well check doc. | 09:10 |
| gadb | anyway have a good dinner and good time | 09:10 |
| hugokuo | gadb: For the total partitions in a cluster, it will be 3*2**part_power . | 09:11 |
| *** sungju_ has joined #openstack-swift | 09:14 | |
| *** sungju_ has quit IRC | 09:14 | |
| *** gvernik has quit IRC | 09:16 | |
| gadb | hugokuo : yes. i know it. but i concern that swift much many partition. | 09:17 |
| gadb | sorry. i must go out. have a good day. | 09:17 |
| *** ashish_ has joined #openstack-swift | 09:19 | |
| *** gadb has quit IRC | 09:21 | |
| *** mlipchuk has quit IRC | 09:25 | |
| *** mkollaro has joined #openstack-swift | 09:27 | |
| *** mlipchuk has joined #openstack-swift | 09:27 | |
| *** saju_m has quit IRC | 09:27 | |
| *** ashish_ has quit IRC | 09:30 | |
| *** mlipchuk1 has joined #openstack-swift | 09:32 | |
| *** mkollaro has quit IRC | 09:35 | |
| *** mlipchuk has quit IRC | 09:35 | |
| *** mkollaro has joined #openstack-swift | 09:42 | |
| openstackgerrit | A change was merged to openstack/swift: Move module level setup under setup_package method https://review.openstack.org/84339 | 09:44 |
| openstackgerrit | A change was merged to openstack/python-swiftclient: Make bin/swift testable part 2 https://review.openstack.org/76489 | 09:45 |
| *** Trixboxer has joined #openstack-swift | 09:52 | |
| *** tovecs has quit IRC | 10:03 | |
| *** gvernik has joined #openstack-swift | 10:08 | |
| *** matsuhashi has quit IRC | 10:16 | |
| *** Longgeek_ has quit IRC | 10:16 | |
| *** matsuhashi has joined #openstack-swift | 10:16 | |
| *** matsuhashi has quit IRC | 10:20 | |
| *** saju_m has joined #openstack-swift | 10:32 | |
| *** [1]gvernik has joined #openstack-swift | 10:42 | |
| *** gvernik has quit IRC | 10:44 | |
| *** [1]gvernik is now known as gvernik | 10:44 | |
| *** saju_m has quit IRC | 10:49 | |
| *** jamie_h_ has joined #openstack-swift | 10:55 | |
| *** [1]gvernik has joined #openstack-swift | 10:57 | |
| *** jamie_h has quit IRC | 10:57 | |
| *** gvernik has quit IRC | 10:58 | |
| *** [1]gvernik is now known as gvernik | 10:58 | |
| *** [1]gvernik has joined #openstack-swift | 11:02 | |
| *** gvernik has quit IRC | 11:03 | |
| *** tdasilva has joined #openstack-swift | 11:05 | |
| *** [1]gvernik has quit IRC | 11:07 | |
| *** scohen has left #openstack-swift | 11:07 | |
| *** gvernik has joined #openstack-swift | 11:08 | |
| *** RockKuo has quit IRC | 11:14 | |
| *** jamie_h_ has quit IRC | 11:21 | |
| *** jamie_h has joined #openstack-swift | 11:21 | |
| *** saju_m has joined #openstack-swift | 11:24 | |
| *** mlipchuk has joined #openstack-swift | 11:26 | |
| *** mlipchuk1 has quit IRC | 11:26 | |
| *** matsuhashi has joined #openstack-swift | 11:26 | |
| *** saschpe has joined #openstack-swift | 11:32 | |
| *** foexle has quit IRC | 11:40 | |
| *** annegentle has quit IRC | 11:43 | |
| *** matsuhashi has quit IRC | 11:49 | |
| *** matsuhashi has joined #openstack-swift | 11:50 | |
| *** foexle has joined #openstack-swift | 11:52 | |
| *** matsuhas_ has joined #openstack-swift | 11:53 | |
| *** matsuhashi has quit IRC | 11:54 | |
| *** mmcardle has quit IRC | 11:57 | |
| *** nosnos has quit IRC | 11:59 | |
| *** dmsimard has joined #openstack-swift | 12:00 | |
| *** cheri has quit IRC | 12:03 | |
| *** ppai has quit IRC | 12:06 | |
| *** JuanManuelOlle has joined #openstack-swift | 12:10 | |
| *** mlipchuk1 has joined #openstack-swift | 12:13 | |
| *** mlipchuk has quit IRC | 12:15 | |
| *** mlipchuk1 has quit IRC | 12:18 | |
| *** mlipchuk has joined #openstack-swift | 12:18 | |
| *** mmcardle has joined #openstack-swift | 12:20 | |
| *** cheri has joined #openstack-swift | 12:21 | |
| *** ppai has joined #openstack-swift | 12:23 | |
| *** zigo has quit IRC | 12:26 | |
| *** tdasilva has left #openstack-swift | 12:37 | |
| *** cheri has quit IRC | 12:42 | |
| *** ppai has quit IRC | 12:43 | |
| *** cheri has joined #openstack-swift | 12:55 | |
| *** rustlebee is now known as russellb | 13:08 | |
| *** changbl has quit IRC | 13:13 | |
| *** RockKuo has joined #openstack-swift | 13:18 | |
| *** cheri has quit IRC | 13:21 | |
| *** matsuhas_ has quit IRC | 13:27 | |
| *** matsuhashi has joined #openstack-swift | 13:28 | |
| *** ashish_ has joined #openstack-swift | 13:28 | |
| *** matsuhas_ has joined #openstack-swift | 13:31 | |
| *** matsuhashi has quit IRC | 13:32 | |
| *** matsuhas_ has quit IRC | 13:34 | |
| *** tanee is now known as tanee-away | 13:42 | |
| *** tdasilva has joined #openstack-swift | 13:52 | |
| *** piyush has joined #openstack-swift | 13:54 | |
| *** chandan_kumar has quit IRC | 13:56 | |
| *** zackf has quit IRC | 13:59 | |
| *** dmsimard1 has joined #openstack-swift | 14:02 | |
| *** zackf1 has joined #openstack-swift | 14:03 | |
| *** dmsimard2 has joined #openstack-swift | 14:03 | |
| *** dmsimard has quit IRC | 14:04 | |
| *** dmsimard1 has quit IRC | 14:06 | |
| *** changbl has joined #openstack-swift | 14:11 | |
| *** zackf1 has quit IRC | 14:12 | |
| *** psharma has quit IRC | 14:17 | |
| *** zackf has joined #openstack-swift | 14:23 | |
| *** zackf has quit IRC | 14:26 | |
| *** bill_az has joined #openstack-swift | 14:32 | |
| *** changbl has quit IRC | 14:34 | |
| *** tellesnobrega has joined #openstack-swift | 14:45 | |
| tellesnobrega | hi, im starting to use sahara and it uses swift for storage, im trying to start a devstack installation with swift, but im getting the following error http://paste.openstack.org/show/75216/ | 14:46 |
| tellesnobrega | can anyone give me a hint how to fix it | 14:46 |
| tellesnobrega | ? | 14:46 |
| cschwede | tellesnobrega: try to set the tenant name, for example: export OS_TENANT_NAME=mytenantname | 14:48 |
| tellesnobrega | cschwede: doesn't devstack set tenant names when installing? | 14:49 |
| *** saju_m has quit IRC | 14:50 | |
| cschwede | afaik not automatically. you can set them using „source ~/devstack/openrc“ (given that ~/devstack is the cloned repo). have a look at http://devstack.org/openrc.html | 14:50 |
| tellesnobrega | cschwede: i see. i will try that | 14:53 |
| *** nshaikh has quit IRC | 14:54 | |
| *** zaitcev has joined #openstack-swift | 15:02 | |
| *** ChanServ sets mode: +v zaitcev | 15:02 | |
| *** bada has joined #openstack-swift | 15:09 | |
| *** nacim has quit IRC | 15:15 | |
| *** Guest_ has joined #openstack-swift | 15:16 | |
| *** cschwede has quit IRC | 15:16 | |
| *** cschwede has joined #openstack-swift | 15:17 | |
| *** gyee has joined #openstack-swift | 15:20 | |
| wer | my delete performance is not that great. I am not sure where the bottleneck is. | 15:23 |
| *** zackf has joined #openstack-swift | 15:24 | |
| *** mwstorer has joined #openstack-swift | 15:28 | |
| *** zul has quit IRC | 15:28 | |
| *** zul has joined #openstack-swift | 15:33 | |
| wer | large batches of deletes begin to take about 3 seconds to complete... and I get ERROR container update failed with 10.9.2.133:6001/d69 timeouts. It seems limited to a 3 drives that timeout.... But they appear healthy. | 15:36 |
| creiht | wer: delete performance has a similar path to small object puts | 15:37 |
| wer | hard on container performance? | 15:37 |
| creiht | the delete's have to be sent to all replicas, and all containers have to be updated | 15:37 |
| openstackgerrit | Greg Lange proposed a change to openstack/swift: Unify backend logging https://review.openstack.org/83237 | 15:38 |
| creiht | yeah the deletes taking about 3 seconds are waiting on the container update | 15:38 |
| creiht | most likely | 15:38 |
| wer | ah! so the container is stored on those three drives probably? | 15:38 |
| creiht | just like if you tried to put a bunch of 0-byte files, you would probably see similar behavior | 15:38 |
| creiht | yeah | 15:38 |
| wer | shit. So I probably should have sharded this. | 15:38 |
| creiht | heh | 15:38 |
| wer | ok that makes some sense at least.... hrm. | 15:39 |
| wer | The other thing I noted is that the proxy server doesn't log the deletes. Which is driving me a bit batty. It logs all the puts but I only get timing info for the deletes. | 15:40 |
| *** dmsimard2 has quit IRC | 15:41 | |
| wer | ty creiht. That bottleneck was staring me right in the face :/ | 15:42 |
| *** dmsimard has joined #openstack-swift | 15:42 | |
| *** foexle has quit IRC | 15:43 | |
| wer | I don't have a lot of containers. I have a lot of objects though. I used the same ring layout for both. So maybe a different layout for the containers is justified. If I could say move them to where there is less io until I can fix this? | 15:44 |
| notmyname | good morning world | 15:47 |
| portante | notmyname: good morning | 15:50 |
| portante | any swiftclient folks hanging around in here? | 15:50 |
| portante | got a question about swiftclient and threads | 15:51 |
| portante | or rather swiftclient with threads vs eventlet | 15:51 |
| swifterdarrell | portante: maybe kinda... | 15:51 |
| notmyname | I was disconnected all weekend (camping in a redwood forest). what did I miss? | 15:51 |
| portante | notmyname: did you camp up in a tree at all? | 15:52 |
| swifterdarrell | portante: I heard swiftclient w/requests doesn't eventlet properly with SSL | 15:52 |
| notmyname | that wood (sic) have been cool | 15:52 |
| portante | ugh | 15:52 |
| swifterdarrell | lol | 15:52 |
| swifterdarrell | portante: otoh, I haven't used the new v2+ swiftclient much | 15:52 |
| notmyname | portante: creiht is working on that, I think | 15:52 |
| portante | notmyname: does that sound like a manager joke to anybody? ;) | 15:52 |
| creiht | portante: yeah I've been working on that | 15:52 |
| notmyname | portante: lol | 15:52 |
| portante | ;) | 15:52 |
| creiht | so monkey patching socket works reasonably | 15:52 |
| portante | creiht: so I am running into a problem with the in-process functional tests | 15:53 |
| creiht | but swiftclient doesn't use a session, so it recreates connections every time | 15:53 |
| portante | creiht: ugh | 15:53 |
| portante | so that is what you are fixing? | 15:53 |
| swifterdarrell | creiht: oh yeah, I saw a crapton of auth requests w/the new swiftclient but haven't had a chance to look into it | 15:53 |
| creiht | portante: well at this point I'm just trying to completly understand the problem | 15:53 |
| swifterdarrell | creiht: would that be a symptom? (was straight http for me, not ssl) | 15:54 |
| portante | creiht: cool | 15:54 |
| creiht | https://review.openstack.org/#/c/74444/ | 15:54 |
| creiht | lunch time but will bbl | 15:54 |
| swifterdarrell | creiht: i mean, WHO CAN TRUST THAT DIRTY AUTH TOKEN, ANYWAY?! | 15:54 |
| creiht | swifterdarrell: oh dunno | 15:55 |
| portante | creiht: thanks | 15:55 |
| swifterdarrell | creiht: aight, I'll keep an eye out for it, and if it's still around next time swiftclient hits my plate, I'll fix it | 15:55 |
| creiht | swifterdarrell: my main motivation right now is fixing swift-bench performance :) | 15:56 |
| swifterdarrell | creiht: *nod* | 15:56 |
| creiht | ok really going to lunch now :) | 15:56 |
| portante | creiht, swifterdarrell: what I am trying to do with https://review.openstack.org/#/c/66108/ is allow the functional tests to run in-process when requested or no-external config found so that we can run coverage with them and debug problems easier | 15:57 |
| swifterdarrell | creiht: authing every time instead of re-using a token seems like a good way to b0rk performance... | 15:57 |
| swifterdarrell | portante: what's the issue you're seeing there w/swiftclient? | 15:58 |
| portante | creiht, swifterdarrell: what is happening is that the "testFileSizeLimit" functional test creates a separate thread for running the PUT method, where it has a content-length header, but sends no body | 15:58 |
| portante | in the presence of eventlet, that method call fails to actually send any headers, so that in-process server never responds | 15:59 |
| portante | what I want to do is change the functional tests to either: | 15:59 |
| swifterdarrell | portante: could that be running afoul of requests' lack of support for expect-100-continue? | 15:59 |
| swifterdarrell | portante: (nevermind, sounds different) | 16:00 |
| portante | swifterdarrell: I don't think it is sending that header by default, but not sure | 16:00 |
| portante | 1. always use eventlet for the timing of that method | 16:00 |
| portante | 2. run the in-process servers in a separate thread so that the main thread can behave normally (monkey patching sockets needed) | 16:01 |
| swifterdarrell | portante: so the test used to run w/OS threads w/o eventlet monkeypatchy, but now there's OS threads + monkeypatchy and it doesn't work in that new state? | 16:01 |
| portante | swifterdarrell: when it is run in the in-process mode of operation proposed by the above patch | 16:02 |
| portante | and only that test, all other tests work as is | 16:02 |
| portante | #1 would mean a change to how the existing functional tests run, which may not be desirable, hence the question about swiftclient and eventlet vs threads | 16:03 |
| portante | #2 would mean more complicated mechanism for the in-process tests, which is what I am leaning towards, but looking for feedback | 16:04 |
| swifterdarrell | portante: is that the only test using OS threads? or are there other tests doing that which still work w/the in-process biz for some reason? | 16:04 |
| portante | the only test | 16:04 |
| swifterdarrell | ah | 16:04 |
| *** changbl has joined #openstack-swift | 16:04 | |
| portante | for a while I failed to even notice the problem because of the way the in-process functional tests constructed their configuration, the test would be skipped | 16:05 |
| portante | which I thought was normal | 16:05 |
| portante | the test avoids sending any data because we don't want to generate 5GB of data 6 times to test the boundary conditions around the max_file_size setting | 16:06 |
| swifterdarrell | portante: which functional test file is that test in? | 16:06 |
| portante | test/functional/tests.py:TestFile.testFileSizeLimit | 16:06 |
| portante | It is actually run twice, with TestFileUTF8 as well | 16:07 |
| portante | swifterdarrell: I've gotta grab lunch before it gets too late, will be back in a bit | 16:08 |
| swifterdarrell | portante: k | 16:08 |
| *** Longgeek_ has joined #openstack-swift | 16:10 | |
| *** saju_m has joined #openstack-swift | 16:12 | |
| *** tellesnobrega has left #openstack-swift | 16:15 | |
| swifterdarrell | portante: you might try turning that "t.join(seconds)" call in timeout() into a loop whose runtime is bounded by the "seconds" value, but in which a smaller (0.1s?) value is passed into t.join() and also eventlet.sleep() is called to trampoline | 16:15 |
| *** chandan_kumar has joined #openstack-swift | 16:16 | |
| swifterdarrell | portante: I'm guessing that t.join(seconds) is causing the main thread to starve the eventlet hub, making your in-proc servers not read what the child thread's writing | 16:16 |
| swifterdarrell | portante: that seems like a pretty safe change that might clear up the problem. | 16:16 |
| *** RockKuo has quit IRC | 16:19 | |
| zackmdavis | http://zackmdavis.net/blog/2014/04/ode-to-swift/ | 16:19 |
| *** ashish_ has quit IRC | 16:20 | |
| *** dmsimard1 has joined #openstack-swift | 16:25 | |
| *** Guest_ has quit IRC | 16:25 | |
| *** Guest_ has joined #openstack-swift | 16:25 | |
| *** dmsimard has quit IRC | 16:26 | |
| *** zigo has joined #openstack-swift | 16:28 | |
| *** zigo has quit IRC | 16:33 | |
| *** madhuri_ has joined #openstack-swift | 16:33 | |
| *** Longgeek_ has quit IRC | 16:34 | |
| *** ashish_ has joined #openstack-swift | 16:34 | |
| *** zigo has joined #openstack-swift | 16:34 | |
| *** tanee-away is now known as tanee | 16:35 | |
| *** saju_m has quit IRC | 16:35 | |
| *** Guest_ has quit IRC | 16:36 | |
| *** Guest_ has joined #openstack-swift | 16:36 | |
| portante | swifterdarrell: I thought a t.join() would block the main thread, allowing the created thread running the method to proceed | 16:40 |
| swifterdarrell | portante: do your in-proc servers run int he main thread? | 16:40 |
| portante | yes | 16:41 |
| *** zigo has quit IRC | 16:41 | |
| portante | I believe so | 16:41 |
| swifterdarrell | portante: then how will they do anything while the main thread is sitting in a t.join(seconds)? | 16:41 |
| swifterdarrell | portante: ...like read their sockets | 16:41 |
| portante | I think that is what is happenning | 16:41 |
| swifterdarrell | portante: I think you need to simultaneously await that thread /and/ service eventlet in that main thread | 16:42 |
| swifterdarrell | portante: inside the definition of timeout() | 16:42 |
| swifterdarrell | portante: timeout() may just need a tweak to be "eventlet-friendly" while still successfully doing his job | 16:42 |
| swifterdarrell | portante: that's my hypothesis, anyway | 16:42 |
| portante | swifterdarrell: yes, that is what I was thinking of doing with #1 above | 16:43 |
| swifterdarrell | portante: I read your #1 as getting rid of the OS thread | 16:44 |
| portante | the other way to consider this is to create one thread that represents the in-process servers, where it is running an eventlet hub there, and the main thread and other threads can use the normal swiftclient interfaces with threads, no code changes to timeout() | 16:45 |
| portante | that is what I meant by #2 | 16:45 |
| *** madhuri_ has quit IRC | 16:45 | |
| portante | it sounds easier the way you describe #1, so I might try that first | 16:45 |
| portante | swifterdarrell: :) | 16:45 |
| *** zigo has joined #openstack-swift | 16:46 | |
| *** shri has joined #openstack-swift | 16:46 | |
| swifterdarrell | portante: hrm, the only part of swiftclient which uses OS threads is bin/swift, right? | 16:46 |
| portante | I am hoping, but I have not investigated it that deeply to be sure | 16:47 |
| swifterdarrell | portante: iirc, the python libarary part was agnostic... it was "eventlet safe" but could be used w/OS threads (a la bin/swift) or eventlet (a la swift-bench) | 16:47 |
| portante | that is how I understood it | 16:47 |
| swifterdarrell | portante: i think just allowing timeout() to service an eventlet hub instead of going into a long/full t.join() will fix the prob w/little downside | 16:48 |
| swifterdarrell | portante: you shoudl probably use wall-time to exit the loop though | 16:48 |
| portante | swifterdarrell: thanks, I'll try that and see how it works | 16:48 |
| *** mmcardle has quit IRC | 16:48 | |
| portante | swifterdarrell: agreed regarding wall-time | 16:48 |
| swifterdarrell | portante: since with eventlet, your t.join() and evenetlet.sleep() will probably both eat wall time | 16:48 |
| swifterdarrell | portante: and w/o eventlet active, I think the eventlet.sleep() may NOP | 16:49 |
| swifterdarrell | portante: but giving t.join() and eventlet.sleep() each a chance to eat 0.1s in a wall-time-bounded-loop should be more than sufficient precision for this style of testing. | 16:49 |
| portante | swifterdarrell: agreed | 16:50 |
| swifterdarrell | portante: but I wouldn't go so far as to remove the OS threading entirely (which is how I read your #1) | 16:50 |
| portante | I was thinking of doing just that because testFileSizeLimit is the only test that uses it, and it *should* not affect the test | 16:51 |
| *** Guest_ has quit IRC | 16:52 | |
| swifterdarrell | portante: I guess eventlet is already a pre-req of running the functional tests? could one theoretically run just the func tests on a server pointed at a swift cluster w/o eventlet installed and have it work? | 16:52 |
| *** Guest_ has joined #openstack-swift | 16:52 | |
| portante | so usually, you do a python setup.py develop to get the test environment setup properly | 16:53 |
| portante | so in that case, one would have eventlet | 16:53 |
| portante | I am not sure if one could take a plain, non-eventlet install system and run the functional tests only | 16:54 |
| portante | I know that eventlet is required for running the unit tests | 16:54 |
| swifterdarrell | portante: *nod* | 16:54 |
| portante | test/unit/proxy/test_server.py requires it for the same reason | 16:54 |
| swifterdarrell | portante: ya, functional tests would be the only ones able to maybe possibly escape a req. on eventlet because they could talk to Swift on a different host (unlike the unit tests) | 16:55 |
| swifterdarrell | portante: and even if that were the case, adding eventlet as a req for running func tests seems ok | 16:55 |
| swifterdarrell | portante: so i'm indifferent to a full eventlet replacement or a eventlet-threading play-nice loop | 16:56 |
| portante | I'll see how it works out using eventlet with that testFileSizeLimit test and propose that as a separate independent patch for review and consideration | 16:56 |
| swifterdarrell | portante: slight pref for the play-nice, but it's probably more overall complexity for the dubious gain of less-code-change | 16:56 |
| swifterdarrell | portante: k | 16:56 |
| portante | thanks | 16:57 |
| *** Guest_ has quit IRC | 16:58 | |
| *** Guest_ has joined #openstack-swift | 16:58 | |
| *** Guest_ has quit IRC | 16:59 | |
| *** Guest_ has joined #openstack-swift | 16:59 | |
| *** tanee is now known as tanee-away | 17:01 | |
| *** piyush has quit IRC | 17:03 | |
| openstackgerrit | Peter Portante proposed a change to openstack/swift: Move module level func to unit test that uses it https://review.openstack.org/85757 | 17:07 |
| *** ashish_ has quit IRC | 17:13 | |
| *** ashish_ has joined #openstack-swift | 17:14 | |
| openstackgerrit | Peter Portante proposed a change to openstack/swift: Move module level funcs to unit tests that use 'em https://review.openstack.org/85757 | 17:18 |
| *** piousbox has joined #openstack-swift | 17:23 | |
| *** d89 has quit IRC | 17:25 | |
| *** saschpe has quit IRC | 17:34 | |
| notmyname | seagate announced 6TB drives today | 17:35 |
| openstackgerrit | Jenkins proposed a change to openstack/swift: Updated from global requirements https://review.openstack.org/81207 | 17:38 |
| *** saschpe has joined #openstack-swift | 17:41 | |
| *** Guest_ has quit IRC | 17:42 | |
| *** Guest_ has joined #openstack-swift | 17:43 | |
| *** dmsimard has joined #openstack-swift | 17:46 | |
| *** dmsimard1 has quit IRC | 17:46 | |
| *** dmsimard1 has joined #openstack-swift | 17:47 | |
| creiht | zackmdavis: nice! :) | 17:47 |
| *** piyush has joined #openstack-swift | 17:49 | |
| *** Guest_ has quit IRC | 17:49 | |
| *** mmcardle has joined #openstack-swift | 17:49 | |
| *** dmsimard has quit IRC | 17:50 | |
| *** piyush1 has joined #openstack-swift | 17:50 | |
| *** piyush has quit IRC | 17:53 | |
| *** mmcardle has quit IRC | 17:54 | |
| *** ashish_ has quit IRC | 17:54 | |
| *** mlipchuk has quit IRC | 17:55 | |
| *** ashish_ has joined #openstack-swift | 17:55 | |
| *** cheri has joined #openstack-swift | 17:57 | |
| *** annegentle_ has joined #openstack-swift | 17:58 | |
| ashish_ | Hey everyone.Could anyone tell me where in the ssbench is the scenrio directory located. | 17:58 |
| notmyname | ashish_: I think swifterdarrell is on the bus right now. let me talk a look | 18:05 |
| swifterdarrell | ashish_: I usually "sudo updatedb; locate scenarios" | 18:10 |
| notmyname | ashish_: do you mean in the codebase or looking for some other directory somewhere | 18:10 |
| notmyname | swifterdarrell: :-) | 18:10 |
| swifterdarrell | ashish_: notmyname: that's not even a joke response, that's literally what I do and I wrote ssbench | 18:11 |
| *** jamie_h has quit IRC | 18:11 | |
| clayg | glange: you still poking at unify backend logging (saw the rebase) - if not I'll go ahead and stick my +2 back on it | 18:15 |
| glange | clayg: yeah, I'm going to do 2 of the 3 things that notmyname wanted | 18:17 |
| portante | clayg: seems like notmyname had a few comments that could go in future patches? | 18:17 |
| portante | would be nice to get those changes in | 18:17 |
| glange | clayg: update the docs and put a '-' in no additional information | 18:17 |
| clayg | glange: yeah both of those are bonus - thanks! | 18:17 |
| notmyname | glange: thanks! | 18:18 |
| notmyname | I was talking to clayg about a complete overhaul of the storage node log format, but he wisely said that the time to do that is when we (re)address some of the proxy server logging, and thus add some sort of toggle, and then unify everything so it's all the same log format | 18:19 |
| clayg | where wise ~= lazy | 18:19 |
| portante | ah grasshoppah | 18:19 |
| notmyname | :-) | 18:19 |
| portante | boston accent applied | 18:20 |
| clayg | peluse: is 83840 ready to go? | 18:21 |
| clayg | patchbot: 83840 | 18:21 |
| patchbot | clayg: Error: "83840" is not a valid command. | 18:21 |
| clayg | accahpachaphcpahc | 18:21 |
| clayg | patchbot: p 83840 | 18:21 |
| patchbot | clayg: https://review.openstack.org/#/c/83840/ | 18:21 |
| clayg | notmyname: patchbot should include the subject | 18:21 |
| notmyname | clayg: https://github.com/notmyname/Patches/pulls | 18:21 |
| openstackgerrit | Peter Portante proposed a change to openstack/swift: Use eventlet instead of threading for timeout https://review.openstack.org/85782 | 18:26 |
| portante | swifterdarrell: ^^^ | 18:26 |
| portante | gotta head out for a bit, back in a couple to tree hours | 18:26 |
| swifterdarrell | portante: lgtmbidari (looks good to be but i didn't actually run it) | 18:27 |
| *** dmsimard1 has quit IRC | 18:28 | |
| *** dmsimard has joined #openstack-swift | 18:29 | |
| openstackgerrit | Greg Lange proposed a change to openstack/swift: Unify backend logging https://review.openstack.org/83237 | 18:30 |
| *** judd7 has joined #openstack-swift | 18:30 | |
| *** patchbot has quit IRC | 18:32 | |
| *** patchbot has joined #openstack-swift | 18:32 | |
| ashish_ | swiftdarrel thanks could you also tell me what is the process of creation of a JSOn file in ssbench | 18:33 |
| swifterdarrell | ashish_: sure, the "Scenarios" section at https://pypi.python.org/pypi/ssbench/0.2.23 has all the details of what can be defined in a scenario file and their semantics | 18:34 |
| clayg | yuan: i was poking at 82515 and the unittest still seems to pass even w/o the new code? | 18:34 |
| notmyname | glange: you didn't add additional_info to the end of line 102 in logs.rst | 18:35 |
| swifterdarrell | ashish_: note that many of the things defined in teh scenario file may also be overridden on the command-line of ssbench-master | 18:35 |
| *** csd has joined #openstack-swift | 18:35 | |
| glange | notmyname: ok, I'll fix that | 18:35 |
| notmyname | glange: thanks. otherwise, +2 | 18:35 |
| openstackgerrit | Greg Lange proposed a change to openstack/swift: Unify backend logging https://review.openstack.org/83237 | 18:37 |
| glange | that was just a test of the swift review process, if that had been an actual mistake it would have been harder to spot :) | 18:38 |
| clayg | glange: olo | 18:38 |
| ashish_ | swiftdarrel I am now able to locate my scenarios directory.Thanks | 18:38 |
| ashish_ | I also want to know can I create my own JSON file. | 18:39 |
| ashish_ | swiftdarrel I also want to know can I create my own JSON file. | 18:39 |
| *** dmsimard_ has joined #openstack-swift | 18:43 | |
| *** cheri has quit IRC | 18:43 | |
| *** jamie_h has joined #openstack-swift | 18:49 | |
| *** mmcardle has joined #openstack-swift | 18:50 | |
| *** saju_m has joined #openstack-swift | 18:51 | |
| *** patchbot has quit IRC | 18:51 | |
| swifterdarrell | ashish_: any text editor should do, just make sure it's valid JSON | 18:52 |
| swifterdarrell | ashish_: sometimes I like to leave trailing commas in lists and dictionaries but JSON doesn't like that | 18:52 |
| *** patchbot has joined #openstack-swift | 18:52 | |
| ashish_ | Thanks swiftdarrel.Could you tell me the various parameters on the basis of which I can carry my test cases. | 18:54 |
| *** mmcardle has quit IRC | 18:54 | |
| ashish_ | swiftdarrel.one parameter I have chosen is to test ssbench performance when I change the number of replication.Could you tell me other such factors. | 18:55 |
| notmyname | ashish_: there are a lot of factors that can affect performance in swift. most of them should be orthogonal to the replica count. ie choose the replica count based on your durability and dispersion requirements, not performance | 18:57 |
| *** Guest_ has joined #openstack-swift | 18:58 | |
| notmyname | ashish_: you shoudl read through http://docs.openstack.org/developer/swift/deployment_guide.html and http://docs.openstack.org/developer/swift/admin_guide.html for info on some general things that you can tune and how to monitor it | 18:58 |
| ashish_ | notmyname Thanks.But my changing the replicas i want to see the change in the iops. | 18:59 |
| notmyname | ashish_: you want to confirm that 3 replicas has 50% more disk requirements than 2 replicas? | 19:00 |
| ashish_ | notmyname What I want to see is how the number of replicas affect my CRUD profile. | 19:02 |
| peluse | clayg: 83840, yes | 19:04 |
| clayg | peluse: sweet - I'll take a look | 19:04 |
| *** Guest_ has quit IRC | 19:09 | |
| *** csd has quit IRC | 19:10 | |
| *** joeljwright2 has joined #openstack-swift | 19:13 | |
| clayg | peluse: seems on the right track, don't like how the policy_stat seed happens on the container table migration instead of the policy_stat table creation... hrm.... they're in the same method - maybe you can't get half upgraded there... | 19:13 |
| peluse | clayg: I could move that SQL udpate into create_policy_stat table I guess | 19:16 |
| peluse | SQL statement I mean :) | 19:16 |
| clayg | peluse: i'm trying one other thing too | 19:16 |
| peluse | clayg: K, BRB | 19:17 |
| *** mmcardle has joined #openstack-swift | 19:26 | |
| *** dmsimard has quit IRC | 19:29 | |
| *** dmsimard_ is now known as dmsimard | 19:29 | |
| *** dmsimard1 has joined #openstack-swift | 19:30 | |
| notmyname | openssl security bug announced today. seems that ubuntu precise is affected and doesn't yet have a new package available http://www.openssl.org/news/secadv_20140407.txt | 19:32 |
| *** dmsimard has quit IRC | 19:36 | |
| *** dmsimard1 is now known as dmsimard | 19:37 | |
| *** dmsimard has quit IRC | 19:38 | |
| *** dmsimard has joined #openstack-swift | 19:38 | |
| joeljwright2 | Hoping for a bit of help with the gerrit workflow: I submitted a patch to python-swiftclient on Friday last week, but it has failed the check-grenade-dsvm test. It seems to fail when running bin/swift because it fails to import the concurrent.futures module, but futures>=2.1.3 is specified in requirements.txt, so I'm at a bit of a loss. | 19:39 |
| joeljwright2 | https://review.openstack.org/85453 | 19:39 |
| notmyname | joeljwright2: futures isn't in the swiftclient [test-]requirements.txt | 19:42 |
| *** jamie_h has quit IRC | 19:43 | |
| notmyname | joeljwright2: you're saying that it's in the global requirements file? | 19:44 |
| joeljwright2 | yes | 19:44 |
| joeljwright2 | it's added to requirements.txt alongside requests, simplejson and six | 19:44 |
| notmyname | joeljwright2: so while I'm a little surprised that it fails (I figured it would have been imported other places in devstack), I'm glad it does. looks like your patch needs to add it to requirements.txt | 19:45 |
| notmyname | joeljwright2: ah. and I see you already have :-) | 19:45 |
| joeljwright2 | :) | 19:45 |
| * notmyname is unfamiliar with futures | 19:46 | |
| notmyname | joeljwright2: does futures provide the concurrent namespace? | 19:46 |
| notmyname | hmm | 19:47 |
| joeljwright2 | notmyname: yes it should do - it's a backport of concurrent.futures from python >=3.2 | 19:47 |
| notmyname | wfm on my SAIO https://gist.github.com/notmyname/679a08aa0515d0c15527 | 19:47 |
| notmyname | joeljwright2: so that would seem that the problem is with getting it installed | 19:47 |
| notmyname | clarkb: maybe you can help make some searching shorter. do you know why a package added to requirements.txt in a patch would fail grenade? | 19:50 |
| clarkb | notmyname: if there is a non backward compatible path from old version to new version with that requirement I think that may happen | 19:50 |
| clarkb | old version -> new version being of the running openstack cloud | 19:50 |
| notmyname | clarkb: no, this is a net new in this patch | 19:51 |
| clarkb | in that case no I don't know why that may happen | 19:51 |
| *** csd has joined #openstack-swift | 19:54 | |
| *** Longgeek_ has joined #openstack-swift | 19:56 | |
| peluse | clayg: I'm back, let me know if you want me to udpate 83840 to move the policy_table population w/legacy data to the table creation routine... | 19:57 |
| *** Longgeek_ has quit IRC | 19:57 | |
| *** Longgee__ has joined #openstack-swift | 19:57 | |
| *** Longgee__ has quit IRC | 19:58 | |
| *** Longgeek_ has joined #openstack-swift | 19:58 | |
| *** Longgeek_ has quit IRC | 19:59 | |
| *** Longgeek_ has joined #openstack-swift | 19:59 | |
| *** Longgeek_ has quit IRC | 20:00 | |
| *** judd7 has quit IRC | 20:01 | |
| *** dfg_ has joined #openstack-swift | 20:05 | |
| *** Anticime1 has joined #openstack-swift | 20:06 | |
| *** ashish_ has quit IRC | 20:06 | |
| *** mhu1 has joined #openstack-swift | 20:09 | |
| *** wayneseguin has joined #openstack-swift | 20:09 | |
| *** russellb_ has joined #openstack-swift | 20:09 | |
| *** fbo has joined #openstack-swift | 20:09 | |
| *** wayneeseguin has quit IRC | 20:10 | |
| *** russellb has quit IRC | 20:10 | |
| *** leews has quit IRC | 20:10 | |
| *** dfg has quit IRC | 20:10 | |
| *** fbo_away has quit IRC | 20:10 | |
| *** Anticimex has quit IRC | 20:10 | |
| *** mhu has quit IRC | 20:10 | |
| *** russellb_ is now known as russellb | 20:10 | |
| *** mhu1 is now known as mhu | 20:10 | |
| *** wayneseguin is now known as wayneeseguin | 20:10 | |
| *** Longgeek_ has joined #openstack-swift | 20:12 | |
| *** lpabon has joined #openstack-swift | 20:13 | |
| *** Longgeek_ has quit IRC | 20:14 | |
| *** mmcardle has quit IRC | 20:15 | |
| *** Longgeek_ has joined #openstack-swift | 20:16 | |
| *** Longgeek_ has quit IRC | 20:17 | |
| *** Longgeek_ has joined #openstack-swift | 20:17 | |
| *** Longgeek_ has quit IRC | 20:18 | |
| *** Longgeek_ has joined #openstack-swift | 20:18 | |
| *** Longgeek_ has quit IRC | 20:21 | |
| *** Longgeek_ has joined #openstack-swift | 20:23 | |
| *** Longgeek_ has quit IRC | 20:23 | |
| *** Longgeek_ has joined #openstack-swift | 20:24 | |
| *** Longgeek_ has quit IRC | 20:24 | |
| *** Longgeek_ has joined #openstack-swift | 20:25 | |
| *** Longgeek_ has quit IRC | 20:25 | |
| *** Longgeek_ has joined #openstack-swift | 20:26 | |
| *** Longgeek_ has quit IRC | 20:26 | |
| *** occupant has joined #openstack-swift | 20:27 | |
| *** lpabon has quit IRC | 20:28 | |
| *** joeljwright2 has quit IRC | 20:28 | |
| *** csd has quit IRC | 20:28 | |
| *** joeljwright has joined #openstack-swift | 20:33 | |
| *** saju_m has quit IRC | 20:41 | |
| *** Longgeek_ has joined #openstack-swift | 20:42 | |
| *** Longgeek_ has quit IRC | 20:43 | |
| *** Longgeek_ has joined #openstack-swift | 20:44 | |
| *** Longgee__ has joined #openstack-swift | 20:44 | |
| *** tdasilva has left #openstack-swift | 20:45 | |
| *** Longgee__ has quit IRC | 20:45 | |
| *** Longgee__ has joined #openstack-swift | 20:45 | |
| *** Longgee__ has quit IRC | 20:46 | |
| *** ChanServ sets mode: +v dfg_ | 20:47 | |
| *** dfg_ is now known as dfg | 20:47 | |
| *** Longgeek_ has quit IRC | 20:48 | |
| *** csd has joined #openstack-swift | 20:49 | |
| *** Longgeek_ has joined #openstack-swift | 20:50 | |
| *** Longgeek_ has quit IRC | 20:50 | |
| *** Longgeek_ has joined #openstack-swift | 20:51 | |
| *** Longgeek_ has quit IRC | 20:51 | |
| *** Longgeek_ has joined #openstack-swift | 20:52 | |
| *** Longgeek_ has quit IRC | 20:52 | |
| *** Longgeek_ has joined #openstack-swift | 20:53 | |
| *** Longgeek_ has quit IRC | 20:53 | |
| *** Longgeek_ has joined #openstack-swift | 20:55 | |
| *** Anticime1 is now known as Anticimex | 20:55 | |
| *** Longgeek_ has quit IRC | 20:56 | |
| *** Longgee__ has joined #openstack-swift | 20:56 | |
| *** Longgee__ has quit IRC | 20:56 | |
| *** Longgeek_ has joined #openstack-swift | 20:57 | |
| *** Longgeek_ has quit IRC | 20:57 | |
| *** Longgeek_ has joined #openstack-swift | 20:58 | |
| *** Longgeek_ has quit IRC | 20:59 | |
| *** Longgeek_ has joined #openstack-swift | 21:01 | |
| *** Longgeek_ has quit IRC | 21:01 | |
| *** Longgeek_ has joined #openstack-swift | 21:02 | |
| *** Longgeek_ has quit IRC | 21:02 | |
| *** Longgeek_ has joined #openstack-swift | 21:03 | |
| *** JuanManuelOlle has quit IRC | 21:03 | |
| *** Longgeek_ has quit IRC | 21:03 | |
| *** Longgeek_ has joined #openstack-swift | 21:04 | |
| *** Longgeek_ has quit IRC | 21:05 | |
| *** joeljwright has quit IRC | 21:10 | |
| *** Longgeek_ has joined #openstack-swift | 21:11 | |
| *** Longgeek_ has quit IRC | 21:12 | |
| *** Longgeek_ has joined #openstack-swift | 21:12 | |
| *** Longgeek_ has quit IRC | 21:13 | |
| *** Longgeek_ has joined #openstack-swift | 21:13 | |
| openstackgerrit | paul luse proposed a change to openstack/swift: Add Storage Policy Documentation https://review.openstack.org/85824 | 21:13 |
| *** Longgeek_ has quit IRC | 21:14 | |
| clayg | peluse: nice | 21:18 |
| *** mmcardle has joined #openstack-swift | 21:19 | |
| *** Longgeek_ has joined #openstack-swift | 21:21 | |
| *** Longgee__ has joined #openstack-swift | 21:22 | |
| *** Longgeek_ has quit IRC | 21:22 | |
| *** Longgeek_ has joined #openstack-swift | 21:23 | |
| *** mmcardle has quit IRC | 21:24 | |
| *** Longgeek_ has quit IRC | 21:24 | |
| *** Longgee__ has quit IRC | 21:27 | |
| peluse | clayg: just a start :) | 21:39 |
| openstackgerrit | Clay Gerrard proposed a change to openstack/swift: Fix issue with account broker and pre-SPI databases https://review.openstack.org/83840 | 21:39 |
| *** joeljwright has joined #openstack-swift | 21:40 | |
| peluse | ^ way cool clayg | 21:41 |
| clayg | i thought it turned out ok - can you push it through the paces and I'll get notmyname to merge it? | 21:42 |
| clayg | peluse: when you get a chance | 21:42 |
| clayg | peluse: I think i also got him to commit to looking at the policy validation patch and merging master back onto feature/ec | 21:42 |
| *** Longgeek_ has joined #openstack-swift | 21:42 | |
| notmyname | doing it now | 21:42 |
| *** joeljwright1 has joined #openstack-swift | 21:42 | |
| notmyname | Longgeek_: you keep leaving and joining (and filling the channel with connect/disconnect messages) | 21:42 |
| *** Longgeek_ has quit IRC | 21:43 | |
| peluse | clayg: yup, I'll do a little more manual functinoal testing real quick and make sure I can't break it | 21:43 |
| clayg | peluse: even better if you can - then we'll fix THAT and know we have one less way to wreck things! | 21:44 |
| *** joeljwright has quit IRC | 21:45 | |
| peluse | well, we could add a delete at the end of the unit test and assert correct rollups in policy_stat... about to check that manually | 21:45 |
| *** Longgeek_ has joined #openstack-swift | 21:46 | |
| *** Longgeek_ has quit IRC | 21:46 | |
| *** Longgeek_ has joined #openstack-swift | 21:47 | |
| clayg | peluse: oh yeah that'd be nice | 21:47 |
| *** Longgeek_ has quit IRC | 21:47 | |
| *** Longgeek_ has joined #openstack-swift | 21:48 | |
| clayg | peluse: I was also thinking the policy_stat's trigger should go ahead and count containers - at least I *think* that would be useful right? | 21:48 |
| peluse | I asked torgomatic about that a while back and he didn't think we should include it right now | 21:49 |
| *** Longgee__ has joined #openstack-swift | 21:49 | |
| peluse | but he's at home watching cartoons so its up to you I guess :) | 21:49 |
| clayg | peluse: maybe? i think it'll be harder to add later... | 21:49 |
| *** piyush1 has left #openstack-swift | 21:49 | |
| clayg | lol | 21:49 |
| * peluse wonders if torgomatic is lurking... | 21:49 | |
| *** Longgee__ has quit IRC | 21:49 | |
| *** joeljwright1 has quit IRC | 21:50 | |
| *** Longgeek_ has quit IRC | 21:53 | |
| *** changbl has quit IRC | 21:54 | |
| *** Longgeek_ has joined #openstack-swift | 22:03 | |
| *** Longgeek_ has quit IRC | 22:04 | |
| *** ChanServ sets mode: -b godb!*@* | 22:07 | |
| *** joeljwright has joined #openstack-swift | 22:09 | |
| *** Longgeek_ has joined #openstack-swift | 22:14 | |
| *** joeljwright has quit IRC | 22:14 | |
| *** ChanServ sets mode: +b Longgeek_!*@* | 22:14 | |
| *** Longgeek_ was kicked by ChanServ (Banned: continual join/leave spam !T 18h) | 22:14 | |
| *** mmcardle has joined #openstack-swift | 22:15 | |
| *** ChanServ sets mode: -b Longgeek_!*@* | 22:15 | |
| *** mmcardle has quit IRC | 22:17 | |
| *** mmcardle1 has joined #openstack-swift | 22:17 | |
| *** mmcardle1 has quit IRC | 22:21 | |
| openstackgerrit | paul luse proposed a change to openstack/swift: Fix issue with account broker and pre-SPI databases https://review.openstack.org/83840 | 22:23 |
| *** Longgeek_ has joined #openstack-swift | 22:26 | |
| *** ChanServ sets mode: +b Longgeek_!*@* | 22:26 | |
| *** Longgeek_ was kicked by ChanServ (Banned: continual join/leave spam) | 22:26 | |
| peluse | clayg: will be in transit for a bit. wrt adding container count, say the word and I'll add it in. Just added the extra delete test with the above patch change | 22:26 |
| clayg | peluse: we should track per policy container_count, and also include storage_policy_index in the container listings - but we an merge this as is I think... | 22:31 |
| *** Longgee__ has joined #openstack-swift | 22:34 | |
| *** zhiyan_ has quit IRC | 22:34 | |
| *** Longgee__ has quit IRC | 22:35 | |
| *** zhiyan_ has joined #openstack-swift | 22:35 | |
| *** early has quit IRC | 22:42 | |
| *** Trixboxer has quit IRC | 22:43 | |
| *** early has joined #openstack-swift | 22:46 | |
| *** saschpe has quit IRC | 23:00 | |
| *** joeljwright has joined #openstack-swift | 23:10 | |
| *** mkollaro has quit IRC | 23:14 | |
| *** joeljwright has quit IRC | 23:14 | |
| *** mmcardle has joined #openstack-swift | 23:18 | |
| *** gyee has quit IRC | 23:21 | |
| *** mmcardle has quit IRC | 23:22 | |
| openstackgerrit | Peter Portante proposed a change to openstack/swift: Use eventlet instead of threading for timeout https://review.openstack.org/85782 | 23:28 |
| portante | swifterdarrell: ^^^ the above uses eventlet instead of threads for that one test timeout, and it seems like it is a bit less code. | 23:29 |
| notmyname | that sound you hear is all the sysadmins updating their boxes right now (BTW canonical pushed a patch for precise, so go update things) | 23:38 |
| *** joearnold has quit IRC | 23:40 | |
| *** mlanner has quit IRC | 23:40 | |
| *** otherjon has quit IRC | 23:40 | |
| *** swifterdarrell has quit IRC | 23:40 | |
| *** alpha_ori has quit IRC | 23:40 | |
| *** minnear has quit IRC | 23:40 | |
| *** hugokuo has quit IRC | 23:40 | |
| *** ctennis has quit IRC | 23:40 | |
| *** anderstj has quit IRC | 23:40 | |
| *** amandap has quit IRC | 23:40 | |
| *** acorwin has quit IRC | 23:40 | |
| *** zackmdavis has quit IRC | 23:40 | |
| *** acorwin has joined #openstack-swift | 23:43 | |
| *** minnear has joined #openstack-swift | 23:43 | |
| *** ctennis has joined #openstack-swift | 23:43 | |
| *** swifterdarrell has joined #openstack-swift | 23:45 | |
| *** ChanServ sets mode: +v swifterdarrell | 23:45 | |
| *** hugokuo has joined #openstack-swift | 23:47 | |
| *** joearnold has joined #openstack-swift | 23:47 | |
| *** mlanner has joined #openstack-swift | 23:48 | |
| *** alpha_ori has joined #openstack-swift | 23:50 | |
| *** otherjon has joined #openstack-swift | 23:50 | |
| *** zackmdavis has joined #openstack-swift | 23:51 | |
| *** amandap has joined #openstack-swift | 23:52 | |
| *** anderstj has joined #openstack-swift | 23:53 | |
| *** gyee has joined #openstack-swift | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!