Tuesday, 2014-07-22

*** tgohad has joined #openstack-swift00:03
*** tusharsg has quit IRC00:06
*** rameshg87 has left #openstack-swift00:09
*** dmsimard_away is now known as dmsimard00:09
*** tusharsg has joined #openstack-swift00:13
*** annegent_ has joined #openstack-swift00:16
*** tgohad has quit IRC00:16
*** dmsimard is now known as dmsimard_away00:20
*** tongli has quit IRC00:20
*** tongli has joined #openstack-swift00:21
*** annegent_ has quit IRC00:24
*** tdasilva has joined #openstack-swift00:24
*** dmsimard_away is now known as dmsimard00:25
*** dmorita has joined #openstack-swift00:29
*** dmorita has quit IRC00:31
*** dmorita has joined #openstack-swift00:35
*** shri has quit IRC00:39
*** dmsimard is now known as dmsimard_away00:47
*** dmsimard_away is now known as dmsimard01:35
*** nosnos has joined #openstack-swift01:40
*** bkopilov has quit IRC01:41
*** dmsimard is now known as dmsimard_away01:42
*** haomaiwang has joined #openstack-swift01:50
*** kype has joined #openstack-swift01:51
kypeHi guys, when you do a swift update. Does it get pushed to all 3 primary locations as well, or only pushed to one location and then the replicator takes care of updating the replicates?01:51
*** haomaiwang has quit IRC01:52
*** bkopilov has joined #openstack-swift01:53
*** haomaiwa_ has joined #openstack-swift01:53
*** tdasilva has quit IRC01:56
*** haomaiw__ has joined #openstack-swift02:08
*** haomaiwa_ has quit IRC02:11
*** haomaiw__ has quit IRC02:12
*** haomaiwang has joined #openstack-swift02:12
*** zaitcev has quit IRC02:13
*** haomaiwang has quit IRC02:13
*** haomaiwang has joined #openstack-swift02:13
*** tusharsg has quit IRC02:21
openstackgerritA change was merged to openstack/python-swiftclient: Fix context sensitive help for info and tempurl  https://review.openstack.org/10600902:25
*** haomaiw__ has joined #openstack-swift02:29
*** haomaiwang has quit IRC02:32
*** bkopilov has quit IRC02:41
*** bkopilov has joined #openstack-swift02:41
*** tongli has quit IRC03:03
*** tsg has joined #openstack-swift03:12
mattoliveraukype: it will attempt to push to all 3 at once.03:40
*** nosnos has quit IRC03:40
kypemattoliverau: In that case its a consistent storage policy isnt it? Because all updates are synced at once03:41
kypeThere is no chance for one file to remain unupdated03:41
*** ho has joined #openstack-swift03:45
notmynamekype: no that is incorrect. it's possible that not all of the replicas get updated (there are various reasons why). swift is eventually consistent03:50
mattoliveraukype: its not consistent, it'll try and update all three at once and will return a success so long as two are successful. The third will be fixed by the replicator and auditor. Eventually consistent.03:52
kypeOh okay in the case an update to one node fails03:52
kypeAh i understand now, thanks notmyname03:53
kypeand mattoliverau03:53
notmynamemattoliverau: kype: (not to mention what happens to handoff replicas during error conditions or capacity adjustments)03:54
mattoliverauWhat notmyname said :) hopefully my sentences make sense, over been traveling for over 30 hours and my last flight was redirected.. Am now in a different Australian city waiting for a new flight! Woo jet lag!03:55
*** chandankumar has joined #openstack-swift04:03
*** shakamunyi has joined #openstack-swift04:05
*** nosnos has joined #openstack-swift04:06
*** shakamunyi has quit IRC04:08
*** haomaiw__ has quit IRC04:30
*** haomaiwa_ has joined #openstack-swift04:31
*** tsg has quit IRC04:43
*** tsg has joined #openstack-swift04:48
*** anand_ts has joined #openstack-swift04:56
*** ppai has joined #openstack-swift05:16
*** bvandenh has joined #openstack-swift05:19
*** chandankumar has quit IRC05:25
*** patchbot has quit IRC05:28
*** notmyname has quit IRC05:28
*** notmyname has joined #openstack-swift05:30
*** ChanServ sets mode: +v notmyname05:30
*** haomaiwa_ has quit IRC05:30
*** haomaiwang has joined #openstack-swift05:30
*** psharma has joined #openstack-swift05:43
*** bvandenh has quit IRC05:44
*** tsg has quit IRC05:49
*** bkopilov has quit IRC05:54
*** haomai___ has joined #openstack-swift05:55
*** haomaiwang has quit IRC05:58
*** k4n0 has joined #openstack-swift05:59
*** haomai___ has quit IRC06:00
*** chandankumar has joined #openstack-swift06:01
*** haomaiwang has joined #openstack-swift06:01
*** haomaiw__ has joined #openstack-swift06:02
*** haomaiwang has quit IRC06:05
*** gyee has quit IRC06:17
*** chandankumar has quit IRC06:39
*** kype has quit IRC06:42
*** mmcardle has joined #openstack-swift06:46
*** JelleB is now known as a1|away07:10
*** bkopilov has joined #openstack-swift07:33
*** a1|away is now known as JelleB07:34
*** foexle has joined #openstack-swift07:43
*** k4n0 has quit IRC07:51
*** foexle has quit IRC07:54
*** nacim has joined #openstack-swift07:55
*** foexle has joined #openstack-swift07:56
*** k4n0 has joined #openstack-swift08:04
*** bvandenh has joined #openstack-swift08:12
*** els13 has joined #openstack-swift08:29
els13hi everyone, is this the right place to ask newbie swift questions?08:30
*** k4n0 has quit IRC08:32
*** acoles_away is now known as acoles08:33
charzels13: welcome08:36
els13well I hope it is. I'm following some cookbook to set up a openstack test environment. in the chapter of swift, one creates a swift server with all the swift services on it, like proxy, container  object server etc. one also creates four nodes on this server to simulate replication. after running swift-dispersion-populate, the replicator processes go kinda ham and the server is at a load of 6008:37
els13any idea on how I can troubleshoot this, or is this to be expected if 4 nodes run on one VM?08:37
els13charz: thanks and hi :-)08:39
psharmaels13, SAIO setup is definitely not suitable for load testing , however i am not sure what you meant by 'load of 60'08:44
els13the cpu load as reported by linux08:44
els13as in system crawls almost to a halt08:45
*** tsg has joined #openstack-swift08:45
*** k4n0 has joined #openstack-swift08:48
*** tsg has quit IRC08:51
*** ppai has quit IRC09:07
*** k4n0 has quit IRC09:19
*** ppai has joined #openstack-swift09:25
*** openstackgerrit has quit IRC09:31
*** openstackgerrit has joined #openstack-swift09:32
*** haomaiw__ has quit IRC09:33
*** haomaiwang has joined #openstack-swift09:34
els13hmm, actually it's only some part of the container that causes such high cpu load09:37
*** k4n0 has joined #openstack-swift09:37
*** DisneyRicky has quit IRC09:38
*** DisneyRicky has joined #openstack-swift09:46
*** DisneyRicky has quit IRC09:51
*** IRTermite has quit IRC09:54
*** amandap has quit IRC09:54
*** ahale_ has quit IRC09:54
*** dosaboy has quit IRC09:54
*** dmsimard_away has quit IRC09:54
*** creiht has quit IRC09:54
*** yliu has quit IRC09:54
*** bsdkurt has quit IRC09:54
*** mariusv has quit IRC09:54
*** charz has quit IRC09:54
*** ctennis has quit IRC09:54
*** yliu has joined #openstack-swift09:54
*** creiht has joined #openstack-swift09:54
*** ChanServ sets mode: +v creiht09:54
*** mariusv has joined #openstack-swift09:55
*** dosaboy has joined #openstack-swift09:55
*** IRTermite has joined #openstack-swift09:55
*** bsdkurt has joined #openstack-swift09:55
*** ctennis has joined #openstack-swift09:55
*** dmsimardz has joined #openstack-swift09:56
*** dmsimardz is now known as dmsimard09:56
*** ahale has joined #openstack-swift09:56
*** amandap has joined #openstack-swift09:56
*** charz has joined #openstack-swift09:56
*** nosnos has quit IRC09:58
*** nosnos has joined #openstack-swift09:58
*** zslot has joined #openstack-swift10:00
*** nosnos has quit IRC10:03
*** zslot has quit IRC10:07
*** zslot has joined #openstack-swift10:08
*** ctennis has quit IRC10:10
*** ctennis has joined #openstack-swift10:10
*** haomaiw__ has joined #openstack-swift10:19
*** haomaiwang has quit IRC10:23
openstackgerritPawel Palucki proposed a change to openstack/swift: type tradeoff -> trade off  https://review.openstack.org/10863810:33
openstackgerritPawel Palucki proposed a change to openstack/swift: typo tradeoff -> trade off  https://review.openstack.org/10863810:35
*** ho has quit IRC10:43
*** DisneyRicky has joined #openstack-swift11:10
*** nosnos has joined #openstack-swift11:13
*** k4n0 has quit IRC11:28
*** nosnos has quit IRC11:40
*** tongli has joined #openstack-swift11:48
*** dmorita has quit IRC12:13
*** psharma has quit IRC12:24
*** mrsnivvel has joined #openstack-swift12:28
*** mkollaro has joined #openstack-swift12:44
*** erlon has joined #openstack-swift12:48
*** mmcardle has quit IRC12:51
*** anand_ts has quit IRC13:00
*** mmcardle has joined #openstack-swift13:03
*** DisneyRicky has quit IRC13:11
*** tdasilva has joined #openstack-swift13:14
acolesportante: some keystone v3 setup steps here: https://gist.github.com/alistairncoles/ae9d5f92063b58afeb8813:24
acolesportante: if you have keystone running then just the openstackclient pieces are relevant (and the bash script) to setup non-default domain user & project13:25
acolesportante: i am just working through the steps on a devstack machine...(but hit the pki token too long issue :( )13:26
*** minisot has joined #openstack-swift13:28
*** ppai has quit IRC13:32
portante:( thanks acoles13:37
minisothi all, is there any possibility to mark the zone as read-only?13:39
*** sts has joined #openstack-swift13:42
stshello folks. is there a way to specify the region for a user in keystone?13:42
openstackgerritOpenStack Proposal Bot proposed a change to openstack/python-swiftclient: Updated from global requirements  https://review.openstack.org/8925013:48
*** mmcardle has quit IRC13:55
minisotwhat this error "ERROR Container update failed (saving for async update later): 411 response from", how to fix?14:06
*** chandankumar has joined #openstack-swift14:06
acolesportante: added comment to gist https://gist.github.com/alistairncoles/ae9d5f92063b58afeb88 showing two config changes needed to run devstack swift with keystone v314:10
acolesportante: that plus the script on same gist should get you setup with user/project in a non-default domain14:11
*** bszabo_ has joined #openstack-swift14:17
*** annegent_ has joined #openstack-swift14:20
*** annegent_ is now known as annegentle_14:21
*** mrmoje has joined #openstack-swift14:27
*** mmcardle has joined #openstack-swift14:28
*** rook_ has joined #openstack-swift14:37
*** mrsnivvel has quit IRC14:41
*** ujjain has quit IRC14:44
*** bvandenh has quit IRC14:45
*** chandankumar has quit IRC14:48
*** ujjain has joined #openstack-swift14:49
*** bkopilov has quit IRC14:52
*** mrsnivvel has joined #openstack-swift14:54
*** mrsnivvel has quit IRC15:04
*** chandankumar has joined #openstack-swift15:07
*** corvus is now known as jeblair15:07
*** nacim has quit IRC15:09
*** kevinc_ has joined #openstack-swift15:13
*** DisneyRicky has joined #openstack-swift15:16
*** mrsnivvel has joined #openstack-swift15:20
*** rando has quit IRC15:29
*** mdonohoe has joined #openstack-swift15:29
*** aerwin has joined #openstack-swift15:47
*** els13 has quit IRC15:49
*** foexle has quit IRC15:53
notmynamegood morning16:03
notmynameacoles: I see (via email) that you reviewed the EC docs. thanks. if it's good, go ahead and land it to move that along16:06
acolesnotmyname: ok16:08
acolesnotmyname: good morning btw!16:08
*** wer is now known as werbrb16:08
*** annegentle_ has quit IRC16:10
*** annegent_ has joined #openstack-swift16:11
notmynamethis may be the most awesomely ridiculous thing I've seen in a long time: https://www.stackmonkey.com/blog/for-bitcoin-from-a-twitter-marketing-bot/16:12
notmynameacoles: looks like your fix for the context sensitive help has landed in swiftclient. time for a release16:14
*** tdasilva has quit IRC16:17
acolesnotmyname: yup, and the SP one for upload.16:20
*** mrmoje has quit IRC16:21
*** mwstorer has joined #openstack-swift16:25
*** annegent_ has quit IRC16:26
notmynameso according to twitter, the opening oscon keynote was given by a 14 year old who just recently open-sourced a bioinformatics library in scala16:27
*** bszabo_ has quit IRC16:31
*** bszabo has joined #openstack-swift16:31
*** gyee has joined #openstack-swift16:34
openstackgerritA change was merged to openstack/swift: Add X-Delete-At/After support to FormPost  https://review.openstack.org/10410116:36
*** bszabo has quit IRC16:37
notmynamerelease done. https://pypi.python.org/pypi/python-swiftclient/2.2.016:41
*** bkopilov has joined #openstack-swift16:43
openstackgerritOpenStack Proposal Bot proposed a change to openstack/swift: Updated from global requirements  https://review.openstack.org/8873616:45
*** acoles is now known as acoles_away16:49
*** aerwin has quit IRC16:51
*** tdasilva has joined #openstack-swift16:59
notmynameok, so swiftclient 2.2.0 now can generate tempurls. and ironic needs that. so they bump the global requirements file to specify python-swiftclient>=2.2.0. and now every project is supposed to upgrade their requirements file17:05
swifterdarrellnotmyname: lol?17:15
torgomaticnotmyname: you know, I think that's the silver lining to losing the auto-abandon bot... I can mute those global-requirements threads in Gmail, and I never have to see those messages :)17:18
*** mmcardle has quit IRC17:23
*** shri has joined #openstack-swift17:28
*** chandankumar has quit IRC17:34
openstackgerritA change was merged to openstack/swift: Initial Erasure Code Docs  https://review.openstack.org/10471317:35
*** fifieldt has joined #openstack-swift17:35
notmynamepeluse_: ^^ yay!17:35
notmynameacoles_away: thanks :-)17:35
*** zslot has quit IRC17:59
*** annegent_ has joined #openstack-swift18:10
*** annegent_ has quit IRC18:22
*** MooingLemur has quit IRC18:31
*** mrmoje has joined #openstack-swift18:33
*** werbrb is now known as wer18:33
*** fifieldt has quit IRC18:39
*** kevinc_ has quit IRC18:44
*** shakamunyi has joined #openstack-swift18:56
*** fifieldt has joined #openstack-swift18:58
*** MooingLemur has joined #openstack-swift19:10
*** pberis has quit IRC19:15
*** fifieldt has quit IRC19:17
*** fifieldt has joined #openstack-swift19:17
*** fifieldt has joined #openstack-swift19:18
*** fifieldt has quit IRC19:18
*** fifieldt has joined #openstack-swift19:18
*** kevinc_ has joined #openstack-swift19:20
*** pberis has joined #openstack-swift19:22
*** mkollaro has quit IRC19:24
*** pberis has quit IRC19:31
*** pberis has joined #openstack-swift19:32
*** kevinc_ has quit IRC19:41
*** shakamunyi has quit IRC19:42
*** fifieldt has quit IRC19:43
*** pberis has quit IRC19:46
*** pberis has joined #openstack-swift19:49
*** fausto has joined #openstack-swift20:13
*** fifieldt has joined #openstack-swift20:15
peluse_notmyname:  awesome!  having docs as the first patch on the EC branch is a great start this time I think!20:31
peluse_BTW nice job everyone who worked on the latest swiftclient, some good stuff in there....20:32
*** DisneyRicky has quit IRC20:37
*** DisneyRicky has joined #openstack-swift20:43
*** kevinc_ has joined #openstack-swift21:04
*** foexle has joined #openstack-swift21:14
*** dmsimard is now known as dmsimard_away21:26
*** rook_ has quit IRC21:26
*** zaitcev has joined #openstack-swift21:30
*** ChanServ sets mode: +v zaitcev21:30
mattoliverauMorning all21:32
*** shakamunyi has joined #openstack-swift21:32
notmynamemattoliverau: good morning21:33
mattoliverauBack in oz and in my normal time zone (well in psudo German Australian timezone) :p21:35
mattoliverauNice work on swiftclient all :)21:35
notmynameya, it's nice to have those last features landed. :-)21:35
*** foexle has quit IRC21:39
openstackgerritPawel Palucki proposed a change to openstack/swift: typo tradeoff -> trade-off  https://review.openstack.org/10863821:47
peluse_^ so not big deal but I'm pretty sure "tradeoff" is a valid word no?21:48
openstackgerritPawel Palucki proposed a change to openstack/swift: typo tradeoff -> trade-off  https://review.openstack.org/10863821:49
*** shakamunyi has quit IRC21:56
*** tongli has quit IRC21:59
openstackgerritJohn Dickinson proposed a change to openstack/swift: Multithread optimization for object updater  https://review.openstack.org/7594122:05
notmyname^^ "just" a rebase, but please check that I did it correctly (tests pass, ship it!)22:06
*** tdasilva has quit IRC22:07
*** fifieldt has quit IRC22:13
torgomaticnotmyname: it worries me a little that there were two dict.copy() calls in the old code but only one in the new22:15
*** pberis has quit IRC22:18
notmyname(tradeoff vs trade-off) vs (talking about it). I merged it22:27
*** fifieldt has joined #openstack-swift22:28
peluse_well-done :)22:35
*** fausto has quit IRC22:57
*** bszabo has joined #openstack-swift23:01
*** bszabo has quit IRC23:06
openstackgerritA change was merged to openstack/swift: Add v2 API to list endpoints middleware  https://review.openstack.org/10166523:33
notmynamemattoliverau: around?23:41
mattoliveraunotmyname: yup :)23:42
notmynamemattoliverau: tell me about https://review.openstack.org/#/c/101743/23:42
mattoliveraunotmyname: was about to ask if I could add a topic to the next meeting agenda?23:43
mattoliverauI'll take a look23:43
notmynamemattoliverau: it's your patch23:43
notmynamemattoliverau: did we not have any tests testing a 499 response?23:43
mattoliverauHold on let me move to a laptop were I can type faster :)23:44
mattoliveraunotmyname: it was a resonse to a bug, where there were no tests covering partial writes being handled (or tested) in the object sever. Seeing as there maybe seperate implementations of DiskFile, I thought it mights be imported to still make sure partial wirtes are handled correctly. So I've added 2 tests, mocking the number of bytes written. One which failes in the first chuck and one that fails in the second (to test23:49
mattoliverauchuncking in the object server).23:49
mattoliveraus/imported/important.23:50
notmynamemattoliverau: there is already the test_short_body() test in there that seems to do the same thing. well, I like your's a little better23:51
mattoliveraunotmyname: oh yeah, that probably does something similar, I've just been working through swift bugs and attempting to fix them, I should look a bit more closely to see if there is already some cross over existing. Sorry.23:53
notmynamemattoliverau: and there is only one place in the object server that returns the HTTPClientDisconnect. what do you mean byt he 2 tests failing in 2 different places?23:53
notmynamemattoliverau: I'm looking through swifterdarrell's bug report now23:54
mattoliveraunotmyname: I'm testing the one place twice, but wanted to do better bounds checking, so I fail the first straight away. And the second in a later chunk.23:55
mattoliverauThus I need to make the second test body chunk_size + 6. And then partial write of chunk_size + 5.23:56
notmynameya that makes sense and sounds reasonable23:57
*** fifieldt has quit IRC23:57
openstackgerritA change was merged to openstack/swift: typo tradeoff -> trade-off  https://review.openstack.org/10863823:58
*** kevinc_ has quit IRC23:59

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!