Thursday, 2015-10-15

*** proteusguy_ has quit IRC00:00
*** klrmn1 has quit IRC00:00
*** pberis has quit IRC00:00
*** jodah has quit IRC00:00
*** remix_tj has quit IRC00:00
*** pberis1 is now known as pberis00:00
*** jodah has joined #openstack-swift00:01
*** remix_tj has joined #openstack-swift00:02
*** jrichli_ has joined #openstack-swift00:08
openstackgerritOpenStack Proposal Bot proposed openstack/swift: Updated from global requirements  https://review.openstack.org/8873600:09
*** jrichli has quit IRC00:10
*** setmason has joined #openstack-swift00:11
openstackgerritMerged openstack/swift: Fix replicator intersection exception when sync data to remote regions.  https://review.openstack.org/23269600:12
*** [1]eranrom has joined #openstack-swift00:14
*** eranrom has quit IRC00:16
*** [1]eranrom is now known as eranrom00:16
*** wbhuber_ has joined #openstack-swift00:17
*** setmason has quit IRC00:17
*** diogogmt has quit IRC00:21
*** thumpba has quit IRC00:22
*** thumpba has joined #openstack-swift00:30
*** _hrou_ has quit IRC00:31
*** garthb has quit IRC00:32
*** thumpba has quit IRC00:34
*** m_kazuhiro has joined #openstack-swift00:41
*** setmason has joined #openstack-swift00:54
*** thumpba has joined #openstack-swift01:03
*** klrmn2 has quit IRC01:09
*** wbhuber has quit IRC01:11
*** thumpba has quit IRC01:12
*** hrou has joined #openstack-swift01:14
*** lpabon has quit IRC01:23
*** setmason has quit IRC01:27
*** thumpba has joined #openstack-swift01:27
*** yuan has quit IRC01:43
*** thumpba has quit IRC01:43
*** yuan has joined #openstack-swift01:44
*** links has joined #openstack-swift01:46
*** thumpba has joined #openstack-swift01:50
*** diogogmt has joined #openstack-swift01:57
*** kei_yama has quit IRC02:04
*** kei_yama has joined #openstack-swift02:05
*** jkugel has joined #openstack-swift02:09
*** yunpengli has joined #openstack-swift02:09
*** yunpengli has quit IRC02:09
*** wbhuber_ has quit IRC02:10
*** setmason has joined #openstack-swift02:14
*** janonymous_ has quit IRC02:15
*** dmorita has joined #openstack-swift02:19
*** setmason has quit IRC02:33
*** klrmn1 has joined #openstack-swift02:34
*** blmartin_ has quit IRC02:46
*** siva_krishnan has joined #openstack-swift02:49
openstackgerritRobert Francis proposed openstack/swift: WIP: Symlink implementation. (work in progress)  https://review.openstack.org/23216202:49
*** sanchitmalhotra has joined #openstack-swift02:58
*** silor has joined #openstack-swift03:01
*** trifon has joined #openstack-swift03:06
*** kei_yama has quit IRC03:07
*** kei_yama has joined #openstack-swift03:08
*** setmason has joined #openstack-swift03:10
*** setmason has quit IRC03:15
*** DericHorn-HP has quit IRC03:16
*** darrenc is now known as darrenc_afk03:24
*** diogogmt has quit IRC03:27
*** haomaiwang has joined #openstack-swift03:44
*** m_kazuhiro has quit IRC03:47
*** m_kazuhiro has joined #openstack-swift03:48
*** trifon has quit IRC03:53
*** haomaiwang has quit IRC04:01
*** haomaiwang has joined #openstack-swift04:01
*** jrichli_ has quit IRC04:02
*** haomaiwang has quit IRC04:06
*** DericHorn-HP has joined #openstack-swift04:06
*** setmason has joined #openstack-swift04:11
*** DericHorn-HP has quit IRC04:11
*** yuan has quit IRC04:16
*** yuan has joined #openstack-swift04:17
*** ppai has joined #openstack-swift04:17
*** darrenc_afk is now known as darrenc04:24
*** zhill has joined #openstack-swift04:24
*** klrmn1 has quit IRC04:29
*** setmason has quit IRC04:33
*** hrou has quit IRC04:40
*** setmason has joined #openstack-swift04:44
*** setmason_ has joined #openstack-swift04:47
*** setmason has quit IRC04:49
*** setmason_ is now known as setmason04:49
*** trifon has joined #openstack-swift04:57
*** sanchitmalhotra1 has joined #openstack-swift05:06
*** sanchitmalhotra has quit IRC05:07
*** DericHorn-HP has joined #openstack-swift05:12
*** bing-hu has joined #openstack-swift05:12
*** sanchitmalhotra has joined #openstack-swift05:15
*** sanchitmalhotra1 has quit IRC05:17
*** sanchitmalhotra1 has joined #openstack-swift05:22
*** sanchitmalhotra has quit IRC05:24
*** thumpba has quit IRC05:26
*** zaitcev has quit IRC05:27
*** breitz has quit IRC05:36
*** breitz has joined #openstack-swift05:36
*** thumpba has joined #openstack-swift05:36
*** thumpba has quit IRC05:36
*** thumpba has joined #openstack-swift05:37
*** SkyRocknRoll has joined #openstack-swift05:38
*** thumpba has quit IRC05:38
*** setmason has quit IRC05:46
*** setmason has joined #openstack-swift05:47
*** bing-hu has quit IRC05:52
*** sanchitmalhotra has joined #openstack-swift05:56
*** sanchitmalhotra1 has quit IRC05:58
*** 16WAAIQMV has joined #openstack-swift06:12
*** haomaiwang has joined #openstack-swift06:19
*** 16WAAIQMV has quit IRC06:21
*** rohit_ has quit IRC06:24
*** setmason has quit IRC06:25
*** hezhiqiang has joined #openstack-swift06:26
*** breitz has quit IRC06:28
*** breitz has joined #openstack-swift06:28
*** silor1 has joined #openstack-swift06:29
*** silor has quit IRC06:29
*** silor1 is now known as silor06:29
*** eranrom has quit IRC06:38
*** jamielennox is now known as jamielennox|away06:39
*** siva_krishnan has quit IRC06:43
*** rledisez has joined #openstack-swift07:10
*** jmccarthy has quit IRC07:21
*** jmccarthy has joined #openstack-swift07:21
*** janonymous has quit IRC07:24
*** jmccarthy has quit IRC07:24
*** jmccarthy has joined #openstack-swift07:30
*** zhill has quit IRC07:32
*** haomaiwang has quit IRC07:48
*** haomaiwa_ has joined #openstack-swift07:50
*** geaaru has joined #openstack-swift07:51
*** peterlisak has joined #openstack-swift07:52
*** zhill has joined #openstack-swift07:57
*** zhill has quit IRC08:02
*** marzif has joined #openstack-swift08:12
*** jordanP has joined #openstack-swift08:17
*** marzif has quit IRC08:19
*** marzif has joined #openstack-swift08:20
*** acoles_ is now known as acoles08:23
*** sanchitmalhotra1 has joined #openstack-swift08:26
*** sanchitmalhotra has quit IRC08:28
*** jistr has joined #openstack-swift08:30
*** aix has joined #openstack-swift08:35
*** haomaiwa_ has quit IRC08:44
*** haomaiwang has joined #openstack-swift08:47
*** silor has quit IRC08:56
*** haomaiwang has quit IRC09:01
*** marzif_ has joined #openstack-swift09:04
*** marzif has quit IRC09:04
*** ppai has quit IRC09:05
*** haomaiwa_ has joined #openstack-swift09:06
*** sanchitmalhotra has joined #openstack-swift09:11
*** sanchitmalhotra1 has quit IRC09:13
*** ppai has joined #openstack-swift09:18
*** SkyRocknRoll has quit IRC09:23
*** ppai has quit IRC09:34
*** SkyRocknRoll has joined #openstack-swift09:36
*** haomaiwa_ has quit IRC09:39
*** haomaiwang has joined #openstack-swift09:39
*** aix has quit IRC09:43
*** ppai has joined #openstack-swift09:46
*** haomaiwang has quit IRC09:46
*** links has quit IRC09:47
*** haomaiwang has joined #openstack-swift09:47
*** sanchitmalhotra1 has joined #openstack-swift09:52
*** sanchitmalhotra has quit IRC09:54
*** aix has joined #openstack-swift09:58
openstackgerritRico Lin proposed openstack/python-swiftclient: improve readme contents  https://review.openstack.org/23520909:58
openstackgerritAlistair Coles proposed openstack/swift: Add backend subrequest tests for staticweb  https://review.openstack.org/23401210:00
*** ppai has quit IRC10:03
*** hezhiqiang has quit IRC10:04
*** ppai has joined #openstack-swift10:16
*** janonymous has joined #openstack-swift10:17
*** silor has joined #openstack-swift10:21
*** dmorita has quit IRC10:25
*** aix has quit IRC10:26
*** haomaiwang has quit IRC10:29
*** ppai has quit IRC10:36
*** aix has joined #openstack-swift10:37
*** m_kazuhiro has quit IRC10:45
*** ppai has joined #openstack-swift10:48
*** weihan has joined #openstack-swift10:51
*** sanchitmalhotra1 has quit IRC10:55
*** ppai has quit IRC11:03
*** petertr7_away has quit IRC11:03
*** petertr7_away has joined #openstack-swift11:06
*** petertr7_away is now known as petertr711:06
*** lpabon has joined #openstack-swift11:07
*** SkyRocknRoll has quit IRC11:07
*** ppai has joined #openstack-swift11:17
*** mahatic has joined #openstack-swift11:17
*** ppai_ has joined #openstack-swift11:17
*** ppai_ has quit IRC11:18
*** mahatic has quit IRC11:26
*** ppai has quit IRC11:26
*** ppai has joined #openstack-swift11:39
*** ho has quit IRC11:40
*** ppai has quit IRC11:51
*** weihan_ has joined #openstack-swift11:55
*** weihan has quit IRC11:56
*** marzif_ has quit IRC11:57
*** marzif_ has joined #openstack-swift11:58
*** ppai has joined #openstack-swift12:05
*** haomaiwang has joined #openstack-swift12:26
*** Guest22646 has quit IRC12:28
*** marzif_ has quit IRC12:28
*** marzif_ has joined #openstack-swift12:28
*** haomaiwang has quit IRC12:31
*** mahatic has joined #openstack-swift12:31
*** aix has quit IRC12:37
*** mahatic has quit IRC12:37
*** aix has joined #openstack-swift12:37
*** mahatic has joined #openstack-swift12:39
*** jkugel has quit IRC12:41
*** haomaiwa_ has joined #openstack-swift12:48
*** resker has joined #openstack-swift12:52
*** haomaiwa_ has quit IRC12:55
*** haomaiwang has joined #openstack-swift12:55
*** esker has quit IRC12:56
*** kei_yama has quit IRC12:59
*** resker has quit IRC12:59
*** haomaiwang has quit IRC13:01
*** haomaiwang has joined #openstack-swift13:01
*** marzif_ has quit IRC13:02
*** marzif_ has joined #openstack-swift13:03
*** marzif_ has quit IRC13:08
*** silor has quit IRC13:20
*** weihan_ has quit IRC13:23
*** mahatic has quit IRC13:24
*** janonymous_ has joined #openstack-swift13:24
*** hezhiqiang has joined #openstack-swift13:26
*** pchng64 has quit IRC13:27
*** jkugel has joined #openstack-swift13:28
*** hrou has joined #openstack-swift13:29
*** rohit_ has joined #openstack-swift13:32
*** setmason has joined #openstack-swift13:33
*** trifon has quit IRC13:33
*** trifon has joined #openstack-swift13:35
*** dustins has joined #openstack-swift13:46
*** hrou_ has joined #openstack-swift13:49
*** janonymous_ has quit IRC13:49
openstackgerritRobert Francis proposed openstack/swift: WIP: Symlink implementation. (work in progress)  https://review.openstack.org/23216213:51
*** trifon has quit IRC13:51
*** pchng has joined #openstack-swift13:51
*** DericHorn-HP has quit IRC13:56
*** haomaiwang has quit IRC14:01
*** haomaiwang has joined #openstack-swift14:01
*** peterlisak has quit IRC14:02
*** hrou has quit IRC14:03
*** hrou_ is now known as hrou14:04
*** breitz has quit IRC14:05
*** ppai has quit IRC14:08
*** janonymous_ has joined #openstack-swift14:10
*** setmason has quit IRC14:10
*** peterlisak has joined #openstack-swift14:11
*** setmason has joined #openstack-swift14:12
*** wbhuber has joined #openstack-swift14:16
*** pchng64 has joined #openstack-swift14:21
*** jrichli has joined #openstack-swift14:24
*** pchng has quit IRC14:24
*** Ailus has joined #openstack-swift14:32
*** siva_krishnan has joined #openstack-swift14:33
*** DericHorn-HP has joined #openstack-swift14:34
*** garthb has joined #openstack-swift14:38
*** jrichli has quit IRC14:47
*** setmason has quit IRC14:47
*** petertr7 is now known as petertr7_away14:48
*** petertr7_away is now known as petertr714:51
acolescschwede: oic! (staticweb patch)14:57
*** minwoob has joined #openstack-swift14:58
*** jlhinson has joined #openstack-swift15:00
*** haomaiwang has quit IRC15:01
*** haomaiwa_ has joined #openstack-swift15:01
*** aix has quit IRC15:02
*** setmason has joined #openstack-swift15:04
openstackgerritRobert Francis proposed openstack/swift: WIP: Symlink implementation. (work in progress)  https://review.openstack.org/23216215:06
*** dleute has joined #openstack-swift15:07
*** dleute has quit IRC15:13
*** dleute has joined #openstack-swift15:14
*** siva_krishnan has quit IRC15:19
*** trifon has joined #openstack-swift15:20
openstackgerritRobert Francis proposed openstack/swift: WIP: Symlink implementation. (work in progress)  https://review.openstack.org/23216215:21
*** pdardeau has joined #openstack-swift15:21
*** diogogmt has joined #openstack-swift15:22
*** jrichli has joined #openstack-swift15:23
*** zhill has joined #openstack-swift15:25
*** mahatic has joined #openstack-swift15:30
*** openstackgerrit has quit IRC15:31
*** openstackgerrit has joined #openstack-swift15:31
*** nadeem_ has joined #openstack-swift15:32
*** nadeem_ has quit IRC15:33
*** siva_krishnan has joined #openstack-swift15:33
*** nadeem_ has joined #openstack-swift15:35
*** minwoob has quit IRC15:47
*** gyee has joined #openstack-swift15:48
*** diogogmt has quit IRC15:48
*** diogogmt has joined #openstack-swift15:49
*** peterlisak has quit IRC15:55
*** haomaiwa_ has quit IRC16:01
*** haomaiwang has joined #openstack-swift16:01
*** peluse has quit IRC16:14
*** peluse_ is now known as peluse16:14
*** minwoob has joined #openstack-swift16:20
*** zhill has quit IRC16:21
*** jistr has quit IRC16:33
*** dleute has quit IRC16:38
*** zhill has joined #openstack-swift16:40
*** mac_ified has joined #openstack-swift16:41
*** zhill has quit IRC16:41
*** petertr7 is now known as petertr7_away16:41
*** zhill has joined #openstack-swift16:41
*** lcurtis has joined #openstack-swift16:46
*** gaurangt has joined #openstack-swift16:46
*** khoah64 has joined #openstack-swift16:49
*** dleute has joined #openstack-swift16:53
*** rohit_ has quit IRC16:54
*** eariasn has joined #openstack-swift16:54
notmynameliberty is officially released16:55
*** dustins has quit IRC16:56
notmynamehurricanerix: what needs to happen to get the testr tests going? I had hoped to get that a while back, but you've got it marked WIP16:57
*** dustins has joined #openstack-swift16:58
*** DericHorn-HP has quit IRC17:00
*** haomaiwang has quit IRC17:01
*** garthb_ has joined #openstack-swift17:01
*** haomaiwang has joined #openstack-swift17:01
*** garthb has quit IRC17:03
openstackgerritRobert Francis proposed openstack/swift: WIP: Symlink implementation. (work in progress)  https://review.openstack.org/23216217:06
*** mac_ified has quit IRC17:07
hurricanerixnotmyname: sorry, I got distracted by other things.  I meant to look over it a little bit more because I am paranoid about breaking things.17:09
*** jordanP has quit IRC17:10
hurricanerixI'll see if I can find some time to go over it again.  and if not, it's a long flight to Japan... :)17:11
notmynameheh :-)17:11
*** bitblt has joined #openstack-swift17:13
*** gyee has quit IRC17:14
*** rledisez has quit IRC17:14
*** jrichli has quit IRC17:18
*** gaurangt1 has joined #openstack-swift17:19
*** gaurangt has quit IRC17:21
*** dleute has quit IRC17:25
*** dleute has joined #openstack-swift17:26
*** klrmn1 has joined #openstack-swift17:29
*** jrichli has joined #openstack-swift17:30
*** dleute has quit IRC17:31
*** dleute has joined #openstack-swift17:31
*** mahatic has quit IRC17:34
janonymous_Wow.. Liberty!!17:34
janonymous_notmyname: Mitaka started now ?17:35
*** dleute_ has joined #openstack-swift17:35
*** dleute has quit IRC17:35
notmynamejanonymous_: sortof. the openstack release is more of a time on the calendar for dev work. roughly starts with a summit. ends with a release (like today's). but aside for having one release that lines up with the end of that calendar time, swift's release doesn't have "cycles" per se.17:37
notmynameswift's releases are more of "work on stuff that needs to be done and we'll release it when it's done."17:37
*** petertr7_away is now known as petertr717:38
*** janonymous__ has joined #openstack-swift17:38
*** acoles is now known as acoles_17:39
*** jrichli has quit IRC17:39
*** janonymous_ has quit IRC17:39
janonymous__Ohhkay.. and the design sessions include all the things needs to be done in next release if i am ryt17:39
*** dleute_ has quit IRC17:40
*** dleute has joined #openstack-swift17:41
janonymous__I am sorry to ask such basic questions but i havn't attended ny openstack summit before....EC was the most imp and major thing of this release, is there something like that planned in M  also?17:42
claygtdasilva: can you dig up the old versioned write upgrade test spreadsheet?17:43
claygtdasilva: heh nm, https://docs.google.com/spreadsheets/d/1hyo3Q7SS70kG0ZCR3NvXJZ6ikeqqQ04RcRYjOLb44m8/edit#gid=017:45
tdasilvaclayg: problems?17:45
*** dleute_ has joined #openstack-swift17:46
*** dustins has quit IRC17:47
*** dleute_ has quit IRC17:47
*** dleute has quit IRC17:47
*** gaurangt has joined #openstack-swift17:51
*** dustins has joined #openstack-swift17:53
*** gaurangt has left #openstack-swift17:53
*** gaurangt1 has quit IRC17:54
*** geaaru has quit IRC17:55
*** zhill has quit IRC17:57
*** zhill has joined #openstack-swift18:00
*** haomaiwang has quit IRC18:01
*** haomaiwang has joined #openstack-swift18:01
*** david-lyle has quit IRC18:01
*** siva_krishnan has quit IRC18:02
*** david-lyle has joined #openstack-swift18:02
*** earlephilhower has joined #openstack-swift18:03
*** siva_krishnan has joined #openstack-swift18:05
claygtdasilva: nah it's *all* good!18:08
*** barker has joined #openstack-swift18:09
*** zhill_ has joined #openstack-swift18:10
*** zhill has quit IRC18:10
*** alejandrito has joined #openstack-swift18:14
*** bitblt has quit IRC18:16
*** zhill_ has quit IRC18:17
*** zhill has joined #openstack-swift18:17
*** garthb_ has quit IRC18:21
claygtdasilva: well then again - i'm reading that matrix on https://docs.google.com/spreadsheets/d/1hyo3Q7SS70kG0ZCR3NvXJZ6ikeqqQ04RcRYjOLb44m8/edit#gid=0 and I don't think it's 100% correct with what ended up ladning?18:23
clayge.g. G13 - when x-versions-location is *set* and container-server allow_versions is *on* - regadless of the middleware config option versioned writes *will* be allowed?18:24
claygtdasilva: like it's not a ternary state (enabled, disabled, do-whatever-container-config-setting-says) - it's just enabled/disabled - and disabled always does what the container-server says - right?18:25
claygor is there a difference between the config of the automatically inserted middleware and the behavior of a middleware explicitly published in the pipleline and set to allow_versioned_writes = false?18:26
*** barker has quit IRC18:26
tdasilvaclayg: one sec...I remember kota_ found some scenarios towards the end of the review cycle that forced us to make some changes, not sure if it's on the spreadsheet18:27
tdasilvabut we did document it18:27
tdasilvaone sec18:27
*** barker has joined #openstack-swift18:27
claygtdasilva: you're the best!18:27
tdasilvaclayg: I'm reading the following link and comparing to what you wrote: http://docs.openstack.org/developer/swift/middleware.html#how-to-enable-object-versioning-in-a-swift-cluster18:29
*** barker has quit IRC18:32
tdasilvaclayg: if admin sets allow_versioned_writes in the middleware config, then middleware is disabled regardless of x-versions-location being set and container-server allow_versions being *on*18:33
tdasilvaclayg: if admin sets allow_versioned_writes to False18:33
claygtdasilva: whoa - so it *is* ternary18:33
clayghrmm... i'm not sure I see immediately where the code differentiates from "None" and "False" ?18:34
*** marzif has joined #openstack-swift18:35
claygcomment L#420 - "if allow_versioned_writes is not set in the configuration files but 'versions' is configured, enable feature to maintain backwards compatibility"18:36
tdasilvacorrect18:36
*** marzif has quit IRC18:36
claygthe check is just "not allow_versioned_writes" - I don't think it's differentiating between "automatically inserted/not set" and "explicitly inserted/disabled"18:36
*** marzif has joined #openstack-swift18:36
*** setmason has quit IRC18:37
claygwhich is good for me - i hate ternary options - and I don't think I see anything in the docs that conflict with that behavior - but the spreadsheet really did seem to be making a distinction - and you also just echo'd the same...18:38
clayg... maybe I should test it :P18:38
tdasilvatrying to get my head around it agian18:38
tdasilvaone sec18:39
openstackgerritSamuel Merritt proposed openstack/swift: Interpolate the explanation string not whole HTML body  https://review.openstack.org/23330018:39
*** setmason has joined #openstack-swift18:39
tdasilvaclayg: so the way I'm reading the code is...if allow_versioned_writes is not set, then is_enabled is False in line 392, unless it is set to True in line 424 (which would mean it was enabled in the container info)18:42
tdasilvaclayg: sorry...you are correct18:44
tdasilvathis backwards compatibility code is confusing18:45
*** garthb_ has joined #openstack-swift18:51
*** hezhiqiang has quit IRC18:54
claygtdasilva: agreed!18:55
claygtdasilva: can you update that chart and I'll double check you?  (or vice versa?)18:55
claygtdasilva: https://docs.google.com/spreadsheets/d/1hyo3Q7SS70kG0ZCR3NvXJZ6ikeqqQ04RcRYjOLb44m8/edit#gid=018:55
*** subscope has joined #openstack-swift18:56
*** setmason has quit IRC18:56
*** haomaiwang has quit IRC18:57
*** lastops has joined #openstack-swift18:57
*** lastops has quit IRC18:57
*** setmason has joined #openstack-swift18:58
*** zhill has quit IRC19:00
*** jkugel has quit IRC19:00
*** dkleute has joined #openstack-swift19:00
*** zhill has joined #openstack-swift19:06
*** eikke has quit IRC19:08
*** eikke has joined #openstack-swift19:09
*** lyrrad has quit IRC19:11
*** lyrrad has joined #openstack-swift19:22
claygtdasilva: nm, spreadsheet is correct - i'm an idiot - it's totally ternary and well described in the comments in https://github.com/openstack/swift/blob/e11a901d3cc1d433f595aad2b357b0656c2bbb6f/swift/common/middleware/versioned_writes.py#L46119:22
tdasilvaclayg: i'm the idiot who wrote the messy code :/19:25
claygtdasilva: working messy code is better than clean broken code?19:27
*** dkleute_ has joined #openstack-swift19:47
*** zhill has quit IRC19:49
*** dkleute has quit IRC19:50
*** marzif has quit IRC19:50
*** dkleute_ has quit IRC19:54
*** zhill has joined #openstack-swift19:59
*** hezhiqiang has joined #openstack-swift20:00
openstackgerritAlistair Coles proposed openstack/swift: Encrypter and Decrypter middleware - object body only  https://review.openstack.org/19419120:09
hrouacoles_, wow a ton of great refactoring, thanks !  Will review that now20:11
*** hezhiqiang has quit IRC20:17
*** subscope has quit IRC20:25
*** zhill has quit IRC20:34
*** lifeless has quit IRC20:34
*** siva_krishnan has quit IRC20:40
*** siva_krishnan has joined #openstack-swift20:41
*** zhill has joined #openstack-swift20:43
*** hezhiqiang has joined #openstack-swift20:47
*** petertr7 is now known as petertr7_away20:53
MooingLemurIs there anything obvious that would make it so a proxy server wouldn't honor an "Expect: 100-continue" request header?  I have a set of proxies that don't seem to return it, and another set that does... and the 2nd set is literally a cloned OS image taken from ones that don't!20:54
MooingLemurI don't think there's anything in the config that would differentiate them20:54
*** DericHorn-HP has joined #openstack-swift20:54
claygMooingLemur: that's nuts!20:54
claygMooingLemur: nothing in swift - is there anything else (load balancer, ssl termination) between the swift-proxy-server process and the client?20:55
MooingLemurpound is on both, the configs are identical.  I'm testing without SSL but still through pound20:55
claygMooingLemur: *maybe* some eventlet version thing, but it'd have to be *super* old?!  and if you cloned it i'm guessing it's all the same20:55
*** johanfi has joined #openstack-swift20:56
johanfiHi.. is it possible to have multiple swift backends?20:56
MooingLemurload balancer is LVS, uninvolved in L420:56
claygMooingLemur: any possibility your test is invalid?  how do you know one does send the expect and the other does not?  how do you know your client is telling the proxy to send it?20:56
claygjohanfi: like 3-replica, 2-replica and EC - you bet20:57
MooingLemurclayg: argh.. you're right.. my test is invalid..  the ones that work are not going through pound but rather localhost:8080!20:57
MooingLemurthanks20:57
johanfiI'm asking because I have a deployment using Mirantis 7.0, and the current state installs swift on the controller-nodes, but does not allow for standalone proxies or storage-nodes20:57
claygjohanfi: oh bummer :'(20:58
claygjohanfi: yeah pretty common to have proxies and storage nodes be different sku's20:58
MooingLemurPound is not translating the 100-continue stuff.20:58
johanfimeaning, if I manually add proxy-nodes, the will be overwritten if I add controllers in fuel20:58
claygjohanfi: some times even the object storage nodes and metadata tiers are different nodes - some times the metadata tier gets lumped on the proxies - everything depends - swift doesn't care ho you lay it out20:59
claygjohanfi: sorry I don't have any experience with fuel - but that sounds like a bummer20:59
johanfiso you can have two, totally seperate swift-installations with one openstack (ie keystone/horizon)?20:59
claygMooingLemur: is there a don't_be_so_fucking_stupid_pound option?21:00
claygjohanfi: that is also true - but orthagonal to what I was talking about - which was cluster topology w/i a single swift-installation21:00
notmynameMooingLemur: correct. Pound doesn't support 100-continue. (that's why I like HAProxy better)21:00
claygnotmyname: wow rly!?21:00
claygnotmyname: crazy... https://bitbucket.org/amplidata/pound/src/f2affa5f867a9718c9dc864e5f1c2e0eddd91b9c/debian/patches/expect_100_continued.patch?at=default&fileviewer=file-view-default21:02
MooingLemuryeah, I can't find patches for 2.721:02
notmynameya, IIRC. that's one of the reasons we dropped it in swiftstack for haproxy21:02
clayghaproxy!?21:03
claygstud21:03
notmynameyeah. that's what I meant ;-)21:04
*** hezhiqiang has quit IRC21:05
MooingLemurThanks :)21:05
claygMooingLemur: I *think* this might be canonical upstream now -> https://github.com/goochjj/pound  - maybe open an issue and try to ask about those debian patches for expect 100 continue support?21:08
johanficlayg: If I recall correctly, you was as at one of the Openstack Swift-sessions in Vancouver.. You talked about having a community "graphite"-environment as swift-operators, correct?21:10
clayguhhhh... not exactly21:10
clayga) that was cschwede21:10
johanfiah, my mistake21:11
claygb) we talked about adding someting baked into vagrant-swift-all-in-one (or something similar) so we could get some community collaborate based on a "standard" configuraiton that would make it easier to share/collaborate on ideas21:11
claygc) ... but we did dick all with that good idea21:12
clayg... well21:12
claygto be fair I'm not sure if cschwede worked on it or now - he's the german guy - i was in the room - I'm the fat obnoxious american21:12
johanfiAs I see it I have some limits on my current fuel-deployment. I can: a) deploy openstack with fuel as is, and severly limit the storage capacity for swift21:14
johanfib) write a plugin in fuel that does the install (probably the correct way)21:14
johanfic) install a standalone swift-cluster seperate from the fuel-deployment and integrate that in my environment21:15
openstackgerritMerged openstack/swift: Script for running unit, func and probe tests at once  https://review.openstack.org/23336821:21
*** mac_ified has joined #openstack-swift21:21
*** trifon has quit IRC21:22
*** lcurtis has quit IRC21:25
*** dustins has quit IRC21:26
mattoliverauMorning21:26
*** hezhiqiang has joined #openstack-swift21:34
*** lpabon has quit IRC21:35
*** zhill has quit IRC21:36
*** DericHorn-HP has quit IRC21:40
claygjohanfi: I can't speak to if a fuel plugin that knows how to swift good is a resonable or correct path21:41
claygjohanfi: but setting up swift stand alone and integrating with other infrastructure is totally a thing you could do21:42
*** aix has joined #openstack-swift21:42
*** zhill has joined #openstack-swift21:43
johanficlayg: thanks21:45
*** wbhuber has quit IRC21:45
johanficlayg: probably test a standalone setup first, see how it behaves in the lab21:45
johanfi*goint to21:46
*** wbhuber has joined #openstack-swift21:46
*** acoles_ is now known as acoles21:46
*** lpabon has joined #openstack-swift21:49
*** gyee has joined #openstack-swift21:49
*** wbhuber has quit IRC21:50
*** hezhiqiang has quit IRC21:53
claygjohanfi: BTW, what are you doing with swift?  just like vm images?21:54
*** lpabon has quit IRC21:58
*** acoles is now known as acoles_21:58
*** pchng64 is now known as pchng22:00
*** proteusguy__ has quit IRC22:13
*** setmason has quit IRC22:15
*** pdardeau has quit IRC22:16
*** jamielennox|away is now known as jamielennox22:16
*** pdardeau has joined #openstack-swift22:16
openstackgerritHamdi Roumani proposed openstack/swift: WIP: Symlink implementation. (work in progress)  https://review.openstack.org/23216222:17
*** hezhiqiang has joined #openstack-swift22:22
*** jlhinson has quit IRC22:26
*** proteusguy__ has joined #openstack-swift22:27
*** briancline has quit IRC22:39
*** zhill has quit IRC22:40
*** zhill has joined #openstack-swift22:41
*** setmason has joined #openstack-swift22:41
johanficlayg: images also, but mainly the goal is to use it for apps in an s3-fasion22:41
johanfiso the initial thought would be to have to have the two controllers from fuel serve as a image-store, and the second cluster for apps22:42
*** hezhiqiang has quit IRC22:42
claygjohanfi: that sounds awesome!22:43
johanficlayg: thanks.. btw, has anyone here used spark together with swift?22:44
*** siva_krishnan has quit IRC22:44
johanfiit seems like sort of the match made in heaven, using spark and swift together.. the only caveat I see is that you have the issue of data locality22:44
johanfiand the potential bottleneck of proxy-nodes22:45
*** nadeem_ has quit IRC22:45
claygjohanfi: not I, but it sure looks like someone's thought about it -> http://spark.apache.org/docs/latest/storage-openstack-swift.html22:46
claygI honestly thought of splunk when i read that the first time :\22:46
*** siva_krishnan has joined #openstack-swift22:46
claygjohanfi: I'm not sure, they mention the list_endpoints middleware - which is a good indication that they're at least thinking about it the right way22:47
*** setmason_ has joined #openstack-swift22:47
setmason_Wouldn’t that crush Swift?22:47
notmynamewhat would?22:48
johanfiI still haven't heard anyone who actually uses swift this way though..22:49
torgomaticuh, so the spark page there talks about auth, credentials, and such...22:49
torgomaticthen it says this "Although not mandatory, it is recommended to configure the proxy server of Swift with list_endpoints to have better data locality."22:49
torgomaticlolwut22:49
torgomaticif you're not bypassing the proxy and going straight to the storage nodes, list_endpoints buys you exactly nothing22:50
*** setmason has quit IRC22:50
*** setmason_ is now known as setmason22:50
*** siva_krishnan has quit IRC22:50
johanfibecause in my mind, if you place spark inside the network of swift, you would have to do keystone authorization on the storage-layer22:50
torgomaticthe swift proxy does authentication; the storage layer does not22:51
torgomaticif you get inside Swift's internal network, there's no authen or authz any more; just full access to absolutely everything22:52
johanfithat's how it works now, yes, but if you would use a 1-1 releationship between storage-nodes and spark-nodes, meaning one storage server and one compute server22:53
*** setmason has quit IRC22:53
torgomaticI mean, some folks went to the trouble of writing list_endpoints, so clearly it's not an issue if you've got a single-purpose cluster and you don't mind bypassing auth22:53
johanfibut does list-endpoint meaning bypass auth?22:53
torgomaticwell, list_endpoints lets you make a request that gives you back some URLs22:54
johanfifrom the storagenodes?22:54
torgomaticso you request something like /list_endpoints/AUTH_me/container/object, and it looks that up in the rings and tells you which storage nodes would have it if it existed22:54
torgomatic(also some headers to send along, yada yada yada)22:54
*** setmason has joined #openstack-swift22:54
torgomaticand then you go talk to the best storage node from that set and get the data22:55
torgomatic(where "best" probably means "closest network-wise")22:55
johanfiah.. so spark connects to the proxy, sends where it is, then how to connect it to the storage-ndoe22:55
johanfibut the storage-nodes and spark still have to be on the same cluster22:55
johanfi*network22:55
johanfifor that to make sense22:55
torgomaticI mean... maybe? The docs on that page are quite confusing to me; my impression is that spark talks to the proxy for everything, but they suggest enabling list_endpoints for something that doesn't make sense22:56
torgomaticnow, if spark *does* talk to the storage nodes directly, then that could work, though at that point I wonder why bother with the proxy at all22:56
*** Ailus has quit IRC22:56
*** km has joined #openstack-swift22:57
*** km is now known as Guest4368422:57
johanficlayg: in terms of adding another swift, how would that look like on the horizon side?23:01
johanficlayg: ie, how could users look at containers, create, enforce quotaes etc?23:01
claygaww man - i keep meaing to look at horizon - won't it just use the same catalog that any of the clients receive?  surely it supports - what does keystone call them?  "regions"23:03
clayg"service regions"23:03
claygidk23:03
mattoliveraujohanfi: doesn't that come down keystone intergration, ie the endpoints you configure23:03
mattoliverauI think horizon looks at the software catalog from keystone, tho I could be wrong23:04
johanfiand keystone supports multiple swift backends?23:04
*** Ailus has joined #openstack-swift23:04
*** Ailus has quit IRC23:05
claygtorgomatic: well - storage nodes can't have list_endpoints in the pipeline - so there's that I guess?23:05
claygjohanfi: yeah for sure23:05
johanfisilly question perhaps, but would having multiple swift be like having different regions?23:06
claygwell - heh23:06
claygyou can have two swift clusters and there's some terminology in keystone and the like to have mutiple services of the same type listed in the catalog - rax calls them regions (DFW, IAD, LON, etc)23:07
claygthen there's *swift* regions - which have to do with GEO clusters23:07
claygand those two things are not related23:07
johanfiwas thinking of this one "keystone endpoint-create --region RegionOne"23:08
claygyeah that's the ticket!23:08
*** david-lyle has quit IRC23:09
*** david-lyle has joined #openstack-swift23:09
johanfiok, so install a standalone swift-cluster, point it towards the correct keystone. then add the vip/LB ip of the proxy-nodes and then I'm good?23:10
johanfiadd it in keystone that is23:10
*** khoah64 has quit IRC23:10
claygit'll be great!23:11
johanfishould be fun! :)23:11
claygparty time!23:11
*** delattec has quit IRC23:12
*** cdelatte has quit IRC23:12
*** MooingLemur has quit IRC23:12
*** hezhiqiang has joined #openstack-swift23:12
johanficlayg: thanks for all the help.. it was really useful.23:12
*** MooingLemur has joined #openstack-swift23:13
*** chlong has quit IRC23:16
*** wshao has joined #openstack-swift23:17
*** johanfi has quit IRC23:19
*** setmason has quit IRC23:20
*** darrenc is now known as darrenc_afk23:20
*** darrenc has joined #openstack-swift23:28
*** darrenc_afk has quit IRC23:30
*** kei_yama has joined #openstack-swift23:30
*** minwoob_ has joined #openstack-swift23:31
*** scotticus has quit IRC23:32
*** hezhiqiang has quit IRC23:33
*** minwoob has quit IRC23:34
*** hogepodge has quit IRC23:34
*** jbonjean has quit IRC23:34
*** baffle has quit IRC23:34
openstackgerritOpenStack Proposal Bot proposed openstack/swift: Updated from global requirements  https://review.openstack.org/8873623:34
*** clayg has quit IRC23:34
*** bapalm has quit IRC23:34
*** mtreinish has quit IRC23:34
*** clayg has joined #openstack-swift23:36
*** ChanServ sets mode: +v clayg23:36
*** scott_ has joined #openstack-swift23:39
*** mtreinish has joined #openstack-swift23:39
*** bapalm has joined #openstack-swift23:39
*** scott_ is now known as Guest4012023:39
*** blmartin has joined #openstack-swift23:42
blmartinGood evening23:42
*** baffle has joined #openstack-swift23:43
*** pdardeau has quit IRC23:54
*** blmartin has quit IRC23:54
*** remix_tj has quit IRC23:54
*** ajiang has quit IRC23:54
*** baffle has quit IRC23:54
*** wshao has quit IRC23:54
*** aix has quit IRC23:54
*** pchng has quit IRC23:54
*** Guest62185 has quit IRC23:54
*** dmsimard has quit IRC23:54
*** BAKfr has quit IRC23:54
*** gmmaha has quit IRC23:54
*** darrenc has quit IRC23:54
*** jodah has quit IRC23:54
*** chrisnelson has quit IRC23:54
*** CaioBren_ has quit IRC23:54
*** portante has quit IRC23:54
*** fbo has quit IRC23:54
*** serverascode has quit IRC23:54
*** philipw has quit IRC23:54
*** jamielennox has quit IRC23:54
*** peluse has quit IRC23:54
*** mitz_ has quit IRC23:54
*** ctennis has quit IRC23:54
*** cppforlife_ has quit IRC23:54
*** zigo has quit IRC23:54
*** tonyb has quit IRC23:54
*** sayali has quit IRC23:54
*** zacksh has quit IRC23:54
*** clyps_ has quit IRC23:54
*** jroll has quit IRC23:54
*** mattoliverau has quit IRC23:54
*** sudorandom has quit IRC23:54
*** CrackerJackMack has quit IRC23:54
*** donagh has quit IRC23:54
*** ChanServ has quit IRC23:54
*** proteusguy__ has quit IRC23:54
*** gyee has quit IRC23:54
*** mac_ified has quit IRC23:54
*** marcusvrn_ has quit IRC23:54
*** briancurtin has quit IRC23:54
*** StevenK has quit IRC23:54
*** chmouel has quit IRC23:54
*** amandap has quit IRC23:54
*** diogogmt has quit IRC23:54
*** dosaboy has quit IRC23:54
*** nottrobin has quit IRC23:54
*** balajir has quit IRC23:54
*** shakamunyi has quit IRC23:54
*** ahale has quit IRC23:54
*** hrou has quit IRC23:54
*** ujjain has quit IRC23:54
*** zackmdavis has quit IRC23:54
*** timburke has quit IRC23:54
*** bapalm has quit IRC23:54
*** zhill has quit IRC23:54
*** petertr7_away has quit IRC23:54
*** jmccarthy has quit IRC23:54
*** j_king has quit IRC23:54
*** dabukalam has quit IRC23:54
*** timur has quit IRC23:54
*** mandarine has quit IRC23:55
*** tristanC has quit IRC23:55
*** csmart has quit IRC23:55
*** anteaya has quit IRC23:55
*** minwoob_ has quit IRC23:55
*** MooingLemur has quit IRC23:55
*** david-lyle has quit IRC23:55
*** wasmum- has quit IRC23:55
*** guitarzan has quit IRC23:55
*** acorwin_ has quit IRC23:55
*** then3rd has quit IRC23:55
*** zhiyan has quit IRC23:55
*** acoles_ has quit IRC23:55
*** swat30 has quit IRC23:55
*** logan2 has quit IRC23:55
*** morgan has quit IRC23:55
*** dfg_ has quit IRC23:55
*** tdasilva has quit IRC23:55
*** swifterdarrell has quit IRC23:55
*** kota_ has quit IRC23:55
*** klrmn has quit IRC23:55
*** thurloat_isgone has quit IRC23:55
*** Guest40120 has quit IRC23:55
*** kei_yama has quit IRC23:55
*** alejandrito has quit IRC23:55
*** openstackgerrit has quit IRC23:55
*** vinsh has quit IRC23:55
*** treyd_ has quit IRC23:55
*** okdas has quit IRC23:55
*** amit213 has quit IRC23:55
*** blair has quit IRC23:55
*** hugespoon has quit IRC23:55
*** dhellmann has quit IRC23:55
*** saltsa has quit IRC23:55
*** aerwin3_ has quit IRC23:55
*** jeblair has quit IRC23:55
*** EmilienM has quit IRC23:55
*** notmyname has quit IRC23:55
*** mtreinish has quit IRC23:55
*** daemontool_ has quit IRC23:55
*** onovy has quit IRC23:55
*** HenryG has quit IRC23:55
*** elmo has quit IRC23:55
*** charz has quit IRC23:55
*** hurricanerix has quit IRC23:55
*** cebruns has quit IRC23:55
*** bwall has quit IRC23:55
*** tamizh_geek has quit IRC23:55
*** mlanner has quit IRC23:55
*** patchbot has quit IRC23:55
*** mfalatic has quit IRC23:55
*** zul has quit IRC23:55
*** torgomatic has quit IRC23:55
*** hugokuo has quit IRC23:55
*** sileht has quit IRC23:55
*** mplex has quit IRC23:55
*** wolsen has quit IRC23:55
*** kragniz has quit IRC23:55
*** sc68cal has quit IRC23:55
*** _fortis has quit IRC23:55
*** ondergetekende has quit IRC23:55
*** glange has quit IRC23:55
*** eikke has quit IRC23:55
*** eariasn has quit IRC23:55
*** yuan has quit IRC23:55
*** bhakta has quit IRC23:55
*** alpha_ori has quit IRC23:55
*** tanee has quit IRC23:55
*** anderstj has quit IRC23:55
*** goodygum has quit IRC23:55
*** clayg has quit IRC23:55
*** Guest43684 has quit IRC23:55
*** klrmn1 has quit IRC23:55
*** pberis has quit IRC23:55
*** pgbridge has quit IRC23:55
*** ndk has quit IRC23:55
*** ekarlso has quit IRC23:55
*** Anticimex has quit IRC23:55
*** bobby2 has quit IRC23:55
*** wer has quit IRC23:55
*** ejat has quit IRC23:55
*** raginbajin has quit IRC23:55
*** cschwede has quit IRC23:55
*** early has quit IRC23:55
*** joearnold has quit IRC23:55
*** earlephilhower has quit IRC23:55
*** janonymous__ has quit IRC23:55
*** janonymous has quit IRC23:55

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