Wednesday, 2015-09-30

*** openstackgerrit has quit IRC00:01
*** openstackgerrit has joined #openstack-swift00:01
*** openstackstatus has quit IRC00:02
*** openstack has joined #openstack-swift00:05
*** devlaps has quit IRC00:09
*** CR7 has joined #openstack-swift00:18
*** minwoob has quit IRC00:19
*** zhill has quit IRC00:29
*** m_kazuhiro has joined #openstack-swift00:29
*** chsc has quit IRC00:33
*** dustins has quit IRC00:36
*** zhill has joined #openstack-swift00:39
*** kota_ has joined #openstack-swift00:39
*** ChanServ sets mode: +v kota_00:39
kota_good morning00:40
timburkegood morning kota_!00:42
*** lyrrad has quit IRC00:43
*** zhill has quit IRC00:43
kota_Hi, Tim :)00:45
*** CR7 has quit IRC00:47
notmynameI just saw a picture of a tshirt proof. looks good :-)00:50
kota_notmyname: sounds good00:50
kota_notmyname: and thanks for fixing my misspelling for s/2xx/1xx/00:51
kota_ah, s/1xx/2xx/00:55
*** duobei has joined #openstack-swift00:57
*** zhill has joined #openstack-swift01:00
*** kota_ has quit IRC01:00
*** duobei has quit IRC01:02
*** NM has quit IRC01:09
*** jkugel has joined #openstack-swift01:20
jrichlinotmyname: cool!  Is it ok if we see it too :-)01:22
*** morgan is now known as notreallymorgan01:30
*** notreallymorgan is now known as morgan01:30
mattoliveraunotmyname: confirmed bug and patch fix on patch 227133 shorcut to results: http://paste.openstack.org/show/474772/01:31
patchbotmattoliverau: https://review.openstack.org/#/c/227133/ - Fix copy requests to service accounts in Keystone01:31
*** cdelatte has quit IRC01:31
mattoliverauusing keystone I mean ^^01:31
*** bill_az has quit IRC01:33
*** Kennnan2 is now known as Kennan01:37
openstackgerritKota Tsuyuzaki proposed openstack/swift: WIP: PUT/GET path for global ec cluster  https://review.openstack.org/21916501:47
*** m_kazuhiro has quit IRC02:03
*** zhill has quit IRC02:12
*** m_kazuhiro has joined #openstack-swift02:12
*** haomaiwang has joined #openstack-swift02:15
*** jkugel has quit IRC02:16
*** kota_ has joined #openstack-swift02:19
*** ChanServ sets mode: +v kota_02:19
*** haomaiwang has quit IRC02:22
*** haomaiwa_ has joined #openstack-swift02:22
*** CaioBrentano has joined #openstack-swift02:26
*** CaioBrentano has quit IRC02:30
*** haomaiwa_ has quit IRC02:41
*** haomaiwang has joined #openstack-swift02:42
*** sanchitmalhotra has joined #openstack-swift02:45
*** gyee has quit IRC02:52
*** haomaiwang has quit IRC03:01
*** haomaiwang has joined #openstack-swift03:01
notmynamemattoliverau: thanks for putting that in gerrit. and for working on it :-)03:11
*** jkugel has joined #openstack-swift03:16
*** chsc has joined #openstack-swift03:37
*** jrichli has quit IRC03:39
*** sanchitmalhotra1 has joined #openstack-swift03:41
*** sanchitmalhotra has quit IRC03:43
*** trifon has joined #openstack-swift03:44
*** klrmn1 has quit IRC03:49
*** trifon has quit IRC03:49
*** links has joined #openstack-swift03:53
*** haomaiwang has quit IRC04:01
*** haomaiwang has joined #openstack-swift04:01
*** SkyRocknRoll has joined #openstack-swift04:02
*** Zyric2 has quit IRC04:04
*** kota_ has quit IRC04:11
*** chsc has quit IRC04:31
*** links has quit IRC04:36
*** sanchitmalhotra has joined #openstack-swift04:41
*** sanchitmalhotra1 has quit IRC04:43
*** mahatic has joined #openstack-swift04:56
*** haomaiwang has quit IRC05:01
*** haomaiwang has joined #openstack-swift05:01
*** SkyRocknRoll has quit IRC05:02
*** links has joined #openstack-swift05:12
*** mahatic has quit IRC05:15
*** mac_ified has quit IRC05:20
*** tellesnobrega is now known as tellesnobrega_af05:22
*** vinsh has quit IRC05:26
*** SkyRocknRoll has joined #openstack-swift05:27
*** mahatic has joined #openstack-swift05:28
*** silor has joined #openstack-swift05:32
*** mahatic_ has joined #openstack-swift05:34
*** mahatic has quit IRC05:34
*** janonymous has quit IRC05:37
*** hrou has quit IRC05:40
*** silor1 has joined #openstack-swift05:41
*** silor has quit IRC05:43
*** silor1 is now known as silor05:43
*** ppai has joined #openstack-swift05:50
*** sanchitmalhotra1 has joined #openstack-swift05:56
*** sanchitmalhotra has quit IRC05:58
*** haomaiwang has quit IRC06:01
*** haomaiwang has joined #openstack-swift06:01
*** vinsh has joined #openstack-swift06:08
*** sanchitmalhotra has joined #openstack-swift06:15
*** sileht has quit IRC06:17
*** sanchitmalhotra1 has quit IRC06:18
*** sileht has joined #openstack-swift06:28
*** haomaiwang has quit IRC07:01
*** haomaiwang has joined #openstack-swift07:01
*** m_kazuhiro has quit IRC07:09
*** rledisez has joined #openstack-swift07:14
*** eranrom has joined #openstack-swift07:15
*** ppai has quit IRC07:17
*** Kennan has quit IRC07:23
*** ppai has joined #openstack-swift07:24
*** Kennan has joined #openstack-swift07:25
*** geaaru has joined #openstack-swift07:32
*** acoles_ is now known as acoles07:33
*** eranrom has quit IRC07:43
acolesminwoob: yes i came up with a workaround. thanks for looking into it.07:50
acolescschwede: shall i update patch 227133 to address mine and mattoliverau comments? looks like mattoliverau is happy that it fixes the bug.07:52
patchbotacoles: https://review.openstack.org/#/c/227133/ - Fix copy requests to service accounts in Keystone07:52
*** ccavanna__ has joined #openstack-swift07:52
*** ccavanna_ has quit IRC07:55
*** geaaru has quit IRC07:58
*** haomaiwang has quit IRC08:00
*** haomaiwang has joined #openstack-swift08:01
*** jistr has joined #openstack-swift08:02
*** geaaru has joined #openstack-swift08:11
*** breitz has quit IRC08:16
*** breitz has joined #openstack-swift08:17
*** m_kazuhiro has joined #openstack-swift08:27
*** kei_yama has quit IRC08:35
*** kota_ has joined #openstack-swift08:36
*** ChanServ sets mode: +v kota_08:36
*** km has quit IRC08:37
*** sanchitmalhotra1 has joined #openstack-swift08:43
*** sanchitmalhotra has quit IRC08:46
openstackgerritAlistair Coles proposed openstack/swift: Update EC overview doc for PUT path  https://review.openstack.org/22930008:48
acolespeluse: wbhuber: ^^ an easy review08:49
acolesnotmyname: ^^ not critical but worth having for liberty08:51
*** jordanP has joined #openstack-swift08:51
*** ppai has quit IRC08:57
*** haomaiwang has quit IRC09:01
*** haomaiwang has joined #openstack-swift09:01
*** aix has joined #openstack-swift09:06
*** ppai has joined #openstack-swift09:11
mahatic_acoles: hi, just a question on a card you added to trello: https://trello.com/c/aSNZ1M3G/54-guess-content-type-when-missing-from-put-request09:24
mahatic_acoles: Did you something failing because of this? I ask coz, I have added the content-type guessing part to encrypter and see no change. Did this mean to solve/address something?09:25
mahatic_did you mean*09:26
*** sanchitmalhotra has joined #openstack-swift09:37
*** sanchitmalhotra1 has quit IRC09:40
*** marzif has joined #openstack-swift09:51
*** silor has quit IRC09:52
*** ppai has quit IRC10:00
*** haomaiwang has quit IRC10:01
*** haomaiwang has joined #openstack-swift10:01
*** marzif has quit IRC10:03
*** silor has joined #openstack-swift10:05
*** ppai has joined #openstack-swift10:13
*** haomaiwang has quit IRC10:15
*** haomaiwang has joined #openstack-swift10:16
*** haomaiwang has quit IRC10:21
*** sanchitmalhotra1 has joined #openstack-swift10:35
*** sanchitmalhotra has quit IRC10:37
*** mahatic_ has quit IRC10:41
acolesmahatic: sorry been on a call, just replied on the trello card.10:46
*** mahatic has joined #openstack-swift10:56
openstackgerritAlistair Coles proposed openstack/swift: Update EC overview doc for GET path  https://review.openstack.org/22935310:58
*** jordanP has quit IRC10:59
*** pberis has joined #openstack-swift11:05
mahaticacoles: thanks. what you said sounds good11:08
*** pberis has quit IRC11:09
*** BAKfr has quit IRC11:18
*** BAKfr has joined #openstack-swift11:21
*** ccavanna__ has quit IRC11:21
*** itlinux has joined #openstack-swift11:25
*** itlinux has quit IRC11:26
*** itlinux has joined #openstack-swift11:26
*** SkyRocknRoll_ has joined #openstack-swift11:27
*** cdelatte has joined #openstack-swift11:39
*** sanchitmalhotra has joined #openstack-swift11:51
*** sanchitmalhotra1 has quit IRC11:53
*** jkugel has quit IRC11:57
*** m_kazuhiro has quit IRC12:04
*** ppai has quit IRC12:06
*** MVenesio has joined #openstack-swift12:10
*** tellesnobrega_af is now known as tellesnobrega12:12
*** ppai has joined #openstack-swift12:20
*** aix has quit IRC12:22
*** mahatic has quit IRC12:29
*** mahatic has joined #openstack-swift12:30
*** marcusvrn_ has joined #openstack-swift12:30
*** gustavo_ has joined #openstack-swift12:31
*** Kennan has quit IRC12:38
*** Kennan has joined #openstack-swift12:39
*** juzuluag has joined #openstack-swift12:42
*** marzif has joined #openstack-swift12:43
*** haomaiwang has joined #openstack-swift12:43
*** ppai has quit IRC12:46
*** peterlisak has joined #openstack-swift12:47
*** marekkaleta has joined #openstack-swift12:47
*** marekkaleta_ has joined #openstack-swift12:47
*** vinsh_ has joined #openstack-swift12:48
*** CaioBrentano has joined #openstack-swift12:50
*** vinsh has quit IRC12:52
*** aix has joined #openstack-swift12:55
*** kota_ has quit IRC12:57
*** tellesnobrega is now known as tellesnobrega_af12:59
*** ppai has joined #openstack-swift12:59
*** haomaiwang has quit IRC13:01
*** haomaiwang has joined #openstack-swift13:01
*** haomaiwang has quit IRC13:02
*** haomaiwang has joined #openstack-swift13:02
*** jkugel has joined #openstack-swift13:07
*** hrou has joined #openstack-swift13:12
openstackgerritPete Zaitcev proposed openstack/swift: Enable object server to return non-durable data  https://review.openstack.org/21527613:15
*** remix_tj has quit IRC13:16
*** dustins has joined #openstack-swift13:18
*** remix_tj has joined #openstack-swift13:19
*** vinsh has joined #openstack-swift13:21
*** vinsh_ has quit IRC13:21
*** SkyRocknRoll has quit IRC13:29
*** wbhuber has joined #openstack-swift13:29
*** SkyRocknRoll_ has quit IRC13:29
*** alejandrito has joined #openstack-swift13:30
*** juzuluag has quit IRC13:31
*** mahatic has quit IRC13:35
*** ppai has quit IRC13:36
*** tellesnobrega_af is now known as tellesnobrega13:37
*** marzif has quit IRC13:43
*** william has joined #openstack-swift13:44
*** william is now known as Guest4963913:45
*** BAKfr has quit IRC13:45
*** mahatic has joined #openstack-swift13:47
*** ccavanna has joined #openstack-swift13:50
*** haomaiwang has quit IRC14:01
*** sanchitmalhotra has quit IRC14:01
*** haomaiwang has joined #openstack-swift14:01
*** BAKfr has joined #openstack-swift14:05
*** itlinux has quit IRC14:08
*** jrichli has joined #openstack-swift14:10
*** j_king has joined #openstack-swift14:12
*** chsc has joined #openstack-swift14:12
*** haomaiwang has quit IRC14:14
*** haomaiwa_ has joined #openstack-swift14:15
*** marekkaleta_ has quit IRC14:17
*** marekkaleta has quit IRC14:17
*** j_king has left #openstack-swift14:18
*** lcurtis has joined #openstack-swift14:22
*** jlhinson has joined #openstack-swift14:26
*** pberis has joined #openstack-swift14:30
tdasilvaacoles: I just put a +A on the PUT doc, looked pretty straight forward14:30
acolestdasilva: thanks!14:30
tdasilvaacoles: are you planning to squash the GET doc?14:30
*** SkyRocknRoll has joined #openstack-swift14:33
*** SkyRocknRoll has quit IRC14:33
*** SkyRocknRoll has joined #openstack-swift14:33
*** mahatic has quit IRC14:35
acolestdasilva: yes most likely, when i have other changes for patch 215276. i just did its as a dependent patch to avoid yet another version of 215276 but get it up on gerrit.14:41
patchbotacoles: https://review.openstack.org/#/c/215276/ - Enable object server to return non-durable data14:41
*** haomaiwa_ has quit IRC14:42
tdasilvaacoles: going to take a look at 215276 now14:42
*** haomaiwang has joined #openstack-swift14:43
acolestdasilva: thanks and good luck! (its big, but lots of comments)14:43
tdasilvaacoles: even the commit message is big :)14:43
*** mahatic has joined #openstack-swift14:44
acolestdasilva: yeah i know :/ actually some of the commit msg content went into the doc patch14:45
acolestdasilva: the commit msg is a spec in disguise :P14:45
*** itlinux has joined #openstack-swift14:46
*** ppai has joined #openstack-swift14:47
*** ppai has quit IRC14:52
tdasilvaacoles: hehe..it's a very good description. I'm just a bit iffy on the "X-Backend-Fragment-Preferences", i'm hoping looking at the code will help understand better14:53
acolestdasilva: ah, in the code its still X-Backend-Frag-Prefs - thats a TODO14:54
acolestdasilva: I can't decide if I prefer brief or precise header names! And I assumed someone would have a strong opinion so I am waiting for that :)14:54
*** trifon has joined #openstack-swift14:54
tdasilvaacoles: I was a bit confused because it says "proxy's current preference with respect to those fragments that it would have object servers  return", but then the next sentence says: "list of fragment indexes that the proxy does NOT require"14:55
acolestdasilva: oic. yes, its a negative preference on frag indexes, but positive wrt to timestamp order.14:56
*** haomaiwang has quit IRC15:01
*** haomaiwang has joined #openstack-swift15:01
acolestdasilva: so i think clayg suggested X-Backend-If-Frag-Not-In or similar - let me know your preference.15:02
acolesugh! "preference" pun not intended15:02
tdasilvalol15:02
*** janonymous has joined #openstack-swift15:04
*** jistr has quit IRC15:07
*** haomaiwang has quit IRC15:10
*** haomaiwa_ has joined #openstack-swift15:10
*** chsc has quit IRC15:19
*** haomaiwa_ has quit IRC15:19
*** haomaiwang has joined #openstack-swift15:20
*** links has quit IRC15:21
*** itlinux has quit IRC15:23
*** minwoob has joined #openstack-swift15:24
*** chsc has joined #openstack-swift15:24
*** zhill has joined #openstack-swift15:25
*** klrmn1 has joined #openstack-swift15:27
*** BAKfr has quit IRC15:36
*** haomaiwang has quit IRC15:37
*** haomaiwang has joined #openstack-swift15:38
openstackgerritJanie Richling proposed openstack/swift: Encrypter and Decrypter middleware - object body only  https://review.openstack.org/19419115:40
*** BAKfr has joined #openstack-swift15:41
*** marzif has joined #openstack-swift15:41
*** lyrrad has joined #openstack-swift15:42
notmynamegood morning15:50
*** haomaiwang has quit IRC15:53
*** BAKfr has quit IRC15:53
janonymousnotmyname: o/15:53
*** haomaiwang has joined #openstack-swift15:53
*** dosaboy_ is now known as dosaboy15:54
*** BAKfr has joined #openstack-swift15:55
*** aix has quit IRC15:57
*** haomaiwang has quit IRC16:01
*** peterlisak has quit IRC16:01
*** haomaiwa_ has joined #openstack-swift16:01
acolesnotmyname: morning16:03
notmynamewhere are we with patch 227133 ?16:04
patchbotnotmyname: https://review.openstack.org/#/c/227133/ - Fix copy requests to service accounts in Keystone16:04
*** marzif has quit IRC16:04
*** haomaiwa_ has quit IRC16:04
notmynamecschwede: will you be doing another patch set? I'd love to see it land this week16:05
acolesnotmyname: i'm just working on that unless cschwede is too ^^16:06
notmynameacoles: great, thanks16:06
*** haomaiwang has joined #openstack-swift16:07
*** DjamOlsky has joined #openstack-swift16:09
*** cazino has left #openstack-swift16:10
*** haomaiwang has quit IRC16:11
*** haomaiwa_ has joined #openstack-swift16:11
DjamOlskyIs there anyone who implemented ELK to parse swift logs?16:13
DjamOlskyreally hard (or maybe unlucky!) to find any filter/patterns for swift online16:13
*** breitz has quit IRC16:13
hugespoonDjamOlsky: http://docs.openstack.org/developer/swift/logs.html probably would be helpful to you, also check out #logstash16:14
*** breitz has joined #openstack-swift16:14
DjamOlskyhugespoon: thanks. Will have a look :)16:15
*** rledisez has quit IRC16:15
*** haomaiwa_ has quit IRC16:18
*** haomaiwang has joined #openstack-swift16:19
*** klrmn1 has quit IRC16:19
*** haomaiwang has quit IRC16:20
*** haomaiwa_ has joined #openstack-swift16:20
*** gyee has joined #openstack-swift16:22
*** chsc has quit IRC16:31
*** haomaiwa_ has quit IRC16:33
*** garthb has joined #openstack-swift16:36
openstackgerritAlistair Coles proposed openstack/swift: Fix copy requests to service accounts in Keystone  https://review.openstack.org/22713316:39
acolesnotmyname: ^^ hopefully cschwede is happy with that16:40
*** devlaps has joined #openstack-swift16:50
notmynamethanks. I'm going to spend some more time with it today16:50
openstackgerritSamuel Merritt proposed openstack/swift: Improve SLO PUT error checking  https://review.openstack.org/22809916:59
*** jrichli has quit IRC17:03
*** Guest49639 has quit IRC17:04
*** william has joined #openstack-swift17:04
*** william is now known as Guest1560817:04
*** earlephilhower has joined #openstack-swift17:08
*** mahatic has quit IRC17:10
openstackgerritOndrej Novy proposed openstack/swift: swift-init: SIGKILL daemon after kill_wait period.  https://review.openstack.org/22953217:11
onovynotmyname, hi, do you have time for small review please? https://review.openstack.org/#/c/210736/3 thanks17:14
*** chsc has joined #openstack-swift17:15
*** klrmn1 has joined #openstack-swift17:17
*** aix has joined #openstack-swift17:19
onovyacoles, or you? :)17:19
*** trifon has quit IRC17:29
openstackgerritMerged openstack/swift: Update EC overview doc for PUT path  https://review.openstack.org/22930017:34
openstackgerritMerged openstack/swift: Update EC overview doc for PUT path  https://review.openstack.org/22930017:34
*** shakamunyi has quit IRC17:35
*** trifon has joined #openstack-swift17:36
*** garthb has quit IRC17:46
openstackgerritAlistair Coles proposed openstack/swift: Add POST capability to ssync for .meta files  https://review.openstack.org/13849817:50
*** jrichli has joined #openstack-swift17:57
*** shakamunyi has joined #openstack-swift18:01
*** eranrom has joined #openstack-swift18:05
*** dmsimard is now known as dmsimard|afk18:11
*** openstackgerrit has quit IRC18:16
*** openstackgerrit has joined #openstack-swift18:16
*** garthb has joined #openstack-swift18:22
*** Guest15608 has quit IRC18:24
openstackgerritOndřej Nový proposed openstack/swift-bench: Try to detect HTTP proxy and warn about it.  https://review.openstack.org/21073618:26
onovytamizh_geek, thanks18:27
tamizh_geekonovy: You are welcome :)18:28
*** aix has quit IRC18:30
*** geaaru has quit IRC18:40
*** SkyRocknRoll has quit IRC18:41
*** Guest15608 has joined #openstack-swift18:42
*** Guest15608 has quit IRC18:43
*** william has joined #openstack-swift18:44
*** william is now known as Guest8237018:45
*** openstackgerrit has quit IRC18:46
*** openstackgerrit has joined #openstack-swift18:46
*** aix has joined #openstack-swift18:46
*** gmmaha has joined #openstack-swift18:49
*** [1]eranrom has joined #openstack-swift18:53
*** eranrom has quit IRC18:53
*** [1]eranrom is now known as eranrom18:53
claygok, i give up - i'm not going to fix part-replica placement in time to review by friday18:54
*** aix has quit IRC18:55
*** david-lyle has quit IRC18:55
*** david-lyle has joined #openstack-swift18:55
claygI'm going to break off back to validate-rings and reduce the critical error to a warning - do what I can to put a tourniquet around it - then review stuff other stuff that can acctually land18:56
claygonce we have an RC1 - i'll go back to my placement planning branch and try to fix-rings-for-real18:56
notmynameclayg: thanks for working on it18:59
*** dmsimard|afk is now known as dmsimard19:03
*** janonymous has quit IRC19:07
*** david-lyle has quit IRC19:22
*** david-lyle has joined #openstack-swift19:22
*** silor has quit IRC19:30
*** ccavanna has quit IRC19:30
openstackgerritOndřej Nový proposed openstack/swift: swift-init: SIGKILL daemon after kill_wait period.  https://review.openstack.org/22953219:35
*** david-lyle has quit IRC19:35
*** marzif has joined #openstack-swift19:40
*** david-lyle has joined #openstack-swift19:52
*** eranrom has quit IRC19:52
*** eranrom has joined #openstack-swift19:53
*** esker has joined #openstack-swift19:53
openstackgerritMinwoo Bae proposed openstack/swift: EC: Handoff node to push existing fragment to the correct location.  https://review.openstack.org/19684819:58
*** david-ly_ has joined #openstack-swift20:07
*** david-lyle has quit IRC20:07
*** david-ly_ is now known as david-lyle20:08
*** gyee has quit IRC20:16
acolesclayg: you there?20:24
claygacoles: sure20:25
*** thumpba has joined #openstack-swift20:25
acolesclayg: sorry, vbox networking died on me.20:32
claygnp20:32
acolesclayg: i got derailed on patch 215276 today :/20:32
patchbotacoles: https://review.openstack.org/#/c/215276/ - Enable object server to return non-durable data20:32
claygacoles: np, I'm going to start looking at it after the meeting20:33
acolesclayg: my response buckets are keyed by timestamp, so all frags with same timestamp go into same bucket. only with ssync and fast-post, we have frags for same object at different timestamps on different nodes after reconstruction :( so i can't GET the object.20:34
claygacoles: sigh20:36
claygacoles: you shouldn't have done that json thing20:36
acolesclayg: exactly20:36
claygacoles: you should have just used datatimestamp#frag_index20:36
acolesclayg: thats the least of my worries20:36
*** esker has quit IRC20:37
claygacoles: you called it coupling with ondisk format - but it's more like coupling with the default implemenation - which includes the fact that datafiles have timestamps independent of the "objects" timestamp20:37
*** garthb has quit IRC20:38
clayg... maybe I'm trivializing the problem, I'm sort of in a bad mood.20:38
claygbut either you have enough frags of a particular timestamp to rebuild the data from that timestamp or you don't - so i'm not quire sure I follow exactly the issue with the .meta files/timestamps unless the object server is just lying about what (timestamp, frag_index)'s it has20:39
*** blmartin has joined #openstack-swift20:40
claygX-Backed-Timestamp-Frag: <ts>#<frag_index>, X-Backend-Alt-Timestamp-Frag: <ts>#<frag_index>, <ts>#<frag_index>, X-Backend-Not-Timestamp-Frag: <ts>#<frag_index>, <ts>#<frag_index>, ... <= SEEMS LIKE IT WOULD WORK20:41
acolesclayg: that equivalent to what i have, no?20:43
acolesthe problem is what timestamp is  X-Backed-Timestamp-Frag?20:43
claygacoles: the timestamp of the frag20:44
acoles.meta file or .data file timestamp?20:44
claygthe .meta is not relevant in any fashion20:44
acolesI agree, but ssync makes it relevant when it takes meta timstamp and applies it to the data file it PUTs :/20:45
clayggd20:45
acolesso t2.meta + t1.data on one node is t2.data on another20:45
claygisn't there alreayd another patch that fixes ssync to POST .meta's?!20:45
acolesyeah!20:45
claygAHPU[OSIDHJOF[IHAEHI80[]WHI80]G32HNP20:45
claygG320:45
acolesque?20:45
acolespatch 13849820:46
patchbotacoles: https://review.openstack.org/#/c/138498/ - Add POST capability to ssync for .meta files20:46
openstackgerritMinwoo Bae proposed openstack/swift: Optimization of the reconstructor for handling of displaced fragments  https://review.openstack.org/19684820:46
claygwell, we should probably merge that20:46
claygacoles: it seems if t2.meta + t1.data on one node was correctly replicated as t2.meta + t1.data on another - you could expose the frag's timestamp correctly20:47
acolesclayg: that would imho be the right answer.  my only other "solution" is to go back to etag keys for buckets and give up trying to match alt-frags to buckets20:47
claygacoles: great let's do it!20:47
*** eranrom has quit IRC20:47
acolesclayg: yes and i just rebased 215276 on 138498 nd the failing probe test passed20:48
clayggreat push it - we'll merge everything - it's end of the release - no one will argue20:48
acoleslol20:48
acolesthats what scares me20:48
*** MVenesio has quit IRC20:48
claygit'll be fine - you're super smart20:49
claygand it's *just* a release *candidate* - there's always hope the final release is less worse20:49
acolesclayg: well if that were true i'd have run the probe test with fast-post switched on, like, a few days back at least20:49
acolesthank goodness i did it today!20:50
acolesclayg: ok i have to clean up the rebase merge resolutions before i can push two sensible patches.20:51
*** DjamOlsky has quit IRC20:51
acolesclayg: thanks.20:51
*** kota_ has joined #openstack-swift20:53
*** ChanServ sets mode: +v kota_20:53
*** hrou has quit IRC20:53
kota_good morning20:54
notmynamehello kota_20:54
acolesclayg: so (at risk of fanning the flames) I actually went with json in the end not because of any implementation coupling concern, but just cos it was less lines to write and test. I have no strong feeling about it otherwise.20:54
acoleshey kota_ you need an irc bouncer - i wanted to thank you for your review while you were sleeping :)20:54
claygacoles: ok, i'm surprised it was *less* code - stuff like keys that want to be floats seem like they would be annoying - but w/e I'm sure I'll see why it's better when I start looking at the change (or at least agree it's not important)20:55
kota_acoles: :) I'm going to scrole back20:55
acoleskota_: you may want to read the scrollback re review 215276 ^^20:55
notmynamehttp://wiki.znc.in/ZNC is a good bouncer, IMO20:56
acoleskota_: thanks for your review! :P20:56
openstackgerritClay Gerrard proposed openstack/swift: Validate against duplicate device part replica assignment  https://review.openstack.org/22279920:56
kota_acoles: you're welcome :920:57
kota_:)20:57
kota_that's awesome patch, imo but need a lot of time to understand new behavior :\20:58
notmynamemeeting time in #openstack-meeting20:59
mattoliveraumorning20:59
kota_mattoliverau: morning20:59
*** thumpba has quit IRC21:01
*** dustins has quit IRC21:01
*** david-lyle has quit IRC21:03
*** thumpba has joined #openstack-swift21:03
*** david-ly_ has joined #openstack-swift21:03
*** Guest82370 has quit IRC21:04
*** david-ly_ is now known as david-lyle21:05
*** thumpba has quit IRC21:08
*** thumpba has joined #openstack-swift21:09
*** Fin1te has quit IRC21:09
*** blmartin_ has joined #openstack-swift21:17
*** blmartin__ has joined #openstack-swift21:20
*** blmartin__ has quit IRC21:20
*** blmartin has quit IRC21:23
*** david-lyle has quit IRC21:26
*** david-lyle has joined #openstack-swift21:29
*** albertom has quit IRC21:33
*** pberis has quit IRC21:39
*** thumpba_ has joined #openstack-swift21:40
*** garthb has joined #openstack-swift21:41
*** thumpba has quit IRC21:43
*** thumpba_ has quit IRC21:46
*** thumpba has joined #openstack-swift21:46
*** david-lyle has quit IRC21:48
* mattoliverau goes to find breakfast21:53
notmynamehttps://wiki.openstack.org/wiki/Swift/PriorityReviews has been updated21:54
*** david-lyle has joined #openstack-swift21:54
*** kota_ has quit IRC22:00
EmilienMnotmyname: I would like your thoughts on https://review.openstack.org/22964622:00
*** ccavanna has joined #openstack-swift22:00
*** jkugel has left #openstack-swift22:02
*** garthb has quit IRC22:03
openstackgerritClay Gerrard proposed openstack/swift: wip: maybe squash this with the other  https://review.openstack.org/22965822:03
torgomaticEmilienM: why are you using swift-informant? Is there something in there that Swift's own statsd events doesn't cover?22:04
EmilienMI don't use it22:04
EmilienMI'm a Puppet OpenStack reviewer and I have to deal with this kind of patch22:04
EmilienMI'm not the author of the patch22:05
torgomaticoh22:05
*** thumpba has quit IRC22:05
EmilienMhttps://github.com/pandemicsyn/swift-informant/issues/1722:05
dfgmattoliverau: hey22:06
dfgmattoliverau: i kinda want to try out https://review.openstack.org/#/c/117710/ . are you working on other things? i wouldn't mind trying to figure out whats up with it if you're busy22:08
*** jrichli has quit IRC22:11
*** david-lyle has quit IRC22:15
*** hrou has joined #openstack-swift22:17
*** garthb has joined #openstack-swift22:19
claygdfg: yay!22:22
claygdfg: all the cool kids use patch 117710 now22:22
patchbotclayg: https://review.openstack.org/#/c/117710/ - Add concurrent reads option to proxy22:22
dfgoh ya22:22
*** chsc has quit IRC22:23
*** marcusvrn_ has quit IRC22:25
mattoliveraudfg: I'm working in all the things.. Actually working on liberty reviews and trying to get sharding take 3 to a place we can discuss at summit. If you want to play, then go for your life :) but keep me in the loop so I can stay involved :)22:29
*** devlaps has quit IRC22:30
mattoliverau+ Id appreciate you expertise in the matter :)22:30
dfgmattoliverau: ok cool. just didn't us both working on it at the same time :) if i figure anything out i'll just put it up on the review. thx22:30
mattoliverauCools, thanks for looking at it.22:30
dfgmattoliverau: ya- i hope you're not expecting much :p22:31
mattoliveraudfg: just a fresh set of eyes would be awesome22:31
*** trifon has quit IRC22:32
notmynameEmilienM: I don't have a particularly strong opinion on that patch. If your project wants to do the configuration for stuff that isn't managed as part of openstack, that's great. If you chose not to, that's great too.22:32
*** earlephilhower has quit IRC22:32
dfgmattoliverau: cool- i'll let you know when i have something22:33
EmilienMnotmyname: #222:33
mattoliveraunotmyname: if you simply need a keystone server you can test the patch with, you can use one of mine, theyre just RS cloud servers I built for testing this stuff :)22:33
notmynamemattoliverau: ah, cool22:34
EmilienMnotmyname: I would rather encourage upstream contribution to openstack/swift and -1 this kind of patch.22:34
*** jlhinson has quit IRC22:36
*** david-lyle has joined #openstack-swift22:36
notmynameEmilienM: I think there's value in promoting the broad ecosystem of stuff that doesn't necessarily need or want to move to the openstack namespace. just the fact that someone has proposed that patch might mean there are users out there needing it22:37
EmilienMnotmyname: I'm not sure it's sane for our Puppet modules to support anything not in OpenStack.22:39
notmynameEmilienM: but I can also understand how the "openstack puppet" project would want to limit scope to just configuring code projects that are part of openstack22:39
*** hrou has quit IRC22:42
acoleskota has gone away again!22:43
acoleskota - if you read this - if you want to develop the tests you had pasted for the ECResponseBucketController than that would be great!22:44
acoleskota - i use panicbnc22:45
*** proteusguy_ has quit IRC22:47
*** lyrrad has quit IRC22:47
*** lyrrad has joined #openstack-swift22:47
openstackgerritAlistair Coles proposed openstack/swift: Add POST capability to ssync for .meta files  https://review.openstack.org/13849822:47
openstackgerritClay Gerrard proposed openstack/swift: wip: maybe squash this with the other  https://review.openstack.org/22965822:48
*** djc_ has joined #openstack-swift22:52
djc_can swift write files encrypted?22:53
torgomaticif you feed it an encrypted file, swift can write it down22:53
djc_torgomatic so encryption would have to be done client side?22:54
djc_swift won't encrypt files for you as it is writes to disk22:55
notmynamedjc_: yes, but there is some ongoing work for swift to do server-side encryption22:55
djc_notmyname: ok. thanks.22:56
*** kota_ has joined #openstack-swift22:56
*** ChanServ sets mode: +v kota_22:56
notmynamekota_ is back :-)22:56
acoleskota_: !22:56
kota_notmyname, acoles: morning, again22:56
kota_something, up?22:57
notmynameacoles: when do you think kota_ will get an IRC bouncer?22:57
acoleskota_: you are 3 minutes too early for 'morning' :P22:57
kota_:\22:57
acolesnotmyname: i think we should all club together and take one to tokyo for him :P22:57
notmyname:-)22:58
*** chsc has joined #openstack-swift22:59
kota_acoles: ah, yeah I'll take a ticket to write some tests for ResponseBuckets22:59
mattoliverauIf he uses quassel I can add him an account to my quassel server in the cloud ;)22:59
*** proteusguy_ has joined #openstack-swift23:00
kota_quassel?23:00
*** dencaval has joined #openstack-swift23:00
acolescloud?23:00
acoles;)23:01
acoleskota_: super, thanks.23:01
mattoliveraulol23:01
kota_maybe I have to google about IRC bouncer23:02
mattoliveraukota_: http://quassel-irc.org/ <-- it is way people can sometimes wake me up if I forget to silence my phone (there an android or iphone app) so am always connected even if I'm out and about.. the wife loves it :P23:04
claygkota_: irssi in screen on a cloud server ftw23:04
mattoliveraubut yeah a standard IRC bouncer is awesome to23:04
mattoliverauI just like having it buzz and use it on my phone23:04
kota_thx23:04
*** alejandrito has quit IRC23:04
*** bill_az has joined #openstack-swift23:04
acolesmattoliverau: i keep forgetting to ping you each night to check you have your phone silenced :)23:07
claygheh23:07
mattoliverauacoles: lol, and I appreciate that you keep forgetting :P23:07
*** km has joined #openstack-swift23:07
* acoles goes to make calendar reminder23:07
*** marzif has quit IRC23:10
notmynamemattoliverau: just gets you ready for the kid. at that point acoles can turn off his calendar reminder ;-)23:11
*** ccavanna_ has joined #openstack-swift23:22
*** djc_ has quit IRC23:25
*** ccavanna has quit IRC23:25
*** kei_yama has joined #openstack-swift23:26
openstackgerritAlistair Coles proposed openstack/swift: Enable object server to return non-durable data  https://review.openstack.org/21527623:27
acoles^^ rebased on ssync fix23:27
claygacoles: nice23:27
mattoliveraunotmyname: lol,  training, I like it :)23:37
*** hrou has joined #openstack-swift23:38
acolesok i'm done, have good days/afternoon/whatever you have left23:39
*** acoles is now known as acoles_23:41
*** mac_ified has joined #openstack-swift23:52
openstackgerritClay Gerrard proposed openstack/swift: Validate against duplicate device part replica assignment  https://review.openstack.org/22279923:54
claygok patch 222799 is as far along as I'm hoping to get it23:55
patchbotclayg: https://review.openstack.org/#/c/222799/ - Validate against duplicate device part replica ass...23:55
claygand it's not *total* ass23:55
claygit might even be "ok" by some definitions - one might even go out on a limb and describe it as "better" than what we have now23:55
claygacctually - i have some empirical results that lead me to believe it's better - although I'm not really sure how to get them in a form that's appropriate to share - maybe I can get something together for Toyko - esp. if I manage to do something better after we get "done" with L23:57
claygI need to break off for a bit23:57
claygI'll start reviewing the other changes when the kids are in bed23:57
claygg'night23:57

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