| *** minwoob has quit IRC | 00:01 | |
| *** sorrison_laptop has joined #openstack-swift | 00:06 | |
| ho | good morning! | 00:14 |
|---|---|---|
| clayg | good morning ho! | 00:14 |
| ho | clayg: hello! | 00:16 |
| *** kota_ has joined #openstack-swift | 00:18 | |
| *** ChanServ sets mode: +v kota_ | 00:18 | |
| notmyname | hello | 00:19 |
| kota_ | hello | 00:20 |
| *** asettle has joined #openstack-swift | 00:21 | |
| ho | mita sounds like m-eat-a (but 2 syllables), what do you think kota_? | 00:22 |
| notmyname | mee-ta | 00:23 |
| kota_ | ho: hmm, sounds true | 00:24 |
| ho | notmyname: yes! | 00:24 |
| kota_ | ho: Was the release name determined? | 00:25 |
| ho | kota_: I don't know but notmyname wrote it here. "-) | 00:26 |
| ho | s/"-)/:-)/ | 00:26 |
| kota_ | ho: ah, ok. | 00:27 |
| kota_ | check out the irc logs... | 00:27 |
| *** CR7 has joined #openstack-swift | 00:28 | |
| ho | zaitcev: you answered it (good japanese skill) :-) | 00:28 |
| *** dmorita has joined #openstack-swift | 00:28 | |
| *** jasondotstar has quit IRC | 00:29 | |
| *** jamielennox|away is now known as jamielennox | 00:29 | |
| kota_ | ok, no official anouncement about Mita release... | 00:31 |
| ho | kota_: yeah, what was your preferable name? | 00:32 |
| *** esker has quit IRC | 00:32 | |
| kota_ | ho: Musashi or Musashino | 00:32 |
| kota_ | ho: second one is the location of my office (and home) | 00:33 |
| ho | kota_: Mitaka? I voted Musashi :-) | 00:33 |
| clayg | ho: Musashi would have been better for sure | 00:33 |
| ho | clayg: you love japanese monkey name right? :-) I forgot the english name. | 00:35 |
| *** ktsuyuzaki has joined #openstack-swift | 00:36 | |
| *** resker has joined #openstack-swift | 00:36 | |
| ktsuyuzaki | kota_ -> ktsuyuzaki, unfortunately my NW disabled a bit. | 00:37 |
| clayg | ho: Yeah I liked macaque it because saying it sounded like an american from the north east saying "my cock" - where "cock" (the rooster) is slang for penis - i'm easily amused | 00:38 |
| *** kota_ has quit IRC | 00:38 | |
| clayg | ho: I liked Musashi because - bad ass ninja | 00:38 |
| ho | clayg: lol | 00:39 |
| clayg | ho: I have litteraly nothing invested in mita in anyway - it's another name I couldn't care less about - I liked Grizzly - bears are cool | 00:40 |
| ho | clayg: you should invest to find out interestings for mita :-) | 00:45 |
| clayg | ho: oh yeah!? is it interesting? | 00:47 |
| clayg | ho: someone said kyocera used to be called mita? | 00:47 |
| clayg | ho: so... that's crappy phones... what else? | 00:47 |
| openstackgerrit | Zack M. Davis proposed openstack/python-swiftclient: add tempurl command to swift.1 man page https://review.openstack.org/194461 | 00:51 |
| *** DericHorn-HP has joined #openstack-swift | 00:53 | |
| ho | clayg: yeah, it's interesting for me. :-) I checked kyocera's history http://global.kyocera.com/company/summary/history/2000yrs.html | 00:54 |
| clayg | ho: (: that's awesome! | 00:57 |
| *** resker has quit IRC | 01:00 | |
| openstackgerrit | John Dickinson proposed openstack/swift: do container listing updates in another (green)thread https://review.openstack.org/189080 | 01:02 |
| notmyname | I'm having a tough time getting any tests to show measurable differences | 01:02 |
| notmyname | on that patch | 01:02 |
| notmyname | but I'm also not going to do much on it for a while (or I'm not counting on that) | 01:03 |
| *** esker has joined #openstack-swift | 01:10 | |
| *** jasondotstar has joined #openstack-swift | 01:15 | |
| *** Kennan has quit IRC | 01:16 | |
| *** Kennan has joined #openstack-swift | 01:16 | |
| *** esker has quit IRC | 01:17 | |
| *** ktsuyuzaki has quit IRC | 01:20 | |
| *** kota_ has joined #openstack-swift | 01:20 | |
| *** ChanServ sets mode: +v kota_ | 01:20 | |
| *** peluse has joined #openstack-swift | 01:26 | |
| *** esker has joined #openstack-swift | 01:27 | |
| *** kota_ has quit IRC | 01:31 | |
| *** kota_ has joined #openstack-swift | 01:32 | |
| *** ChanServ sets mode: +v kota_ | 01:32 | |
| *** esker has quit IRC | 01:37 | |
| *** asettle has quit IRC | 01:44 | |
| openstackgerrit | Pradeep Kumar Singh proposed openstack/python-swiftclient: Added check for negative segment-size https://review.openstack.org/194464 | 01:50 |
| *** CaioBrentano has joined #openstack-swift | 01:50 | |
| openstackgerrit | Pradeep Kumar Singh proposed openstack/python-swiftclient: Added check for negative segment-size https://review.openstack.org/194464 | 01:51 |
| *** CaioBrentano has quit IRC | 01:54 | |
| *** mwheckmann has joined #openstack-swift | 01:55 | |
| *** CR7 has quit IRC | 02:00 | |
| *** kei_yama has quit IRC | 02:10 | |
| *** sorrison_laptop has quit IRC | 02:20 | |
| *** sorrison_laptop has joined #openstack-swift | 02:21 | |
| *** sorrison_laptop has quit IRC | 02:22 | |
| *** bkopilov has quit IRC | 02:29 | |
| *** DericHorn-HP has quit IRC | 02:30 | |
| *** DericHorn-HP has joined #openstack-swift | 02:39 | |
| *** jasondotstar has quit IRC | 02:47 | |
| *** kei_yama has joined #openstack-swift | 02:52 | |
| *** DericHorn-HP has quit IRC | 02:57 | |
| zaitcev | https://twitter.com/MirantisIT/status/613147417081348096 - it's a promoted tweet but funny | 03:05 |
| *** silor has joined #openstack-swift | 03:09 | |
| *** blmartin has quit IRC | 03:13 | |
| *** mwheckmann has quit IRC | 03:21 | |
| *** mwheckmann has joined #openstack-swift | 03:32 | |
| *** mgarza has joined #openstack-swift | 03:35 | |
| *** DericHorn-HP has joined #openstack-swift | 03:38 | |
| *** DericHorn-HP has quit IRC | 03:38 | |
| *** wbhuber_ has quit IRC | 03:45 | |
| *** bkopilov has joined #openstack-swift | 03:47 | |
| *** mwheckmann has quit IRC | 03:51 | |
| *** mwheckmann has joined #openstack-swift | 03:54 | |
| *** mgarza has quit IRC | 03:58 | |
| *** mwheckmann has quit IRC | 04:02 | |
| *** mgarza_ has joined #openstack-swift | 04:38 | |
| *** proteusguy has quit IRC | 04:40 | |
| *** mgarza__ has joined #openstack-swift | 04:41 | |
| *** mgarza_ has quit IRC | 04:43 | |
| *** zaitcev has quit IRC | 04:51 | |
| *** jith_ has joined #openstack-swift | 04:55 | |
| jith_ | hi all.. i have configured SAIO in a VM and devstack without swift in another VM... then using CLI i tried to configure the devstack keystone for authenticating the swift(SAIO).. but only swift stat works.. but not the list and all... | 05:21 |
| jith_ | is it necessary i should execute swift enabled devstack script in SAIO for enabling the swift in dashboard? | 05:23 |
| jith_ | ho: u there? | 05:23 |
| *** SkyRocknRoll has joined #openstack-swift | 05:25 | |
| *** SkyRocknRoll has joined #openstack-swift | 05:25 | |
| *** asettle has joined #openstack-swift | 05:28 | |
| *** proteusguy has joined #openstack-swift | 05:32 | |
| openstackgerrit | Merged openstack/swift: EC Ssync: Update parms to include node and frag indices https://review.openstack.org/191521 | 05:32 |
| *** proteusguy_ has joined #openstack-swift | 05:34 | |
| *** proteusguy has quit IRC | 05:38 | |
| *** proteusguy_ has quit IRC | 05:39 | |
| *** proteusguy_ has joined #openstack-swift | 05:52 | |
| *** proteusguy_ has quit IRC | 05:56 | |
| *** proteusguy has joined #openstack-swift | 05:56 | |
| *** kota_ has quit IRC | 05:58 | |
| *** yuan has joined #openstack-swift | 06:04 | |
| *** Kennan has quit IRC | 06:11 | |
| *** Kennan2 has joined #openstack-swift | 06:11 | |
| *** mgarza__ has quit IRC | 06:16 | |
| ho | jith_: i'm back. (but I only have 15 min for next meeting) | 06:16 |
| jith_ | ho: thanks for ur response... yes in keystone i have created swift credentials and endpoints and configured keystone auth and auth token in proxy-server.conf.. so i can able to see the swift stat.. but not list | 06:20 |
| ho | jith_: you have a SAIO env and devstack without swift. then you registered swift related users and roles to keystone in the devstack | 06:20 |
| jith_ | list remains empty | 06:20 |
| jith_ | ho: exactly | 06:20 |
| jith_ | in devstack dashboard i couldnt see the swift enabled.. so should i run devstack script with swift in SAIO... i m sorry if i m wrong | 06:22 |
| ho | jith_: thanks for the explanation! i don't know about dashborad but i think that dashboard gets swift's endoint from keystone, it might not work. i'm not sure... | 06:25 |
| ho | jith_: can you get info from swift by swift cli/api? | 06:26 |
| jith_ | ho: thanks :) | 06:26 |
| ho | jith_: sorry, i have to go to meeting. | 06:29 |
| jith_ | ho: its ok... no issues... thanks for the response... | 06:29 |
| *** rledisez has joined #openstack-swift | 06:34 | |
| *** ianbrown has quit IRC | 06:38 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift: Add one more test for ssync_receiver https://review.openstack.org/194518 | 06:39 |
| *** Kennan2 is now known as Kennan | 06:53 | |
| openstackgerrit | Kuo-tung Kao proposed openstack/swift: Use testtools for base test class https://review.openstack.org/192960 | 06:57 |
| *** pksingh has joined #openstack-swift | 06:58 | |
| *** ppai has joined #openstack-swift | 07:04 | |
| *** bkopilov has quit IRC | 07:05 | |
| *** bkopilov has joined #openstack-swift | 07:21 | |
| *** haypo has joined #openstack-swift | 07:28 | |
| openstackgerrit | Christian Schwede proposed openstack/swift: Replace xrange() with six.moves.range() https://review.openstack.org/185460 | 07:29 |
| haypo | cschwede, ah, thanks for the updated commit title :) | 07:39 |
| cschwede | haypo: you’re welcome :) hope to see this merged real soon now! | 07:39 |
| haypo | cschwede, yeah me too, i start hating rebasing each morning | 07:40 |
| *** kota_ has joined #openstack-swift | 07:58 | |
| *** ChanServ sets mode: +v kota_ | 07:58 | |
| *** kota_ has quit IRC | 08:07 | |
| openstackgerrit | Merged openstack/swift: Get better at closing WSGI iterables. https://review.openstack.org/193303 | 08:16 |
| *** SkyRocknRoll has quit IRC | 08:24 | |
| *** occupant has quit IRC | 08:34 | |
| *** yaakov_g has joined #openstack-swift | 08:34 | |
| *** occupant has joined #openstack-swift | 08:36 | |
| yaakov_g | . | 08:43 |
| *** SkyRocknRoll has joined #openstack-swift | 08:48 | |
| *** SkyRocknRoll has quit IRC | 08:48 | |
| *** SkyRocknRoll has joined #openstack-swift | 08:48 | |
| *** aix has quit IRC | 08:53 | |
| *** aix has joined #openstack-swift | 09:07 | |
| *** yaakov_g has quit IRC | 09:08 | |
| *** bkopilov has quit IRC | 09:17 | |
| *** dmorita has quit IRC | 09:27 | |
| *** bkopilov has joined #openstack-swift | 09:29 | |
| *** proteusguy has quit IRC | 09:31 | |
| *** jasondotstar has joined #openstack-swift | 09:37 | |
| *** bkopilov has quit IRC | 09:38 | |
| *** bkopilov has joined #openstack-swift | 09:38 | |
| *** kjelly has joined #openstack-swift | 09:39 | |
| *** silor has quit IRC | 09:44 | |
| *** marzif has joined #openstack-swift | 09:48 | |
| *** proteusguy has joined #openstack-swift | 09:49 | |
| openstackgerrit | Kuo-tung Kao proposed openstack/swift: Use testtools for base test class https://review.openstack.org/192960 | 09:52 |
| kjelly | I send a patch. https://review.openstack.org/#/c/192960/ . And I found that it's easy to conflict or make my patch failed if other patch change test codes. Are there good ideas to solve the problem ? | 09:55 |
| *** mwheckmann has joined #openstack-swift | 09:59 | |
| kjelly | testtools library requires that the subclass of testcase should call super.setup or super.teardown if you override setup or teardown. The requirement often break my patch if the test code of other patch forget to call super.setup or super.teardown. | 10:00 |
| *** mwheckmann has quit IRC | 10:04 | |
| *** pksingh has quit IRC | 10:09 | |
| *** SkyRocknRoll_ has joined #openstack-swift | 10:38 | |
| *** silor has joined #openstack-swift | 10:55 | |
| *** ho has quit IRC | 10:57 | |
| *** joeljwright has joined #openstack-swift | 11:00 | |
| *** ChanServ sets mode: +v joeljwright | 11:00 | |
| *** jamielennox is now known as jamielennox|away | 11:00 | |
| *** breitz1 has joined #openstack-swift | 11:02 | |
| *** breitz has quit IRC | 11:02 | |
| *** aix has quit IRC | 11:15 | |
| *** silor1 has joined #openstack-swift | 11:16 | |
| *** silor has quit IRC | 11:19 | |
| *** silor1 has quit IRC | 11:19 | |
| *** silor has joined #openstack-swift | 11:22 | |
| *** cdelatte has quit IRC | 11:23 | |
| *** silor1 has joined #openstack-swift | 11:26 | |
| *** silor has quit IRC | 11:27 | |
| *** aix has joined #openstack-swift | 11:28 | |
| *** silor1 has quit IRC | 11:32 | |
| *** SkyRocknRoll has quit IRC | 11:36 | |
| *** cdelatte has joined #openstack-swift | 11:50 | |
| *** mmcardle has quit IRC | 11:54 | |
| *** jith_ has quit IRC | 12:01 | |
| *** jasondotstar has quit IRC | 12:06 | |
| *** SkyRocknRoll_ has quit IRC | 12:13 | |
| *** knl has joined #openstack-swift | 12:15 | |
| *** km has quit IRC | 12:16 | |
| *** mgarza_ has joined #openstack-swift | 12:19 | |
| *** agentle has joined #openstack-swift | 12:26 | |
| *** aix has quit IRC | 12:31 | |
| *** mmcardle has joined #openstack-swift | 12:34 | |
| *** agentle has quit IRC | 12:35 | |
| *** agentle has joined #openstack-swift | 12:36 | |
| *** bkopilov has quit IRC | 12:36 | |
| *** MVenesio has joined #openstack-swift | 12:37 | |
| *** amoturi has joined #openstack-swift | 12:39 | |
| *** MVenesio has quit IRC | 12:39 | |
| *** janonymous_ has joined #openstack-swift | 12:47 | |
| *** aix has joined #openstack-swift | 12:47 | |
| *** CaioBrentano has joined #openstack-swift | 12:48 | |
| *** petertr7_away is now known as petertr7 | 12:59 | |
| *** amoturi has quit IRC | 13:17 | |
| *** jasondotstar has joined #openstack-swift | 13:17 | |
| *** kei_yama has quit IRC | 13:17 | |
| *** openstack has quit IRC | 13:17 | |
| *** openstack has joined #openstack-swift | 13:19 | |
| *** petertr7_away is now known as petertr7 | 13:20 | |
| *** esker has joined #openstack-swift | 13:20 | |
| *** silor has joined #openstack-swift | 13:28 | |
| *** aix has quit IRC | 13:29 | |
| *** janonymous_ has quit IRC | 13:38 | |
| *** aix has joined #openstack-swift | 13:42 | |
| *** agentle has joined #openstack-swift | 13:53 | |
| *** agentle has quit IRC | 14:01 | |
| *** bkopilov has joined #openstack-swift | 14:02 | |
| *** thurloat_isgone is now known as thurloat | 14:08 | |
| *** blmartin has joined #openstack-swift | 14:13 | |
| *** blmartin has quit IRC | 14:13 | |
| *** mgarza_ has quit IRC | 14:17 | |
| *** blmartin has joined #openstack-swift | 14:24 | |
| *** jrichli has joined #openstack-swift | 14:30 | |
| *** jasondotstar has quit IRC | 14:31 | |
| *** pgbridge has joined #openstack-swift | 14:40 | |
| *** ppai has quit IRC | 14:44 | |
| *** jasondotstar has joined #openstack-swift | 14:46 | |
| *** mgarza_ has joined #openstack-swift | 14:48 | |
| *** minwoob has joined #openstack-swift | 14:55 | |
| *** mwheckmann has joined #openstack-swift | 15:02 | |
| *** vinsh has joined #openstack-swift | 15:08 | |
| haypo | cschwede, ah! https://review.openstack.org/#/c/185460/ got two +2 but it's not approved yet :-p | 15:15 |
| *** esker has quit IRC | 15:16 | |
| haypo | cschwede, hum, i wanted to ask Pete Zaitcev why he didn't approve the patch but i don't see him connected on this channel | 15:16 |
| *** mwheckmann has quit IRC | 15:24 | |
| *** petertr7 is now known as petertr7_away | 15:31 | |
| *** haypo has quit IRC | 15:41 | |
| *** mgarza_ has quit IRC | 15:42 | |
| *** wbhuber_ has joined #openstack-swift | 15:43 | |
| *** mgarza_ has joined #openstack-swift | 15:43 | |
| *** mgarza_ has quit IRC | 15:47 | |
| *** wbhuber has quit IRC | 15:47 | |
| *** mgarza_ has joined #openstack-swift | 15:51 | |
| *** zaitcev has joined #openstack-swift | 15:59 | |
| *** ChanServ sets mode: +v zaitcev | 15:59 | |
| *** mmcardle has left #openstack-swift | 16:03 | |
| *** vinsh_ has joined #openstack-swift | 16:03 | |
| *** joeljwright has quit IRC | 16:03 | |
| *** nadeem has joined #openstack-swift | 16:05 | |
| *** vinsh has quit IRC | 16:06 | |
| *** thurloat is now known as thurloat_isgone | 16:08 | |
| *** jlhinson has joined #openstack-swift | 16:09 | |
| *** knl has quit IRC | 16:16 | |
| *** esker has joined #openstack-swift | 16:18 | |
| *** esker has quit IRC | 16:21 | |
| *** rledisez has quit IRC | 16:22 | |
| *** vinsh_ has quit IRC | 16:22 | |
| *** vinsh has joined #openstack-swift | 16:23 | |
| *** joeljwright has joined #openstack-swift | 16:24 | |
| *** ChanServ sets mode: +v joeljwright | 16:24 | |
| *** delattec has joined #openstack-swift | 16:33 | |
| *** mfalatic has joined #openstack-swift | 16:33 | |
| *** proteusguy has quit IRC | 16:33 | |
| cschwede | haypo: well, I think Pete didn’t add his +A because all of us three are from the same company - someone else should have a look to | 16:34 |
| *** cdelatte has quit IRC | 16:36 | |
| *** petertr7_away is now known as petertr7 | 16:36 | |
| *** blmartin has quit IRC | 16:40 | |
| *** blmartin has joined #openstack-swift | 16:42 | |
| *** jasondotstar has quit IRC | 16:43 | |
| notmyname | good morning | 16:43 |
| blmartin | good day | 16:44 |
| *** proteusguy has joined #openstack-swift | 16:46 | |
| *** vinsh_ has joined #openstack-swift | 16:48 | |
| *** amoturi has joined #openstack-swift | 16:50 | |
| *** marzif_ has joined #openstack-swift | 16:50 | |
| *** vinsh has quit IRC | 16:52 | |
| *** jasondotstar has joined #openstack-swift | 16:52 | |
| *** delatte has joined #openstack-swift | 16:53 | |
| *** mfalatic has quit IRC | 16:54 | |
| *** delattec has quit IRC | 16:55 | |
| *** marzif has quit IRC | 16:56 | |
| *** esker has joined #openstack-swift | 17:00 | |
| *** dontalton has joined #openstack-swift | 17:01 | |
| *** esker has quit IRC | 17:04 | |
| *** DericHorn-HP has joined #openstack-swift | 17:08 | |
| *** mfalatic has joined #openstack-swift | 17:09 | |
| MooingLemur | good 🐄rning | 17:11 |
| openstackgerrit | Michael Barton proposed openstack/swift: go: add a few unit tests https://review.openstack.org/194750 | 17:15 |
| *** jlhinson has quit IRC | 17:15 | |
| *** esker has joined #openstack-swift | 17:23 | |
| *** esker has quit IRC | 17:23 | |
| *** esker has joined #openstack-swift | 17:24 | |
| *** esker has quit IRC | 17:27 | |
| *** gyee_ has quit IRC | 17:33 | |
| *** zhill has joined #openstack-swift | 17:33 | |
| *** gyee has joined #openstack-swift | 17:36 | |
| *** breitz1 is now known as breitz | 17:37 | |
| *** jlhinson has joined #openstack-swift | 17:41 | |
| *** pgbridge has quit IRC | 17:44 | |
| clayg | on the debate going on in https://bugs.launchpad.net/swift/+bug/1467668 | 18:03 |
| openstack | Launchpad bug 1467668 in OpenStack Object Storage (swift) "Multi Range Read with with server side copy results in a MIME" [Undecided,Confirmed] | 18:03 |
| clayg | 1) we probably can't change existing behavior - we'd need a new qs arg for copy or header to enable this multi-range-copy-prune behavior | 18:03 |
| clayg | ... not that anyone was using multi-range copy on purpose - because ... | 18:03 |
| clayg | do we store the content-type (including the boundry) correctly with the new mime object? | 18:04 |
| clayg | like if you multipart copy into a new mime document - on a GET do you have enough info in the response to pull the envelope apart? | 18:05 |
| *** jasondotstar has quit IRC | 18:07 | |
| *** blmartin has quit IRC | 18:08 | |
| clayg | hrmm... idk - looks like it's all there | 18:10 |
| clayg | so... not night and day | 18:12 |
| clayg | it's probably just a feature request - and with an optional qs or header - I agree with notmyname - it's something the client couldn't easily do on their own and has some interesting properties | 18:13 |
| notmyname | clayg: I'm not sure if "existing behavior" is a thing for this one. I feel that what's there now is kindof and accident, and it's certainly not supported in any client cli/sdk | 18:14 |
| *** jasondotstar has joined #openstack-swift | 18:14 | |
| clayg | yeah maybe the question is if it's ever useful to *store* a multi-part mime document - or if that's really just ment to be a protocol for transfer | 18:16 |
| clayg | well i suppose if you were doing email archving you might backup a multi-part mime document | 18:17 |
| clayg | maybe that doesn't help | 18:18 |
| notmyname | I'm not sure if that matters for this case | 18:18 |
| notmyname | ie content of object you might store with other operations | 18:18 |
| clayg | notmyname: I *just* said it's not helpful - man guy can't think outloud ;) | 18:21 |
| *** pgbridge has joined #openstack-swift | 18:22 | |
| notmyname | lol | 18:23 |
| clayg | notmyname: i'm trying to imagine someone doing multipart range requests with video/audio - get header info also some bytes... | 18:24 |
| clayg | notmyname: say there's some fancy caching layer mixed into swift where a youtube like link to offset 4:32 in video xyz goes viral - and they do a COPY of that request into a "hot" storage policy | 18:25 |
| clayg | the response of the "hot copy" should be exactly what came out of the GET | 18:25 |
| notmyname | yeah, that's a cool use case | 18:26 |
| clayg | ... unless that's not what you want - then you say COPY?multipart_manifest=false | 18:26 |
| clayg | notmyname: i totally made it up | 18:26 |
| clayg | audio/video streaming is the only usecase I know of for multipart range requests | 18:26 |
| notmyname | I was thinking more like extracting bits of text files | 18:26 |
| clayg | anyway - I think it's pretty intresting that there is an API command that can create new data that didn't exist before | 18:28 |
| clayg | any the ability to tweak that command via query string (either multipart of concact) sees pretty cool | 18:28 |
| occupant | so we're pushing loads of video into our swift cluster and seemingly hitting its max and I'm trying to understand where it's bottlenecking. looking at iostat, it seems like 2-3 drives (out of 30) will be used at 99% whereas the rest are sitting relatively idle, causing iowait and proxy-to-storage timeouts. I kinda expected that disk access would be more distributed. Any thoughts? | 18:28 |
| clayg | notmyname: ^ weird timing?! | 18:29 |
| clayg | occupant: do you have some "hot" videos :) | 18:29 |
| wer | occupant: sounds like you are hotspotting on container lookup or something. | 18:29 |
| occupant | clayg: the load is us doing mass uploads, not user activity. | 18:29 |
| clayg | notmyname: i wonder if EC might help distribute IO load a little bit?! | 18:30 |
| clayg | occupant: oh - just use slo's | 18:30 |
| occupant | SLOs? | 18:30 |
| clayg | wer: oh - yeah that's a good point too | 18:30 |
| wer | :) been there | 18:31 |
| clayg | occupant: how big are the videos? do you upload whole files or use the large object api to split them up into 10-100M chunks? | 18:31 |
| occupant | videos are variable but can be in the gigs range, but we're chunking. brb, meeting | 18:32 |
| notmyname | clayg: doug says the multi-range copy use case is for pruning (dedupe) backups and for slicing out parts of videos to give the smaller data to other people | 18:32 |
| *** pgbridge has quit IRC | 18:34 | |
| clayg | and the video people *also* want to just slam the data into a single object? I guess I could see that if you know the video format really well and are super careful with your frame offsets... | 18:34 |
| clayg | notmyname: doesn't really matter I suppose - I'd probaby still arguee we can't get rid of the multi-part behavior or change it to do concact w/o an extra qs or header | 18:35 |
| clayg | wer: yeah if occupant is chunking already it probably is the container updates - use more contaiers or move the containers onto their own disks I suppose. | 18:36 |
| wer | shard! :) | 18:36 |
| clayg | wer: do you run containers on ssd's? | 18:37 |
| wer | no clayg. I just run a 24 disk spinny jbod setup. | 18:37 |
| notmyname | clayg: yeah, maybe the video use case isn't as big of a thing (after talking with doug more) | 18:38 |
| clayg | wer: and the same disks are in both the object and a&c rings? | 18:39 |
| wer | yeah, I keep it really simple. Assume all rings are the same. | 18:39 |
| clayg | wer: yeah I really like that - cool setup | 18:40 |
| wer | at one point I separated the container stuff.... but I dropped the complexity in favor of sharding via some middleware. | 18:40 |
| clayg | oh interesting! | 18:40 |
| wer | thanks! we've been really happy with it. doing a lot of data on ~20mb - 1gig files. | 18:41 |
| notmyname | :-) | 18:41 |
| wer | The only thing I didn't do... was replicate the cluster. I manually sync to another cluster. | 18:41 |
| wer | I was too scared. | 18:41 |
| wer | I'm in the dark ages and things have moved a lot since I deployed. | 18:42 |
| *** aix has quit IRC | 18:42 | |
| wer | it's all guids... so sharding was easy. human chosen names could easily be a candidate for hotspotting IMO | 18:43 |
| *** janonymous_ has joined #openstack-swift | 18:43 | |
| *** marzif_ has quit IRC | 18:43 | |
| *** DericHorn-HP has quit IRC | 19:00 | |
| *** DericHorn-HP has joined #openstack-swift | 19:02 | |
| *** mgarza_ has quit IRC | 19:04 | |
| *** petertr7 is now known as petertr7_away | 19:05 | |
| openstackgerrit | Greg Lange proposed openstack/swift: go: add targeted replication https://review.openstack.org/194787 | 19:05 |
| *** DericHorn-HP_ has joined #openstack-swift | 19:07 | |
| openstackgerrit | Greg Lange proposed openstack/swift: go: check for empty path parts in object server https://review.openstack.org/194789 | 19:07 |
| *** DericHorn-HP has quit IRC | 19:07 | |
| *** DericHorn-HP_ is now known as DericHorn-HP | 19:07 | |
| *** petertr7_away is now known as petertr7 | 19:09 | |
| janonymous_ | Hi , I would like to have your opinion on https://blueprints.launchpad.net/swift/+spec/swift-bandit bp .? i donno it is much of relevance or not but must give a look at it. | 19:09 |
| *** mgarza_ has joined #openstack-swift | 19:15 | |
| *** openstackgerrit has quit IRC | 19:21 | |
| *** openstackgerrit has joined #openstack-swift | 19:22 | |
| *** janonymous_ has quit IRC | 19:29 | |
| *** fifieldt_ has joined #openstack-swift | 19:31 | |
| *** fifieldt has quit IRC | 19:34 | |
| *** lyrrad has joined #openstack-swift | 19:40 | |
| *** aix has joined #openstack-swift | 19:43 | |
| *** silor has quit IRC | 19:43 | |
| *** peluse has quit IRC | 19:46 | |
| *** peluse has joined #openstack-swift | 19:49 | |
| *** peluse has quit IRC | 19:57 | |
| notmyname | janonymous: yeah, i saw that. I'm not sure what to think of it yet | 19:57 |
| notmyname | janonymous: what sort of changes would be required in swift? I'd hope that it would be an external tool | 19:58 |
| *** annegentle has joined #openstack-swift | 20:04 | |
| *** delatte has quit IRC | 20:05 | |
| *** zhill has quit IRC | 20:07 | |
| *** zhill has joined #openstack-swift | 20:07 | |
| *** amoturi has quit IRC | 20:07 | |
| *** greghaynes has quit IRC | 20:08 | |
| *** jlhinson has quit IRC | 20:12 | |
| *** annegentle has quit IRC | 20:18 | |
| *** annegentle has joined #openstack-swift | 20:18 | |
| *** jasondotstar has quit IRC | 20:22 | |
| *** dontalton has quit IRC | 20:32 | |
| *** peluse has joined #openstack-swift | 20:33 | |
| *** dontalton has joined #openstack-swift | 20:33 | |
| *** zhill has quit IRC | 20:33 | |
| notmyname | mattoliverau: I'm looking at my notes. did you get a large container DB to test with? | 20:38 |
| *** peluse has quit IRC | 20:39 | |
| *** esker has joined #openstack-swift | 20:39 | |
| *** DericHorn-HP has quit IRC | 20:41 | |
| hurricanerix | notmyname: I am working on getting one to him. :) | 20:41 |
| notmyname | hurricanerix: cool. thanks :-) | 20:42 |
| *** DericHorn-HP has joined #openstack-swift | 20:42 | |
| notmyname | This morning I got a phone call from someone working on behalf of Red Hat and HP. wanted to know why I didn't go to this year's red hat summit and if I could share my thoughts on last year's summit. and if I had the appropriate connections at red hat and hp for all for my technology needs | 20:44 |
| notmyname | I thought it was pretty funny | 20:44 |
| notmyname | "yes, I know people at red hat and hp who can help me if I need anything" | 20:44 |
| notmyname | acoles_away: tdasilva: ^ | 20:44 |
| tdasilva | lol | 20:45 |
| tdasilva | you can catch the keynote streaming live right here: http://www.redhat.com/summit | 20:45 |
| tdasilva | cool demo going on right now | 20:45 |
| notmyname | I feel sorry for the low-level sales people who do cold calls and get my name/number. I'm such a terrible prospect for them :-) | 20:45 |
| hurricanerix | notmyname: maybe that is how they haze the new sales guys. =P | 20:54 |
| notmyname | hurricanerix: heh, maybe :-) | 20:55 |
| elmiko | notmyname: how much software would you like to buy? ;) | 20:58 |
| notmyname | elmiko: I'd like 3 software, please | 20:58 |
| elmiko | \o/ a sale! | 20:58 |
| notmyname | see how easy that was? | 20:58 |
| elmiko | hehe | 20:58 |
| *** nadeem has quit IRC | 20:59 | |
| *** DericHorn-HP has quit IRC | 20:59 | |
| *** annegentle has quit IRC | 21:03 | |
| *** JTen has joined #openstack-swift | 21:05 | |
| *** nadeem_ has joined #openstack-swift | 21:05 | |
| notmyname | hi JTen :-) | 21:06 |
| notmyname | JTen: just out of curiosity, anything you can share about how you're using swift? | 21:06 |
| *** petertr7 is now known as petertr7_away | 21:11 | |
| *** ruind has quit IRC | 21:14 | |
| *** kota_ has joined #openstack-swift | 21:15 | |
| *** ChanServ sets mode: +v kota_ | 21:15 | |
| *** annegentle has joined #openstack-swift | 21:16 | |
| *** peluse has joined #openstack-swift | 21:19 | |
| notmyname | torgomatic: do you remember this patch you wrote over a year ago to add the trans-id-suffix? https://review.openstack.org/#/c/93985/ | 21:28 |
| torgomatic | notmyname: it merged and I promptly forgot about it. why? | 21:28 |
| notmyname | looks like the rest of openstack is trying to reinvent it. https://review.openstack.org/#/c/156508/ | 21:28 |
| notmyname | timburke: joeljwright: does swiftclient return the transaction id on error? | 21:30 |
| *** ChanServ sets mode: +v peluse | 21:33 | |
| timburke | notmyname: what do we mean by "return"? i don't think it appears in the normal output... | 21:33 |
| *** annegentle has quit IRC | 21:34 | |
| notmyname | yeah, that's all. so if there's an error does the CLi print the transaction id? or if using the sdk can the user grab the transaction id of the failed request? | 21:34 |
| *** annegentle has joined #openstack-swift | 21:34 | |
| notmyname | seems people are asking that of various openstack clients | 21:35 |
| timburke | notmyname: if you use --debug, you'll get it, but that's the only way i'm aware of | 21:35 |
| notmyname | ok | 21:35 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-swiftclient: Updated from global requirements https://review.openstack.org/89250 | 21:50 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 21:50 |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift: Add policy support to dispersion tools https://review.openstack.org/185459 | 21:52 |
| *** esker has quit IRC | 21:53 | |
| *** ianbrown has joined #openstack-swift | 21:54 | |
| *** peluse has quit IRC | 21:54 | |
| *** DericHorn-HP has joined #openstack-swift | 21:56 | |
| *** pberis has joined #openstack-swift | 21:57 | |
| notmyname | https://review.openstack.org/#/c/186635/ looks like it's a big change on how openstack tests and identifies dependencies. probably would be worth a few of us looking over it from a swift perspective | 21:57 |
| *** blmartin has joined #openstack-swift | 21:58 | |
| *** pberis has quit IRC | 22:04 | |
| *** pberis has joined #openstack-swift | 22:06 | |
| *** kota_ has quit IRC | 22:08 | |
| *** annegent_ has joined #openstack-swift | 22:09 | |
| *** peluse has joined #openstack-swift | 22:13 | |
| *** annegentle has quit IRC | 22:13 | |
| *** jrichli has quit IRC | 22:17 | |
| *** CaioBrentano has quit IRC | 22:17 | |
| *** peluse has quit IRC | 22:18 | |
| *** CaioBrentano has joined #openstack-swift | 22:18 | |
| notmyname | reading over that patch (^) it seems that we'll get something we've wanted in swift for a while: global requirements compatibility checking instead of exact string matching | 22:18 |
| *** peluse has joined #openstack-swift | 22:18 | |
| *** dontalton2 has joined #openstack-swift | 22:19 | |
| *** annegent_ has quit IRC | 22:22 | |
| *** CaioBrentano has quit IRC | 22:22 | |
| *** annegentle has joined #openstack-swift | 22:22 | |
| *** gyee has quit IRC | 22:23 | |
| *** gyee has joined #openstack-swift | 22:27 | |
| *** peluse has quit IRC | 22:40 | |
| *** km has joined #openstack-swift | 22:57 | |
| notmyname | mattoliverau: what's the status of container sharding? when can we see something and play with it? | 23:01 |
| notmyname | cool https://twitter.com/ercarter/status/613476742053769216 | 23:02 |
| *** annegentle has quit IRC | 23:03 | |
| *** annegentle has joined #openstack-swift | 23:05 | |
| zaitcev | 186635 - "4. Not be fiction" | 23:06 |
| notmyname | :-) | 23:07 |
| zaitcev | periodic job that shifts gate practices sounds good | 23:09 |
| notmyname | mattoliverau: FYI I'm hearing more about problems that would be solved with the container sharding idea. ie existing software apps integrating with swift | 23:09 |
| notmyname | zaitcev: gate chaos monkey? | 23:10 |
| *** annegentle has quit IRC | 23:10 | |
| mattoliverau | notmyname: Morning, been sick all week, so need to get my bearings. But container shardings going well, haven't done as much as I liked over the last 2 weeks (sick and internal work got in the way). Hurricanerix has been awesome and organizing getting me we large containers to play with.. In the meantime playing with large containers I've built. I've hacked ssbench to allow me to specify which container to use, so I | 23:10 |
| mattoliverau | can benchmark reading and writing to and from shared and non shared containers. There are some pull requests and some testing thanks to blmartin. | 23:10 |
| notmyname | nice | 23:11 |
| mattoliverau | notmyname: awesome. Once I get some numbers together I'll tackle shrinking | 23:11 |
| blmartin | speaking of, let me know when you want me to post more bugfixes and tests mattoliverau. | 23:12 |
| blmartin | I don't want to flood your pull request page :) | 23:12 |
| mattoliverau | blmartin, will do, thanks man! Sorry I've been dead to the world, easing back into work today, so will try and look at them today. | 23:14 |
| notmyname | mattoliverau: thanks for the update. I hope you feel better soon | 23:14 |
| notmyname | I'm going to commute home now, but I'll definitely be online tonight. lots to do before tomorrow :-) | 23:15 |
| mattoliverau | I'm much better then I was.. Plans for the rest of the week, is getting some benchmarking numbers and pull requests :) | 23:15 |
| mattoliverau | Kk | 23:15 |
| *** mfalatic has quit IRC | 23:16 | |
| *** dontalton2 has quit IRC | 23:16 | |
| *** dontalton has quit IRC | 23:16 | |
| *** kei_yama has joined #openstack-swift | 23:17 | |
| *** mgarza_ has quit IRC | 23:22 | |
| *** jasondotstar has joined #openstack-swift | 23:23 | |
| *** blmartin is now known as blmartin_away | 23:29 | |
| *** DericHorn-HP has quit IRC | 23:45 | |
| *** ho has joined #openstack-swift | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!