| openstackgerrit | Tim Burke proposed openstack/swift master: Version DLOs, just like every other type of object https://review.openstack.org/446142 | 00:12 |
|---|---|---|
| openstackgerrit | Tim Burke proposed openstack/swift master: Factor out a bunch of common testing setup https://review.openstack.org/446185 | 00:22 |
| openstackgerrit | Tim Burke proposed openstack/swift master: Factor out a bunch of common testing setup https://review.openstack.org/446185 | 00:30 |
| *** tonanhngo has quit IRC | 00:33 | |
| *** chsc has quit IRC | 00:41 | |
| *** tonanhngo has joined #openstack-swift | 00:41 | |
| *** zhurong has joined #openstack-swift | 00:43 | |
| *** tonanhngo has quit IRC | 00:45 | |
| *** tonanhngo has joined #openstack-swift | 00:47 | |
| *** tonanhngo has quit IRC | 00:52 | |
| *** lifeless_ is now known as lifeless | 01:00 | |
| *** _JZ_ has quit IRC | 01:00 | |
| *** vinsh has joined #openstack-swift | 01:06 | |
| *** vinsh has quit IRC | 01:09 | |
| *** klrmn has quit IRC | 01:13 | |
| *** m_kazuhiro has joined #openstack-swift | 01:18 | |
| m_kazuhiro | good morning | 01:19 |
| *** zhurong has quit IRC | 01:23 | |
| *** NM has joined #openstack-swift | 01:24 | |
| *** NM has quit IRC | 01:40 | |
| *** zhurong has joined #openstack-swift | 01:42 | |
| *** ntata has joined #openstack-swift | 01:48 | |
| *** mmotiani has joined #openstack-swift | 01:51 | |
| *** vinsh has joined #openstack-swift | 01:57 | |
| *** sgundur- has joined #openstack-swift | 01:58 | |
| tone_z | Good morning | 01:59 |
| *** vinsh has quit IRC | 02:01 | |
| *** _JZ_ has joined #openstack-swift | 02:04 | |
| mattoliverau | m_kazuhiro, tone_z: morning | 02:06 |
| tone_z | :) | 02:07 |
| *** _JZ_ has quit IRC | 02:08 | |
| m_kazuhiro | mattoliverau, tone_z: morning! | 02:14 |
| mattoliverau | Here is one backport to fix one itermittant stable/newton swift error. https://review.openstack.org/448854 | 02:15 |
| *** sams-gleb has joined #openstack-swift | 02:16 | |
| mattoliverau | Although it isn't the same failure that timburke was seeing. infact none of them are the same.. they are all different errors. But will conintue running tests to see if I can hit another one. | 02:16 |
| mattoliverau | also, gotta love git bisect to find these patches to back port :) | 02:17 |
| *** dmorita has quit IRC | 02:17 | |
| *** sams-gleb has quit IRC | 02:20 | |
| m_kazuhiro | mattoliverau: Is the url link [1] in your commit message correct? The link shows me "File Not Found". | 02:21 |
| tone_z | me too | 02:23 |
| mattoliverau | m_kazuhiro: ahh, well it's a back port, so it's cschwede's old commit message, and we only keep logs for so long, so there probably gone now. | 02:23 |
| mattoliverau | Problem is, I hit this locally, so don't have a link, but maybe I'll update the commit message to point to the error in question. | 02:23 |
| mattoliverau | m_kazuhiro, tone_z: thanks both of you :) | 02:24 |
| tone_z | :-) | 02:25 |
| openstackgerrit | junboli proposed openstack/swift master: keep consistent naming convention of swift and urls https://review.openstack.org/447387 | 02:28 |
| openstackgerrit | junboli proposed openstack/python-swiftclient master: Be consistent in using example.com instead of identity.api.rackspacecloud.com https://review.openstack.org/448856 | 02:31 |
| mattoliverau | k, changed it to a paste.o.o link with the failed test error you receive | 02:32 |
| tone_z | It is OK now. ;) | 02:36 |
| m_kazuhiro | me too :) | 02:36 |
| mattoliverau | ta :) | 02:41 |
| *** dmorita has joined #openstack-swift | 02:42 | |
| *** dmorita has quit IRC | 02:43 | |
| *** dmorita_ has joined #openstack-swift | 02:43 | |
| *** dmorita_ has quit IRC | 02:44 | |
| *** dmorita has joined #openstack-swift | 02:44 | |
| *** dmorita_ has joined #openstack-swift | 02:46 | |
| *** dmorita has quit IRC | 02:49 | |
| *** dmorita_ has quit IRC | 02:50 | |
| *** dmorita has joined #openstack-swift | 02:55 | |
| *** dmorita_ has joined #openstack-swift | 02:56 | |
| *** dmorita_ has quit IRC | 02:57 | |
| *** dmorita_ has joined #openstack-swift | 02:57 | |
| *** dmorita has quit IRC | 02:59 | |
| *** dmorita_ has quit IRC | 03:02 | |
| *** dmorita has joined #openstack-swift | 03:14 | |
| *** sams-gleb has joined #openstack-swift | 03:18 | |
| *** dmorita has quit IRC | 03:18 | |
| *** dmorita has joined #openstack-swift | 03:20 | |
| *** dmorita has quit IRC | 03:22 | |
| *** sams-gleb has quit IRC | 03:22 | |
| *** dmorita has joined #openstack-swift | 03:22 | |
| *** zaitcev has quit IRC | 03:22 | |
| *** zaitcev has joined #openstack-swift | 03:23 | |
| *** ChanServ sets mode: +v zaitcev | 03:23 | |
| *** dmorita_ has joined #openstack-swift | 03:24 | |
| *** dmorita has quit IRC | 03:26 | |
| *** dmorita_ has quit IRC | 03:28 | |
| *** links has joined #openstack-swift | 03:37 | |
| *** dmorita has joined #openstack-swift | 03:39 | |
| *** dmorita has quit IRC | 03:44 | |
| *** m_kazuhiro has quit IRC | 03:45 | |
| *** zhurong has quit IRC | 03:48 | |
| *** dmorita has joined #openstack-swift | 04:05 | |
| *** dmorita_ has joined #openstack-swift | 04:07 | |
| *** dmorita has quit IRC | 04:07 | |
| *** dmorita has joined #openstack-swift | 04:08 | |
| *** dmorita__ has joined #openstack-swift | 04:10 | |
| *** dmorita_ has quit IRC | 04:11 | |
| *** dmorita has quit IRC | 04:13 | |
| *** dmorita__ has quit IRC | 04:15 | |
| *** chsc has joined #openstack-swift | 04:19 | |
| *** chsc has joined #openstack-swift | 04:19 | |
| *** chsc has quit IRC | 04:26 | |
| *** psachin has joined #openstack-swift | 04:28 | |
| kota_ | good afternoon | 04:31 |
| mattoliverau | kota_: o/ | 04:34 |
| mahatic | kota_: mattoliverau o/ morning | 04:36 |
| mattoliverau | mahatic: good morning | 04:36 |
| kota_ | mahatic: mattoliverau: o/ | 04:38 |
| kota_ | mattoliverau: thanks for your input for composite ring CLI etherpad, I was looking at it briefly in the last night | 04:38 |
| mattoliverau | no problem :) I feel both sides have pros and cons.. and I'd like to see it all apart of ringbuilder, but feel its better to move forward then get stuck at this juncture, but that might just be my own opinion :) | 04:40 |
| mattoliverau | but worried whatever we add to ringbuilder now, will have to stay for a while, so zipper makes sense in the short term | 04:41 |
| mattoliverau | until we can get it right | 04:41 |
| mattoliverau | anyway thats what my comments boiled down to | 04:41 |
| kota_ | mattoliverau: sounds good opinion, thanks. | 04:42 |
| *** dmorita has joined #openstack-swift | 04:54 | |
| *** Jeffrey4l_ has quit IRC | 04:55 | |
| *** Jeffrey4l has joined #openstack-swift | 04:56 | |
| *** gkadam has joined #openstack-swift | 04:57 | |
| *** dmorita_ has joined #openstack-swift | 04:58 | |
| *** dmorita has quit IRC | 04:59 | |
| *** dmorita_ has quit IRC | 05:00 | |
| *** dmorita has joined #openstack-swift | 05:00 | |
| *** dmorita_ has joined #openstack-swift | 05:01 | |
| *** dmorita has quit IRC | 05:01 | |
| *** dmorita_ has quit IRC | 05:02 | |
| *** dmorita has joined #openstack-swift | 05:03 | |
| *** dmorita_ has joined #openstack-swift | 05:04 | |
| *** dmorita__ has joined #openstack-swift | 05:05 | |
| *** dmorita_ has quit IRC | 05:05 | |
| *** dmorita has quit IRC | 05:07 | |
| *** dmorita__ has quit IRC | 05:10 | |
| *** dmorita has joined #openstack-swift | 05:12 | |
| *** dmorita_ has joined #openstack-swift | 05:16 | |
| *** dmorita has quit IRC | 05:17 | |
| openstackgerrit | Jeremy Zhang proposed openstack/swift master: Using assertIsNone() instead of assertEqual(None) https://review.openstack.org/448894 | 05:17 |
| *** JimCheung has joined #openstack-swift | 05:17 | |
| *** sams-gleb has joined #openstack-swift | 05:20 | |
| *** JimCheung has quit IRC | 05:21 | |
| *** adriant has quit IRC | 05:23 | |
| *** dmorita_ has quit IRC | 05:24 | |
| *** sams-gleb has quit IRC | 05:25 | |
| *** dmorita has joined #openstack-swift | 05:36 | |
| *** dmorita_ has joined #openstack-swift | 05:38 | |
| *** dmorita has quit IRC | 05:38 | |
| *** dmorita_ has quit IRC | 05:42 | |
| *** dmorita has joined #openstack-swift | 05:43 | |
| *** dmorita_ has joined #openstack-swift | 05:44 | |
| *** dmorita has quit IRC | 05:44 | |
| *** dmorita_ has quit IRC | 05:46 | |
| *** dmorita has joined #openstack-swift | 05:46 | |
| *** dmorita has quit IRC | 05:47 | |
| *** dmorita has joined #openstack-swift | 05:47 | |
| *** dmorita_ has joined #openstack-swift | 05:49 | |
| *** dmorita has quit IRC | 05:49 | |
| *** dmorita_ has quit IRC | 05:50 | |
| *** dmorita has joined #openstack-swift | 05:50 | |
| *** dmorita has quit IRC | 05:51 | |
| *** dmorita_ has joined #openstack-swift | 05:52 | |
| *** zhurong has joined #openstack-swift | 05:53 | |
| *** dmorita_ has quit IRC | 05:56 | |
| openstackgerrit | Jeremy Zhang proposed openstack/swift master: Using assertIsNone() instead of assertEqual(None) https://review.openstack.org/448894 | 06:04 |
| *** tonyb_ has joined #openstack-swift | 06:06 | |
| *** mlanner_ has joined #openstack-swift | 06:12 | |
| *** tonyb has quit IRC | 06:13 | |
| *** swifterdarrell has quit IRC | 06:13 | |
| *** mlanner has quit IRC | 06:13 | |
| *** mlanner_ is now known as mlanner | 06:13 | |
| *** swifterdarrell has joined #openstack-swift | 06:13 | |
| *** ChanServ sets mode: +v swifterdarrell | 06:13 | |
| openstackgerrit | Jeremy Zhang proposed openstack/swift master: Using assertIsNone() instead of assertEqual(None) https://review.openstack.org/448894 | 06:30 |
| *** jaosorior has joined #openstack-swift | 06:30 | |
| *** ChubYann has quit IRC | 06:32 | |
| openstackgerrit | Kazuhiro MIYAHARA proposed openstack/swift master: Another proposal of Automated Tiering's hidden account name rule https://review.openstack.org/448941 | 06:46 |
| openstackgerrit | Kazuhiro MIYAHARA proposed openstack/swift master: Another proposal of Automated Tiering's hidden account name rule https://review.openstack.org/448941 | 06:51 |
| *** dmellado has quit IRC | 07:03 | |
| *** pdardeau_ has quit IRC | 07:03 | |
| *** mmmucky has quit IRC | 07:04 | |
| *** pdardeau_ has joined #openstack-swift | 07:04 | |
| *** mmmucky has joined #openstack-swift | 07:05 | |
| *** dmellado has joined #openstack-swift | 07:08 | |
| *** bkopilov has joined #openstack-swift | 07:18 | |
| *** sams-gleb has joined #openstack-swift | 07:22 | |
| *** sams-gleb has quit IRC | 07:26 | |
| *** dmorita has joined #openstack-swift | 07:38 | |
| *** dmorita_ has joined #openstack-swift | 07:39 | |
| *** dmorita has quit IRC | 07:43 | |
| *** amoralej|off is now known as amoralej | 07:43 | |
| *** dmorita_ has quit IRC | 07:43 | |
| *** tesseract has joined #openstack-swift | 07:44 | |
| *** sams-gleb has joined #openstack-swift | 07:50 | |
| *** dmorita has joined #openstack-swift | 07:52 | |
| *** dmorita has quit IRC | 07:56 | |
| openstackgerrit | junboli proposed openstack/python-swiftclient master: remove py34 from envlist and its reference from classifier in setup.cfg https://review.openstack.org/448978 | 07:56 |
| *** pcaruana has joined #openstack-swift | 08:06 | |
| *** PavelK has joined #openstack-swift | 08:13 | |
| *** Administrator_ has quit IRC | 08:20 | |
| *** Administrator_ has joined #openstack-swift | 08:21 | |
| *** dmorita has joined #openstack-swift | 08:21 | |
| *** dmorita_ has joined #openstack-swift | 08:22 | |
| *** dmorita__ has joined #openstack-swift | 08:24 | |
| *** dmorita_ has quit IRC | 08:24 | |
| *** dmorita has quit IRC | 08:24 | |
| *** dmorita__ has quit IRC | 08:28 | |
| *** dmorita has joined #openstack-swift | 08:33 | |
| acoles | good morning | 08:34 |
| *** oshritf_ has joined #openstack-swift | 08:37 | |
| *** dmorita has quit IRC | 08:38 | |
| acoles | rledisez: did you find what you need re wsgicontext? | 08:38 |
| rledisez | acoles: yes, i think so. it looks like i need to fix the tests after using wsgicontext. i'll look into it today | 08:40 |
| acoles | rledisez: ah, that's the problem with tests :) let me know if you have any more questions | 08:43 |
| *** geaaru has joined #openstack-swift | 08:44 | |
| acoles | rledisez: BTW, the issue w.r.t. duplicate header names/nit putting headers into a dict - I should have said that swift does not produce any duplicate header names, but in the past I have heard of third party middlewares that do. In fact, that's how i learnt about this - one of my first swift patches did exactly this with response headers and broke someone's middleware :/ | 08:45 |
| rledisez | acoles: sure. no matter what, we should respect RFC when possible :) | 08:45 |
| *** cbartz has joined #openstack-swift | 08:53 | |
| *** jaosorior is now known as jaosorior_brb | 08:55 | |
| *** ma9_ has quit IRC | 08:56 | |
| *** openstackgerrit has quit IRC | 09:03 | |
| *** ma9_ has joined #openstack-swift | 09:04 | |
| *** openstackgerrit has joined #openstack-swift | 09:07 | |
| openstackgerrit | Jeremy Zhang proposed openstack/swift master: Using assertIsNone() instead of assertEqual(None) https://review.openstack.org/448894 | 09:07 |
| *** ma9_ has quit IRC | 09:07 | |
| acoles | mattoliverau: I just added sharding to the Boston topics etherpad so now you must come :) | 09:14 |
| *** ma9_ has joined #openstack-swift | 09:15 | |
| mattoliverau | acoles: well now that its written I have no choice ;) | 09:20 |
| acoles | book those flights! | 09:20 |
| *** jordanP has joined #openstack-swift | 09:21 | |
| *** gabor_antal has quit IRC | 09:23 | |
| *** gabor_antal has joined #openstack-swift | 09:24 | |
| *** ma9_ has quit IRC | 09:30 | |
| *** ma9_ has joined #openstack-swift | 09:33 | |
| *** links has quit IRC | 09:39 | |
| *** links has joined #openstack-swift | 09:56 | |
| acoles | alecuyer: rledisez: I left some comments on https://etherpad.openstack.org/p/swift-losf-base | 10:03 |
| openstackgerrit | Merged openstack/swift master: keep consistent naming convention of swift and urls https://review.openstack.org/447387 | 10:06 |
| openstackgerrit | Romain LE DISEZ proposed openstack/swift master: Rewrite redirection in cname_lookup & domain_remap https://review.openstack.org/435771 | 10:06 |
| *** geaaru has quit IRC | 10:11 | |
| *** links has quit IRC | 10:11 | |
| *** links has joined #openstack-swift | 10:23 | |
| *** geaaru has joined #openstack-swift | 10:23 | |
| openstackgerrit | Alistair Coles proposed openstack/swift master: Add more doc and test for cors_expose_headers option https://review.openstack.org/445021 | 10:25 |
| *** links has quit IRC | 10:35 | |
| *** links has joined #openstack-swift | 10:51 | |
| *** jaosorior_brb is now known as jaosorior | 11:02 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift master: Add Composite Ring Functionality https://review.openstack.org/441921 | 11:02 |
| *** jistr is now known as jistr|afk | 11:06 | |
| openstackgerrit | Romain LE DISEZ proposed openstack/swift master: Replace replication_one_per_device by custom count https://review.openstack.org/390781 | 11:06 |
| *** sams-gleb has quit IRC | 11:08 | |
| *** sams-gleb has joined #openstack-swift | 11:09 | |
| *** sams-gleb has quit IRC | 11:13 | |
| openstackgerrit | junboli proposed openstack/swift master: fixed get suffix of filename https://review.openstack.org/449077 | 11:17 |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift master: Eliminate node_index in Putter https://review.openstack.org/443072 | 11:27 |
| kota_ | just rebased | 11:29 |
| * kota_ is saying about patch 443072 | 11:29 | |
| *** mvk has quit IRC | 11:36 | |
| *** ntata has quit IRC | 11:36 | |
| *** sgundur- has quit IRC | 11:36 | |
| *** mmotiani has quit IRC | 11:36 | |
| *** sams-gleb has joined #openstack-swift | 11:52 | |
| *** mvk has joined #openstack-swift | 12:06 | |
| *** bkopilov has quit IRC | 12:12 | |
| *** zhurong has quit IRC | 12:22 | |
| *** NM has joined #openstack-swift | 12:24 | |
| *** jistr|afk is now known as jistr | 12:27 | |
| *** catintheroof has joined #openstack-swift | 12:44 | |
| *** klamath has joined #openstack-swift | 12:46 | |
| *** klamath has quit IRC | 12:47 | |
| *** klamath has joined #openstack-swift | 12:47 | |
| *** psachin has quit IRC | 12:52 | |
| kota_ | acoles: I updated some my patches, will look at your great improvement (e.g. around node election) tommorow, thanks! | 12:57 |
| kota_ | and it's time I should leave my office. | 12:57 |
| kota_ | bye | 12:57 |
| acoles | kota_: I saw - but you should be sleeping ! | 12:58 |
| kota_ | ;-) | 12:58 |
| acoles | thanks - btw the node iter patches are WIP - I have some more thoughts for improvements on those. but still appreciate any comments on the direction there. | 12:59 |
| kota_ | ok | 12:59 |
| acoles | kota_: I am looking at your updated composite builder again today | 12:59 |
| *** catintheroof has quit IRC | 13:03 | |
| kota_ | acoles: thx! (now on the way home) | 13:03 |
| *** gkadam has quit IRC | 13:04 | |
| tdasilva | kota_: wow, that's a late night for you! | 13:05 |
| * acoles was thinking the same | 13:06 | |
| *** amoralej is now known as amoralej|lunch | 13:07 | |
| *** kei_yama has quit IRC | 13:11 | |
| *** silor has joined #openstack-swift | 13:17 | |
| alecuyer | acoles: thanks a lot for your comments on the etherpad. I will submit a small patch soon | 13:27 |
| *** oshritf_ has quit IRC | 13:35 | |
| *** oshritf_ has joined #openstack-swift | 13:37 | |
| *** Italux has joined #openstack-swift | 14:05 | |
| *** Italux has left #openstack-swift | 14:05 | |
| *** vint_bra has joined #openstack-swift | 14:10 | |
| *** ChanServ sets mode: +v tdasilva | 14:12 | |
| *** dmorita has joined #openstack-swift | 14:18 | |
| *** dmorita_ has joined #openstack-swift | 14:21 | |
| *** dmorita_ has quit IRC | 14:21 | |
| *** dmorita_ has joined #openstack-swift | 14:22 | |
| *** dmorita has quit IRC | 14:22 | |
| *** amoralej|lunch is now known as amoralej | 14:22 | |
| acoles | alecuyer: hope it helps | 14:23 |
| *** dmorita has joined #openstack-swift | 14:26 | |
| *** dmorita_ has quit IRC | 14:28 | |
| *** oshritf_ has quit IRC | 14:29 | |
| *** links has quit IRC | 14:38 | |
| *** oshritf_ has joined #openstack-swift | 14:41 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift master: Updated from global requirements https://review.openstack.org/88736 | 14:42 |
| *** _JZ_ has joined #openstack-swift | 14:58 | |
| *** silor has quit IRC | 15:04 | |
| *** vinsh has joined #openstack-swift | 15:05 | |
| *** dmorita has quit IRC | 15:09 | |
| *** vinsh has quit IRC | 15:19 | |
| *** dmorita has joined #openstack-swift | 15:24 | |
| *** dmorita_ has joined #openstack-swift | 15:28 | |
| *** dmorita has quit IRC | 15:29 | |
| *** dmorita_ has quit IRC | 15:31 | |
| *** dmorita has joined #openstack-swift | 15:32 | |
| *** dmorita_ has joined #openstack-swift | 15:33 | |
| *** dmorita has quit IRC | 15:33 | |
| *** vinsh has joined #openstack-swift | 15:35 | |
| *** dmorita has joined #openstack-swift | 15:37 | |
| *** dmorita_ has quit IRC | 15:38 | |
| *** mmotiani has joined #openstack-swift | 15:38 | |
| *** vinsh has quit IRC | 15:40 | |
| notmyname | good morning | 15:40 |
| *** dmorita_ has joined #openstack-swift | 15:40 | |
| *** dmorita_ has quit IRC | 15:41 | |
| *** dmorita_ has joined #openstack-swift | 15:41 | |
| *** dmorita has quit IRC | 15:42 | |
| *** sgundur- has joined #openstack-swift | 15:44 | |
| *** dmorita has joined #openstack-swift | 15:45 | |
| *** dmorita_ has quit IRC | 15:46 | |
| *** dmorita has quit IRC | 15:46 | |
| *** dmorita has joined #openstack-swift | 15:46 | |
| *** ntata has joined #openstack-swift | 15:48 | |
| *** dmorita_ has joined #openstack-swift | 15:49 | |
| *** dmorita has quit IRC | 15:50 | |
| *** sams-gleb has quit IRC | 15:50 | |
| *** sams-gleb has joined #openstack-swift | 15:51 | |
| *** dmorita_ has quit IRC | 15:54 | |
| *** amoralej is now known as amoralej|brb | 15:54 | |
| *** sams-gleb has quit IRC | 15:55 | |
| *** dmorita has joined #openstack-swift | 15:57 | |
| *** dmorita_ has joined #openstack-swift | 15:58 | |
| *** dmorita_ has quit IRC | 15:59 | |
| *** dmorita_ has joined #openstack-swift | 15:59 | |
| *** dmorita_ has quit IRC | 16:00 | |
| *** dmorita_ has joined #openstack-swift | 16:00 | |
| *** dmorita has quit IRC | 16:01 | |
| *** dmorita_ has quit IRC | 16:05 | |
| *** silor has joined #openstack-swift | 16:05 | |
| *** chsc has joined #openstack-swift | 16:07 | |
| *** chsc has joined #openstack-swift | 16:07 | |
| *** bkopilov has joined #openstack-swift | 16:11 | |
| *** dmorita has joined #openstack-swift | 16:24 | |
| *** JimCheung has joined #openstack-swift | 16:25 | |
| *** sams-gleb has joined #openstack-swift | 16:25 | |
| *** dmorita has quit IRC | 16:27 | |
| *** dmorita has joined #openstack-swift | 16:27 | |
| *** SkyRocknRoll has joined #openstack-swift | 16:33 | |
| notmyname | IRC is less lonely when you don't filter the join/part messages | 16:36 |
| rledisez | IRC is never lonely, just need to trigger reactions | 16:39 |
| rledisez | good morning :) | 16:39 |
| *** amoralej|brb is now known as amoralej | 16:41 | |
| *** oshritf_ has quit IRC | 16:47 | |
| notmyname | but now that I killed patchbot, I can't inadvertently trigger nonsense reactions! | 16:47 |
| timburke | good morning | 16:53 |
| acoles | I miss patchbot | 16:54 |
| rledisez | i miss reading clayg complaining about patchbot ;) | 16:55 |
| timburke | :] | 16:55 |
| notmyname | timburke: ...and nothing happened | 16:56 |
| timburke | notmyname: what? i was just smiling :P | 16:56 |
| acoles | last ain't what it used to be | 16:57 |
| *** cbartz has left #openstack-swift | 16:58 | |
| *** dmorita has quit IRC | 17:01 | |
| *** dmorita has joined #openstack-swift | 17:02 | |
| *** catintheroof has joined #openstack-swift | 17:07 | |
| notmyname | please continue to add stuff to https://etherpad.openstack.org/p/BOS-Swift-brainstorming | 17:08 |
| openstackgerrit | Pavel Kvasnička proposed openstack/swift master: Container/account disk drive error results double space usage on rest drives https://review.openstack.org/448480 | 17:14 |
| *** pcaruana has quit IRC | 17:14 | |
| *** mvk has quit IRC | 17:17 | |
| *** gyee has joined #openstack-swift | 17:20 | |
| *** PavelK has quit IRC | 17:27 | |
| *** jaosorior has quit IRC | 17:30 | |
| *** tonanhngo has joined #openstack-swift | 17:30 | |
| *** klrmn has joined #openstack-swift | 17:31 | |
| *** gyee has quit IRC | 17:33 | |
| *** gyee has joined #openstack-swift | 17:34 | |
| *** gyee has quit IRC | 17:39 | |
| tdasilva | I've been looking at this file and can't quite figure out what's going on, maybe somebody can help me: https://github.com/openstack-infra/project-config/blob/master/jenkins/jobs/swift.yaml | 17:47 |
| tdasilva | that first job (line 2) is to run the swift functional tests | 17:47 |
| tdasilva | but i think we run them against swiftclient patches??? | 17:48 |
| tdasilva | am i reading this wrong? | 17:48 |
| acoles | tdasilva: there are two templates - one for swift then one for swiftclient | 17:48 |
| tdasilva | right | 17:49 |
| acoles | so one is used in swift gate job and the other in swiftclient gate jobs | 17:49 |
| acoles | AFAIK they are reference by the name attr | 17:50 |
| timburke | tdasilva: yeah, i remember noticing us running swift dsvm for swiftclient too :-/ but iirc swift *does* import swiftclient for func tests, so maybe there was a reason for it? | 17:50 |
| tdasilva | acoles, timburke: right, so that's what I noticed | 17:51 |
| timburke | https://github.com/openstack-infra/project-config/blob/master/zuul/layout.yaml#L16625-L16636 | 17:51 |
| tdasilva | the swift func tests are run both on swift and swiftclient patches | 17:51 |
| tdasilva | and the swiftclient tests are run only against the swiftclient patches | 17:51 |
| *** links has joined #openstack-swift | 17:52 | |
| tdasilva | what does dsvm stand for? | 17:52 |
| *** mvk has joined #openstack-swift | 17:53 | |
| timburke | devstack vm as i recall | 17:53 |
| acoles | do we do that so that swiftclient change can't cause swift func tests to break? | 17:54 |
| timburke | i think the swift tests are intended to ensure so we don't break get_auth for https://github.com/openstack/swift/blob/2.13.0/test/functional/swift_test_client.py#L29 (but i could certainly be wrong) | 17:54 |
| acoles | i.e. run the swift dsvm job against swiftclient patches | 17:54 |
| timburke | i'm not sure the dsvm job actually *covers* that line, and even if it did, it seems like overkill; something more like gate-swift-tox-xfs-tmp-func-ubuntu-xenial should be perfectly sufficient and run in 1/6 the time | 17:56 |
| tdasilva | oh, i see, so even thou it is running the swift func tests, that swiftclient patch is what gets installed on the node for the test | 17:56 |
| timburke | maybe *that's* why we went with dsvm -- to ensure the in-review version gets installed... | 17:56 |
| tdasilva | maybe | 17:57 |
| tdasilva | hehe | 17:57 |
| tdasilva | timburke: which actually gives me another piece to the puzzle | 17:57 |
| tdasilva | the reason i'm asking all this is because i'd like to have a job that tests libec and pyeclib together | 17:58 |
| acoles | tdasilva: I think that's the goal - check swift func tests pass with the patched version of swiftclient | 17:58 |
| *** links has quit IRC | 17:58 | |
| tdasilva | acoles: right | 17:59 |
| timburke | i'd be all for paring down the services that get brought up as part of that, though. we've already got gate-tempest-dsvm-neutron-src-python-swiftclient-ubuntu-xenial for cross-project coverage, right? seems unnecessary to bring up all of devstack when all we really want is swift (and maybe keystone) to ensure we don't break swift func tests... | 18:00 |
| timburke | tdasilva: yeah... i think there's some devstack-y goodness to make everything work right with Depends-On and all of that... not sure where the magic is, though | 18:01 |
| tdasilva | timburke: ack, will keep playing with it | 18:04 |
| tdasilva | unfortunately i can't find much documentation so have to rely on trial-and-error | 18:04 |
| timburke | tdasilva: yeah, i've always had that feeling when looking at devstack... it's a little hard to trace where things happen and what affects a change will have... | 18:06 |
| *** catintheroof has quit IRC | 18:07 | |
| *** ujjain has quit IRC | 18:13 | |
| acoles | tdasilva: timburke devstack also helpfully installs (a subset of) the swift test projects/users into keystone so func tests can run | 18:13 |
| tdasilva | acoles: oh nice, so this is probably the reason for this particular job: | 18:15 |
| tdasilva | https://github.com/openstack-infra/project-config/blob/master/jenkins/jobs/swift.yaml#L45 | 18:15 |
| acoles | yeah and so that swiftclient func tests use a similar set of test accounts as swift tests, see https://github.com/openstack/python-swiftclient/blob/2a62dd88ff48c2e281679c470d47dc72c82707b6/tests/sample.conf | 18:18 |
| tdasilva | acoles: but what about this line here: | 18:20 |
| tdasilva | https://github.com/openstack-infra/project-config/blob/master/jenkins/jobs/swift.yaml#L78 | 18:20 |
| tdasilva | it's actually referring to the sample.conf in the swift dir | 18:20 |
| tdasilva | not the swiftclient dir | 18:20 |
| tdasilva | plus, is that sample.conf modified when it gets installed in devstack? because the default is v1 | 18:21 |
| tdasilva | not v2 | 18:21 |
| *** jordanP has quit IRC | 18:21 | |
| acoles | yes, so here https://github.com/openstack/python-swiftclient/blob/2a62dd88ff48c2e281679c470d47dc72c82707b6/tests/sample.conf you see the func tests are actually run twice, once using a test file written by devstack IIRC (/etc/swift/test.conf) and once using the sample file in tree. | 18:26 |
| acoles | tdasilva: for example here you'll see one run has a bunch of skips, most likely for v2 only tests when run against v1. http://logs.openstack.org/04/443104/4/check/gate-swiftclient-dsvm-functional-ubuntu-xenial/780af79/console.html | 18:27 |
| acoles | hmmm, no line by line test results in the logs | 18:27 |
| acoles | same happen with the swift func tests | 18:28 |
| tdasilva | acoles: got it | 18:32 |
| openstackgerrit | Tim Burke proposed openstack/swift master: Drop support for auth-server from common/manager https://review.openstack.org/449270 | 18:37 |
| *** patchbot has joined #openstack-swift | 18:37 | |
| *** patchbot has quit IRC | 18:38 | |
| notmyname | tdasilva: are you at vault this week? | 18:47 |
| *** geaaru has quit IRC | 18:49 | |
| *** tesseract has quit IRC | 19:28 | |
| notmyname | has anyone gotten slqite to work on osx? | 19:36 |
| notmyname | *sqlite | 19:36 |
| *** amoralej is now known as amoralej|off | 19:46 | |
| *** ChubYann has joined #openstack-swift | 19:48 | |
| clayg | notmyname: since when does sqlite not work on osx!? | 19:51 |
| notmyname | clayg: no idea! | 19:51 |
| notmyname | https://gist.github.com/notmyname/dff0c20eee60ef1a1190e0c02be3f368 | 19:52 |
| *** SkyRocknRoll has quit IRC | 19:52 | |
| clayg | notmyname: so i installed python from brew I think? https://gist.github.com/clayg/7e5c64c3d99d88b9fadeadcf5cdeca43 | 19:55 |
| notmyname | clayg: what versuon of osx do you have? | 19:55 |
| notmyname | hmm, I've got a 16.4.0 kernel | 19:56 |
| notmyname | it's not too important. just inconvenient | 19:56 |
| *** NM has quit IRC | 19:56 | |
| clayg | oh yeah *thats* probably it - Version 10.12.2 - Sierra? | 19:57 |
| clayg | the updates store w/e says I'm supposed o restart and get 10.12.3 - no idea how long I've been putting that off | 19:57 |
| notmyname | hmm | 19:58 |
| notmyname | I'm on 10.12.3 | 19:58 |
| notmyname | oh well. like I said, not critical, just inconvenient | 19:58 |
| clayg | well I have no idea how to make mac's built in python do *anything* useful | 19:58 |
| clayg | so I install python from brew and go from there - AFAIK that's just "how it's done" | 19:58 |
| notmyname | surely you mean "ssh to a vm or remote box running linux" right? | 19:59 |
| clayg | you're the one trying to make python do something useful on your mac | 19:59 |
| notmyname | lol | 20:00 |
| notmyname | and my solution now is to add an scp step to copy my python file to a remote vm running linux! | 20:00 |
| clayg | ok, well i'm repeating myself a third time - but your other option is to install python with brew and use that primarlly for dev/python stuff | 20:01 |
| clayg | but your other option works well too! so don't listen to me! | 20:01 |
| notmyname | actually this is all your fault! | 20:01 |
| notmyname | I killed patchbot (your fault), and now i'm looking at sopel (again, your fault), and sopel uses sqlite3 (that part isn't your fault) | 20:02 |
| notmyname | ;-) | 20:02 |
| notmyname | WINKY FACE! | 20:02 |
| timburke | notmyname: `pip install sqlite3`?? i thought that was stdlib! https://docs.python.org/2/library/sqlite3.html | 20:02 |
| notmyname | OUTRAGE! INJUSTICE! (and my kids' favorite new word) PREPOSTEROUS! | 20:03 |
| timburke | notmyname: literally *every* version of python i have is happy to import it: http://paste.openstack.org/show/603979/ | 20:05 |
| *** NM has joined #openstack-swift | 20:06 | |
| tdasilva | notmyname: I didn't go to vault this year | 20:12 |
| notmyname | tdasilva: is this like you not going to the boston summit? isn't it *right there*? | 20:13 |
| tdasilva | notmyname: heh, at least the summit I get a free pass | 20:13 |
| notmyname | :-) | 20:13 |
| tdasilva | vault was like $700 or something | 20:13 |
| notmyname | ouch | 20:13 |
| tdasilva | plus hotel and all that | 20:14 |
| tdasilva | it's a bit of a hike for me to drive in... | 20:14 |
| notmyname | you don't get to expense a week of mortgage for a conference ;-) | 20:14 |
| tdasilva | probably more than 2hrs in the morning | 20:14 |
| notmyname | (although that would be awesome for local events) | 20:14 |
| timburke | between https://review.openstack.org/#/c/232162/53/swift/common/middleware/symlink.py@141 and https://review.openstack.org/#/c/423906/7/swift/common/request_helpers.py@717 and https://github.com/openstack/swift/blob/2.13.0/swift/common/middleware/crypto/decrypter.py#L350 ... what do people think about moving a/c listing response-type mangling to proxy middleware, and having the proxy *always* do `?format=json` during requests? and pro | 20:15 |
| timburke | bably removing xml support from backend servers in the process | 20:15 |
| notmyname | timburke: in general, I'm a fan of the storage nodes only doing one kind of response and the proxy mutating it for the client | 20:16 |
| timburke | notmyname: yeah, i'd kinda like to get rid of txt responses, too, but that'd probably break some internal clients. xml seems safe to get rid of, though? | 20:17 |
| notmyname | timburke: from the storage node, yeah. and we should look into the internal client usage. | 20:18 |
| jrichli | timburke: so that means that there must be a required mw that will make the necessary transformation. Both symlinks and encryption are optional. where would the required transformation live? | 20:19 |
| jrichli | or would it take place in the proxy controller or somewhere else in the proxy? | 20:20 |
| timburke | jrichli: i was thinking far left; just right of cache or that first proxy-logging | 20:20 |
| jrichli | ok, so you are thinking a new required mw? | 20:21 |
| notmyname | timburke: in the proxy controller itself might not be a bad idea | 20:21 |
| notmyname | instead of middleware | 20:21 |
| timburke | with the container controller still forcing a json listing | 20:21 |
| timburke | notmyname: proxy controller doesn't help the encryption & symlink cases, though. i want middleware to only need to think about one format | 20:21 |
| *** ujjain has joined #openstack-swift | 20:22 | |
| *** ujjain has joined #openstack-swift | 20:22 | |
| jrichli | oic. so one format to worry about in the response. but if you have both symlinks and encryption, wouldn't symlinks see the already processed response (xml)? so would still need to worry about all options? | 20:24 |
| tdasilva | timburke: and would that middleware know about symlink and encryption changes? | 20:25 |
| jrichli | i am a little confused how we things will work if we don't only have one spot to transform from the one obj-server format. | 20:26 |
| timburke | tdasilva: not if we do it right :-) | 20:26 |
| jrichli | oh - so like a method they all add to? | 20:26 |
| tdasilva | timburke: so each middleware would still need to iterate over the list? | 20:26 |
| timburke | jrichli: with both symlinks and encryption, there'd be some extra json deserialization/serialization (so each could do their processing), but punt to the new mw for xml formatting | 20:26 |
| timburke | tdasilva: ya | 20:27 |
| jrichli | oh, right. i see. | 20:27 |
| jrichli | the new mw is the one spot | 20:27 |
| openstackgerrit | Alistair Coles proposed openstack/swift master: WIP use composite ring metadata to prevent bad things https://review.openstack.org/449310 | 20:28 |
| tdasilva | timburke: wouldn't we gain by having just one place doing any changes necessary to the listing? I guess that would not be good for third-party middleware | 20:28 |
| jrichli | makes sense. a middleware to handle listing formats. | 20:28 |
| tdasilva | but just trying to save on multiple mws traversing through the same list multiple times | 20:28 |
| jrichli | true: third-party middleware could be a concern. other than that, i think it makes a lot of sense to encapsulate listing format into a mw. | 20:29 |
| timburke | tdasilva: either we have each middleware handle its own list iteration, or we create some new, non-WSGI callback pipeline. the former seems preferable to having a new copy_hook look-alike? | 20:32 |
| acoles | kota_: I wrote some *early* ideas for composite ring metadata in patch 449310 | 20:34 |
| acoles | good night | 20:34 |
| timburke | g'night acoles! | 20:34 |
| tdasilva | right, that's sort of i was thinking. e.g., encryption provides the callback function to decrypt data, but that gets called from this new mw | 20:34 |
| tdasilva | acoles: good night! | 20:34 |
| acoles | cheers guys, have a good evening | 20:34 |
| jrichli | good night acoles | 20:35 |
| timburke | jrichli: agreed, third-party middleware may be a problem. but we've broken them before for less-obvious wins. and i *think* most middlewares that do their own container listings would typically go for json anyway? certainly versioned_writes and dlo do | 20:36 |
| jrichli | tdasilva: I guess that would be assuming that processing is a line-by-line independent thing. probably true now, but it technically adds that restriction. unless it was still ok for a mw to do its own loop | 20:38 |
| *** silor1 has joined #openstack-swift | 20:44 | |
| *** silor has quit IRC | 20:46 | |
| *** silor1 has quit IRC | 20:49 | |
| *** clarkb has joined #openstack-swift | 21:00 | |
| *** JimCheung has quit IRC | 21:01 | |
| clarkb | Hello. I have been looking at memory consumption of various portions of our integration testing ebcause OOMs have become a problem in some gate jobs. One thing I noticed was that our python35 integration job (which doesn't run swift) uses significantly less memory than the job which does use swift | 21:01 |
| clarkb | digging into that more it appears that we have workers + 1 processes for most of the swift processes (proxy, container server, object server, etc) | 21:01 |
| clarkb | for the container server and object server this is 3 processes each as they have 2 workers. My question is can that be dialed back to 1 worker safely without losing test coverage or degrading the service? | 21:02 |
| clarkb | and potentially is there a way to drop the extra fork and just have N processes instead of N+1? | 21:02 |
| clarkb | notmyname: clayg ^ | 21:02 |
| notmyname | clarkb: most likely, yes | 21:02 |
| notmyname | clarkb: the "workers" setting in the config files | 21:03 |
| clarkb | notmyname: ya I see that, so it should be fine to set that down to 1 for container server and object server? | 21:04 |
| notmyname | yeah | 21:04 |
| clarkb | oh and account server. Cool I can get that pushed up then. Thanks | 21:04 |
| clarkb | as for the N+1 situation I'm guessing thats beacuse the parent is the coordinator of a bunch of workers? so you can't just get rid of it? | 21:04 |
| *** JimCheung has joined #openstack-swift | 21:05 | |
| notmyname | we might could. but TBH it's not what's used in any prod environment, so I haven't spend much (any?) time thinking about that | 21:05 |
| clarkb | notmyname: prod would use the N+1 regardless? or prod uses a completely different process management setup? | 21:06 |
| notmyname | prod would use N+1. I don't know of any prod setups with only one core per server. we want at least one worker per core | 21:07 |
| clarkb | gotcha | 21:07 |
| clayg | zaitcev: another thing to think about if get get to make a hard cutover to a v2 internal storage api | 21:14 |
| clayg | we should design them so that a/c/o apis can all run on the same port/worker pool | 21:14 |
| clayg | be useful for the swift-in-the tiny use-case - and not terrible/unreasonable for a sort of thought out api | 21:15 |
| *** NM has quit IRC | 21:15 | |
| zaitcev | use /av2 /cv2 /ov1 or a header? | 21:15 |
| clayg | ^ cc redbo /<v>/dev/<service>/part/ | 21:15 |
| clayg | NOT header ;) | 21:15 |
| zaitcev | heh | 21:15 |
| clayg | i'm sick of headers that dramatically change the nature of the api - stupid x-backend-storage-policy-index | 21:16 |
| clayg | /<v>/dev/<service>[-spi]/part/ maybe | 21:16 |
| *** NM has joined #openstack-swift | 21:18 | |
| jrichli | Question on SAIO at https://docs.openstack.org/developer/swift/development_saio.html#common-post-device-setup | 21:23 |
| jrichli | The /etc/rc.local file is modified, but nothing is explicitly run to invoke that file. I know a reboot runs it, but what in the instructions are supposed to cause that file to be run? is the next call to 'service' supposed to run it? | 21:23 |
| jrichli | This is related to patch https://review.openstack.org/443162 | 21:24 |
| zaitcev | There have to be limits of what we can teach people about basics of Linux. | 21:25 |
| zaitcev | It's like these days they teach students to write in complete sentences in colleges, because they enjoyed self-esteem improvements in K-12 systems instead of studying. | 21:26 |
| notmyname | maybe it's annoying to add a "source rc.local" comment because "kids these days...", but there's always new people learning stuff (and sometimes even older people learning stuff) | 21:28 |
| clarkb | speaking of learning things, on the N+1 with N == numcpu, do you not tend to oversaturate IO that way? | 21:29 |
| clarkb | seems like you'd run out of pci lanes before you did cpu cycles in that setup | 21:29 |
| notmyname | clarkb: normally there's a lot more drives than cores, and disk IO being what it is, a thread doing stuff on the disk is blocked. so having more workers allows us to handle more requests while the other requests are busy spinning platters | 21:32 |
| notmyname | clarkb: and because python and eventlet, we need to fork child processes (hello, golang!) | 21:32 |
| notmyname | clarkb: you are hitting a very important port. or raising a very important question. "how does swift schedule IO to make most efficient use of all of the available disk iops" | 21:34 |
| clayg | jrichli: it's a sysV thing - I'm sorta glad you brought it up because I've never really considered how systemd and rc.local get along | 21:34 |
| clarkb | notmyname: ya I recall the email explaining how golang was useful around this | 21:35 |
| notmyname | clarkb: and the answer is "meh. but we want to get better and have some reasonable ideas on how to get there." clayg's working on some great stuff about scaling the background daemons to do better in relation to this | 21:35 |
| clarkb | clayg: occasionally they don't get along and then things break in weird places (like puppet) | 21:36 |
| clarkb | clayg: but usually you jsut drop fiels in the usual places and reload the systemd config and can start and stop services like any other systemd unit | 21:36 |
| notmyname | jrichli: I think you mean ^ | 21:36 |
| clayg | clarkb: yeah I think the answer to my question was "systemd tries to be backwards compatible with crufty bits like that - you should test it!" | 21:36 |
| notmyname | bah! sorry. | 21:37 |
| notmyname | both clarkb and clayg are the same color in my irc client (and very similar) | 21:37 |
| clarkb | :) | 21:38 |
| notmyname | clarkb: there's a lot of facets to the solution for better iops. much of it is language agnostic (determining the work to be done and scheduling it effectively) and some isn't (what's in the data path for moving bits from nic to disk and back) | 21:38 |
| notmyname | http://d.not.mn/very_confusing_colors.png | 21:39 |
| *** patchbot has joined #openstack-swift | 21:46 | |
| notmyname | patch 1234 | 21:46 |
| patchbot | https://review.openstack.org/#/c/1234/ - tempest - Removing link doctoring in test_images (MERGED) | 21:46 |
| notmyname | ok, a *limited* version of patchbot is back | 21:46 |
| jrichli | yay! welcome back patchbot | 21:47 |
| *** NM has quit IRC | 21:47 | |
| clayg | notmyname: thanks bro - sorry to be an ass - but you're awesome | 21:47 |
| jrichli | limited how? can I do all the poking now to see what happens? | 21:47 |
| clayg | patchbot: do you hate me? | 21:47 |
| notmyname | limitations include (1) not being drunk (2) not responding to patch URLS (3) only responding to one patch number per line | 21:47 |
| clayg | notmyname: timburke will fix that eventually - you rock | 21:48 |
| jrichli | i can live with that | 21:48 |
| notmyname | lol | 21:48 |
| timburke | last time patchbot was here, he seemed to hate clayg... | 21:48 |
| timburke | yay! :] | 21:48 |
| clayg | timburke: mutual | 21:48 |
| clayg | this new cat sems cool so far | 21:48 |
| notmyname | I want the multi-response-per-line thing | 21:48 |
| clayg | yeah that was not pretty not terrible :\ | 21:48 |
| notmyname | patch 1234 patch 5678 | 21:49 |
| patchbot | https://review.openstack.org/#/c/1234/ - tempest - Removing link doctoring in test_images (MERGED) | 21:49 |
| timburke | notmyname: still gonna live at https://github.com/notmyname/Patches/ ? | 21:49 |
| notmyname | p 6543 | 21:49 |
| patchbot | https://review.openstack.org/#/c/6543/ - keystone - Just for testing (ABANDONED) | 21:49 |
| notmyname | timburke: dunno | 21:49 |
| notmyname | I gotta say that I *love* sopel relative to supybot though | 21:50 |
| notmyname | timburke: I'll get it up somewhere and let you know | 21:50 |
| timburke | notmyname: that's the part i really care about :-) | 21:50 |
| timburke | and do links like https://review.openstack.org/#/c/446185/ still work? | 21:51 |
| timburke | :-( | 21:52 |
| mattoliverau | Morning | 21:52 |
| notmyname | timburke: that's the other thing I want to get working. that one should be close to trivial | 21:52 |
| jrichli | timburke: that was limitation (2) ;-) | 21:52 |
| timburke | jrichli: fah! reading! | 21:53 |
| jrichli | lol | 21:53 |
| notmyname | I'm thinking of adding a "if time.now() == 3am: bot.tell(clayg, "I'm still watching you...") | 21:53 |
| jrichli | mismatched quotes | 21:54 |
| jrichli | :-) | 21:54 |
| clayg | lol @ jrichli | 21:54 |
| clayg | notmyname: no funny | 21:54 |
| jrichli | I know timburke is dying to test some more | 21:55 |
| timburke | notmyname: you've gotta get some randomness in there. like, spin up a separate thread to sleep(random.randint(2, 8) * 3600 + random.randint(0, 3599)) then pick one of at least a dozen messages to tell | 21:58 |
| *** dmorita has quit IRC | 21:58 | |
| notmyname | of course. it was just pseudocode ;-) | 21:58 |
| timburke | bonus points for doing it in some random channel clayg is known to frequent rather than a pm | 21:58 |
| clayg | :D | 21:59 |
| clayg | I have the best friends | 21:59 |
| * clayg 's heart grows three sizes | 22:00 | |
| *** patchbot has quit IRC | 22:02 | |
| timburke | :-( | 22:02 |
| timburke | patchbot, i miss you already! | 22:02 |
| notmyname | ...testing... | 22:03 |
| zaitcev | No, better use an unbalanced ' | 22:09 |
| zaitcev | My height is 6'5" | 22:10 |
| *** sams-gleb has quit IRC | 22:42 | |
| *** patchbot has joined #openstack-swift | 22:43 | |
| notmyname | zaitcev: how tall are you? | 22:43 |
| notmyname | p 1234 p 5678 | 22:43 |
| patchbot | https://review.openstack.org/#/c/1234/ - tempest - Removing link doctoring in test_images (MERGED) | 22:43 |
| patchbot | https://review.openstack.org/#/c/5678/ - nova - Check that DescribeInstance works with deleted image (MERGED) | 22:43 |
| notmyname | https://review.openstack.org/#/c/5678/ https://review.openstack.org/#/c/56799/ | 22:43 |
| patchbot | patch 5678 - nova - Check that DescribeInstance works with deleted image (MERGED) | 22:43 |
| patchbot | patch 56799 - nova - Compact pre-Icehouse database migrations <= 170. (MERGED) | 22:43 |
| zaitcev | 6'5" | 22:43 |
| notmyname | no errors from patchbot! | 22:43 |
| zaitcev | hmm. Needs a command, I suspect. | 22:43 |
| timburke | wooo! | 22:43 |
| *** sams-gleb has joined #openstack-swift | 22:43 | |
| notmyname | patch 1234 https://review.openstack.org/#/c/5678/ | 22:44 |
| patchbot | patch 1234 - tempest - Removing link doctoring in test_images (MERGED) | 22:44 |
| notmyname | mixing the two forms doesn't work | 22:44 |
| zaitcev | Wearing a nicotine patch at 6'5" | 22:44 |
| *** sams-gleb has quit IRC | 22:44 | |
| zaitcev | (which I'm not, but) | 22:44 |
| timburke | p 1234 patch 4567 | 22:44 |
| patchbot | https://review.openstack.org/#/c/1234/ - tempest - Removing link doctoring in test_images (MERGED) | 22:44 |
| patchbot | https://review.openstack.org/#/c/4567/ - python-keystoneclient - Fix ec2-credentials-list to print 'access'. (ABANDONED) | 22:44 |
| timburke | ppp 1234 | 22:44 |
| timburke | good | 22:45 |
| notmyname | timburke: https://github.com/notmyname/Patches2 | 22:47 |
| notmyname | timburke: and you've got commit access there | 22:48 |
| timburke | p qwer | 22:51 |
| *** tonanhngo has quit IRC | 22:51 | |
| timburke | p -1 | 22:52 |
| timburke | p 0 | 22:53 |
| patchbot | AttributeError: 'NoneType' object has no attribute 'get' (file "/home/ubuntu/.sopel/modules/patchbot.py", line 39, in get_response) | 22:53 |
| mattoliverau | timburke: you did it | 22:53 |
| timburke | yay! | 22:53 |
| timburke | notmyname: ...is that a... p-0 bug? | 22:54 |
| timburke | :P | 22:54 |
| *** ujjain has quit IRC | 22:57 | |
| *** tonanhngo has joined #openstack-swift | 22:58 | |
| timburke | tdasilva, jrichli (and i suppose mattoliverau may be interested): fwiw, functests pass with something like https://gist.github.com/tipabu/513dfeabe471e652d17474424328bc74 in the proxy pipeline... | 22:59 |
| *** ujjain has joined #openstack-swift | 23:01 | |
| *** ujjain has joined #openstack-swift | 23:01 | |
| *** tonanhngo has quit IRC | 23:02 | |
| *** catintheroof has joined #openstack-swift | 23:13 | |
| notmyname | timburke: bah | 23:16 |
| *** catintheroof has quit IRC | 23:20 | |
| *** patchbot has quit IRC | 23:20 | |
| *** patchbot has joined #openstack-swift | 23:20 | |
| notmyname | timburke: patch 0 | 23:21 |
| *** catintheroof has joined #openstack-swift | 23:21 | |
| *** klamath has quit IRC | 23:23 | |
| clayg | timburke: that was *after* you got to look at the source code - all bugs are shallow | 23:25 |
| clayg | timburke: wtf!? https://gist.github.com/tipabu/513dfeabe471e652d17474424328bc74 is *awesome* - have you had that laying around?! | 23:26 |
| clayg | I mean, don't get me wrong - i didn't understand wtf it was doing for like the first 5 mins - but ... well whatever docstrings - psffttt | 23:27 |
| timburke | clayg: ...i just wrote it. like, today. no tests (beyond swift's func tests) and no immediate plans to write them (because otherwise it's just time taken away from getting sharding loaded into my head) | 23:28 |
| timburke | but it seemed like we were hitting issues with this sort of thing a decent bit (crypto, symlinks, sharding...) so i figured i'd take a stab at it. if mattoliverau thinks it'll make his life easier with sharding, i could write some tests and we could get a chain going... | 23:29 |
| clayg | *gotcha* - nice work | 23:30 |
| *** kei_yama has joined #openstack-swift | 23:31 | |
| mattoliverau | timburke: so yeah, taking the listing away from the account and container server (and proxy in the case of sharding) is kinda cool. So what all we'll really care about is getting stuff in json then :) | 23:33 |
| *** catintheroof has quit IRC | 23:33 | |
| mattoliverau | only downside I see, which isn't much of one. If that json listings means more data then the default text listings. However, I assume json listing are very common anyway, and sharding will always use em. | 23:34 |
| mattoliverau | that is assuming if we have a listing format middleware, we could remove the code from the container/account server. If you want to get non json from them, then just add the middleware to thier pipline too. | 23:37 |
| *** catintheroof has joined #openstack-swift | 23:38 | |
| mattoliverau | \o/ after 272 runs I finally have a second stable/newton swift failure (first one is already up as a backport). | 23:41 |
| notmyname | mattoliverau: nice! | 23:41 |
| *** sams-gleb has joined #openstack-swift | 23:44 | |
| *** catintheroof has quit IRC | 23:45 | |
| *** sams-gleb has quit IRC | 23:49 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
