| *** avarner has quit IRC | 00:00 | |
| *** jerrygb has quit IRC | 00:00 | |
| *** jerrygb has joined #openstack-glance | 00:02 | |
| *** sgotliv_ has quit IRC | 00:02 | |
| *** jamielennox|away is now known as jamielennox | 00:05 | |
| *** takedakn has quit IRC | 00:07 | |
| *** takedakn has joined #openstack-glance | 00:07 | |
| *** jerrygb has quit IRC | 00:18 | |
| *** jerrygb has joined #openstack-glance | 00:18 | |
| *** ianbrown has quit IRC | 00:20 | |
| *** ianbrown has joined #openstack-glance | 00:20 | |
| *** smatzek has quit IRC | 00:20 | |
| *** jerrygb has quit IRC | 00:23 | |
| *** ianbrown has quit IRC | 00:25 | |
| *** ianbrown_ has joined #openstack-glance | 00:25 | |
| *** mtanino has quit IRC | 00:33 | |
| *** ducttape_ has quit IRC | 00:40 | |
| *** ducttape_ has joined #openstack-glance | 00:40 | |
| *** takedakn has quit IRC | 00:43 | |
| *** alejandrito has quit IRC | 00:46 | |
| *** RichardRaseley has quit IRC | 00:56 | |
| *** changbl has quit IRC | 01:00 | |
| *** changbl has joined #openstack-glance | 01:13 | |
| *** changbl has quit IRC | 01:14 | |
| *** ianbrown_ has quit IRC | 01:20 | |
| *** nikhil_k has joined #openstack-glance | 01:23 | |
| *** nikhil has quit IRC | 01:25 | |
| *** ducttape_ has quit IRC | 01:25 | |
| *** links has joined #openstack-glance | 01:27 | |
| *** ducttape_ has joined #openstack-glance | 01:33 | |
| *** changbl has joined #openstack-glance | 01:34 | |
| *** wxy has quit IRC | 01:37 | |
| *** wxy has joined #openstack-glance | 01:38 | |
| *** yasu has joined #openstack-glance | 01:39 | |
| *** ducttape_ has quit IRC | 01:42 | |
| *** r-daneel has quit IRC | 01:43 | |
| *** changbl has quit IRC | 01:47 | |
| *** ianbrown_ has joined #openstack-glance | 01:48 | |
| *** 7F1AA32U1 has joined #openstack-glance | 01:58 | |
| *** changbl has joined #openstack-glance | 02:00 | |
| *** 7F1AA32U1 has quit IRC | 02:01 | |
| *** haomaiwang has joined #openstack-glance | 02:01 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/glance: Updated from global requirements https://review.openstack.org/254469 | 02:05 |
|---|---|---|
| *** jerrygb has joined #openstack-glance | 02:14 | |
| *** jerrygb has quit IRC | 02:16 | |
| *** itisha has quit IRC | 02:26 | |
| *** ayoung has quit IRC | 02:29 | |
| *** ducttape_ has joined #openstack-glance | 02:41 | |
| *** houming has joined #openstack-glance | 02:42 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/glance: Updated from global requirements https://review.openstack.org/254469 | 02:50 |
| *** ianbrown_ has quit IRC | 02:57 | |
| *** ducttape_ has quit IRC | 03:00 | |
| *** haomaiwang has quit IRC | 03:01 | |
| *** azbiswas has quit IRC | 03:01 | |
| *** haomaiwang has joined #openstack-glance | 03:01 | |
| *** azbiswas has joined #openstack-glance | 03:01 | |
| openstackgerrit | wangxiyuan proposed openstack/glance: Update the cache middleware flavor guideline. https://review.openstack.org/254517 | 03:02 |
| *** azbiswas has quit IRC | 03:06 | |
| *** edmondsw has quit IRC | 03:07 | |
| *** haomaiwang has quit IRC | 03:15 | |
| *** haomaiwang has joined #openstack-glance | 03:15 | |
| *** haomaiwang has quit IRC | 03:16 | |
| *** haomaiwang has joined #openstack-glance | 03:16 | |
| *** haomaiwang has quit IRC | 03:17 | |
| *** ayoung has joined #openstack-glance | 03:18 | |
| *** ayoung has quit IRC | 03:18 | |
| *** haomaiwang has joined #openstack-glance | 03:18 | |
| *** haomaiwang has quit IRC | 03:19 | |
| *** 17WAA5E9R has joined #openstack-glance | 03:19 | |
| *** 17WAA5E9R has quit IRC | 03:20 | |
| *** haomaiwang has joined #openstack-glance | 03:20 | |
| *** haomaiwang has quit IRC | 03:21 | |
| *** haomaiwang has joined #openstack-glance | 03:24 | |
| *** haomaiwang has quit IRC | 03:25 | |
| *** julim has quit IRC | 03:25 | |
| *** haomaiwang has joined #openstack-glance | 03:25 | |
| *** haomaiwang has quit IRC | 03:26 | |
| *** haomaiwang has joined #openstack-glance | 03:26 | |
| *** haomaiwang has quit IRC | 03:27 | |
| *** haomaiwang has joined #openstack-glance | 03:27 | |
| *** haomaiwang has quit IRC | 03:28 | |
| *** ducttape_ has joined #openstack-glance | 03:28 | |
| *** gberginc has joined #openstack-glance | 03:28 | |
| *** haomaiwang has joined #openstack-glance | 03:28 | |
| *** haomaiwang has quit IRC | 03:29 | |
| *** haomaiwang has joined #openstack-glance | 03:29 | |
| *** haomaiwang has quit IRC | 03:30 | |
| *** 7F1AA34GJ has joined #openstack-glance | 03:30 | |
| *** 7F1AA34GJ has quit IRC | 03:31 | |
| *** haomaiwang has joined #openstack-glance | 03:31 | |
| *** haomaiwang has quit IRC | 03:32 | |
| *** haomaiwa_ has joined #openstack-glance | 03:33 | |
| *** haomaiwa_ has quit IRC | 03:34 | |
| *** haomaiwang has joined #openstack-glance | 03:34 | |
| *** haomaiwang has quit IRC | 03:35 | |
| *** jerrygb has joined #openstack-glance | 03:35 | |
| *** haomaiwa_ has joined #openstack-glance | 03:38 | |
| *** ducttape_ has quit IRC | 03:54 | |
| *** ducttape_ has joined #openstack-glance | 03:54 | |
| *** ianbrown has joined #openstack-glance | 03:55 | |
| *** haomaiwa_ has quit IRC | 04:01 | |
| *** haomaiwang has joined #openstack-glance | 04:01 | |
| *** azbiswas has joined #openstack-glance | 04:04 | |
| *** darrenc is now known as darrenc_afk | 04:06 | |
| *** azbiswas has quit IRC | 04:08 | |
| *** ducttape_ has quit IRC | 04:17 | |
| *** ianbrown has quit IRC | 04:35 | |
| *** darrenc_afk is now known as darrenc | 04:42 | |
| *** boris-42_ has quit IRC | 04:43 | |
| *** azbiswas has joined #openstack-glance | 04:51 | |
| *** abhishekk has joined #openstack-glance | 05:00 | |
| *** haomaiwang has quit IRC | 05:01 | |
| *** haomaiwang has joined #openstack-glance | 05:01 | |
| *** ianbrown has joined #openstack-glance | 05:02 | |
| *** gberginc has quit IRC | 05:22 | |
| *** GB21 has joined #openstack-glance | 05:30 | |
| *** jerrygb has quit IRC | 05:50 | |
| *** ianbrown has quit IRC | 05:57 | |
| *** ianbrown has joined #openstack-glance | 05:57 | |
| *** ianbrown_ has joined #openstack-glance | 05:57 | |
| *** ianbrown_ has quit IRC | 05:57 | |
| *** dims has quit IRC | 05:57 | |
| *** ianbrown_ has joined #openstack-glance | 06:00 | |
| *** ianbrown has quit IRC | 06:00 | |
| *** haomaiwang has quit IRC | 06:01 | |
| *** haomaiwang has joined #openstack-glance | 06:01 | |
| *** wolsen has quit IRC | 06:04 | |
| *** wolsen has joined #openstack-glance | 06:05 | |
| *** kebray has quit IRC | 06:08 | |
| *** gberginc has joined #openstack-glance | 06:13 | |
| *** ianbrown_ has quit IRC | 06:17 | |
| *** dims has joined #openstack-glance | 06:20 | |
| *** sgotliv_ has joined #openstack-glance | 06:22 | |
| *** ianbrown has joined #openstack-glance | 06:23 | |
| *** haomaiwang has quit IRC | 06:32 | |
| *** 18WABB396 has joined #openstack-glance | 06:33 | |
| openstackgerrit | wangxiyuan proposed openstack/glance: Update the cache middleware flavor guideline. https://review.openstack.org/254517 | 06:34 |
| *** e0ne has joined #openstack-glance | 06:38 | |
| *** azbiswas has quit IRC | 06:38 | |
| *** ianbrown has quit IRC | 06:40 | |
| *** ianbrown has joined #openstack-glance | 06:40 | |
| *** ianbrown has quit IRC | 06:41 | |
| *** ianbrown_ has joined #openstack-glance | 06:41 | |
| *** ianbrown__ has joined #openstack-glance | 06:43 | |
| *** ianbrown_ has quit IRC | 06:43 | |
| *** e0ne has quit IRC | 06:45 | |
| *** e0ne has joined #openstack-glance | 06:49 | |
| *** e0ne has quit IRC | 06:52 | |
| *** groen692 has joined #openstack-glance | 06:52 | |
| openstackgerrit | Jinxing Fang proposed openstack/glance: Remove the qpid message driver from config file https://review.openstack.org/254580 | 06:53 |
| *** azbiswas has joined #openstack-glance | 06:54 | |
| *** 18WABB396 has quit IRC | 07:01 | |
| *** haomaiwang has joined #openstack-glance | 07:01 | |
| *** e0ne has joined #openstack-glance | 07:02 | |
| *** houming has quit IRC | 07:08 | |
| *** haomaiwang has quit IRC | 07:09 | |
| *** haomaiwang has joined #openstack-glance | 07:09 | |
| *** dims has quit IRC | 07:11 | |
| *** houming has joined #openstack-glance | 07:11 | |
| *** houming has quit IRC | 07:16 | |
| *** e0ne has quit IRC | 07:20 | |
| *** e0ne has joined #openstack-glance | 07:21 | |
| *** haomaiwang has quit IRC | 07:28 | |
| *** haomaiwa_ has joined #openstack-glance | 07:28 | |
| *** dims has joined #openstack-glance | 07:31 | |
| *** houming has joined #openstack-glance | 07:42 | |
| *** dims has quit IRC | 07:52 | |
| *** GB21 has quit IRC | 07:52 | |
| *** e0ne has quit IRC | 07:56 | |
| *** jerrygb has joined #openstack-glance | 08:06 | |
| *** azbiswas has quit IRC | 08:07 | |
| *** azbiswas has joined #openstack-glance | 08:08 | |
| *** jerrygb has quit IRC | 08:12 | |
| *** GB21 has joined #openstack-glance | 08:12 | |
| *** azbiswas has quit IRC | 08:13 | |
| *** MattMan has quit IRC | 08:26 | |
| *** MattMan has joined #openstack-glance | 08:32 | |
| *** markus_z has joined #openstack-glance | 08:37 | |
| *** groen692 has quit IRC | 08:47 | |
| *** groen692 has joined #openstack-glance | 08:48 | |
| *** links has quit IRC | 08:55 | |
| *** e0ne has joined #openstack-glance | 09:03 | |
| *** links has joined #openstack-glance | 09:09 | |
| *** bunting has quit IRC | 09:11 | |
| *** bunting has joined #openstack-glance | 09:13 | |
| *** shakamunyi has joined #openstack-glance | 09:14 | |
| *** shakamunyi has quit IRC | 09:23 | |
| *** gberginc has quit IRC | 09:48 | |
| *** houming has quit IRC | 09:59 | |
| *** aix has joined #openstack-glance | 10:01 | |
| *** jistr has joined #openstack-glance | 10:14 | |
| *** aix has quit IRC | 10:21 | |
| *** gberginc has joined #openstack-glance | 10:26 | |
| openstackgerrit | Stuart McLaren proposed openstack/python-glanceclient: Handle 403 forbidden on download https://review.openstack.org/254328 | 10:31 |
| *** aix has joined #openstack-glance | 10:33 | |
| *** links has quit IRC | 10:37 | |
| *** links has joined #openstack-glance | 10:58 | |
| *** openstackgerrit has quit IRC | 11:02 | |
| *** openstackgerrit has joined #openstack-glance | 11:02 | |
| *** haomaiwa_ has quit IRC | 11:06 | |
| *** Kennan has quit IRC | 11:14 | |
| *** Kennan has joined #openstack-glance | 11:17 | |
| *** smatzek has joined #openstack-glance | 11:28 | |
| *** Kennan has quit IRC | 11:29 | |
| flaper87 | sabari: I'm working on creating tasks for that spec. We're good as long as you've time | 11:31 |
| *** Kennan has joined #openstack-glance | 11:32 | |
| *** azbiswas has joined #openstack-glance | 11:32 | |
| *** azbiswas has quit IRC | 11:37 | |
| *** Kennan has quit IRC | 11:39 | |
| *** Kennan has joined #openstack-glance | 11:42 | |
| *** Kennan has quit IRC | 11:43 | |
| *** abhishekk has quit IRC | 11:44 | |
| *** Kennan has joined #openstack-glance | 11:46 | |
| *** Kennan has quit IRC | 11:47 | |
| *** yasu has quit IRC | 11:47 | |
| *** Kennan has joined #openstack-glance | 11:50 | |
| openstackgerrit | Alexander Tivelkov proposed openstack/glance-specs: Glare Public (plugin-independent) API https://review.openstack.org/254710 | 11:53 |
| *** houming has joined #openstack-glance | 11:59 | |
| openstackgerrit | Alexander Tivelkov proposed openstack/glance-specs: Deprecate v3 API https://review.openstack.org/254163 | 12:00 |
| *** dshakhray has joined #openstack-glance | 12:00 | |
| *** e0ne has quit IRC | 12:03 | |
| *** jerrygb has joined #openstack-glance | 12:09 | |
| *** e0ne has joined #openstack-glance | 12:11 | |
| *** jerrygb has quit IRC | 12:13 | |
| *** Kennan has quit IRC | 12:20 | |
| *** GB21 has quit IRC | 12:21 | |
| *** Kennan has joined #openstack-glance | 12:27 | |
| *** links has quit IRC | 12:36 | |
| *** Kennan has quit IRC | 12:48 | |
| *** Kennan has joined #openstack-glance | 12:54 | |
| *** ducttape_ has joined #openstack-glance | 12:58 | |
| *** ducttape_ has quit IRC | 13:02 | |
| *** ducttape_ has joined #openstack-glance | 13:06 | |
| *** ducttape_ has quit IRC | 13:16 | |
| *** dims has joined #openstack-glance | 13:16 | |
| flaper87 | kairat: btw, you abandoned yourfix for: https://bugs.launchpad.net/glance/+bug/1519402 | 13:18 |
| openstack | Launchpad bug 1519402 in python-glanceclient "Add docker container format to defaults" [Wishlist,In progress] - Assigned to Atsushi SAKAI (sakaia) | 13:18 |
| flaper87 | do you have the other link handy? | 13:18 |
| flaper87 | I mean, the link to the patch that superseded yours ? | 13:19 |
| kairat | flaper87, https://review.openstack.org/#/c/252806/ | 13:19 |
| kairat | flaper87, I proposed to update the patch with the reference to the bug but it seems that the progress is not so fast as expected =) | 13:20 |
| flaper87 | kairat: FWIW, feel free to take over and add yourself as a co-author. If you do a commit, the original author will be preserved | 13:22 |
| *** ianbrown__ has quit IRC | 13:22 | |
| *** ianbrown_ has joined #openstack-glance | 13:22 | |
| kairat | flaper87, ok, will do this if there will be no updates today | 13:22 |
| *** edmondsw has joined #openstack-glance | 13:31 | |
| *** itisha has joined #openstack-glance | 13:33 | |
| *** mingdang1 has joined #openstack-glance | 13:34 | |
| *** jerrygb has joined #openstack-glance | 13:44 | |
| *** julim has joined #openstack-glance | 13:55 | |
| *** nikhil_k is now known as nikhil | 14:05 | |
| *** ducttape_ has joined #openstack-glance | 14:08 | |
| openstackgerrit | Dina Belova proposed openstack/glance: Do not use api-paste.ini osprofiler options https://review.openstack.org/245060 | 14:09 |
| *** changbl has quit IRC | 14:10 | |
| *** ducttape_ has quit IRC | 14:15 | |
| *** gberginc has quit IRC | 14:16 | |
| kairat | nikhil, see comments here https://review.openstack.org/#/c/246943/ to understand why we can't fail with authrization error in swiftclient | 14:29 |
| kairat | nikhil, unfortunately we can lose data here | 14:29 |
| *** mclaren_ has joined #openstack-glance | 14:30 | |
| kairat | I will refine spec description to explain why we can't use that approach | 14:30 |
| rosmaita | mclaren_: hello | 14:30 |
| flaper87 | -1 on starting on theswift case | 14:30 |
| mclaren_ | ok, why? | 14:30 |
| flaper87 | I'm working on a series of tasks for us to start working on this spec | 14:30 |
| flaper87 | but the first scenario we should cover is direct upload | 14:30 |
| mclaren_ | why? | 14:31 |
| flaper87 | that's what's mostly broken now and that's what's part of the compute group in defcore | 14:31 |
| mclaren_ | right, but we're more ready to start on the swift one | 14:31 |
| rosmaita | mclaren_: what flaper87 said, that's the feedback from the rest of the community | 14:31 |
| flaper87 | (sorry, as I mentioned I've some lag, my messages go out a bit late) | 14:31 |
| nikhil | kairat: ack, will look. I a curious about overlap with https://review.openstack.org/#/c/120866 | 14:31 |
| flaper87 | mclaren_: I'm sorry but no. whatever time we will put on coding hte swift part I'd rather put it in figuring out the solution ofr the direct upload one | 14:32 |
| rosmaita | mclaren_: i think the key thing is that we don't do anything that fucks up the swift-local workflow | 14:32 |
| mclaren_ | ok, what can we make a start on before finalizing the bikeshed side of things? | 14:32 |
| mclaren_ | gotta go, back in ~15mins | 14:32 |
| flaper87 | mclaren_: as I said, I'm breaking the spec down into tasks. Started yesterday and I'll finish today | 14:32 |
| nikhil | -1 on starting now | 14:32 |
| kairat | nikhil, i reviewed that, there will be no overlap, will explain this in spec also. | 14:33 |
| rosmaita | mclaren_: ping when you get back | 14:33 |
| flaper87 | that should give us several places to start on that don't require bikeshed to be ready | 14:33 |
| flaper87 | mclaren_: ^ | 14:33 |
| nikhil | there's been so much time spent discussing this, starting w/o full idea is going to make things worse | 14:33 |
| nikhil | kairat: thanks! | 14:33 |
| flaper87 | nikhil: there are things we can start with. There are required policy changes, api version bumps and other cosmetic things that will be needed anyway | 14:34 |
| rosmaita | flaper87: mclaren_: my feeling is that we start with no end-user exposure of bikeshed, and then if it turns out that it woudl be really useful, we can rethink it | 14:34 |
| flaper87 | thaose are the things I'm mentioning that we could get done now | 14:34 |
| flaper87 | nothing big that requires discussion | 14:34 |
| flaper87 | rosmaita: ++ exactly my thoughts now | 14:34 |
| flaper87 | it'd be easier to add the bikeshed resource later than adding it now and having to maintain it forever | 14:35 |
| *** mingdang1 has quit IRC | 14:35 | |
| flaper87 | w/ the discoverability in place, it shouldn't be hard to add it later | 14:35 |
| rosmaita | i think all the end-user wants to know is (1) glance has my data, and (2) glance gives me a checksum so i know my data wasn't messed up in transit | 14:36 |
| rosmaita | other than that, i've handed my data over to glance to do its magic and turn it into image data | 14:37 |
| rosmaita | so i shouldn't care what happens after that | 14:37 |
| nikhil | rosmaita: I think (1) is probably more like "glance has my data in the format that works for the operator" | 14:37 |
| nikhil | and (2) would change accordingly | 14:37 |
| rosmaita | nikhil: well, for (2) i'm thinking of the checksum of what glance received, before glance manipulates it as part of import process | 14:38 |
| *** smatzek has quit IRC | 14:38 | |
| nikhil | rosmaita: I see, so we need to add that separately somewhere? | 14:38 |
| rosmaita | nikhil: i think so, maybe as part of the Image "message" element | 14:39 |
| nikhil | gotcha | 14:39 |
| rosmaita | so for your use case, user (1) creates Image object, (2) does a PUT of an OVA, (3) does GET on Image to see status indicating "ready to import" and the checksum of what is in the bikeshed | 14:40 |
| *** ayoung has joined #openstack-glance | 14:40 | |
| nikhil | makes sense | 14:41 |
| rosmaita | then user (4) does an import call, and (5) does GET on Image until status==active, and at that point the 'checksum' in the Image is the checksum of what's actually stored in glance, i.e., the image in the OVA | 14:41 |
| mclaren_ | ping | 14:41 |
| rosmaita | hi stuart | 14:41 |
| flaper87 | kinda pong | 14:41 |
| flaper87 | :D | 14:41 |
| mclaren_ | sorry, scrum meeting | 14:41 |
| mclaren_ | back now | 14:42 |
| flaper87 | Ill have to leave in 10/15 mins | 14:43 |
| nikhil | and I need to be back in 15, sorry | 14:43 |
| mclaren_ | I'm not sure I've much to add at this point, will think about the 'invisible' bikeshed :-) | 14:43 |
| rosmaita | mclaren_: i think we can always make it visibile if that turns out to be the right thing | 14:44 |
| flaper87 | right | 14:44 |
| rosmaita | i'd like to try seeing how far we get without making it visible | 14:44 |
| flaper87 | invisible -> visible is easier than the other way around | 14:44 |
| rosmaita | mclaren_: i think we can keep separation of concerns "behind the scenes" | 14:45 |
| rosmaita | hopefully | 14:45 |
| mclaren_ | well if you've a new state, not really | 14:45 |
| mclaren_ | and if v1 delete deletes the bikeshed, not really either | 14:45 |
| mclaren_ | you're modifying your existing implementation | 14:46 |
| *** edmondsw has quit IRC | 14:48 | |
| rosmaita | mclaren_: ok, you are right | 14:48 |
| rosmaita | so now i will say separation of concerns is overrated | 14:48 |
| rosmaita | something like "too much consistency" | 14:48 |
| rosmaita | :) | 14:48 |
| rosmaita | mclaren_: keep pushing, i really don't want to do something stupid here (as i have apparently done already with tasks) | 14:49 |
| mclaren_ | I really don't like being pushy! | 14:50 |
| rosmaita | mclaren_: ok, you can nudge, then | 14:51 |
| mclaren_ | and I don't think I have it all figured out either btw | 14:51 |
| *** dims_ has joined #openstack-glance | 14:52 | |
| rosmaita | mclaren_: ok, your v1-delete point is well taken ... but how does making the bikeshed visible help that? | 14:52 |
| mclaren_ | But when, for example, I see what AWS have done -- they didn't hide the checksum. That feels a bit weird. I uploaded that thing yesterday. Did the checksum verification work? I can't remember. | 14:52 |
| mclaren_ | how does making the bikeshed visible help that? By not having it fused with an image. A la the S3 implementation. | 14:53 |
| mclaren_ | I guess my gut feeling here is this | 14:53 |
| mclaren_ | we have some code that's a bit like a wobbly Jenga tower | 14:53 |
| rosmaita | so if we took this to an extreme, you could set up 2 glances -- one for bootable images, and one for bikeshedding | 14:54 |
| rosmaita | (sorry, didn't mean to interrupt) | 14:54 |
| mclaren_ | we have the option of either A) swapping some blocks in that tower around and then building more stuff on top of that or B) not touching it and building a separate tower | 14:55 |
| *** dims has quit IRC | 14:55 | |
| rosmaita | note to self: want to ask sigmavirus24_awa if he knows what a yak pen is called, we could use that term instead of 'bikeshed' | 14:55 |
| rosmaita | mclaren_: i guess i'm not convinced that we can really do B | 14:55 |
| rosmaita | that's basically what the import task did | 14:56 |
| rosmaita | and it has been enthusiastically rejected | 14:56 |
| mclaren_ | well the import task only worked with external data | 14:56 |
| mclaren_ | so it was a task, not data | 14:56 |
| mclaren_ | I don't see it as the same at all | 14:56 |
| mclaren_ | but I do get what you're saying, the community doesn't necessarily listen to rational arguments | 14:58 |
| *** ducttape_ has joined #openstack-glance | 14:59 | |
| mclaren_ | I can try to ping Jay maybe | 14:59 |
| mclaren_ | see what he thinks | 15:00 |
| rosmaita | i would be interested | 15:00 |
| rosmaita | particularly aobut separation of concerns | 15:00 |
| mclaren_ | ok, let me try to reach out to him. | 15:01 |
| rosmaita | cool | 15:01 |
| mclaren_ | I'm happy that we've decided on strategy and have moved on to tactics | 15:01 |
| rosmaita | i guess so | 15:02 |
| rosmaita | i don't know what i think anymore! | 15:02 |
| mclaren_ | yeah it's been a slog, we'll get there though | 15:02 |
| *** ducttape_ has quit IRC | 15:03 | |
| *** ducttape_ has joined #openstack-glance | 15:03 | |
| sigmavirus24_awa | rosmaita: lol | 15:05 |
| sigmavirus24_awa | I've never used a yak pen before | 15:05 |
| rosmaita | it was your "shaving yaks" comment that got me thinking | 15:05 |
| *** sigmavirus24_awa is now known as sigmavirus24 | 15:06 | |
| rosmaita | maybe this will have something: http://www.fao.org/docrep/006/ad347e/ad347e0j.htm | 15:07 |
| rosmaita | apparently yak pens are sometimes made of feces ... is that appropriate to our use case, or what? | 15:08 |
| *** azbiswas has joined #openstack-glance | 15:08 | |
| *** TravT has quit IRC | 15:09 | |
| *** jerrygb has quit IRC | 15:09 | |
| *** jerrygb has joined #openstack-glance | 15:10 | |
| jokke_ | I would really really like to just stamp the bikeshed as bikeshed ... lets just make it official ... "Dump your data to the bikeshed, let us know and we'll take it from there" | 15:11 |
| *** edmondsw has joined #openstack-glance | 15:13 | |
| *** smatzek has joined #openstack-glance | 15:14 | |
| *** azbiswas has quit IRC | 15:14 | |
| *** jerrygb has quit IRC | 15:15 | |
| sigmavirus24 | jokke_: cloud providers determine the color of the bikeshed? | 15:20 |
| sigmavirus24 | a.k.a., what the image is in the end | 15:20 |
| sigmavirus24 | Hey, I have an idea | 15:20 |
| sigmavirus24 | What if we make a new standard that handles all the other image formats and adds new goodies on top to unify everything? | 15:21 |
| * sigmavirus24 hides in a corner | 15:21 | |
| *** _ducttape_ has joined #openstack-glance | 15:21 | |
| *** vijendar has joined #openstack-glance | 15:22 | |
| flaper87 | rosmaita: commented w/ a list of tasks | 15:22 |
| rosmaita | flaper87: thanks, i will get to work | 15:23 |
| * sigmavirus24 cracks whip over rosmaita's back | 15:23 | |
| rosmaita | sigmavirus24: glad to have you back! it wasn't the same without you | 15:24 |
| *** jerrygb has joined #openstack-glance | 15:24 | |
| sigmavirus24 | Whoops. I keep confusing myself for jcook | 15:24 |
| sigmavirus24 | sorry rosmaita | 15:24 |
| * sigmavirus24 puts whip away | 15:24 | |
| *** ducttape_ has quit IRC | 15:24 | |
| *** mtanino has joined #openstack-glance | 15:33 | |
| *** erhudy has joined #openstack-glance | 15:36 | |
| *** jistr is now known as jistr|mtg | 15:43 | |
| *** gokrokve has joined #openstack-glance | 15:59 | |
| jcook | ? | 16:02 |
| jcook | what about me :) | 16:02 |
| sigmavirus24 | Oh never you mind :P | 16:03 |
| *** avarner has joined #openstack-glance | 16:03 | |
| jcook | :) | 16:06 |
| flaper87 | nikhil: sanity check, pls, https://bugs.launchpad.net/glance/+bug/1518431 | 16:07 |
| openstack | Launchpad bug 1518431 in Glance "Glance failed to upload image to swift storage" [Undecided,Confirmed] | 16:07 |
| flaper87 | ops | 16:07 |
| flaper87 | nikhil: I meant, https://review.openstack.org/#/c/250774/ | 16:07 |
| *** aix has quit IRC | 16:08 | |
| *** smatzek has quit IRC | 16:09 | |
| *** ninag has joined #openstack-glance | 16:09 | |
| *** ninag has quit IRC | 16:09 | |
| sigmavirus24 | flaper87: that bug is interesting | 16:11 |
| flaper87 | sigmavirus24: it is, indeed. I think we finally found the root cause | 16:13 |
| flaper87 | let's see what fix cyril comes up with | 16:13 |
| flaper87 | :D | 16:13 |
| sigmavirus24 | wooo for "compatible" APIs, amiright? | 16:13 |
| kairat | I noticed that swiftclient seems not send content-length or transfer-encoding to rados | 16:14 |
| kairat | It seems that apache FastCGI blocks the request because it is not compatible with HTTP specification | 16:14 |
| kairat | We ar sending our request here: https://github.com/openstack/glance_store/blob/master/glance_store/_drivers/swift/store.py#L550 | 16:15 |
| *** changbl has joined #openstack-glance | 16:15 | |
| kairat | There is no content-length specified: content_length = None | 16:15 |
| kairat | all that we have are container, name and reader | 16:16 |
| *** ianbrown_ has quit IRC | 16:17 | |
| *** spzala has quit IRC | 16:18 | |
| kairat | After that swiftclient prepares a request here: https://github.com/openstack/python-swiftclient/blob/master/swiftclient/client.py#L1117 | 16:18 |
| kairat | *prepares data | 16:18 |
| *** pumaranikar has joined #openstack-glance | 16:22 | |
| kairat | Hmm, it looks like swiftclient do not specify transfer encoding nor content-length if image size is not defined | 16:23 |
| kairat | I am wondering who need to specify this then | 16:24 |
| *** dshakhray has quit IRC | 16:24 | |
| *** changbl has quit IRC | 16:25 | |
| *** changbl has joined #openstack-glance | 16:26 | |
| openstackgerrit | Merged openstack/glance-specs: Revert "Cap Pillow to fix doc gates" https://review.openstack.org/239259 | 16:27 |
| *** ianbrown_ has joined #openstack-glance | 16:29 | |
| *** changbl has quit IRC | 16:30 | |
| *** smatzek has joined #openstack-glance | 16:32 | |
| openstackgerrit | Merged openstack/glance-specs: Fix reviewers list render (db purge) https://review.openstack.org/250774 | 16:33 |
| kairat | mclaren_, I know you also participated in swift development | 16:37 |
| kairat | mclaren_, it looks like that swiftclient do not specify transfer encoding nor content-length if image size is not defined | 16:38 |
| *** gberginc has joined #openstack-glance | 16:40 | |
| *** groen692 has quit IRC | 16:42 | |
| *** changbl has joined #openstack-glance | 16:43 | |
| *** boris-42_ has joined #openstack-glance | 16:44 | |
| sigmavirus24 | kairat: link to the code? This might be handled by requests for us | 16:44 |
| *** openstackgerrit has quit IRC | 16:47 | |
| *** openstackgerrit has joined #openstack-glance | 16:47 | |
| kairat | sigmavirus24, https://github.com/openstack/glance_store/blob/master/glance_store/_drivers/swift/store.py#L550 | 16:47 |
| kairat | that is the request to swift if we don't know content-length | 16:47 |
| sigmavirus24 | So swiftclient is handling this case then | 16:48 |
| kairat | image_size is 0 so content-length is None | 16:48 |
| kairat | yep, but if you look at the paste in the bug | 16:48 |
| kairat | https://bugs.launchpad.net/glance/+bug/1518431 | 16:49 |
| openstack | Launchpad bug 1518431 in Glance "Glance failed to upload image to swift storage" [Undecided,Confirmed] | 16:49 |
| kairat | it seems that not length or transfer encoding specified | 16:49 |
| kairat | So apparently apache do not allow that | 16:49 |
| kairat | I raised this question in openstack-swift | 16:49 |
| nikhil | flaper87: ack | 16:50 |
| sigmavirus24 | kairat: in https://github.com/openstack/python-swiftclient/blob/master/swiftclient/client.py#L1547 content_length is passed on | 16:51 |
| kairat | sigmavirus24, we are passing None AFAIK | 16:51 |
| sigmavirus24 | kairat: which is the default for that method | 16:52 |
| kairat | yep | 16:52 |
| kairat | but swift do not handle this correctly then | 16:52 |
| sigmavirus24 | kairat: while you're in openstack-swift, ask them how they approved https://github.com/openstack/python-swiftclient/blob/master/swiftclient/client.py#L1096..L1106 | 16:52 |
| kairat | heh | 16:52 |
| sigmavirus24 | retry calls https://github.com/openstack/python-swiftclient/blob/master/swiftclient/client.py#L1034 | 16:53 |
| sigmavirus24 | kairat: I'd guess this is a bug in those lines actually that I linked | 16:54 |
| kairat | Will do as soon as receive an answer about the bug) | 16:54 |
| sigmavirus24 | Because the put_object method calls that put_object function | 16:54 |
| sigmavirus24 | So, they appear to always be assigning an `int` to the header value which afaik neither urllib3 or requests like | 16:55 |
| sigmavirus24 | oh gosh | 16:55 |
| sigmavirus24 | nevermind | 16:55 |
| sigmavirus24 | I'm confusing content-length and content-type | 16:55 |
| kairat | sigmavirus24, I don't think so | 16:55 |
| kairat | headers is None | 16:55 |
| *** ianbrown__ has joined #openstack-glance | 16:55 | |
| *** alejandrito has joined #openstack-glance | 16:55 | |
| kairat | so finally we should be here: https://github.com/openstack/python-swiftclient/blob/master/swiftclient/client.py#L1117 | 16:56 |
| sigmavirus24 | So here is what happens https://github.com/openstack/python-swiftclient/blob/master/swiftclient/client.py#L1117 | 16:56 |
| sigmavirus24 | yeah | 16:56 |
| sigmavirus24 | Right | 16:56 |
| kairat | And I didn't find any code where they specifies transfer-encoding | 16:56 |
| kairat | So it looks like violation of HTTP spec | 16:56 |
| sigmavirus24 | so if that gets passed to requests requests should emit transfer-encoding: chunked | 16:56 |
| kairat | Perhaps | 16:57 |
| sigmavirus24 | kairat: did you do a packet trace to show that it's not actually being set by swiftclient? | 16:57 |
| sigmavirus24 | The ReadableToIterable is a class that makes requests think its handling a generator | 16:57 |
| *** ianbrown_ has quit IRC | 16:57 | |
| sigmavirus24 | Because of that requests handles it as a chunked transfer-encoding | 16:57 |
| sigmavirus24 | (Trust me, I know this very well) | 16:57 |
| sigmavirus24 | kairat: if you're able to reproduce this, please send me a packet trace showing that we're not setting the t-e header | 16:58 |
| kairat | nope, unfortunately, need some time to configure ceph back-end:( | 16:58 |
| kairat | sigmavirus24, all that I have is this: http://paste.openstack.org/show/479621/ | 16:58 |
| kairat | IIUC, there is no transfer-encoding generated in request | 16:59 |
| kairat | although, requests can add this header after logging | 17:00 |
| kairat | so you can be right then | 17:00 |
| sigmavirus24 | requests has no logging of its own and urllib3 has logging but doesn't log headers because of potentially sensitive information | 17:00 |
| openstackgerrit | Cyril Roelandt proposed openstack/glance_store: Fix a race condition in the Swift store https://review.openstack.org/254873 | 17:00 |
| openstackgerrit | Cyril Roelandt proposed openstack/glance_store: Fix a race condition in the Swift store https://review.openstack.org/254873 | 17:03 |
| *** alex_xu has quit IRC | 17:04 | |
| kairat | so yes | 17:04 |
| sigmavirus24 | kairat: ? | 17:05 |
| kairat | need to trace requests to ensure if this is the problem with transfer-encoding: chunked | 17:05 |
| kairat | I asked guys about that | 17:05 |
| kairat | unfortunately i do not have ceph installed | 17:05 |
| kairat | waiting for answer | 17:05 |
| *** kebray has joined #openstack-glance | 17:05 | |
| *** azbiswas has joined #openstack-glance | 17:05 | |
| *** Kennan has quit IRC | 17:06 | |
| *** kebray has quit IRC | 17:06 | |
| *** houming has quit IRC | 17:07 | |
| *** kebray has joined #openstack-glance | 17:07 | |
| *** Kennan has joined #openstack-glance | 17:07 | |
| *** ayoung has quit IRC | 17:10 | |
| openstackgerrit | Cyril Roelandt proposed openstack/glance_store: Fix a race condition in the Swift store https://review.openstack.org/254873 | 17:10 |
| *** alex_xu has joined #openstack-glance | 17:10 | |
| *** azbiswas has quit IRC | 17:11 | |
| *** RichardRaseley has joined #openstack-glance | 17:17 | |
| *** dshakhray has joined #openstack-glance | 17:19 | |
| *** jistr|mtg is now known as jistr | 17:21 | |
| *** ayoung has joined #openstack-glance | 17:25 | |
| *** ayoung has quit IRC | 17:31 | |
| openstackgerrit | Niall Bunting proposed openstack/python-glanceclient: Disable suggestion of v1 help for v2 commands https://review.openstack.org/254883 | 17:32 |
| *** smatzek has quit IRC | 17:39 | |
| *** TravT has joined #openstack-glance | 17:43 | |
| *** ayoung has joined #openstack-glance | 17:44 | |
| *** ctina_ has joined #openstack-glance | 17:46 | |
| *** TravT has quit IRC | 17:48 | |
| *** TravT has joined #openstack-glance | 17:48 | |
| *** e0ne has quit IRC | 17:50 | |
| *** ianbrown__ has quit IRC | 17:52 | |
| openstackgerrit | Merged openstack/glance: Fix default value with postgreSQL https://review.openstack.org/208851 | 17:53 |
| *** ianbrown__ has joined #openstack-glance | 17:56 | |
| *** markus_z has quit IRC | 17:57 | |
| *** kebray has quit IRC | 18:00 | |
| *** smatzek has joined #openstack-glance | 18:02 | |
| *** kebray has joined #openstack-glance | 18:03 | |
| *** TravT has quit IRC | 18:03 | |
| *** TravT has joined #openstack-glance | 18:03 | |
| *** jistr has quit IRC | 18:05 | |
| *** alejandrito has quit IRC | 18:19 | |
| *** alejandrito has joined #openstack-glance | 18:19 | |
| *** vijendar has left #openstack-glance | 18:21 | |
| *** kebray has quit IRC | 18:22 | |
| *** gokrokve has quit IRC | 18:24 | |
| *** gokrokve has joined #openstack-glance | 18:25 | |
| *** shakamunyi has joined #openstack-glance | 18:25 | |
| *** pumaranikar has quit IRC | 18:29 | |
| *** gokrokve has quit IRC | 18:30 | |
| *** dims_ has quit IRC | 18:30 | |
| *** gokrokve has joined #openstack-glance | 18:31 | |
| *** azbiswas has joined #openstack-glance | 18:32 | |
| *** gokrokve has quit IRC | 18:33 | |
| *** gokrokve has joined #openstack-glance | 18:34 | |
| *** jerrygb has quit IRC | 18:39 | |
| *** e0ne has joined #openstack-glance | 18:39 | |
| *** pumaranikar has joined #openstack-glance | 18:43 | |
| *** Kennan has quit IRC | 18:43 | |
| *** Kennan has joined #openstack-glance | 18:44 | |
| *** etoews has joined #openstack-glance | 18:44 | |
| *** etoews has left #openstack-glance | 18:45 | |
| *** ctina_ has quit IRC | 18:46 | |
| *** jerrygb has joined #openstack-glance | 18:46 | |
| *** david-ly_ has joined #openstack-glance | 18:50 | |
| *** david-ly_ is now known as david-lyle_ | 18:52 | |
| *** david-lyle has quit IRC | 18:54 | |
| *** sgotliv_ has quit IRC | 18:55 | |
| *** openstackgerrit has quit IRC | 19:02 | |
| *** openstackgerrit has joined #openstack-glance | 19:02 | |
| *** TravT has quit IRC | 19:05 | |
| *** avarner has quit IRC | 19:07 | |
| *** avarner has joined #openstack-glance | 19:07 | |
| *** gokrokve has quit IRC | 19:08 | |
| *** kebray has joined #openstack-glance | 19:11 | |
| *** david-lyle_ is now known as david-lyle | 19:14 | |
| *** gokrokve has joined #openstack-glance | 19:16 | |
| *** gokrokve has quit IRC | 19:31 | |
| *** sgotliv_ has joined #openstack-glance | 19:34 | |
| *** kebray has quit IRC | 19:38 | |
| *** harlowja has quit IRC | 19:38 | |
| *** harlowja has joined #openstack-glance | 19:38 | |
| *** erhudy has quit IRC | 19:40 | |
| *** TravT has joined #openstack-glance | 19:54 | |
| openstackgerrit | Bertrand Lallau proposed openstack/glance: Remove iso8601 dependency https://review.openstack.org/246565 | 19:58 |
| *** kebray has joined #openstack-glance | 20:06 | |
| *** kebray_ has joined #openstack-glance | 20:07 | |
| *** kebray has quit IRC | 20:07 | |
| *** kebray_ has quit IRC | 20:18 | |
| *** vijendar has joined #openstack-glance | 20:23 | |
| *** kebray has joined #openstack-glance | 20:38 | |
| *** avarner has quit IRC | 20:53 | |
| *** kebray has quit IRC | 21:01 | |
| *** ianbrown__ has quit IRC | 21:19 | |
| *** ianbrown has joined #openstack-glance | 21:23 | |
| *** cdelatte has quit IRC | 21:23 | |
| *** smatzek has quit IRC | 21:23 | |
| *** TravT has quit IRC | 21:25 | |
| *** ianbrown has quit IRC | 21:27 | |
| *** ianbrown has joined #openstack-glance | 21:28 | |
| *** dims has joined #openstack-glance | 21:31 | |
| *** ayoung has quit IRC | 21:31 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/glance: Updated from global requirements https://review.openstack.org/254469 | 21:31 |
| *** mingdang1 has joined #openstack-glance | 21:33 | |
| *** ianbrown_ has joined #openstack-glance | 21:33 | |
| *** ianbrown has quit IRC | 21:34 | |
| *** ianbrown has joined #openstack-glance | 21:41 | |
| *** cdelatte has joined #openstack-glance | 21:49 | |
| openstackgerrit | Merged openstack/glance_store: Updated from global requirements https://review.openstack.org/254087 | 21:49 |
| *** azbiswas has quit IRC | 21:51 | |
| *** gonzalo2kx has joined #openstack-glance | 21:58 | |
| *** dims has quit IRC | 21:59 | |
| *** sgotliv_ has quit IRC | 22:00 | |
| *** gonzalo2kx has quit IRC | 22:03 | |
| *** gonzalo2kx has joined #openstack-glance | 22:05 | |
| *** gonzalo2kx has quit IRC | 22:05 | |
| *** gokrokve has joined #openstack-glance | 22:05 | |
| *** gokrokve has quit IRC | 22:07 | |
| *** gokrokve has joined #openstack-glance | 22:07 | |
| *** dshakhray has quit IRC | 22:08 | |
| *** david-ly_ has joined #openstack-glance | 22:10 | |
| *** david-lyle has quit IRC | 22:13 | |
| *** david-ly_ is now known as david-lyle | 22:13 | |
| *** belmoreira has joined #openstack-glance | 22:19 | |
| *** julim has quit IRC | 22:21 | |
| *** e0ne has quit IRC | 22:26 | |
| *** kebray has joined #openstack-glance | 22:26 | |
| *** changbl has quit IRC | 22:31 | |
| *** kebray has quit IRC | 22:31 | |
| *** mingdang1 has quit IRC | 22:33 | |
| openstackgerrit | Brianna Poulos proposed openstack/glance: Add sign-the-hash deprecation warning https://review.openstack.org/254987 | 22:34 |
| *** pumaranikar has quit IRC | 22:35 | |
| *** _ducttape_ has quit IRC | 22:38 | |
| *** gokrokve has quit IRC | 22:39 | |
| *** ayoung has joined #openstack-glance | 22:43 | |
| *** gokrokve has joined #openstack-glance | 22:49 | |
| *** gokrokve has quit IRC | 22:50 | |
| *** gokrokve has joined #openstack-glance | 22:53 | |
| *** ianbrown_ has quit IRC | 22:55 | |
| *** gokrokve_ has joined #openstack-glance | 22:55 | |
| *** gokrokve has quit IRC | 22:56 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 22:56 | |
| *** ianbrown_ has joined #openstack-glance | 22:57 | |
| *** burgerk has quit IRC | 22:58 | |
| *** ianbrown has quit IRC | 23:02 | |
| *** gokrokve_ has quit IRC | 23:03 | |
| *** chlong has quit IRC | 23:10 | |
| *** edmondsw has quit IRC | 23:21 | |
| *** burgerk has joined #openstack-glance | 23:29 | |
| *** vijendar has quit IRC | 23:31 | |
| *** alex_xu has quit IRC | 23:36 | |
| *** alex_xu has joined #openstack-glance | 23:41 | |
| *** jerrygb has quit IRC | 23:55 | |
| *** jerrygb has joined #openstack-glance | 23:56 | |
| *** burgerk has quit IRC | 23:56 | |
| *** takedakn has joined #openstack-glance | 23:56 | |
| *** mingdang1 has joined #openstack-glance | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!