| *** sdake has joined #openstack-glance | 00:08 | |
| *** rrecio has quit IRC | 00:20 | |
| *** buzztroll has quit IRC | 00:40 | |
| *** ducttape_ has joined #openstack-glance | 00:41 | |
| *** ducttape_ has quit IRC | 00:41 | |
| *** ducttape_ has joined #openstack-glance | 00:41 | |
| *** david-lyle has quit IRC | 00:42 | |
| *** david-lyle__ has joined #openstack-glance | 00:45 | |
| *** david-lyle__ has quit IRC | 00:45 | |
| *** sdake has quit IRC | 00:49 | |
| *** ducttape_ has quit IRC | 00:51 | |
| *** ducttape_ has joined #openstack-glance | 00:53 | |
| *** ducttape_ has quit IRC | 01:07 | |
| openstackgerrit | Sabari proposed openstack/python-glanceclient: Fix warlock model creation https://review.openstack.org/338630 | 01:25 |
|---|---|---|
| *** gjayavelu has quit IRC | 01:25 | |
| *** ducttape_ has joined #openstack-glance | 01:28 | |
| *** YifanZeng has joined #openstack-glance | 01:28 | |
| sabari | mfedosin: ^ Ah, I thought the bug was unassigned and posted the patch but looks like that was for a different project. Hope you don't mind !!! | 01:29 |
| *** mtanino has joined #openstack-glance | 01:33 | |
| *** bpokorny has quit IRC | 01:38 | |
| *** bpokorny has joined #openstack-glance | 01:38 | |
| *** sdake has joined #openstack-glance | 01:50 | |
| *** wxy__ has quit IRC | 01:50 | |
| *** bpokorny_ has joined #openstack-glance | 01:51 | |
| *** bpokorny has quit IRC | 01:55 | |
| *** bpokorny_ has quit IRC | 01:57 | |
| *** btully has joined #openstack-glance | 02:03 | |
| *** julim has joined #openstack-glance | 02:05 | |
| *** btully has quit IRC | 02:07 | |
| *** cdelatte has quit IRC | 02:14 | |
| *** gjayavelu has joined #openstack-glance | 02:21 | |
| *** ducttape_ has quit IRC | 02:34 | |
| *** gjayavelu has quit IRC | 02:40 | |
| *** buzztroll has joined #openstack-glance | 02:43 | |
| *** buzztroll has quit IRC | 02:43 | |
| *** gjayavelu has joined #openstack-glance | 03:11 | |
| *** ducttape_ has joined #openstack-glance | 03:22 | |
| *** buzztroll has joined #openstack-glance | 03:28 | |
| *** gjayavelu has quit IRC | 03:28 | |
| *** sgotliv__ has joined #openstack-glance | 03:32 | |
| *** ducttape_ has quit IRC | 03:35 | |
| *** sgotliv__ has quit IRC | 03:38 | |
| *** MattMan has quit IRC | 03:42 | |
| *** MattMan has joined #openstack-glance | 03:42 | |
| *** jamielennox is now known as jamielennox|away | 03:43 | |
| *** sdake has quit IRC | 03:48 | |
| *** jamielennox|away is now known as jamielennox | 03:56 | |
| *** mtanino has quit IRC | 03:57 | |
| *** gjayavelu has joined #openstack-glance | 03:59 | |
| *** gjayavelu has quit IRC | 03:59 | |
| *** links has joined #openstack-glance | 04:03 | |
| *** buzztroll has quit IRC | 04:21 | |
| *** sdake has joined #openstack-glance | 04:32 | |
| *** sgotliv__ has joined #openstack-glance | 04:35 | |
| *** jamielennox is now known as jamielennox|away | 04:35 | |
| *** sgotliv__ has quit IRC | 04:41 | |
| *** jamielennox|away is now known as jamielennox | 04:42 | |
| *** sgotliv__ has joined #openstack-glance | 04:44 | |
| *** prateek has joined #openstack-glance | 04:50 | |
| *** sdake has quit IRC | 04:51 | |
| *** sgotliv__ has quit IRC | 04:52 | |
| *** sdake has joined #openstack-glance | 04:55 | |
| *** ratailor has joined #openstack-glance | 05:01 | |
| *** gjayavelu has joined #openstack-glance | 05:06 | |
| *** ducttape_ has joined #openstack-glance | 05:06 | |
| *** flwang1 has quit IRC | 05:09 | |
| *** gjayavelu has quit IRC | 05:10 | |
| *** ducttape_ has quit IRC | 05:12 | |
| *** wxy has quit IRC | 05:20 | |
| *** btully has joined #openstack-glance | 05:21 | |
| *** gjayavelu has joined #openstack-glance | 05:25 | |
| *** sdake has quit IRC | 05:26 | |
| *** M00nr41n has joined #openstack-glance | 05:37 | |
| *** tshefi has joined #openstack-glance | 05:41 | |
| *** flwang1 has joined #openstack-glance | 05:51 | |
| *** sdake has joined #openstack-glance | 06:10 | |
| *** gjayavelu has quit IRC | 06:18 | |
| *** flwang1 has quit IRC | 06:18 | |
| *** rcernin has joined #openstack-glance | 06:18 | |
| *** sdake has quit IRC | 06:19 | |
| *** sdake has joined #openstack-glance | 06:20 | |
| *** sdake has quit IRC | 06:29 | |
| *** gjayavelu has joined #openstack-glance | 06:38 | |
| *** sdake has joined #openstack-glance | 06:39 | |
| *** flwang1 has joined #openstack-glance | 06:47 | |
| *** sdake has quit IRC | 06:55 | |
| *** giri has joined #openstack-glance | 06:56 | |
| *** gjayavelu has quit IRC | 06:57 | |
| *** tesseract- has joined #openstack-glance | 07:03 | |
| *** flwang1 has quit IRC | 07:04 | |
| *** flwang1 has joined #openstack-glance | 07:09 | |
| *** flwang1 has quit IRC | 07:20 | |
| openstackgerrit | Kairat Kushaev proposed openstack/glance: Correct reraising of exception https://review.openstack.org/338040 | 07:31 |
| *** groen692 has joined #openstack-glance | 07:54 | |
| *** mvk has joined #openstack-glance | 07:57 | |
| *** ratailor is now known as ratailor|Lunch | 07:57 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/glance: [WIP] Testing latest u-c https://review.openstack.org/318436 | 08:10 |
| *** ducttape_ has joined #openstack-glance | 08:12 | |
| *** ducttape_ has quit IRC | 08:17 | |
| *** openstackgerrit has quit IRC | 08:18 | |
| *** openstackgerrit has joined #openstack-glance | 08:18 | |
| *** pcaruana has joined #openstack-glance | 08:30 | |
| *** ratailor|Lunch is now known as ratailor | 08:41 | |
| *** btully has quit IRC | 08:41 | |
| *** ntt has joined #openstack-glance | 08:49 | |
| *** YifanZeng has quit IRC | 09:01 | |
| *** YifanZeng has joined #openstack-glance | 09:02 | |
| openstackgerrit | Ilya Menkov proposed openstack/glance: Refactoring of tests https://review.openstack.org/337530 | 09:03 |
| *** mfedosin has joined #openstack-glance | 09:04 | |
| *** flwang1 has joined #openstack-glance | 09:05 | |
| *** ducttape_ has joined #openstack-glance | 09:13 | |
| *** ducttape_ has quit IRC | 09:17 | |
| *** flwang1 has quit IRC | 09:20 | |
| openstackgerrit | Flavio Percoco proposed openstack/glance-specs: Don't deprecate `/file` endpoint https://review.openstack.org/313947 | 09:38 |
| *** giri has quit IRC | 10:08 | |
| *** ducttape_ has joined #openstack-glance | 10:13 | |
| *** ducttape_ has quit IRC | 10:18 | |
| *** btully has joined #openstack-glance | 10:27 | |
| *** btully has quit IRC | 10:31 | |
| *** sgotliv__ has joined #openstack-glance | 10:33 | |
| openstackgerrit | Niall Bunting proposed openstack/glance: Perform a cleanup of configuring.rst https://review.openstack.org/338861 | 10:39 |
| *** pdeore__ has joined #openstack-glance | 10:48 | |
| openstackgerrit | Kairat Kushaev proposed openstack/glance: [WIP][Glare] add tests for create and update artifacts https://review.openstack.org/338868 | 10:58 |
| *** ccesario has joined #openstack-glance | 11:06 | |
| openstackgerrit | Kairat Kushaev proposed openstack/glance: [WIP][Glare] add tests for create and update artifacts https://review.openstack.org/338868 | 11:14 |
| *** ducttape_ has joined #openstack-glance | 11:14 | |
| *** ducttape_ has quit IRC | 11:19 | |
| *** ayoung has quit IRC | 11:27 | |
| *** pdeore__ has quit IRC | 11:29 | |
| *** GB21 has joined #openstack-glance | 11:32 | |
| *** ntt has quit IRC | 11:39 | |
| *** ayoung has joined #openstack-glance | 11:40 | |
| openstackgerrit | Merged openstack/glance_store: Check that size is a number https://review.openstack.org/335074 | 11:42 |
| *** cdelatte has joined #openstack-glance | 11:48 | |
| *** ayoung has quit IRC | 11:48 | |
| *** rsjethani has joined #openstack-glance | 11:58 | |
| *** ducttape_ has joined #openstack-glance | 12:03 | |
| rsjethani | Hi jokke_ | 12:04 |
| *** ducttape_ has quit IRC | 12:04 | |
| rsjethani | The glance request id issue is pending from quite sometime | 12:05 |
| rsjethani | can you please take a look at latest developements | 12:05 |
| *** ducttape_ has joined #openstack-glance | 12:05 | |
| rsjethani | Latest: https://review.openstack.org/#/c/316052/ | 12:05 |
| openstackgerrit | Mike Fedosin proposed openstack/glance: [WIP] Refactoring of tests https://review.openstack.org/337530 | 12:05 |
| rsjethani | Previous patch if want to look at other approaches: https://review.openstack.org/#/c/261288 | 12:07 |
| *** itisha has joined #openstack-glance | 12:07 | |
| rsjethani | Hi Kairat__, flaper87 can you also take a look | 12:08 |
| rsjethani | hanks | 12:08 |
| rsjethani | Thanks | 12:08 |
| jokke_ | rsjethani: hi, will do | 12:13 |
| jokke_ | rsjethani: so what do you expect from us on that? It's in merge conflict and fails the gate (might be just the merge conflict) | 12:15 |
| rsjethani | jokke_: I am fixing it | 12:19 |
| rsjethani | just wanted to point out about the two approaches being taken here | 12:19 |
| *** gjayavelu has joined #openstack-glance | 12:20 | |
| *** mdavidson has quit IRC | 12:21 | |
| *** ratailor has quit IRC | 12:21 | |
| rsjethani | the latest patch is implementing Cao's suggestion but this approach has on disadvantage. For example the update call makes three requests to do its job but we will only get the last request id if we use this approach | 12:22 |
| *** buzztroll has joined #openstack-glance | 12:23 | |
| *** gjayavelu has quit IRC | 12:25 | |
| rsjethani | whereas in earlier approach https://review.openstack.org/#/c/261288/11 we do not have this disadvantage | 12:26 |
| jokke_ | rsjethani: right | 12:28 |
| *** ducttape_ has quit IRC | 12:28 | |
| jokke_ | rsjethani: I'm almost sorry to say, but I'm tempted to block the work on glanceclient until there is amendment to that specs specifying consistent approach how this will be done across the projects. There is clearly a gap in the design it even allowing this kind of problem to raise | 12:30 |
| *** mdavidson has joined #openstack-glance | 12:37 | |
| rsjethani | jokke_: kindly refer this ML thread: http://lists.openstack.org/pipermail/openstack-dev/2015-December/081842.html as to why we need a different approach for glance | 12:37 |
| *** abhishekk has joined #openstack-glance | 12:38 | |
| rsjethani | cinder, nova,neutron implementation is according to the cross project specs | 12:38 |
| jokke_ | rsjethani: ah, no wonder that discussion has dropped off my radar in the first place ... we normally keep all glance discussion tagged as [glace] not [python-glanceclient] so my filters did not catch that | 12:40 |
| rsjethani | glance implementation is according to this lite-spec: https://bugs.launchpad.net/glance/+bug/1525259 | 12:40 |
| openstack | Launchpad bug 1525259 in Glance "Add request_ids attribute to v2 schemas" [Wishlist,Triaged] - Assigned to Abhishek Kekane (abhishek-kekane) | 12:40 |
| openstackgerrit | Niall Bunting proposed openstack/glance: Perform a cleanup of configuring.rst https://review.openstack.org/338861 | 12:40 |
| *** wxy has joined #openstack-glance | 12:41 | |
| *** vikram has joined #openstack-glance | 12:44 | |
| vikram | Hi there | 12:44 |
| vikram | I want to upload an image using ReST APi | 12:44 |
| vikram | Can someone please point to an example | 12:44 |
| rsjethani | vikram: There is a minimal example in api reference page: http://developer.openstack.org/api-ref-image-v2.html#storeImageFile-v2 | 12:48 |
| openstackgerrit | Abhishek Kekane proposed openstack/python-glanceclient: WIP: Add request id to returned objects https://review.openstack.org/316052 | 12:53 |
| rsjethani | jokke_: the patch has been rebased | 12:54 |
| hemanthm | 3~/win 22 | 12:55 |
| *** links has quit IRC | 13:02 | |
| *** buzztroll has quit IRC | 13:04 | |
| *** pdeore__ has joined #openstack-glance | 13:04 | |
| rsjethani | jokke_: in summary patch https://review.openstack.org/#/c/316052/ implements Caos's approach whereas the patch https://review.openstack.org/#/c/261288 implements Stuart's appraoch | 13:04 |
| openstackgerrit | Niall Bunting proposed openstack/glance: Perform a cleanup of configuring.rst https://review.openstack.org/338861 | 13:07 |
| *** sdake has joined #openstack-glance | 13:07 | |
| jokke_ | rsjethani: I actually had responded to that mailchain ;) | 13:08 |
| *** ducttape_ has joined #openstack-glance | 13:12 | |
| *** ducttape_ has quit IRC | 13:13 | |
| *** M00nr41n has quit IRC | 13:15 | |
| *** GB21 has quit IRC | 13:18 | |
| *** mtanino has joined #openstack-glance | 13:19 | |
| jokke_ | rsjethani: just quick question: did I understand correctly that the code adds up all request IDs ... so when you are listing images it will provide the request ID for each page when paginating and add them to the list | 13:21 |
| jokke_ | ? | 13:21 |
| *** sdake_ has joined #openstack-glance | 13:23 | |
| rsjethani | yes | 13:25 |
| rsjethani | jokke_:yes | 13:25 |
| openstackgerrit | Darja Shakhray proposed openstack/glance: [WIP] Add schemas for artifacts https://review.openstack.org/334404 | 13:26 |
| jokke_ | I was afraid so, thanks | 13:26 |
| *** sdake has quit IRC | 13:27 | |
| *** takedakn has joined #openstack-glance | 13:29 | |
| *** abhishekk has quit IRC | 13:36 | |
| *** mtanino has quit IRC | 13:43 | |
| *** sigmavirus_away is now known as sigmavirus | 13:45 | |
| *** ducttape_ has joined #openstack-glance | 13:45 | |
| *** tsymanczyk has joined #openstack-glance | 13:50 | |
| *** buzztroll has joined #openstack-glance | 13:51 | |
| nikhil | Courtesy meeting reminder: ativelkov, cpallares, flaper87, flwang1, hemanthm, jokke_, kragniz, lakshmiS, mclaren, mfedosin, nikhil_k, Nikolay_St, Olena, pennerc, rosmaita, sigmavirus24, sabari, TravT, ajayaa, GB21, bpoulos, harshs, abhishek, bunting, dshakhray, wxy, dhellmann, kairat | 13:59 |
| openstackgerrit | Mike Fedosin proposed openstack/glance: [WIP] Glare https://review.openstack.org/292327 | 14:01 |
| *** rsjethani has quit IRC | 14:03 | |
| *** btully has joined #openstack-glance | 14:05 | |
| *** btully has quit IRC | 14:10 | |
| *** prateek has quit IRC | 14:20 | |
| *** mtanino has joined #openstack-glance | 14:28 | |
| *** burgerk has joined #openstack-glance | 14:43 | |
| *** ayoung has joined #openstack-glance | 14:48 | |
| *** takedakn has quit IRC | 14:50 | |
| *** pushkaru has joined #openstack-glance | 14:51 | |
| *** ametts has joined #openstack-glance | 14:51 | |
| *** groen692 has quit IRC | 14:58 | |
| *** rrecio has joined #openstack-glance | 14:59 | |
| nikhil | rosmaita: jokke_ sabari sigmavirus hemanthm wxy tsymanczyk bunting mfedosin kairat__ : so my findinds on critical concers on glare until now (something I haven't had time to test it thorough):- | 15:02 |
| * sigmavirus listens | 15:02 | |
| nikhil | 1. glare db will be separate table from glance images. there is potential conflict of primary key - we can never use images as artifacts. also, we need to add consitraint for that (didn't find in the spec). also, we need to make sure ops things it right when glare is deployed. | 15:03 |
| nikhil | 2. glare can't use the same user/config for swift (other backend stores) when uploading data. the current state of operations expect primary key to be assigned to the object. this could to data loss/corruption. | 15:04 |
| *** sskripnick1 has joined #openstack-glance | 15:05 | |
| kairat__ | 1. primary_key must be unique with artifact type | 15:05 |
| nikhil | 3. glare API needs to be service API only -- something we discussed earlier but the app-catalog priority makes it confusion. the large scale practice we are pushing towards defcore and tc for avoiding direct streaming of data to glare API makes it non-adherernt to what glance is proposing for upload mechanisms. | 15:06 |
| kairat__ | so I think there will be no dupes | 15:06 |
| mfedosin | yes, uniqueness is artifact_type_name + id | 15:06 |
| jokke_ | nikhil: 1. images as artifacts would need migration between the tables and solving possible conflicts. Not pretty but possible, I don't personally see that as blocker 2. this needs to be documented well, again don't see as blocker | 15:06 |
| kairat__ | ++ to jokke_ for 2 | 15:07 |
| mfedosin | glare has different config from glance | 15:07 |
| jokke_ | what comes to 3. I think I have missed some part of the discussion ... I understood that the glare API had 2 parts. The service api and the public api | 15:08 |
| mfedosin | so, documentaion about swift is enough | 15:08 |
| jokke_ | mfedosin: ++ makes life lots easier when it is it's own service :D | 15:08 |
| jokke_ | and not sharing config files | 15:08 |
| mfedosin | jokke_: yeah, much better | 15:09 |
| nikhil | 4. dependency management isn't clear in the spec. we can't expect external links in the glare blobs or we break the glance immutability construct. | 15:09 |
| mfedosin | dependencies are soft | 15:10 |
| kairat__ | so we dependency are much like linux | 15:10 |
| mfedosin | we'll add hard dependencies in the future | 15:10 |
| kairat__ | symbolic link and hard links | 15:10 |
| mfedosin | kairat__: ++ | 15:10 |
| jokke_ | nikhil: 4. I specifically raised the immutability point in the spec review. I think needs clarification and clear documentation but glance image immutability and glare blob/dependency immutability are totally different discussions | 15:10 |
| kairat__ | soft behaves like symbolic link | 15:10 |
| mfedosin | now they are symbolic | 15:10 |
| kairat__ | BTW, hard dependencies was asked by folks | 15:11 |
| kairat__ | but they will be implemented later | 15:11 |
| nikhil | 5. upgrade path: how oslo.vo will affect them and how we will handle a change in the type scheme when object definition is asked to be changed. | 15:13 |
| *** vikram has quit IRC | 15:13 | |
| nikhil | (I can't remember 6 & 7 atm) | 15:13 |
| *** sdake_ has quit IRC | 15:13 | |
| *** sdake has joined #openstack-glance | 15:14 | |
| *** pumarani__ has joined #openstack-glance | 15:14 | |
| *** pushkaru has quit IRC | 15:14 | |
| nikhil | rosmaita: even now you think that people should expect all the info from spec? I am asking sincerely because I'm not the right judge for this and people have complained to me offline. | 15:15 |
| mfedosin | 5. Stan asked me about it in the spec | 15:15 |
| *** rrecio_ has joined #openstack-glance | 15:17 | |
| kairat__ | nikhil, so they can go to spec and ask questions | 15:17 |
| *** rrecio has quit IRC | 15:17 | |
| rosmaita | nikhil: am in searchlight meeting, i will read back the discussion in a few min | 15:17 |
| kairat__ | We will be welcome to answer | 15:17 |
| nikhil | rosmaita: k | 15:17 |
| jokke_ | kairat__: ++ the spec is still in review ... I would assume people bringing their concerns up there, not via back channels and complain later | 15:17 |
| nikhil | oh yes, the most important one: | 15:18 |
| mfedosin | 5. http://paste.openstack.org/show/526982/ | 15:18 |
| nikhil | we discussed glare to be a common service for images, volumes, baremeta images, etc. at the summit and with the current direction, priorities and architecture that seems improbable (API requirement for images, dependency management, etc.). so, the concept of glare to be a generic catalog isn't completely accomplished but it's too late to change that now. | 15:19 |
| kairat__ | Hm, could you please elaborate? | 15:20 |
| kairat__ | from what I see now Glare is catalog | 15:20 |
| kairat__ | and it is designed to be safe catalog items storage for all services | 15:21 |
| *** david-lyle__ has joined #openstack-glance | 15:21 | |
| kairat__ | unlike Glance because Glance tries to satisfy two goals | 15:21 |
| nikhil | kairat__: that's what I am trying to tell above, it's not designed to be compat with glance | 15:21 |
| kairat__ | or two areas: image catalog and image service | 15:22 |
| nikhil | compatibility is flawed at fundamental level | 15:22 |
| nikhil | (and I was not involved in _any_ of those design decisions when such approach was taken, mostly my worry and reason for sadness) | 15:23 |
| kairat__ | I still think we can do that | 15:23 |
| kairat__ | but it was not our focus since we decided not to implement af type for images | 15:23 |
| kairat__ | IIUC | 15:24 |
| kairat__ | we need translation level at middleware | 15:24 |
| kairat__ | and adapter to DB | 15:24 |
| *** david-lyle__ is now known as david-lyle | 15:24 | |
| nikhil | I don't understand all that | 15:24 |
| kairat__ | so | 15:24 |
| kairat__ | tecnhnically it is possible | 15:25 |
| kairat__ | but there is no interest in that | 15:25 |
| nikhil | (I think you mean theoratically) | 15:25 |
| kairat__ | and we decides not to implement artifact type for images | 15:25 |
| kairat__ | nikhil, i guess you participated in this discussion | 15:25 |
| nikhil | (we decided not now) | 15:25 |
| nikhil | (in the midcycle we decided that we won't use glare for images) | 15:26 |
| kairat__ | so we had image artifact tyoe | 15:27 |
| kairat__ | and it worked | 15:27 |
| kairat__ | but we deleted it because of mid-cycle | 15:27 |
| kairat__ | so not only theoretically | 15:27 |
| kairat__ | but of course there can be troubles | 15:27 |
| nikhil | exactly, it's theoretically possible but not in practice | 15:28 |
| kairat__ | But I think Glare is pretty-well satisfies catalog goals | 15:28 |
| nikhil | firstly, no one from glare team has participated in defcore discussions | 15:28 |
| kairat__ | Sorry, but mfedosin participated in meetings | 15:29 |
| kairat__ | so it is better to ask him about that | 15:29 |
| * kairat__ didn't participate, true=) | 15:29 | |
| nikhil | also, some of the latest requests suggest that there is difference in understanding on what API (specifically upload API) needs to be for cataloging | 15:30 |
| nikhil | there's difference in understanding the implications of implementing a client | 15:30 |
| nikhil | I'm waiting for someone from glare to lead and come to consensus because the energy involved in those discussions hasn't been productive for me. | 15:32 |
| kairat__ | nikhil, so your feedback about that would be welcome, that's why we asked you to review | 15:32 |
| nikhil | kairat__: yeah, these are things I think are beyond the spec which talks mostly about API | 15:32 |
| nikhil | anyway, I will give my review before monday. | 15:33 |
| kairat__ | so that about up;oading | 15:33 |
| kairat__ | what the concerns here? | 15:33 |
| nikhil | (I'm just a bit disappointed on what the vision of glare was and what it is today) | 15:33 |
| mfedosin | nikhil: glare always was a catalog (or repository) of artifacts | 15:34 |
| kairat__ | nikhil, I tried to respond to all your concerns. so if you have any other points I (and mfedosin ) would be happy to share and understand them | 15:35 |
| kairat__ | *concerns above | 15:35 |
| nikhil | kairat__: unfortunately, each of these points I raise are not a matter of irc discussions because we can expect a lot of back and forth; especially upload because the defcore discussions are plenty. I can schedule a separate meeting with you guys if you are interested in learning. | 15:35 |
| mfedosin | what learning? | 15:36 |
| nikhil | kairat__: mfedosin : like I said, glare isn't a possible/recommended alternative for glance deployment for the above reasons. | 15:36 |
| kairat__ | ++ to mfedosin | 15:36 |
| bunting | mfedosin: Can I ask a question, why don't you move Glare into it's own project? | 15:36 |
| mfedosin | bunting: good question :) | 15:37 |
| nikhil | kairat__: do you know why we are having such long and complicated discussions with defcore? | 15:38 |
| bunting | mfedosin: I have outlined a few arguments for this on the spec | 15:38 |
| mfedosin | "Would an alternative be to create a new project to do this?" lol | 15:39 |
| *** bpokorny has joined #openstack-glance | 15:39 | |
| * nikhil leaves irc and reads comments on spec | 15:39 | |
| kairat__ | nikhil, could you please elaborate? | 15:40 |
| kairat__ | so yeah | 15:40 |
| * kairat__ goes to dinner=) | 15:40 | |
| *** GB21 has joined #openstack-glance | 15:42 | |
| *** TravT_ is now known as TravT | 15:43 | |
| *** tshefi has quit IRC | 15:44 | |
| *** gjayavelu has joined #openstack-glance | 15:44 | |
| *** sdake has quit IRC | 15:48 | |
| *** pcaruana has quit IRC | 15:49 | |
| *** sdake has joined #openstack-glance | 15:51 | |
| *** pumarani__ has quit IRC | 15:53 | |
| *** dharinic_ has joined #openstack-glance | 15:56 | |
| openstackgerrit | Kairat Kushaev proposed openstack/glance: [WIP][Glare] add tests for create and update artifacts https://review.openstack.org/338868 | 16:00 |
| openstackgerrit | Dharini Chandrasekar proposed openstack/glance: Improving help text for context middleware opts https://review.openstack.org/335671 | 16:05 |
| *** sdake has quit IRC | 16:09 | |
| *** rcernin has quit IRC | 16:10 | |
| *** buzztroll has quit IRC | 16:12 | |
| *** gjayavelu has quit IRC | 16:17 | |
| *** gjayavelu has joined #openstack-glance | 16:18 | |
| *** buzztroll has joined #openstack-glance | 16:18 | |
| *** sdake has joined #openstack-glance | 16:20 | |
| *** M00nr41n has joined #openstack-glance | 16:21 | |
| *** tesseract- has quit IRC | 16:23 | |
| *** sdake has quit IRC | 16:25 | |
| *** sdake has joined #openstack-glance | 16:27 | |
| *** ducttape_ has quit IRC | 16:29 | |
| *** julim has quit IRC | 16:38 | |
| *** julim has joined #openstack-glance | 16:38 | |
| *** bpokorny has quit IRC | 16:38 | |
| *** julim has quit IRC | 16:42 | |
| *** bpokorny has joined #openstack-glance | 16:51 | |
| *** cdelatte has quit IRC | 16:58 | |
| *** gjayavelu has quit IRC | 17:00 | |
| *** mick_m has joined #openstack-glance | 17:02 | |
| *** tsymanczyk has quit IRC | 17:02 | |
| *** wxy has quit IRC | 17:11 | |
| *** sdake has quit IRC | 17:11 | |
| *** ducttape_ has joined #openstack-glance | 17:11 | |
| *** ducttape_ has quit IRC | 17:15 | |
| openstackgerrit | Marc Abramowitz proposed openstack/glance: replicator: dump: Display more info https://review.openstack.org/339146 | 17:16 |
| rosmaita | nikhil: just realized i never answered your question ... i think you raise excellent points about some issues that need to be settled in glare implementation ... but as far as people who want to start designing stuff to use glare, i think the API spec pretty much gives a complete rundown of the public interface | 17:16 |
| *** ducttape_ has joined #openstack-glance | 17:18 | |
| bunting | rosmaita: What do you think about moving glare into a separate project? | 17:19 |
| rosmaita | bunting: not sure, it's been discussed before, very seriously, and decided to keep it as part of glance | 17:20 |
| *** julim has joined #openstack-glance | 17:20 | |
| rosmaita | i've always thought of artifacts as the next generation of images, so glare seems a natural fit in glance | 17:20 |
| rosmaita | but, if your goal is simplicity of deployment, maybe not | 17:21 |
| *** julim has quit IRC | 17:21 | |
| kragniz | bunting: don't start that one again :P | 17:21 |
| rosmaita | kragniz: you were there in vancouver, i think? | 17:21 |
| kragniz | yeah | 17:21 |
| bunting | That would be why i missed it then | 17:22 |
| bunting | From what I understand the idea was not as concrete as it was now tohugh? | 17:22 |
| *** julim has joined #openstack-glance | 17:23 | |
| rosmaita | well, not sure about that ... all i remember is that before we had the future of glance session, i was all for keeping artifacts in glance and stuart was against it ... and after, i thought no reason to retain artifacts in glance, and stuart wanted to keep them! | 17:23 |
| *** GB21 has quit IRC | 17:23 | |
| kragniz | haha, I'd forgotten about that | 17:24 |
| rosmaita | bunting: https://etherpad.openstack.org/p/liberty-what-is-glance | 17:24 |
| bunting | I think that first line probably highlights the problem in glance | 17:25 |
| rosmaita | at that session, TravT announced that the experimental search api would become searchlight | 17:26 |
| bunting | Image service vs data assets | 17:26 |
| rosmaita | i think that was a good move, it's been able to move much faster as its own project | 17:26 |
| rosmaita | bunting: i can't decide if it's a problem or an opportunity ... a "data asset" seems like a slight generalization of an image | 17:29 |
| bunting | It is a bit ambiguous | 17:30 |
| *** sdake has joined #openstack-glance | 17:32 | |
| *** sdake_ has joined #openstack-glance | 17:34 | |
| *** sdake has quit IRC | 17:37 | |
| openstackgerrit | Marc Abramowitz proposed openstack/glance: Add support for using os_client_config cloud names https://review.openstack.org/339155 | 17:39 |
| openstackgerrit | Marc Abramowitz proposed openstack/glance: Add support for using os_client_config cloud names https://review.openstack.org/339155 | 17:41 |
| *** btully has joined #openstack-glance | 17:42 | |
| *** btully has quit IRC | 17:46 | |
| *** M00nr41n has quit IRC | 17:47 | |
| *** sgotliv__ has quit IRC | 17:50 | |
| *** catintheroof has joined #openstack-glance | 17:53 | |
| *** ducttape_ has quit IRC | 17:53 | |
| openstackgerrit | Marc Abramowitz proposed openstack/glance: replicator: livecopy: Display more info https://review.openstack.org/339161 | 17:54 |
| *** M00nr41n has joined #openstack-glance | 17:54 | |
| *** M00nr41n has quit IRC | 17:55 | |
| *** sdake_ has quit IRC | 17:57 | |
| openstackgerrit | Marc Abramowitz proposed openstack/glance: replicator: livecopy: Display more info https://review.openstack.org/339161 | 17:59 |
| *** gjayavelu has joined #openstack-glance | 18:01 | |
| *** david-lyle has quit IRC | 18:04 | |
| *** ducttape_ has joined #openstack-glance | 18:05 | |
| *** gjayavelu has quit IRC | 18:05 | |
| *** ducttape_ has quit IRC | 18:05 | |
| *** david-lyle has joined #openstack-glance | 18:06 | |
| *** itisha has quit IRC | 18:10 | |
| *** TravT_ has joined #openstack-glance | 18:15 | |
| *** gjayavelu has joined #openstack-glance | 18:17 | |
| *** prateek has joined #openstack-glance | 18:17 | |
| *** gjayavelu has quit IRC | 18:17 | |
| *** TravT has quit IRC | 18:19 | |
| *** sskripnick1 has quit IRC | 18:21 | |
| *** TravT_ has quit IRC | 18:21 | |
| *** buzztroll has quit IRC | 18:21 | |
| *** buzztroll has joined #openstack-glance | 18:26 | |
| nikhil | rosmaita: I am not against the API spec in itself but I think the expectations seems rather distinct. the proposal is that if glare API merges then the glare code is okay to merge and that seems like the idea being floated. this is where we as a community fail to operate -- we've set a expectation for a spec for an API but a commit like this can't claim to | 18:38 |
| nikhil | have "implements glare api spec" like things in commit message. https://review.openstack.org/#/c/330456/ | 18:38 |
| nikhil | we can +2W the "what" of glare but we've failed to have consensus on "why" and "how" | 18:38 |
| *** rcernin has joined #openstack-glance | 18:39 | |
| nikhil | so any claim/request to merge the glare code based on approval for the glare API spec approval would be totally redundant. | 18:39 |
| nikhil | following up on the comments from bunting and kragniz , I can say that the idea of artifacts as future generation/generalization of images is failing and that's my real argument or the ask from community to think things through before pushing blindly on the spec and code merge. | 18:40 |
| nikhil | since liberty, we've divereged on the idea of artifacts, the architecture of implementation, the priorities of the features/aspects, the people involved and their approach towards this goal, the vision of glance being central information store and it's practical possiblity, the discussions on how we are planning to achieve that, and the considerate | 18:42 |
| nikhil | expectations accorindly. | 18:42 |
| rosmaita | nikhil: you make good points, and i don't disagree | 18:42 |
| rosmaita | you're managing multiple projects at this point | 18:43 |
| openstackgerrit | Sirushti Murugesan proposed openstack/python-glanceclient: py3: Fix encoding and use sys.stdin.buffer https://review.openstack.org/331417 | 18:43 |
| rosmaita | and i think you are right about a communication problem and expectations | 18:43 |
| rosmaita | my view is that the api spec is proposing an interface but not an implementation | 18:43 |
| rosmaita | but there are implementation issues involved, so i guess there's not really a clean separation | 18:44 |
| rosmaita | and i guess in the long run, people don't want to start developing against api contracts that are "only" in a spec | 18:44 |
| *** dharinic_ has quit IRC | 18:45 | |
| nikhil | rosmaita: I didn't get the last part of 'only in a spec' | 18:47 |
| nikhil | rosmaita: only in spec vs. in the ap-ref doc ? | 18:47 |
| *** julim has quit IRC | 18:51 | |
| rosmaita | nikhil: i meant "only" because it's always possible that something might have to change during implementation | 18:52 |
| *** itisha has joined #openstack-glance | 18:53 | |
| rosmaita | i'm not sure how much people trust specs | 18:53 |
| rosmaita | i mean people in other projects trusting specs from different projects | 18:53 |
| rosmaita | i think we trust our own specs! | 18:53 |
| nikhil | rosmaita: agreed and I hope our community follow lite-spec process to understand if it is ship of theseus. | 18:54 |
| nikhil | *keeps following | 18:54 |
| nikhil | *understand and communicate | 18:55 |
| nikhil | rosmaita: also, one very important issue, to tell the truth -- something I've tried to shield our team from (and good time to reveal here): we've been 'blamed' to have been developing and encouraging development of glare as a 'proprietary project' and something that's dumped upstream rather being developed openly. This breaks our the governance charter set | 18:55 |
| nikhil | by TC. I've made quite a few attempts (as per my obligations and more from my side as professional courtesy) to ensure we don't fall under that radar again. While the API spec has good collaboration but the community around glare has failed. This leads to some serious issues and as a mature project as well as a established and recognized community, we want | 18:55 |
| nikhil | to avoid more of such feedback. Since, this was a highly negative comments, people have given it privately rather than directing upstream. But that's the reason why I keep saying people give feedback privately as it has a lot more to do than simple an API to consume. | 18:55 |
| rosmaita | nikhil: wow, that is disappointing criticism given how much discussion over artifacts has occurred since icehouse? certainly juno | 18:57 |
| *** rrecio_ has quit IRC | 18:57 | |
| nikhil | rosmaita: yeah, but glare as far as community is concerned is totally new thing. a cost we are paying for marketing it like that. there's more to it, because artifacts (glance v3) is very very different from currently proposal glare architecture which people are failing to understand the 'whys'. | 18:58 |
| *** rrecio has joined #openstack-glance | 18:58 | |
| nikhil | (FWIW, I was against a new name, separate service, etc. from the beginning) | 18:59 |
| *** mfedosin has quit IRC | 19:00 | |
| rosmaita | well, the separate service was a compromise for operators who only wanted to deal with images (at least that's what i remember) | 19:01 |
| nikhil | yeah, it just has adverse consequences like these. | 19:01 |
| * nikhil finds a failure in establishing consensus at multiple fronts rather than simply an approval, direction or concept of the glare api spec. | 19:05 | |
| *** prateek has quit IRC | 19:12 | |
| *** dramakri has quit IRC | 19:18 | |
| *** ametts has quit IRC | 19:23 | |
| *** cdelatte has joined #openstack-glance | 19:25 | |
| *** flwang1 has joined #openstack-glance | 19:29 | |
| *** btully has joined #openstack-glance | 19:31 | |
| *** TravT has joined #openstack-glance | 19:39 | |
| *** aslaen has joined #openstack-glance | 19:40 | |
| *** flwang1 has quit IRC | 19:48 | |
| *** mfedosin has joined #openstack-glance | 20:05 | |
| *** aslaen has left #openstack-glance | 20:05 | |
| *** aslaen has joined #openstack-glance | 20:19 | |
| *** aslaen has joined #openstack-glance | 20:19 | |
| *** aslaen has quit IRC | 20:20 | |
| *** ducttape_ has joined #openstack-glance | 20:41 | |
| *** pdeore__ has quit IRC | 20:41 | |
| *** hemanthm is now known as hemanthm|afk | 20:45 | |
| *** buzztroll has quit IRC | 20:59 | |
| openstackgerrit | Brian Rosmaita proposed openstack/glance: WADL to RST migration (part 2 - images) https://review.openstack.org/332950 | 21:06 |
| *** buzztroll has joined #openstack-glance | 21:08 | |
| *** pt_15 has joined #openstack-glance | 21:12 | |
| *** mtanino has quit IRC | 21:14 | |
| *** rcernin has quit IRC | 21:17 | |
| *** agireud has quit IRC | 21:20 | |
| *** david-lyle has quit IRC | 21:21 | |
| *** david-lyle has joined #openstack-glance | 21:23 | |
| *** flwang1 has joined #openstack-glance | 21:28 | |
| *** burgerk has quit IRC | 21:30 | |
| *** ayoung has quit IRC | 21:34 | |
| *** ozialien10 has joined #openstack-glance | 21:35 | |
| *** mick_m has quit IRC | 21:55 | |
| *** btully has quit IRC | 22:02 | |
| *** darrenc is now known as darrenc_afk | 22:06 | |
| *** david-lyle has quit IRC | 22:07 | |
| *** TravT has quit IRC | 22:15 | |
| *** ducttape_ has quit IRC | 22:17 | |
| openstackgerrit | Marc Abramowitz proposed openstack/glance: replicator: livecopy: Display more info https://review.openstack.org/339161 | 22:20 |
| *** sigmavirus is now known as sigmavirus_away | 22:20 | |
| *** mfedosin has quit IRC | 22:21 | |
| *** ozialien10 has quit IRC | 22:22 | |
| openstackgerrit | Marc Abramowitz proposed openstack/glance: Add support for using os_client_config cloud names https://review.openstack.org/339155 | 22:24 |
| *** TravT has joined #openstack-glance | 22:26 | |
| *** darrenc_afk is now known as darrenc | 22:26 | |
| *** catintheroof has quit IRC | 22:40 | |
| openstackgerrit | Marc Abramowitz proposed openstack/glance: replicator: dump: Display more info https://review.openstack.org/339146 | 22:42 |
| *** julim has joined #openstack-glance | 22:44 | |
| openstackgerrit | Akhila Kishore proposed openstack/glance: Nitpick spell change https://review.openstack.org/338594 | 22:50 |
| *** bpokorny_ has joined #openstack-glance | 22:58 | |
| *** buzztroll has quit IRC | 22:58 | |
| *** bpokorny has quit IRC | 23:02 | |
| openstackgerrit | Akhila Kishore proposed openstack/glance: Nitpick spell change https://review.openstack.org/338594 | 23:02 |
| *** buzztroll has joined #openstack-glance | 23:06 | |
| *** flwang has joined #openstack-glance | 23:14 | |
| flwang | rosmaita: any luck you around? | 23:14 |
| flwang | re the location strategy ristrict | 23:14 |
| *** julim has quit IRC | 23:27 | |
| *** julim has joined #openstack-glance | 23:32 | |
| *** ayoung has joined #openstack-glance | 23:36 | |
| openstackgerrit | Sabari proposed openstack/glance_store: Remove unused variable in vmware store https://review.openstack.org/339277 | 23:39 |
| *** ducttape_ has joined #openstack-glance | 23:46 | |
| *** btully has joined #openstack-glance | 23:47 | |
| openstackgerrit | qinchunhua proposed openstack/glance: Correct reraising of exception https://review.openstack.org/338040 | 23:48 |
| *** btully has quit IRC | 23:52 | |
| *** rrecio has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!