| *** medberry is now known as med_out | 00:29 | |
| *** heckj has quit IRC | 00:36 | |
| *** jakedahn has quit IRC | 00:56 | |
| *** danwent_ has joined #openstack-meeting | 00:57 | |
| *** danwent_ has quit IRC | 01:01 | |
| *** salv has quit IRC | 01:10 | |
| *** littleidea has joined #openstack-meeting | 01:23 | |
| *** dragondm has quit IRC | 01:33 | |
| *** jakedahn has joined #openstack-meeting | 01:47 | |
| *** Cyns has joined #openstack-meeting | 02:31 | |
| *** Cyns has quit IRC | 02:31 | |
| *** ewanmellor has quit IRC | 02:33 | |
| *** martine_ has joined #openstack-meeting | 03:29 | |
| *** shwetaap has quit IRC | 03:37 | |
| *** martine_ has quit IRC | 03:43 | |
| *** dolphm_ has joined #openstack-meeting | 03:51 | |
| *** dolphm_ has quit IRC | 03:52 | |
| *** littleidea has quit IRC | 04:20 | |
| *** adjohn has joined #openstack-meeting | 06:01 | |
| *** termie has quit IRC | 06:01 | |
| *** ke4qqq has quit IRC | 06:02 | |
| *** termie has joined #openstack-meeting | 06:02 | |
| *** ke4qqq has joined #openstack-meeting | 06:02 | |
| *** royh has joined #openstack-meeting | 06:57 | |
| *** ewanmellor has joined #openstack-meeting | 07:07 | |
| *** ewanmellor has quit IRC | 07:14 | |
| *** darraghb has joined #openstack-meeting | 08:58 | |
| *** yamahata has joined #openstack-meeting | 10:49 | |
| jaypipes | yamahata: morning/evening :) | 11:27 |
|---|---|---|
| yamahata | jaypipes, morning | 11:40 |
| jaypipes | yamahata: so, I got confirmation last night that s1rp, bcwaldon, and I think jkoelker will be here (in 15 mins) | 11:44 |
| *** cbeck has quit IRC | 11:45 | |
| yamahata | jaypipes, Great. Thank you for arranging the meeting | 11:46 |
| *** cbeck has joined #openstack-meeting | 11:46 | |
| jaypipes | yamahata: no problem :) sorry it took so long to set it up! | 11:47 |
| blamar | jaypipes: morning | 11:52 |
| blamar | yamahata: evening :) | 11:52 |
| jaypipes | blamar: morning! :) | 11:53 |
| *** bcwaldon has joined #openstack-meeting | 11:53 | |
| jaypipes | bcwaldon: morning :) | 11:53 |
| bcwaldon | jaypipes: hey there | 11:53 |
| blamar | bcwaldon, thanks for the coffee | 11:53 |
| bcwaldon | bcwaldon: you're welcome? | 11:53 |
| bcwaldon | ha | 11:54 |
| bcwaldon | blamar ^ | 11:54 |
| jaypipes | bcwaldon: too early :) | 11:54 |
| bcwaldon | exactly | 11:54 |
| jaypipes | hehe | 11:54 |
| bcwaldon | I don't envy CDT right now (s1rp) | 11:55 |
| jaypipes | bcwaldon: indeed. | 11:56 |
| jaypipes | all: I'm keeping notes here: http://etherpad.openstack.org/StructuredImageMetadata | 11:57 |
| bcwaldon | jaypipes: discuss here, notes there? I don't want to mess it up, it looks so pretty in one color ;) | 11:58 |
| jaypipes | bcwaldon: hehe, no worries. what would folks prefer? we can discuss here or there. I just wanted to make sure we took notes on everything... | 11:59 |
| bcwaldon | Let's discuss here, kinda unclear if all through etherpad. We can follow up there | 12:00 |
| jaypipes | yamahata has done a good job summarizing the discussion so far; that's what I pasted in that etherpad. | 12:00 |
| bcwaldon | awesome | 12:00 |
| s1rp | morning | 12:01 |
| bcwaldon | hey there, s1rp | 12:01 |
| yamahata | s1rp, morning | 12:01 |
| jaypipes | so, it sounds like yamahata doesn't really care which approach is taken (he just wants to get it done :) ). I'm on the side of natural filters/Atom subresource. s1rp is on the side of JSON blob in existing custom properties. | 12:01 |
| jaypipes | bcwaldon, blamar: just off the cuff, where do you guys stand? | 12:02 |
| bcwaldon | I'd prefer anything generic over specific. Give me a sec and I'll look at the Atom idea. I also had the thought that we could wait until we break apart the image mover and registry, then we get complex properties for free | 12:03 |
| bcwaldon | ...since we don't have to send data through headers | 12:03 |
| jaypipes | bcwaldon: could you explain that last part a bit more? | 12:03 |
| jaypipes | bcwaldon: ah, yes. | 12:03 |
| jaypipes | bcwaldon: that thought occurred to me as well. | 12:03 |
| s1rp | filtering is still complex though | 12:03 |
| jaypipes | s1rp: right. | 12:03 |
| bcwaldon | definitely, I think it will have to be complex | 12:04 |
| s1rp | is filtering arbitrarily nested data-structures a hard requirement though? | 12:04 |
| jaypipes | s1rp: I like the "naturalness" of an API like /images/<ID>/manifest/devices/ | 12:04 |
| blamar | jaypipes: is it really the purview of glance to handle that though? | 12:05 |
| jaypipes | blamar: I hear you. | 12:05 |
| jaypipes | blamar: not sure. that's why we're having this discussion ;) | 12:05 |
| *** chuck_ has joined #openstack-meeting | 12:05 | |
| jaypipes | blamar: put another way... if Glance does *not* handle that kind of information, what does? | 12:06 |
| blamar | jaypipes: yeah, I'm still wondering what the *ideal* place for this data | 12:06 |
| bcwaldon | Honestly, this feels like a horribly complex feature that may not be top priority right now. Would it be crazy to suggest we use json blobs and ignore filtering for now? | 12:06 |
| *** chuck_ has quit IRC | 12:06 | |
| blamar | jaypipes: I deal in ideals a lot, and then try and work my way towards that slowly | 12:06 |
| *** chuck_ has joined #openstack-meeting | 12:06 | |
| *** zul has quit IRC | 12:07 | |
| *** chuck_ is now known as zul | 12:07 | |
| *** zul has joined #openstack-meeting | 12:07 | |
| jaypipes | yamahata: what's your opinion on what is more important: a) ability to easily search for images that have a specific block-device-mapping through the API or b) just storing the information so that it is attached in output of GET /images/<ID>? | 12:08 |
| yamahata | b) | 12:08 |
| jaypipes | ok, that's a good guide. | 12:08 |
| jaypipes | so, if b) is the case, I think the natural solution would be the more generic one -- i.e. we shove a JSON blob into image_properties with a key of 'manifest' (or similar)? | 12:09 |
| yamahata | What I wanted is b. When I implemented, I generalized it a bit. | 12:09 |
| s1rp | if only some of the structured metadata needs to be filtered on, we could denormalize it: have structured metdata stored as a blob, and split out any thing we need to filter on as separate metdata properties | 12:09 |
| jaypipes | s1rp: good point. | 12:09 |
| s1rp | originally when we were thinking of OVF this was an idea... basically read the whole manifest in as a JSON or XML blob... keep that around if needed, but anything we *really* use would be duplicated in easy to access fields like `os_type` | 12:10 |
| *** jkoelker has joined #openstack-meeting | 12:10 | |
| jaypipes | yeah | 12:10 |
| s1rp | that provides filtering, but doesn't lose the structured aspect, allowing us to recreate the OVF manifest if needed | 12:11 |
| jaypipes | s1rp: so, given yamahata's use case (priority b) ), do you think that bcwaldon's point above about the split of move and registry in the API should be prioritized? | 12:11 |
| jaypipes | s1rp: in other words, splitting the mover and registry in the APIs would essentially solve yamahata's main use case, as complex JSON could be returned in the call to the registry GET /images/<ID> that would not need to be converted into headers... | 12:12 |
| s1rp | jaypipes: yeah, that may be the simplest/fastest way to get him what he needs | 12:12 |
| s1rp | it has the bonus of also getting us closer to where need to be as well | 12:12 |
| jaypipes | yamahata: let me give you some background... so this makes more sense :) | 12:12 |
| yamahata | jaypipes, yes please. In fact I lost about spliting mover and registry. | 12:13 |
| jaypipes | yamahata: so, we had a conversation a couple weeks ago about the future of the Glance architecture and API. One problem we've run into by having a Swift-style API (one that attached metadata about an object/image in HTTP headers) is that a) you are limited in the amount of metadata you can put in HTTP headers and b) you cannot cache static image data separately from non-static image metadata... | 12:14 |
| jaypipes | yamahata: so we discussed splitting out the unified Glance API into a "mover" API node and a registry API node, with the mover API returning the image file itself on GET /images/<ID> and the registry returning a serialized representation of the image's metadata on GET /images/<ID> | 12:15 |
| jaypipes | yamahata: and we would get rid of the API calls to HEAD /images/<ID> and the use of HTTP headers to store image metadata. | 12:16 |
| jaypipes | yamahata: this actually won't be all that difficult to do, since, as you know, the registry is already a separate server daemon and the unified API server simply speaks to the registry through the registry client. | 12:17 |
| jaypipes | yamahata: however, this would of course break backwards compatibility, to we would need to bump the Glance API to a v2.0. | 12:17 |
| jaypipes | yamahata: let me know if this makes sense or if you need more explanation. | 12:17 |
| yamahata | I see. So we want metadata to be 1st class citizon from http header. | 12:18 |
| yamahata | That makes much sense. | 12:18 |
| yamahata | API compatibility would be an issue. | 12:18 |
| jaypipes | yamahata: yes, we want no more metadata in HTTP headers. and the glance client would speak directly to the registry node (instead of the API node) to get metadata about an image. | 12:19 |
| jaypipes | yamahata: so... due to the API compatibility thing, we've been wondering when a "good" time to do this would be? ;) | 12:19 |
| jaypipes | s1rp, bcwaldon, blamar: thoughts? | 12:20 |
| s1rp | s1rp: if that approach will work for yamahata (i can't think of a reason it wouldn't), i'd favor it | 12:21 |
| s1rp | heh, talking to myself (it's early!) | 12:21 |
| yamahata | the API would be implemented in server code. But when for nova to use. | 12:21 |
| jaypipes | yamahata: Nova would continue to communicate with Glance via the glance client. | 12:21 |
| bcwaldon | jaypipes: I'd say let's break it up asap | 12:22 |
| jaypipes | yamahata: so most of the code changes will be around the client communicating with both a registry and a mover node directly (instead of only talking to the API node as it currently does) | 12:22 |
| yamahata | jaypipes, right. | 12:22 |
| jaypipes | bcwaldon, s1rp: what kind of timeframe do you think we need to copmlete this? | 12:23 |
| jaypipes | blamar: want to hear from you on this, too.. | 12:23 |
| blamar | jaypipes: no worries, I'm tired and sitting next to bcwaldon so I'm letting him do the talking | 12:23 |
| jaypipes | blamar: :) | 12:23 |
| blamar | jaypipes: really, whatever he says is just me dictating | 12:23 |
| bcwaldon | I'd say it could absolutely get done for d4. I could take a look at it this evening | 12:23 |
| jaypipes | lol | 12:24 |
| bcwaldon | blamar: excuse me? | 12:24 |
| s1rp | jaypipes: famous last words, but i don't think it would be too difficult | 12:24 |
| bcwaldon | its not complex, there would just be a lot of things to nail down | 12:24 |
| s1rp | it's going to be time consuming testing end-to-end... | 12:24 |
| jaypipes | s1rp: ok. well clearly, this would block all other work that touched the API. | 12:24 |
| s1rp | yea | 12:25 |
| bcwaldon | jaypipes: what else is going on with the api? | 12:25 |
| jaypipes | should we put this as a proposal to the community? | 12:25 |
| bcwaldon | At least say it's happening ;) | 12:25 |
| jaypipes | bcwaldon: authentication is ongoing | 12:25 |
| * jaypipes will type up a draft proposal for what we want to do | 12:25 | |
| * jaypipes will email all people here to review the draft | 12:26 | |
| bcwaldon | sounds awesome | 12:26 |
| s1rp | cool | 12:26 |
| yamahata | great | 12:26 |
| jaypipes | Once good, we'll ask not for suggestions, but more as a notification of what is coming? | 12:26 |
| jaypipes | in other words, a warning vs. a proposal? | 12:26 |
| bcwaldon | probably a good direction | 12:26 |
| s1rp | ++ | 12:26 |
| jaypipes | otherwise, the discussion could seriously derail... | 12:27 |
| bcwaldon | we all know how that goes | 12:27 |
| s1rp | yeah, really worried about drawing this out... | 12:27 |
| s1rp | we know we need to do this for a bunch of reasons (caching, single glance registry, OVF, structured metadata) | 12:27 |
| s1rp | let's just do it | 12:27 |
| jaypipes | yamahata: ok, so you are good with this decision? It would essentially mean that much of your work would be scratched :( | 12:28 |
| bcwaldon | we can go ahead and get started on it, assuming community approval | 12:28 |
| bcwaldon | anybody here want to take a lead on this? | 12:28 |
| jaypipes | yamahata: I feel bad since you put so much effort into it :( | 12:28 |
| yamahata | I'm quite fine with the conclusion. | 12:28 |
| jaypipes | yamahata: ok, thank you! | 12:28 |
| jaypipes | bcwaldon: I think s1rp should take the lead on this. | 12:28 |
| bcwaldon | fine! | 12:29 |
| jaypipes | rick, you ok with that? | 12:29 |
| yamahata | Anyway we can reach much better solution. | 12:29 |
| s1rp | ill need to run it by comstud first to make sure this won't impact our team priorities too much, but yeah, would love to | 12:29 |
| jaypipes | yamahata: yes, I think long-term this will definitely pay off. | 12:29 |
| jaypipes | s1rp: k. could you get with him this morning on that? | 12:29 |
| s1rp | jaypipes: will do | 12:29 |
| jaypipes | cheers | 12:29 |
| jaypipes | OK, expect to see a draft of the email to the mailing list sometime before noonish, EDT, today. | 12:30 |
| jaypipes | anything else anyone want to bring up? | 12:30 |
| jaypipes | alright, see y'all later! | 12:30 |
| bcwaldon | thanks, Jay | 12:31 |
| yamahata | jaypipes, thank you. | 12:31 |
| s1rp | thanks guys | 12:31 |
| * s1rp trying to decide whether to snooze for another hour or just stay up | 12:31 | |
| * blamar zzzzzzzzz | 12:32 | |
| *** jkoelker has quit IRC | 12:32 | |
| s1rp | hehe, seeing as it will be 108 today, think ill enjoy the relatively cool 80 degrees out on my balcony ) | 12:32 |
| s1rp | :) | 12:32 |
| blamar | niiiiice | 12:32 |
| *** jkoelker has joined #openstack-meeting | 12:32 | |
| *** yamahata has quit IRC | 12:36 | |
| *** sandywalsh has joined #openstack-meeting | 12:40 | |
| *** bcwaldon has left #openstack-meeting | 12:43 | |
| *** edconzel has joined #openstack-meeting | 13:45 | |
| *** toobulkeh has quit IRC | 13:54 | |
| *** dragondm has joined #openstack-meeting | 14:04 | |
| *** annegentle has joined #openstack-meeting | 14:10 | |
| *** toobulkeh has joined #openstack-meeting | 14:20 | |
| *** dprince has joined #openstack-meeting | 14:33 | |
| *** Cyns has joined #openstack-meeting | 14:56 | |
| *** rnirmal has joined #openstack-meeting | 14:57 | |
| *** msinhore has joined #openstack-meeting | 15:04 | |
| *** msinhore has quit IRC | 15:33 | |
| *** ewanmellor has joined #openstack-meeting | 15:48 | |
| *** heckj has joined #openstack-meeting | 15:49 | |
| *** nati has joined #openstack-meeting | 15:51 | |
| *** Cyns has quit IRC | 15:51 | |
| *** msinhore has joined #openstack-meeting | 16:29 | |
| *** ewanmellor has quit IRC | 16:31 | |
| *** clayg has left #openstack-meeting | 16:35 | |
| *** msinhore has quit IRC | 17:06 | |
| *** mattray has joined #openstack-meeting | 17:06 | |
| *** jakedahn has quit IRC | 17:11 | |
| *** joearnold has joined #openstack-meeting | 17:19 | |
| *** darraghb has quit IRC | 17:20 | |
| *** dprince has quit IRC | 17:22 | |
| *** heckj has quit IRC | 17:25 | |
| *** jakedahn has joined #openstack-meeting | 17:51 | |
| *** dendro-afk is now known as dendrobates | 17:59 | |
| *** rnirmal has quit IRC | 18:00 | |
| *** rnirmal_ has joined #openstack-meeting | 18:01 | |
| *** edconzel has quit IRC | 18:05 | |
| *** jakedahn_ has joined #openstack-meeting | 18:17 | |
| *** jakedahn has quit IRC | 18:19 | |
| *** jakedahn_ is now known as jakedahn | 18:19 | |
| *** msinhore has joined #openstack-meeting | 18:30 | |
| *** jakedahn has quit IRC | 18:33 | |
| *** KnuckleSangwich has joined #openstack-meeting | 18:36 | |
| *** jakedahn has joined #openstack-meeting | 18:39 | |
| *** msinhore has quit IRC | 18:45 | |
| *** jakedahn_ has joined #openstack-meeting | 18:47 | |
| *** jakedahn has quit IRC | 18:48 | |
| *** jakedahn_ is now known as jakedahn | 18:48 | |
| *** nati has quit IRC | 18:55 | |
| *** nati has joined #openstack-meeting | 18:56 | |
| *** rnirmal_ has quit IRC | 19:02 | |
| *** rnirmal has joined #openstack-meeting | 19:02 | |
| *** edconzel has joined #openstack-meeting | 19:05 | |
| *** dendrobates is now known as dendro-afk | 19:06 | |
| *** mdomsch has joined #openstack-meeting | 19:11 | |
| *** msinhore has joined #openstack-meeting | 19:16 | |
| *** nati has quit IRC | 19:23 | |
| *** Cyns has joined #openstack-meeting | 19:31 | |
| *** cynb has quit IRC | 19:33 | |
| *** cynb has joined #openstack-meeting | 19:34 | |
| *** nati_ has joined #openstack-meeting | 19:50 | |
| *** toobulkeh has quit IRC | 20:34 | |
| *** sandywalsh has quit IRC | 20:38 | |
| *** littleidea has joined #openstack-meeting | 20:42 | |
| *** cynb has quit IRC | 21:00 | |
| *** jjm has quit IRC | 21:06 | |
| *** _cerberus_ has quit IRC | 21:07 | |
| *** mdomsch has quit IRC | 21:09 | |
| *** _cerberus_ has joined #openstack-meeting | 21:10 | |
| *** jaypipes has quit IRC | 21:11 | |
| *** Cyns has quit IRC | 21:11 | |
| *** cynb has joined #openstack-meeting | 21:15 | |
| *** jjm has joined #openstack-meeting | 21:19 | |
| *** jaypipes has joined #openstack-meeting | 21:20 | |
| *** dolphm has quit IRC | 21:25 | |
| *** jkoelker has quit IRC | 21:41 | |
| *** sandywalsh has joined #openstack-meeting | 21:47 | |
| *** nati_ has quit IRC | 22:12 | |
| *** primeministerp1 is now known as primeministerp|h | 22:41 | |
| *** mattray has quit IRC | 22:43 | |
| *** edconzel has quit IRC | 22:53 | |
| *** Cyns has joined #openstack-meeting | 22:58 | |
| *** msinhore has quit IRC | 23:02 | |
| *** Cyns_ has joined #openstack-meeting | 23:17 | |
| *** Cyns has quit IRC | 23:18 | |
| *** Cyns_ is now known as Cyns | 23:18 | |
| *** joearnold has quit IRC | 23:18 | |
| *** jakedahn has quit IRC | 23:19 | |
| *** Cyns has quit IRC | 23:31 | |
| *** msinhore has joined #openstack-meeting | 23:39 | |
| *** rnirmal has quit IRC | 23:42 | |
| *** littleidea has quit IRC | 23:55 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!