| *** brinzhang has joined #openstack-glance | 00:32 | |
| *** brinzhang_ has joined #openstack-glance | 00:45 | |
| *** brinzhang has quit IRC | 00:49 | |
| *** goldyfruit_ has joined #openstack-glance | 01:07 | |
| *** Liang__ has joined #openstack-glance | 01:08 | |
| *** lifeless has quit IRC | 01:34 | |
| *** lifeless has joined #openstack-glance | 01:49 | |
| *** Liang__ is now known as LiangFang | 01:54 | |
| *** awalende has joined #openstack-glance | 01:57 | |
| *** goldyfruit_ has quit IRC | 01:58 | |
| *** awalende has quit IRC | 02:02 | |
| *** lifeless has quit IRC | 02:12 | |
| *** lifeless has joined #openstack-glance | 02:24 | |
| *** gyee has quit IRC | 02:53 | |
| *** goldyfruit_ has joined #openstack-glance | 03:28 | |
| *** bhagyashris has joined #openstack-glance | 04:31 | |
| *** abhishekk has joined #openstack-glance | 04:57 | |
| *** goldyfruit_ has quit IRC | 05:00 | |
| *** bhagyashris has quit IRC | 05:04 | |
| *** brinzhang has joined #openstack-glance | 05:09 | |
| *** bhagyashris has joined #openstack-glance | 05:09 | |
| *** brinzhang has quit IRC | 05:10 | |
| *** brinzhang has joined #openstack-glance | 05:11 | |
| *** brinzhang_ has quit IRC | 05:12 | |
| *** pcaruana has joined #openstack-glance | 05:25 | |
| *** ratailor has joined #openstack-glance | 05:27 | |
| *** jmlowe has quit IRC | 05:46 | |
| *** brinzhang_ has joined #openstack-glance | 05:52 | |
| *** brinzhang has quit IRC | 05:55 | |
| *** awalende has joined #openstack-glance | 05:58 | |
| *** brinzhang has joined #openstack-glance | 06:00 | |
| openstackgerrit | Abhishek Kekane proposed openstack/glance-specs master: Copying existing image in multiple stores https://review.opendev.org/694724 | 06:01 |
|---|---|---|
| *** awalende has quit IRC | 06:02 | |
| *** brinzhang_ has quit IRC | 06:03 | |
| *** abhishekk has quit IRC | 07:06 | |
| *** brinzhang has quit IRC | 07:15 | |
| *** brinzhang has joined #openstack-glance | 07:16 | |
| *** abhishekk has joined #openstack-glance | 07:18 | |
| *** udesale has joined #openstack-glance | 07:25 | |
| *** tosky has joined #openstack-glance | 07:35 | |
| *** priteau has quit IRC | 07:38 | |
| *** awalende has joined #openstack-glance | 07:47 | |
| *** tesseract has joined #openstack-glance | 07:58 | |
| *** tkajinam has quit IRC | 08:17 | |
| openstackgerrit | zhangboye proposed openstack/glance master: Migrate grenade jobs to py3 https://review.opendev.org/695149 | 09:02 |
| *** jawad_axd has joined #openstack-glance | 09:03 | |
| tosky | duplicate review... ^ | 09:06 |
| *** tosky has quit IRC | 09:13 | |
| *** tosky has joined #openstack-glance | 09:14 | |
| openstackgerrit | Abhishek Kekane proposed openstack/glance-specs master: Make cinder driver compatible with multiple stores https://review.opendev.org/695152 | 09:36 |
| *** awalende has quit IRC | 09:43 | |
| *** rcernin has quit IRC | 09:46 | |
| *** awalende has joined #openstack-glance | 09:48 | |
| *** priteau has joined #openstack-glance | 09:52 | |
| *** awalende has quit IRC | 09:52 | |
| *** awalende has joined #openstack-glance | 09:52 | |
| *** priteau has quit IRC | 10:06 | |
| *** abhishekk has quit IRC | 10:25 | |
| *** LiangFang has quit IRC | 10:32 | |
| *** brinzhang_ has joined #openstack-glance | 10:38 | |
| *** brinzhang has quit IRC | 10:42 | |
| *** brinzhang has joined #openstack-glance | 11:01 | |
| *** brinzhang_ has quit IRC | 11:05 | |
| *** bhagyashris has quit IRC | 11:10 | |
| *** lpetrut has joined #openstack-glance | 11:26 | |
| openstackgerrit | Luigi Toscano proposed openstack/glance stable/stein: Periodic jobs are failing for python 3.5 https://review.opendev.org/691308 | 11:28 |
| openstackgerrit | Luigi Toscano proposed openstack/glance stable/stein: Run functional jobs against Python 3.6 https://review.opendev.org/691313 | 11:28 |
| openstackgerrit | Luigi Toscano proposed openstack/glance stable/stein: Don't target broken ssl tests to specific py3.minor https://review.opendev.org/691367 | 11:28 |
| openstackgerrit | Luigi Toscano proposed openstack/glance stable/stein: Bring py27 tests back https://review.opendev.org/695176 | 11:28 |
| openstackgerrit | Luigi Toscano proposed openstack/glance stable/stein: Dropping the py35 testing https://review.opendev.org/695177 | 11:28 |
| tosky | this should hopefully fix the tests in that branch | 11:28 |
| tosky | this stack of patches | 11:29 |
| *** abhishekk has joined #openstack-glance | 11:41 | |
| *** brinzhang_ has joined #openstack-glance | 11:45 | |
| *** ratailor has quit IRC | 11:46 | |
| openstackgerrit | Luigi Toscano proposed openstack/glance stable/stein: Dropping the py35 testing https://review.opendev.org/695177 | 11:48 |
| *** brinzhang has quit IRC | 11:49 | |
| *** brinzhang_ has quit IRC | 12:16 | |
| *** abhishekk has quit IRC | 12:20 | |
| *** pcaruana has quit IRC | 12:24 | |
| *** awalende has quit IRC | 12:26 | |
| *** pcaruana has joined #openstack-glance | 12:50 | |
| *** awalende has joined #openstack-glance | 12:52 | |
| *** awalende has quit IRC | 12:56 | |
| *** goldyfruit_ has joined #openstack-glance | 12:59 | |
| *** priteau has joined #openstack-glance | 13:05 | |
| *** goldyfruit_ has quit IRC | 13:16 | |
| *** bhagyashris has joined #openstack-glance | 13:18 | |
| *** bhagyashris has quit IRC | 13:46 | |
| *** bhagyashris has joined #openstack-glance | 13:46 | |
| *** priteau has quit IRC | 14:11 | |
| *** priteau has joined #openstack-glance | 14:12 | |
| *** priteau has quit IRC | 14:13 | |
| *** bhagyashris has quit IRC | 14:30 | |
| *** goldyfruit_ has joined #openstack-glance | 14:43 | |
| *** jawad_axd has quit IRC | 14:45 | |
| *** jawad_axd has joined #openstack-glance | 14:46 | |
| *** jawad_ax_ has joined #openstack-glance | 14:49 | |
| *** bhagyashris has joined #openstack-glance | 14:49 | |
| *** jawad_axd has quit IRC | 14:50 | |
| *** jawad_axd has joined #openstack-glance | 14:51 | |
| *** jawad_ax_ has quit IRC | 14:53 | |
| *** jawad_axd has quit IRC | 14:55 | |
| *** davee_ has joined #openstack-glance | 15:07 | |
| *** baojg has joined #openstack-glance | 15:08 | |
| *** johnthetubaguy has joined #openstack-glance | 15:09 | |
| *** udesale has quit IRC | 15:20 | |
| dansmith | can anyone answer a dumb question for me? | 15:20 |
| *** udesale has joined #openstack-glance | 15:20 | |
| dansmith | a long while ago, there was some discussion about a way to have multiple payloads for a glance image, | 15:20 |
| dansmith | like a special meta key that referenced another glance upload that was't actually a disk image | 15:21 |
| dansmith | I want to say it was for storing a LUKS key or something like that, or maybe some hypervisor-specific payload | 15:21 |
| dansmith | is that a formal thing and/or am I making that up? | 15:21 |
| davee_ | I believe you may be talking about multi store possibly | 15:24 |
| davee_ | @dansmith is this what you are referencing? https://review.opendev.org/#/c/661676/4 | 15:27 |
| dansmith | no, not multi-store | 15:28 |
| dansmith | I'm talking about an image that has some sidecar along with it.. so the instance disk and some binary blob that goes along with it | 15:28 |
| dansmith | like firmware, or an fpga image, or a securitykey | 15:28 |
| *** bhagyashris has quit IRC | 15:30 | |
| davee_ | https://www.mirantis.co.jp/blog/openstack-glances-artifacts/ possibly | 15:37 |
| dansmith | davee_: was that glare? | 15:37 |
| *** efried has joined #openstack-glance | 15:38 | |
| dansmith | is there any prior art around storing binary blobs in glance that aren't disk miages? | 15:38 |
| dansmith | because I'm kinda wondering if polluting glance with non-bootable images that are all bare/raw would be confusing to users | 15:38 |
| davee_ | Looks like glare would do what you are describing | 15:39 |
| efried | dansmith: the last time I asked a similar question (I think I was asking whether swift or glance would be a better option for $something) folks basically said you could store whatever you want. | 15:39 |
| efried | glare doesn't exist though? | 15:39 |
| dansmith | no, glare's not an option AFAIK | 15:39 |
| dansmith | efried: yeah, so I'm sure we technically can, but I'm just thinking as a user, I go to list images and I get 500, 50 of which are actually things I can boot from, | 15:40 |
| dansmith | and if I'm using all raw images, there's no good way to filter that list right? | 15:40 |
| * dansmith pokes rosmaita | 15:41 | |
| rosmaita | ouch | 15:41 |
| rosmaita | well, we do the metadata thing for amazon-style images | 15:42 |
| efried | dansmith: I'm not sure "might be bootable" is really a thing. Seems like there's a zillion ways you could fail to boot if you pick the wrong image anyway. | 15:42 |
| rosmaita | but there's nothing in glance enforcing that the metadata actually refers to anything | 15:42 |
| dansmith | rosmaita: to link the root with the kernel yeah? | 15:42 |
| rosmaita | yeah, we expect nova to know how to do that | 15:43 |
| rosmaita | to answer your question though | 15:43 |
| dansmith | efried: I think most people doing an image-list expect that the things in that list are bootable | 15:43 |
| rosmaita | you can put anythinng into glance, but you should also expect that end users will try to boot it | 15:43 |
| dansmith | efried: ^ | 15:43 |
| rosmaita | also, the days of anything into glance may be numbered | 15:43 |
| rosmaita | as people switch from simple upload to image import | 15:44 |
| dansmith | rosmaita: meaning glance being more particular about you uploading an *actual* disk image right? | 15:44 |
| dansmith | yeah | 15:44 |
| rosmaita | where you can have a plugin chain to validate stuff | 15:44 |
| rosmaita | dansmith: right | 15:44 |
| rosmaita | but that would be an operator choice | 15:44 |
| dansmith | yeah | 15:44 |
| dansmith | yeah, but we would not want to commit to a long-term path that was going to eliminate the ability for some operators to have strong image validation | 15:45 |
| rosmaita | arbitrary blobs should go into the object store IMO | 15:46 |
| dansmith | rosmaita: presumably it might make sense for glance to gain a new container or image format of "not a bootable thing" if glance really wants to be in the business of storing such things | 15:46 |
| dansmith | rosmaita: yeah, in general that's definitely true, but building a core feature into nova that requires a swift deployment isn't great | 15:46 |
| rosmaita | yes, that would be a possibility | 15:46 |
| rosmaita | dansmith: i am doing last minute prep for cinder meeting at top of the hour, will read through this later | 15:47 |
| rosmaita | would prob be best to talk to erno and abhishek | 15:47 |
| dansmith | rosmaita: no problem, thanks, I think you've answered the question that we shouldn't just shove this into glance without more discussion | 15:47 |
| dansmith | efried: right? ^ | 15:47 |
| efried | I didn't quite pick that up, but let me reread... | 15:48 |
| efried | (and go find the logs from before I jumped on the channel) | 15:48 |
| dansmith | efried: two key takeaways: 1. users expect everything to be bootable in glance and 2. glance may soon start doing more of an image-import than just saving your blob to disk, which would mean uploading random stuff and calling it a disk image may stop working | 15:49 |
| dansmith | efried: if we're going to do that, we likely need a new container/image format of "blob" that i explicitly not bootable for things like this | 15:49 |
| efried | which seems like what raw/bare ought to have been from the start, but whatevs. | 15:50 |
| dansmith | efried: no, bare/raw means a byte-for-byte copy of a disk partition | 15:50 |
| dansmith | efried: which is what you have to use if you're on ceph, so... not like it's not a thing :) | 15:50 |
| efried | dansmith: and is glance not supposed to be used to store data disks? | 15:50 |
| efried | only boot disks? | 15:50 |
| dansmith | efried: it's intended to store bootable things, so an AWS image (set) is not even just a bootable disk | 15:51 |
| dansmith | efried: but yeah, we don't ever snapshot a data-only disk like ephemeral to glance that I know of | 15:51 |
| dansmith | efried: <rosmaita>you can put anythinng into glance, but you should also expect that end users will try to boot it | 15:52 |
| efried | I mean, | 15:53 |
| efried | I'm having a hard time caring if users try to boot this thing. | 15:53 |
| efried | Like I would have a hard time caring if they tried to boot a ppc image on x86 | 15:53 |
| efried | it just... wouldn't work. And it's their fault. So what? | 15:54 |
| dansmith | but that's the thing, | 15:54 |
| efried | maybe I'm missing a fundamental architectural philosophy here. | 15:54 |
| dansmith | they don't try to boot ppc on x86 | 15:54 |
| dansmith | if the cloud has both images in it, they boot an *image* and the scheduler selects a host that matches the arch | 15:54 |
| dansmith | efried: I think what you're missing is that glance contains bootable things, from the user's perspective | 15:54 |
| efried | okay, noted. | 15:55 |
| *** jmlowe has joined #openstack-glance | 16:00 | |
| *** jmlowe has quit IRC | 16:10 | |
| *** udesale has quit IRC | 16:14 | |
| *** gyee has joined #openstack-glance | 16:16 | |
| *** jawad_axd has joined #openstack-glance | 16:29 | |
| *** jawad_axd has quit IRC | 16:34 | |
| *** goldyfruit___ has joined #openstack-glance | 16:53 | |
| *** goldyfruit___ is now known as goldyfruit | 16:54 | |
| *** goldyfruit_ has quit IRC | 16:56 | |
| *** takamatsu has joined #openstack-glance | 17:23 | |
| *** jawad_axd has joined #openstack-glance | 17:26 | |
| *** goldyfruit has quit IRC | 17:27 | |
| *** goldyfruit has joined #openstack-glance | 17:30 | |
| *** jawad_axd has quit IRC | 17:31 | |
| *** goldyfruit has quit IRC | 17:31 | |
| *** goldyfruit has joined #openstack-glance | 17:34 | |
| *** goldyfruit has joined #openstack-glance | 17:35 | |
| *** jawad_axd has joined #openstack-glance | 17:47 | |
| *** jawad_axd has quit IRC | 17:52 | |
| *** jmlowe has joined #openstack-glance | 18:17 | |
| *** tesseract has quit IRC | 18:18 | |
| *** tosky has quit IRC | 18:55 | |
| *** goldyfruit has quit IRC | 19:18 | |
| *** goldyfruit has joined #openstack-glance | 19:18 | |
| *** goldyfruit has quit IRC | 19:20 | |
| *** goldyfruit has joined #openstack-glance | 19:20 | |
| *** awalende has joined #openstack-glance | 19:31 | |
| *** lpetrut has quit IRC | 19:32 | |
| *** lpetrut has joined #openstack-glance | 19:32 | |
| *** awalende_ has joined #openstack-glance | 19:36 | |
| *** awalende has quit IRC | 19:39 | |
| *** awalende_ has quit IRC | 19:41 | |
| *** awalende has joined #openstack-glance | 19:41 | |
| *** awalende_ has joined #openstack-glance | 19:51 | |
| *** awalende has quit IRC | 19:54 | |
| *** awalende has joined #openstack-glance | 20:05 | |
| *** awalende_ has quit IRC | 20:06 | |
| *** awalende_ has joined #openstack-glance | 20:06 | |
| *** awalende has quit IRC | 20:10 | |
| *** awalende has joined #openstack-glance | 20:17 | |
| *** awalende has quit IRC | 20:19 | |
| *** awalende_ has quit IRC | 20:20 | |
| *** jmlowe has quit IRC | 20:43 | |
| *** awalende has joined #openstack-glance | 21:14 | |
| *** awalende has quit IRC | 21:19 | |
| *** tosky has joined #openstack-glance | 21:41 | |
| *** pcaruana has quit IRC | 21:42 | |
| *** awalende has joined #openstack-glance | 22:11 | |
| *** awalende has quit IRC | 22:15 | |
| *** rcernin has joined #openstack-glance | 22:31 | |
| *** tkajinam has joined #openstack-glance | 23:05 | |
| *** goldyfruit has quit IRC | 23:48 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!