| *** markvoelker has joined #openstack-glance | 00:05 | |
| *** markvoelker has quit IRC | 00:10 | |
| *** brinzhang has joined #openstack-glance | 01:04 | |
| *** brinzhang has joined #openstack-glance | 01:05 | |
| *** markvoelker has joined #openstack-glance | 01:05 | |
| *** markvoelker has quit IRC | 01:09 | |
| *** brinzhang has quit IRC | 01:10 | |
| *** brinzhang has joined #openstack-glance | 01:11 | |
| *** brinzhang has quit IRC | 01:12 | |
| *** brinzhang has joined #openstack-glance | 01:13 | |
| *** imacdonn has quit IRC | 01:22 | |
| *** imacdonn has joined #openstack-glance | 01:22 | |
| *** LiangFang has joined #openstack-glance | 01:23 | |
| *** zhanglong has joined #openstack-glance | 01:32 | |
| *** MattMan_ has quit IRC | 01:46 | |
| *** MattMan_ has joined #openstack-glance | 01:46 | |
| *** bhagyashris has joined #openstack-glance | 01:46 | |
| LiangFang | Hi cores, regarding the Solution2 for BP: Specify image size to avoid glance backend resize | 02:54 |
|---|---|---|
| LiangFang | if you want to take a try | 02:55 |
| LiangFang | just need to cherry-pick the patch, and it can works | 02:55 |
| LiangFang | 1. cherry-pick the cinder patch: https://review.openstack.org/#/c/609994/ | 02:56 |
| LiangFang | 2. cherry-pick the glance patch: https://review.openstack.org/#/c/609997/ | 02:57 |
| LiangFang | I just use ceph as backend, and it works | 02:58 |
| *** rosmaita has left #openstack-glance | 02:58 | |
| *** rcernin has quit IRC | 03:02 | |
| *** brinzhang has quit IRC | 03:17 | |
| *** brinzhang has joined #openstack-glance | 03:18 | |
| *** rcernin has joined #openstack-glance | 03:28 | |
| *** brinzhang has quit IRC | 04:20 | |
| *** udesale has joined #openstack-glance | 04:45 | |
| *** bhagyashris has quit IRC | 04:59 | |
| *** lbragstad_503 has quit IRC | 05:27 | |
| *** lbragstad_503 has joined #openstack-glance | 05:27 | |
| *** ratailor has joined #openstack-glance | 06:01 | |
| *** Luzi has joined #openstack-glance | 06:15 | |
| *** mosulica has joined #openstack-glance | 06:26 | |
| *** bhagyashris has joined #openstack-glance | 06:29 | |
| *** rcernin has quit IRC | 07:07 | |
| *** jcosmao has quit IRC | 07:22 | |
| *** openstackgerrit has quit IRC | 07:35 | |
| *** dpawlik has quit IRC | 07:55 | |
| *** dpawlik has joined #openstack-glance | 08:13 | |
| *** ratailor has quit IRC | 08:43 | |
| *** ratailor has joined #openstack-glance | 08:43 | |
| *** pooja-jadhav has joined #openstack-glance | 08:46 | |
| *** pooja_jadhav has quit IRC | 08:46 | |
| *** dpawlik has quit IRC | 09:02 | |
| *** dpawlik has joined #openstack-glance | 09:04 | |
| jokke_ | LiangFang: I'll try to have a look of it today | 09:19 |
| *** LiangFang has quit IRC | 09:31 | |
| *** pooja-jadhav is now known as pooja_jadhav | 09:33 | |
| *** bhagyashris has quit IRC | 09:39 | |
| *** priteau has joined #openstack-glance | 09:52 | |
| *** zhanglong has quit IRC | 10:59 | |
| *** udesale has quit IRC | 11:04 | |
| *** ratailor has quit IRC | 11:32 | |
| *** rosmaita has joined #openstack-glance | 11:59 | |
| *** dims has quit IRC | 12:30 | |
| *** dims has joined #openstack-glance | 12:33 | |
| *** mvkr has quit IRC | 13:27 | |
| *** lbragstad_503 is now known as lbragstad | 13:32 | |
| *** LiangFang has joined #openstack-glance | 13:42 | |
| *** jcosmao has joined #openstack-glance | 13:47 | |
| LiangFang | thanks erno | 13:52 |
| *** mvkr has joined #openstack-glance | 13:57 | |
| imacdonn | jokke_: I need to run to a dentist appointment ... please re-review this https://review.openstack.org/597648 | 14:48 |
| imacdonn | rosmaita: thanks for your +2 | 14:49 |
| *** jmlowe has quit IRC | 14:57 | |
| *** Luzi has quit IRC | 15:01 | |
| *** itlinux has quit IRC | 15:01 | |
| LiangFang | @jokke_, just to correct one thing, glance image-show can show store id, like "| stores | ceph" | 15:02 |
| *** lixiaoy1 has joined #openstack-glance | 15:04 | |
| jokke_ | LiangFang: ok, I thought it should have been there | 15:04 |
| jokke_ | thanks for the confirmation | 15:04 |
| jokke_ | so I assume the -v image-list does not? | 15:04 |
| LiangFang | I guess image-list should also know the store info, but not displayed | 15:05 |
| LiangFang | glance -v image-list only list a few columns | 15:06 |
| LiangFang | not including store id | 15:07 |
| LiangFang | I can take a look if -v image-list can show store id or not | 15:08 |
| *** mosulica has quit IRC | 15:13 | |
| LiangFang | I just added store id to the column, -v image-list can show store id | 15:14 |
| LiangFang | like: | 15:14 |
| LiangFang | | 4804a6b7-6392-4232-9a9e-897d1eeef589 | img2-ceph | ceph | rbd | raw | bare | 13267968 | active | 60a12b1788ad44468afd983f89a5f8dc | | 15:14 |
| LiangFang | | e9af91ba-4d8c-4a3c-b2d7-5fbb5421f065 | img2-file | locallvm | file | raw | bare | 13267968 | active | 60a12b1788ad44468afd983f89a5f8dc | | 15:15 |
| *** openstackgerrit has joined #openstack-glance | 15:18 | |
| openstackgerrit | Merged openstack/glance_store master: Add statement explaining "tips" job configuration https://review.openstack.org/609756 | 15:18 |
| LiangFang | so it seems we don't need to save backend type anymore, store id can describe the backend type instead. | 15:18 |
| LiangFang | what we may can do is just add store id to the column of "-v image-list" | 15:19 |
| LiangFang | I don't know whether I confused you or not:) | 15:20 |
| LiangFang | one of my review is to save store info, this one may don't be needed anymore. because store id already there. https://review.openstack.org/#/c/605006/ | 15:22 |
| rosmaita | LiangFang: that sounds the way to go ... maybe a spec-lite for python-glanceclient, about adding the column? It might be a good idea to add a separate switch, maybe --show-stores or something, to the image-list response instead of cluttering up the current -v (or maybe make it -vv) | 15:24 |
| LiangFang | another on is to display store info in "-v image-list", for this one, we just need to add store id to the display. https://review.openstack.org/#/c/605014/ | 15:24 |
| rosmaita | i'm just worried about most clouds showing all 'default' in that store id column, that's why i'm not sure it should go in the "normal" -v response | 15:26 |
| LiangFang | @rosmaita, sounds good. | 15:27 |
| rosmaita | LiangFang: have you done a glance spec-lite before? it's pretty lightweight, and will give us a place to discuss your proposal | 15:27 |
| LiangFang | no, I'm new to openstack | 15:28 |
| rosmaita | ok, give me a sec and i'll get you a pointer to some info | 15:28 |
| rosmaita | here's the basic info: https://docs.openstack.org/glance/latest/contributor/blueprints.html | 15:29 |
| LiangFang | ok, thanks | 15:30 |
| *** lixiaoy1 has left #openstack-glance | 15:30 | |
| rosmaita | LiangFang: you put up a patch, just like for code. You'd propose it here: https://git.openstack.org/cgit/openstack/glance-specs/tree/specs/stein/approved/python-glanceclient | 15:30 |
| rosmaita | just copy the spec-lite-template.rst file that's in that directory and fill it out with your proposal, and put up the patch | 15:31 |
| LiangFang | OK.. | 15:32 |
| rosmaita | ask in here if you have any questions | 15:33 |
| imacdonn | LiangFang: example of a spec-lite: https://review.openstack.org/597648 | 15:33 |
| LiangFang | thanks | 15:34 |
| imacdonn | good luck getting it approved *cough* *grin* | 15:34 |
| rosmaita | imacdonn: i think yours is close to approval, erno's going to put some comments (you can read through the meeting log if he hasn't posted them yet), just 2 points to address | 15:35 |
| imacdonn | rosmaita: OK. Just got back from dentist .. will check the meeting log | 15:36 |
| rosmaita | imacdonn: hope it was a pleasant visit! | 15:36 |
| imacdonn | hah, yes, as much as can be expected .. this was just a followup from a procedure I had done last week | 15:36 |
| rosmaita | i once missed a faculty meeting because i was having a root canal, and i can honestly say that i enjoyed the root canal more than the typical faculty meeting! | 15:37 |
| imacdonn | lol | 15:38 |
| LiangFang | guys, I need to drop now, it's late in asia:) | 15:39 |
| rosmaita | LiangFang: good night | 15:39 |
| imacdonn | GN | 15:39 |
| LiangFang | good day, bye.. | 15:39 |
| *** LiangFang has left #openstack-glance | 15:41 | |
| *** gyee has joined #openstack-glance | 15:43 | |
| *** itlinux has joined #openstack-glance | 15:51 | |
| imacdonn | rosmaita: I don't understand the concern about os_hash_algo having to be the same as the one glance is configured with | 15:56 |
| imacdonn | glance is neither computing not validating the hash, so why should it have to match? All that matters is that any *consumer* of the image can utilise the hash | 15:57 |
| rosmaita | right, but if all images in an installation are using the same hash algo, it's more likely any consumer will have the algo available | 15:57 |
| rosmaita | i'm not sure how big a deal it is, but i think that's the concern | 15:59 |
| imacdonn | I believe that the multihash spec states (or at least implies) that anything supported by hashlib is suitable | 16:00 |
| imacdonn | so presumably any consumer is expected to utilise hashlib, or some equivalent | 16:00 |
| rosmaita | yes, but it also says somewhere that the hash in use isn't specified by the consumer | 16:00 |
| imacdonn | yeah, but also the value isn't specified by the consumer in the cases considered there | 16:05 |
| rosmaita | yes, you are providing an extension to existing functionality | 16:05 |
| rosmaita | i think that's the source of the consistency concern | 16:05 |
| rosmaita | did erno bring that up, or was it me? | 16:06 |
| imacdonn | 14:07:28 <jokke_> and how is this gonna affect if there is no supported hash-algo configured in glance? | 16:07 |
| imacdonn | my answer would have been "it doesn't" | 16:07 |
| rosmaita | i think erno's concern is what if the user has "os_hash_algo": "donald_trump", what do we do in that case? | 16:08 |
| rosmaita | i think you have some validation, but i couldn't remember | 16:09 |
| imacdonn | as stated in the spec, it must be one of the ones that hashlib supports | 16:10 |
| rosmaita | right, but the key thing is to make sure glance rejects an algo it doesn't recognize | 16:10 |
| imacdonn | validated at https://review.openstack.org/#/c/597368/7/glance/api/v2/images.py lines 382-387 | 16:10 |
| rosmaita | since it's not going to actually compute it | 16:10 |
| rosmaita | i thought you did, just need to mention on the spec | 16:11 |
| imacdonn | spec says "... os_hash_algo must be one of the algorithms supported by hashlib. Any violations of these rules will result in a HTTPConflict exception (HTTP status 409)." | 16:12 |
| rosmaita | ok, cool | 16:12 |
| *** mvkr has quit IRC | 16:58 | |
| *** mvkr has joined #openstack-glance | 17:41 | |
| *** priteau has quit IRC | 17:52 | |
| *** lbragstad has quit IRC | 18:17 | |
| *** lbragstad has joined #openstack-glance | 18:48 | |
| *** itlinux has quit IRC | 19:16 | |
| *** lbragstad has quit IRC | 19:21 | |
| *** lbragstad has joined #openstack-glance | 19:24 | |
| imacdonn | rosmaita: I added a couple of comments to the spec review. Maybe you could "hash it out" (get it?) with jokke_ tomorrow (since you seem to have more timezone overlap) | 19:34 |
| rosmaita | imacdonn: sure thing | 19:34 |
| imacdonn | thanks!@ | 19:34 |
| *** openstackgerrit has quit IRC | 20:36 | |
| *** pcaruana has quit IRC | 20:44 | |
| *** itlinux has joined #openstack-glance | 21:06 | |
| *** rosmaita has left #openstack-glance | 21:42 | |
| *** rcernin has joined #openstack-glance | 22:33 | |
| *** jiaopengju has quit IRC | 23:05 | |
| *** jiaopengju has joined #openstack-glance | 23:08 | |
| *** jiaopengju has quit IRC | 23:35 | |
| *** jiaopengju has joined #openstack-glance | 23:36 | |
| *** markvoelker has joined #openstack-glance | 23:57 | |
| *** gyee has quit IRC | 23:59 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!