*** chlong has joined #openstack-glance | 00:00 | |
*** ducttape_ has quit IRC | 00:03 | |
*** harshs has quit IRC | 00:07 | |
*** manous has quit IRC | 00:07 | |
*** r-daneel has joined #openstack-glance | 00:37 | |
*** r-daneel has quit IRC | 00:44 | |
*** dims has joined #openstack-glance | 00:58 | |
*** TravT has joined #openstack-glance | 01:04 | |
*** TravT has quit IRC | 01:09 | |
*** annashen has joined #openstack-glance | 01:11 | |
*** jasondotstar has joined #openstack-glance | 01:26 | |
*** mingdang1 has joined #openstack-glance | 01:51 | |
*** ducttape_ has joined #openstack-glance | 01:58 | |
*** TravT has joined #openstack-glance | 02:05 | |
*** TravT has quit IRC | 02:09 | |
*** shalq has joined #openstack-glance | 02:10 | |
*** ayoung has quit IRC | 02:17 | |
*** ducttape_ has quit IRC | 02:35 | |
*** annashen has quit IRC | 02:35 | |
*** annashen has joined #openstack-glance | 02:37 | |
*** annashen_ has joined #openstack-glance | 02:40 | |
*** annashen_ has quit IRC | 02:41 | |
*** annashen has quit IRC | 02:42 | |
*** annashen has joined #openstack-glance | 02:42 | |
*** annashen has quit IRC | 02:45 | |
*** annashen has joined #openstack-glance | 02:46 | |
*** annashen has quit IRC | 02:50 | |
*** annashen has joined #openstack-glance | 02:50 | |
*** snapdey has joined #openstack-glance | 02:58 | |
*** dims has quit IRC | 02:59 | |
*** ducttape_ has joined #openstack-glance | 03:01 | |
*** snapdey has quit IRC | 03:02 | |
*** kcaj has quit IRC | 03:07 | |
*** annashen has quit IRC | 03:09 | |
*** annashen has joined #openstack-glance | 03:12 | |
*** kcaj has joined #openstack-glance | 03:14 | |
*** ducttape_ has quit IRC | 03:14 | |
*** ducttape_ has joined #openstack-glance | 03:15 | |
*** changbl has quit IRC | 03:19 | |
*** jasondotstar has quit IRC | 03:20 | |
*** ducttape_ has quit IRC | 03:32 | |
*** changbl has joined #openstack-glance | 03:32 | |
*** ducttape_ has joined #openstack-glance | 03:35 | |
*** ducttape_ has quit IRC | 03:40 | |
*** dramakri has joined #openstack-glance | 03:42 | |
*** dramakri has quit IRC | 03:46 | |
*** annashen has quit IRC | 03:48 | |
*** annashen has joined #openstack-glance | 03:48 | |
*** zul has quit IRC | 03:52 | |
*** zul has joined #openstack-glance | 03:53 | |
*** dims has joined #openstack-glance | 04:00 | |
*** haomaiwang has quit IRC | 04:02 | |
*** haomaiwang has joined #openstack-glance | 04:04 | |
*** TravT has joined #openstack-glance | 04:06 | |
*** dims has quit IRC | 04:06 | |
*** TravT has quit IRC | 04:10 | |
*** annashen has quit IRC | 04:16 | |
*** annashen has joined #openstack-glance | 04:16 | |
*** kebray has joined #openstack-glance | 04:17 | |
*** harshs has joined #openstack-glance | 04:17 | |
*** annashen has quit IRC | 04:19 | |
*** annashen has joined #openstack-glance | 04:28 | |
*** annashen has quit IRC | 04:35 | |
*** tpeoples has quit IRC | 04:55 | |
*** TravT has joined #openstack-glance | 05:21 | |
*** TravT has quit IRC | 05:26 | |
*** TravT has joined #openstack-glance | 05:26 | |
*** TravT has quit IRC | 05:30 | |
*** harshs has quit IRC | 05:34 | |
*** annashen has joined #openstack-glance | 05:35 | |
*** lakshmiS has joined #openstack-glance | 05:38 | |
*** annashen has quit IRC | 05:40 | |
*** frickler_ is now known as frickler | 05:42 | |
*** groen692 has joined #openstack-glance | 05:59 | |
*** fifieldt has joined #openstack-glance | 06:00 | |
*** dims has joined #openstack-glance | 06:03 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/glance: Imported Translations from Transifex https://review.openstack.org/204905 | 06:04 |
---|---|---|
*** dims has quit IRC | 06:09 | |
*** markus_z has joined #openstack-glance | 06:10 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/glance_store: Imported Translations from Transifex https://review.openstack.org/204920 | 06:13 |
*** annashen has joined #openstack-glance | 06:14 | |
*** markus_z has quit IRC | 06:17 | |
*** annashen has quit IRC | 06:23 | |
*** lan has quit IRC | 06:27 | |
*** annashen has joined #openstack-glance | 06:29 | |
*** annashen has quit IRC | 06:31 | |
*** annashen has joined #openstack-glance | 06:33 | |
*** sgotliv_ has quit IRC | 06:33 | |
*** annashen has quit IRC | 06:39 | |
*** annashen has joined #openstack-glance | 06:46 | |
*** annashen has quit IRC | 06:49 | |
*** annashen has joined #openstack-glance | 06:51 | |
*** annashen has quit IRC | 06:53 | |
*** markus_z has joined #openstack-glance | 07:11 | |
*** kebray has quit IRC | 07:11 | |
*** [1]shalq has joined #openstack-glance | 07:25 | |
*** shalq has quit IRC | 07:25 | |
*** [1]shalq is now known as shalq | 07:25 | |
*** TravT has joined #openstack-glance | 07:27 | |
*** chlong has quit IRC | 07:28 | |
*** TravT has quit IRC | 07:31 | |
*** fifieldt has quit IRC | 07:42 | |
*** jistr has joined #openstack-glance | 07:44 | |
*** aarefiev has joined #openstack-glance | 08:07 | |
*** sgotliv has joined #openstack-glance | 08:09 | |
openstackgerrit | makocchi proposed openstack/glance: Keeping the configuration file with convention https://review.openstack.org/204800 | 08:10 |
*** GB21 has joined #openstack-glance | 08:11 | |
*** GB21_ has joined #openstack-glance | 08:11 | |
*** lakshmiS has quit IRC | 08:20 | |
*** MarcF has joined #openstack-glance | 08:23 | |
*** MattMan has joined #openstack-glance | 08:23 | |
openstackgerrit | Teruaki Ishizaki proposed openstack/glance_store: Sheepdog: improve operations for storing glance-image https://review.openstack.org/205942 | 08:27 |
*** Jhelf has joined #openstack-glance | 08:29 | |
jokke_ | sigmavirus24_awa: yep | 08:30 |
*** aix has joined #openstack-glance | 08:31 | |
*** e0ne has joined #openstack-glance | 08:32 | |
openstackgerrit | Teruaki Ishizaki proposed openstack/glance_store: Sheepdog: improve operations for storing glance-image https://review.openstack.org/205942 | 08:34 |
*** annashen has joined #openstack-glance | 08:38 | |
*** eandersson has joined #openstack-glance | 09:00 | |
*** dims has joined #openstack-glance | 09:06 | |
*** dims has quit IRC | 09:11 | |
*** annashen has quit IRC | 09:18 | |
*** TravT has joined #openstack-glance | 09:28 | |
*** jasondotstar has joined #openstack-glance | 09:32 | |
*** TravT has quit IRC | 09:32 | |
*** dims has joined #openstack-glance | 10:04 | |
*** dims_ has joined #openstack-glance | 10:06 | |
*** dims has quit IRC | 10:10 | |
*** mancdaz has quit IRC | 10:21 | |
*** mancdaz has joined #openstack-glance | 10:41 | |
*** lakshmiS has joined #openstack-glance | 10:54 | |
*** mingdang1 has quit IRC | 11:09 | |
*** shalq has quit IRC | 11:19 | |
*** shalq has joined #openstack-glance | 11:19 | |
*** TravT has joined #openstack-glance | 11:29 | |
*** marzif has joined #openstack-glance | 11:31 | |
*** TravT has quit IRC | 11:33 | |
*** markus_z has quit IRC | 11:35 | |
openstackgerrit | Darja Shakhray proposed openstack/glance: Glance to handle exceptions from glance_store https://review.openstack.org/200000 | 11:43 |
*** shalq has quit IRC | 11:48 | |
*** markus_z has joined #openstack-glance | 11:48 | |
*** marcusvrn has joined #openstack-glance | 12:02 | |
*** jasondotstar has quit IRC | 12:10 | |
*** annashen has joined #openstack-glance | 12:19 | |
*** annashen has quit IRC | 12:23 | |
openstackgerrit | Darja Shakhray proposed openstack/python-glanceclient: Add the ability client to create unicode properties in v2 https://review.openstack.org/206037 | 12:24 |
*** mingdang1 has joined #openstack-glance | 12:28 | |
*** edmondsw has joined #openstack-glance | 12:30 | |
*** chlong has joined #openstack-glance | 12:30 | |
*** jasondotstar has joined #openstack-glance | 12:32 | |
*** lakshmiS has quit IRC | 12:42 | |
*** lakshmiS has joined #openstack-glance | 12:43 | |
*** annegentle has joined #openstack-glance | 12:54 | |
openstackgerrit | Darja Shakhray proposed openstack/python-glanceclient: Add the ability client to create unicode properties in v2 https://review.openstack.org/206037 | 12:56 |
*** markus_z has quit IRC | 12:59 | |
*** markus_z has joined #openstack-glance | 13:03 | |
*** julim has joined #openstack-glance | 13:05 | |
*** edmondsw has quit IRC | 13:08 | |
*** annegentle has quit IRC | 13:10 | |
*** annegentle has joined #openstack-glance | 13:11 | |
*** GB21_ has quit IRC | 13:12 | |
flaper87 | o/ | 13:12 |
kragniz | the return of the flaper! | 13:13 |
*** ducttape_ has joined #openstack-glance | 13:20 | |
openstackgerrit | Merged openstack/glance: Keeping the configuration file with convention https://review.openstack.org/204800 | 13:23 |
*** ericpeterson has joined #openstack-glance | 13:26 | |
*** ducttape_ has quit IRC | 13:26 | |
*** TravT has joined #openstack-glance | 13:30 | |
*** mjturek1 has joined #openstack-glance | 13:31 | |
*** TravT has quit IRC | 13:34 | |
*** annegent_ has joined #openstack-glance | 13:38 | |
*** annegentle has quit IRC | 13:41 | |
*** chlong has quit IRC | 13:42 | |
*** pbourke has quit IRC | 13:45 | |
*** pbourke has joined #openstack-glance | 13:45 | |
*** jasondotstar has quit IRC | 13:48 | |
*** jistr has quit IRC | 13:52 | |
*** jistr has joined #openstack-glance | 13:54 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/glance: Updated from global requirements https://review.openstack.org/204393 | 13:55 |
*** julim_ has joined #openstack-glance | 13:57 | |
*** edmondsw has joined #openstack-glance | 13:57 | |
*** mfedosin_ has joined #openstack-glance | 13:59 | |
*** julim has quit IRC | 14:00 | |
ativelkov | Courtesy artifacts-meeting reminder: nikhil, ativelkov, mfedosin, sigmavirus24, flaper87, kragniz, dshakhray | 14:00 |
mfedosin_ | write the same message in openstack-glance :) | 14:01 |
mfedosin_ | I mean meeting-3, too | 14:02 |
kragniz | mfedosin_: this is openstack-glance :P | 14:02 |
ativelkov | it is in meeting-4 | 14:02 |
*** boris-42 has joined #openstack-glance | 14:02 | |
ativelkov | at least the therpad says so | 14:02 |
kragniz | ativelkov: yeah, it was changed from -3 | 14:03 |
*** exploreshaifali has joined #openstack-glance | 14:03 | |
*** sigmavirus24_awa is now known as sigmavirus24 | 14:04 | |
*** ericpeterson has quit IRC | 14:08 | |
*** mingdang1 has quit IRC | 14:13 | |
*** exploreshaifali_ has joined #openstack-glance | 14:26 | |
*** ayoung has joined #openstack-glance | 14:27 | |
*** jasondotstar has joined #openstack-glance | 14:28 | |
*** exploreshaifali has quit IRC | 14:28 | |
*** TravT has joined #openstack-glance | 14:31 | |
*** kodoku has joined #openstack-glance | 14:32 | |
*** changbl has quit IRC | 14:32 | |
kodoku | Hi, I upgrade juno to kilo but now I can't upload new image ==> Error in store configuration. Adding images to store is disabled. | 14:33 |
kodoku | I have filesystem_store_datadir = /var/lib/glance/images/ | 14:33 |
kodoku | Somebody can help me ? | 14:33 |
*** mfedosin_ has quit IRC | 14:35 | |
*** TravT has quit IRC | 14:35 | |
*** ducttape_ has joined #openstack-glance | 14:36 | |
*** r-daneel has joined #openstack-glance | 14:46 | |
*** mtanino has joined #openstack-glance | 14:48 | |
*** shaifali__ has joined #openstack-glance | 14:50 | |
*** exploreshaifali_ has quit IRC | 14:52 | |
*** annegent_ has quit IRC | 14:54 | |
*** tpeoples has joined #openstack-glance | 14:56 | |
*** kodoku has quit IRC | 15:03 | |
*** exploreshaifali_ has joined #openstack-glance | 15:04 | |
*** shaifali__ has quit IRC | 15:04 | |
*** exploreshaifali_ has quit IRC | 15:06 | |
*** kodoku has joined #openstack-glance | 15:16 | |
kodoku | Hi, I have this issue https://bugs.launchpad.net/python-glanceclient/+bug/1442664 . So I upgrade glance-client 0.17.0 to 0.17.2 But I have always trace AttributeError: 'PoolManager' object has no attribute 'connection_kw' in nova-api | 15:18 |
openstack | Launchpad bug 1442664 in python-glanceclient "Since 0.16.1 client breaks nova when using https" [Critical,Fix released] - Assigned to Stuart McLaren (stuart-mclaren) | 15:18 |
*** TravT has joined #openstack-glance | 15:18 | |
*** dims_ has quit IRC | 15:21 | |
*** tsekiyama has joined #openstack-glance | 15:21 | |
*** dims has joined #openstack-glance | 15:21 | |
*** aix has quit IRC | 15:22 | |
edmondsw | flaper87, can you explain https://github.com/openstack/glance/blob/master/glance/db/sqlalchemy/migrate_repo/versions/017_quote_encrypted_swift_credentials.py#L31-32 ? | 15:31 |
edmondsw | that module isn't referenced after it is imported... is it doing some init? | 15:32 |
edmondsw | this is currently the only reason to have swiftclient in requirements.txt... would like to remove that requirement if possible | 15:32 |
*** kebray has joined #openstack-glance | 15:33 | |
edmondsw | see https://review.openstack.org/#/c/203242] | 15:33 |
flaper87 | edmondsw: it was doing it when I added that line. I added it when we pulled glance/store out of glance. | 15:34 |
flaper87 | edmondsw: not sure if things changed after it on the swift side | 15:34 |
flaper87 | IIRC, it was registering some swift options require for the tests | 15:35 |
sigmavirus24 | Does anyone know the story behind glance.common.auth? Why are we using that instead of keystoneclient.auth? | 15:35 |
edmondsw | flaper87, so it may or may not still be needed? | 15:35 |
flaper87 | edmondsw: correct | 15:37 |
flaper87 | sigmavirus24: could it be related to the "no auth" thingy ? | 15:37 |
flaper87 | I don't recall, honestly. Just guessing | 15:37 |
sigmavirus24 | flaper87: ah | 15:37 |
sigmavirus24 | the keystone strategy should be using keystoneclient.auth because we currently can't handle v3 | 15:38 |
sigmavirus24 | (Keystone v30 | 15:38 |
sigmavirus24 | *v3) | 15:38 |
*** lakshmiS has quit IRC | 15:39 | |
openstackgerrit | Matthew Edmonds proposed openstack/glance: (WIP) remove store-specific requirements https://review.openstack.org/203242 | 15:42 |
sigmavirus24 | jokke_: ping | 15:43 |
kragniz | edmondsw: I think there should be more discussion before removing swiftclient from the dependancies, since I think it's currently assumed glance will work with swift as the default store out of the box | 15:44 |
edmondsw | kragniz, sure... let's discuss | 15:44 |
edmondsw | removing it from requirements shouldn't mean it can't be the default.. just that it's not required | 15:45 |
*** groen692 has quit IRC | 15:46 | |
openstackgerrit | Jesse J. Cook proposed openstack/glance-specs: Duplicate downloads spec https://review.openstack.org/206120 | 15:47 |
mfedosin | sigmavirus24, afair there will be oslo meeting in 10 minutes. are we going to ask about oslo.rbd there? | 15:48 |
kragniz | mfedosin: I was going to | 15:49 |
mfedosin | kragniz, cool! thanks | 15:49 |
kragniz | edmondsw: the issue is that someone installs glance, then gets a bunch of errors when they try to use it | 15:49 |
edmondsw | kragniz, if they didn't install the dependencies for the backend they want to use, yes... that's to be expected, not a problem | 15:50 |
edmondsw | they will need to understand what the dependencies are for their choice of config | 15:50 |
edmondsw | they need to understand that for non-default configs. They need to understand it for the default config as well | 15:51 |
edmondsw | listing things in requirements that aren't really requirements causes someone to have to install things that they don't need | 15:51 |
edmondsw | an operator who is not going to use the swift backend should not have to install swiftclient | 15:52 |
kragniz | this is true, and if we were starting from scratch I'd agree with you | 15:53 |
kragniz | I just don't want to make life harder for people updating to liberty for no reason | 15:53 |
edmondsw | kragniz, why would this be an upgrade issue? If they are upgrading, and they were using swift, they already have swift | 15:54 |
sigmavirus24 | mfedosin: I was goign to attend as well | 15:54 |
sigmavirus24 | edmondsw: except that we may require a higher version of swiftclient in liberty | 15:55 |
sigmavirus24 | And if we develop against features not present in the kilo swiftclient then things will break for no apparent reason | 15:55 |
sigmavirus24 | That's an upgrade issue | 15:55 |
edmondsw | test-requirements will still list swiftclient, so anyone can still see what version of swiftclient is necessary by looking at that | 15:56 |
kragniz | edmondsw: if, for example, they have a playbook to deploy glance they'll need to add an extra step to manually install python-swiftclient | 15:56 |
kragniz | I'm not stongly opposed to it, mind you | 15:56 |
kragniz | I just want to think about what issues we could be causing if we remove it | 15:56 |
sigmavirus24 | edmondsw: yes that's a really operator friendly way of doing things | 15:56 |
edmondsw | sigmavirus24, very little of what we do is terribly operator-friendly, unfortunately... :( | 15:57 |
edmondsw | I very much wish there was a better way of handling this for operators... but there isn't, today | 15:57 |
sigmavirus24 | edmondsw: yes so let's do more of it to be even less friendly | 15:57 |
edmondsw | I have to run... back later | 15:57 |
kragniz | there is http://specs.openstack.org/openstack/oslo-specs/specs/liberty/optional-deps.html | 15:57 |
kragniz | I'm not sure if it's only for libraries, though | 15:57 |
sigmavirus24 | kragniz: yeah I was thinking about using extras for this | 15:59 |
sigmavirus24 | but I'm not sure if we want extras in glance or in glance_store or in both | 15:59 |
sigmavirus24 | Also, I'm more concerned with how to add keystone v3 support in a way that we can possibly backport it to kilo | 16:00 |
kragniz | doesn't that sort of count as a feature? | 16:00 |
kragniz | (so a no-no for backporting) | 16:00 |
*** mgarza_ has joined #openstack-glance | 16:00 | |
jokke_ | sigmavirus24: pong | 16:05 |
sigmavirus24 | jokke_: can you imagine a way that we could add Keystone v3 support to Glance and make it backportable to Kilo? | 16:07 |
jokke_ | sigmavirus24: nope | 16:07 |
jokke_ | sigmavirus24: there is no way we're gonna break our api demanding new auth fields just to support that | 16:08 |
sigmavirus24 | We don't need to "demand" new field | 16:08 |
sigmavirus24 | *fields | 16:08 |
sigmavirus24 | We can collect them if they're present and use them only if the URL is for v3 | 16:09 |
sigmavirus24 | Kind of like how we do that for v2.0 in glance/common/auth.py | 16:09 |
jokke_ | sigmavirus24: didn't Keystone v3 need that domain field for auth? | 16:09 |
*** e0ne has quit IRC | 16:10 | |
sigmavirus24 | jokke_: it renames most fields, but we're not going to require people to use v3 for auth | 16:10 |
sigmavirus24 | That's not how this works (that's not how any of this works) | 16:10 |
jokke_ | sigmavirus24: I really don't see any of these being ticked https://wiki.openstack.org/wiki/StableBranch#Appropriate_Fixes for enabling the keystone v3 auth in Kilo | 16:11 |
sigmavirus24 | so I can imagine a way that this would be a tiny change that could be refactored in a dependent change for liberty to use keystoneclient.auth (which is what we should have been using anyway). There is user visible benefit (people trying to federate keystones, which requires v3 and will thusly make glance completely unusable) | 16:13 |
*** kodoku has quit IRC | 16:17 | |
jokke_ | sigmavirus24: So it _is_ new feature & it _does_ change API these two are alone big no-no. It does have quite a risk of regression as v3 is not even closely as well field proven as v2, it does not fix a user visible bug but enables deployer to utilize new feature. Seriously it belongs to new release, not stable backports :) | 16:18 |
*** mragupat has joined #openstack-glance | 16:18 | |
sigmavirus24 | jokke_: that's what I thought | 16:24 |
jokke_ | sigmavirus24: this is example of tiny fix that has big impact https://review.openstack.org/#/c/205432/ and yet it is not trivial to drop in. | 16:24 |
sigmavirus24 | let me guess. oslo.utils? | 16:24 |
sigmavirus24 | ah, no | 16:25 |
jokke_ | and I think only reason to backport that in is that otherwise we will have hell of a planning to upgrade to Liberty _if_ nova gets the v2 support together | 16:25 |
jokke_ | you cannot trust that new Nova would be able to talk to old Glance node and that ain't great result either | 16:26 |
jokke_ | and even with that change it gets messy for where the old schema is cached | 16:26 |
*** jasondotstar has quit IRC | 16:34 | |
openstackgerrit | Jesse J. Cook proposed openstack/glance-specs: Duplicate downloads spec https://review.openstack.org/206120 | 16:37 |
*** jasondotstar has joined #openstack-glance | 16:40 | |
*** snapdey has joined #openstack-glance | 16:41 | |
*** harshs has joined #openstack-glance | 16:47 | |
*** sgotliv has quit IRC | 16:49 | |
*** markus_z has quit IRC | 16:50 | |
*** jistr has quit IRC | 16:55 | |
*** snapdey has quit IRC | 16:57 | |
*** annashen has joined #openstack-glance | 16:59 | |
*** snapdey has joined #openstack-glance | 17:01 | |
kragniz | http://lists.openstack.org/pipermail/openstack-operators/2015-July/007767.html | 17:10 |
*** mingdang1 has joined #openstack-glance | 17:14 | |
*** mingdang1 has quit IRC | 17:19 | |
*** e0ne has joined #openstack-glance | 17:22 | |
*** snapdey has quit IRC | 17:24 | |
*** snapdey has joined #openstack-glance | 17:26 | |
mfedosin | sigmavirus24, kragniz I had a talk with e0ne and he likes the idea of oslo.rbd | 17:26 |
kragniz | mfedosin: cool | 17:27 |
kragniz | mfedosin: we should probably sit down and have a look at what code can be shared between the cinder driver and the glance store | 17:27 |
mfedosin | yep | 17:28 |
mfedosin | currently have no time on it, but next week is free and I can fill the etherpad with e0ne | 17:29 |
kragniz | sounds good | 17:30 |
*** snapdey has quit IRC | 17:36 | |
*** boris-42 has quit IRC | 17:40 | |
*** mgarza_ has quit IRC | 17:44 | |
*** harlowja has joined #openstack-glance | 17:52 | |
jokke_ | kragniz: ++ | 17:54 |
jokke_ | for removing CIS in Liberty | 17:54 |
mfedosin | kragniz, there is code in Nova to work with rbd https://github.com/openstack/nova/blob/master/nova/virt/libvirt/rbd_utils.py | 17:58 |
*** annegentle has joined #openstack-glance | 17:58 | |
mfedosin | so oslo.rbd is useful there too! | 17:58 |
kragniz | mfedosin: at the very least, we can move that parse_url function! | 17:59 |
kragniz | I think they all implement the same thing | 18:00 |
kragniz | in cinder it's called _parse_location, though | 18:00 |
mfedosin | and get_connection too | 18:00 |
kragniz | there's so much in common between nova and cinder | 18:00 |
*** e0ne has quit IRC | 18:01 | |
mfedosin | cinder was a part of Nova :) so, yes | 18:02 |
kragniz | in our hearts, we're all part of nova | 18:02 |
mfedosin | but we gemmated earlier | 18:05 |
*** drwahl has joined #openstack-glance | 18:05 | |
drwahl | we have multiple clusters. we're looking for a convenient way to take images in cluster A and uploading them to cluster B | 18:06 |
drwahl | any suggestions on how to accomplish this conveniently? | 18:07 |
*** snapdey has joined #openstack-glance | 18:10 | |
*** changbl has joined #openstack-glance | 18:12 | |
kragniz | drwahl: I think glance-replicator is what you're looking for | 18:13 |
kragniz | (I've not needed to do this myself, though, so I could be wrong) | 18:13 |
*** annegentle has quit IRC | 18:15 | |
*** annashen has quit IRC | 18:16 | |
*** annashen has joined #openstack-glance | 18:18 | |
drwahl | kragniz: i'll take a look at that.thanks! | 18:21 |
*** jasondot_ has joined #openstack-glance | 18:25 | |
*** annegentle has joined #openstack-glance | 18:27 | |
*** jasondotstar has quit IRC | 18:27 | |
*** ayoung has quit IRC | 18:30 | |
*** kebray has quit IRC | 18:30 | |
*** eandersson has quit IRC | 18:32 | |
*** snapdey has quit IRC | 18:36 | |
*** boris-42 has joined #openstack-glance | 18:42 | |
*** annegentle has quit IRC | 18:43 | |
*** kebray has joined #openstack-glance | 18:46 | |
*** sgotliv has joined #openstack-glance | 18:46 | |
*** mragupat_ has joined #openstack-glance | 18:50 | |
*** kebray has quit IRC | 18:51 | |
*** mragupat has quit IRC | 18:54 | |
openstackgerrit | Merged openstack/glance: Glance to handle exceptions from glance_store https://review.openstack.org/200000 | 18:54 |
*** spzala has joined #openstack-glance | 18:58 | |
*** annegentle has joined #openstack-glance | 19:02 | |
*** kebray has joined #openstack-glance | 19:11 | |
*** annegentle has quit IRC | 19:11 | |
*** annegentle has joined #openstack-glance | 19:12 | |
*** GB21 has quit IRC | 19:22 | |
*** snapdey has joined #openstack-glance | 19:22 | |
*** ayoung has joined #openstack-glance | 19:43 | |
*** harlowja has quit IRC | 19:45 | |
*** kebray has quit IRC | 19:56 | |
*** harlowja has joined #openstack-glance | 20:01 | |
*** mfedosin_ has joined #openstack-glance | 20:04 | |
*** mfedosin_ has quit IRC | 20:08 | |
*** dramakri has joined #openstack-glance | 20:09 | |
edmondsw | flaper87, Jenkins passed without https://github.com/openstack/glance/blob/master/glance/db/sqlalchemy/migrate_repo/versions/017_quote_encrypted_swift_credentials.py#L31-32 | 20:09 |
edmondsw | so I assume that means we could take it out... yes? | 20:10 |
edmondsw | kragniz, that was the last technical hurdle to removing swiftclient from requirements. How do we close on the concern you raised about glance using swift by default? | 20:12 |
edmondsw | sigmavirus24, ^^ | 20:12 |
*** snapdey has quit IRC | 20:14 | |
*** snapdey_ has joined #openstack-glance | 20:16 | |
*** flwang has quit IRC | 20:19 | |
*** kebray has joined #openstack-glance | 20:21 | |
*** kebray has quit IRC | 20:21 | |
*** spzala has quit IRC | 20:22 | |
*** kebray has joined #openstack-glance | 20:31 | |
lifeless | edmondsw: sigmavirus24: bystander comment - but extras might be appropriate. | 20:31 |
lifeless | assuming I read between the lines appropriately | 20:31 |
sigmavirus24 | lifeless: right, but we A) don't have them and this B) breaks default behaviour | 20:32 |
sigmavirus24 | I would say "add extras" with a warning in documentation that we're removing swiftclient in M | 20:32 |
sigmavirus24 | to ease upgrade concerns | 20:32 |
edmondsw | sigmavirus24, I would not say that this breaks anything | 20:32 |
lifeless | edmondsw: it absolutely will :) | 20:32 |
*** flwang has joined #openstack-glance | 20:32 | |
lifeless | edmondsw: someone somewhere will a_ depend on swiftclient being present and b) only have it installed due to the transitive deps | 20:33 |
lifeless | I don't think its major-version-bump bad | 20:33 |
lifeless | but its 'be gentle about it' territory | 20:33 |
*** dmsimard has joined #openstack-glance | 20:34 | |
*** dmsimard has left #openstack-glance | 20:34 | |
*** mragupat_ has quit IRC | 20:36 | |
*** mragupat has joined #openstack-glance | 20:37 | |
*** mragupat has quit IRC | 20:37 | |
edmondsw | lifeless, maybe... but they'll have it installed in an upgrade case because it was required in previous releases. | 20:37 |
edmondsw | They will only be broken if a) glance requires a newer version in liberty and b) they don't upgrade swiftclient for liberty | 20:37 |
*** annegentle has quit IRC | 20:37 | |
*** mragupat has joined #openstack-glance | 20:37 | |
*** julim_ has quit IRC | 20:38 | |
*** mragupat has quit IRC | 20:38 | |
edmondsw | b) assumes that global-requirements isn't also updated to have the Liberty-required version | 20:38 |
*** mragupat has joined #openstack-glance | 20:39 | |
edmondsw | if all that goes just wrong for someone... they'd quickly figure this out, right? | 20:39 |
stevelle | "quickly" | 20:39 |
stevelle | not so sure. | 20:40 |
sigmavirus24 | edmondsw: yes because everyone wants to debug this in production | 20:41 |
edmondsw | sigmavirus24 which is why you shouldn't do an upgrade in production until you've tried one not in production | 20:41 |
sigmavirus24 | edmondsw: because everyone has perfect mirrors for test of exactly every behaviour that happens in production, right? | 20:41 |
kragniz | lifeless: so could we use optional dependancies for this? | 20:43 |
edmondsw | this isn't some odd detail somewhere... this is basic stuff that they should have in a non-prod env | 20:43 |
kragniz | (they seem to be more focused on libraries) | 20:43 |
sigmavirus24 | kragniz: optional depenencies are essentially setting an extra for this which would be opt-in | 20:43 |
*** snapdey_ has quit IRC | 20:43 | |
sigmavirus24 | edmondsw: if they upgrade and do very light testing and we're relying on a new feature of a more up-to-date swiftclient, then they might not catch it, no | 20:44 |
sigmavirus24 | The old version would stay because the upgrade isn't forcing an upgrade to a new version of the client | 20:44 |
edmondsw | why are we worried about this case more than about the same case for non-default backends? | 20:45 |
edmondsw | which would have all the issues you're bringing up... today | 20:45 |
*** snapdey has joined #openstack-glance | 20:45 | |
edmondsw | if this is a problem that must be solved, it must be solved for non-default backends just as much as for the default one | 20:46 |
sigmavirus24 | edmondsw: except that people deploying non-default backends already handle this themselves through various means | 20:47 |
*** harlowja_ has joined #openstack-glance | 20:47 | |
edmondsw | right... and those same various means can be used for people deploying with the default backend | 20:47 |
sigmavirus24 | That said, I'm +1 on utilizing extras, and -1 on removing swiftclient before Liberty is released | 20:47 |
sigmavirus24 | edmondsw: except you're fundamentally changing default behaviour by removing our handling of the swiftclient dependency for them in requirements.txt | 20:48 |
sigmavirus24 | You're changing *default* behaviour without a deprecation cycle | 20:48 |
edmondsw | i.e., you're ok with removing swiftclient from requirements, but only in M, after we do something in Liberty (what?) to warn them that this will be removed in M? | 20:48 |
kragniz | edmondsw: I totally agree it should be solved, but I don't want to just leave ops out in the cold for the most common way of deploying glance | 20:49 |
edmondsw | I don't want to leave ops in the cold any more than you guys... I think we're just on different pages as to what leaves them in the cold and how to avoid that | 20:49 |
edmondsw | so... how would you guys suggest fixing this? | 20:50 |
edmondsw | we don't have extras yet, right? So... wait for that? And then ops have to know to specify that, right? | 20:50 |
*** harlowja has quit IRC | 20:51 | |
kragniz | maybe we should take this to the ML | 20:52 |
kragniz | others may have had this same discussion | 20:52 |
edmondsw | I'm fine with that | 20:53 |
kragniz | I personally think we should not change until we have a better option, but I'd like to hear what others have to say about it | 20:53 |
edmondsw | kragniz, in the meantime, I think we can go ahead and agree to proceed with your https://review.openstack.org/#/c/203200 just removing oslo.vmware | 20:54 |
kragniz | edmondsw: also, it's the glance mid-cycle this week, so a finding cores around online might be a little harder | 20:56 |
kragniz | edmondsw: good idea | 20:56 |
kragniz | jokke_: mind updating your review on that? ^ | 20:56 |
*** annegentle has joined #openstack-glance | 21:06 | |
*** annashen has quit IRC | 21:14 | |
*** annashen has joined #openstack-glance | 21:39 | |
openstackgerrit | Matthew Edmonds proposed openstack/glance: (WIP) remove python-swiftclient from requirements https://review.openstack.org/203242 | 21:40 |
*** ccrouch has quit IRC | 21:42 | |
edmondsw | ^ that is just cleanup to remove the olso.vmware change, for which we'll use https://review.openstack.org/#/c/203200 | 21:43 |
edmondsw | kragniz, sigmavirus24 http://lists.openstack.org/pipermail/openstack-dev/2015-July/070699.html | 21:45 |
*** julim has joined #openstack-glance | 21:46 | |
edmondsw | sigmavirus24, why did you remove +1 on https://review.openstack.org/#/c/203200 ? We should be able to go ahead and fix the oslo.vmware bug without waiting on the ML discussion | 21:46 |
*** snapdey has quit IRC | 21:50 | |
*** 16WAADQEM has joined #openstack-glance | 21:52 | |
openstackgerrit | Matthew Edmonds proposed openstack/glance: (WIP) remove python-swiftclient from requirements https://review.openstack.org/203242 | 21:52 |
*** sgotliv has quit IRC | 21:55 | |
*** mragupat has quit IRC | 21:57 | |
*** chlong has joined #openstack-glance | 22:09 | |
*** edmondsw has quit IRC | 22:14 | |
*** jasondot_ has quit IRC | 22:15 | |
*** dramakri has quit IRC | 22:17 | |
jamielennox | kragniz: do you know what's happenign with the glance_store gate? https://review.openstack.org/#/c/193422/ is still failing and i don't think it's me | 22:18 |
jamielennox | or anyone ^ | 22:20 |
jamielennox | i would like to get that merged | 22:20 |
*** harlowja_ has quit IRC | 22:21 | |
*** harlowja has joined #openstack-glance | 22:21 | |
sigmavirus24 | jamielennox: looks like you have two failing tempest tests there: http://logs.openstack.org/22/193422/3/check/gate-tempest-dsvm-full-ceph-src-glance_store/098988f/console.html#_2015-07-25_10_43_05_982 and http://logs.openstack.org/22/193422/3/check/gate-tempest-dsvm-full-ceph-src-glance_store/098988f/console.html#_2015-07-25_10_46_50_027 | 22:23 |
*** tpeoples has quit IRC | 22:23 | |
jamielennox | sigmavirus24: yep, but it looked like a general failure in glance_store: v | 22:24 |
jamielennox | https://review.openstack.org/#/c/204920/ | 22:24 |
jamielennox | was wondering if anyone else had looked | 22:25 |
*** marcusvrn has quit IRC | 22:25 | |
sigmavirus24 | that's also the ceph gate | 22:25 |
sigmavirus24 | probably a problem with the ceph gates | 22:25 |
sigmavirus24 | jamielennox: I think mriedem added that a couple weeks ago, I'm not sure it's passed since then | 22:26 |
sigmavirus24 | might be worth bugging them about | 22:26 |
jamielennox | sigmavirus24: bugging who? | 22:26 |
sigmavirus24 | mriedem | 22:26 |
jamielennox | because of nova or infra? | 22:27 |
sigmavirus24 | Because I think mriedem was the reason the ceph gate was added | 22:27 |
jamielennox | ok, cool | 22:27 |
sigmavirus24 | And they seem to know a lot about ceph because glance_store apparently broke ceph gates previously which is why it was added to glance_store | 22:27 |
sigmavirus24 | But those failures are unrelated to glance_store at the moment it seems | 22:28 |
*** dramakri has joined #openstack-glance | 22:39 | |
*** dramakri has quit IRC | 22:39 | |
openstackgerrit | Jesse J. Cook proposed openstack/glance-specs: Upload retry spec https://review.openstack.org/206250 | 22:46 |
*** annegentle has quit IRC | 22:50 | |
lifeless | kragniz: optional deps let you say 'for feature Y, bring in deps A,B,C,D' | 22:50 |
lifeless | kragniz: so if Y is something folk 'opt in to' | 22:51 |
lifeless | then an extra makes sense | 22:51 |
lifeless | if Y is something folk expect to Just Work, then it doesn't, and it should be a hard dep | 22:51 |
*** 16WAADQEM has quit IRC | 23:02 | |
*** r-daneel has quit IRC | 23:02 | |
*** snapdey has joined #openstack-glance | 23:05 | |
*** ducttape_ has quit IRC | 23:05 | |
*** arahal has joined #openstack-glance | 23:06 | |
*** jasondotstar has joined #openstack-glance | 23:08 | |
*** annashen has quit IRC | 23:12 | |
*** jasondotstar has quit IRC | 23:13 | |
*** annashen has joined #openstack-glance | 23:17 | |
*** annashen has quit IRC | 23:23 | |
jokke_ | lifeless: still there? | 23:27 |
*** jasondot_ has joined #openstack-glance | 23:28 | |
jokke_ | I'd like to remind that swift store as backend is not any more default than vmware or ceph, our only default store is file | 23:30 |
*** arahal has quit IRC | 23:37 | |
*** arahal has joined #openstack-glance | 23:37 | |
*** arahal has quit IRC | 23:37 | |
*** harshs has quit IRC | 23:41 | |
*** jasondotstar has joined #openstack-glance | 23:44 | |
*** jasondot_ has quit IRC | 23:47 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!