| *** claudiub|2 has quit IRC | 00:02 | |
| *** mlavalle has quit IRC | 00:10 | |
| *** hongbin has quit IRC | 00:10 | |
| *** jwcroppe has quit IRC | 00:13 | |
| *** zul has quit IRC | 00:13 | |
| *** oanson has quit IRC | 00:16 | |
| *** ijw has quit IRC | 00:18 | |
| *** ijw has joined #openstack-nova | 00:19 | |
| *** hfu has joined #openstack-nova | 00:20 | |
| *** lamt has quit IRC | 00:25 | |
| *** hfu has quit IRC | 00:28 | |
| *** hfu has joined #openstack-nova | 00:29 | |
| *** lamt has joined #openstack-nova | 00:30 | |
| *** sach-m_ has quit IRC | 00:31 | |
| *** ijw has quit IRC | 00:31 | |
| *** sach-m has joined #openstack-nova | 00:32 | |
| *** sach-m has quit IRC | 00:38 | |
| *** rama_y has quit IRC | 00:38 | |
| *** sach-m has joined #openstack-nova | 00:39 | |
| *** browne has quit IRC | 00:42 | |
| *** gmann_ has joined #openstack-nova | 00:43 | |
| *** zul has joined #openstack-nova | 00:43 | |
| *** takashin has joined #openstack-nova | 00:45 | |
| *** wanghao_ has joined #openstack-nova | 00:45 | |
| *** Swami has quit IRC | 00:46 | |
| *** tlian has quit IRC | 00:49 | |
| *** bvanhav has joined #openstack-nova | 00:50 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova-specs: Abort Cold Migration https://review.openstack.org/334732 | 00:50 |
|---|---|---|
| openstackgerrit | melanie witt proposed openstack/nova: Detach volume after deleting instance with no host https://review.openstack.org/340614 | 00:52 |
| *** amotoki has joined #openstack-nova | 00:52 | |
| *** ijw has joined #openstack-nova | 00:52 | |
| *** hoangcx has joined #openstack-nova | 00:54 | |
| *** owalsh_ has joined #openstack-nova | 00:55 | |
| *** bdeschenes has quit IRC | 00:55 | |
| *** owalsh has quit IRC | 00:56 | |
| *** ijw has quit IRC | 00:57 | |
| *** amotoki has quit IRC | 00:57 | |
| *** gjayavelu1 has joined #openstack-nova | 00:58 | |
| *** jwcroppe has joined #openstack-nova | 00:59 | |
| *** ijw has joined #openstack-nova | 01:02 | |
| *** gjayavelu has quit IRC | 01:02 | |
| *** phuongnh has joined #openstack-nova | 01:02 | |
| *** ijw has quit IRC | 01:07 | |
| *** Sukhdev has quit IRC | 01:22 | |
| *** gmann has joined #openstack-nova | 01:22 | |
| *** gmann_ has quit IRC | 01:23 | |
| *** walharthi has quit IRC | 01:23 | |
| *** TravT has joined #openstack-nova | 01:29 | |
| *** yuntongjin has joined #openstack-nova | 01:29 | |
| *** victorhe has joined #openstack-nova | 01:29 | |
| *** amotoki has joined #openstack-nova | 01:29 | |
| *** owalsh_ is now known as owalsh | 01:30 | |
| *** yuntongjin has quit IRC | 01:30 | |
| *** victorhe has quit IRC | 01:32 | |
| *** victorhe has joined #openstack-nova | 01:33 | |
| *** browne has joined #openstack-nova | 01:33 | |
| *** browne has quit IRC | 01:34 | |
| *** ijw has joined #openstack-nova | 01:37 | |
| *** sree has joined #openstack-nova | 01:37 | |
| *** bvanhav has quit IRC | 01:37 | |
| *** takashin has quit IRC | 01:38 | |
| *** rrecio has joined #openstack-nova | 01:38 | |
| *** rrecio_ has joined #openstack-nova | 01:40 | |
| *** hfu has quit IRC | 01:40 | |
| *** thorst has joined #openstack-nova | 01:41 | |
| openstackgerrit | Leehom Li proposed openstack/nova: Support detach interface with same MAC from instance https://review.openstack.org/372243 | 01:41 |
| *** sree has quit IRC | 01:42 | |
| *** rrecio has quit IRC | 01:43 | |
| *** thorst has quit IRC | 01:45 | |
| *** mriedem has joined #openstack-nova | 01:45 | |
| *** shuquan has joined #openstack-nova | 01:48 | |
| *** thorst has joined #openstack-nova | 01:49 | |
| *** gcb has joined #openstack-nova | 01:50 | |
| *** hfu has joined #openstack-nova | 01:50 | |
| *** xinliang has quit IRC | 01:50 | |
| *** thorst has quit IRC | 01:51 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add microversion cap information https://review.openstack.org/396514 | 01:52 |
| mriedem | dansmith: do you know if nhicher is still planning on working on this? https://blueprints.launchpad.net/nova/+spec/purge-deleted-instances-cmd | 01:52 |
| dansmith | mriedem: is that a redhat person? | 01:54 |
| mriedem | yeah | 01:54 |
| mriedem | nicolas hicher | 01:55 |
| *** links has joined #openstack-nova | 01:55 | |
| dansmith | I don't think I recognize that name | 01:55 |
| dansmith | oh I see, on the assignee | 01:56 |
| dansmith | huh | 01:56 |
| dansmith | dunno | 01:56 |
| dansmith | kinda doubt it | 01:56 |
| *** lamt has quit IRC | 01:57 | |
| *** tonytan4ever has joined #openstack-nova | 02:01 | |
| *** RuiChen has joined #openstack-nova | 02:02 | |
| melwitt | mriedem: is that like a special case of archive_deleted_rows? only instances and no move to shadow tables? | 02:02 |
| *** shuquan_ has joined #openstack-nova | 02:02 | |
| *** shuquan has quit IRC | 02:03 | |
| *** xinliang has joined #openstack-nova | 02:03 | |
| mriedem | melwitt: right no move to shadow tables, just purge | 02:04 |
| mriedem | with a time window i think was proposed, | 02:04 |
| mriedem | like all instances and related records after x days | 02:04 |
| melwitt | gotcha | 02:05 |
| *** links has quit IRC | 02:07 | |
| mriedem | ayoung: jamielennox: on this spec for nova to pass a service token to other services to re-auth in case the user token expires, https://review.openstack.org/#/c/387711/7 | 02:08 |
| mriedem | ayoung: what happens if keystone API isn't new enough for that new header? | 02:08 |
| jamielennox | mriedem: it'll be ignored | 02:09 |
| jamielennox | so essentially the same behavior as today | 02:09 |
| *** shuquan_ has quit IRC | 02:09 | |
| mriedem | "We add a flag ?allow_expired=1 to the existing GET /auth/tokens validation route that allows fetching an expired token." | 02:09 |
| mriedem | so if keystonemiddleware is new enough it can send that query paramter to the tokens REST API, | 02:10 |
| mriedem | but if keystone API isn't new enough it's just going to not know how to handle that query parameter | 02:10 |
| mriedem | and nothing happens | 02:10 |
| mriedem | rigth? | 02:10 |
| mriedem | *right | 02:10 |
| *** siva_krish has joined #openstack-nova | 02:11 | |
| jamielennox | yes | 02:13 |
| *** tjones has quit IRC | 02:13 | |
| jamielennox | meaning it will be no worse than now | 02:13 |
| mriedem | raj_singh: got some things that could be cleaned up in here in a follow up, but +W https://review.openstack.org/#/c/387711/ | 02:13 |
| mriedem | jamielennox: awesome, thanks for clarifying | 02:14 |
| *** shuquan has joined #openstack-nova | 02:15 | |
| *** thorst has joined #openstack-nova | 02:20 | |
| *** shuquan has quit IRC | 02:20 | |
| *** nic has quit IRC | 02:21 | |
| *** walharthi has joined #openstack-nova | 02:23 | |
| *** hfu has quit IRC | 02:29 | |
| *** owalsh has quit IRC | 02:29 | |
| *** owalsh has joined #openstack-nova | 02:31 | |
| *** victorhe has quit IRC | 02:32 | |
| *** hfu has joined #openstack-nova | 02:32 | |
| *** victorhe has joined #openstack-nova | 02:35 | |
| *** sree has joined #openstack-nova | 02:36 | |
| *** kenperkins has quit IRC | 02:38 | |
| *** Apoorva has quit IRC | 02:39 | |
| *** thorst has quit IRC | 02:40 | |
| *** sree has quit IRC | 02:41 | |
| *** armax has quit IRC | 02:44 | |
| *** yuntongjin has joined #openstack-nova | 02:45 | |
| openstackgerrit | int32bit proposed openstack/python-novaclient: Change fake server id as str to fit real server id type https://review.openstack.org/398050 | 02:50 |
| *** shuquan has joined #openstack-nova | 02:50 | |
| mriedem | mikal: comments in https://review.openstack.org/#/c/395959/ - mostly ok but i think we could use some more details on the user role thing | 02:53 |
| mriedem | i didn't grok most of that discussion during the summit session | 02:53 |
| *** unicell has quit IRC | 02:56 | |
| *** ijw has quit IRC | 03:02 | |
| *** baoli has joined #openstack-nova | 03:03 | |
| *** zhangyufei has joined #openstack-nova | 03:04 | |
| *** tbachman has quit IRC | 03:07 | |
| *** mfeoktistov has joined #openstack-nova | 03:07 | |
| *** armax has joined #openstack-nova | 03:07 | |
| *** ijw has joined #openstack-nova | 03:08 | |
| openstackgerrit | Li Yingjun proposed openstack/nova: No return for flavor destroy https://review.openstack.org/398060 | 03:09 |
| *** victorhe has quit IRC | 03:12 | |
| *** victorhe has joined #openstack-nova | 03:13 | |
| *** ijw has quit IRC | 03:13 | |
| *** victorhe has quit IRC | 03:17 | |
| *** mriedem has quit IRC | 03:18 | |
| *** siva_krish has quit IRC | 03:19 | |
| *** gjayavelu1 has quit IRC | 03:20 | |
| *** yuntongjin has quit IRC | 03:28 | |
| *** hongbin has joined #openstack-nova | 03:29 | |
| *** hfu has quit IRC | 03:35 | |
| *** zhangyufei has quit IRC | 03:35 | |
| *** shuquan has quit IRC | 03:35 | |
| *** hfu has joined #openstack-nova | 03:35 | |
| *** tjones has joined #openstack-nova | 03:39 | |
| *** armax has quit IRC | 03:39 | |
| *** rrecio_ has quit IRC | 03:41 | |
| *** fragatin_ has joined #openstack-nova | 03:47 | |
| *** manasm has joined #openstack-nova | 03:49 | |
| *** fragatina has quit IRC | 03:51 | |
| *** fragatin_ has quit IRC | 03:52 | |
| *** baoli has quit IRC | 03:52 | |
| *** psachin has joined #openstack-nova | 03:54 | |
| *** siva_krish has joined #openstack-nova | 04:01 | |
| *** siva_krish has quit IRC | 04:01 | |
| *** Jeffrey4l has joined #openstack-nova | 04:03 | |
| *** amotoki has quit IRC | 04:09 | |
| *** GB21 has joined #openstack-nova | 04:12 | |
| *** kwathore__ has joined #openstack-nova | 04:14 | |
| *** Sukhdev has joined #openstack-nova | 04:14 | |
| *** armax has joined #openstack-nova | 04:16 | |
| *** manasm has quit IRC | 04:18 | |
| *** takedakn has joined #openstack-nova | 04:21 | |
| *** hongbin has quit IRC | 04:23 | |
| *** tonytan4ever has quit IRC | 04:23 | |
| *** tonytan4ever has joined #openstack-nova | 04:23 | |
| *** sudipto has joined #openstack-nova | 04:24 | |
| *** bks has joined #openstack-nova | 04:24 | |
| *** walharthi has quit IRC | 04:25 | |
| *** takedakn has quit IRC | 04:26 | |
| *** victorhe has joined #openstack-nova | 04:26 | |
| *** kwathore__ has quit IRC | 04:26 | |
| *** kwathore__ has joined #openstack-nova | 04:27 | |
| *** tonytan4ever has quit IRC | 04:28 | |
| *** nicolasbock has quit IRC | 04:28 | |
| *** victorhe has quit IRC | 04:30 | |
| *** takedakn has joined #openstack-nova | 04:31 | |
| *** yamahata has quit IRC | 04:32 | |
| *** rdo has quit IRC | 04:34 | |
| *** rdo has joined #openstack-nova | 04:36 | |
| *** khamtamtun has joined #openstack-nova | 04:40 | |
| *** khamtamtun has quit IRC | 04:44 | |
| *** krtaylor has joined #openstack-nova | 04:45 | |
| *** ratailor has joined #openstack-nova | 04:51 | |
| *** sree has joined #openstack-nova | 04:51 | |
| *** sudipto has quit IRC | 04:53 | |
| *** archies has joined #openstack-nova | 04:54 | |
| *** manasm has joined #openstack-nova | 04:55 | |
| *** zhangyufei has joined #openstack-nova | 04:55 | |
| *** kwathore__ has quit IRC | 04:55 | |
| *** kwathore has joined #openstack-nova | 04:55 | |
| *** walharthi has joined #openstack-nova | 04:56 | |
| openstackgerrit | Huan Xie proposed openstack/nova: xenapi: support the hotplug of a neutron port https://review.openstack.org/365505 | 04:56 |
| *** kenperkins has joined #openstack-nova | 04:57 | |
| *** shuquan has joined #openstack-nova | 04:57 | |
| *** tjones has left #openstack-nova | 04:59 | |
| *** ijw has joined #openstack-nova | 05:01 | |
| *** kwathore has quit IRC | 05:07 | |
| *** kwathore has joined #openstack-nova | 05:07 | |
| *** yuntongjin has joined #openstack-nova | 05:07 | |
| *** takedakn has quit IRC | 05:14 | |
| *** claudiub has joined #openstack-nova | 05:15 | |
| *** tbachman has joined #openstack-nova | 05:16 | |
| *** darrenc is now known as darrenc_afk | 05:19 | |
| *** dims has quit IRC | 05:20 | |
| *** takedakn has joined #openstack-nova | 05:22 | |
| *** Sukhdev has quit IRC | 05:25 | |
| *** prateek has joined #openstack-nova | 05:27 | |
| *** tjones has joined #openstack-nova | 05:32 | |
| *** gongysh has joined #openstack-nova | 05:32 | |
| *** rama_y has joined #openstack-nova | 05:34 | |
| *** tbachman_ has joined #openstack-nova | 05:37 | |
| *** tbachman has quit IRC | 05:38 | |
| *** tbachman_ is now known as tbachman | 05:38 | |
| *** tbachman has quit IRC | 05:39 | |
| *** gjayavelu has joined #openstack-nova | 05:40 | |
| *** manasm has quit IRC | 05:40 | |
| *** manasm has joined #openstack-nova | 05:42 | |
| *** tbachman has joined #openstack-nova | 05:42 | |
| *** sandanar has joined #openstack-nova | 05:42 | |
| *** moshele has joined #openstack-nova | 05:44 | |
| *** gjayavelu1 has joined #openstack-nova | 05:48 | |
| *** gjayavelu has quit IRC | 05:49 | |
| *** zhangyufei has quit IRC | 05:50 | |
| *** sandanar has quit IRC | 05:51 | |
| *** tjones has quit IRC | 05:51 | |
| *** walharthi has quit IRC | 05:51 | |
| *** harlowja has quit IRC | 05:55 | |
| *** gongysh has quit IRC | 05:56 | |
| *** kenperkins has quit IRC | 05:57 | |
| *** darrenc_afk is now known as darrenc | 05:57 | |
| *** unicell has joined #openstack-nova | 05:58 | |
| *** gcb has quit IRC | 05:58 | |
| *** dims has joined #openstack-nova | 05:59 | |
| *** gcb has joined #openstack-nova | 05:59 | |
| oomichi | alex_xu: https://review.openstack.org/#/c/388518/ seems good way, did we get a consensus about that in the summit? | 05:59 |
| *** hshiina has joined #openstack-nova | 06:00 | |
| alex_xu | oomichi: yea, people think this is priority task in the release | 06:00 |
| *** kwathore_ has joined #openstack-nova | 06:01 | |
| oomichi | alex_xu: cool, I prefer that :) | 06:01 |
| alex_xu | oomichi: thanks, also checking your comment now :) | 06:02 |
| oomichi | alex_xu: basically, I just want to know the status code(400?) | 06:02 |
| oomichi | in invalid query cases | 06:03 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/397916 | 06:03 |
| *** kwathore has quit IRC | 06:04 | |
| alex_xu | oomichi: yea, it should be 400, I clarify that | 06:04 |
| oomichi | +1 on 400 | 06:05 |
| alex_xu | oomichi: are you agree with Chris's point? | 06:05 |
| oomichi | alex_xu: cdent points are many. maybe list structure point? | 06:07 |
| alex_xu | oomichi: ah, sorry, the line 113 one | 06:07 |
| *** yuntongjin has quit IRC | 06:08 | |
| oomichi | alex_xu: as your code, the list structure seems good for me because the query itself should be simple and we don't need to take care of multi-level structure | 06:08 |
| *** abalutoiu has joined #openstack-nova | 06:09 | |
| alex_xu | oomichi: yea | 06:09 |
| *** pbandark has joined #openstack-nova | 06:09 | |
| *** amotoki has joined #openstack-nova | 06:10 | |
| alex_xu | oomichi: actually the Chris concern on the single value should be list also. just line 117, will be {'sort_key': ['created_at']} | 06:11 |
| oomichi | alex_xu: but I can see cdent point also because a little difficult to know available queries(eg. "a=1&a=2" doesn't seem a list) from the list structure of json-schema | 06:11 |
| *** victorhe has joined #openstack-nova | 06:11 | |
| oomichi | alex_xu: ah, I see | 06:12 |
| oomichi | alex_xu: maybe the list implementation for a single value would become easy I guess | 06:12 |
| *** lpetrut has joined #openstack-nova | 06:12 | |
| alex_xu | I think finally we want to return 400 for 'a' only accpet single value, but user specify multiple one in the url like 'a=1&a=2' | 06:13 |
| alex_xu | oomichi: if that is a list, we can implement ^ above validation | 06:13 |
| alex_xu | s/can/can't/ | 06:13 |
| openstackgerrit | Leehom Li proposed openstack/nova: Support detach interface with same MAC from instance https://review.openstack.org/372243 | 06:14 |
| oomichi | alex_xu: you are saing "a=1&a=2" becomes 400, right? | 06:14 |
| *** abalutoiu has quit IRC | 06:15 | |
| alex_xu | oomichi: yes | 06:15 |
| oomichi | alex_xu: so that should be enum instead of list? | 06:15 |
| oomichi | or oneOf | 06:15 |
| oomichi | As common json-schema, oneOf seems clear meaning for this case | 06:16 |
| *** zhangyufei has joined #openstack-nova | 06:16 | |
| *** jaosorior has joined #openstack-nova | 06:16 | |
| alex_xu | oomichi: let us use one realy example. The 'name' in the server list API. If the request is "name=abc&name=def", actually the API only filter by the last one 'name=def'. The first one was ignored | 06:17 |
| *** victorhe has quit IRC | 06:17 | |
| *** ekuris has joined #openstack-nova | 06:18 | |
| alex_xu | oomichi: in the future whether we want to return 400, when user request "name=abc&name=def", then tell user, you only can pass only one "name" parameter in the url | 06:18 |
| alex_xu | s/realy/real/ | 06:18 |
| oomichi | alex_xu: oh, good explanation. easy understanding for me | 06:19 |
| alex_xu | oomichi: :) | 06:19 |
| openstackgerrit | Leehom Li proposed openstack/nova: Support detach interface with same MAC from instance https://review.openstack.org/372243 | 06:21 |
| gmann | alex_xu: oomichi : yea, but not sure its doable with json schema ? | 06:22 |
| gmann | alex_xu: oomichi with array we can check uniqeness but on object there is no field yet | 06:22 |
| alex_xu | gmann: we can implement it, by convert the query parameters into a dict, just like the line 86 at https://review.openstack.org/#/c/388518/9/specs/ocata/approved/consistent-query-parameters-validation.rst | 06:23 |
| oomichi | gmann: yeah, I guess general json-schema doesn't support such way | 06:24 |
| gmann | alex_xu: yea for query param we can do in our conversion method | 06:24 |
| gmann | alex_xu: oomichi but same issue we have for request body also | 06:24 |
| gmann | alex_xu: oomichi i was thinking to add this support in json schema lib | 06:25 |
| gmann | which we can used in tempest also | 06:25 |
| gmann | but need to check their process for adding that or they can add | 06:25 |
| *** tbachman has quit IRC | 06:26 | |
| *** markvoelker has quit IRC | 06:28 | |
| alex_xu | oomichi: gmann, fyi, there is second part of this task https://review.openstack.org/#/c/393205/5, appreciate feedback also! | 06:29 |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add microversion cap information https://review.openstack.org/396514 | 06:30 |
| openstackgerrit | Leehom Li proposed openstack/nova: Support detach interface with same MAC from instance https://review.openstack.org/372243 | 06:31 |
| *** harlowja_at_home has joined #openstack-nova | 06:32 | |
| *** harlowja_at_home has quit IRC | 06:33 | |
| oomichi | alex_xu: about the above spec, do we want to put the whitelist of available filters as the decorators on API method? | 06:33 |
| alex_xu | oomichi: yea, the schema is in the REST API impact section | 06:33 |
| oomichi | alex_xu: humm, that seems your spec can cover that, I just feel. Do we need to separate that from yours? | 06:34 |
| alex_xu | oomichi: yea, johnthetubaguy suggest we have separate one to clear about the white list. you can find there are a lot of query parameters are removed in that proposal | 06:35 |
| oomichi | alex_xu: ok, I will check it deeply later, thanks for picking it up | 06:36 |
| alex_xu | oomichi: thanks for the review also | 06:36 |
| gmann | alex_xu: thanks, ll check that | 06:39 |
| alex_xu | gmann: thanks | 06:39 |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add microversion cap information https://review.openstack.org/396514 | 06:39 |
| *** jaosorior has quit IRC | 06:41 | |
| *** jaosorior has joined #openstack-nova | 06:41 | |
| *** lpetrut has quit IRC | 06:42 | |
| *** oanson has joined #openstack-nova | 06:44 | |
| *** rama_y has quit IRC | 06:45 | |
| *** walharthi has joined #openstack-nova | 06:48 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Add microversion cap information https://review.openstack.org/396514 | 06:50 |
| *** moshele has quit IRC | 06:52 | |
| *** armax has quit IRC | 06:52 | |
| *** walharthi has quit IRC | 06:53 | |
| *** cfriesen has quit IRC | 06:54 | |
| *** ljxiash has joined #openstack-nova | 07:01 | |
| *** liuyulong__ has joined #openstack-nova | 07:04 | |
| *** liuyulong_ has quit IRC | 07:07 | |
| *** edand has joined #openstack-nova | 07:08 | |
| *** unicell has quit IRC | 07:10 | |
| *** tuanluong has joined #openstack-nova | 07:10 | |
| *** unicell has joined #openstack-nova | 07:11 | |
| *** mjura has joined #openstack-nova | 07:11 | |
| *** GB21 has quit IRC | 07:16 | |
| *** mjura has quit IRC | 07:19 | |
| *** mjura has joined #openstack-nova | 07:19 | |
| *** edand has quit IRC | 07:26 | |
| *** alexchadin has joined #openstack-nova | 07:28 | |
| *** andreas_s has joined #openstack-nova | 07:29 | |
| *** edand has joined #openstack-nova | 07:29 | |
| *** GB21 has joined #openstack-nova | 07:29 | |
| *** manasm has quit IRC | 07:32 | |
| *** manasm has joined #openstack-nova | 07:33 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Remove mox in tests/unit/objects/test_instance_info_cache https://review.openstack.org/307155 | 07:33 |
| *** jaosorior has quit IRC | 07:34 | |
| *** nherciu has joined #openstack-nova | 07:35 | |
| openstackgerrit | zhaolihui proposed openstack/nova: Fix failed to boot vm with pci_alias's name including space https://review.openstack.org/392077 | 07:36 |
| *** houming has joined #openstack-nova | 07:36 | |
| *** lpetrut has joined #openstack-nova | 07:37 | |
| *** manasm has quit IRC | 07:38 | |
| *** manasm has joined #openstack-nova | 07:38 | |
| *** pcaruana has joined #openstack-nova | 07:41 | |
| *** lpetrut has quit IRC | 07:41 | |
| *** rha has joined #openstack-nova | 07:42 | |
| *** markus_z has joined #openstack-nova | 07:42 | |
| *** moshele has joined #openstack-nova | 07:42 | |
| *** walharthi has joined #openstack-nova | 07:49 | |
| *** ihrachys has joined #openstack-nova | 07:53 | |
| *** jaosorior has joined #openstack-nova | 07:54 | |
| *** yuntongjin has joined #openstack-nova | 07:58 | |
| *** gjayavelu1 has quit IRC | 07:59 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Remove mox in tests/unit/objects/test_pci_device https://review.openstack.org/307615 | 08:00 |
| *** ihrachys has quit IRC | 08:02 | |
| *** houming has quit IRC | 08:02 | |
| *** gcb has quit IRC | 08:02 | |
| *** zhs has quit IRC | 08:02 | |
| *** takedakn has quit IRC | 08:06 | |
| *** karthiks has quit IRC | 08:07 | |
| openstackgerrit | Li Yingjun proposed openstack/nova: Notifications on flavor operations https://review.openstack.org/398171 | 08:07 |
| markus_z | sfinucan: JFYI, I'm starting the review of the config options and I go from oldest to newest. | 08:09 |
| *** houming has joined #openstack-nova | 08:12 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: conf: Remove deprecated service manager opts https://review.openstack.org/319532 | 08:12 |
| *** gcb has joined #openstack-nova | 08:15 | |
| kaisers | @mriedem Looking at it, an issue with the attached RDO cloud. | 08:19 |
| *** ralonsoh has joined #openstack-nova | 08:20 | |
| *** karthiks has joined #openstack-nova | 08:22 | |
| *** paul-carlton2 has joined #openstack-nova | 08:23 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Remove mox in tests/unit/objects/test_pci_device https://review.openstack.org/307615 | 08:24 |
| *** tovin07 has quit IRC | 08:24 | |
| *** alexpilotti has joined #openstack-nova | 08:26 | |
| *** gcb has quit IRC | 08:27 | |
| *** markvoelker has joined #openstack-nova | 08:29 | |
| *** takedakn has joined #openstack-nova | 08:29 | |
| *** ekuris has quit IRC | 08:30 | |
| *** gcb has joined #openstack-nova | 08:30 | |
| *** alexpilotti has quit IRC | 08:30 | |
| *** alexpilotti has joined #openstack-nova | 08:30 | |
| *** markvoelker has quit IRC | 08:34 | |
| *** owalsh has quit IRC | 08:38 | |
| *** GB21 has quit IRC | 08:38 | |
| *** owalsh has joined #openstack-nova | 08:39 | |
| *** lpetrut has joined #openstack-nova | 08:39 | |
| *** ekuris has joined #openstack-nova | 08:44 | |
| openstackgerrit | Paul Carlton proposed openstack/nova: Attach/detach root volume for shelved_offloaded https://review.openstack.org/317446 | 08:46 |
| *** manasm has quit IRC | 08:48 | |
| *** GB21 has joined #openstack-nova | 08:51 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: api-ref: body verification for os-hypervisors https://review.openstack.org/398197 | 08:55 |
| *** andreas_s has quit IRC | 08:57 | |
| *** tovin07 has joined #openstack-nova | 08:57 | |
| *** paul-carlton2 has quit IRC | 08:58 | |
| *** andreas_s has joined #openstack-nova | 09:00 | |
| *** zzzeek has quit IRC | 09:00 | |
| openstackgerrit | Lee Yarwood proposed openstack/nova: encryptors: Introduce encryption provider constants https://review.openstack.org/393901 | 09:01 |
| openstackgerrit | Alex Xu proposed openstack/nova: api-ref: Complete all the verifications of remote consoles https://review.openstack.org/338051 | 09:02 |
| *** victorhe has joined #openstack-nova | 09:03 | |
| *** zzzeek has joined #openstack-nova | 09:03 | |
| *** karthiks has quit IRC | 09:06 | |
| *** victorhe has quit IRC | 09:07 | |
| openstackgerrit | Timofey Durakov proposed openstack/nova: Add handle for 2 exceptions in force_delete https://review.openstack.org/397373 | 09:12 |
| *** gouthamr has joined #openstack-nova | 09:13 | |
| *** hshiina has quit IRC | 09:14 | |
| *** gouthamr has quit IRC | 09:17 | |
| *** gouthamr has joined #openstack-nova | 09:17 | |
| *** karthiks has joined #openstack-nova | 09:19 | |
| *** takedakn has quit IRC | 09:19 | |
| *** edand has quit IRC | 09:22 | |
| *** edand has joined #openstack-nova | 09:22 | |
| *** manasm has joined #openstack-nova | 09:23 | |
| *** tovin07 has quit IRC | 09:23 | |
| *** bdeschenes has joined #openstack-nova | 09:24 | |
| *** yassine has joined #openstack-nova | 09:26 | |
| *** yassine is now known as Guest45910 | 09:26 | |
| *** sree has quit IRC | 09:27 | |
| *** sree has joined #openstack-nova | 09:27 | |
| *** huanxie has joined #openstack-nova | 09:28 | |
| *** ekuris has quit IRC | 09:28 | |
| *** sree_ has joined #openstack-nova | 09:31 | |
| *** lucas-afk is now known as lucasagomes | 09:32 | |
| *** sree has quit IRC | 09:32 | |
| *** sree_ is now known as Guest40710 | 09:32 | |
| *** manasm has quit IRC | 09:33 | |
| *** paul-carlton2 has joined #openstack-nova | 09:35 | |
| *** alexchadin has quit IRC | 09:35 | |
| *** alexchad_ has joined #openstack-nova | 09:35 | |
| *** matrohon has joined #openstack-nova | 09:36 | |
| *** Guest40710 has quit IRC | 09:36 | |
| openstackgerrit | John Hua proposed openstack/nova: WIP: fix memory usage miscalculation https://review.openstack.org/394857 | 09:37 |
| *** alexchad_ is now known as alexchadin | 09:39 | |
| *** derekh has joined #openstack-nova | 09:41 | |
| *** walharthi has quit IRC | 09:41 | |
| *** oanson has quit IRC | 09:42 | |
| *** gouthamr has quit IRC | 09:43 | |
| *** ekuris has joined #openstack-nova | 09:43 | |
| *** alexchadin has left #openstack-nova | 09:44 | |
| *** roeyc1 has joined #openstack-nova | 09:45 | |
| *** yuntongjin has quit IRC | 09:46 | |
| *** oanson has joined #openstack-nova | 09:47 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: api-ref: verification for os-server-actions(1) https://review.openstack.org/398233 | 09:49 |
| *** gouthamr has joined #openstack-nova | 09:50 | |
| openstackgerrit | Paul Carlton proposed openstack/nova: Enabling detach root volume for shelved_offloaded https://review.openstack.org/326899 | 09:51 |
| openstackgerrit | Paul Carlton proposed openstack/nova: Attach/detach root volume for shelved_offloaded https://review.openstack.org/317446 | 09:51 |
| openstackgerrit | Paul Carlton proposed openstack/nova: Adding device_present field to bdm table https://review.openstack.org/268053 | 09:51 |
| *** hfu has quit IRC | 09:51 | |
| *** yamahata has joined #openstack-nova | 09:54 | |
| openstackgerrit | Arne Recknagel proposed openstack/nova: libvirt: handle os-brick InvalidConnectorProtocol on init https://review.openstack.org/395067 | 09:56 |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: api-ref: verification for os-server-actions(1) https://review.openstack.org/398233 | 10:01 |
| *** victorhe has joined #openstack-nova | 10:02 | |
| *** zhangyufei has quit IRC | 10:02 | |
| openstackgerrit | zhaolihui proposed openstack/nova: Fix failed to boot vm with pci_alias's name including space https://review.openstack.org/392077 | 10:04 |
| *** victorhe has quit IRC | 10:06 | |
| *** yamahata has quit IRC | 10:07 | |
| openstackgerrit | Pawel Koniszewski proposed openstack/nova: Refactor console checks in live migration process https://review.openstack.org/373264 | 10:07 |
| *** zhs has joined #openstack-nova | 10:07 | |
| tdurakov | johnthetubaguy: hi, are you around? | 10:09 |
| *** ljxiash has quit IRC | 10:11 | |
| *** ljxiash has joined #openstack-nova | 10:12 | |
| *** shuquan has quit IRC | 10:14 | |
| *** ekuris has quit IRC | 10:16 | |
| *** hoangcx has quit IRC | 10:16 | |
| *** ljxiash has quit IRC | 10:17 | |
| *** phuongnh has quit IRC | 10:20 | |
| *** GB21 has quit IRC | 10:22 | |
| *** tuanluong has quit IRC | 10:22 | |
| *** sahid has joined #openstack-nova | 10:23 | |
| *** cdent has joined #openstack-nova | 10:27 | |
| *** sambetts|afk is now known as sambetts | 10:28 | |
| *** liusheng has quit IRC | 10:29 | |
| *** mvk has quit IRC | 10:30 | |
| *** liusheng has joined #openstack-nova | 10:30 | |
| *** markvoelker has joined #openstack-nova | 10:30 | |
| sahid | hello johnthetubaguy, when you have a moment can you re+2 https://review.openstack.org/#/c/284094/, I have addressed couple of minors typos | 10:31 |
| sahid | I'd expect dan smith to ack it by this afternoon | 10:31 |
| *** ekuris has joined #openstack-nova | 10:31 | |
| *** victorhe has joined #openstack-nova | 10:31 | |
| *** sree has joined #openstack-nova | 10:32 | |
| *** GB21 has joined #openstack-nova | 10:34 | |
| *** markvoelker has quit IRC | 10:35 | |
| *** victorhe has quit IRC | 10:36 | |
| *** bdeschenes has quit IRC | 10:37 | |
| *** sree has quit IRC | 10:37 | |
| *** walharthi has joined #openstack-nova | 10:38 | |
| *** gouthamr has quit IRC | 10:38 | |
| *** gouthamr has joined #openstack-nova | 10:39 | |
| cdent | thanks for the comments on the max_unit tests johnthetubaguy, will endeavor to integrate that into my evolving "tests-like-what-nova-likes" guidelines | 10:40 |
| *** brault has quit IRC | 10:41 | |
| *** BobBall_AWOL is now known as BobBall | 10:41 | |
| *** BobBall is now known as Guest34042 | 10:41 | |
| *** victorhe has joined #openstack-nova | 10:42 | |
| johnthetubaguy | cdent: mind you, some evolution of "what-nova-likes" is no bad thing | 10:42 |
| *** walharthi has quit IRC | 10:43 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Fix backwards compatibility for InstanceNUMACell https://review.openstack.org/396184 | 10:43 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Remove 'super' call in obj_make_compatible https://review.openstack.org/398264 | 10:43 |
| openstackgerrit | Moshe Levi proposed openstack/nova: [WIP] OVS acceleration https://review.openstack.org/398265 | 10:44 |
| johnthetubaguy | sahid: I am curious around dan's take on the quota issues, not sure thats been resolved yet | 10:44 |
| cdent | johnthetubaguy: stress on the evolving! :) | 10:44 |
| cdent | continuously | 10:44 |
| *** brault has joined #openstack-nova | 10:45 | |
| *** victorhe has quit IRC | 10:46 | |
| *** owalsh has quit IRC | 10:46 | |
| johnthetubaguy | cdent: +100 | 10:48 |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/nova: DNM PoC Enable VNC console access to Ironic nodes https://review.openstack.org/398270 | 10:50 |
| *** omnipresent has joined #openstack-nova | 10:52 | |
| *** zhangyufei has joined #openstack-nova | 10:53 | |
| *** ihrachys has joined #openstack-nova | 10:54 | |
| *** mvk has joined #openstack-nova | 10:54 | |
| *** omnipresent has quit IRC | 10:57 | |
| openstackgerrit | Moshe Levi proposed openstack/nova: [WIP] OVS acceleration https://review.openstack.org/398265 | 11:00 |
| openstackgerrit | pangliye proposed openstack/nova: Typo error allocations.yaml https://review.openstack.org/398275 | 11:00 |
| openstackgerrit | Moshe Levi proposed openstack/os-vif: [WIP] add OVS acceleration https://review.openstack.org/398277 | 11:02 |
| *** zenoway has joined #openstack-nova | 11:02 | |
| openstackgerrit | pangliye proposed openstack/nova: Typo error servers.py https://review.openstack.org/398280 | 11:02 |
| sahid | johnthetubaguy: well the question is not clear for me | 11:02 |
| johnthetubaguy | sahid: the question is how to we handle the user's quota for vCPUs (note that happens in the API, so does't have access to the virt driver overhead method) | 11:03 |
| sahid | we certainly charge for the pCPU consumed that is the point of extending the method 'estimate_instance_overhead' | 11:03 |
| *** bks has quit IRC | 11:04 | |
| johnthetubaguy | there is resource usage on the compute node, and user's quota, which are separate | 11:04 |
| johnthetubaguy | agreed on the resource usage plan | 11:04 |
| *** houming has left #openstack-nova | 11:04 | |
| sahid | i see... | 11:04 |
| *** ekuris has quit IRC | 11:05 | |
| sahid | johnthetubaguy: how do you handle that point in xenapi fro the memroy overhead ? | 11:07 |
| *** gouthamr has quit IRC | 11:07 | |
| johnthetubaguy | sahid: it doesn't consume extra quota, as the overhead scales with the initial quota, so it doesn't really affect capacity planning | 11:08 |
| openstackgerrit | int32bit proposed openstack/nova: List support for force-completing a live migration in Feature support matrix https://review.openstack.org/397871 | 11:09 |
| sahid | johnthetubaguy: i guess for emulator_threads is different, we are in context od cpu_policy dedicated so a vCPU consumed is bassically a pCPU consumed | 11:09 |
| sahid | we should consume 1 vCPU on the user quota, let me add a note about that specific point | 11:10 |
| johnthetubaguy | sahid: I am not sure how that will work, code wise, but thats probably the correct thing to do | 11:10 |
| johnthetubaguy | sahid: honestly its tempting to just not affect the user quota | 11:11 |
| johnthetubaguy | as its confusing that metadata on the flavor affects the quota usage | 11:11 |
| *** victorhe has joined #openstack-nova | 11:11 | |
| johnthetubaguy | (we need to rethink quota, generally, but thats a different discussion) | 11:11 |
| *** Guest34042 is now known as bobba | 11:12 | |
| sahid | hum.. if we don't do that an user with a quota of 10vCPUs is able to consume 20pCPUs | 11:15 |
| *** gszasz has joined #openstack-nova | 11:15 | |
| *** victorhe has quit IRC | 11:16 | |
| *** coreywright has quit IRC | 11:16 | |
| *** GB21 has quit IRC | 11:18 | |
| *** edand has quit IRC | 11:20 | |
| *** adisky has joined #openstack-nova | 11:20 | |
| *** rfolco has joined #openstack-nova | 11:22 | |
| *** ekuris has joined #openstack-nova | 11:23 | |
| *** jaosorior is now known as jaosorior_lunch | 11:25 | |
| johnthetubaguy | there are lots of reasons quota are not that strict, think ssd vs non-ssd disks, ironic vs virtual, GPU quota, etc Its only really ever giving you a fuzzy upper bound on what folks can use | 11:26 |
| *** gouthamr has joined #openstack-nova | 11:26 | |
| johnthetubaguy | sahid: think about the user that can't launch as their quota says they have 1 vCPU left, and the flavor API says it uses 1vCPU | 11:27 |
| johnthetubaguy | sahid: honestly I could be argued into supporting either direction, but I am leaning towards the keep it simple direction | 11:27 |
| sahid | johnthetubaguy: yes that are good points | 11:29 |
| *** amotoki_ has joined #openstack-nova | 11:30 | |
| *** nicolasbock has joined #openstack-nova | 11:31 | |
| *** GB21 has joined #openstack-nova | 11:31 | |
| sahid | johnthetubaguy: ok so as you said let's try to stay simple in a first attempt, it should be easy to update that point anyway | 11:32 |
| *** amotoki has quit IRC | 11:33 | |
| *** sree has joined #openstack-nova | 11:35 | |
| *** gongysh has joined #openstack-nova | 11:36 | |
| *** hfu has joined #openstack-nova | 11:44 | |
| *** gongysh has quit IRC | 11:45 | |
| openstackgerrit | sahid proposed openstack/nova-specs: Libvirt driver emulator threads placement policy https://review.openstack.org/284094 | 11:45 |
| openstackgerrit | Thomas Bechtold proposed openstack/nova: Don't refresh info_cache while a instance gets deleted https://review.openstack.org/398308 | 11:47 |
| alex_xu | cdent: using the max length of the list to validate the single value parameter is better and simpler, thanks a lot for point that out | 11:50 |
| *** gouthamr has quit IRC | 11:51 | |
| *** gouthamr has joined #openstack-nova | 11:52 | |
| cdent | alex_xu: I'm glad it led somewhere good. I ended up just confusing myself :) | 11:53 |
| alex_xu | cdent: :) | 11:54 |
| *** zenoway has quit IRC | 11:59 | |
| *** zenoway has joined #openstack-nova | 11:59 | |
| openstackgerrit | Mikhail Feoktistov proposed openstack/nova: libvirt: ephemeral disk support for virtuozzo containers https://review.openstack.org/312488 | 12:02 |
| *** owalsh has joined #openstack-nova | 12:03 | |
| *** zenoway has quit IRC | 12:04 | |
| *** ekuris has quit IRC | 12:04 | |
| *** zenoway has joined #openstack-nova | 12:07 | |
| openstackgerrit | Paul Carlton proposed openstack/nova: Attach/detach root volume for shelved_offloaded https://review.openstack.org/317446 | 12:08 |
| *** cdent has quit IRC | 12:08 | |
| *** archies has quit IRC | 12:08 | |
| *** jaosorior_lunch is now known as jaosorior | 12:09 | |
| *** lucasagomes is now known as lucas-hungry | 12:11 | |
| markus_z | mikal: I think you can abandon your change https://review.openstack.org/#/c/271354/4/nova/conf/scheduler.py , this is already done by another change. | 12:15 |
| openstackgerrit | int32bit proposed openstack/nova: Fix data error in api samples doc https://review.openstack.org/397821 | 12:17 |
| *** ekuris has joined #openstack-nova | 12:20 | |
| *** GB21 has quit IRC | 12:22 | |
| *** catintheroof has joined #openstack-nova | 12:23 | |
| openstackgerrit | Gábor Antal proposed openstack/nova: Transform instance.create.error notification https://review.openstack.org/397715 | 12:24 |
| *** hferenc has joined #openstack-nova | 12:24 | |
| *** victorhe has joined #openstack-nova | 12:30 | |
| *** novafire99 has joined #openstack-nova | 12:31 | |
| *** gouthamr has quit IRC | 12:32 | |
| *** markus_z has quit IRC | 12:33 | |
| openstackgerrit | Mikhail Feoktistov proposed openstack/nova: raise exception if create virtuozzo container with swap disk https://review.openstack.org/397829 | 12:33 |
| *** rtheis has joined #openstack-nova | 12:34 | |
| *** victorhe has quit IRC | 12:34 | |
| *** ekuris has quit IRC | 12:34 | |
| *** talanar has quit IRC | 12:35 | |
| openstackgerrit | Alex Xu proposed openstack/nova-specs: Consistent query parameters validation https://review.openstack.org/388518 | 12:36 |
| openstackgerrit | Alex Xu proposed openstack/nova-specs: Consistent query parameters validation https://review.openstack.org/388518 | 12:38 |
| *** gcb has quit IRC | 12:38 | |
| *** walharthi has joined #openstack-nova | 12:39 | |
| *** owalsh has quit IRC | 12:40 | |
| *** owalsh has joined #openstack-nova | 12:40 | |
| *** gcb has joined #openstack-nova | 12:42 | |
| *** gmann_ has joined #openstack-nova | 12:44 | |
| alex_xu | nova api meeting is in 15 mins at #openstack-meeting-4 | 12:45 |
| *** owalsh has quit IRC | 12:46 | |
| *** cdent has joined #openstack-nova | 12:49 | |
| *** moshele has quit IRC | 12:49 | |
| *** ekuris has joined #openstack-nova | 12:50 | |
| openstackgerrit | Mikhail Feoktistov proposed openstack/nova: fix error message https://review.openstack.org/398334 | 12:52 |
| *** thorst_ has joined #openstack-nova | 12:53 | |
| *** thorst_ has quit IRC | 12:54 | |
| *** thorst_ has joined #openstack-nova | 12:54 | |
| *** ratailor has quit IRC | 12:54 | |
| *** pradk has joined #openstack-nova | 12:55 | |
| *** sree has quit IRC | 12:57 | |
| *** sree has joined #openstack-nova | 12:57 | |
| openstackgerrit | int32bit proposed openstack/nova: Fix data error in api samples doc https://review.openstack.org/397821 | 12:58 |
| alex_xu | nova api meeting is running at #openstack-meeting-4 | 13:00 |
| *** sree has quit IRC | 13:01 | |
| cdent | edleafe, jaypipes: Im available to chat whenever, if we still want to do that, but will need 30 minutes warning to get somewhere talkable | 13:02 |
| *** moshele has joined #openstack-nova | 13:03 | |
| *** snikitin has quit IRC | 13:04 | |
| *** tblakes has joined #openstack-nova | 13:06 | |
| *** moshele has quit IRC | 13:06 | |
| *** ducttape_ has joined #openstack-nova | 13:08 | |
| *** moshele has joined #openstack-nova | 13:09 | |
| *** kylek3h has quit IRC | 13:09 | |
| *** kylek3h has joined #openstack-nova | 13:09 | |
| *** ducttape_ has quit IRC | 13:09 | |
| *** ducttape_ has joined #openstack-nova | 13:10 | |
| openstackgerrit | Lee Yarwood proposed openstack/nova: compute: Do not create bdms before successfully reserving a volume https://review.openstack.org/290793 | 13:10 |
| openstackgerrit | Lee Yarwood proposed openstack/nova: Make BlockDeviceMapping object support uuid https://review.openstack.org/242603 | 13:10 |
| openstackgerrit | Lee Yarwood proposed openstack/nova: Add uuid column to BlockDeviceMapping https://review.openstack.org/242602 | 13:10 |
| *** prateek has quit IRC | 13:10 | |
| *** hfu has quit IRC | 13:13 | |
| *** kylek3h has quit IRC | 13:13 | |
| openstackgerrit | Alex Szarka proposed openstack/nova: Reduce code complexity - servers.py https://review.openstack.org/359861 | 13:14 |
| openstackgerrit | Alex Szarka proposed openstack/nova: Reduce code complexity - manager.py https://review.openstack.org/359868 | 13:15 |
| openstackgerrit | Alex Szarka proposed openstack/nova: Reduce code complexity - linux_net.py https://review.openstack.org/359871 | 13:15 |
| *** brault has quit IRC | 13:17 | |
| *** brault has joined #openstack-nova | 13:18 | |
| *** lucas-hungry is now known as lucasagomes | 13:20 | |
| *** bdeschenes has joined #openstack-nova | 13:20 | |
| *** lamt has joined #openstack-nova | 13:23 | |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Extend get_all_by_filters to support resource criteria https://review.openstack.org/386242 | 13:24 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: WIP: Expose a REST API method for a specific list of RPs https://review.openstack.org/392569 | 13:24 |
| *** jheroux has joined #openstack-nova | 13:24 | |
| *** jamielennox is now known as jamielennox|away | 13:25 | |
| *** dave-mccowan has joined #openstack-nova | 13:26 | |
| *** bdeschenes has quit IRC | 13:27 | |
| *** GB21 has joined #openstack-nova | 13:27 | |
| *** talanar has joined #openstack-nova | 13:28 | |
| *** bkopilov_ is now known as bkopilov | 13:28 | |
| schegi | Hi got a little question related to migration. configuring migration to not use ssh instead qemu+tcp, works fine for live migration (e.g nova live-migration server host) but when i do a 'cold' migration (just nova migrate server) nova still tries to migrate using ssh and fails. All storage is ceph or iscsi so shared. | 13:28 |
| schegi | This issue also hits me during resizeing. | 13:29 |
| cdent | bauzas: did you know your push to https://review.openstack.org/#/c/386242/ just clobbered edleafe's changes? | 13:30 |
| bauzas | cdent: nope, I left them | 13:30 |
| *** brault has quit IRC | 13:30 | |
| bauzas | cdent: check the diff between PS7 and PS8 | 13:31 |
| cdent | I am checking the diff betwene ps7 and ps8 | 13:31 |
| *** siva_krish has joined #openstack-nova | 13:31 | |
| cdent | I guess you moved the comment back to the start? | 13:31 |
| *** novafire99 has quit IRC | 13:31 | |
| bauzas | cdent: nope, placed the comment at the right place rather | 13:32 |
| bauzas | cdent: and modified it because it was wrong | 13:32 |
| *** xyang1 has joined #openstack-nova | 13:32 | |
| cdent | okay | 13:32 |
| bauzas | just removed the Co-Authored-By tag which looks to me like a nit | 13:33 |
| bauzas | given I worked on the SQL stuff since 2 weeks, and edleafe just modified a comment... | 13:33 |
| bauzas | which is why I was really angry yesterday | 13:33 |
| cdent | bauzas: edleafe didn't just add a comment | 13:35 |
| cdent | he changed the sql | 13:35 |
| bauzas | cdent: sure, but he just added jaypipes's comment, that's it | 13:35 |
| *** brault has joined #openstack-nova | 13:35 | |
| *** takedakn has joined #openstack-nova | 13:35 | |
| bauzas | so, for example, I'm building a house | 13:35 |
| bauzas | since 12 months | 13:35 |
| *** amotoki_ has quit IRC | 13:36 | |
| bauzas | and someone just painting the walls would be co-authoring my house ? | 13:36 |
| openstackgerrit | Leehom Li proposed openstack/nova: Support detach interface with same MAC from instance https://review.openstack.org/372243 | 13:36 |
| cdent | bauzas: we work together on this stuff, don't we, as a team, and you've been busy with other important things and edleafe had some time, so he tried to help so that we can move things, made some useful changes, so co-authored seems appropriate. Why does authoriship/ownership matter so much? The tag is there to help find people who can repsond to question on the code. | 13:37 |
| *** ducttape_ has quit IRC | 13:38 | |
| bauzas | cdent: I won't go this way and discuss about that | 13:38 |
| bauzas | I just provided my humble opinion | 13:38 |
| cdent | wow, okay. | 13:38 |
| *** ducttape_ has joined #openstack-nova | 13:38 | |
| *** zhangyufei has quit IRC | 13:39 | |
| bauzas | cdent: if I would co-author, I would put jay given his comments | 13:40 |
| *** gmann__ has joined #openstack-nova | 13:40 | |
| bauzas | anyway | 13:41 |
| *** kylek3h has joined #openstack-nova | 13:42 | |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: Adds view builders for keypairs controller https://review.openstack.org/347289 | 13:42 |
| *** rodrigods has quit IRC | 13:42 | |
| *** esberglu has quit IRC | 13:42 | |
| *** rodrigods has joined #openstack-nova | 13:42 | |
| *** gmann_ has quit IRC | 13:42 | |
| *** ducttape_ has quit IRC | 13:43 | |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: [WIP] scheduler: strictly isolate group of hosts for image/flavor https://review.openstack.org/370458 | 13:43 |
| *** edmondsw has joined #openstack-nova | 13:43 | |
| *** markvoelker has joined #openstack-nova | 13:44 | |
| *** esberglu has joined #openstack-nova | 13:45 | |
| *** shuquan has joined #openstack-nova | 13:47 | |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: [WIP] scheduler: strictly isolate group of hosts for image/flavor https://review.openstack.org/370458 | 13:49 |
| openstackgerrit | Pawel Koniszewski proposed openstack/nova: live-mig: Add claims and proper resource tracking https://review.openstack.org/244489 | 13:50 |
| *** gouthamr has joined #openstack-nova | 13:50 | |
| *** ekuris has quit IRC | 13:50 | |
| *** victorhe has joined #openstack-nova | 13:51 | |
| *** liuyulong_ has joined #openstack-nova | 13:52 | |
| *** rtheis has quit IRC | 13:52 | |
| *** esberglu has quit IRC | 13:53 | |
| *** sree has joined #openstack-nova | 13:53 | |
| *** tlian has joined #openstack-nova | 13:54 | |
| *** gouthamr has quit IRC | 13:54 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Fix backwards compatibility for InstanceNUMACell https://review.openstack.org/396184 | 13:54 |
| *** takedakn has quit IRC | 13:55 | |
| *** liuyulong__ has quit IRC | 13:55 | |
| edleafe | bauzas: Adding "co-authored-by" isn't stealing credit for someone else's work. It simply a way to show that someone other than the patch owner changed the patch in a significant way. | 13:55 |
| *** rtheis has joined #openstack-nova | 13:55 | |
| *** eliqiao has quit IRC | 13:56 | |
| edleafe | bauzas: I don't understand why you feel so offended by that | 13:56 |
| *** victorhe has quit IRC | 13:56 | |
| bauzas | edleafe: for two reasons : | 13:56 |
| bauzas | 1/ I think the modification wasn't significant, the only real change was provided by someone different than you | 13:56 |
| *** eliqiao has joined #openstack-nova | 13:57 | |
| bauzas | 2/ I usually provide new patchsets for others, but I first ask them if they agree, and then, only unless I'm really working on something needing my mind, I don't add a tag | 13:57 |
| bauzas | call it a gentleman's agreement | 13:58 |
| bauzas | but I'm done discussing with that | 13:58 |
| edleafe | bauzas: All I want to make sure is that you understand that there was no offense intended. We can disagree on the technical merits. | 13:59 |
| bauzas | edleafe: okay, I appreciate that | 13:59 |
| *** amotoki has joined #openstack-nova | 13:59 | |
| Kevin_Zheng | it is pretty common to filter by userid | 14:00 |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: [WIP] scheduler: strictly isolate group of hosts for image/flavor https://review.openstack.org/370458 | 14:02 |
| *** alextricity25 has left #openstack-nova | 14:02 | |
| Kevin_Zheng | johnthetubaguy: it is pretty common to filter by user_id | 14:03 |
| johnthetubaguy | Kevin_Zheng: project_id is more common I guess? | 14:03 |
| bauzas | edleafe: cdent: about the SQL-ish, you can ask for printing the query using pdb | 14:03 |
| johnthetubaguy | Kevin_Zheng: for most it seems like this is the key one: Index('instances_project_id_deleted_idx', 'project_id', 'deleted') | 14:03 |
| edleafe | bauzas: yes, I do that, but I wish there was a simple way of seeing the binding parameters | 14:04 |
| bauzas | edleafe: cdent: it's possible to just str(query) but I tend to use the MySQL dialect | 14:04 |
| Kevin_Zheng | johnthetubaguy: hmm | 14:04 |
| *** liuyulong_ has quit IRC | 14:05 | |
| bauzas | edleafe: cdent: print str(q.statement.compile(dialect=sqlalchemy.dialects.mysql.dialect())) | 14:05 |
| *** liuyulong_ has joined #openstack-nova | 14:05 | |
| cdent | bauzas: I'm not sure what you're trying to explain? The need (or lack thereof) of the comment, or just the general confusion over the query? If it's the latter the issue isn't that the sqlA is confusing, it's that the generated SQL is | 14:05 |
| bauzas | cdent: I try to help you correctly understanding the SQL query | 14:06 |
| Kevin_Zheng | johnthetubaguy: so do exactly as what currently have indexes？ | 14:06 |
| johnthetubaguy | Kevin_Zheng: I am not sure | 14:06 |
| bauzas | given SQLA can be difficult to review, you can just use pdb and run the print statement above | 14:06 |
| bauzas | then, either you play with MySQL Workbench like I did | 14:06 |
| johnthetubaguy | Kevin_Zheng: I think we should decide what is the bare minimum, then check we have the correct indexes in place | 14:06 |
| *** ekuris has joined #openstack-nova | 14:07 | |
| bauzas | or you pimp your classic favorite sql client to use it | 14:07 |
| Kevin_Zheng | yes, actually | 14:07 |
| Kevin_Zheng | I | 14:07 |
| *** mriedem has joined #openstack-nova | 14:07 | |
| edleafe | bauzas: that doesn't show the values of the parameters, unfortunately | 14:07 |
| Kevin_Zheng | I'm not sure what should be included | 14:07 |
| bauzas | edleafe: right, but you get them from the dict | 14:07 |
| edleafe | bauzas: which dict? | 14:08 |
| Kevin_Zheng | johnthetubaguy: could you help on that？ | 14:08 |
| edleafe | IOW, I want to make sure that the passed values to the method are the ones that sqla is getting | 14:08 |
| bauzas | edleafe: I suppose you wanna understand how to pass the amounts and the resource classes ? | 14:08 |
| johnthetubaguy | Kevin_Zheng: right now, I can't really, I have been trying to find a DB expert to take a look at this for us | 14:08 |
| *** gmann__ has quit IRC | 14:09 | |
| Kevin_Zheng | johnthetubaguy: hmm sure ok | 14:09 |
| bauzas | edleafe: that's why I'm iterating on the "resources" dict | 14:09 |
| bauzas | keyed by the RC id | 14:09 |
| edleafe | bauzas: no, I want to verify that the values passed in the filters are winding up in the right places in the sqla query | 14:09 |
| bauzas | edleafe: then, pdb it | 14:09 |
| edleafe | I can easily see the dict | 14:09 |
| openstackgerrit | Lucian Petrut proposed openstack/nova: HyperV: use os-brick for volume related operations https://review.openstack.org/273504 | 14:10 |
| bauzas | edleafe: I also wrote the functional tests for the exact reason to have a TDD | 14:10 |
| edleafe | bauzas: that's great, but I'm not just concerned with this particular query | 14:10 |
| bauzas | edleafe: I'm not sure I'm getting your point | 14:10 |
| edleafe | bauzas: my difficulty with sqla is that it so effectively hides the query parameters | 14:11 |
| *** rtheis_ has joined #openstack-nova | 14:11 | |
| edleafe | You can see the SQL, but you can't see the values that are being passed along to the engine | 14:11 |
| *** gouthamr has joined #openstack-nova | 14:11 | |
| *** victorhe has joined #openstack-nova | 14:12 | |
| *** esberglu has joined #openstack-nova | 14:12 | |
| edleafe | It's hard to debug one-off errors | 14:12 |
| *** rtheis has quit IRC | 14:12 | |
| *** baoli has joined #openstack-nova | 14:13 | |
| bauzas | edleafe: IIRC, there is some way to do that | 14:13 |
| bauzas | edleafe: using SQLA | 14:13 |
| bauzas | edleafe: but you can also modify your mysql server to add a specific log :) | 14:14 |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: [WIP] scheduler: strictly isolate group of hosts for image/flavor https://review.openstack.org/370458 | 14:14 |
| bauzas | I mean the query log | 14:14 |
| edleafe | bauzas: I've only seen complex programs that jump through all sorts of hoops | 14:14 |
| *** novafire has joined #openstack-nova | 14:14 | |
| *** tbachman has joined #openstack-nova | 14:14 | |
| *** baoli has quit IRC | 14:14 | |
| *** timello has quit IRC | 14:14 | |
| bauzas | edleafe: yeah, I remember this now http://stackoverflow.com/questions/5631078/sqlalchemy-print-the-actual-query | 14:15 |
| *** baoli has joined #openstack-nova | 14:15 | |
| bauzas | edleafe: I usually don't need to get the bind parameters because I can read the SQLA statement, but in case you wanna debug, ths ^ can help | 14:16 |
| *** talanar has quit IRC | 14:16 | |
| bauzas | edleafe: I mean the literal_binds kwarg | 14:16 |
| *** timello has joined #openstack-nova | 14:17 | |
| edleafe | bauzas: yeah, that's the one I remember seeing | 14:17 |
| *** prateek has joined #openstack-nova | 14:17 | |
| edleafe | I just wish there was a way to: print query.binding_param_dict or something like that | 14:18 |
| *** baoli_ has joined #openstack-nova | 14:18 | |
| *** ducttape_ has joined #openstack-nova | 14:20 | |
| *** baoli has quit IRC | 14:21 | |
| *** mdrabe has joined #openstack-nova | 14:22 | |
| *** lamt has quit IRC | 14:23 | |
| *** victorhe_ has joined #openstack-nova | 14:25 | |
| *** READ10 has joined #openstack-nova | 14:26 | |
| *** bvanhav has joined #openstack-nova | 14:28 | |
| *** victorhe has quit IRC | 14:28 | |
| edleafe | jaypipes: are you around? | 14:28 |
| mriedem | bauzas: there is a config option in oslo.db to trace the sql calls | 14:29 |
| bauzas | mriedem: <3 | 14:29 |
| bauzas | mriedem: /me looking, wasn't aware of it | 14:29 |
| mriedem | one of these | 14:29 |
| mriedem | # Verbosity of SQL debugging information: 0=None, 100=Everything. (integer # value) #connection_debug = 0 # Add Python stack traces to SQL as comment strings. (boolean value) #connection_trace = false | 14:29 |
| *** victorhe_ has quit IRC | 14:29 | |
| bauzas | voila | 14:29 |
| bauzas | edleafe: ^ | 14:30 |
| *** dave-mccowan has quit IRC | 14:30 | |
| *** timello has quit IRC | 14:30 | |
| bauzas | mriedem: the problem we were discussing is that only jaypipes and me were understanding the patch I wrote :p | 14:30 |
| edleafe | but we're using sqla directly, not through oslo.db, right? | 14:31 |
| *** timello has joined #openstack-nova | 14:31 | |
| *** wanghao_ has quit IRC | 14:31 | |
| *** amotoki_ has joined #openstack-nova | 14:31 | |
| cdent | edleafe: the connection is still administered via olso.db | 14:31 |
| *** ducttape_ has quit IRC | 14:31 | |
| cdent | the oslo stuff just hooks into the engine within sql | 14:31 |
| cdent | a | 14:31 |
| edleafe | cdent: ah, thanks | 14:31 |
| cdent | (I think) | 14:31 |
| bauzas | edleafe: cdent: correct | 14:32 |
| *** walharthi has quit IRC | 14:32 | |
| *** ducttape_ has joined #openstack-nova | 14:32 | |
| *** markus_z has joined #openstack-nova | 14:32 | |
| bauzas | we use the oslo.db connection managers | 14:32 |
| bauzas | edleafe: don't confound using SQLA direct types (instead of SQLA ORM) and the use of oslo.db for calling the database | 14:33 |
| mriedem | rigtht, we're not doing straight sql command execution | 14:34 |
| mriedem | we're using sqla, | 14:34 |
| mriedem | just not the ORM | 14:34 |
| *** amotoki has quit IRC | 14:34 | |
| mriedem | i agree that if you're not jaypipes it's not easy to parse | 14:35 |
| *** jschwarz has joined #openstack-nova | 14:35 | |
| mriedem | i think jay dreams in sql | 14:35 |
| mriedem | pugs writing db queries | 14:35 |
| jschwarz | Hey guys - trying to boot a VM with an image I downloaded fails with "Block Device Mapping is Invalid" - any ideas? | 14:35 |
| jschwarz | This is the image I used: https://cloud-images.ubuntu.com/trusty/current/trusty-server-cloudimg-amd64-disk1.img | 14:36 |
| mriedem | your bdm is invalid :) | 14:36 |
| mriedem | check the request | 14:36 |
| mriedem | likely the block device mapping parameters to the request are wrong somehow | 14:36 |
| *** ducttape_ has quit IRC | 14:36 | |
| mriedem | look for errors in the n-api logs | 14:36 |
| *** ekuris has quit IRC | 14:37 | |
| jschwarz | no error in n-api, the parameters I see are here: http://pastebin.com/RaNitUa1 | 14:38 |
| *** gouthamr has quit IRC | 14:39 | |
| *** ar1el has joined #openstack-nova | 14:40 | |
| *** cfriesen has joined #openstack-nova | 14:41 | |
| *** amotoki_ has quit IRC | 14:41 | |
| *** jaosorior has quit IRC | 14:41 | |
| sdatko | mriedem: hello :) | 14:41 |
| *** lamt has joined #openstack-nova | 14:41 | |
| *** jaosorior has joined #openstack-nova | 14:41 | |
| mriedem | jschwarz: i guess look through https://github.com/openstack/nova/blob/master/nova/compute/api.py#L1287 - does the image with that id exist? | 14:42 |
| *** shuquan has quit IRC | 14:42 | |
| mriedem | sdatko: hi | 14:42 |
| *** tongli has joined #openstack-nova | 14:44 | |
| *** shuquan has joined #openstack-nova | 14:44 | |
| jschwarz | mriedem, it does: http://pastebin.com/psUMB55Z | 14:45 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: Acquire TCP ports for console during live migration https://review.openstack.org/275801 | 14:45 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: do not return serial address if disabled on destination https://review.openstack.org/397276 | 14:45 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: clean serial listen addr to always deal with string https://review.openstack.org/398389 | 14:45 |
| *** mgoddard_ has joined #openstack-nova | 14:46 | |
| *** rama_y has joined #openstack-nova | 14:46 | |
| *** jaosorior has quit IRC | 14:47 | |
| *** cleong has joined #openstack-nova | 14:47 | |
| mriedem | jschwarz: which release? master? newton? | 14:47 |
| mriedem | do you have debug logging enabled? | 14:47 |
| jschwarz | mriedem, it's a master from about 2 months ago | 14:48 |
| jschwarz | mriedem, I can re-stack with an updated master though | 14:48 |
| *** victorhe has joined #openstack-nova | 14:48 | |
| mriedem | i'd enable debug logging first and retry and see if there is anything logged | 14:48 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: api-ref: body verification for abort live migration https://review.openstack.org/397407 | 14:48 |
| jschwarz | mriedem, debug logging is on, n-api doesn't show anything suspicious to me | 14:48 |
| jschwarz | granted, I'm not a nova guy though :) | 14:49 |
| *** mgoddard has quit IRC | 14:49 | |
| dansmith | mriedem: easly is a word | 14:49 |
| mriedem | as is mistakently | 14:50 |
| dansmith | clearly you are mistakent | 14:50 |
| openstackgerrit | Paul Carlton proposed openstack/nova: Attach/detach root volume for shelved_offloaded https://review.openstack.org/317446 | 14:50 |
| *** victorhe has quit IRC | 14:50 | |
| artom | jschwarz, you're getting a 400 bad request response? What's your command look like? | 14:51 |
| *** psachin has quit IRC | 14:51 | |
| *** victorhe has joined #openstack-nova | 14:51 | |
| sdatko | mriedem: sorry for responding so late, but I had a problem with my computer | 14:51 |
| sdatko | self.flags is not possible in FakeLibvirtFixture, as it not inherits from Tests | 14:52 |
| mriedem | jschwarz: the other thing i don't get is the rest api will return the more specific error message, so you should get more details in the response than just 'Block Device Mapping is Invalid' | 14:52 |
| *** swamireddy has quit IRC | 14:52 | |
| sdatko | however importing nova.conf and performing CONF.set_override() works fine | 14:52 |
| mriedem | sdatko: ok you might just need to do that then, | 14:53 |
| sdatko | mriedem: is that okay for you? | 14:53 |
| mriedem | the only thing i worry about with that is concurrency issues with multiple tests running at the same time that are using that flag | 14:53 |
| mriedem | but i think the fixture is applied on setup for each test case so it shouldn't be an issue | 14:53 |
| jschwarz | artom, using Horizon to boot it, actually | 14:54 |
| jschwarz | artom, so I'm guessing something standard | 14:54 |
| *** ducttape_ has joined #openstack-nova | 14:54 | |
| sdatko | not sure about each tests | 14:54 |
| jschwarz | mriedem, let me run a 'nova boot' | 14:54 |
| sdatko | but they are applied for eeach class | 14:54 |
| sdatko | ok, i will push it to review then :) | 14:54 |
| *** GB21 has quit IRC | 14:54 | |
| *** nherciu has quit IRC | 14:54 | |
| sdatko | mriedem: thank you for help! | 14:54 |
| *** sree has quit IRC | 14:55 | |
| artom | jschwarz, is your image OK? Have you checked the checksum? | 14:55 |
| *** andymaier has joined #openstack-nova | 14:55 | |
| jschwarz | artom, I think it is, I'll make sure | 14:55 |
| *** sree has joined #openstack-nova | 14:55 | |
| macsz | mriedem dansmith : this can be restored now https://review.openstack.org/#/c/392236/ | 14:57 |
| *** andymaier has quit IRC | 14:57 | |
| jschwarz | mriedem, artom, 'git pull' seems to have fixed it | 14:58 |
| jschwarz | weird | 14:58 |
| *** rmart04 has joined #openstack-nova | 14:58 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Make n-net refuse to start unless using CellsV1 https://review.openstack.org/392236 | 14:58 |
| *** hongbin has joined #openstack-nova | 14:58 | |
| *** mgoddard has joined #openstack-nova | 14:58 | |
| *** mgoddard_ has quit IRC | 14:59 | |
| mriedem | macsz: we'll see | 14:59 |
| *** sree has quit IRC | 14:59 | |
| mriedem | dansmith: are you going to add any compute api unit test coverage for https://review.openstack.org/#/c/394656/ or you just want to tell me to bug off? | 15:01 |
| mriedem | we could also do it later on top, i think i know a place to add that into an existing test | 15:02 |
| dansmith | mriedem: I'm working on it now, although there don't seem to be any obvious places that aren't super contrived | 15:02 |
| dansmith | so just writing a new one | 15:02 |
| mriedem | ok | 15:02 |
| dansmith | and trying to write it so it won't need to immediately change | 15:02 |
| mriedem | mdbooth is out all week right? | 15:05 |
| mriedem | PTO? | 15:05 |
| cfriesen | https://review.openstack.org/#/c/346263 is repeatedly failing nova-dvsm-full-tempest for Microsoft Hyper-V CI. I can't quite figure out what the actual problem is...could someone take a look? | 15:05 |
| mriedem | cfriesen: that wouldn't be related at all | 15:06 |
| cfriesen | mriedem: okay, cool | 15:06 |
| dansmith | mriedem: yeah | 15:07 |
| mriedem | looks like the hyper-v ci is cherry picking https://review.openstack.org/#/c/383049/ | 15:07 |
| mriedem | from tmpest | 15:07 |
| mriedem | *tempest | 15:07 |
| mriedem | claudiub: is anyone working on adding the vif plugging timeout wait code to the hyper-v driver? | 15:08 |
| mriedem | lpetrut: ^ | 15:09 |
| *** mdrabe has quit IRC | 15:11 | |
| mriedem | cfriesen: doesn't look like it's actually testing your change | 15:11 |
| mriedem | HEAD is now at 431e019 Merge commit 'refs/changes/04/273504/19' of ssh://review.openstack.org:29418/openstack/nova into HEAD | 15:11 |
| mriedem | which is https://review.openstack.org/#/c/273504 | 15:11 |
| mriedem | seems the hyper-v ci is a bit jacked up | 15:12 |
| *** cdent has quit IRC | 15:12 | |
| openstackgerrit | Thomas Bechtold proposed openstack/nova: Don't refresh info_cache while a instance gets deleted https://review.openstack.org/398308 | 15:13 |
| *** swamireddy_ has joined #openstack-nova | 15:15 | |
| *** pbandark has quit IRC | 15:17 | |
| *** hieulq_ has joined #openstack-nova | 15:17 | |
| *** kenperkins has joined #openstack-nova | 15:18 | |
| *** prateek has quit IRC | 15:18 | |
| claudiub | checking logs | 15:19 |
| *** victorhe has quit IRC | 15:23 | |
| openstackgerrit | Alex Xu proposed openstack/nova-specs: Consistent query parameters validation https://review.openstack.org/388518 | 15:23 |
| *** cdent has joined #openstack-nova | 15:24 | |
| *** baoli_ has quit IRC | 15:24 | |
| claudiub | cfriesen: seems to be failing during build, while setting up devstack. it'll be fixed shortly. your patch has no impact on the hyper-v driver, so, no need to worry about it. :) | 15:25 |
| *** mlavalle has joined #openstack-nova | 15:26 | |
| alex_xu | johnthetubaguy: I updated the spec https://review.openstack.org/#/c/388518/12, thanks in advance if you still have time to review it today :) | 15:26 |
| claudiub | mriedem: yeah, i was, but someone put him/her self as the assignee of the bug on the nova project, so, I was waiting to see a patch for it. Tomorrow is a week since the assignment, I'll reassign myself for it, and send a patch for it. | 15:26 |
| mriedem | claudiub: ok | 15:27 |
| dansmith | mriedem: I wrote you the awesomest test | 15:28 |
| dansmith | just fyi | 15:28 |
| openstackgerrit | Dan Smith proposed openstack/nova: Return build_requests instead of instances https://review.openstack.org/374267 | 15:28 |
| openstackgerrit | Dan Smith proposed openstack/nova: Refactor two nearly useless secgroup tests https://review.openstack.org/393438 | 15:28 |
| openstackgerrit | Dan Smith proposed openstack/nova: Make RPCFixture support multiple connections https://review.openstack.org/396417 | 15:28 |
| openstackgerrit | Dan Smith proposed openstack/nova: Store security groups in RequestSpec https://review.openstack.org/394656 | 15:28 |
| cfriesen | claudiub: thanks for taking a look | 15:29 |
| mriedem | crap i have a meeting to run, bbiab | 15:29 |
| jaypipes | bauzas, mriedem, edleafe, cdent: sorry, got a few morning meetings this mornig. | 15:29 |
| bauzas | jaypipes: fortunately, those morning meetings were during the morning, they could be happening during the afternoons :p | 15:30 |
| *** hieulq_ has quit IRC | 15:30 | |
| jroll | any nova-specs cores have time for a second +2 on the ironic bonding spec today? https://review.openstack.org/#/c/387534/ | 15:30 |
| *** annegentle has joined #openstack-nova | 15:32 | |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: [WIP] scheduler: strictly isolate group of hosts for image/flavor https://review.openstack.org/370458 | 15:33 |
| openstackgerrit | Alex Xu proposed openstack/nova-specs: Consistent query parameters validation https://review.openstack.org/388518 | 15:34 |
| *** ducttape_ has quit IRC | 15:35 | |
| *** jwcroppe has quit IRC | 15:37 | |
| sahid | johnthetubaguy, dansmith, jaypipes: any chance you take a second look at https://review.openstack.org/#/c/284094/ ? | 15:37 |
| sahid | so I can try to fix any comments before end of the day (in my time) | 15:38 |
| mlavalle | jaypipes: ping | 15:39 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/397916 | 15:41 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/os-vif: Updated from global requirements https://review.openstack.org/395347 | 15:43 |
| *** shuquan has quit IRC | 15:45 | |
| openstackgerrit | Maciej Szankin proposed openstack/nova: conf: remove deprecated cells driver option https://review.openstack.org/398425 | 15:45 |
| *** annegentle has quit IRC | 15:45 | |
| dansmith | sahid: I'm still not sure about the numa object change, but let johnthetubaguy +2 it and then I will hit it and we can argue about the code in the code | 15:47 |
| *** armax has joined #openstack-nova | 15:49 | |
| vladikr | ajo, dansmith, could you please take a look at the pf vlan tags spec, hope all the comments been addressed | 15:50 |
| vladikr | https://review.openstack.org/#/c/381370/5 | 15:50 |
| *** gjayavelu has joined #openstack-nova | 15:51 | |
| ajo | ack, I will review again | 15:51 |
| *** oanson has quit IRC | 15:51 | |
| *** ociuhandu has joined #openstack-nova | 15:51 | |
| cdent | alex_xu: if you're still around/awake, if you could have a look at https://review.openstack.org/#/c/355263/ , the one above it is already +W | 15:52 |
| *** snikitin_ has joined #openstack-nova | 15:53 | |
| *** jwcroppe has joined #openstack-nova | 15:55 | |
| *** lpetrut has quit IRC | 15:55 | |
| *** Sukhdev_ has joined #openstack-nova | 15:58 | |
| *** rrecio has joined #openstack-nova | 16:02 | |
| *** ducttape_ has joined #openstack-nova | 16:02 | |
| *** ar1el has quit IRC | 16:04 | |
| *** dave-mccowan has joined #openstack-nova | 16:04 | |
| moshele | mriedem: hi are you planing to go over the specs https://etherpad.openstack.org/p/nova-ocata-spec-review-sprint in tomorrow nova meeting to see which one is approved? | 16:04 |
| mriedem | moshele: no | 16:05 |
| *** ababich has joined #openstack-nova | 16:05 | |
| *** rrecio_ has joined #openstack-nova | 16:05 | |
| sahid | dansmith: ok we can do that, thanks for your reviews | 16:05 |
| *** mjura has quit IRC | 16:06 | |
| *** gjayavelu has quit IRC | 16:06 | |
| *** rrecio has quit IRC | 16:07 | |
| *** yamahata has joined #openstack-nova | 16:07 | |
| moshele | mriedem: so if they don't get merge by tomorrow they will be rejected, right?. I just want to understand the status of my User-controlled SR-IOV ports allocation (which I know you are not cray about :) ) | 16:07 |
| *** coreywright has joined #openstack-nova | 16:08 | |
| mriedem | moshele: they'll likely be frozen for ocata unless we make some exceptions | 16:08 |
| *** whenry has joined #openstack-nova | 16:08 | |
| openstackgerrit | Maciej Szankin proposed openstack/nova: conf: remove deprecated exception option https://review.openstack.org/397823 | 16:09 |
| *** gouthamr has joined #openstack-nova | 16:09 | |
| *** clenimar has joined #openstack-nova | 16:11 | |
| *** annegentle has joined #openstack-nova | 16:11 | |
| wznoinsk | moshele, mriedem we (our division in Intel) are also interested in getting User-controlled SR-IOV ports (surpise surprise :-)) and would like to help | 16:13 |
| wznoinsk | mriedem, what is there that could make it more possible to land in Ocata? more code-work, testing? | 16:14 |
| mriedem | who is going to do 3rd party CI testing on this feature? | 16:15 |
| mriedem | mellanox, intel? | 16:15 |
| moshele | mriedem: we will do, I wrote it in the spec, but it requires some tempest changes | 16:16 |
| *** ar1el has joined #openstack-nova | 16:16 | |
| wznoinsk | we (Intel) do 3rd party testing for all vendor specific features we develop so we can join the development, we'll run 3rdparty testing for it too | 16:16 |
| *** armax has quit IRC | 16:16 | |
| wznoinsk | s/we can/if we can | 16:17 |
| *** rtheis_ is now known as rtheis | 16:17 | |
| *** dtp has joined #openstack-nova | 16:21 | |
| *** whenry has quit IRC | 16:21 | |
| *** ducttape_ has quit IRC | 16:22 | |
| *** ducttape_ has joined #openstack-nova | 16:23 | |
| markus_z | trivial fix: https://review.openstack.org/#/c/393694/1 | 16:23 |
| *** armax has joined #openstack-nova | 16:23 | |
| mriedem | but i actually like those whitespaces! | 16:24 |
| *** erhudy has joined #openstack-nova | 16:24 | |
| markus_z | It's for the greater good | 16:24 |
| mriedem | oh well in that case | 16:24 |
| wznoinsk | mriedem, so it looks like there will be at least one company testing it, if there's any work we can help with there will be 3rdpartytesting by both companies done on that | 16:24 |
| *** moshele has quit IRC | 16:26 | |
| mriedem | wznoinsk: ok, i understand, i'm not blocking it, | 16:26 |
| *** roeyc1 has quit IRC | 16:26 | |
| mriedem | i'm just trying to wade through several dozen other specs too | 16:26 |
| macsz | i like those spaces as well :( i guess few weeks from now there will be a trivial fix that brings them back :D | 16:26 |
| mriedem | and juggle the changes and reviews and bug fixes we're making priorities for ocata already | 16:27 |
| *** tojuvone has quit IRC | 16:27 | |
| *** rama_y has quit IRC | 16:29 | |
| *** krtaylor has quit IRC | 16:29 | |
| wznoinsk | mriedem, roger that | 16:31 |
| mriedem | markus_z: sfinucan: why can't we get the oslo.db options for the api_database and placement config groups in https://review.openstack.org/#/c/393695/ rather than re-define them ourselves? | 16:31 |
| mriedem | like, we get keystoneauth options from ksa but for other groups in nova.conf, like ironic and neutron | 16:31 |
| *** venkhari has joined #openstack-nova | 16:33 | |
| markus_z | mriedem: You mean re-using them? AFAIK it would be hard/impossible to deprecate/change one of those options if we did. | 16:33 |
| mriedem | markus_z: well for example https://github.com/openstack/nova/blob/master/nova/conf/ironic.py#L105 | 16:34 |
| mriedem | that puts the ksa options in the ironic group in nova.conf | 16:34 |
| mriedem | why can't we do similar for the placement and api database groups in nova.conf, but use the options from oslo.db? | 16:34 |
| openstackgerrit | Jay Pipes proposed openstack/nova: rt: pass nodename to internal methods https://review.openstack.org/398469 | 16:34 |
| openstackgerrit | Jay Pipes proposed openstack/nova: rt: pass the nodename to public methods https://review.openstack.org/398470 | 16:34 |
| openstackgerrit | Jay Pipes proposed openstack/nova: rt: remove fluff from test_resource_tracker https://review.openstack.org/398471 | 16:34 |
| openstackgerrit | Jay Pipes proposed openstack/nova: rt: explicitly pass compute node to _update() https://review.openstack.org/398472 | 16:34 |
| openstackgerrit | Jay Pipes proposed openstack/nova: rt: use a single ResourceTracker object instance https://review.openstack.org/398473 | 16:34 |
| markus_z | mriedem: I understand what you want, let me dig a little for a proper answer | 16:35 |
| mriedem | thanks, | 16:35 |
| mriedem | i just don't want to copy the values from oslo.db in nova and then oslo.db changes and we're out in the wind | 16:35 |
| mriedem | markus_z: btw, do you miss triaging bugs like this? https://bugs.launchpad.net/nova/+bug/1642234 | 16:38 |
| openstack | Launchpad bug 1642234 in OpenStack Compute (nova) "error nova.api.openstack.extensions" [Undecided,Invalid] | 16:38 |
| markus_z | mriedem: mhm, that's how I like them, enough room for my wild spirit to to guesswork | 16:39 |
| *** rmart04 has quit IRC | 16:39 | |
| markus_z | s/to/do | 16:39 |
| openstackgerrit | Lee Yarwood proposed openstack/nova: WIP libvirt: Detach from the domain before detaching encryptors https://review.openstack.org/398479 | 16:40 |
| mriedem | sahid: you might be interested in https://bugs.launchpad.net/nova/+bug/1641517 | 16:40 |
| openstack | Launchpad bug 1641517 in OpenStack Compute (nova) "spice can't support usb redirection" [Undecided,New] - Assigned to wang jiedong (jeery) | 16:40 |
| *** paul-carlton2 has quit IRC | 16:42 | |
| *** siva_krish has quit IRC | 16:42 | |
| *** siva_krish has joined #openstack-nova | 16:43 | |
| *** krtaylor has joined #openstack-nova | 16:43 | |
| *** venkhari has quit IRC | 16:44 | |
| *** zenoway has quit IRC | 16:47 | |
| *** zenoway has joined #openstack-nova | 16:47 | |
| *** jistr is now known as jistr|biab | 16:48 | |
| *** slaweq_ has joined #openstack-nova | 16:48 | |
| openstackgerrit | Thomas Bechtold proposed openstack/nova: Handle deleted instances when refreshing the info_cache https://review.openstack.org/398308 | 16:49 |
| mriedem | omg poison, bad pandora, bad | 16:50 |
| *** mnestratov has quit IRC | 16:50 | |
| mriedem | alex_xu: gmann: oomichi: we allow a string for a port id when creating a server apparently https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/schemas/servers.py#L44 | 16:51 |
| mriedem | that should only ever be a uuid or none | 16:51 |
| sahid | mriedem: looking | 16:51 |
| mriedem | do you remember at all why we allow that to be a string? | 16:51 |
| cdent | \m/>.<\m/ | 16:51 |
| *** zenoway has quit IRC | 16:51 | |
| mriedem | no, there are no horns for poison | 16:51 |
| cdent | sarcastic horns? | 16:52 |
| mriedem | maybe ironic hipster horns | 16:52 |
| mriedem | i'm sure there is a guy with a hat, beard, suspenders, skinny jeans and poison t-shirt in a coffee shop somewhere | 16:52 |
| cdent | typing on a manual typewriter | 16:53 |
| sdatko | mriedem: hey... setting | 16:53 |
| sdatko | mriedem: hey... setting CONF.set_override('sysinfo_serial', 'none', 'libvirt') fixes the original problem | 16:54 |
| mriedem | ok | 16:54 |
| mriedem | but...? | 16:54 |
| sdatko | mriedem: but there are 8 new problems | 16:54 |
| mriedem | probably tests that require you to set that to some other value because they rely on it | 16:54 |
| mriedem | but were too lazy to set it specifically | 16:55 |
| mriedem | and relied on the default | 16:55 |
| sdatko | all the tests are from LibvirtConnTestCase | 16:55 |
| mriedem | alex_xu: oomichi: actually i'm reading that jsonschema incorrectly, | 16:55 |
| mriedem | it says it must be a string of format uuid, else none | 16:56 |
| *** gjayavelu has joined #openstack-nova | 16:56 | |
| mriedem | so i'm not sure why that didn't blow up early in this bug https://bugs.launchpad.net/nova/+bug/1642125 | 16:56 |
| openstack | Launchpad bug 1642125 in OpenStack Compute (nova) "Unexpected API Error? (unable to launch instance)" [Undecided,Invalid] | 16:56 |
| sdatko | test_xml_and_uri and all their variants | 16:56 |
| sdatko | mriedem: http://pastebin.com/mECsUhbL | 16:57 |
| *** tjones has joined #openstack-nova | 17:00 | |
| *** Zer0Byte__ has joined #openstack-nova | 17:01 | |
| mriedem | sdatko: so that means you need to explicitly set self.flags(sysinfo_serial=something, group='libvirt') in those specific test cases that are failing | 17:01 |
| mriedem | probably just set it to 'auto' for those tests as that's the default | 17:01 |
| *** gjayavelu has quit IRC | 17:02 | |
| *** swamireddy_ has quit IRC | 17:02 | |
| sdatko | shall I do that for all eight test_xml_and_uri_* functions | 17:04 |
| mriedem | yes | 17:04 |
| sdatko | or just in _check_xml_and_uri function? | 17:04 |
| mriedem | whichever is best | 17:04 |
| mriedem | if doing it in one spot fixes all, then do that | 17:04 |
| sdatko | all therese eight functions | 17:04 |
| mriedem | i'm assuming the code is ran before that check function is called though | 17:04 |
| mriedem | s/ran/run/ | 17:04 |
| sdatko | are in fact a call to this _check_xml_and_uri with different arguments only | 17:05 |
| *** Sukhdev_ has quit IRC | 17:05 | |
| mriedem | yes! https://bugs.launchpad.net/nova/+bug/1641508 | 17:07 |
| openstack | Launchpad bug 1641508 in OpenStack Compute (nova) "image_api.get() in nova ignores custom properties of images." [Undecided,Won't fix] | 17:07 |
| *** unicell1 has joined #openstack-nova | 17:07 | |
| *** unicell has quit IRC | 17:08 | |
| *** andreas_s has quit IRC | 17:10 | |
| *** rtheis_ has joined #openstack-nova | 17:14 | |
| *** ociuhandu has quit IRC | 17:14 | |
| openstackgerrit | Maciej Szankin proposed openstack/nova: conf: remove deprecated image url options https://review.openstack.org/397800 | 17:15 |
| *** rtheis has quit IRC | 17:15 | |
| *** nic has joined #openstack-nova | 17:23 | |
| *** matrohon has quit IRC | 17:23 | |
| *** rmart04 has joined #openstack-nova | 17:24 | |
| *** nic has quit IRC | 17:24 | |
| *** nic has joined #openstack-nova | 17:24 | |
| *** nic has quit IRC | 17:24 | |
| *** nic has joined #openstack-nova | 17:25 | |
| *** swamireddy has joined #openstack-nova | 17:26 | |
| *** raildo has quit IRC | 17:26 | |
| *** mfeoktistov has quit IRC | 17:26 | |
| *** Swami has joined #openstack-nova | 17:29 | |
| *** baoli has joined #openstack-nova | 17:30 | |
| *** gjayavelu has joined #openstack-nova | 17:31 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova-specs: Adds Hyper-V Shielded VMs spec https://review.openstack.org/274709 | 17:32 |
| *** raildo has joined #openstack-nova | 17:34 | |
| *** lennyb has quit IRC | 17:35 | |
| *** walharthi has joined #openstack-nova | 17:35 | |
| *** jistr|biab is now known as jistr | 17:36 | |
| *** markus_z has quit IRC | 17:36 | |
| *** walharthi has quit IRC | 17:37 | |
| *** walharthi has joined #openstack-nova | 17:37 | |
| *** esberglu has quit IRC | 17:39 | |
| *** esberglu has joined #openstack-nova | 17:39 | |
| sfinucan | sahid: RE: commit 2935436b - is 'huge' an alias for 2 MB or 1 GB huge pages? | 17:39 |
| sfinucan | sahid: Similarly, is 'small' an alias for 4 KB or 2 MB huge pages? | 17:40 |
| sfinucan | (the tests don't make it obvious) | 17:40 |
| sean-k-mooney | huge means anyting other the 4k | 17:40 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: WIP: hyper-v: Adds support for VMs with nested virtualization https://review.openstack.org/398509 | 17:40 |
| sean-k-mooney | small was an alise for 4k | 17:41 |
| *** fragatina has joined #openstack-nova | 17:41 | |
| sean-k-mooney | sorry not huge | 17:41 |
| sean-k-mooney | large is anything other then 4k | 17:41 |
| *** Apoorva has joined #openstack-nova | 17:41 | |
| sfinucan | sean-k-mooney: Excellent. Thanks, Sean :) | 17:42 |
| *** walharthi has quit IRC | 17:42 | |
| sean-k-mooney | sfinucan: im not sure huge is a valid value | 17:42 |
| sean-k-mooney | sfinucan: it was not in the original spec but maybe it was added at a later date | 17:42 |
| sfinucan | sean-k-mooney: Ah, yes - I meant 'large' | 17:42 |
| *** unicell1 has quit IRC | 17:42 | |
| sfinucan | That's what's in the code | 17:42 |
| sean-k-mooney | huge pages on non intel plathforms can be other sizes then 2m and 1G amd i think suppport 512mb hugepages or maybe that was power or arm | 17:43 |
| *** rama_y has joined #openstack-nova | 17:43 | |
| sfinucan | sean-k-mooney: So if I requested 'large' then I might get 2 kB or 1 GB, but if I request '2048' then I'll only get 1 GB, right? | 17:44 |
| *** esberglu has quit IRC | 17:44 | |
| sean-k-mooney | sfinucan: i thikn you mixed those up | 17:44 |
| sfinucan | *if I request '2048' then I'll only get 2 kB | 17:44 |
| sfinucan | :) | 17:44 |
| sean-k-mooney | if you request 2048 you will get 2MB hugepages | 17:45 |
| *** whenry has joined #openstack-nova | 17:45 | |
| sean-k-mooney | we specified the value in MB i think but maybe it is KB | 17:46 |
| sfinucan | ...and if I request 'large'? | 17:46 |
| sean-k-mooney | the spec is your best bet. large on intel systems will be 2MB or 1GB at libverts/qemus discretion | 17:46 |
| *** lennyb has joined #openstack-nova | 17:47 | |
| *** lucasagomes is now known as lucas-afk | 17:47 | |
| *** derekh has quit IRC | 17:47 | |
| sean-k-mooney | sfinucan: personally i allways just use 'large' | 17:47 |
| sean-k-mooney | sfinucan: if i want hugepages that is | 17:47 |
| *** david-lyle has quit IRC | 17:48 | |
| *** david-lyle has joined #openstack-nova | 17:48 | |
| sean-k-mooney | sfinucan: generally unless you have profiled your application an know there is a strong reason to be more precise there is no reason to be more specific | 17:48 |
| sahid | sfinucan: you can be more specific byt using MB, KB, GB | 17:51 |
| *** esberglu has joined #openstack-nova | 17:51 | |
| sfinucan | sahid: For example? | 17:52 |
| sahid | but as sean-k-mooney said, on x86 only 2MiB and 1GiB are available | 17:52 |
| *** esberglu_ has joined #openstack-nova | 17:52 | |
| *** claudiub has quit IRC | 17:52 | |
| sfinucan | sahid: So would 'hw:mem_page_size=2M' be valid? | 17:52 |
| sahid | sfinucan: if for example on your host you have a mix of 1GiB and 2MiB and you want to use 2MiB you will have to specify the size | 17:52 |
| sahid | sfinucan: yes, i can't recall exaclty the syntaxe | 17:53 |
| sahid | let me double check | 17:53 |
| sean-k-mooney | sfinucan: this might help https://specs.openstack.org/openstack/nova-specs/specs/kilo/implemented/virt-driver-large-pages.html | 17:54 |
| sahid | it's the syntax used in 'string_to_bytes' | 17:54 |
| *** adisky has quit IRC | 17:55 | |
| sahid | sfinucan: hw:mem_page_size=2MB | 17:55 |
| sahid | you can see some examples in test_hardware.py | 17:55 |
| *** esberglu has quit IRC | 17:55 | |
| nicolasbock | Hi! I am running into an issue with oslo.rootwrap I don't understand. | 17:56 |
| sfinucan | sean-k-mooney: I've learnt not to trust specs as they frequently lie :) | 17:57 |
| nicolasbock | I am running nova-compute on a xen node | 17:57 |
| sfinucan | sahid: Ahhh, gotcha. That's the one | 17:57 |
| nicolasbock | when I create a server the allocation fails eventually with an exception | 17:57 |
| bobba | nicolasbock: Xen+libvirt or XenAPI (e.g. XenServer) | 17:57 |
| nicolasbock | libvirt | 17:57 |
| sahid | sfinucan, sean-k-mooney yeah t's difficult to keep them up-to-date the code is moving quickly | 17:58 |
| nicolasbock | the backtrace goes down to /usr/lib/python2.7/site-packages/oslo_rootwrap/client.py | 17:58 |
| nicolasbock | and then | 17:58 |
| *** ralonsoh has quit IRC | 17:58 | |
| nicolasbock | Unserializable message: ('#ERROR', ValueError('I/O operation on closed file',)) | 17:58 |
| nicolasbock | the command was 'xend status' | 17:58 |
| *** mvk has quit IRC | 17:58 | |
| nicolasbock | this command should fail, I don't have xend installed | 17:58 |
| sfinucan | sahid: Aye, that it is. I'm writing (hopefully) up-to-date huge page docs now, hence the questions :) | 17:58 |
| nicolasbock | but that exception doesn't look like the command wasn't found | 17:59 |
| bobba | That might be one for #xen or anthonyper | 17:59 |
| nicolasbock | ok | 17:59 |
| nicolasbock | bobba, I'll try it there | 17:59 |
| nicolasbock | bobba, thanks! | 17:59 |
| sahid | that is nice, thanks sfinucan | 17:59 |
| *** cdelatte has joined #openstack-nova | 18:00 | |
| *** walharthi has joined #openstack-nova | 18:00 | |
| *** yamahata has quit IRC | 18:01 | |
| kashyap | sfinucan: Appreciate your love for docs. | 18:02 |
| sfinucan | kashyap: If only "tech writer" paid better ;) | 18:02 |
| kashyap | sfinucan: Developer-doc-writer is a deadly combination. | 18:02 |
| kashyap | (In a good sense) | 18:03 |
| *** fragatin_ has joined #openstack-nova | 18:04 | |
| openstackgerrit | STEW TY proposed openstack/nova: Complete verification of servers-action-fixed-ip.inc https://review.openstack.org/338606 | 18:05 |
| *** moshele has joined #openstack-nova | 18:05 | |
| *** pkoniszewski has quit IRC | 18:07 | |
| *** fragatina has quit IRC | 18:08 | |
| *** bobba is now known as BobBall_AWOL | 18:08 | |
| moshele | jaypipes: around? | 18:09 |
| *** pkoniszewski has joined #openstack-nova | 18:10 | |
| *** sahid has quit IRC | 18:11 | |
| *** diogogmt has joined #openstack-nova | 18:13 | |
| *** catinthe_ has joined #openstack-nova | 18:13 | |
| *** harlowja has joined #openstack-nova | 18:14 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Make n-net refuse to start unless using CellsV1 https://review.openstack.org/392236 | 18:15 |
| *** catintheroof has quit IRC | 18:15 | |
| *** catintheroof has joined #openstack-nova | 18:15 | |
| *** unicell has joined #openstack-nova | 18:16 | |
| *** sambetts is now known as sambetts|afk | 18:18 | |
| mriedem | bauzas: are you pumped to get this in or what?! https://review.openstack.org/#/c/394656/ | 18:18 |
| *** catinthe_ has quit IRC | 18:18 | |
| *** victorhe_ has joined #openstack-nova | 18:18 | |
| bauzas | mriedem: probably 10-4 | 18:19 |
| bauzas | mriedem: lemme look quickly | 18:19 |
| openstackgerrit | STEW TY proposed openstack/nova: Complete verification for os-floating-ips https://review.openstack.org/329704 | 18:20 |
| bauzas | mriedem: dansmith: I don't live in Australia, but Wallaby'd | 18:21 |
| dansmith | \o/ | 18:22 |
| mriedem | you don't have to come from a land down under | 18:22 |
| mriedem | to love vegemite | 18:22 |
| * bauzas goes afk for a couple of hours before going back | 18:22 | |
| bauzas | ++ | 18:22 |
| openstackgerrit | Andrey Volkov proposed openstack/nova: PoC: add hierarchy quota driver https://review.openstack.org/391072 | 18:31 |
| mriedem | @all i'm going to be pushing the release request for the o-1 tag soonish, let me know if there are any changes that are in the gate queue which we should wait to get into that tag, as it's treated like a beta | 18:32 |
| mriedem | i.e. nasty bug fixes | 18:33 |
| jlk | Anybody around that wants to talk nova upgrades and the HUP step? | 18:36 |
| *** yamahata has joined #openstack-nova | 18:39 | |
| *** kenperkins has quit IRC | 18:39 | |
| dansmith | jlk: want? no. | 18:40 |
| *** kaisers_ has joined #openstack-nova | 18:40 | |
| jlk | dansmith: heh, I'm trying to understand how the upgrade levels code works internally. If I'm doing a rolling upgrade, and I've got [upgrade_levels]compute=auto, and I'm to the point where APIs/conductor/scheduler/et al have been restarted to new code, and all the computes have restarted, the docs say I need to send HUP to all the processes so they clear any cached service version data | 18:45 |
| jlk | Given that HUPs will happen non-atomically, I'm assuming what's happening is each service advertises the maximum it could handle, and what it's currently sending out (due to older service versions at one point). The HUPS make each individual process re-evaluate what the maximums everybody could handle are, and thus switch to a newer version to talk. | 18:46 |
| jlk | I'm also assuming that if the HUP is never sent, the services will forever (for the life of the process) talk the older versions, even though everybody could talk newer. | 18:47 |
| jlk | Am I correct in my understanding? | 18:47 |
| dansmith | jlk: right, you can send old stuff as long as everyone is newer, | 18:47 |
| dansmith | so kicking them to send newer stuff need not be atomic | 18:47 |
| *** rmart04_ has joined #openstack-nova | 18:48 | |
| openstackgerrit | Sarafraj Singh proposed openstack/nova: conf: remove deprecated cells driver option https://review.openstack.org/398425 | 18:49 |
| *** rmart04 has quit IRC | 18:49 | |
| *** rmart04_ is now known as rmart04 | 18:49 | |
| *** efried1 has quit IRC | 18:50 | |
| *** huhaoran has joined #openstack-nova | 18:50 | |
| jlk | dansmith: is a HUP disruptive? | 18:50 |
| dansmith | jlk: nope | 18:50 |
| jlk | like, do I need to sprinkle them out, or can I blast them all at once? | 18:51 |
| dansmith | jlk: it will cause each service to re-query for the minimum service version, which is a very simple query, but... | 18:52 |
| *** mdrabe has joined #openstack-nova | 18:54 | |
| openstackgerrit | STEW TY proposed openstack/nova: Transform instance.resize_revert notification https://review.openstack.org/396811 | 18:55 |
| *** whenry has quit IRC | 18:57 | |
| *** gouthamr has quit IRC | 18:57 | |
| *** efried has joined #openstack-nova | 18:58 | |
| dtp | is there a cells v2 meeting today? this is out of date - https://wiki.openstack.org/wiki/Meetings/NovaCellsv2 | 18:59 |
| *** gouthamr has joined #openstack-nova | 18:59 | |
| *** nherciu has joined #openstack-nova | 19:00 | |
| mriedem | dtp: that's up to dansmith | 19:00 |
| mriedem | i don't think there is news, mostly reviews | 19:00 |
| *** jwcroppe has quit IRC | 19:00 | |
| mriedem | making some progress on this series https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/cells-sched-staging | 19:01 |
| dansmith | mriedem: yeah no news really, from me at least | 19:03 |
| *** penick has joined #openstack-nova | 19:03 | |
| dansmith | dtp: do you have things to discuss? | 19:03 |
| jlk | dansmith: another question, nova is configured to start N processes for things like nova-api. Should I be HUPing all of them? | 19:04 |
| dansmith | jlk: yeah, I don't think oslo.service cascades HUPs down to children | 19:04 |
| dansmith | I mean, maybe it does, I have no idea | 19:04 |
| *** gszasz has quit IRC | 19:06 | |
| jlk | fair enough | 19:06 |
| jlk | ugh, pkill, why you gotta be. | 19:06 |
| *** Sukhdev has joined #openstack-nova | 19:06 | |
| dansmith | jlk: pkill is what you want, right? | 19:07 |
| jlk | dansmith: yes, but it's having difficulty finding "nova-consoleauth" | 19:07 |
| jlk | it wants to find "nova-consoleaut" | 19:07 |
| dansmith | use -f | 19:07 |
| dansmith | oh | 19:07 |
| jlk | I'm testing that now | 19:07 |
| *** kaisers_ has quit IRC | 19:07 | |
| jlk | yeah that works. | 19:07 |
| jlk | lame that without -f it truncates the executable | 19:08 |
| dansmith | I dunno what that's about, yeah, but -f is pretty important for most interpreted language executables | 19:08 |
| dansmith | since they often show up as "python" for the name | 19:08 |
| jlk | I always remember Alf. pgrep -alf | 19:09 |
| mriedem | he easts cats | 19:09 |
| mriedem | *eats | 19:09 |
| mriedem | and the dad on that show was the lead singer from judas priest | 19:09 |
| dansmith | hmm, what is -a? | 19:09 |
| jlk | list the full comamnd line | 19:10 |
| jlk | shortcut for --list-full | 19:10 |
| *** ihrachys has quit IRC | 19:10 | |
| dansmith | ah | 19:10 |
| jlk | useful when your command line involve version paths | 19:10 |
| dansmith | yah | 19:10 |
| jlk | 12436 /opt/bbc/openstack-2016.2-newton/nova/bin/python /usr/local/bin/nova-api --config-dir /etc/nova | 19:10 |
| openstackgerrit | Jay Pipes proposed openstack/nova: placement: implement ResourceClass.create() https://review.openstack.org/386844 | 19:11 |
| openstackgerrit | Jay Pipes proposed openstack/nova: placement: REST API for resource classes https://review.openstack.org/391959 | 19:11 |
| openstackgerrit | Jay Pipes proposed openstack/nova: placement: adds ResourceClass.get_by_name() https://review.openstack.org/391917 | 19:11 |
| openstackgerrit | Jay Pipes proposed openstack/nova: placement: implement destroy() resource class https://review.openstack.org/391918 | 19:11 |
| *** jwcroppe has joined #openstack-nova | 19:15 | |
| dansmith | mriedem: gdi, that patch is #2 in the gate and getting kicked out again | 19:15 |
| dansmith | that cinder bug is popping up 300% more than the next one below it | 19:16 |
| mriedem | https://review.openstack.org/#/c/394656/ ? | 19:16 |
| openstackgerrit | Sarafraj Singh proposed openstack/nova: conf: Trivial fix of indentation in 'api' https://review.openstack.org/397834 | 19:16 |
| dansmith | yeah | 19:16 |
| mriedem | there is also a post failure in the gate on that patch | 19:17 |
| mriedem | and yeah http://status.openstack.org//elastic-recheck/index.html#1642111 is the top check queue failure | 19:18 |
| mriedem | smcginnis: ^ are you aware of that? | 19:18 |
| smcginnis | mriedem: No, i had not noticed that one. | 19:18 |
| *** walharthi has quit IRC | 19:18 | |
| smcginnis | mriedem: That's alarming. | 19:18 |
| mriedem | i sounded the alarm in the cinder channel last night but you guys were all out partying or sleeping or otherwise not working late :) | 19:19 |
| *** jamesdenton has joined #openstack-nova | 19:19 | |
| mriedem | it's master and newton so i'm suspecting updated packages on the xenial nodes | 19:19 |
| smcginnis | mriedem: Oh, good clue. | 19:20 |
| *** haplo37_ has quit IRC | 19:20 | |
| mriedem | logstash only goes back 10 days too and it started before that | 19:20 |
| jlk | melwitt: Hi there. Your fix for root_gb being 0 on boot from volume, is that something that could be (easily) backported to Newton (and Mitaka?)? | 19:20 |
| mriedem | jlk: it's not in master right? | 19:21 |
| mriedem | oh i see it has a +2 on it now https://review.openstack.org/#/c/355091/ | 19:21 |
| jlk | mriedem: correct it hasn't merged yet. Just looking for a general assessment on backportability so I can inform some of our customers. | 19:22 |
| mriedem | so i thought we weren't doing this because we had resource providers coming | 19:22 |
| dansmith | mriedem: yeah | 19:23 |
| mriedem | or has this patch substantially changed since newton? | 19:24 |
| dansmith | mriedem: the only argument worth bringing up was to get it into newton because RPs was Ocata, but.. merging it now makes even less sense to me | 19:24 |
| jlk | looks like it changed a lot | 19:24 |
| jlk | and that question, re resource providers, was asked and answered | 19:24 |
| mriedem | i'd definitely not backport this to mitaka | 19:25 |
| dansmith | jlk: not sure what you mean | 19:25 |
| mriedem | that's phase 3 now | 19:25 |
| *** jamielennox|away is now known as jamielennox | 19:26 | |
| mriedem | landing this in master just to rip it out for resource providers is nasty, but it'd have to be done if we were going to backport it to newton | 19:26 |
| dansmith | sigh | 19:26 |
| melwitt | jaypipes had asked me about the patch at summit and recently reviewed it. maybe he can share his thoughts on it regarding RPs | 19:26 |
| mriedem | i'm also not sure what ripping this out after RP looks like, | 19:26 |
| mriedem | or the upgrade implications | 19:26 |
| *** matrohon has joined #openstack-nova | 19:26 | |
| jlk | hrm, I thought there was a question on RPs on this PR, but maybe it was different one | 19:26 |
| mriedem | these aren't the RP PRs you're looking for | 19:27 |
| jlk | indeed | 19:27 |
| dansmith | mriedem: ripping the actual code out isn't hard, it's the fact that we'll have residue of the hack in our DB | 19:27 |
| melwitt | the only residue will be on RequestSpec | 19:27 |
| dansmith | mriedem: not the worst thing, but it just doesn't seem worth it for such a short period | 19:27 |
| dansmith | yeah I know | 19:27 |
| melwitt | okay | 19:27 |
| jlk | it would be worth it for Newton consumers... | 19:28 |
| jlk | they won't all be jumping on Ocata the moment it releases either | 19:28 |
| jlk | ops has a long tail :( | 19:28 |
| dansmith | jlk: I didn't agree that it would, even when we had it proposed for newton | 19:28 |
| *** cdent has quit IRC | 19:28 | |
| *** walharthi has joined #openstack-nova | 19:28 | |
| mriedem | the hack == overwriting the requestspec.flavor.root_gb right? | 19:28 |
| *** haplo37_ has joined #openstack-nova | 19:29 | |
| mriedem | goofing with reqspec does worry me given the small amount of that code i've looked at recently | 19:29 |
| jlk | We're talking about months and months in production, where resources are not appropriately accounted for | 19:29 |
| *** lpetrut has joined #openstack-nova | 19:29 | |
| mriedem | jlk: this is also a latent issue, not a regression | 19:29 |
| dansmith | and reqspec is what we rebuild instances from, so we basically have to account for it forever | 19:30 |
| dansmith | latent since the beginning of time | 19:30 |
| dansmith | or beginning of BFV at least | 19:30 |
| mriedem | so now instance.flavor.root_gb == 0 means something different from reqspec.flavor.root_gb? | 19:30 |
| mriedem | is that right? | 19:30 |
| mriedem | i also wonder how this would ever muddy these waters even more https://specs.openstack.org/openstack/nova-specs/specs/newton/approved/flavor-root-disk-none.html | 19:31 |
| mriedem | if that got in | 19:31 |
| *** READ10 has quit IRC | 19:32 | |
| jaypipes | mriedem, dansmith, jlk: regardling melwitt's patch, it fixes the latent issue and can be fairly easily removed once we complete resource providers, but remember that is a ways in the future (we need to get flavors and other things constructing a set of requested resource amounts). I think it's worth getting melanie's patch in for Ocata. | 19:33 |
| dansmith | jaypipes: it can't be removed because the residue remains in persisted objects forever | 19:34 |
| *** betherly_ has joined #openstack-nova | 19:35 | |
| openstackgerrit | Gábor Antal proposed openstack/nova: Transform instance.create.error notification https://review.openstack.org/397715 | 19:35 |
| *** ijw has quit IRC | 19:35 | |
| jaypipes | dansmith: not following you... the residue in this case is the setting of root_gb to zero for the purposes of resource accounting and scheduling. | 19:35 |
| dtp | dansmith nothing to discuss, just trying to stay informed :) | 19:35 |
| dansmith | jaypipes: in the req spec of any instance created in the window in which this hack is in place | 19:36 |
| *** lpetrut has quit IRC | 19:36 | |
| mriedem | can we migrate the old hack data when we encounter it in any way? | 19:36 |
| dansmith | mriedem: it's in a serialized object, which means it's not easy to convert in batch, so we just have to support it on load somehow forever, | 19:37 |
| jaypipes | dansmith: still not following you... the request spec has a flavor.root_gb set to zero when the instance is booted from volume. how is that "hack data"? | 19:37 |
| dansmith | or take the pain of touching every reqspec in the database to inspect/fix | 19:37 |
| dansmith | jaypipes: I'm not sure how to answer that | 19:38 |
| mriedem | could we work that into our nova-manage db online_data_migrations command at some point? you'd still have to pull all of those records to inspect them, which sucks, i'm not sure if you could mark that somehow to say you've processed a record so don't look again | 19:38 |
| mriedem | or, do something like ^ in pike but then drop it in queens, so then it's only a one time hit on upgrade | 19:39 |
| dansmith | mriedem: sure, but that kind of migration would be pretty heavy | 19:39 |
| mriedem | not a schema migration, so not downtime, but yeah | 19:39 |
| mriedem | what happens if after we start using resource providers, we just stop looking at reqspec.flavor.root_gb? | 19:40 |
| *** tongli has quit IRC | 19:40 | |
| *** fragatin_ has quit IRC | 19:41 | |
| mriedem | i get we'd have dirty data in the reqspec, but if we don't read it after RPs, do we care? | 19:41 |
| jaypipes | mriedem: in resource providers, it's just DISK_GB -- there's no separation of root, swap or ephemeral disk space. | 19:41 |
| dansmith | the point of reqspec.flavor is to recreate instances from there | 19:41 |
| *** fragatina has joined #openstack-nova | 19:41 | |
| jaypipes | dansmith, mriedem: guys, what am I missing here? are you in favor of adding an online migration that changes the request_spec.flavor.root_gb for existing instances? is that the main hangup? sorry for not following :( | 19:45 |
| dansmith | no | 19:45 |
| dansmith | my major hangup is that this solves a problem that has been around forever, and that RPs is poised to solve *correctly* | 19:47 |
| openstackgerrit | Sarafraj Singh proposed openstack/nova: conf: Move api options to a group https://review.openstack.org/384490 | 19:47 |
| *** TravT has quit IRC | 19:47 | |
| dansmith | we have multiple overloaded meanings for root_gb=<=0 and this adds another, which will persist in our DB, admittedly in fewer places than originally proposed | 19:48 |
| mriedem | jaypipes: my hangup is not knowing the implications around dealing with this after RPs, as noted in the review | 19:48 |
| mriedem | but yeah the multiple semantically different meanins for flavor.root_gb = 0 worries me | 19:48 |
| dansmith | we're reporting resources to placement properly about BFV instances already | 19:48 |
| dansmith | mriedem: right | 19:48 |
| dansmith | I feel like we've discussed this to death | 19:49 |
| dansmith | ya'll feel free to merge with my -1 on there | 19:49 |
| mriedem | so i was exploring options around doing online data migrations to remove those after RPs | 19:49 |
| mriedem | i also -1ed | 19:49 |
| jaypipes | dansmith: even after RPs, we still will need to have flavor.root_gb, flavor.swap_gb and flavor.ephemeral_gb, though. Since the RP only cares about DISK_GB in its entirety. | 19:50 |
| dansmith | yeah I know | 19:50 |
| *** oanson has joined #openstack-nova | 19:50 | |
| mriedem | so dumb question, but to check my understanding, a BFV instance RP consumer is just going to have a DISK_GB allocation of 0 right? | 19:51 |
| jaypipes | if a reqspec is read to rebuild an existing instance, the flavor.root_gb can/should be re-evaluated to determine if the instance was booted from volume and set it to 0 if it was. | 19:51 |
| jaypipes | mriedem: yes, and that is already in the codebase now. | 19:51 |
| jaypipes | dan added that. | 19:51 |
| jaypipes | IIRC | 19:51 |
| *** pcaruana has quit IRC | 19:51 | |
| mriedem | ok i just wanted to make sure i'm speaking the same language | 19:51 |
| mriedem | so... | 19:52 |
| mriedem | let's say we land this, | 19:52 |
| mriedem | and i have a BFV instance, so reqspec.flavor.root_gb = 0 | 19:52 |
| mriedem | then i rebuild with a different flavor, | 19:53 |
| jaypipes | it would still be root_gb = 0. | 19:54 |
| mriedem | oh i guess we don't allow changing volumes or the flavor with a rebuild | 19:54 |
| jaypipes | right. | 19:54 |
| mriedem | b/c changing the flavor would be a resize... | 19:54 |
| jaypipes | yes. | 19:54 |
| *** gszasz has joined #openstack-nova | 20:00 | |
| *** rmart04 has left #openstack-nova | 20:02 | |
| *** penick has quit IRC | 20:09 | |
| openstackgerrit | Ludovic Beliveau proposed openstack/nova: PCI: Release PCI devices on drop_move_claim() https://review.openstack.org/370374 | 20:11 |
| openstackgerrit | Eric Fried proposed openstack/nova: Introduce PowerVMLiveMigrateData https://review.openstack.org/391284 | 20:12 |
| mriedem | i was poking around to see if this would impact any sort of resize with a volume-backed instance, doesn't look like it would | 20:12 |
| mriedem | at least it's not obvoius | 20:12 |
| *** sean-k-mooney has quit IRC | 20:12 | |
| mriedem | *obvious | 20:12 |
| *** kaisers_ has joined #openstack-nova | 20:12 | |
| *** sean-k-mooney has joined #openstack-nova | 20:13 | |
| *** ijw_ has joined #openstack-nova | 20:13 | |
| *** annegentle has quit IRC | 20:15 | |
| *** cdent has joined #openstack-nova | 20:16 | |
| *** kaisers_ has quit IRC | 20:17 | |
| *** annegentle has joined #openstack-nova | 20:17 | |
| *** tbachman has quit IRC | 20:18 | |
| *** ijw_ has quit IRC | 20:19 | |
| openstackgerrit | Peter Hamilton proposed openstack/nova-specs: Add support for certificate validation https://review.openstack.org/357151 | 20:21 |
| moshele | jaypipes: hi | 20:23 |
| *** Sukhdev has quit IRC | 20:23 | |
| jaypipes | moshele: hey | 20:24 |
| *** penick has joined #openstack-nova | 20:24 | |
| *** slaweq_ has quit IRC | 20:25 | |
| moshele | jaypipes: Just wandering if you +2 https://review.openstack.org/#/c/182242/ User-controlled SR-IOV ports allocation also can I help with the https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/nested-resource-providers patches | 20:26 |
| *** slaweq_ has joined #openstack-nova | 20:26 | |
| jaypipes | moshele: I am not on nova-drivers team. | 20:26 |
| *** rtheis_ has quit IRC | 20:26 | |
| jaypipes | moshele: so I cannot +2 any spec. that said, I'm really +0 on that and told mriedem so. | 20:26 |
| *** slaweq_ has quit IRC | 20:29 | |
| *** slaweq_ has joined #openstack-nova | 20:29 | |
| moshele | jaypipes: I see. and regarding my second question can I work on the :bp/nested-resource-providers patches ( I mean fix my own comments) | 20:30 |
| *** kaisers_ has joined #openstack-nova | 20:30 | |
| *** jwcroppe has quit IRC | 20:30 | |
| *** ijw has joined #openstack-nova | 20:31 | |
| *** jwcroppe has joined #openstack-nova | 20:32 | |
| diana_clarke | mriedem: Apologies for leaving the questions raised on that tempest review hanging. I didn't want to detract from spec reviews, so I figured I'd circle back next week for that task. | 20:32 |
| mriedem | diana_clarke: i was just digging into that patch | 20:32 |
| *** TravT has joined #openstack-nova | 20:34 | |
| mriedem | i'm having a hard time sort out some of the failures i'm seeing | 20:35 |
| mriedem | like in test_shelve_volume_backed_instance it fails creating the volume-backed instance, not during the shelve operatoin | 20:35 |
| mriedem | *operation | 20:35 |
| *** rtheis has joined #openstack-nova | 20:36 | |
| openstackgerrit | ijw-ubuntu proposed openstack/nova: Fix incorrect port unbind request in nova on instance delete https://review.openstack.org/387687 | 20:37 |
| *** TravT has quit IRC | 20:39 | |
| openstackgerrit | Dharini Chandrasekar proposed openstack/nova: Improving help text for xenapi_vmops_opts https://review.openstack.org/347085 | 20:40 |
| *** unicell has quit IRC | 20:46 | |
| openstackgerrit | Ludovic Beliveau proposed openstack/nova: PCI: Release PCI devices on drop_move_claim() https://review.openstack.org/370374 | 20:46 |
| mriedem | diana_clarke: sorting it out in cinder | 20:47 |
| mriedem | it's other bugs | 20:47 |
| *** matrohon has quit IRC | 20:47 | |
| *** raildo has quit IRC | 20:50 | |
| jaypipes | moshele: yep, still working on it :) | 20:51 |
| jaypipes | moshele: if you would like to push a fix, go for it. | 20:52 |
| moshele | jaypipes: ok thanks :) | 20:52 |
| *** unicell has joined #openstack-nova | 20:53 | |
| *** rtheis has quit IRC | 20:53 | |
| *** slaweq_ has quit IRC | 20:54 | |
| *** slaweq_ has joined #openstack-nova | 20:56 | |
| *** Swami has quit IRC | 20:56 | |
| *** rcernin has joined #openstack-nova | 20:58 | |
| *** krtaylor has quit IRC | 20:58 | |
| *** Swami has joined #openstack-nova | 20:59 | |
| diana_clarke | mriedem: Yeah, this all seems fishy (except maybe the snapshot skip). I'm not at all confident that my skips are appropriate until we know more about the why... | 21:01 |
| melwitt | mriedem: cells meeting if you're around | 21:02 |
| mriedem | diana_clarke: https://bugs.launchpad.net/cinder/+bug/1642394 | 21:02 |
| openstack | Launchpad bug 1642394 in Cinder "NFS: Creating multiple volumes from the same image fails when image caching is turned on by default" [Undecided,New] | 21:02 |
| mriedem | diana_clarke: so devstack enabled the image cache in cinder by default, | 21:02 |
| mriedem | but that uses cloning | 21:02 |
| mriedem | which is why we're going down that path and it blows up | 21:03 |
| mriedem | quick fix is to disable the cinder image cache in my devstack-plugin-nfs patch | 21:03 |
| *** hughsaunders has quit IRC | 21:03 | |
| *** whenry has joined #openstack-nova | 21:04 | |
| diana_clarke | ah, I see... | 21:07 |
| *** huhaoran has quit IRC | 21:07 | |
| *** hughsaunders has joined #openstack-nova | 21:07 | |
| *** mvk has joined #openstack-nova | 21:11 | |
| *** Guest45910 has quit IRC | 21:11 | |
| mriedem | diana_clarke: ok updated https://review.openstack.org/#/c/395887/5/devstack/override-defaults | 21:12 |
| mriedem | we'll see how that goes | 21:12 |
| *** tbachman has joined #openstack-nova | 21:13 | |
| *** volenbovsky has joined #openstack-nova | 21:13 | |
| *** rfolco has quit IRC | 21:14 | |
| alaski | dansmith: melwitt I think it makes sense to move consoleauth into the cells, https://review.openstack.org/#/c/390660/1/doc/source/cells.rst@266 . Just MHO | 21:19 |
| dansmith | alaski: there was an upcall concern | 21:19 |
| alaski | I was thinking that having it in the cell eliminated any upcall. I may be missing something of course | 21:20 |
| melwitt | yeah, as long as we can put any related tables to the DB-only token validation in the cell it wouldn't be an up call AFAICT | 21:21 |
| melwitt | I intend to gather the detail, link to the spec, notes etc on that and put it on our etherpad. I can't remember the details off the top of my head | 21:22 |
| melwitt | of what PaulMurray had proposed | 21:22 |
| alaski | IIRC he was working on keeping the token tables in the cell db. But you've got it covered so I'll stop butting in :) | 21:27 |
| *** nherciu has quit IRC | 21:28 | |
| melwitt | alaski: wait are you just thinking of moving consoleauth as-is, one instance per cell? | 21:28 |
| mriedem | alaski: your butt is always welcome here | 21:28 |
| melwitt | alaski: I don't have it covered, so I'm glad you're commenting. the only plan I have is to aggregate the spec and notes that are out on the various summit/midcycle etherpads and hope that it's clear enough :P | 21:29 |
| alaski | mriedem: hah. I will wield it when necessary then | 21:30 |
| alaski | melwitt: yeah, pretty much as is. The only catch I thought was that the API needed to take an instance uuid so it could proxy requests to the right cell. That's what PaulMurray was working on I thought and I don't know where it ended up | 21:31 |
| *** catinthe_ has joined #openstack-nova | 21:32 | |
| melwitt | alaski: I see. I had thought the plan involved getting rid of the consoleauth service altogether, so I was thinking of something more complicated | 21:32 |
| *** krtaylor has joined #openstack-nova | 21:33 | |
| *** Jeffrey4l has quit IRC | 21:33 | |
| alaski | gotcha | 21:34 |
| *** cdelatte has quit IRC | 21:34 | |
| *** catintheroof has quit IRC | 21:36 | |
| melwitt | alaski: I found this about the instance uuid https://review.openstack.org/#/c/325382/ abandoned | 21:37 |
| *** krtaylor has quit IRC | 21:37 | |
| *** gouthamr has quit IRC | 21:37 | |
| melwitt | https://review.openstack.org/#/q/topic:bp/convert-consoles-to-objects | 21:38 |
| alaski | yeah. I thought it got further than that, but that was the basic plan. | 21:39 |
| melwitt | okay, so maybe I can figure out how to pick these back up | 21:40 |
| mriedem | fuuuke yeah let's do this https://review.openstack.org/#/c/392236/ | 21:42 |
| melwitt | mriedem: on that, are we really sure n-net isn't the default in devstack still? I was trying to figure that out last week and then forgot about it | 21:43 |
| melwitt | I could have sworn I started a new devstack last week and got n-net | 21:43 |
| *** kaisers_ has quit IRC | 21:43 | |
| mriedem | i'd have to try it out. neutron is definitely the default in the CI system....but that's devstack-gate and doesn't mean anything to joe developer | 21:43 |
| mriedem | melwitt: https://github.com/openstack-dev/devstack/blob/master/stackrc#L61 | 21:44 |
| mriedem | https://github.com/openstack-dev/devstack/commit/6a008fa74bce0497b7902dd95300599a3026e2dd | 21:44 |
| melwitt | weird, I wonder what I did then | 21:45 |
| dansmith | it's been the default in devstack for a couple months right? | 21:45 |
| dansmith | melwitt: blank config or something else? | 21:45 |
| *** krtaylor has joined #openstack-nova | 21:45 | |
| melwitt | blank | 21:45 |
| dansmith | huh | 21:45 |
| *** xinliang has quit IRC | 21:46 | |
| melwitt | yeah, I'm questioning myself now but I did set up a lot of new devstack VMs last week and thought I got n-net. I have to try it again to be certain | 21:47 |
| dansmith | are you on current devstack? | 21:47 |
| *** krtaylor has quit IRC | 21:47 | |
| melwitt | yeah, new VM, new cloned devstack, blank local.conf except log dir and stack.sh | 21:48 |
| melwitt | I'm going to do it again because it doesn't make sense | 21:48 |
| dansmith | huh | 21:48 |
| melwitt | now that I have my super server, I can have a lot of devstacks | 21:49 |
| *** Sukhdev has joined #openstack-nova | 21:49 | |
| * melwitt wipes fingernails on lapel | 21:49 | |
| dansmith | heh | 21:49 |
| openstackgerrit | Ludovic Beliveau proposed openstack/nova: PCI: Update MAC address for NICs PF passthrough devices https://review.openstack.org/361438 | 21:49 |
| *** jwcroppe has quit IRC | 21:50 | |
| *** krtaylor has joined #openstack-nova | 21:50 | |
| *** irenab has joined #openstack-nova | 21:50 | |
| *** catintheroof has joined #openstack-nova | 21:51 | |
| *** catinthe_ has quit IRC | 21:51 | |
| *** krtaylor has quit IRC | 21:53 | |
| *** ayoung has quit IRC | 21:55 | |
| *** irenab has quit IRC | 21:55 | |
| *** krtaylor has joined #openstack-nova | 21:56 | |
| *** sree has joined #openstack-nova | 21:56 | |
| mriedem | edleafe: jaypipes: is https://review.openstack.org/#/c/386844/ ready for review? confused as to whether or not edleafe is going to make changes there | 21:58 |
| mriedem | looks like a follow up patch if anything | 21:58 |
| *** krtaylor has quit IRC | 21:59 | |
| *** tjones has quit IRC | 21:59 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/397916 | 21:59 |
| *** krtaylor has joined #openstack-nova | 21:59 | |
| *** xinliang has joined #openstack-nova | 22:00 | |
| *** sree has quit IRC | 22:00 | |
| *** slaweq_ has quit IRC | 22:06 | |
| edleafe | mriedem: yes, in a follow-up | 22:06 |
| *** jwcroppe has joined #openstack-nova | 22:07 | |
| *** slaweq_ has joined #openstack-nova | 22:08 | |
| cdent | edleafe: you have a bit of spare brain, see my comment on: https://review.openstack.org/#/c/398473/ | 22:09 |
| *** kylek3h has quit IRC | 22:09 | |
| cdent | s/you/if you/ | 22:09 |
| cdent | i'm really struggling to untangle things | 22:09 |
| openstackgerrit | Matthew Edmonds proposed openstack/nova: update policy UT fixtures https://review.openstack.org/398610 | 22:09 |
| edleafe | cdent: it's late, go to bed. I can take a crack at it | 22:10 |
| cdent | naw mate I'm larnin' | 22:10 |
| *** betherly has quit IRC | 22:10 | |
| *** betherly_ is now known as betherly | 22:10 | |
| cdent | it's only 10 after all | 22:10 |
| *** lifeless has quit IRC | 22:11 | |
| *** lifeless has joined #openstack-nova | 22:12 | |
| *** annegentle has quit IRC | 22:14 | |
| *** thorst_ has quit IRC | 22:18 | |
| *** thorst_ has joined #openstack-nova | 22:19 | |
| cdent | edleafe, i think i've figured it out at least a lame fix | 22:21 |
| mriedem | why do we need the custom resource classes to start at id=10000 in https://review.openstack.org/#/c/386844/ ? | 22:21 |
| mriedem | what are we buffering against when the standard resource classes aren't in the db | 22:22 |
| *** tjones has joined #openstack-nova | 22:22 | |
| *** whenry has quit IRC | 22:23 | |
| *** thorst_ has quit IRC | 22:24 | |
| cdent | mriedem: those that are not in in the db are still in the cache by way of an id | 22:24 |
| cdent | iirc | 22:24 |
| *** slaweq_ has quit IRC | 22:25 | |
| cdent | the id of the standard ones were at one point an index into a list | 22:25 |
| *** slaweq_ has joined #openstack-nova | 22:25 | |
| cdent | I'm not sure if that's still the case as I've not looked recently | 22:25 |
| mriedem | not seeing that in the resource cache code | 22:25 |
| mriedem | the caches are updated in _refresh_from_db | 22:25 |
| mriedem | and that's only for the custom resource calsses | 22:25 |
| mriedem | *classes | 22:25 |
| * cdent refreshes his state | 22:25 | |
| *** jamesdenton has quit IRC | 22:26 | |
| edleafe | mriedem: the code checks for the standard first, and then the db | 22:28 |
| mriedem | edleafe: yeah i know | 22:28 |
| *** walharthi has quit IRC | 22:28 | |
| mriedem | but the standard rc's aren't in the db | 22:28 |
| edleafe | mriedem: nope | 22:28 |
| mriedem | so the rc ids in the db could start at 1 | 22:28 |
| mriedem | right? | 22:28 |
| cdent | mriedem: they are in the cache though | 22:28 |
| mriedem | cdent: they aren't in the cache | 22:28 |
| edleafe | the standards have IDs, that are based on their index in the code | 22:28 |
| cdent | see around 125 in nova/db/sqlalchemy/resource_class_cache.py | 22:29 |
| mriedem | standard rcs aren't ever put into id_cache or str_cache | 22:29 |
| mriedem | b/c those are only populated in _refresh_from_db | 22:29 |
| *** penick has quit IRC | 22:29 | |
| mriedem | which is only ever called if the thing we're looking for isn't in the standard set | 22:29 |
| cdent | that's true, but they are in ResourceClassCache object | 22:29 |
| mriedem | which is the field type thing | 22:29 |
| mriedem | cdent: how? | 22:30 |
| cdent | and the interface to ResourceClassCache is used everywhere, with string_from_id and id_from_string | 22:30 |
| mriedem | so i guess what we're saying is, if i have a custom resource class with id=1, | 22:30 |
| mriedem | that could return STANDARD.MEMORY_MB b/c of the id conflict | 22:31 |
| edleafe | yes | 22:31 |
| mriedem | well gfdi | 22:31 |
| mriedem | ok :) | 22:31 |
| edleafe | it's all in the name of interop | 22:31 |
| edleafe | any ID less than 10000 is guaranteed to be on all clouds | 22:31 |
| edleafe | s/to be/to be the same/ | 22:32 |
| cdent | it is all bit hard to process (mentally) in the name of being somewhat easier to process code-ally | 22:33 |
| cdent | I guess | 22:33 |
| mriedem | i hope i'm not around when that 10000 limit is not enough... | 22:33 |
| mriedem | ok so -1 is just for some missing tests in https://review.openstack.org/#/c/386844/ then i'm +2 | 22:34 |
| edleafe | 640K should be enough for anybody | 22:34 |
| cdent | if we let the standard classes get that big we've done something very very very wrong | 22:34 |
| mriedem | tell that to jamespd's baremetal + NFV cloud | 22:34 |
| mriedem | where he has a million unique types | 22:34 |
| cdent | he's the cause of all this malarkey | 22:34 |
| cdent | sic? | 22:35 |
| edleafe | they'll all be custom, though, so standard classes are safe from him | 22:36 |
| mriedem | he'll find a way | 22:36 |
| mriedem | to throw a flaming ball of suck into this | 22:37 |
| melwitt | heh. I agree the standard classes should be safe. the custom classes should be wary | 22:39 |
| *** tblakes has quit IRC | 22:39 | |
| *** volenbovsky has quit IRC | 22:40 | |
| openstackgerrit | Chris Dent proposed openstack/nova: rt: use a single ResourceTracker object instance https://review.openstack.org/398473 | 22:40 |
| cdent | edleafe: I think that ^ fixes it | 22:40 |
| *** oanson has quit IRC | 22:41 | |
| cdent | it == the thing I was trying to untangle | 22:41 |
| mriedem | alright my stomach is eating itself so it's time to leave | 22:41 |
| *** venkhari has joined #openstack-nova | 22:41 | |
| *** mriedem has quit IRC | 22:41 | |
| *** kylek3h has joined #openstack-nova | 22:43 | |
| *** bvanhav has quit IRC | 22:43 | |
| *** ayoung has joined #openstack-nova | 22:44 | |
| edleafe | cdent: congrats | 22:45 |
| *** venkhari has quit IRC | 22:46 | |
| *** pradk has quit IRC | 22:48 | |
| *** rama_y has quit IRC | 22:50 | |
| *** catintheroof has quit IRC | 22:54 | |
| openstackgerrit | Ed Leafe proposed openstack/nova: placement: REST API for resource classes https://review.openstack.org/391959 | 22:56 |
| edleafe | jaypipes: ^^ small refactor and added a test | 22:56 |
| *** victorhe_ has quit IRC | 23:01 | |
| *** jheroux has quit IRC | 23:04 | |
| *** dave-mccowan has quit IRC | 23:07 | |
| *** penick has joined #openstack-nova | 23:07 | |
| *** mdrabe has quit IRC | 23:08 | |
| *** gszasz has quit IRC | 23:09 | |
| *** krtaylor has quit IRC | 23:09 | |
| *** jwcroppe has quit IRC | 23:09 | |
| *** jwcroppe has joined #openstack-nova | 23:10 | |
| *** jwcroppe has quit IRC | 23:14 | |
| *** annegentle has joined #openstack-nova | 23:15 | |
| *** xyang1 has quit IRC | 23:15 | |
| jaypipes | edleafe: cheers! | 23:16 |
| *** ducttape_ has quit IRC | 23:16 | |
| *** edmondsw has quit IRC | 23:17 | |
| *** penick has quit IRC | 23:20 | |
| *** penick has joined #openstack-nova | 23:20 | |
| *** cleong has quit IRC | 23:20 | |
| *** penick has quit IRC | 23:23 | |
| *** diogogmt has quit IRC | 23:25 | |
| *** jwcroppe has joined #openstack-nova | 23:26 | |
| *** jwcroppe has quit IRC | 23:31 | |
| *** Jeffrey4l has joined #openstack-nova | 23:31 | |
| *** lamt has quit IRC | 23:33 | |
| gmann | mriedem: yea so port id is allowed uuid or null only. In schema definition will be string but with format uuid | 23:35 |
| openstackgerrit | Moshe Levi proposed openstack/nova: placement: adds REST API for nested providers https://review.openstack.org/384807 | 23:36 |
| openstackgerrit | Moshe Levi proposed openstack/nova: placement: add get_all_by_root_provider_uuid() https://review.openstack.org/377215 | 23:36 |
| openstackgerrit | Moshe Levi proposed openstack/nova: placement: update client to set parent provider https://review.openstack.org/385693 | 23:36 |
| openstackgerrit | Moshe Levi proposed openstack/nova: placement: add nested resource providers https://review.openstack.org/377138 | 23:36 |
| *** claudiub has joined #openstack-nova | 23:38 | |
| *** irenab has joined #openstack-nova | 23:38 | |
| *** khamtamtun has joined #openstack-nova | 23:39 | |
| *** rfolco has joined #openstack-nova | 23:40 | |
| *** mlavalle has quit IRC | 23:40 | |
| *** khamtamtun has quit IRC | 23:42 | |
| *** irenab has quit IRC | 23:43 | |
| *** kaisers_ has joined #openstack-nova | 23:44 | |
| *** kaisers_ has quit IRC | 23:48 | |
| *** cdelatte has joined #openstack-nova | 23:49 | |
| *** david-lyle_ has joined #openstack-nova | 23:50 | |
| *** erhudy has quit IRC | 23:51 | |
| *** david-lyle has quit IRC | 23:51 | |
| *** rcernin has quit IRC | 23:52 | |
| *** walharthi has joined #openstack-nova | 23:53 | |
| openstackgerrit | melanie witt proposed openstack/nova: WIP Experimenting with CellDatabases test fixture https://review.openstack.org/398645 | 23:54 |
| *** Swami has quit IRC | 23:58 | |
| *** baoli has quit IRC | 23:59 | |
| melwitt | dansmith: I just put up my experiment ^ but the series underneath needs a rebase | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!