| *** yamahata has quit IRC | 00:04 | |
| *** yuntongjin has joined #openstack-nova | 00:05 | |
| *** ssurana has quit IRC | 00:05 | |
| *** yamahata has joined #openstack-nova | 00:06 | |
| openstackgerrit | Eric Brown proposed openstack/nova-specs: VMware: store vCenter user/password in Keystone https://review.openstack.org/234480 | 00:11 |
|---|---|---|
| *** EinstCrazy has quit IRC | 00:11 | |
| *** otter768 has quit IRC | 00:11 | |
| *** asettle has quit IRC | 00:12 | |
| *** otter768 has joined #openstack-nova | 00:12 | |
| *** zhangjn has quit IRC | 00:12 | |
| *** vilobhmm11 has quit IRC | 00:12 | |
| openstackgerrit | Eric Brown proposed openstack/nova-specs: VMware: store vCenter user/password in Keystone https://review.openstack.org/234480 | 00:14 |
| *** pm90_ has joined #openstack-nova | 00:19 | |
| *** vilobhmm11 has joined #openstack-nova | 00:22 | |
| *** vilobhmm11 has quit IRC | 00:22 | |
| *** angdraug has quit IRC | 00:26 | |
| *** tongli has joined #openstack-nova | 00:27 | |
| *** thorst has joined #openstack-nova | 00:29 | |
| openstackgerrit | Eric Brown proposed openstack/nova-specs: VMware: store vCenter user/password in Keystone https://review.openstack.org/234480 | 00:31 |
| *** tongli has quit IRC | 00:32 | |
| *** vilobhmm11 has joined #openstack-nova | 00:33 | |
| *** rfolco has quit IRC | 00:35 | |
| *** su_zhang has quit IRC | 00:36 | |
| *** electrocucaracha has quit IRC | 00:36 | |
| *** darrenc_afk is now known as darrenc | 00:36 | |
| *** dave-mccowan has joined #openstack-nova | 00:40 | |
| *** otter768 has quit IRC | 00:41 | |
| *** browne has quit IRC | 00:41 | |
| *** otter768 has joined #openstack-nova | 00:42 | |
| *** diogogmt has quit IRC | 00:42 | |
| *** mnestratov has quit IRC | 00:42 | |
| *** baoli has joined #openstack-nova | 00:43 | |
| *** haigang has joined #openstack-nova | 00:45 | |
| *** baoli_ has joined #openstack-nova | 00:46 | |
| *** haomaiwang has joined #openstack-nova | 00:47 | |
| *** markvoelker has quit IRC | 00:47 | |
| *** baoli has quit IRC | 00:49 | |
| *** haomaiwang has quit IRC | 00:51 | |
| *** baoli_ has quit IRC | 00:52 | |
| *** links has joined #openstack-nova | 00:53 | |
| *** eliqiao has joined #openstack-nova | 00:54 | |
| *** lykinsbd_ has quit IRC | 00:59 | |
| *** eliqiao has quit IRC | 01:00 | |
| *** baoli has joined #openstack-nova | 01:00 | |
| *** jcoufal_ has quit IRC | 01:01 | |
| *** shangxdy has joined #openstack-nova | 01:04 | |
| *** lykinsbd_ has joined #openstack-nova | 01:04 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Resize: Return to initial state instead of ACTIVE when resize fails https://review.openstack.org/202379 | 01:05 |
| *** EinstCrazy has joined #openstack-nova | 01:05 | |
| *** sneti has joined #openstack-nova | 01:11 | |
| *** vilobhmm11 has quit IRC | 01:13 | |
| *** zhenguo has joined #openstack-nova | 01:15 | |
| *** baoli has quit IRC | 01:15 | |
| *** shuquan has joined #openstack-nova | 01:17 | |
| *** eliqiao has joined #openstack-nova | 01:18 | |
| *** baoli has joined #openstack-nova | 01:18 | |
| *** mrkz has quit IRC | 01:19 | |
| *** haomaiwa_ has joined #openstack-nova | 01:20 | |
| *** wanghua has joined #openstack-nova | 01:20 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: Ignore errorcode=4 when executing `cryptsetup remove` command https://review.openstack.org/215252 | 01:22 |
| *** asettle has joined #openstack-nova | 01:23 | |
| *** haomaiwa_ has quit IRC | 01:24 | |
| *** mmedvede has quit IRC | 01:26 | |
| *** mmedvede has joined #openstack-nova | 01:26 | |
| *** zhangjn has joined #openstack-nova | 01:28 | |
| *** mdrabe has joined #openstack-nova | 01:28 | |
| *** shuquan has quit IRC | 01:29 | |
| *** shuquan has joined #openstack-nova | 01:29 | |
| *** dims has quit IRC | 01:33 | |
| *** dims has joined #openstack-nova | 01:34 | |
| *** alex_klimov has quit IRC | 01:34 | |
| *** baoli has quit IRC | 01:34 | |
| *** baoli has joined #openstack-nova | 01:35 | |
| *** apoorvad has quit IRC | 01:36 | |
| *** shuquan has quit IRC | 01:40 | |
| *** ijw has quit IRC | 01:41 | |
| *** jwcroppe has joined #openstack-nova | 01:45 | |
| *** shuquan has joined #openstack-nova | 01:46 | |
| *** markvoelker has joined #openstack-nova | 01:48 | |
| *** baoli has quit IRC | 01:49 | |
| *** baoli has joined #openstack-nova | 01:50 | |
| openstackgerrit | Rui Chen proposed openstack/nova: Only return latest fault of each instance in db api https://review.openstack.org/223392 | 01:50 |
| *** markvoelker has quit IRC | 01:53 | |
| *** nic has quit IRC | 01:56 | |
| *** baoli has quit IRC | 02:00 | |
| *** armax has quit IRC | 02:00 | |
| *** baoli has joined #openstack-nova | 02:00 | |
| *** haomaiwang has joined #openstack-nova | 02:01 | |
| *** haomaiwang has quit IRC | 02:01 | |
| *** haomaiwang has joined #openstack-nova | 02:01 | |
| *** jwcroppe has quit IRC | 02:01 | |
| *** alex_xu has joined #openstack-nova | 02:02 | |
| *** annegentle has quit IRC | 02:03 | |
| *** baoli has quit IRC | 02:05 | |
| *** pm90_ has quit IRC | 02:05 | |
| *** gyee has quit IRC | 02:14 | |
| *** pdardeau has joined #openstack-nova | 02:16 | |
| *** browne has joined #openstack-nova | 02:19 | |
| *** mtanino has quit IRC | 02:20 | |
| *** tpeoples has joined #openstack-nova | 02:20 | |
| *** boris-42 has quit IRC | 02:28 | |
| *** VW has joined #openstack-nova | 02:28 | |
| *** baoli has joined #openstack-nova | 02:31 | |
| *** mdrabe has quit IRC | 02:36 | |
| *** mdrabe has joined #openstack-nova | 02:36 | |
| openstackgerrit | Li Yingjun proposed openstack/python-novaclient: Add pagination params for flavor list https://review.openstack.org/234520 | 02:38 |
| *** yamahata has quit IRC | 02:40 | |
| *** mdrabe has quit IRC | 02:44 | |
| *** mdrabe has joined #openstack-nova | 02:45 | |
| *** haigang has quit IRC | 02:45 | |
| *** armax has joined #openstack-nova | 02:46 | |
| *** markvoelker has joined #openstack-nova | 02:49 | |
| *** penick has joined #openstack-nova | 02:50 | |
| *** markvoelker has quit IRC | 02:53 | |
| *** unicell has quit IRC | 02:53 | |
| *** mdrabe has quit IRC | 02:53 | |
| *** sneti has quit IRC | 02:54 | |
| *** achanda has joined #openstack-nova | 02:57 | |
| *** pdardeau has quit IRC | 02:58 | |
| *** haomaiwang has quit IRC | 03:01 | |
| *** 18WAAWN0R has joined #openstack-nova | 03:01 | |
| *** rfolco has joined #openstack-nova | 03:07 | |
| *** tpeoples is now known as tdp | 03:10 | |
| *** amotoki has joined #openstack-nova | 03:14 | |
| *** tdp is now known as tpeoples | 03:14 | |
| *** 18WAAWN0R has quit IRC | 03:15 | |
| *** haomaiwang has joined #openstack-nova | 03:16 | |
| *** reed_ has joined #openstack-nova | 03:16 | |
| openstackgerrit | Hiroyuki Eguchi proposed openstack/nova: Cannot rebuild a instance booted from volume https://review.openstack.org/201458 | 03:17 |
| *** rfolco has quit IRC | 03:18 | |
| *** penick has quit IRC | 03:19 | |
| *** otter768 has quit IRC | 03:25 | |
| *** otter768 has joined #openstack-nova | 03:25 | |
| *** VW has quit IRC | 03:26 | |
| *** VW has joined #openstack-nova | 03:26 | |
| *** VW_ has joined #openstack-nova | 03:27 | |
| *** VW has quit IRC | 03:27 | |
| *** VW_ has quit IRC | 03:28 | |
| *** VW has joined #openstack-nova | 03:29 | |
| *** amotoki has quit IRC | 03:30 | |
| *** VW has quit IRC | 03:33 | |
| *** penick has joined #openstack-nova | 03:35 | |
| *** su_zhang has joined #openstack-nova | 03:35 | |
| *** amotoki has joined #openstack-nova | 03:41 | |
| *** yuntongjin has quit IRC | 03:43 | |
| *** baoli has quit IRC | 03:46 | |
| *** haomaiwang has quit IRC | 03:47 | |
| *** haomaiwang has joined #openstack-nova | 03:48 | |
| *** markvoelker has joined #openstack-nova | 03:50 | |
| *** amotoki has quit IRC | 03:50 | |
| *** annegentle has joined #openstack-nova | 03:52 | |
| *** markvoelker has quit IRC | 03:54 | |
| *** amotoki has joined #openstack-nova | 03:55 | |
| *** annegentle has quit IRC | 03:57 | |
| openstackgerrit | Nimish proposed openstack/nova: *Nova can incorrectly think an instance is volume backed if instance is booted with "nova boot...", it ends up with no image_ref set & an entry in the block_device_mappingstable. This confuses api.py & assume if image_ref isn't set then the instance is vo https://review.openstack.org/234165 | 03:57 |
| *** otter768 has quit IRC | 03:57 | |
| openstackgerrit | Nimish proposed openstack/nova: *Nova can incorrectly think an instance is volume backed if instance is booted with "nova boot...", it ends up with no image_ref set & an entry in the block_device_mappingstable. This confuses api.py & assume if image_ref isn't set then the instance is vo https://review.openstack.org/234165 | 03:58 |
| *** otter768 has joined #openstack-nova | 03:58 | |
| *** haomaiwang has quit IRC | 04:01 | |
| *** haomaiwang has joined #openstack-nova | 04:01 | |
| *** david-lyle has quit IRC | 04:01 | |
| alex_xu | Kevin_Zheng: gmann , I flush out few bugs for wadl doc in https://etherpad.openstack.org/p/nova-v2.1-api-doc under The wadl API ref bug | 04:03 |
| alex_xu | I probably need go through more detail the full doc again, to check if there are some more detail problem | 04:04 |
| alex_xu | feel free take the item from that etherpad and mark the your name after the item, thanks | 04:04 |
| *** Murali has joined #openstack-nova | 04:06 | |
| gmann | alex_xu: Thanks | 04:09 |
| gmann | alex_xu: just added VIF one | 04:09 |
| alex_xu | gmann: that is good one, thanks a lot! | 04:09 |
| *** shuquan has quit IRC | 04:09 | |
| alex_xu | gmann: Kevin_Zheng free feel to add item to the list if there is something I missed | 04:10 |
| *** sudipto has joined #openstack-nova | 04:10 | |
| gmann | alex_xu: I will check more, Thanks for putting all together :) | 04:10 |
| alex_xu | gmann: np :) | 04:10 |
| alex_xu | gmann: few item need change our api sample test | 04:11 |
| alex_xu | gmann: but there isn't too much, just few items | 04:11 |
| gmann | alex_xu: oh, which one | 04:12 |
| alex_xu | gmann: for example, multiple create one https://bugs.launchpad.net/openstack-api-site/+bug/1505881 | 04:12 |
| openstack | Launchpad bug 1505881 in openstack-api-site "Multiple create api for servers should use all extension enabled api sample in nova api ref" [Undecided,New] | 04:12 |
| *** seanw1 has quit IRC | 04:13 | |
| gmann | alex_xu: but those are for other place also right? where we run tests based on extension. | 04:13 |
| gmann | i mean when we have our sample tests with all extension enabled then it will be fine | 04:14 |
| *** seanw1 has joined #openstack-nova | 04:14 | |
| *** oomichi has joined #openstack-nova | 04:14 | |
| *** tongli has joined #openstack-nova | 04:14 | |
| alex_xu | gmann: I guess most of item is ok, like server start/stop, etc | 04:14 |
| alex_xu | gmann: is it possible we just change few api sample for currently? | 04:15 |
| gmann | alex_xu: yea that we can do.. | 04:16 |
| alex_xu | gmann: most of api didn't have extension, most of extension are extend the servers/flavor/image. So for fixing the wadl doc, we only need fix those related api sample test | 04:16 |
| alex_xu | gmann: and the good news is we already have all extension enabled sample test for servers and flavors https://github.com/openstack/nova/tree/master/doc/api_samples/all_extensions | 04:16 |
| gmann | alex_xu: i saw for server also we do not have tests and sample file with all extension | 04:18 |
| *** amotoki has quit IRC | 04:18 | |
| gmann | like - disk config and server extended attribute extensions tests | 04:18 |
| *** Sree has joined #openstack-nova | 04:19 | |
| *** Sree has quit IRC | 04:20 | |
| *** Murali has quit IRC | 04:20 | |
| *** Sree has joined #openstack-nova | 04:20 | |
| gmann | alex_xu: os-extended-availability-zone, os-extended-status, os-server-usage etc | 04:21 |
| alex_xu | gmann: those are in https://github.com/openstack/nova/blob/master/doc/api_samples/all_extensions/servers-details-resp.json#L54 | 04:21 |
| *** tpeoples has quit IRC | 04:21 | |
| gmann | alex_xu: ohh, so api-site takes those from all_extensions? | 04:21 |
| alex_xu | gmann: yeah | 04:22 |
| *** ssurana has joined #openstack-nova | 04:22 | |
| gmann | alex_xu: ahh got it | 04:22 |
| alex_xu | gmann: then in the api ref doc, we only have one servers entry instead of those separated items: os-extended-availability-zone, os-extended-status, os-server-usage etc | 04:22 |
| gmann | alex_xu: and which are left (as you mentioned) we just need to have all extension tests for those | 04:22 |
| gmann | alex_xu: yea | 04:22 |
| alex_xu | gmann: yea | 04:23 |
| gmann | alex_xu: so that each extension gets included in all_extension sample file and so in doc | 04:23 |
| gmann | alex_xu: cool | 04:23 |
| gmann | alex_xu: i will do those quickly | 04:23 |
| alex_xu | gmann: thanks a lot | 04:24 |
| gmann | alex_xu: thats nice plan to fix those as of now and later anyways we will have only all extension tests | 04:24 |
| alex_xu | gmann: yea, and we also have swagger, hope we won't spend too much time on the wadl | 04:24 |
| alex_xu | gmann: I think the goal is find easy and quick fix the current wadl | 04:25 |
| gmann | alex_xu: yea that will be nice. honestly saying i created my first patch on api-site the vif one and i do not like updating there | 04:25 |
| gmann | alex_xu: +1 | 04:25 |
| alex_xu | gmann: yea | 04:26 |
| alex_xu | gmann: if you work on any item, don't forget mark your name after the item | 04:26 |
| *** garyk has quit IRC | 04:26 | |
| *** Murali has joined #openstack-nova | 04:28 | |
| *** yuntongjin1 has joined #openstack-nova | 04:29 | |
| *** thorst has quit IRC | 04:29 | |
| gmann | alex_xu: yea, done | 04:30 |
| alex_xu | gmann: cool, thanks :) | 04:30 |
| *** Murali_ has joined #openstack-nova | 04:31 | |
| *** Murali has quit IRC | 04:33 | |
| *** Murali_ is now known as Murali | 04:33 | |
| *** shuquan has joined #openstack-nova | 04:35 | |
| *** deepthi has joined #openstack-nova | 04:36 | |
| *** Murali has quit IRC | 04:38 | |
| *** ijw has joined #openstack-nova | 04:40 | |
| *** Kennan2 has quit IRC | 04:40 | |
| *** Kennan has joined #openstack-nova | 04:41 | |
| *** dave-mccowan has quit IRC | 04:44 | |
| *** markvoelker has joined #openstack-nova | 04:50 | |
| *** markvoelker has quit IRC | 04:55 | |
| *** penick has quit IRC | 04:57 | |
| *** tbh has joined #openstack-nova | 04:59 | |
| *** jwcroppe has joined #openstack-nova | 05:00 | |
| *** haomaiwang has quit IRC | 05:01 | |
| *** haomaiwang has joined #openstack-nova | 05:01 | |
| *** wanghua has quit IRC | 05:02 | |
| *** pradipta has joined #openstack-nova | 05:04 | |
| *** asettle has quit IRC | 05:04 | |
| *** su_zhang has quit IRC | 05:11 | |
| *** ssurana has quit IRC | 05:13 | |
| *** su_zhang has joined #openstack-nova | 05:22 | |
| *** boris-42 has joined #openstack-nova | 05:23 | |
| *** tongli has quit IRC | 05:24 | |
| *** moshele has joined #openstack-nova | 05:25 | |
| *** VikasC has joined #openstack-nova | 05:25 | |
| *** ankit_ag has quit IRC | 05:30 | |
| *** vilobhmm11 has joined #openstack-nova | 05:34 | |
| *** breitz has quit IRC | 05:35 | |
| *** breitz has joined #openstack-nova | 05:36 | |
| *** haomaiwang has quit IRC | 05:38 | |
| *** haomaiwang has joined #openstack-nova | 05:38 | |
| *** changbl has quit IRC | 05:40 | |
| *** annegentle has joined #openstack-nova | 05:41 | |
| *** nikhil_k has joined #openstack-nova | 05:42 | |
| *** nikhil has quit IRC | 05:42 | |
| *** yuntongjin1 has quit IRC | 05:44 | |
| *** changbl has joined #openstack-nova | 05:44 | |
| *** ijw has quit IRC | 05:45 | |
| *** annegentle has quit IRC | 05:46 | |
| *** asettle has joined #openstack-nova | 05:48 | |
| *** armax has quit IRC | 05:49 | |
| *** pixelbeat has quit IRC | 05:51 | |
| *** markvoelker has joined #openstack-nova | 05:51 | |
| *** haomaiwang has quit IRC | 05:51 | |
| *** garyk has joined #openstack-nova | 05:52 | |
| *** haomaiwa_ has joined #openstack-nova | 05:52 | |
| *** yuntongjin has joined #openstack-nova | 05:53 | |
| *** pixelbeat has joined #openstack-nova | 05:54 | |
| *** markvoelker has quit IRC | 05:56 | |
| *** haomaiwa_ has quit IRC | 06:01 | |
| *** 32NAAJ35R has joined #openstack-nova | 06:02 | |
| *** VikasC_ has joined #openstack-nova | 06:02 | |
| *** VikasC has quit IRC | 06:05 | |
| *** nkrinner has joined #openstack-nova | 06:05 | |
| *** irenab has joined #openstack-nova | 06:06 | |
| *** Miouge has joined #openstack-nova | 06:07 | |
| openstackgerrit | garyk proposed openstack/nova: Config drive: convert to use instance.image_meta property https://review.openstack.org/234224 | 06:11 |
| *** unicell has joined #openstack-nova | 06:13 | |
| *** vilobhmm11 has quit IRC | 06:18 | |
| *** asettle has quit IRC | 06:19 | |
| *** Miouge has quit IRC | 06:19 | |
| *** Sree has quit IRC | 06:20 | |
| *** vilobhmm11 has joined #openstack-nova | 06:20 | |
| *** Sree has joined #openstack-nova | 06:20 | |
| *** Miouge has joined #openstack-nova | 06:20 | |
| *** asettle has joined #openstack-nova | 06:23 | |
| *** Sree has quit IRC | 06:25 | |
| *** su_zhang has quit IRC | 06:27 | |
| *** vlaza has joined #openstack-nova | 06:29 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Ensure Glance image 'size' attribute is 0, not 'None' https://review.openstack.org/215947 | 06:30 |
| *** scheuran has joined #openstack-nova | 06:31 | |
| *** lykinsbd_ has quit IRC | 06:36 | |
| *** ankit_ag has joined #openstack-nova | 06:39 | |
| *** Marga_ has quit IRC | 06:42 | |
| * alex_xu hate wadl | 06:43 | |
| *** heyongli has joined #openstack-nova | 06:45 | |
| *** vilobhmm11 has quit IRC | 06:45 | |
| *** lykinsbd_ has joined #openstack-nova | 06:46 | |
| *** moshele has quit IRC | 06:48 | |
| *** moshele has joined #openstack-nova | 06:48 | |
| *** Murali has joined #openstack-nova | 06:49 | |
| *** sahid has joined #openstack-nova | 06:50 | |
| *** leimon has joined #openstack-nova | 06:50 | |
| *** leimon has left #openstack-nova | 06:51 | |
| *** leimon has joined #openstack-nova | 06:52 | |
| *** markvoelker has joined #openstack-nova | 06:52 | |
| *** heyongli has quit IRC | 06:52 | |
| *** stevemar_ has quit IRC | 06:53 | |
| *** heyongli has joined #openstack-nova | 06:53 | |
| *** stevemar_ has joined #openstack-nova | 06:54 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Fix multiple create server tests for all extension https://review.openstack.org/234602 | 06:54 |
| *** stevemar_ has quit IRC | 06:56 | |
| *** markvoelker has quit IRC | 06:56 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Fix multiple create server tests for all extension https://review.openstack.org/234602 | 06:57 |
| *** subscope has joined #openstack-nova | 06:59 | |
| *** Savemech has quit IRC | 06:59 | |
| *** romainh has joined #openstack-nova | 07:00 | |
| *** 32NAAJ35R has quit IRC | 07:01 | |
| *** openstackgerrit has quit IRC | 07:01 | |
| *** garyk has quit IRC | 07:01 | |
| *** Savemech has joined #openstack-nova | 07:01 | |
| *** sudipto has quit IRC | 07:01 | |
| *** openstackgerrit has joined #openstack-nova | 07:02 | |
| *** haomaiwang has joined #openstack-nova | 07:02 | |
| *** heyongli has quit IRC | 07:02 | |
| *** heyongli has joined #openstack-nova | 07:03 | |
| *** Sree has joined #openstack-nova | 07:05 | |
| *** Sree has quit IRC | 07:05 | |
| *** Sree has joined #openstack-nova | 07:06 | |
| *** unicell1 has joined #openstack-nova | 07:09 | |
| *** unicell has quit IRC | 07:10 | |
| *** Marga_ has joined #openstack-nova | 07:11 | |
| *** heyongli has quit IRC | 07:13 | |
| gmann | alex_xu: this only should fix the multiple create response in doc - https://review.openstack.org/#/c/234610/ | 07:13 |
| gmann | alex_xu: do we need nova one too? - https://review.openstack.org/#/c/234602/ | 07:13 |
| *** Marga_ has quit IRC | 07:14 | |
| *** heyongli has joined #openstack-nova | 07:14 | |
| *** Marga_ has joined #openstack-nova | 07:14 | |
| gmann | alex_xu: i like to fix multiple sample tests for all extension together with other as we discussed yesterday meeting | 07:14 |
| alex_xu | gmann: is it easy for fixing them all? | 07:17 |
| gmann | alex_xu: checking one by one, if it need some extra sample file or update then we need nova side fix also | 07:18 |
| gmann | alex_xu: actually, after nova fix we need to sync those sample file on api-site also | 07:18 |
| *** leimon has left #openstack-nova | 07:18 | |
| *** leimon has joined #openstack-nova | 07:18 | |
| gmann | alex_xu: for multiple create issue directly we can fix on api-site as https://review.openstack.org/#/c/234610/ | 07:19 |
| *** rdopiera has joined #openstack-nova | 07:19 | |
| *** leseb_ has quit IRC | 07:20 | |
| alex_xu | gmann: so after nova side sample updated, we need change the path back | 07:20 |
| gmann | alex_xu: yea, even when we will run all our sample tests for all extension then there will be path changes for many sample file | 07:21 |
| gmann | alex_xu: as we will share common sample files among tests | 07:22 |
| gmann | alex_xu: but need to see when actually refactoring the tests | 07:22 |
| *** matrohon has joined #openstack-nova | 07:23 | |
| *** heyongli has quit IRC | 07:23 | |
| *** markus_z has joined #openstack-nova | 07:23 | |
| *** heyongli has joined #openstack-nova | 07:24 | |
| alex_xu | gmann: ok, that sounds long term work, we can't finish it soon, so I think https://review.openstack.org/#/c/234610/ it is ok | 07:24 |
| *** browne has quit IRC | 07:24 | |
| *** jwcroppe has quit IRC | 07:24 | |
| gmann | alex_xu: yea, once we will have our final sample files and path with all extension then we can reiterate on api-site side(or may be not needed as per your swagger things :) ) | 07:25 |
| *** abhishekk has joined #openstack-nova | 07:25 | |
| alex_xu | gmann: yea, the best thing is we not need reiterate api-site again :) | 07:26 |
| *** subscope has quit IRC | 07:26 | |
| gmann | alex_xu: yea. ll be much happy :) | 07:26 |
| *** achanda has quit IRC | 07:26 | |
| gmann | even all ll be happy :) | 07:26 |
| alex_xu | :) | 07:26 |
| *** leseb_ has joined #openstack-nova | 07:27 | |
| *** achanda has joined #openstack-nova | 07:27 | |
| *** alex_klimov has joined #openstack-nova | 07:27 | |
| *** mpavone has joined #openstack-nova | 07:28 | |
| *** liu_ has joined #openstack-nova | 07:28 | |
| *** leimon has quit IRC | 07:29 | |
| *** leimon has joined #openstack-nova | 07:29 | |
| *** leimon has quit IRC | 07:29 | |
| *** annegentle has joined #openstack-nova | 07:30 | |
| *** liu_ has quit IRC | 07:31 | |
| *** liu_ has joined #openstack-nova | 07:31 | |
| *** liu_ has quit IRC | 07:31 | |
| *** achanda has quit IRC | 07:33 | |
| *** heyongli has quit IRC | 07:33 | |
| *** heyongli has joined #openstack-nova | 07:34 | |
| *** annegentle has quit IRC | 07:35 | |
| *** ijw has joined #openstack-nova | 07:37 | |
| *** leimeon has joined #openstack-nova | 07:38 | |
| *** jlanoux has joined #openstack-nova | 07:39 | |
| *** leimon has joined #openstack-nova | 07:40 | |
| *** leimon is now known as leimond | 07:41 | |
| *** leimond is now known as leimon | 07:41 | |
| *** leimon has quit IRC | 07:42 | |
| leimeon | HexChat: 2.10.2 ** OS: Linux 4.0.4-301.fc22.x86_64 x86_64 ** Distro: Fedora release 22 (Twenty Two) ** CPU: 8 x AMD FX(tm)-8300 Eight-Core Processor (AuthenticAMD) @ 1.40GHz ** RAM: Physical: 15.6GiB, 49.1% free ** Disk: Total: 1.8TiB, 93.5% free ** VGA: NVIDIA Corporation GT218 [GeForce 210] ** Sound: HDA-Intel - HDA ATI SB1: HDA-Intel - HDA NVidia ** Ethernet: Realtek Semiconductor Co., Ltd. CIe Gigabit | 07:43 |
| leimeon | Ethernet ** Uptime: 5h 7m 50s ** | 07:43 |
| *** heyongli has quit IRC | 07:43 | |
| bauzas | good morning Nova | 07:44 |
| *** heyongli has joined #openstack-nova | 07:44 | |
| *** leimon has joined #openstack-nova | 07:46 | |
| *** Murali has quit IRC | 07:49 | |
| *** leimon has left #openstack-nova | 07:50 | |
| *** leimeon has quit IRC | 07:51 | |
| *** leimeon has joined #openstack-nova | 07:51 | |
| *** marzif has quit IRC | 07:52 | |
| *** markvoelker has joined #openstack-nova | 07:53 | |
| *** heyongli has quit IRC | 07:54 | |
| *** heyongli has joined #openstack-nova | 07:55 | |
| *** Sree_ has joined #openstack-nova | 07:57 | |
| *** david-lyle has joined #openstack-nova | 07:57 | |
| *** Sree_ is now known as Guest6340 | 07:57 | |
| *** markvoelker has quit IRC | 07:57 | |
| *** lpetrut has joined #openstack-nova | 07:58 | |
| *** jcoufal has joined #openstack-nova | 07:59 | |
| *** Sree has quit IRC | 08:00 | |
| *** haomaiwang has quit IRC | 08:01 | |
| *** Sree has joined #openstack-nova | 08:01 | |
| *** Marga_ has quit IRC | 08:01 | |
| *** ijw has quit IRC | 08:01 | |
| *** haomaiwa_ has joined #openstack-nova | 08:02 | |
| openstackgerrit | Li Yingjun proposed openstack/python-novaclient: Add sort_dir/key to flavor list https://review.openstack.org/234623 | 08:02 |
| *** ijw has joined #openstack-nova | 08:03 | |
| *** heyongli has quit IRC | 08:04 | |
| openstackgerrit | Hans Lindgren proposed openstack/nova: Remove and deprecate conductor provider_fw_rule_get_all() https://review.openstack.org/216241 | 08:04 |
| openstackgerrit | Hans Lindgren proposed openstack/nova: Remove unused provider firewall rules functionality in nova https://review.openstack.org/184027 | 08:04 |
| openstackgerrit | Hans Lindgren proposed openstack/nova: Remove provider_fw_rules tables and associated db api methods https://review.openstack.org/216242 | 08:04 |
| *** sahid has quit IRC | 08:05 | |
| *** heyongli has joined #openstack-nova | 08:05 | |
| *** sahid has joined #openstack-nova | 08:05 | |
| *** marzif has joined #openstack-nova | 08:06 | |
| *** sudipto has joined #openstack-nova | 08:07 | |
| *** ijw has quit IRC | 08:07 | |
| *** yassine has joined #openstack-nova | 08:08 | |
| *** mgoddard has joined #openstack-nova | 08:10 | |
| *** heyongli has quit IRC | 08:14 | |
| *** ociuhandu has quit IRC | 08:15 | |
| *** heyongli has joined #openstack-nova | 08:15 | |
| *** pradipta has quit IRC | 08:17 | |
| *** jistr has joined #openstack-nova | 08:18 | |
| *** danpb has joined #openstack-nova | 08:18 | |
| *** bapalm has quit IRC | 08:23 | |
| *** PaulMurray has joined #openstack-nova | 08:23 | |
| *** heyongli has quit IRC | 08:24 | |
| *** heyongli has joined #openstack-nova | 08:25 | |
| PaulMurray | johnthetubaguy, do you have a moment to talk about migration | 08:25 |
| *** bapalm has joined #openstack-nova | 08:26 | |
| *** eglynn has joined #openstack-nova | 08:27 | |
| *** zhangjn has quit IRC | 08:27 | |
| *** ndipanov has joined #openstack-nova | 08:27 | |
| *** zhangjn has joined #openstack-nova | 08:30 | |
| *** wznoinsk has joined #openstack-nova | 08:30 | |
| wznoinsk | hi all | 08:30 |
| wznoinsk | johnthetubaguy, jaypipes let me know when either of you is around, I wanted to have a 2mins chat about intel nfv ci | 08:31 |
| *** sudipto has quit IRC | 08:34 | |
| *** heyongli has quit IRC | 08:35 | |
| johnthetubaguy | PaulMurray: hey, I made a note to reach out to you about the live-migrate stuff | 08:35 |
| johnthetubaguy | PaulMurray: I saw the etherpad of things | 08:35 |
| PaulMurray | johnthetubaguy, I was away last week and missed the meeting - sorry | 08:35 |
| johnthetubaguy | PaulMurray: I am wondering about the best way to make use of summit time on that | 08:35 |
| johnthetubaguy | PaulMurray: no worries | 08:35 |
| *** heyongli has joined #openstack-nova | 08:36 | |
| *** Amichayp has joined #openstack-nova | 08:36 | |
| PaulMurray | same here - there are a couple of things that seem to be quite clear, and a couple that are less so | 08:36 |
| PaulMurray | The clear ones are migrating instances with a mix of volumes, ephemeral and config drive | 08:37 |
| *** rotbeard has joined #openstack-nova | 08:37 | |
| PaulMurray | So not much to talk about there - there is a patch already being tried | 08:37 |
| johnthetubaguy | PaulMurray: so you can totally have time on the friday, I am just wondering if we need a session for whats left | 08:37 |
| johnthetubaguy | PaulMurray: I think its worth creating a subteam in the etherpad for this live-migrate stuff, to help focus people on the right patches | 08:38 |
| *** marzif has quit IRC | 08:38 | |
| PaulMurray | That's what I was thinking - I'm not sure there is need for a lot of in depth discussion, but there is a need for coordination | 08:38 |
| *** marzif has joined #openstack-nova | 08:39 | |
| PaulMurray | Also its not so clear what can be done this cycle vs taking longer | 08:39 |
| *** Sree has quit IRC | 08:39 | |
| PaulMurray | What do you suggest? | 08:39 |
| *** Sree has joined #openstack-nova | 08:39 | |
| *** garyk has joined #openstack-nova | 08:40 | |
| johnthetubaguy | honestly, I think its worth assuming infinite time and resources, just to make a list with all the dependencies | 08:40 |
| johnthetubaguy | we can work out whats blocked by what, etc | 08:40 |
| johnthetubaguy | then its a case of who does what, and when, I guess | 08:40 |
| *** jamielennox is now known as jamielennox|away | 08:40 | |
| johnthetubaguy | if we could review that at the summit, on the friday, I think that would help move things along | 08:40 |
| *** lucasagomes has joined #openstack-nova | 08:41 | |
| PaulMurray | ok - I thought what you had put down in the etherpad looked good to me - just remember that the friday was a long list of stuff last time | 08:41 |
| johnthetubaguy | wznoinsk: quite note, I noticed your message, just PaulMurray got in the queue first, and my multi-tasking is terrible, be with you in a bit | 08:41 |
| johnthetubaguy | PaulMurray: yeah, agreed, that comes up the top of the list, I feel | 08:42 |
| *** lucasagomes_ has joined #openstack-nova | 08:42 | |
| PaulMurray | I'll try to structure as best I can with details about what we are trying to do | 08:42 |
| johnthetubaguy | PaulMurray: so, there was some stuff that didn't have a clear plan, what are those bits? | 08:42 |
| PaulMurray | The least detail is in CI - so might need some discussion there | 08:42 |
| PaulMurray | Also there is a little disagreement / discussion around what you should be able to do with an instance while it is being migrated | 08:43 |
| johnthetubaguy | so the CI is already in place, just needs hardening, as I understand it? | 08:44 |
| johnthetubaguy | as in the multi-host test | 08:44 |
| johnthetubaguy | in terms of parallel operations, right now we allow nothing, as I understand it, I vote we get it working before we start worrying about power operations during a live-migrate | 08:45 |
| *** heyongli has quit IRC | 08:45 | |
| *** apuimedo_ has joined #openstack-nova | 08:45 | |
| johnthetubaguy | you should always be able to delete, other than that, I mean | 08:45 |
| *** heyongli has joined #openstack-nova | 08:46 | |
| PaulMurray | there are simple things like getting at vnc and console logs | 08:46 |
| *** lucasagomes has quit IRC | 08:46 | |
| *** sudipto has joined #openstack-nova | 08:46 | |
| johnthetubaguy | ah, true, that, in theory should work all the time, as its not something that changes the VM state | 08:46 |
| johnthetubaguy | now thats probably broken, but yeah, those things should get fixed | 08:47 |
| johnthetubaguy | PaulMurray: so fun idea, maybe we start with updating the API concept guide, to clearly state (and agree) our intentions | 08:47 |
| PaulMurray | I think if the top few items in the "near term" list I made get done we will be in good shape | 08:47 |
| johnthetubaguy | to be honest, getting the CI working is the main thing that worries me | 08:48 |
| johnthetubaguy | we keep fixing it, and promptly breaking it | 08:49 |
| *** Sree_ has joined #openstack-nova | 08:49 | |
| *** Sree has quit IRC | 08:49 | |
| *** Sree_ is now known as Guest75569 | 08:49 | |
| johnthetubaguy | once we have that safety net in place, we can make progress, I feel | 08:49 |
| johnthetubaguy | PaulMurray: do we still have someone pushing on fixing the multi-host job? | 08:50 |
| PaulMurray | So a good target for the summit would be to agree aims for migration and sign people up to near toerm work | 08:50 |
| *** derekh has joined #openstack-nova | 08:50 | |
| openstackgerrit | gongysh proposed openstack/nova: Support keystone V3 for neutron access https://review.openstack.org/234642 | 08:50 |
| johnthetubaguy | PaulMurray: yes, shall we very quickly review that list, let me bring up the etherpad | 08:51 |
| johnthetubaguy | https://etherpad.openstack.org/p/mitaka-live-migration is the one right? | 08:51 |
| PaulMurray | yep | 08:51 |
| *** ZZelle has quit IRC | 08:53 | |
| *** ZZelle has joined #openstack-nova | 08:53 | |
| *** markvoelker has joined #openstack-nova | 08:53 | |
| johnthetubaguy | PaulMurray: I remember seeing a spec for auto-converge, trying to think where that is at | 08:54 |
| *** subscope has joined #openstack-nova | 08:54 | |
| *** zhangjn has quit IRC | 08:54 | |
| PaulMurray | there are a few auto-converge type things - some in qemu rather than nova | 08:55 |
| PaulMurray | what are you thinking of? | 08:55 |
| *** stevemar_ has joined #openstack-nova | 08:55 | |
| *** heyongli has quit IRC | 08:55 | |
| johnthetubaguy | I think its the qemu one, thinking about it, maybe its the same as pause | 08:55 |
| johnthetubaguy | although I think it was sync copying memory for a period, rather than a full pause | 08:56 |
| johnthetubaguy | like all pages dirty, or something like that | 08:56 |
| *** heyongli has joined #openstack-nova | 08:56 | |
| *** zhangjn has joined #openstack-nova | 08:56 | |
| PaulMurray | there is something in qemu about slowing the vm and/or pausing it to guarantee completion | 08:56 |
| PaulMurray | I haven't listed things that are outside of nova | 08:57 |
| johnthetubaguy | yeah, its that stuff | 08:57 |
| johnthetubaguy | just thinking about us adding support for it, thats all really | 08:57 |
| PaulMurray | I think it is in the longer term category because its not impplemented yet | 08:57 |
| PaulMurray | that is what the bullet there is intended to capture | 08:58 |
| *** lucasagomes has joined #openstack-nova | 08:58 | |
| johnthetubaguy | ah, OK, I got the impression it was merged, but maybe that was me making assuptions | 08:58 |
| johnthetubaguy | yeah, thats in the long term, so I think its covered | 08:58 |
| *** stevemar_ has quit IRC | 08:58 | |
| *** markvoelker has quit IRC | 08:58 | |
| *** deepakcs has joined #openstack-nova | 08:58 | |
| PaulMurray | I would have to check - not entirely sure | 08:58 |
| *** rajesht has quit IRC | 08:59 | |
| *** ankit_ag has quit IRC | 08:59 | |
| *** edmondsw has joined #openstack-nova | 08:59 | |
| *** abhishekk has quit IRC | 08:59 | |
| *** rajesht has joined #openstack-nova | 09:00 | |
| *** ankit_ag has joined #openstack-nova | 09:00 | |
| *** abhishekk has joined #openstack-nova | 09:00 | |
| johnthetubaguy | PaulMurray: so I went through that near term list, I re-ordered it a little, does that look OK? | 09:01 |
| *** haomaiwa_ has quit IRC | 09:01 | |
| *** Murali has joined #openstack-nova | 09:01 | |
| *** subscope has quit IRC | 09:01 | |
| *** subscope has joined #openstack-nova | 09:01 | |
| PaulMurray | Might be good to have cancel migration higher - mabe bundle in with pause bullet | 09:02 |
| johnthetubaguy | wznoinsk: hi, did you want to catch up around the intel nfv ci? | 09:02 |
| johnthetubaguy | PaulMurray: yeah, could we not just focus on one of those, or does that not cut it? | 09:03 |
| *** ijw has joined #openstack-nova | 09:03 | |
| PaulMurray | lets dig in a see if there is a lot of work involved - could turn out to be easy to do both together | 09:04 |
| johnthetubaguy | PaulMurray: OK, makes sense | 09:04 |
| PaulMurray | Its a lot of work - I doubt it will all get there | 09:05 |
| johnthetubaguy | PaulMurray: I guess this might be guest dependent thinking about it, cancel would stop the need to pause | 09:05 |
| johnthetubaguy | PaulMurray: agreed, it seems unlikely but lets see who steps up, I guess | 09:05 |
| PaulMurray | yes - I think we have a plan now | 09:05 |
| *** heyongli has quit IRC | 09:05 | |
| PaulMurray | thanks | 09:06 |
| johnthetubaguy | PaulMurray: yeah, certainly worth reviewing this at the summit, but yeah, a full session seems overkill given how much agreement we already have here | 09:06 |
| johnthetubaguy | PaulMurray: are you good setting up a subteam in the review etherpad, just so you all see each others work quickly? | 09:06 |
| *** heyongli has joined #openstack-nova | 09:06 | |
| johnthetubaguy | PaulMurray: I am thinking about in this one: https://etherpad.openstack.org/p/mitaka-nova-priorities-tracking | 09:07 |
| *** alexschm has joined #openstack-nova | 09:07 | |
| PaulMurray | yes, will do | 09:07 |
| bauzas | johnthetubaguy: had you time to discuss with jaypipes about the sched session ? | 09:07 |
| *** lucasagomes has quit IRC | 09:07 | |
| *** lucasagomes_ is now known as lucasagomes | 09:08 | |
| *** jichen has joined #openstack-nova | 09:08 | |
| johnthetubaguy | bauzas: not been online the same time as him recently, I am afraid | 09:08 |
| johnthetubaguy | or not both been free at the same time | 09:08 |
| *** ijw has quit IRC | 09:09 | |
| bauzas | johnthetubaguy: okay, I just want to make sure you're not blocked | 09:09 |
| markus_z | bauzas: I have a (maybe stupid) scheduler question which you might be able to answer, do you have a minute? | 09:09 |
| bauzas | markus_z: there is no stupid question | 09:10 |
| markus_z | bauzas: We'll see :) | 09:10 |
| markus_z | bauzas: Are there events in Nova which might cause the scheduler to migrate an instance, which are NOT driven by the REST API? | 09:10 |
| johnthetubaguy | bauzas: yeah, I am a little blocked on that right now, need to work out what goes in that scheduler session, thats concrete, or cancel it | 09:11 |
| *** aix has joined #openstack-nova | 09:11 | |
| bauzas | markus_z: retries | 09:11 |
| bauzas | markus_z: when something goes bad on the compute side, then it fires a reschedule by calling again the conductor api | 09:12 |
| johnthetubaguy | markus_z: we are resisting adding them | 09:12 |
| bauzas | markus_z: apart from that, no, because we're cloud | 09:12 |
| *** wolsen has quit IRC | 09:12 | |
| johnthetubaguy | never saw retries like that, but I guess thats kinda true | 09:13 |
| bauzas | markus_z: if something needs to orchestrate a migration, it has to be done elsewhere | 09:13 |
| *** yuntongjin has quit IRC | 09:13 | |
| bauzas | johnthetubaguy: well, catching an exception fires a RPC cast from the compute to the conductor, right ? :) | 09:13 |
| bauzas | although I have to check for migrations tho :) | 09:14 |
| wznoinsk | johnthetubaguy: hi, thanks for comin gback | 09:14 |
| markus_z | bauzas: johnthetubaguy: OK, I thought so too, I wasn't sure anymore after I stumbled on https://blueprints.launchpad.net/nova/+spec/utilization-aware-scheduling | 09:14 |
| johnthetubaguy | bauzas: yeah, its more a build again, in my head, still from the initial API call, but yeah, works both ways | 09:14 |
| wznoinsk | johnthetubaguy: according to our last discussion to run standard set of tests with our featured flavors, please check https://review.openstack.org/#/c/229279/ - comments from last night | 09:15 |
| bauzas | johnthetubaguy: agreed, I have to doublecheck re: migrations | 09:15 |
| johnthetubaguy | markus_z: yeah, we would just provide them with APIs to do what they want, I feel | 09:15 |
| wznoinsk | johnthetubaguy: happy to talk to you whenever you have a free(er) moment | 09:15 |
| johnthetubaguy | wznoinsk: I will take a look | 09:15 |
| markus_z | johnthetubaguy: right, "build again" but traceable to one initial REST API call. | 09:16 |
| *** heyongli has quit IRC | 09:16 | |
| johnthetubaguy | wznoinsk: ah, I see lots of tests there, which one should I look at? | 09:16 |
| bauzas | markus_z: I remember some OpenStack POC about that kind of utilisation-aware schedululer | 09:16 |
| wznoinsk | johnthetubaguy: I think we discussed to run as much as possible ;-) | 09:16 |
| wznoinsk | johnthetubaguy: we have 3 features: hugepages, cpupinning, numatopology | 09:17 |
| bauzas | markus_z: but like I said, it should be done by consuming the Nova APIs, not modifying those or orchestrating internally :) | 09:17 |
| *** heyongli has joined #openstack-nova | 09:17 | |
| *** irenab has quit IRC | 09:17 | |
| markus_z | bauzas: johnthetubaguy: OK, thanks, my doubts are gone, I fully agree with your view here. | 09:17 |
| *** dims_ has joined #openstack-nova | 09:17 | |
| wznoinsk | johnthetubaguy: I run 2 types (1. all from tempest/api/compute, 2. all from tempest/scenario) on each types = 6 tests + 1 our custom hugepages/numa/cpupinning tests | 09:17 |
| *** ashishjain has joined #openstack-nova | 09:17 | |
| bauzas | markus_z: FWIW, https://wiki.openstack.org/wiki/Watcher was what I've been heard | 09:17 |
| ashishjain | Hello | 09:18 |
| *** tbh has quit IRC | 09:18 | |
| johnthetubaguy | wznoinsk: this one looks good: http://intel-openstack-ci-logs.ovh/compute-ci/refs/changes/52/215252/7/api-compute-on-cpupinning-flavors/tempest/api-compute-on-cpupinning-flavors/testr_results.html | 09:18 |
| ashishjain | I am using kilo and trying to spin off vms | 09:18 |
| *** irenab has joined #openstack-nova | 09:18 | |
| johnthetubaguy | wznoinsk: although I don't see any of the resize tests being run | 09:18 |
| *** tbh has joined #openstack-nova | 09:18 | |
| *** annegentle has joined #openstack-nova | 09:18 | |
| ashishjain | I see an issue where it says "Unable to mount image /var/lib/nova/instances/5cb913f2-bdec-48c7-9507-741ac4250e41/disk with error libguestfs installed but not usable (cannot find any suitable libguestfs supermin, fixed or old-style appliance on LIBGUESTFS_PATH (search path: /usr/lib/guestfs)). Cannot resize" | 09:19 |
| ashishjain | Do I need to install libguestfs on nova compute node? | 09:19 |
| *** marzif has quit IRC | 09:19 | |
| wznoinsk | johnthetubaguy: I had to disable cinder because there were some issues (mainly with some volume resizing or shutting down servers with volumes attached etc.) I can work on later, there are still only 1 or 2 failing in current tests I need to investigate | 09:19 |
| ashishjain | I see a bug was fixed just before the kilo release https://bugs.launchpad.net/openstack-ansible/+bug/1428639 | 09:19 |
| openstack | Launchpad bug 1428639 in openstack-ansible "Nova requires python-libguestfs in Kilo" [Medium,Fix released] - Assigned to Kevin Carter (kevin-carter) | 09:19 |
| *** njirap has joined #openstack-nova | 09:19 | |
| openstackgerrit | Sergey Nikitin proposed openstack/nova: Added server tags support in nova-api https://review.openstack.org/128940 | 09:19 |
| *** marzif has joined #openstack-nova | 09:19 | |
| wznoinsk | johnthetubaguy: I actually assumed these will be in scenario, looking | 09:20 |
| *** dims has quit IRC | 09:20 | |
| johnthetubaguy | wznoinsk: yeah, I see the tempest conf enables them, it might be me being a bit blind | 09:20 |
| johnthetubaguy | ashishjain: I think libguestfs is required, but I am not an expert on the libvirt driver, it might depend on exactly what features you use though | 09:21 |
| johnthetubaguy | ashishjain: but you should get better answers in #openstack where there are more users about who are better at answering usage questions (please see topic) | 09:22 |
| *** links has quit IRC | 09:22 | |
| ashishjain | johnthetubaguy: thanks I will ask those guys | 09:22 |
| wznoinsk | johnthetubaguy: expand any of the tempest.api.compute.servers.test_server_actions.ServerActionsTestJSON | 09:22 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: hacking check for contextlib.nested for py34 support https://review.openstack.org/228688 | 09:23 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: Enable options for oslo.reports https://review.openstack.org/216917 | 09:23 |
| bauzas | johnthetubaguy: interesting... we do reschedule on build and resize, not on migrations | 09:23 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: Avoid intermediate Token auth plugin https://review.openstack.org/229368 | 09:23 |
| *** annegentle has quit IRC | 09:23 | |
| bauzas | markus_z: nevermind my comment about retries if it was about migrations, there is no retry in place for them | 09:23 |
| wznoinsk | johnthetubaguy: or hit 'All' at the top of the page and ctrl+f for resize | 09:24 |
| johnthetubaguy | bauzas: I don't think we reschedule on anything but builds right now | 09:24 |
| bauzas | johnthetubaguy: we do on resizes | 09:24 |
| johnthetubaguy | bauzas: thats news to me | 09:24 |
| bauzas | johnthetubaguy: hence my ping :) | 09:24 |
| bauzas | johnthetubaguy: I remembered a _reschedule() helper | 09:24 |
| markus_z | bauzas: no worries, I think I got the idea | 09:24 |
| openstackgerrit | Bartosz Fic proposed openstack/nova: Live migration fails when the instance has a config_drive_format=iso9660 https://review.openstack.org/234659 | 09:24 |
| bauzas | johnthetubaguy: but it's only called by prep_resize | 09:25 |
| johnthetubaguy | wznoinsk: ah, yeah, click all, oops! | 09:25 |
| bauzas | johnthetubaguy: FYI https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L1484-L1515 | 09:25 |
| bauzas | johnthetubaguy: which is called in https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L3730 | 09:26 |
| *** heyongli has quit IRC | 09:26 | |
| johnthetubaguy | wznoinsk: so, after looking at all the tests results :) this is totally what I was hoping to see | 09:27 |
| *** heyongli has joined #openstack-nova | 09:27 | |
| wznoinsk | johnthetubaguy: my questions to you and jaypipes(ping) are then: 1. is there anything missing from our tests you want me to include, 2. if no in Q1 then given the 1 or 2 failed tests will be resolved can we start commenting back to gerrit | 09:27 |
| wznoinsk | johnthetubaguy: happy to hear that | 09:27 |
| johnthetubaguy | wznoinsk: yeah, I like how you have tests you need to skip, we just need to publish that list, and work through trying to empty out that list | 09:27 |
| wznoinsk | johnthetubaguy: the only issue with that suite of tests now is it's taking long time to finish all 7, 2h+ in total, I'm adding another server to it today which will be 1-1.5h (still a lot given ~120patchsets in Nova daily) | 09:29 |
| johnthetubaguy | wznoinsk: yeah, its a heavy load | 09:29 |
| wznoinsk | johnthetubaguy: by publishing the skipped tests you mean on 3rdparty ci WIKI or on the webhosting with logs itself? (I'm thinking of an extract of the skipped tests from tempest.log to a single file if testr_results.html is not enough to show them) | 09:30 |
| *** haomaiwang has joined #openstack-nova | 09:30 | |
| johnthetubaguy | wznoinsk: ideally just with the logs, normally its just a list of tests that are skipped/excluded or something like that | 09:30 |
| wznoinsk | johnthetubaguy: if you mean something like an extract of all 'skip' tests from the results (like http://intel-openstack-ci-logs.ovh/compute-ci/refs/changes/79/202379/53/api-compute-on-hugepages-flavors/tempest/api-compute-on-hugepages-flavors/testr_results.html) consider it done | 09:32 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/234673 | 09:32 |
| johnthetubaguy | wznoinsk: just digging for an example, one sec | 09:33 |
| *** changbl has quit IRC | 09:35 | |
| *** lpetrut has quit IRC | 09:36 | |
| *** heyongli has quit IRC | 09:36 | |
| *** alex_klimov has quit IRC | 09:36 | |
| wznoinsk | johnthetubaguy: I think I saw 'tabs' in tempest results html somewhere, you could filter on the skipped tests then, looking | 09:37 |
| *** alex_klimov has joined #openstack-nova | 09:37 | |
| *** heyongli has joined #openstack-nova | 09:37 | |
| johnthetubaguy | wznoinsk: so most CIs report their tempest_exclusion_list file | 09:37 |
| johnthetubaguy | wznoinsk: at least I thought they did, just trying to find it now | 09:37 |
| BobBall | xenserver CI does report it | 09:38 |
| johnthetubaguy | BobBall: I can't find it in the results, what is it called? | 09:38 |
| johnthetubaguy | BobBall: I am just looking at: http://dd6b71949550285df7dc-dda4e480e005aaa13ec303551d2d8155.r49.cf1.rackcdn.com/40/128940/80/12528//logs/index.html | 09:38 |
| BobBall | tempest_exclusion_list... ... ... don't tell me it's gone missing | 09:38 |
| BobBall | grrr | 09:39 |
| BobBall | :) | 09:39 |
| *** subscope has quit IRC | 09:39 | |
| johnthetubaguy | a little bit happy its not just me going mad | 09:39 |
| BobBall | It's not there even vaguely. I'll look at getting that sorted :) | 09:39 |
| *** Miouge has quit IRC | 09:39 | |
| *** lpetrut has joined #openstack-nova | 09:40 | |
| johnthetubaguy | wznoinsk: here is an example from hyper-v: http://64.119.130.115/nova/128940/80/devstack_config/excluded-tests.txt | 09:40 |
| johnthetubaguy | wznoinsk: thats the kind of thing I am thinking | 09:40 |
| wznoinsk | BobBall: would you have a sample of the file somewhere? johnthetubaguy would I follow that standard or once BobBall has to fix it anyway should we improve (how would you like to see that exclusion?) | 09:40 |
| *** ihrachys has joined #openstack-nova | 09:41 | |
| johnthetubaguy | wznoinsk: its basically the list of tests you give to tempest to do the skips | 09:41 |
| BobBall | wznoinsk: ours isn't standard... it's very ugly. | 09:41 |
| ihrachys | tristanC: ping re https://review.openstack.org/#/c/209627/ | 09:41 |
| *** Miouge has joined #openstack-nova | 09:41 | |
| ihrachys | tristanC: I suspect the patch to break lbaas gate for neutron | 09:41 |
| johnthetubaguy | wznoinsk: the cells one is quite messy, and is here: https://github.com/openstack/nova/blob/master/devstack/tempest-dsvm-cells-rc | 09:42 |
| BobBall | wznoinsk: We patch tempest to support an exclusion list https://github.com/stackforge/xenapi-os-testing/blob/master/run_tests.sh#L146 and this is the list https://github.com/stackforge/xenapi-os-testing/blob/master/tempest_exclusion_list | 09:42 |
| wznoinsk | johnthetubaguy, BobBall thanks | 09:45 |
| johnthetubaguy | BobBall: do you know how hyper-v make their list work? | 09:45 |
| johnthetubaguy | BobBall: wondering if there is upstream support for a excluded-tests list now? | 09:45 |
| wznoinsk | johnthetubaguy: I'll work on the separate exclusion list then, in the meantime I'll skip the failing 1 or 2 and would like to start commenting to gerrit, I think majority of the work is done | 09:45 |
| johnthetubaguy | wznoinsk: so I am wondering, could we turn all these features on in a single flavor, and just run that one test? | 09:46 |
| johnthetubaguy | wznoinsk: so hugepages, plus pinning, plus SRIOV? | 09:46 |
| *** moshele has quit IRC | 09:46 | |
| wznoinsk | johnthetubaguy: that's what I'm going to find out too now | 09:46 |
| *** heyongli has quit IRC | 09:46 | |
| johnthetubaguy | wznoinsk: :) | 09:46 |
| johnthetubaguy | wznoinsk: I think thats a good way forward here, try to get a single tempest run that covers all those features | 09:47 |
| openstackgerrit | Tang Chen proposed openstack/nova: Add a Nova API to trigger crash dump in an instance. https://review.openstack.org/202617 | 09:47 |
| openstackgerrit | Tang Chen proposed openstack/nova: Add a compute RPC API to trigger crash dump in instance. https://review.openstack.org/202605 | 09:47 |
| openstackgerrit | Tang Chen proposed openstack/nova: Improve inject_nmi() in libvirt driver and add tests. https://review.openstack.org/208794 | 09:47 |
| openstackgerrit | Tang Chen proposed openstack/nova: Add a cell RPC API to trigger crash dump in instance. https://review.openstack.org/234691 | 09:47 |
| *** Kennan has quit IRC | 09:47 | |
| *** heyongli has joined #openstack-nova | 09:47 | |
| * wznoinsk 's Chrome lost icons on the tabs (getting thru too many open items) | 09:47 | |
| *** Kennan has joined #openstack-nova | 09:48 | |
| johnthetubaguy | wznoinsk: after we have all those running, why do we need those scenario tests? are they now redundant? | 09:48 |
| openstackgerrit | Tomi Juvonen proposed openstack/nova-specs: Get valid server state https://review.openstack.org/192246 | 09:50 |
| *** haomaiwang has quit IRC | 09:50 | |
| johnthetubaguy | wznoinsk: for the ones that are not covered, I just wonder if we can get them upstream | 09:50 |
| *** haomaiwa_ has joined #openstack-nova | 09:51 | |
| wznoinsk | johnthetubaguy: I have to get back to the originator of these tests and we'll compare, atm no harm in running them I think - takes 3mins to finish | 09:52 |
| sudipto | ndipanov, Hi, one quick question: https://github.com/openstack/nova/blob/master/nova/virt/hardware.py#L697 is there a reason we count the siblings to determine the number of hyperthreads instead of relying on the libvirt capabilities - topology section? | 09:53 |
| johnthetubaguy | wznoinsk: yeah, OK, just thinking long term really, thanks for getting in touch, and getting all the regular tempest tests running. | 09:53 |
| openstackgerrit | Ihar Hrachyshka proposed openstack/nova: Revert "virt: Use preexec_fn to ulimit qemu-img info call" https://review.openstack.org/234696 | 09:54 |
| *** markvoelker has joined #openstack-nova | 09:54 | |
| sudipto | wondering, if it's to account for offline threads... | 09:55 |
| *** moshele has joined #openstack-nova | 09:55 | |
| wznoinsk | johnthetubaguy: happy to see it working finally (with only minor things todo), thanks for the time on checking it, I'll talk to you later on findings re flavor with all our nfv features tests and tests exclusions | 09:56 |
| ndipanov | sudipto, I am not sure I understand the question | 09:56 |
| johnthetubaguy | wznoinsk: :) | 09:57 |
| *** heyongli has quit IRC | 09:57 | |
| sudipto | ndipanov, ok - basically we seem to be counting the number of siblings to figure out the number of threads in a given core per the code i cited above. | 09:57 |
| *** kumar has joined #openstack-nova | 09:57 | |
| ndipanov | so there is a bug in that code | 09:57 |
| sudipto | ndipanov, but libvirt also provides something like <topology sockets='1' cores='2' threads='2'/> - which suggests that there are 2 threads in each c ore... | 09:58 |
| ndipanov | sudipto, see these 3 fixes https://review.openstack.org/#/c/229575/ | 09:58 |
| sudipto | ndipanov, ok... | 09:58 |
| *** heyongli has joined #openstack-nova | 09:58 | |
| ndipanov | sudipto, but the idea is to expose the fact that it is running on hyperthreads to the guest OS | 09:58 |
| *** markvoelker has quit IRC | 09:58 | |
| ndipanov | it is not really necessary to use that whole topology structure | 09:59 |
| ndipanov | we could have just added "thread" to the NUMA cell object | 09:59 |
| sudipto | ndipanov, exactly what i thought... | 09:59 |
| ndipanov | so if we pin vcpus to 2 cores with 2 sibling threads | 09:59 |
| ndipanov | we expose that in the topology element in libvirt so that the guest_os is aware of that | 10:00 |
| ndipanov | sudipto, due to the cpu pinning logic which considers only single numa cells | 10:00 |
| *** haomaiwa_ has quit IRC | 10:01 | |
| ndipanov | we do this per cell | 10:01 |
| ndipanov | and then find the "least bad" value when spawning the instance | 10:01 |
| *** haomaiwang has joined #openstack-nova | 10:01 | |
| sudipto | ndipanov, hmm that's fair. I was thinking more specifically to why there's a need to count the number of siblings to determine the hyperthreading... | 10:01 |
| sudipto | ndipanov, it should be more straight forward to determine that via the libvirt capabilities (i thought) | 10:02 |
| ndipanov | sudipto, I don't understand what you mean by count | 10:03 |
| *** ijw has joined #openstack-nova | 10:04 | |
| ndipanov | sudipto, I think what you are missing is that we have to consider exact placement when pinning | 10:05 |
| *** shuquan has quit IRC | 10:06 | |
| ndipanov | so we may end up with 2 vcpus pinned to 2 hyperthreads that are not siblings of eachother | 10:06 |
| sudipto | ndipanov, ok - https://github.com/openstack/nova/blob/master/nova/virt/hardware.py#L702 cores_per_sib is probably something like [1, 2] - and hence threads per core is len([1,2]) which is 2. Instead of deriving it like this, why can't be get the value directly of the libvirt topology - was my uestion | 10:06 |
| sudipto | ndipanov, ah yeah that makes sense. | 10:06 |
| ndipanov | in that case - we don't want the guest to know that there are threads | 10:06 |
| ndipanov | since from the guest POV they are not | 10:07 |
| ndipanov | they are just slower CPUs lol | 10:07 |
| sudipto | ndipanov, got it... | 10:07 |
| *** heyongli has quit IRC | 10:07 | |
| sudipto | yeah :D now i get the reason for accounting the siblings... | 10:07 |
| ndipanov | cool | 10:07 |
| ndipanov | that code could use more comments for sure - that's what mdbooth was complaining about actually | 10:08 |
| *** heyongli has joined #openstack-nova | 10:08 | |
| ndipanov | will do it this week I hope | 10:08 |
| *** PaulMurray has quit IRC | 10:09 | |
| *** Alexus has joined #openstack-nova | 10:09 | |
| *** ijw has quit IRC | 10:09 | |
| sudipto | ndipanov, sure...i also had a different question around this... | 10:09 |
| sudipto | ndipanov, if you remember, i had spoken about a ppc related problem where not all threads of the core are made visible. | 10:10 |
| sudipto | ndipanov, here's another kvm patch that sort of makes it flexible from a KVM per say... https://www.mail-archive.com/kvm@vger.kernel.org/msg115978.html | 10:10 |
| sudipto | ndipanov, this is a ppc specific patch. | 10:10 |
| *** haomaiwang has quit IRC | 10:11 | |
| *** romainh has quit IRC | 10:11 | |
| sudipto | ndipanov, the typical problem now is - in ppc there are 8 threads per core in a whole core mode. But only the primary thread - that is thread0 is made online and the rest of the threads are kept offline. KVM has this logic (via this patch) to wake up threads dynamically. | 10:12 |
| sudipto | so if we have a situation where a VM has been requested with 8 VCPUs, the current pinning logic in openstack would look for the number of hyperthreads via the siblings...and would end up selecting 8 different cores...(even though each core has 7 sleeping threads).. | 10:12 |
| ndipanov | sudipto, yeah - we spoke about this before and it seems to me that we need libvirt to know about this | 10:13 |
| *** zhenguo has quit IRC | 10:13 | |
| sudipto | ndipanov, libvirt can let us know about the number of hyperthreads in a core via the topology information - but it won't be able to update the siblings section against each core - because those threads are offline... | 10:14 |
| sudipto | ndipanov, in a way - the user can't really pin to those offline threads either...but if we have a logic that can determine this condition - we can probably do better packing. | 10:15 |
| sudipto | for example - 2 vcpus pinning request comes in - and we decide it to pin it to core0, next request comes in(for 2 vcpus) we know that core0 has 6 threads still left, so we again pin to core0. | 10:16 |
| *** heyongli has quit IRC | 10:17 | |
| *** heyongli has joined #openstack-nova | 10:18 | |
| ndipanov | sudipto, I get what you are saying - we just can't make that assumption in Nova | 10:18 |
| ndipanov | it's not the right place to do it | 10:18 |
| ndipanov | we need libvirt to tell us what we can and cannot do | 10:18 |
| *** Murali_ has joined #openstack-nova | 10:18 | |
| ndipanov | so even though the code is in kvm | 10:18 |
| ndipanov | we really need it to be exposed via libvirt in some way | 10:18 |
| *** oomichi has quit IRC | 10:19 | |
| ndipanov | otherwise we have no hope of getting itright in the general case | 10:19 |
| sudipto | ndipanov, you mean libvirt should be able to say - how many threads are used inside a core? | 10:19 |
| johnthetubaguy | bauzas: how is your tasks spec coming along? | 10:19 |
| ndipanov | yes - if it's ppc specific - maybe it makes sense to update the libvirt xml with relevant information | 10:20 |
| sudipto | ndipanov, hmm got it... | 10:20 |
| *** marzif has quit IRC | 10:20 | |
| bauzas | johnthetubaguy: I took some notes in there https://review.openstack.org/#/c/221280/ | 10:20 |
| ndipanov | Nova assuming that kvm maybe does something on guest boot when scheduling is just not something we want to do | 10:20 |
| *** marzif has joined #openstack-nova | 10:20 | |
| bauzas | johnthetubaguy: the idea is that we could agree on the items and create BPs accordingly | 10:21 |
| ndipanov | nova codes to it's data model, which is in turn based on the libvirt xml contract | 10:21 |
| sudipto | ndipanov, hmm ok.. | 10:21 |
| *** Murali has quit IRC | 10:21 | |
| *** Murali_ is now known as Murali | 10:21 | |
| sudipto | ndipanov, i believe libvirt today can't alter that capabilities dynamically (i may be wrong - i will go back and check) | 10:21 |
| *** kumar has quit IRC | 10:22 | |
| *** subscope has joined #openstack-nova | 10:22 | |
| bauzas | johnthetubaguy: I'd be focusing on adding more instance actions whenever it's possible and resubmit the Tasks BP as actions for Mitaka | 10:22 |
| bauzas | the "fix reporting of error handling" is more bugfixes | 10:23 |
| johnthetubaguy | bauzas: yeah, I think we want to make that a low hanging fruit task, once we are happy with the testing we have in place, etc | 10:23 |
| ndipanov | well maybe it does not need to do it dynamically, just expose different values from the start | 10:23 |
| bauzas | johnthetubaguy: right, hence me waiting to make sure we're all going to the same direction, then I could fire a couple of bugs and items for newcomers | 10:23 |
| sudipto | ndipanov, you mean different values for each core - telling how many threads are there for each core? (sorry about dragging this) | 10:23 |
| bauzas | johnthetubaguy: tbh, I can't boil the ocean by my own, I'd be more interested in seeing if people could be interested in helping me | 10:24 |
| bauzas | johnthetubaguy: hence me writing that, then decoupling into actionable items | 10:24 |
| bauzas | that could, or could not, be low-hanging-fruits | 10:24 |
| *** wolsen has joined #openstack-nova | 10:26 | |
| *** sahid has quit IRC | 10:26 | |
| ndipanov | sudipto, I don't really know what the right solution would be from libvirt POV | 10:26 |
| johnthetubaguy | bauzas: totally agreed | 10:26 |
| *** Guest75569 has quit IRC | 10:27 | |
| *** heyongli has quit IRC | 10:27 | |
| *** heyongli has joined #openstack-nova | 10:28 | |
| bauzas | johnthetubaguy: mmm, I see etherpads ready for sessions, I guess we can begin drawing ? :) | 10:32 |
| bauzas | ha, you did that already, nvm | 10:32 |
| ihrachys | danpb: ping re revert of qemu patch | 10:34 |
| *** irenab has quit IRC | 10:37 | |
| johnthetubaguy | bauzas: most of those etherpads are blank right now, but yeah feel free to get filling them up with things :) | 10:37 |
| *** heyongli has quit IRC | 10:38 | |
| johnthetubaguy | bauzas: I just uploaded the draft schedule actually: http://mitakadesignsummit.sched.org/type/Nova | 10:38 |
| bauzas | yup, just saw that | 10:38 |
| *** gongysh has quit IRC | 10:38 | |
| bauzas | I'll on vacation next week, so I'd do some stuff this week | 10:38 |
| bauzas | be* | 10:38 |
| johnthetubaguy | bauzas: ah, good point, that was my original plan too, I suspect there are others doing the same | 10:39 |
| *** sahid has joined #openstack-nova | 10:39 | |
| bauzas | yeah, Japan is attractive | 10:39 |
| *** heyongli has joined #openstack-nova | 10:39 | |
| bauzas | johnthetubaguy: I guess you were aware of https://mitakadesignsummit.sched.org/event/b2812a85cbca739432fe0b3d26f5fe69#.Vh4xArzDia4 ? | 10:40 |
| bauzas | that's conflicting the 2nd resource session tho :( | 10:40 |
| *** Murali_ has joined #openstack-nova | 10:40 | |
| *** IanGovett has joined #openstack-nova | 10:40 | |
| johnthetubaguy | bauzas: no, I still need to reach out to armando, failed so far | 10:40 |
| bauzas | ack | 10:41 |
| *** Murali has quit IRC | 10:42 | |
| *** Murali_ is now known as Murali | 10:42 | |
| openstackgerrit | Bartosz Fic proposed openstack/nova: Live migration fails when the instance has a config_drive_format=iso9660 https://review.openstack.org/234659 | 10:43 |
| danpb | ihrachys: ? | 10:44 |
| *** leimeon has quit IRC | 10:45 | |
| ihrachys | danpb: can you check comment at https://review.openstack.org/#/c/234696/ ? | 10:46 |
| *** heyongli has quit IRC | 10:48 | |
| openstackgerrit | Bartosz Fic proposed openstack/nova: Live migration fails when the instance has a config_drive_format=iso9660 https://review.openstack.org/234659 | 10:48 |
| *** heyongli has joined #openstack-nova | 10:49 | |
| danpb | ihrachys: there's really no need to ping people to tell them you've added a comment to gerrit | 10:49 |
| ihrachys | danpb: it's gate failure, I think those are different. | 10:50 |
| *** marzif has quit IRC | 10:50 | |
| *** kmartin has quit IRC | 10:51 | |
| *** Marga_ has joined #openstack-nova | 10:52 | |
| *** ociuhandu has joined #openstack-nova | 10:53 | |
| *** marzif has joined #openstack-nova | 10:53 | |
| *** markvoelker has joined #openstack-nova | 10:55 | |
| *** stevemar_ has joined #openstack-nova | 10:56 | |
| *** e0ne has joined #openstack-nova | 10:56 | |
| *** heyongli has quit IRC | 10:58 | |
| *** PaulMurray has joined #openstack-nova | 10:58 | |
| openstackgerrit | Paul Carlton proposed openstack/nova: Mask the token used to allow access to consoles https://review.openstack.org/220622 | 10:58 |
| *** stevemar_ has quit IRC | 10:59 | |
| *** heyongli has joined #openstack-nova | 10:59 | |
| sdague | bauzas: ++ for getting the ball moving on tasks again | 10:59 |
| *** markvoelker has quit IRC | 11:00 | |
| *** ociuhandu has quit IRC | 11:01 | |
| sdague | ihrachys: I think you should change your logstash query to - message:"segfault at" AND message:"nova-compute" | 11:03 |
| sdague | it makes it a bit more clear that there is a recent event here | 11:03 |
| sdague | http://logstash.openstack.org/#eyJzZWFyY2giOiJtZXNzYWdlOlwic2VnZmF1bHQgYXRcIiBBTkQgbWVzc2FnZTpcIm5vdmEtY29tcHV0ZVwiIiwiZmllbGRzIjpbXSwib2Zmc2V0IjowLCJ0aW1lZnJhbWUiOiI2MDQ4MDAiLCJncmFwaG1vZGUiOiJjb3VudCIsInRpbWUiOnsidXNlcl9pbnRlcnZhbCI6MH0sInN0YW1wIjoxNDQ0ODIwNTI3MzYwfQ== | 11:03 |
| ihrachys | sdague: sure, I just manually looked for the earliest hit - when I referred to Oct 13 | 11:03 |
| danpb | ihrachys: we need a separate bug for this because i looks to be not related to the neutron lbaas bug at all | 11:04 |
| danpb | the logstash failures appear to hit arbitrary jobs/tests | 11:04 |
| ihrachys | danpb: well, I agree you may want another bug, but it's probably related: lbaas uses octavia, that uses nova to start an instance to host the load balancer. but the instance fails before of the nova segfault | 11:05 |
| sdague | yeh, and that signature hits no lbaas jobs | 11:05 |
| *** ijw has joined #openstack-nova | 11:05 | |
| sdague | http://goo.gl/WHZKHS is the job breakdown | 11:06 |
| ihrachys | sdague: it does hit lbaas jobs for neutron repo, see the first log link in the bug | 11:06 |
| danpb | that's just a co-incidence though | 11:06 |
| ihrachys | specifically, lbaasv2-minimal job, and that's because we forgot to disable nova driver for octavia for that job (others work fine) | 11:07 |
| danpb | this is just like any nova bug which causes random failures across all jobs - any of them could randomly cause an lbaas job to fail | 11:07 |
| *** annegentle has joined #openstack-nova | 11:07 | |
| sdague | ihrachys: that's not an lbaasv2-minimal job, it's a postgres job - gate-tempest-dsvm-postgres-full | 11:08 |
| *** ociuhandu has joined #openstack-nova | 11:08 | |
| ihrachys | sdague: I am sorry, where do you look for? I specifically refer to http://logs.openstack.org/43/234343/1/check/gate-neutron-lbaasv2-dsvm-minimal/b767acd/logs/testr_results.html.gz | 11:08 |
| sdague | ihrachys: the link in the review | 11:08 |
| *** heyongli has quit IRC | 11:08 | |
| *** zhangjn has quit IRC | 11:09 | |
| ihrachys | sdague: danpb: well maybe that job is the only one that triggers test failure for the error | 11:09 |
| *** heyongli has joined #openstack-nova | 11:09 | |
| kashyap | ihrachys: This log doesn't exist anymore? http://logs.openstack.org/27/230227/5/gate/gate-tempest-dsvm-postgres-full/27576c8/logs/syslog.txt | 11:10 |
| kashyap | (It was mentioned in the comments.) | 11:10 |
| danpb | i'll just file a new bug myself for this | 11:10 |
| *** ijw has quit IRC | 11:11 | |
| ihrachys | kashyap: how is this relevant? The point is it started to crash on Oct 13, in qemu-img, and that seems to be quite in line with that nova patch merged. | 11:11 |
| *** EinstCrazy has quit IRC | 11:11 | |
| ihrachys | danpb: ok. I think we should still refer to the lbaas bug as Related | 11:11 |
| *** shuquan has joined #openstack-nova | 11:11 | |
| danpb | its really not related - its just co-incidence | 11:12 |
| kashyap | ihrachys: I'm only reading the comments just now, and clicked on the URL mentioned above. If it's not relevant, why was that mentioned? | 11:12 |
| ihrachys | kashyap: I copy-pasted the link from logstash | 11:12 |
| *** annegentle has quit IRC | 11:12 | |
| *** marzif has quit IRC | 11:12 | |
| ihrachys | kashyap: that link was the first that hit nova-compute on Oct 13 | 11:12 |
| danpb | https://bugs.launchpad.net/nova/+bug/1506012 | 11:12 |
| openstack | Launchpad bug 1506012 in OpenStack Compute (nova) "Gate tests randomly failing qemu-img info calls" [Critical,New] | 11:12 |
| ihrachys | not sure why no logs | 11:12 |
| openstackgerrit | Ihar Hrachyshka proposed openstack/nova: Revert "virt: Use preexec_fn to ulimit qemu-img info call" https://review.openstack.org/234696 | 11:13 |
| ihrachys | danpb: cool, thanks. I updated the commit message ^ | 11:13 |
| *** shuquan has quit IRC | 11:13 | |
| *** marzif has joined #openstack-nova | 11:13 | |
| danpb | ihrachys: as i said, its not related to that neutron bug | 11:13 |
| danpb | also your commit message description is still talking about hte neutron bug | 11:14 |
| sdague | so, there is clearly a new issue in nova-compute segfaulting, and it may have hit a few neutron jobs, but it is orthoginal | 11:14 |
| *** deepakcs has quit IRC | 11:14 | |
| ihrachys | danpb: I don't agree with your interpretation of Related-Bug. That said, I am fine to make it any way that would allow it to merge. I will update the commit message once more. | 11:15 |
| openstackgerrit | Ihar Hrachyshka proposed openstack/nova: Revert "virt: Use preexec_fn to ulimit qemu-img info call" https://review.openstack.org/234696 | 11:16 |
| ihrachys | danpb: should be ok now, if not, please feel free to modify the commit message as you see more fit | 11:16 |
| sdague | http://logs.openstack.org/92/194192/2/gate/gate-tempest-dsvm-full/33cef79/logs/screen-n-cpu.txt.gz#_2015-10-14_08_51_29_142 is the real signature here | 11:16 |
| danpb | ok, i've +2 it now | 11:17 |
| *** heyongli has quit IRC | 11:19 | |
| *** heyongli has joined #openstack-nova | 11:19 | |
| sdague | http://goo.gl/ziE4T8 is an updated query with the nova-compute logs that highlights the issue more clearly | 11:20 |
| kashyap | sdague: Added a logstash query for it? | 11:20 |
| sdague | do we have a dedicated bug for this yet (not the neutron one) | 11:20 |
| kashyap | s/logstash/ER/ | 11:20 |
| sdague | ? | 11:20 |
| kashyap | sdague: See above, Dan filed one | 11:21 |
| kashyap | 1506012 | 11:21 |
| sdague | oh, yep, great | 11:21 |
| danpb | sdague: i just updated hte bug with your new query | 11:21 |
| *** manas has joined #openstack-nova | 11:21 | |
| sdague | danpb: thanks | 11:21 |
| sdague | I'll put in an ER signature | 11:21 |
| *** e0ne has quit IRC | 11:23 | |
| ihrachys | sdague: danpb: thanks for turnaround | 11:28 |
| *** heyongli has quit IRC | 11:29 | |
| *** heyongli has joined #openstack-nova | 11:30 | |
| *** _amrith_ is now known as amrith | 11:32 | |
| ihrachys | sdague: it's still interesting why it hits neutron lbaas-minimal job only. I would expect other tempest jobs to be affected too. But it's pretty consistent just for that job in our gate. | 11:36 |
| ihrachys | only == in neutron gate | 11:36 |
| sdague | ihrachys: it doesn't only hit that job | 11:36 |
| sdague | you saw the job break down query right? | 11:37 |
| sdague | you have to be really careful extrapolating from a few things you happened to have noticed | 11:37 |
| *** EinstCrazy has joined #openstack-nova | 11:37 | |
| sdague | because you can see a pattern where there is none because you don't have all the data | 11:37 |
| sdague | http://goo.gl/WHZKHS | 11:37 |
| *** zhangjn has joined #openstack-nova | 11:38 | |
| sdague | I suspect that it tends to hit more neutron jobs in general because there are more guests created during the jobs | 11:38 |
| ihrachys | sdague: ok right, now I see others are affected. it's interesting why I haven't seen any others. or maybe I saw and rechecked, but since minimal was new, I decided to look closer | 11:38 |
| *** VW has joined #openstack-nova | 11:38 | |
| ihrachys | ok cool, now cleared. thanks folks :) | 11:39 |
| sdague | ihrachys: yes, that's often how we dig into these things, but it's really important to then take a step back and look at all the data and see what's going on | 11:39 |
| *** heyongli has quit IRC | 11:39 | |
| *** heyongli has joined #openstack-nova | 11:40 | |
| ihrachys | sdague: btw how do you get that 'quick analysis' page? | 11:41 |
| ihrachys | kashyap: I know why the link did not work: looks like though kibana provides .txt link, in reality the file is gzipped, hence you need to append .gz to it | 11:41 |
| *** gszasz has joined #openstack-nova | 11:43 | |
| kashyap | ihrachys: Ah, yeah, I'm aware of this, but didn't realize it instantly. Thanks. | 11:43 |
| *** irenab has joined #openstack-nova | 11:44 | |
| *** heyongli has quit IRC | 11:45 | |
| sdague | ihrachys: it's the "Score" button on results | 11:45 |
| openstackgerrit | Sergey Nikitin proposed openstack/nova: Added server tags support in nova-api https://review.openstack.org/128940 | 11:45 |
| sdague | click on the > on a column. | 11:45 |
| sdague | and on the popup you can click on Score | 11:46 |
| *** heyongli has joined #openstack-nova | 11:46 | |
| *** aix has quit IRC | 11:46 | |
| sdague | it's basically a groupby | 11:46 |
| sdague | query that gives you the top 25 results | 11:46 |
| kashyap | sdague: That's a nice tip | 11:47 |
| ihrachys | sdague: oh cool thanks, I didn't realize there a lot of useful stuff under that arrow. | 11:47 |
| *** VW has quit IRC | 11:47 | |
| *** VW has joined #openstack-nova | 11:48 | |
| *** baoli has joined #openstack-nova | 11:48 | |
| *** Sree has joined #openstack-nova | 11:48 | |
| *** baoli_ has joined #openstack-nova | 11:49 | |
| *** thorst has joined #openstack-nova | 11:50 | |
| *** VW has quit IRC | 11:52 | |
| *** baoli has quit IRC | 11:52 | |
| *** subscope has quit IRC | 11:54 | |
| *** shuquan has joined #openstack-nova | 11:54 | |
| *** heyongli has quit IRC | 11:55 | |
| *** shuquan has quit IRC | 11:55 | |
| *** markvoelker has joined #openstack-nova | 11:56 | |
| *** heyongli has joined #openstack-nova | 11:56 | |
| *** aysyd has joined #openstack-nova | 11:56 | |
| *** subscope has joined #openstack-nova | 11:57 | |
| *** shuquan has joined #openstack-nova | 11:59 | |
| *** baoli_ has quit IRC | 12:00 | |
| *** markvoelker has quit IRC | 12:00 | |
| *** markvoelker has joined #openstack-nova | 12:00 | |
| *** manas has quit IRC | 12:02 | |
| *** ijw has joined #openstack-nova | 12:03 | |
| *** lucasagomes is now known as lucas-hungry | 12:03 | |
| *** irenab has quit IRC | 12:04 | |
| *** manas has joined #openstack-nova | 12:05 | |
| *** pixelbeat has quit IRC | 12:05 | |
| *** irenab has joined #openstack-nova | 12:06 | |
| *** Miouge has quit IRC | 12:07 | |
| *** ijw has quit IRC | 12:07 | |
| *** subscope has quit IRC | 12:07 | |
| *** pixelbeat has joined #openstack-nova | 12:08 | |
| *** raildo-afk is now known as raildo | 12:12 | |
| *** alexschm has quit IRC | 12:13 | |
| *** Sree has quit IRC | 12:14 | |
| *** Murali has quit IRC | 12:14 | |
| *** Sree has joined #openstack-nova | 12:14 | |
| *** heyongli has quit IRC | 12:15 | |
| *** achanda has joined #openstack-nova | 12:16 | |
| *** alex_klimov has quit IRC | 12:16 | |
| *** alex_klimov has joined #openstack-nova | 12:17 | |
| *** heyongli has joined #openstack-nova | 12:17 | |
| *** achanda has quit IRC | 12:17 | |
| *** achanda has joined #openstack-nova | 12:18 | |
| *** Sree has quit IRC | 12:19 | |
| *** shuquan has quit IRC | 12:20 | |
| *** gongysh has joined #openstack-nova | 12:20 | |
| *** garyk1 has joined #openstack-nova | 12:21 | |
| *** tbh has quit IRC | 12:22 | |
| *** achanda has quit IRC | 12:22 | |
| *** garyk1 has left #openstack-nova | 12:23 | |
| *** garyk1 has joined #openstack-nova | 12:23 | |
| *** shuquan has joined #openstack-nova | 12:24 | |
| garyk1 | johnthetubaguy: can you please look at https://review.openstack.org/230425 i addressed your comments | 12:25 |
| *** heyongli has quit IRC | 12:25 | |
| *** mpavone has quit IRC | 12:26 | |
| *** heyongli has joined #openstack-nova | 12:27 | |
| openstackgerrit | Rajesh Tailor proposed openstack/nova: Fix quota-update when nova-compute startup finish https://review.openstack.org/170118 | 12:29 |
| *** ociuhandu has quit IRC | 12:30 | |
| *** Amichayp has quit IRC | 12:30 | |
| *** garyk has quit IRC | 12:30 | |
| *** marzif has quit IRC | 12:31 | |
| *** marzif has joined #openstack-nova | 12:31 | |
| *** aix has joined #openstack-nova | 12:33 | |
| *** ociuhandu has joined #openstack-nova | 12:35 | |
| *** dave-mccowan has joined #openstack-nova | 12:35 | |
| *** heyongli has quit IRC | 12:36 | |
| *** shuquan has quit IRC | 12:36 | |
| johnthetubaguy | garyk1: thanks for that update, upgrade impact is much clearer now, thank you | 12:36 |
| *** heyongli has joined #openstack-nova | 12:37 | |
| garyk1 | johnthetubaguy: ok, thanks. | 12:37 |
| *** baoli has joined #openstack-nova | 12:37 | |
| *** garyk1 has left #openstack-nova | 12:38 | |
| *** garyk1 has joined #openstack-nova | 12:38 | |
| *** rfolco has joined #openstack-nova | 12:42 | |
| *** subscope has joined #openstack-nova | 12:43 | |
| tristanC | ihrachys: danpb: oh wow, so do we know what went wrong ? are the default limit to strict or something ? | 12:43 |
| *** ccarmack has joined #openstack-nova | 12:43 | |
| ihrachys | tristanC: I personally don't know. | 12:44 |
| markus_z | sdague: Could you help me with https://bugs.launchpad.net/nova/+bug/1505677 please? I think I did the update of the report wrong. I nominated it for "Liberty" and changed it to invalid for master. I'm not sure if this was the correct thing to do. | 12:44 |
| openstack | Launchpad bug 1505677 in openstack-ansible "oslo.versionedobjects 0.11.0 causing KeyError: 'objects' in nova-conductor log" [Critical,In progress] - Assigned to Jesse Pretorius (jesse-pretorius) | 12:45 |
| *** heyongli has quit IRC | 12:46 | |
| johnthetubaguy | markus_z: didn't we fix that in RC3? | 12:47 |
| *** heyongli has joined #openstack-nova | 12:47 | |
| markus_z | johnthetubaguy: yes, sdague and dims did that. I wanted the bug report to reflect that. | 12:47 |
| danpb | tristanC: i'm not sure to be honest - the failures look really random | 12:48 |
| *** rfolco has quit IRC | 12:48 | |
| sdague | johnthetubaguy: we fixed it, the fix was happening before the bug, we didn't delay the patch by changing the commit message | 12:48 |
| *** rfolco has joined #openstack-nova | 12:48 | |
| johnthetubaguy | markus_z: so RC3 is closed now, so its not easy to target https://bugs.launchpad.net/nova/+bug/1505677 for RC3 to mark it as fixed, ttx might now how to update that now its released | 12:48 |
| openstack | Launchpad bug 1505677 in openstack-ansible "oslo.versionedobjects 0.11.0 causing KeyError: 'objects' in nova-conductor log" [Critical,In progress] - Assigned to Jesse Pretorius (jesse-pretorius) | 12:48 |
| danpb | tristanC: i would expect that for any single image the memory & cpu usage should be approx constant | 12:48 |
| danpb | tristanC: so the randomness is rather suprising | 12:49 |
| *** setuid has quit IRC | 12:49 | |
| johnthetubaguy | sdague: yeah, totally, just needs targeting to the closed milestone, I think | 12:49 |
| danpb | tristanC: it could be todo with the way we use a preexec fn callback | 12:50 |
| danpb | tristanC: we are setting the limit between the time we fork() and when we exec() | 12:50 |
| markus_z | johnthetubaguy: Ok, I'll ping ttx later about this. | 12:50 |
| danpb | tristanC: so at that point, we have a copy-on-write image of the main nova process memory still | 12:51 |
| *** setuid has joined #openstack-nova | 12:51 | |
| danpb | tristanC: so our memory limit that is intended to apply to qemu-img, is in fact being applied to whatever we inherit from nova, before we execve | 12:51 |
| danpb | it is conceivable that nova memory usage is randomly exceeding the limit we set | 12:51 |
| tristanC | danpb: arg, yes this might very well be the reason this is failing randomly | 12:52 |
| danpb | tristanC: the only way around this is to *not* use the prexec-fn we just added to oslo.concurrency | 12:52 |
| danpb | tristanC: and instead do something like my original solution which invokes prlimit | 12:52 |
| danpb | since we can't rely on prlimit though, we'd have to do something else | 12:53 |
| tristanC | danpb: yeah prlimit would have been the best | 12:53 |
| danpb | we could create a stupidly small re-impl of prlimit in python for example | 12:53 |
| danpb | ...but that would be horribly slow :L-( | 12:53 |
| danpb | or we could create a nova-prlimit, but only use that if we find the real native prlimit missing | 12:54 |
| danpb | that would limit the perf hit to ancient-ish distros | 12:54 |
| tristanC | danpb: alright, thanks for the clarification | 12:55 |
| *** eharney has joined #openstack-nova | 12:55 | |
| ttx | johnthetubaguy: it's easy, just need to mark the milestone as active again (and remember to close it again once done) | 12:56 |
| * ttx does the job | 12:56 | |
| *** annegentle has joined #openstack-nova | 12:56 | |
| *** su_zhang has joined #openstack-nova | 12:56 | |
| *** heyongli has quit IRC | 12:56 | |
| *** stevemar_ has joined #openstack-nova | 12:57 | |
| *** eliqiao has quit IRC | 12:57 | |
| ttx | johnthetubaguy, markus_z: done | 12:57 |
| *** rlrossit has joined #openstack-nova | 12:58 | |
| *** Miouge has joined #openstack-nova | 13:00 | |
| *** stevemar_ has quit IRC | 13:00 | |
| *** annegentle has quit IRC | 13:01 | |
| *** heyongli has joined #openstack-nova | 13:01 | |
| *** mgoddard has quit IRC | 13:03 | |
| *** shuquan has joined #openstack-nova | 13:04 | |
| *** shangxdy has quit IRC | 13:04 | |
| *** shangxdy has joined #openstack-nova | 13:04 | |
| *** heyongli has quit IRC | 13:06 | |
| *** heyongli has joined #openstack-nova | 13:07 | |
| johnthetubaguy | ttx: ah, that makes sense, thank you | 13:08 |
| openstackgerrit | Sean Dague proposed openstack/python-novaclient: make project_id optional in urls for version discovery https://review.openstack.org/234782 | 13:11 |
| *** subscope has quit IRC | 13:11 | |
| *** dustins has joined #openstack-nova | 13:12 | |
| *** lucas-hungry is now known as lucasagomes | 13:12 | |
| *** marzif has quit IRC | 13:13 | |
| *** romainh has joined #openstack-nova | 13:14 | |
| *** sudipto has quit IRC | 13:14 | |
| *** vladikr has joined #openstack-nova | 13:14 | |
| *** amrith is now known as _amrith_ | 13:15 | |
| *** amotoki has joined #openstack-nova | 13:15 | |
| *** claudiub has joined #openstack-nova | 13:16 | |
| *** marzif has joined #openstack-nova | 13:16 | |
| *** heyongli has quit IRC | 13:17 | |
| *** tbh has joined #openstack-nova | 13:17 | |
| *** stefanb has joined #openstack-nova | 13:19 | |
| *** subscope has joined #openstack-nova | 13:19 | |
| *** annegentle has joined #openstack-nova | 13:20 | |
| *** vladikr has quit IRC | 13:21 | |
| *** vladikr has joined #openstack-nova | 13:21 | |
| *** shuquan has quit IRC | 13:21 | |
| *** Alexus has quit IRC | 13:22 | |
| *** kylek3h has joined #openstack-nova | 13:23 | |
| *** kylek3h has quit IRC | 13:23 | |
| *** kylek3h has joined #openstack-nova | 13:23 | |
| *** shuquan has joined #openstack-nova | 13:25 | |
| *** ijuwang has joined #openstack-nova | 13:25 | |
| *** mriedem has joined #openstack-nova | 13:25 | |
| *** Sree has joined #openstack-nova | 13:25 | |
| sdague | johnthetubaguy: fyi, implementation notes on unwinding project_id for nova - https://etherpad.openstack.org/p/removing-project-id-in-nova-urls | 13:26 |
| mriedem | sdague: simple patch to unblock the hyperv driver https://review.openstack.org/#/c/234437/ | 13:28 |
| *** subscope has quit IRC | 13:28 | |
| sdague | mriedem: don't we normally put that behind a try | 13:29 |
| *** yuntongjin has joined #openstack-nova | 13:29 | |
| sdague | at the global level | 13:29 |
| sdague | instead of sticking in a method | 13:29 |
| mriedem | sdague: that's what i had asked for in ps2 | 13:29 |
| mriedem | i've seen imports at the method level too | 13:29 |
| mriedem | usually to avoid circular deps though, not failed imports | 13:29 |
| *** sudipto has joined #openstack-nova | 13:29 | |
| sdague | the problem with the method level is it's potentially a late failure | 13:30 |
| mriedem | yeah, see ps2 | 13:30 |
| mriedem | it was purely conditional there | 13:30 |
| *** pm90_ has joined #openstack-nova | 13:31 | |
| openstackgerrit | Bartosz Fic proposed openstack/nova: Live migration fails when the instance has a config_drive_format=iso9660 https://review.openstack.org/234659 | 13:31 |
| *** ankit_ag has quit IRC | 13:31 | |
| sdague | ok, so you're with me on the -1 then :) | 13:32 |
| openstackgerrit | Alexey I. Froloff proposed openstack/nova: Fix more inconsistency between Nova-Net and Neutron https://review.openstack.org/226274 | 13:32 |
| claudiub | sdague: putting that import in a try, won't that also ignore the import failure anyways? | 13:32 |
| *** amotoki has quit IRC | 13:32 | |
| mriedem | claudiub: i like the conditional in ps2 | 13:32 |
| sdague | claudiub: right, but then it happens at start time | 13:32 |
| mriedem | where you had the if resource: in _qemu_resources() | 13:32 |
| sdague | not some other random point of time down the road | 13:32 |
| * dansmith yawns | 13:33 | |
| claudiub | ah, I see. makes sense. | 13:33 |
| sdague | and _qemu_resources isn't a singleton, it gets called a bunch. And I don't like import getting called a bunch even if it's supposed to be a noop on all the follow on calls | 13:33 |
| *** marzif has quit IRC | 13:37 | |
| *** eharney has quit IRC | 13:38 | |
| claudiub | sdague: mriedem: question: do we want to log an error message if the import failed? | 13:38 |
| mriedem | claudiub: i don't think so | 13:39 |
| claudiub | or warning | 13:39 |
| sdague | claudiub: I don't think so, because it's normal on some platforms | 13:39 |
| claudiub | k | 13:39 |
| *** subscope has joined #openstack-nova | 13:39 | |
| sdague | a WARN inside _qemu_resources if resource=None might be appropriate | 13:39 |
| *** abhishekk has quit IRC | 13:39 | |
| sdague | to explain that something went very wrong | 13:39 |
| *** amotoki has joined #openstack-nova | 13:40 | |
| *** annegentle has quit IRC | 13:41 | |
| *** subscope has quit IRC | 13:44 | |
| claudiub | k | 13:44 |
| *** mdrabe has joined #openstack-nova | 13:44 | |
| *** subscope has joined #openstack-nova | 13:44 | |
| *** mdrabe_ has joined #openstack-nova | 13:44 | |
| *** mdrabe_ has quit IRC | 13:45 | |
| *** deepthi has quit IRC | 13:47 | |
| *** amotoki has quit IRC | 13:47 | |
| *** tpeoples has joined #openstack-nova | 13:48 | |
| *** eharney has joined #openstack-nova | 13:51 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Ignores ImportError when importing resource module https://review.openstack.org/234437 | 13:51 |
| claudiub | sdague: mriedem: ^ done. Thanks for the help. :) | 13:51 |
| *** ctrath has joined #openstack-nova | 13:51 | |
| *** VW has joined #openstack-nova | 13:52 | |
| dansmith | zzzeek: around? | 13:53 |
| *** dustins has quit IRC | 13:55 | |
| mriedem | claudiub: comments inline | 13:55 |
| *** shuquan has quit IRC | 13:56 | |
| *** jistr is now known as jistr|mtg | 13:56 | |
| markus_z | ttx: thanks! (sorry, I was in a meeting) | 13:58 |
| zzzeek | dansmith: yes | 13:58 |
| *** mrkz has joined #openstack-nova | 13:58 | |
| dansmith | zzzeek: so, I just wanted to see if you can help spot something | 13:58 |
| dansmith | zzzeek: https://github.com/openstack/nova/blob/master/nova/db/sqlalchemy/api.py#L448-L451 | 13:59 |
| dansmith | if you recall that | 13:59 |
| dansmith | zzzeek: running against this dump: http://logs.openstack.org/89/233289/9/check/gate-grenade-dsvm-partial-ncpu/e0c4c5a/logs/grenade.sh.txt.gz#_2015-10-13_23_53_15_824 | 13:59 |
| dansmith | zzzeek: I'm getting either 0 or None back from that call and I don't know why | 13:59 |
| *** eharney has quit IRC | 13:59 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Ignores ImportError when importing resource module https://review.openstack.org/234437 | 13:59 |
| zzzeek | dansmith: None back means that the SELECT statement matched no rows | 14:00 |
| dansmith | zzzeek: version is the last field on each of those lines, so it's easier to spot | 14:00 |
| dansmith | zzzeek: oh crap, does that mean the false() isn't matching the 0 of the forced_down? | 14:00 |
| *** manas has quit IRC | 14:00 | |
| zzzeek | dansmith: i think that it should, howver it would be one of the things I'd suspect to look at | 14:01 |
| *** dims_ has quit IRC | 14:01 | |
| *** ssurana has joined #openstack-nova | 14:01 | |
| *** dustins has joined #openstack-nova | 14:01 | |
| dansmith | zzzeek: dang, so there is a test that passes on this in sqlite.. could that differ between that and mysql? | 14:01 |
| *** dims has joined #openstack-nova | 14:01 | |
| zzzeek | dansmith: sqlite and mysql would work the same way in this regard, let me make sure false() works on mysql in this context | 14:01 |
| *** amotoki has joined #openstack-nova | 14:02 | |
| dansmith | zzzeek: oh god, I just got it | 14:02 |
| dansmith | zzzeek: it's nova-compute, not compute | 14:02 |
| dansmith | zzzeek: I bet we have a bunch of tests that are broken for that | 14:02 |
| zzzeek | dansmith: OK false() does what its supposed to normally, *however*, mysql in fact has a "false" constant | 14:03 |
| zzzeek | so for the mysql backend a query comes out as: SELECT count(%s) AS count_1 | 14:03 |
| zzzeek | WHERE q = false | 14:03 |
| zzzeek | dansmith: sqlite produces: SELECT count(?) AS count_1 | 14:04 |
| zzzeek | WHERE q = 0 | 14:04 |
| *** pm90__ has joined #openstack-nova | 14:04 | |
| dansmith | zzzeek: so should I change that to == 0? | 14:05 |
| zzzeek | dansmith: false should be equated to 0 on mysql, just tried it | 14:06 |
| dansmith | okay | 14:07 |
| *** stevemar_ has joined #openstack-nova | 14:07 | |
| dansmith | zzzeek: thanks for your help, sorry I'm an idiot and can't read :) | 14:07 |
| *** pm90_ has quit IRC | 14:07 | |
| *** lykinsbd has quit IRC | 14:07 | |
| zzzeek | dansmith: im not sure what the problem is yet i still need to look through those rows, this should be working | 14:08 |
| dansmith | zzzeek: it shouldn't because I'm checking for binary == compute | 14:08 |
| zzzeek | dansmith: oh | 14:08 |
| dansmith | zzzeek: it just took you saying "null means no rows" to figure that out :/ | 14:08 |
| zzzeek | dansmith: ok great :) | 14:08 |
| *** marzif has joined #openstack-nova | 14:09 | |
| zzzeek | dansmith: oh didnt see your line there that you found it :) | 14:09 |
| dansmith | heh, yeah | 14:09 |
| *** smatzek_ has joined #openstack-nova | 14:09 | |
| bauzas | johnthetubaguy: if you agree, I can respin https://review.openstack.org/#/c/192260/ | 14:10 |
| dansmith | sdague: I can't depends-on a nova patch to get grenade to use that for the master side, can I? | 14:10 |
| dansmith | sdague: er, depends-on a nova patch, from a grenade patch I mean | 14:10 |
| bauzas | sdague: oh btw. in http://lists.openstack.org/pipermail/openstack-dev/2015-October/076869.html by saying same UUIDs, you meant stable and master changes share the same Change-ID, right? | 14:11 |
| *** moshele has quit IRC | 14:11 | |
| bauzas | I thought it was impossible | 14:11 |
| *** vlaza has quit IRC | 14:11 | |
| *** smurke has joined #openstack-nova | 14:12 | |
| bauzas | (ie. I always ask for a new change-id before submitting a backport) | 14:12 |
| *** Piet has quit IRC | 14:13 | |
| johnthetubaguy | bauzas: yes, that would be great, thanks | 14:13 |
| bauzas | ack | 14:13 |
| johnthetubaguy | bauzas: I suspect its just the index that needs merging | 14:13 |
| bauzas | yup | 14:13 |
| sdague | bauzas: the change-id can be the same | 14:14 |
| *** eharney has joined #openstack-nova | 14:15 | |
| sdague | the stable maint teams were often -1ing them if they weren't | 14:15 |
| *** burgerk has joined #openstack-nova | 14:15 | |
| sdague | dansmith: you can depends-on with grenade and nova | 14:15 |
| dansmith | sdague: orly, sweet | 14:15 |
| sdague | because depends-on doesn't care about branches | 14:15 |
| *** ihrachys has quit IRC | 14:16 | |
| dansmith | sdague: okay | 14:16 |
| bauzas | sdague: okay, good to know | 14:16 |
| *** _amrith_ is now known as amrith | 14:17 | |
| *** ihrachys has joined #openstack-nova | 14:17 | |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: docs: add the scheduler evolution plans https://review.openstack.org/192260 | 14:17 |
| bauzas | johnthetubaguy: ^ | 14:17 |
| mriedem | johnthetubaguy: dansmith: do you have the link handy to the potential summit agenda etherpad? | 14:19 |
| *** burt has joined #openstack-nova | 14:19 | |
| johnthetubaguy | mriedem: sure https://etherpad.openstack.org/p/mitaka-nova-summit-suggestions | 14:20 |
| johnthetubaguy | bauzas: thanks | 14:20 |
| johnthetubaguy | mriedem: I have uploaded the current draft to sched | 14:20 |
| johnthetubaguy | mriedem: just about to send an note to the ML about all that | 14:20 |
| *** asettle has quit IRC | 14:21 | |
| *** sneti has joined #openstack-nova | 14:21 | |
| *** browne has joined #openstack-nova | 14:22 | |
| *** xyang1 has joined #openstack-nova | 14:23 | |
| *** sneti has quit IRC | 14:23 | |
| *** sneti has joined #openstack-nova | 14:24 | |
| *** wznoinsk1 has joined #openstack-nova | 14:25 | |
| *** marcusvrn_ has quit IRC | 14:25 | |
| *** wznoinsk has quit IRC | 14:26 | |
| *** wznoinsk2 has joined #openstack-nova | 14:26 | |
| *** yuntongjin has quit IRC | 14:26 | |
| *** yuntongjin1 has joined #openstack-nova | 14:26 | |
| *** Murali has joined #openstack-nova | 14:27 | |
| openstackgerrit | Dan Smith proposed openstack/nova: Fix service_version minimum calculation for compute RPC https://review.openstack.org/234809 | 14:27 |
| dansmith | sdague: trivial stupid fix required for the auto pinning stuff ^ | 14:27 |
| *** jwcroppe has joined #openstack-nova | 14:27 | |
| *** wznoinsk1 has quit IRC | 14:27 | |
| *** jwcroppe has quit IRC | 14:28 | |
| *** jwcroppe has joined #openstack-nova | 14:28 | |
| *** annegentle has joined #openstack-nova | 14:28 | |
| openstackgerrit | Mark Doffman proposed openstack/nova: libvirt: Don't attempt to convert initrd images. https://review.openstack.org/225289 | 14:29 |
| *** mc_nair has joined #openstack-nova | 14:29 | |
| *** pumaranikar has joined #openstack-nova | 14:31 | |
| diana_clarke | bauzas: Moving the conversation here in case someone else knows why the novasclient gate has been failing. In particular: gate-novaclient-dsvm-functional | 14:31 |
| diana_clarke | I suspect it hasn't passed since Oct 5th, but I could be wrong. | 14:32 |
| *** nihilifer has quit IRC | 14:32 | |
| *** READ10 has joined #openstack-nova | 14:32 | |
| pumaranikar | hi, can you suggest how can i reproduce https://bugs.launchpad.net/nova/+bug/1384228 | 14:33 |
| openstack | Launchpad bug 1384228 in OpenStack Compute (nova) "Exception should be raised if nova has failed to add fixed ip" [Low,Confirmed] - Assigned to Pushkar Umaranikar (pushkar-umaranikar) | 14:33 |
| *** nihilifer has joined #openstack-nova | 14:33 | |
| *** electrocucaracha has joined #openstack-nova | 14:33 | |
| *** weihan has joined #openstack-nova | 14:34 | |
| *** haomaiwa_ has joined #openstack-nova | 14:35 | |
| *** tonytan4ever has joined #openstack-nova | 14:36 | |
| sdague | dansmith: is there a reason not to just throw an exception instead of warn there | 14:36 |
| *** mriedem has quit IRC | 14:37 | |
| sdague | this should be only called with an enumerated set of services, right? | 14:37 |
| dansmith | sdague: should, but it's technically just a query method and it will work if you have something in your DB without nova-, so I kinda feel like being liberal in what we accept will avoid us being bitten in the future | 14:38 |
| *** ihrachys has quit IRC | 14:38 | |
| dansmith | sdague: so, yes, we can raise there if you prefer, but my instinct is to not | 14:38 |
| *** diogogmt has joined #openstack-nova | 14:39 | |
| sdague | I guess my preference is to avoid the typo causing issues and people not noticing. It seems like adding a canonical service in nova is a big enough deal that explicitly calling that fact out shouldn't be too onerous. | 14:40 |
| bauzas | diana_clarke: seen that ? http://lists.openstack.org/pipermail/openstack-dev/2015-October/076805.html | 14:40 |
| diana_clarke | bauzas: No, I hadn't, thanks! That's exactly what I'm talking about. | 14:41 |
| dansmith | sdague: yeah, I was more thinking about the case where we change that prefix some day and we end up pinned because of that check in some code we can't change until after a cycle | 14:41 |
| dansmith | sdague: but that's fine | 14:41 |
| *** Sree has quit IRC | 14:42 | |
| *** Alexus has joined #openstack-nova | 14:42 | |
| dansmith | sdague: I'll leave that log and then raise, so we get the more detailed log information about what happened | 14:42 |
| *** sneti_ has joined #openstack-nova | 14:42 | |
| *** sneti has quit IRC | 14:42 | |
| sdague | dansmith: ok | 14:43 |
| *** lpetrut has quit IRC | 14:43 | |
| *** tpeoples has quit IRC | 14:45 | |
| *** ildikov has joined #openstack-nova | 14:46 | |
| *** mriedem has joined #openstack-nova | 14:47 | |
| mriedem | diana_clarke: Vek was saying in the ML that there were problems in that job | 14:48 |
| openstackgerrit | Anne Gentle proposed openstack/nova: Rearranges to create new Compute API Guide https://review.openstack.org/230186 | 14:48 |
| *** baoli has quit IRC | 14:48 | |
| *** baoli_ has joined #openstack-nova | 14:48 | |
| bauzas | mriedem: diana_clarke: and melwitt started investigating | 14:49 |
| *** changbl has joined #openstack-nova | 14:49 | |
| mriedem | bauzas: melwitt: is there a logstash query? | 14:49 |
| mriedem | or a bug reported? | 14:49 |
| bauzas | mriedem: http://lists.openstack.org/pipermail/openstack-dev/2015-October/076833.html | 14:49 |
| bauzas | I haven't yet digged into that | 14:50 |
| *** tbh has quit IRC | 14:50 | |
| *** stevemar_ has quit IRC | 14:50 | |
| *** stevemar_ has joined #openstack-nova | 14:51 | |
| *** rotbeard has quit IRC | 14:51 | |
| *** stevemar_ has quit IRC | 14:52 | |
| *** stevemar_ has joined #openstack-nova | 14:52 | |
| *** oomichi has joined #openstack-nova | 14:53 | |
| openstackgerrit | Diane Fleming proposed openstack/nova: Fix the bug of "Error spelling of 'explicitely'" https://review.openstack.org/232827 | 14:53 |
| *** sneti_ has quit IRC | 14:54 | |
| *** sneti has joined #openstack-nova | 14:54 | |
| *** erlarese has joined #openstack-nova | 14:55 | |
| *** dustins has quit IRC | 14:56 | |
| *** yuntongjin1 has quit IRC | 14:56 | |
| dansmith | sdague: you suck, and let me explain why | 14:57 |
| *** eharney_ has joined #openstack-nova | 14:57 | |
| dansmith | sdague: raising there found all the wrong unit tests I suspected we had, as mentioned to zzzeek | 14:58 |
| dansmith | sdague: which is a lot of extra work for me, and probably means you were right to make me raise here | 14:58 |
| mriedem | bauzas: melwitt: diana_clarke: Vek: here is a bug to track at least https://bugs.launchpad.net/python-novaclient/+bug/1506103 | 14:58 |
| openstack | Launchpad bug 1506103 in python-novaclient "gate-novaclient-dsvm-functional fails with 300 response code" [Undecided,New] | 14:58 |
| sdague | dansmith: :) | 14:58 |
| diana_clarke | mriedem, bauzas: Thanks! I should start paying attention to that mailing list... | 14:59 |
| diana_clarke | Next, I need to figure out why I keep getting gate failures for my nova patch. I don't think they are related to my patch either. | 15:00 |
| kashyap | diana_clarke: Be aware, it runs like a firehose :-) Though, you can subscribe per-tag basis | 15:00 |
| *** dustins has joined #openstack-nova | 15:00 | |
| *** haomaiwa_ has quit IRC | 15:01 | |
| dansmith | subscribing by tag is dangerous.. filtering in your MUA is better | 15:01 |
| dansmith | because mistyped tags (which happens really often) means you miss things | 15:01 |
| *** haomaiwang has joined #openstack-nova | 15:01 | |
| dansmith | and untagged mail gets dropped, IIRC | 15:01 |
| *** eharney has quit IRC | 15:01 | |
| openstackgerrit | Nimish proposed openstack/nova: *Nova can incorrectly think an instance is volume backed if instance is booted with "nova boot...",it confuses api.py & assume if image_ref isn't set then theinstance is volume backed. closes Bug: #1501851 https://review.openstack.org/234165 | 15:01 |
| openstack | bug 1501851 in OpenStack Compute (nova) "Nova can incorrectly think an instance is volume backed" [Low,In progress] https://launchpad.net/bugs/1501851 - Assigned to Nimish Joshi (jnimish77) | 15:01 |
| *** eharney_ is now known as eharney | 15:01 | |
| *** garyk1 has quit IRC | 15:01 | |
| kashyap | dansmith: Yeah, I filter locally | 15:02 |
| *** jistr|mtg is now known as jistr | 15:02 | |
| kashyap | For the reasons you mention | 15:02 |
| openstackgerrit | Dan Smith proposed openstack/nova: Fix service_version minimum calculation for compute RPC https://review.openstack.org/234809 | 15:02 |
| *** gongysh has quit IRC | 15:02 | |
| openstackgerrit | Sean Dague proposed openstack/nova: WIP: remove project_id enforcement in url https://review.openstack.org/233076 | 15:02 |
| *** baoli_ has quit IRC | 15:02 | |
| dansmith | sdague: ^ | 15:02 |
| sdague | +2 | 15:03 |
| *** ashishjain has quit IRC | 15:03 | |
| *** Piet has joined #openstack-nova | 15:04 | |
| *** pm90__ has quit IRC | 15:04 | |
| *** Miouge has quit IRC | 15:04 | |
| erlarese | Hi All, if I have a neutron port ID and I want to get the corresponding VIF object from within a nova process, what is the best way to do that? | 15:05 |
| dansmith | thanks | 15:05 |
| dansmith | erlarese: hmm, do we store VIF objects if we're using neutron? | 15:06 |
| dansmith | pretty sure not | 15:06 |
| dansmith | erlarese: are you talking about nova.network.model.VIF or nova.objects.VirtualInterface? | 15:07 |
| *** VikasC_ has quit IRC | 15:07 | |
| *** moshele has joined #openstack-nova | 15:07 | |
| *** pdardeau has joined #openstack-nova | 15:08 | |
| mriedem | we have the network info cache | 15:08 |
| mriedem | which has the serialized vif objects | 15:08 |
| *** armax has joined #openstack-nova | 15:09 | |
| dansmith | right | 15:09 |
| dansmith | when I first read it, I thought he meant looking up VirtualInterface objects from the DB | 15:10 |
| dansmith | s/he | 15:10 |
| mriedem | they might | 15:10 |
| sdague | trying to figure out what it would take to remove project_id from nova urls is making me learn more about our wsgi routing infrastructure, which is probably a good thing | 15:10 |
| mriedem | erlarese: are you eric larese? | 15:10 |
| dansmith | sdague: for a masochist | 15:10 |
| erlarese | mriedem: Hi Matt, yes | 15:10 |
| sdague | dansmith: it's not that bad, mostly it's probably obscured through too many layers of abstractions | 15:11 |
| *** pdardeau has quit IRC | 15:11 | |
| *** pm90_ has joined #openstack-nova | 15:11 | |
| *** jwcroppe has quit IRC | 15:11 | |
| *** emagana has joined #openstack-nova | 15:11 | |
| *** jwcroppe has joined #openstack-nova | 15:12 | |
| erlarese | honestly I'm not sure which object I'm after, I'm trying to find the "preserve_on_delete" flag for a given port as a result of this change: https://bugs.launchpad.net/nova/+bug/1158684 | 15:12 |
| openstack | Launchpad bug 1158684 in Group Based Policy "Pre-created ports get deleted on VM delete" [High,Confirmed] - Assigned to Robert Kukura (rkukura) | 15:12 |
| *** zhangjn has quit IRC | 15:12 | |
| mriedem | erlarese: that's in the info cache | 15:12 |
| *** pm90__ has joined #openstack-nova | 15:13 | |
| mriedem | see how it's used here https://github.com/openstack/nova/blob/master/nova/network/neutronv2/api.py#L1632 | 15:13 |
| *** tpeoples has joined #openstack-nova | 15:13 | |
| *** dave-mccowan has quit IRC | 15:13 | |
| mriedem | the nw info cache is a list of vifs which are dicts | 15:13 |
| *** kmartin has joined #openstack-nova | 15:13 | |
| mriedem | stored here https://github.com/openstack/nova/blob/master/nova/network/neutronv2/api.py#L1718 | 15:14 |
| *** pm90_ has quit IRC | 15:15 | |
| erlarese | ok cool, thanks that helps. one more potentially dumb question, what is the best way to get an instance object using a port ID ? | 15:16 |
| *** jwcroppe has quit IRC | 15:16 | |
| mriedem | it's not stored in the db for neutron, so, | 15:16 |
| mriedem | you probably have to get all instances from the db, then filter on nw info cache until you find the matching vif by port id | 15:16 |
| mriedem | which sucks for performance | 15:17 |
| dansmith | the neutron port should h ave device_id, | 15:17 |
| dansmith | which is the instance uuid | 15:17 |
| mriedem | ah yeah | 15:17 |
| mriedem | that's much better | 15:17 |
| dansmith | yeah :) | 15:17 |
| mriedem | i forgot about how the other services stash instances uuid everywhere for our pleasure | 15:17 |
| *** amrith is now known as _amrith_ | 15:18 | |
| erlarese | ok excellent, thanks all for the help | 15:18 |
| *** tdurakov has joined #openstack-nova | 15:18 | |
| tdurakov | mriedem, are you around? | 15:19 |
| mriedem | umm | 15:19 |
| *** jcoufal has quit IRC | 15:19 | |
| mriedem | depends on what you're going to ask me | 15:19 |
| tdurakov | mriedem, got question about /nova/virt/images.py | 15:19 |
| mriedem | i probably don't have answers, but ok | 15:20 |
| tdurakov | why do we have emu specific methods there? | 15:20 |
| tdurakov | qemu* | 15:20 |
| mriedem | because the libvirt guys wrote it? | 15:20 |
| *** _amrith_ is now known as amrith | 15:20 | |
| mriedem | tdurakov: i'd ask pixelbeat | 15:20 |
| tdurakov | ok, why it is no in libvirt package then? | 15:21 |
| *** LukeHinds has joined #openstack-nova | 15:21 | |
| tdurakov | pixelbeat, hi, could you comment above? | 15:21 |
| *** erhudy has joined #openstack-nova | 15:21 | |
| mriedem | tdurakov: the qemu bits could probably be moved | 15:21 |
| *** weihan has quit IRC | 15:22 | |
| openstackgerrit | jichenjc proposed openstack/nova: WIP: add vnc section into nova configuration https://review.openstack.org/234836 | 15:22 |
| *** weihan has joined #openstack-nova | 15:23 | |
| *** odyssey4me has joined #openstack-nova | 15:23 | |
| tdurakov | mriedem, imo, we should keep such things separate, what do you think? | 15:23 |
| mriedem | in general i agree in separation of duties yes | 15:23 |
| mriedem | but i haven't looked at what is all using the qemu methods in nova.virt.images | 15:23 |
| pixelbeat | tdurakov, Yes ideally libvirt specific stuff shouldn't be there | 15:24 |
| *** weihan has quit IRC | 15:25 | |
| *** mdrabe has quit IRC | 15:25 | |
| *** david-lyle has quit IRC | 15:25 | |
| tdurakov | mriedem, so, let me file bug for that and place it in the right package? | 15:25 |
| *** weihan has joined #openstack-nova | 15:25 | |
| mriedem | tdurakov: sure | 15:25 |
| ndipanov | johnthetubaguy, I see that etherpads are up | 15:26 |
| *** mgoddard has joined #openstack-nova | 15:26 | |
| *** david-lyle has joined #openstack-nova | 15:27 | |
| *** dave-mccowan has joined #openstack-nova | 15:27 | |
| *** tdurakov has quit IRC | 15:27 | |
| *** e0ne has joined #openstack-nova | 15:27 | |
| odyssey4me | hi guys - I'm testing liberty here and have hit something new since yesterday, which I hope is simply a config error: the metadata service is spewing 'CRITICAL nova [-] PasteAppNotFound: Could not load paste app 'metadata' from /etc/nova/api-paste.ini' and isn't working as a result | 15:28 |
| *** yamahata has joined #openstack-nova | 15:30 | |
| *** jdurgin1 has joined #openstack-nova | 15:30 | |
| *** tdurakov has joined #openstack-nova | 15:33 | |
| odyssey4me | a full log is here: http://logs.openstack.org/39/227839/16/check/gate-openstack-ansible-dsvm-commit/1f331e8/logs/aio1_nova_api_metadata_container-1d782190/nova-api-metadata.log | 15:34 |
| *** VW has quit IRC | 15:34 | |
| *** VW has joined #openstack-nova | 15:35 | |
| odyssey4me | lol, never mind - found the issue | 15:35 |
| openstackgerrit | Ryan McNair proposed openstack/nova: Add retry logic for removing device with LibVirt https://review.openstack.org/227851 | 15:35 |
| *** artom has joined #openstack-nova | 15:35 | |
| tdurakov | johnthetubaguy, hi, about multinode ci for nova. I've pushed few patches in tempest, that cover more cases for live-migration and cleans up code in test case. now working on job stability. Have you created sub-team for live-migrations in nova? | 15:36 |
| *** scheuran has quit IRC | 15:37 | |
| *** pumaranikar has quit IRC | 15:38 | |
| *** pumaranikar has joined #openstack-nova | 15:38 | |
| *** VW has quit IRC | 15:39 | |
| *** marzif has quit IRC | 15:39 | |
| *** sahid has quit IRC | 15:40 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: Use version convert methods from oslo_utils.versionutils https://review.openstack.org/203322 | 15:40 |
| *** pumaranikar has quit IRC | 15:40 | |
| *** VW has joined #openstack-nova | 15:41 | |
| *** pumaranikar has joined #openstack-nova | 15:41 | |
| johnthetubaguy | ndipanov: yeah, I have created the links, I am slowing filling them up with the boilerplace stuff right now | 15:42 |
| ndipanov | ok so it was you filling stuff up | 15:43 |
| *** markus_z has quit IRC | 15:43 | |
| ndipanov | cool will add something to the SR-IOV one to make it a biti more useful for te session itself | 15:43 |
| johnthetubaguy | ndipanov: thats awesome thank you | 15:43 |
| *** mdrabe has joined #openstack-nova | 15:44 | |
| *** mgoddard_ has joined #openstack-nova | 15:45 | |
| johnthetubaguy | dansmith: did alaski_out catch you about his ideas around having the instance tables in both the cell and API databases? | 15:45 |
| dansmith | johnthetubaguy: nope | 15:45 |
| johnthetubaguy | dansmith: ah, OK, beginning to think I was imagining my conversation I had with him | 15:45 |
| johnthetubaguy | dansmith: I think the idea was a bit like this... | 15:46 |
| dansmith | johnthetubaguy: yesterday he said he might pop in here in the next few days, so.. | 15:46 |
| *** david-ly_ has joined #openstack-nova | 15:46 | |
| johnthetubaguy | dansmith: ah, yes, he did say he wanted to do that | 15:46 |
| *** david-lyle has quit IRC | 15:47 | |
| johnthetubaguy | dansmith: I think the idea is still only have the instance object in one place, but just do a move operation once the cell mapping is added, and the child cell is picked, rather than having a different structure for instances that are not yet fully created | 15:47 |
| *** mgoddard has quit IRC | 15:48 | |
| *** annegentle has quit IRC | 15:48 | |
| dansmith | johnthetubaguy: yeah, it sounds nice, but it seems problematic to me where relationships are concerned | 15:48 |
| *** annegentle has joined #openstack-nova | 15:48 | |
| dansmith | johnthetubaguy: i.e. having to move a bunch of related data, or not having some related data we expect when it's in the API database | 15:48 |
| dansmith | johnthetubaguy: a move operation could do a create of the thing and all sub-things in the other database I guess, so it's doable, I just.. I dunno.. worry about the details | 15:49 |
| johnthetubaguy | dansmith: yeah, in theory the migrate from cells v1 and migrate between two v2 cells, all become very similar operations, at least at a high level | 15:50 |
| johnthetubaguy | dansmith: but you are totally right, the cross links will be tricky | 15:50 |
| bauzas | I'm also a bit concerned by any locking action it could need | 15:50 |
| bauzas | I mean the move | 15:51 |
| *** oomichi has quit IRC | 15:51 | |
| bauzas | because that needs a bit of synchronization, nope ? | 15:51 |
| dansmith | yeah | 15:51 |
| *** jwcroppe has joined #openstack-nova | 15:51 | |
| johnthetubaguy | bauzas: should be lock free, with a tiny state race, if I thought that through correctly | 15:51 |
| bauzas | or we could face all the pain that we have with cells v1 | 15:51 |
| bauzas | johnthetubaguy: sure we could be eventually consistent | 15:51 |
| johnthetubaguy | I was thinking presence of the mapping means you look in the child cell | 15:51 |
| johnthetubaguy | no, there is only ever one copy alive at once | 15:52 |
| dansmith | it seems like just persisting the request spec at the top end makes it easier, because then you don't have to move things, and creating an instance involves creating the spec, then the instance in the cell db, and things are always where you expect | 15:52 |
| dansmith | johnthetubaguy: the problem is if you fail in the middle, you might end up with the instance in both or neither place right? | 15:52 |
| *** pumaranikar has quit IRC | 15:52 | |
| bauzas | johnthetubaguy: but that still needs a bit of implementation for making sure both instances couldn't be trampled | 15:52 |
| dansmith | johnthetubaguy: moving between DBs is harder than moving between tables | 15:52 |
| bauzas | +1 | 15:52 |
| johnthetubaguy | I guess the thing at the top is explicitly read-only, which helps | 15:52 |
| *** ildikov has quit IRC | 15:52 | |
| *** ijw has joined #openstack-nova | 15:53 | |
| *** pm90__ has quit IRC | 15:53 | |
| bauzas | I thought an API conductor could be having this kind of indirection API | 15:53 |
| *** kiran-r has joined #openstack-nova | 15:53 | |
| bauzas | ie. either lookup at this, or that | 15:53 |
| johnthetubaguy | it does replace that | 15:53 |
| *** kiran-r has quit IRC | 15:53 | |
| johnthetubaguy | I guess I am describing this very badly | 15:53 |
| bauzas | johnthetubaguy: do you know if alaski_out already said if he could be remote for the cells session ? | 15:54 |
| johnthetubaguy | its basically using the instance table, rather than a new request spec table for the temporary entry in the top DB | 15:54 |
| *** david-ly_ is now known as david-lyle | 15:54 | |
| dansmith | johnthetubaguy: I'm also doing a bunch of other things at the moment.. can we talk about this in tokyo? | 15:54 |
| johnthetubaguy | dansmith: yes | 15:54 |
| bauzas | for the cells session, yes! | 15:54 |
| dansmith | right | 15:54 |
| bauzas | because that's IMHO the main problem that we need to fix for Mitaka | 15:54 |
| johnthetubaguy | bauzas: he is planning to be remove, I have emailed him about picking a time, so I can juggle that around as needed | 15:55 |
| bauzas | ie. where to place data | 15:55 |
| johnthetubaguy | arg | 15:55 |
| bauzas | yeah remo | 15:55 |
| johnthetubaguy | s/remove/remote/ | 15:55 |
| bauzas | *te* | 15:55 |
| bauzas | ;) | 15:55 |
| bauzas | okay, let us know, because alaski_out made a huge teaser | 15:55 |
| bauzas | by saying he was about thinking that, without explaining what :) | 15:56 |
| bauzas | :D | 15:56 |
| johnthetubaguy | ah, yeah, I think thats the rough idea, above, but I suspect he has refined that somewhat since then | 15:56 |
| * bauzas goes back to specs | 15:56 | |
| *** apoorvad has joined #openstack-nova | 15:58 | |
| *** ildikov has joined #openstack-nova | 15:59 | |
| *** mgoddard_ has quit IRC | 15:59 | |
| *** wznoinsk2 is now known as wznoinsk | 16:00 | |
| *** ihrachys has joined #openstack-nova | 16:00 | |
| *** mrkz has quit IRC | 16:00 | |
| *** mrkz has joined #openstack-nova | 16:00 | |
| *** jwcroppe has quit IRC | 16:00 | |
| *** jwcroppe has joined #openstack-nova | 16:00 | |
| *** haomaiwang has quit IRC | 16:01 | |
| *** pkholkin has quit IRC | 16:01 | |
| *** haomaiwang has joined #openstack-nova | 16:01 | |
| *** jcoufal has joined #openstack-nova | 16:02 | |
| *** su_zhang has quit IRC | 16:03 | |
| *** shangxdy has quit IRC | 16:03 | |
| *** mgoddard has joined #openstack-nova | 16:05 | |
| *** jistr has quit IRC | 16:06 | |
| *** jlanoux has quit IRC | 16:06 | |
| cfriesen | has anyone ever looked at getting rid of nova.compute.resources.vcpu.VCPU? It seems wierd to track vcpus differently from all the other resources. Was that a failed attempt at unifying the resource tracking or something? | 16:06 |
| *** matrohon has quit IRC | 16:06 | |
| ndipanov | cfriesen, exactly | 16:07 |
| johnthetubaguy | cfriesen: that sounds like a hang over from the extensible resource tracker, yeah | 16:07 |
| ndipanov | the so called extensible resource tracker | 16:07 |
| johnthetubaguy | the plan was to move all resources into that new model, so its easier to add new resource types | 16:08 |
| bauzas | cfriesen: feel free to submit a change and I'll be happily +1'ing it | 16:08 |
| *** pumaranikar has joined #openstack-nova | 16:08 | |
| cfriesen | bauzas: :) | 16:08 |
| mriedem | PaulMurray: ^ | 16:08 |
| bauzas | IMHO there are zero reasons to keep that until we somehow respawn an effort for another way to provide resourcres | 16:08 |
| *** ccesario_ has quit IRC | 16:09 | |
| *** subscope has quit IRC | 16:09 | |
| bauzas | given all the work we plan with resource-objects BP, and what could be discussed during the summit, that sounds a nice moveback until we have better ways to provide resources | 16:09 |
| johnthetubaguy | cfriesen: bauzas: we have two sessions at the summit on resource modelling, it would be good to pause until after that | 16:09 |
| *** kmartin has quit IRC | 16:09 | |
| bauzas | johnthetubaguy: honestly, this is dead code | 16:09 |
| cfriesen | bauzas: isnt' it currently used for vcpus though? | 16:10 |
| bauzas | just a reminiscent of something which was reverted | 16:10 |
| *** kmartin has joined #openstack-nova | 16:10 | |
| bauzas | cfriesen: yup, but only for it | 16:10 |
| *** baoli has joined #openstack-nova | 16:10 | |
| bauzas | and we actually decided to stop adding more | 16:10 |
| *** jdurgin1 has quit IRC | 16:10 | |
| bauzas | so the thing is, that's pretty confusing for people coming in and looking to the wonderful world of RT | 16:11 |
| *** testing has joined #openstack-nova | 16:11 | |
| bauzas | because they don't get that tribal knowledge | 16:11 |
| tdurakov | johnthetubaguy, ping | 16:11 |
| bauzas | and honestly, I can't hardly explain why we keep it | 16:11 |
| *** testing is now known as Guest81146 | 16:11 | |
| johnthetubaguy | bauzas: I quite liked that direction, honestly | 16:12 |
| johnthetubaguy | tdurakov: can I help? | 16:12 |
| bauzas | johnthetubaguy: you mean the ERT? | 16:13 |
| *** sneti_ has joined #openstack-nova | 16:13 | |
| johnthetubaguy | bauzas: yep | 16:13 |
| tdurakov | johnthetubaguy, hi, left a message earlier, have you seen it? | 16:13 |
| openstackgerrit | Akhila Chetlapalle proposed openstack/python-novaclient: Support to boot a VM with network name https://review.openstack.org/225326 | 16:13 |
| bauzas | johnthetubaguy: so the thing is, there is the resource-objects BP that jaypipes makes me loving | 16:13 |
| bauzas | johnthetubaguy: we can surely wait for it to be implemented because it would refactor that stuff | 16:14 |
| johnthetubaguy | bauzas: agreed thats a better direction, but I see that as a refactor of the ERT interface | 16:14 |
| *** ihrachys has quit IRC | 16:14 | |
| bauzas | johnthetubaguy: but in the meantime, it's really hard to consider the current situation as the idea lone | 16:14 |
| *** sneti has quit IRC | 16:15 | |
| bauzas | johnthetubaguy: okay, then your idea to wait for the resource sessions is okay with me | 16:15 |
| johnthetubaguy | bauzas: I would rather know its replacement before we remove that, thats all | 16:15 |
| bauzas | johnthetubaguy: we can defer a decision after the talks | 16:15 |
| bauzas | johnthetubaguy: agreed, makes sense | 16:15 |
| johnthetubaguy | tdurakov: oh, sorry, been slogging through all the summit session etherpads | 16:15 |
| johnthetubaguy | tdurakov: anyone can create a subteam, feel free to organise something with PaulMurray | 16:16 |
| *** penick has joined #openstack-nova | 16:16 | |
| tdurakov | johnthetubaguy, ok, sure - do we have some specific process for that?) | 16:16 |
| johnthetubaguy | tdurakov: I don't want to be a blocker for creating a subteam, but I am happy to help create one if thats useful | 16:16 |
| *** ihrachys has joined #openstack-nova | 16:17 | |
| johnthetubaguy | tdurakov: its really up to the subteam, there are etherpads and wiki pages, but whatever works best for you all is the real answer | 16:17 |
| *** Sree has joined #openstack-nova | 16:17 | |
| johnthetubaguy | tdurakov: at this point, I think its worth having an entry in the code review and spec review etherpads, then agreeing how to move forward at the summit | 16:18 |
| *** romainh has left #openstack-nova | 16:19 | |
| tdurakov | johnthetubaguy, ok:) | 16:19 |
| *** odyssey4me has left #openstack-nova | 16:19 | |
| tdurakov | johnthetubaguy, can i also ask you to review this patch again? https://review.openstack.org/#/c/227897/ | 16:19 |
| *** Murali has quit IRC | 16:19 | |
| tdurakov | PaulMurray, hi, are you around? | 16:19 |
| *** ccesario_ has joined #openstack-nova | 16:21 | |
| *** rotbeard has joined #openstack-nova | 16:23 | |
| johnthetubaguy | tdurakov: we should talk through that patch, it seems further away in this latest patchest | 16:24 |
| johnthetubaguy | tdurakov: just going back to what we want, if self._teardown_network_on_destination raises an exception, we still want to attempt self._rollback_driver_on_destination | 16:24 |
| johnthetubaguy | tdurakov: why not add self._rollback_driver_on_destination into a finally clause, would that do what we want? | 16:25 |
| *** nic has joined #openstack-nova | 16:25 | |
| tdurakov | johnthetubaguy, what if driver rollback fails to | 16:26 |
| johnthetubaguy | tdurakov: thats OK I think, what problem are you seeing in that case? | 16:27 |
| tdurakov | it shadow network fail, no? | 16:27 |
| johnthetubaguy | tdurakov: thats OK, given you have @wrap_exception and @wrap_instance_fault both errors will get reported, I think you only actually need those decorators on _rollback_driver_on_destination | 16:28 |
| *** r-daneel has joined #openstack-nova | 16:28 | |
| *** kumar has joined #openstack-nova | 16:28 | |
| *** weihan has quit IRC | 16:29 | |
| tdurakov | johnthetubaguy, so, about decorators. @wrap_exception and @wrap_instance_fault should be applied only to _rollback_driver_on_destination? | 16:31 |
| *** sudipto has quit IRC | 16:31 | |
| *** yassine has quit IRC | 16:31 | |
| tdurakov | and public rollback_live_migration_at_destination i suppose | 16:32 |
| *** electrocucaracha has quit IRC | 16:32 | |
| johnthetubaguy | tdurakov: I am not sure now, just thinking that through | 16:34 |
| *** Guest81146 has quit IRC | 16:37 | |
| *** testing_ has joined #openstack-nova | 16:37 | |
| *** electrocucaracha has joined #openstack-nova | 16:37 | |
| *** ccarmack has left #openstack-nova | 16:37 | |
| *** Murali has joined #openstack-nova | 16:38 | |
| *** pm90_ has joined #openstack-nova | 16:38 | |
| *** NightKhaos_ has joined #openstack-nova | 16:39 | |
| *** electrocucaracha has quit IRC | 16:39 | |
| *** jasondotstar_ has joined #openstack-nova | 16:39 | |
| *** pm90__ has joined #openstack-nova | 16:41 | |
| openstackgerrit | John Garbutt proposed openstack/nova: Wrap setup_networks_on_host with try/except https://review.openstack.org/227897 | 16:42 |
| *** electrocucaracha has joined #openstack-nova | 16:43 | |
| *** Murali has quit IRC | 16:43 | |
| johnthetubaguy | tdurakov: I have attempted a code sketch of an idea, does that work? https://review.openstack.org/#/c/227897/6/nova/compute/manager.py,cm | 16:43 |
| *** thomasem_ has joined #openstack-nova | 16:44 | |
| *** pm90_ has quit IRC | 16:44 | |
| *** cburgess_ has joined #openstack-nova | 16:44 | |
| openstackgerrit | John Garbutt proposed openstack/nova: Wrap setup_networks_on_host with try/except https://review.openstack.org/227897 | 16:45 |
| *** smurke_ has joined #openstack-nova | 16:46 | |
| johnthetubaguy | tdurakov: so my code sketch was rubbish, so I updated it | 16:46 |
| *** smurke has quit IRC | 16:46 | |
| *** Sree has quit IRC | 16:46 | |
| tdurakov | johnthetubaguy, need to check | 16:46 |
| *** zzzeek_ has joined #openstack-nova | 16:46 | |
| cfriesen | question...in virt/libvirt/storage/lvm.py, in _zero_volume(), why do we use "oflag=direct" and "conf=fdatasync" on each iteration? Wouldn't it be enough to skip the direct I/O and call fdatasync() on the file once at the end? | 16:47 |
| *** rfolco has quit IRC | 16:47 | |
| johnthetubaguy | tdurakov: that should mean both methods always get run, the first exception is always logged just incase its hidden by an exception in the finally clause, etc | 16:47 |
| *** rfolco has joined #openstack-nova | 16:47 | |
| tdurakov | johnthetubaguy, so, if we suppose that driver exception has more priority then network exception, than that would be ok, one thing to change:) | 16:48 |
| tdurakov | no need to create private methods then | 16:48 |
| *** aix has quit IRC | 16:48 | |
| johnthetubaguy | tdurakov: yeah, I think its worth removing the private methods | 16:48 |
| johnthetubaguy | tdurakov: the logging should cover both equally, which seems cool | 16:49 |
| *** e0ne_ has joined #openstack-nova | 16:49 | |
| *** ihrachys has quit IRC | 16:50 | |
| *** jwcroppe has quit IRC | 16:50 | |
| *** mrkz has quit IRC | 16:50 | |
| *** e0ne has quit IRC | 16:50 | |
| *** wznoinsk has quit IRC | 16:50 | |
| *** xyang1 has quit IRC | 16:50 | |
| *** alex_klimov has quit IRC | 16:50 | |
| *** jichen has quit IRC | 16:50 | |
| *** lykinsbd_ has quit IRC | 16:50 | |
| *** jaypipes has quit IRC | 16:50 | |
| *** zzzeek has quit IRC | 16:50 | |
| *** cburgess has quit IRC | 16:50 | |
| *** sbadia has quit IRC | 16:50 | |
| *** mirrorbox has quit IRC | 16:50 | |
| *** sileht has quit IRC | 16:50 | |
| *** Vek has quit IRC | 16:50 | |
| *** NightKhaos has quit IRC | 16:50 | |
| *** jdurgin has quit IRC | 16:50 | |
| *** ekhugen has quit IRC | 16:50 | |
| *** thomasem has quit IRC | 16:50 | |
| *** jasondotstar has quit IRC | 16:50 | |
| *** amrith has quit IRC | 16:50 | |
| *** dulek has quit IRC | 16:50 | |
| *** edmondk has quit IRC | 16:50 | |
| *** jcook has quit IRC | 16:50 | |
| *** rlrossit has quit IRC | 16:50 | |
| *** ociuhandu has quit IRC | 16:50 | |
| *** bapalm has quit IRC | 16:50 | |
| *** rdopiera has quit IRC | 16:50 | |
| *** unicell1 has quit IRC | 16:50 | |
| *** lifeless has quit IRC | 16:50 | |
| *** hogepodge has quit IRC | 16:50 | |
| *** pedroalvarez has quit IRC | 16:50 | |
| *** mtreinish has quit IRC | 16:50 | |
| *** isq_ has quit IRC | 16:50 | |
| *** baffle has quit IRC | 16:50 | |
| *** _cerberus_ has quit IRC | 16:50 | |
| *** mleroy has quit IRC | 16:50 | |
| *** NightKhaos_ is now known as NightKhaos | 16:50 | |
| * johnthetubaguy feel like I am on the wrong side of that split... | 16:50 | |
| dansmith | johnthetubaguy: them I am too I guess | 16:51 |
| *** _cerberus_ has joined #openstack-nova | 16:51 | |
| raildo | lol | 16:52 |
| cfriesen | dansmith: johnthetubaguy: either of you know anything about the sparse volume support in virt/libvirt/storage/lvm? I'm suspicious that it's buggy...it seems to unconditionally allocate 64MB of space regardless of the configured volume size. The code calls it "preallocated_space", but looking at the lvm docs it seems like that's all the "real" space for that volume. | 16:53 |
| *** moshele has quit IRC | 16:53 | |
| *** phschwartz is now known as phschwartz_aw | 16:54 | |
| dansmith | cfriesen: no | 16:54 |
| dansmith | cfriesen: git blame | 16:54 |
| *** baoli_ has joined #openstack-nova | 16:54 | |
| *** e0ne_ is now known as e0ne | 16:55 | |
| *** pedroalvarez has joined #openstack-nova | 16:55 | |
| johnthetubaguy | dansmith: yeah, I thought it looked like everyone at first, I got carried away | 16:55 |
| dansmith | johnthetubaguy: by the way, minor oopsie in this code, if you could +A: https://review.openstack.org/#/c/234809/ | 16:56 |
| cfriesen | dansmith: looks like it dates back to the initial lvm-disk-images blueprint in folsom | 16:56 |
| johnthetubaguy | cfriesen: afraid there is very few bits of the libvirt driver know super well, I know the xenapi side better I am afriad | 16:57 |
| dansmith | johnthetubaguy: with that, the grenade partial job auto-selects compute rpc version pins :D | 16:57 |
| johnthetubaguy | dansmith: oh, very sweet | 16:57 |
| *** moshele has joined #openstack-nova | 16:58 | |
| dansmith | johnthetubaguy: http://logs.openstack.org/89/233289/10/check/gate-grenade-dsvm-partial-ncpu/9cd1493/logs/new/screen-n-cond.txt.gz#_2015-10-14_16_05_52_783 | 16:58 |
| *** aysyd has quit IRC | 16:58 | |
| *** mrkz has joined #openstack-nova | 16:59 | |
| *** rlrossit has joined #openstack-nova | 17:00 | |
| *** ociuhandu has joined #openstack-nova | 17:00 | |
| *** bapalm has joined #openstack-nova | 17:00 | |
| *** rdopiera has joined #openstack-nova | 17:00 | |
| *** lifeless has joined #openstack-nova | 17:00 | |
| *** hogepodge has joined #openstack-nova | 17:00 | |
| *** mleroy has joined #openstack-nova | 17:00 | |
| *** mtreinish has joined #openstack-nova | 17:00 | |
| *** isq_ has joined #openstack-nova | 17:00 | |
| *** baffle has joined #openstack-nova | 17:00 | |
| openstackgerrit | Timofey Durakov proposed openstack/nova: Wrap setup_networks_on_host with try/except https://review.openstack.org/227897 | 17:00 |
| tdurakov | johnthetubaguy, done | 17:00 |
| *** haomaiwang has quit IRC | 17:01 | |
| *** moshele has quit IRC | 17:01 | |
| *** haomaiwang has joined #openstack-nova | 17:01 | |
| johnthetubaguy | tdurakov: that looks good to me now, sorry about all the debate there, hopefully others like that now too | 17:02 |
| tdurakov | johnthetubaguy, oops again))) | 17:02 |
| tdurakov | duplicate logging | 17:03 |
| tdurakov | so warning or error for that? | 17:03 |
| johnthetubaguy | tdurakov: where does it get logged again? | 17:03 |
| johnthetubaguy | tdurakov: the error and debug is intentional there | 17:03 |
| tdurakov | johnthetubaguy, copy pasted that LOG.error and LOG.debug... | 17:04 |
| tdurakov | so what would be correct for that? | 17:04 |
| johnthetubaguy | tdurakov: its because of the exc_info | 17:04 |
| tdurakov | so, it's ok? | 17:05 |
| *** xyang1 has joined #openstack-nova | 17:05 | |
| johnthetubaguy | tdurakov: I think so, its because we do not allow exe_info=True in things higher that debug, ideally | 17:05 |
| johnthetubaguy | at least, that was the plan at some point | 17:05 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova-specs: spec document for sriov-physical-function-passthrough blueprint https://review.openstack.org/212472 | 17:06 |
| tdurakov | johnthetubaguy, ok, thank you for helping with this:) | 17:06 |
| *** ihrachys has joined #openstack-nova | 17:06 | |
| johnthetubaguy | tdurakov: having said that, I can't find other examples of that double logging thing any more :S | 17:07 |
| *** mrkz has quit IRC | 17:07 | |
| *** rlrossit has quit IRC | 17:07 | |
| *** ociuhandu has quit IRC | 17:07 | |
| *** bapalm has quit IRC | 17:07 | |
| *** rdopiera has quit IRC | 17:07 | |
| *** lifeless has quit IRC | 17:07 | |
| *** hogepodge has quit IRC | 17:07 | |
| *** mtreinish has quit IRC | 17:07 | |
| *** isq_ has quit IRC | 17:07 | |
| *** baffle has quit IRC | 17:07 | |
| *** mleroy has quit IRC | 17:07 | |
| *** Vek has joined #openstack-nova | 17:07 | |
| *** dulek has joined #openstack-nova | 17:07 | |
| *** jaypipes has joined #openstack-nova | 17:07 | |
| *** lykinsbd has joined #openstack-nova | 17:07 | |
| *** mirrorbox has joined #openstack-nova | 17:07 | |
| *** mirrorbox has joined #openstack-nova | 17:07 | |
| *** jdurgin has joined #openstack-nova | 17:07 | |
| *** amrith has joined #openstack-nova | 17:07 | |
| *** edmondk has joined #openstack-nova | 17:08 | |
| *** jcook has joined #openstack-nova | 17:08 | |
| *** wznoinsk has joined #openstack-nova | 17:08 | |
| *** rfolco has quit IRC | 17:08 | |
| *** sileht has joined #openstack-nova | 17:08 | |
| *** ildikov has quit IRC | 17:08 | |
| *** VW has quit IRC | 17:08 | |
| *** su_zhang has joined #openstack-nova | 17:08 | |
| *** derekh has quit IRC | 17:08 | |
| *** kumar has quit IRC | 17:08 | |
| *** tdurakov has quit IRC | 17:09 | |
| *** VW has joined #openstack-nova | 17:09 | |
| *** ekhugen has joined #openstack-nova | 17:10 | |
| *** mrkz has joined #openstack-nova | 17:10 | |
| *** rlrossit has joined #openstack-nova | 17:10 | |
| *** ociuhandu has joined #openstack-nova | 17:10 | |
| *** bapalm has joined #openstack-nova | 17:10 | |
| *** rdopiera has joined #openstack-nova | 17:10 | |
| *** lifeless has joined #openstack-nova | 17:10 | |
| *** hogepodge has joined #openstack-nova | 17:10 | |
| *** mleroy has joined #openstack-nova | 17:10 | |
| *** mtreinish has joined #openstack-nova | 17:10 | |
| *** isq_ has joined #openstack-nova | 17:10 | |
| *** baffle has joined #openstack-nova | 17:10 | |
| *** ekhugen has quit IRC | 17:10 | |
| *** ekhugen has joined #openstack-nova | 17:10 | |
| *** ctrath has quit IRC | 17:10 | |
| smurke_ | to delete a vm is nova-compute servcice necessary to be running ? | 17:11 |
| *** sbadia has joined #openstack-nova | 17:11 | |
| *** ijw has quit IRC | 17:11 | |
| *** Alexus has quit IRC | 17:12 | |
| *** aysyd has joined #openstack-nova | 17:12 | |
| *** VW has quit IRC | 17:13 | |
| *** ctrath has joined #openstack-nova | 17:13 | |
| *** manas has joined #openstack-nova | 17:13 | |
| johnthetubaguy | smurke_: we have code to work around that, there might be a problem is when the service id down but its not yet detected as down | 17:14 |
| *** yamahata has quit IRC | 17:14 | |
| *** rfolco has joined #openstack-nova | 17:14 | |
| *** tjones has joined #openstack-nova | 17:14 | |
| *** lucasagomes is now known as lucas-dinner | 17:14 | |
| *** amrith is now known as _amrith_ | 17:14 | |
| *** claudiub has quit IRC | 17:15 | |
| *** electroc_ has joined #openstack-nova | 17:15 | |
| smurke_ | johnthetubaguy: yeah i was figuring it out when i disabled my nova-compute service and ran dlete vm command it still worked fine | 17:15 |
| *** jcoufal has quit IRC | 17:15 | |
| *** baoli has quit IRC | 17:15 | |
| smurke_ | johnthetubaguy: should i assume it as a bug ? | 17:15 |
| *** browne has quit IRC | 17:15 | |
| johnthetubaguy | smurke_: thats intentional | 17:16 |
| johnthetubaguy | smurke_: now a disabled host can still be running, which is a slightly confusing case | 17:16 |
| johnthetubaguy | smurke_: at least if I understand whats you are saying is happening correctly its intentional | 17:17 |
| *** unicell has joined #openstack-nova | 17:17 | |
| *** achanda has joined #openstack-nova | 17:17 | |
| *** moshele has joined #openstack-nova | 17:17 | |
| smurke_ | johnthetubaguy: i am new to openstack, just wanted to get clarified with the behavior. Thanx for the info ! | 17:18 |
| *** ccarmack has joined #openstack-nova | 17:18 | |
| *** ctrath has joined #openstack-nova | 17:18 | |
| *** gszasz has quit IRC | 17:18 | |
| *** vilobhmm11 has joined #openstack-nova | 17:18 | |
| *** electrocucaracha has quit IRC | 17:18 | |
| johnthetubaguy | smurke_: no worries, #openstack should be better at helping with more usage things, this is more a dev channel | 17:19 |
| *** pixelbeat has quit IRC | 17:19 | |
| johnthetubaguy | smurke_: fwiw, there is logic that when a compute node stats up, it deletes instances that were fast deleted while it was down, etc, if my memory is correct | 17:19 |
| johnthetubaguy | that might be configurable actually... | 17:19 |
| smurke_ | johnthetubaguy: Thanku ! | 17:21 |
| *** mriedem has quit IRC | 17:21 | |
| *** phschwartz_aw is now known as phschwartz | 17:23 | |
| openstackgerrit | Artom Lifshitz proposed openstack/nova: WIP: vNIC tagging https://review.openstack.org/234904 | 17:25 |
| *** rfolco has quit IRC | 17:26 | |
| *** rfolco has joined #openstack-nova | 17:27 | |
| *** nkrinner has quit IRC | 17:28 | |
| *** baoli has joined #openstack-nova | 17:28 | |
| *** baoli has quit IRC | 17:28 | |
| *** patrickeast_ has joined #openstack-nova | 17:30 | |
| *** hemna has quit IRC | 17:31 | |
| *** baoli has joined #openstack-nova | 17:32 | |
| *** hemna has joined #openstack-nova | 17:34 | |
| *** baoli_ has quit IRC | 17:35 | |
| *** pixelbeat has joined #openstack-nova | 17:36 | |
| *** Marga_ has quit IRC | 17:38 | |
| *** Marga_ has joined #openstack-nova | 17:38 | |
| *** yamahata has joined #openstack-nova | 17:38 | |
| *** testing_ has quit IRC | 17:39 | |
| *** testing has joined #openstack-nova | 17:39 | |
| *** jasondotstar_ is now known as jasondotstar | 17:39 | |
| *** testing is now known as Guest39846 | 17:39 | |
| *** moshele has quit IRC | 17:39 | |
| *** danpb has quit IRC | 17:39 | |
| *** mriedem has joined #openstack-nova | 17:40 | |
| *** Guest39846 has quit IRC | 17:40 | |
| vilobhmm11 | johnthetubaguy : ping | 17:40 |
| vilobhmm11 | 17:40 | |
| *** PaulMurray has quit IRC | 17:40 | |
| *** PaulMurray has joined #openstack-nova | 17:41 | |
| *** PaulMurray has quit IRC | 17:43 | |
| *** pixelbeat has quit IRC | 17:44 | |
| *** Marga_ has quit IRC | 17:44 | |
| *** moshele has joined #openstack-nova | 17:44 | |
| *** nikhil_k has quit IRC | 17:45 | |
| *** rlrossit has quit IRC | 17:47 | |
| *** thumpba has joined #openstack-nova | 17:51 | |
| *** rlrossit has joined #openstack-nova | 17:52 | |
| *** njirap has quit IRC | 17:53 | |
| *** jogo has quit IRC | 17:53 | |
| *** VW has joined #openstack-nova | 17:54 | |
| *** garyk has joined #openstack-nova | 17:54 | |
| *** jogo has joined #openstack-nova | 17:55 | |
| *** browne has joined #openstack-nova | 17:55 | |
| *** tjones has quit IRC | 17:55 | |
| *** moshele has quit IRC | 17:57 | |
| *** VW_ has joined #openstack-nova | 17:58 | |
| *** wasmum- has joined #openstack-nova | 17:58 | |
| *** VW_ has quit IRC | 17:58 | |
| *** su_zhang_ has joined #openstack-nova | 17:58 | |
| *** VW_ has joined #openstack-nova | 17:59 | |
| *** VW has quit IRC | 17:59 | |
| *** _amrith_ is now known as amrith | 18:00 | |
| *** baoli has quit IRC | 18:00 | |
| *** Marga_ has joined #openstack-nova | 18:00 | |
| *** haomaiwang has quit IRC | 18:01 | |
| *** READ10 has quit IRC | 18:01 | |
| *** haomaiwang has joined #openstack-nova | 18:01 | |
| *** su_zhang has quit IRC | 18:01 | |
| vilobhmm11 | dansmith : ping | 18:02 |
| *** smurke_ has quit IRC | 18:02 | |
| *** jogo has quit IRC | 18:02 | |
| *** Murali has joined #openstack-nova | 18:02 | |
| *** moshele has joined #openstack-nova | 18:04 | |
| *** kmartin_ has joined #openstack-nova | 18:04 | |
| *** kmartin has quit IRC | 18:05 | |
| *** jogo has joined #openstack-nova | 18:05 | |
| *** moshele has quit IRC | 18:05 | |
| *** sajeesh has joined #openstack-nova | 18:06 | |
| *** LukeHinds has quit IRC | 18:06 | |
| openstackgerrit | Artom Lifshitz proposed openstack/nova: WIP: BLock device tagging https://review.openstack.org/234917 | 18:08 |
| *** edmondsw has quit IRC | 18:09 | |
| *** smurke has joined #openstack-nova | 18:13 | |
| *** ijw has joined #openstack-nova | 18:14 | |
| *** nikhil has joined #openstack-nova | 18:15 | |
| *** e0ne has quit IRC | 18:15 | |
| *** xyang1 has quit IRC | 18:17 | |
| *** ijw_ has joined #openstack-nova | 18:18 | |
| *** ijw has quit IRC | 18:19 | |
| openstackgerrit | Artom Lifshitz proposed openstack/nova: WIP: Block device tagging https://review.openstack.org/234917 | 18:19 |
| *** haomaiwang has quit IRC | 18:19 | |
| *** haomaiwang has joined #openstack-nova | 18:20 | |
| *** e0ne has joined #openstack-nova | 18:20 | |
| *** Murali has quit IRC | 18:21 | |
| openstackgerrit | Artom Lifshitz proposed openstack/nova: WIP: Block device tagging https://review.openstack.org/234917 | 18:21 |
| dansmith | vilobhmm11: ack | 18:21 |
| *** eglynn has quit IRC | 18:22 | |
| *** ihrachys has quit IRC | 18:22 | |
| *** sneti_ has quit IRC | 18:23 | |
| *** sneti has joined #openstack-nova | 18:23 | |
| bauzas | vilobhmm11: FWIW, currently reviewing your spec, but I think that since it's a refactoring effort, you should not provide any usecase and clean your spec for that | 18:24 |
| *** jichen has joined #openstack-nova | 18:24 | |
| bauzas | currently reviewing the last sections before uploading the review | 18:24 |
| vilobhmm11 | bauzas : ok cool…your comments have been helpful | 18:24 |
| bauzas | vilobhmm11: tbh, I don't see a need to discuss on ZK or MC in the spec | 18:25 |
| bauzas | vilobhmm11: because you're just changing the interface | 18:25 |
| *** achanda has quit IRC | 18:25 | |
| vilobhmm11 | bauzas : I agree and as I clearly mentioned it will be covered in sperate spec | 18:25 |
| bauzas | not the drivers | 18:25 |
| openstackgerrit | Artom Lifshitz proposed openstack/nova: WIP: Block device tagging https://review.openstack.org/234917 | 18:25 |
| vilobhmm11 | just giving a reference | 18:25 |
| bauzas | okay, lemme upload my comments | 18:25 |
| *** patrickeast_ has quit IRC | 18:26 | |
| vilobhmm11 | bauzas : ok | 18:26 |
| vilobhmm11 | dansmith : just wanted to request you to have a look at https://review.openstack.org/#/c/222423/8 ; hides in the corner after this request | 18:27 |
| openstackgerrit | Artom Lifshitz proposed openstack/nova: WIP: vNIC tagging https://review.openstack.org/234904 | 18:27 |
| *** edmondsw has joined #openstack-nova | 18:28 | |
| openstackgerrit | Chuck Carmack proposed openstack/nova-specs: Allow user to set and retrieve the server Description https://review.openstack.org/228908 | 18:28 |
| dansmith | vilobhmm11: sigh :( | 18:28 |
| *** thumpba has quit IRC | 18:29 | |
| *** jborak has joined #openstack-nova | 18:31 | |
| *** mehip has joined #openstack-nova | 18:34 | |
| *** annegentle has quit IRC | 18:35 | |
| *** david-lyle has quit IRC | 18:35 | |
| *** annegentle has joined #openstack-nova | 18:36 | |
| openstackgerrit | Clif Houck proposed openstack/nova: Ensure _cleanup_deploy is called even if _unprovision raises https://review.openstack.org/234468 | 18:37 |
| dansmith | garyk: question for you here: https://review.openstack.org/#/c/195662/8 | 18:37 |
| *** jogo has quit IRC | 18:37 | |
| *** matrohon has joined #openstack-nova | 18:37 | |
| *** mehip has quit IRC | 18:38 | |
| *** pm90_ has joined #openstack-nova | 18:38 | |
| *** david-lyle has joined #openstack-nova | 18:39 | |
| *** mehip has joined #openstack-nova | 18:39 | |
| *** mehip has quit IRC | 18:40 | |
| *** pm90__ has quit IRC | 18:40 | |
| *** jogo has joined #openstack-nova | 18:40 | |
| *** david-ly_ has joined #openstack-nova | 18:40 | |
| jborak | I'm poking around the nova database and even though I have 50+ instances with fixed IPs (and some floating IPs) there are no entries in the fixed_ips table in the nova db. | 18:41 |
| jborak | Does anyone know where the records for fixed IPs are stored in the nova db? | 18:41 |
| dansmith | jborak: are you using neutron? | 18:42 |
| jborak | yup | 18:42 |
| dansmith | jborak: they're in neutron's database :) | 18:42 |
| openstackgerrit | Andrei V. Ostapenko proposed openstack/nova: Fixes bug with notify_decorator bad getattr default value https://review.openstack.org/223890 | 18:43 |
| openstackgerrit | Andrei V. Ostapenko proposed openstack/nova: Fixes bug with notify_decorator bad getattr default value https://review.openstack.org/223890 | 18:43 |
| jborak | dansmith, lol, I'm trying to link the floating_ips to the instances themselves | 18:43 |
| *** david-ly_ has quit IRC | 18:43 | |
| openstackgerrit | Andrei V. Ostapenko proposed openstack/nova: Fixes bug with notify_decorator bad getattr default value https://review.openstack.org/223890 | 18:43 |
| dansmith | jborak: well, the link is still in neutron's db.. the device id is the instance uuid | 18:43 |
| jborak | so I see in neutron.floating_ips there is a tenant_id which helps me narrow down the project. | 18:43 |
| dansmith | jborak: we cache some of that information in our info_cache, but it's not easy to get at | 18:43 |
| *** david-ly_ has joined #openstack-nova | 18:44 | |
| jborak | dansmith, ah, I'm going to check that now, thanks for the tip | 18:44 |
| *** david-lyle has quit IRC | 18:44 | |
| *** lucas-dinner has quit IRC | 18:44 | |
| *** pixelbeat has joined #openstack-nova | 18:44 | |
| *** david-ly_ is now known as david-lyle | 18:45 | |
| *** david-lyle has quit IRC | 18:45 | |
| *** su_zhang_ has quit IRC | 18:45 | |
| *** david-lyle has joined #openstack-nova | 18:47 | |
| *** sneti has quit IRC | 18:47 | |
| *** sneti has joined #openstack-nova | 18:47 | |
| *** apuimedo has quit IRC | 18:50 | |
| openstackgerrit | Chuck Carmack proposed openstack/nova-specs: Allow user to set and retrieve the server Description https://review.openstack.org/228908 | 18:52 |
| *** sneti has quit IRC | 18:53 | |
| *** haomaiwang has quit IRC | 18:53 | |
| *** nic has quit IRC | 18:53 | |
| jborak | dansmith, I wrote the query I needed, thanks for the help! | 18:53 |
| dansmith | jborak: np | 18:54 |
| jborak | dansmith, I'm trying to write a simple program/script where I can query information about instances that have floating IPs and then update my "external DNS" with entries for them. | 18:54 |
| *** su_zhang has joined #openstack-nova | 18:55 | |
| dansmith | jborak: well, there is some hookage for when that stuff happens, instead of scraping the DB | 18:56 |
| dansmith | jborak: scraping it out of the nova database in JSON form is going to suck | 18:56 |
| dansmith | you can listen for notifications and probably get that information so you can do an update the instant it happens | 18:56 |
| jborak | that's what I thought and I've been trying look for good docs on how to write some kind of hook for when ips are associated/released | 18:56 |
| *** su_zhang has quit IRC | 18:57 | |
| jborak | both are good ideas, not sure how to listen for notifications either. I imagine there are some messages or topic on rabbitmq I could subscribe too. | 18:58 |
| dansmith | yes | 18:58 |
| *** penick has quit IRC | 18:58 | |
| jborak | do you have a link handy where i can read which topic I should be subscribing too and I can write a program to use that instead of polling the db | 18:59 |
| dansmith | no, but let me look | 18:59 |
| dansmith | kbyrne: https://wiki.openstack.org/wiki/NotificationEventExamples | 19:00 |
| dansmith | wtf, major tab fail | 19:00 |
| dansmith | jborak: ^ | 19:00 |
| *** david-lyle has quit IRC | 19:00 | |
| dansmith | jborak: https://prosuncsedu.wordpress.com/2014/01/08/notification-of-actions-in-openstack-nova/ | 19:02 |
| dansmith | jborak: basically, I think google will be able to find you examples | 19:03 |
| jborak | dansmith, thanks for those two links. I've been searching on this topic for a while when I can between other things I do (write software), but I've had a hard time. I gave up this afternoon and just wanted to get *something* going. | 19:04 |
| *** thumpba has joined #openstack-nova | 19:04 | |
| jborak | dansmith, but I can look through the nova python code for th events I need to look for, this all helps a lot! | 19:05 |
| dansmith | jborak: just getting an indication of an instance create with uuid means you can poll the neutron api with the client means you should be able to get what you want I think | 19:06 |
| *** rdopiera has quit IRC | 19:06 | |
| jborak | dansmith, yeah I should be able to do that. I use gophercloud to do some openstack automation. It's getting the notifications that will help me close the loop. | 19:07 |
| *** david-lyle has joined #openstack-nova | 19:08 | |
| *** achanda has joined #openstack-nova | 19:09 | |
| *** sneti has joined #openstack-nova | 19:10 | |
| *** artom has quit IRC | 19:10 | |
| *** alex_klimov has joined #openstack-nova | 19:10 | |
| bauzas | jroll: around ? | 19:14 |
| bauzas | jroll: re: your question about evacuation, I can point you some code if you want | 19:15 |
| openstackgerrit | jichenjc proposed openstack/nova: WIP: add vnc section into nova configuration https://review.openstack.org/234836 | 19:15 |
| bauzas | jroll: we basically ask for recreate to the driver,and if the driver doesn't support it, then we raise an exception | 19:15 |
| *** claudiub has joined #openstack-nova | 19:16 | |
| bauzas | jroll: https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L2760-L2762 | 19:16 |
| bauzas | when evacuating, recreate=True | 19:16 |
| *** annegentle has quit IRC | 19:16 | |
| *** annegentle has joined #openstack-nova | 19:17 | |
| *** xyang1 has joined #openstack-nova | 19:24 | |
| *** penick has joined #openstack-nova | 19:25 | |
| *** ociuhandu has quit IRC | 19:25 | |
| *** david-lyle has quit IRC | 19:25 | |
| *** david-ly_ has joined #openstack-nova | 19:25 | |
| *** ctrath has quit IRC | 19:25 | |
| mriedem | ccarmack: http://developer.openstack.org/api-ref-compute-v2.1.html | 19:26 |
| mriedem | already shows up for all_projects in os-server-groups | 19:26 |
| ccarmack | mriedem: I made all_projects a boolena even though it doesn't matter | 19:28 |
| *** su_zhang has joined #openstack-nova | 19:28 | |
| *** sneti has quit IRC | 19:28 | |
| mriedem | yeah | 19:28 |
| mriedem | that's fine | 19:28 |
| *** sneti has joined #openstack-nova | 19:28 | |
| mriedem | until the day that someone reports a bug that it works with all_projects=1 or all_projects=foo | 19:28 |
| mriedem | and we have to explain that we don't care | 19:28 |
| ccarmack | mriedem: is it wrong for the non-admin to get a reply using all_projects and not an error? | 19:29 |
| openstackgerrit | Sean Dague proposed openstack/nova: WIP: remove project_id enforcement in url https://review.openstack.org/233076 | 19:29 |
| openstackgerrit | Sean Dague proposed openstack/nova: WIP: test adding route to simplify some of the wonkiness of the client. https://review.openstack.org/233779 | 19:29 |
| mriedem | ccarmack: kind of misleading, what does all_tenants do if you're not an admin? | 19:29 |
| *** manas has quit IRC | 19:30 | |
| *** ctrath has joined #openstack-nova | 19:30 | |
| *** su_zhang has quit IRC | 19:32 | |
| *** Marga_ has quit IRC | 19:33 | |
| *** ociuhandu has joined #openstack-nova | 19:36 | |
| *** apuimedo has joined #openstack-nova | 19:37 | |
| *** david-ly_ has quit IRC | 19:38 | |
| *** vkmc is now known as vkmc-afk | 19:38 | |
| *** pumaranikar has quit IRC | 19:38 | |
| *** david-lyle has joined #openstack-nova | 19:38 | |
| openstackgerrit | Vilobh Meshram proposed openstack/nova-specs: Servicegroup foundational refactoring for Control Plane https://review.openstack.org/222423 | 19:39 |
| *** david-lyle has quit IRC | 19:39 | |
| *** david-lyle has joined #openstack-nova | 19:41 | |
| ccarmack | mriedem: it looks like if all_tenants has a policy check https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/servers.py#L371 | 19:42 |
| *** heyongli has joined #openstack-nova | 19:48 | |
| *** e0ne has quit IRC | 19:49 | |
| *** leseb_ has quit IRC | 19:50 | |
| *** tonytan4ever has quit IRC | 19:51 | |
| *** stevemar_ is now known as stevemar_jays | 19:55 | |
| *** thumpba has quit IRC | 19:55 | |
| *** j_king has joined #openstack-nova | 19:56 | |
| *** david-lyle has quit IRC | 19:56 | |
| *** leseb_ has joined #openstack-nova | 19:56 | |
| mriedem | ccarmack: yeah https://github.com/openstack/nova/blob/master/etc/nova/policy.json#L256 | 19:57 |
| mriedem | so in that case it'd fail if a non-admin tried by default | 19:57 |
| mriedem | but os-server-groups would be ok | 19:57 |
| openstackgerrit | Sean Dague proposed openstack/nova: WIP: remove project_id enforcement in url https://review.openstack.org/233076 | 19:57 |
| *** heyongli has quit IRC | 19:57 | |
| mriedem | seems like we should make os-server-groups all_projects usage be similar to all_tenants, with policy check and name of the query parameter, but meh | 19:57 |
| mriedem | it'd be a microversion chagne for something that maybe no one cares about | 19:58 |
| *** david-lyle has joined #openstack-nova | 19:58 | |
| ccarmack | mriedem: ok, no need to change existing behaviour | 19:58 |
| *** heyongli has joined #openstack-nova | 19:59 | |
| *** tonytan4ever has joined #openstack-nova | 20:00 | |
| *** achanda has quit IRC | 20:00 | |
| *** david-ly_ has joined #openstack-nova | 20:01 | |
| *** david-lyle has quit IRC | 20:03 | |
| *** david-ly_ is now known as david-lyle__ | 20:03 | |
| *** david-lyle__ is now known as david-lyle___ | 20:03 | |
| *** Marga_ has joined #openstack-nova | 20:03 | |
| *** david-lyle has joined #openstack-nova | 20:06 | |
| *** david-lyle___ has quit IRC | 20:06 | |
| *** heyongli has quit IRC | 20:07 | |
| *** Marga_ has quit IRC | 20:09 | |
| *** angdraug has joined #openstack-nova | 20:09 | |
| *** heyongli has joined #openstack-nova | 20:09 | |
| *** amrith is now known as _amrith_ | 20:10 | |
| *** marcusvrn_ has joined #openstack-nova | 20:10 | |
| *** su_zhang has joined #openstack-nova | 20:12 | |
| *** heyongli has quit IRC | 20:18 | |
| *** heyongli has joined #openstack-nova | 20:19 | |
| openstackgerrit | Ryan McNair proposed openstack/nova: Add retry logic for detaching device using LibVirt https://review.openstack.org/227851 | 20:22 |
| *** matrohon has quit IRC | 20:26 | |
| openstackgerrit | Mike Fedosin proposed openstack/nova: Use helper method to remove version hardcoding https://review.openstack.org/228578 | 20:26 |
| openstackgerrit | Mike Fedosin proposed openstack/nova: Bringing Glance v2 images to a uniform format https://review.openstack.org/231696 | 20:26 |
| openstackgerrit | Mike Fedosin proposed openstack/nova: glance:add helper method to get client version https://review.openstack.org/222150 | 20:26 |
| *** su_zhang has quit IRC | 20:26 | |
| *** smatzek_ has quit IRC | 20:27 | |
| *** dustins has quit IRC | 20:27 | |
| *** heyongli has quit IRC | 20:28 | |
| *** heyongli has joined #openstack-nova | 20:29 | |
| *** su_zhang has joined #openstack-nova | 20:29 | |
| *** eharney has quit IRC | 20:30 | |
| *** baoli has joined #openstack-nova | 20:32 | |
| *** baoli_ has joined #openstack-nova | 20:33 | |
| *** ijw_ has quit IRC | 20:36 | |
| *** baoli has quit IRC | 20:37 | |
| *** heyongli has quit IRC | 20:38 | |
| *** heyongli has joined #openstack-nova | 20:39 | |
| *** achanda has joined #openstack-nova | 20:41 | |
| *** achanda has quit IRC | 20:41 | |
| *** achanda has joined #openstack-nova | 20:44 | |
| openstackgerrit | Sean Dague proposed openstack/python-novaclient: make project_id optional in urls for version discovery https://review.openstack.org/234782 | 20:45 |
| *** gszasz has joined #openstack-nova | 20:46 | |
| *** su_zhang has quit IRC | 20:47 | |
| *** heyongli has quit IRC | 20:48 | |
| *** dave-mccowan has quit IRC | 20:50 | |
| *** changbl has quit IRC | 20:51 | |
| *** Marga_ has joined #openstack-nova | 20:51 | |
| *** pumaranikar has joined #openstack-nova | 20:52 | |
| *** smatzek has joined #openstack-nova | 20:54 | |
| *** Marga_ has quit IRC | 20:55 | |
| jroll | bauzas: interesting, thanks | 20:56 |
| *** achanda has quit IRC | 20:56 | |
| mrkz | question: is there already a defined date to freeze blueprints/specs for the Mitaka release? | 20:58 |
| *** raildo is now known as raildo-afk | 20:59 | |
| *** edmondsw has quit IRC | 21:01 | |
| *** rfolco has quit IRC | 21:01 | |
| pumaranikar | hi, I am adding a fixed ip to nova instance. It does not raise an exception at client if it fails. is it expected behaviour ?I can see error in nova conductor logs | 21:01 |
| *** ZZelle_ has joined #openstack-nova | 21:02 | |
| *** su_zhang has joined #openstack-nova | 21:02 | |
| *** su_zhang has quit IRC | 21:02 | |
| *** su_zhang has joined #openstack-nova | 21:03 | |
| *** su_zhang has quit IRC | 21:03 | |
| *** eharney has joined #openstack-nova | 21:03 | |
| *** jichen has quit IRC | 21:05 | |
| *** thumpba has joined #openstack-nova | 21:05 | |
| mriedem | mrkz: it'll probably be about the same as for liberty | 21:08 |
| mriedem | pumaranikar: i think you've asked this question about 5 times now? do you have a trace of the failure that you can put in a paste? | 21:09 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: Wrap setup_networks_on_host with try/except https://review.openstack.org/227897 | 21:10 |
| mriedem | pumaranikar: adding a fixed ip to an instance is a cast https://github.com/openstack/nova/blob/master/nova/compute/rpcapi.py#L384 | 21:11 |
| mriedem | so the caller is not going to get like a 400 in the response, they get a 202 | 21:11 |
| mriedem | pumaranikar: if it fails, there should be an instance fault associated with the instance | 21:12 |
| *** achanda has joined #openstack-nova | 21:12 | |
| mriedem | if it fails in the compute service, then the instance fault will be recorded here https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L3980 | 21:13 |
| pumaranikar | mriedem : yes instance status is error | 21:13 |
| mriedem | pumaranikar: so, this isn't a support channel. and besides, it's hard to know what you're seeing with a pastebin of the trace. you could report it as a bug with those details, but if it failed in the compute then there should be an instance fault associated with the instance, which you can get back via the api or novaclient. if the instance fault isn't there, then that's a bug. | 21:15 |
| mriedem | *without a pastebin of the trace | 21:15 |
| pumaranikar | maiedem : here is a trace http://paste.openstack.org/show/476312/ | 21:16 |
| *** smurke has quit IRC | 21:16 | |
| *** smurke has joined #openstack-nova | 21:17 | |
| dansmith | pumaranikar: it's a cast, so no error is expected on return | 21:17 |
| dansmith | pumaranikar: and that trace is not related to add-fixed | 21:18 |
| mriedem | umm, that's a failure for building an instance, not adding a fixed ip | 21:18 |
| mriedem | pumaranikar: you should have an error in the nova-compute logs if you're failing to add a fixed ip | 21:18 |
| mriedem | to a built instance | 21:18 |
| pumaranikar | i was working on a bug saying exception should be raised at client side if add_fixed_ip() fails | 21:19 |
| pumaranikar | *am | 21:19 |
| mrkz | mriedem: I read in the F.A.Q that is preferred to be before the summit so ideas could be properly defined/managed but I see that in Liberty schedule the spec revision was +3 weeks after the summit, that said, you mean it would be same case for M ? | 21:19 |
| mriedem | pumaranikar: link the bug here please | 21:19 |
| mriedem | mrkz: our deadlines have been about the same for kilo and liberty, so i don't expect them to change for mitaka | 21:20 |
| pumaranikar | mriedem : https://bugs.launchpad.net/nova/+bug/1384228 | 21:20 |
| openstack | Launchpad bug 1384228 in OpenStack Compute (nova) "Exception should be raised if nova has failed to add fixed ip" [Low,Confirmed] - Assigned to Pushkar Umaranikar (pushkar-umaranikar) | 21:20 |
| mriedem | we start freezing things in the 2nd milestone | 21:20 |
| dansmith | pumaranikar: just because it's filed as a bug doesn't mean it's a bug, nor that it needs fixing :) | 21:20 |
| mriedem | pumaranikar: not a bug | 21:21 |
| mriedem | cyeoh pointed it out long ago | 21:21 |
| dansmith | yeah, but he didn't mark it invalid unfortunatel | 21:21 |
| dansmith | I will do that now | 21:21 |
| mrkz | mriedem: right, and is expected that for that dates all blueprint/specs are completely defined (no draft status), right? | 21:21 |
| *** su_zhang has joined #openstack-nova | 21:22 | |
| pumaranikar | mriedem : Thank you for your comment. | 21:22 |
| *** su_zhang has quit IRC | 21:23 | |
| mriedem | mrkz: there is a spec freeze where you have to have specs up for review, and a feature proposal freeze, where approved specs have to have code up for review, and then a final feature freeze for everything else | 21:24 |
| mriedem | john had a wiki with those dates for liberty, which i can't seem to find | 21:25 |
| * mrkz hands https://wiki.openstack.org/wiki/Nova/Liberty_Release_Schedule#Spec_and_Blueprint_Approval_Freeze | 21:25 | |
| mriedem | https://wiki.openstack.org/wiki/Nova/Liberty_Release_Schedule | 21:25 |
| mriedem | mrkz: so the rule of thumb is get your spec up for review by last week | 21:26 |
| doffm | mriedem: Is there any ci for the zookeeper service group stuff? Basically is this https://github.com/openstack/nova/blob/master/nova/servicegroup/drivers/zk.py#L102 still true? | 21:26 |
| *** pm90_ has quit IRC | 21:26 | |
| mriedem | create a time machine, go back in time, post your spec, and then wait | 21:26 |
| mriedem | doffm: i don't think there is any CI for the zookeeper stuff now | 21:26 |
| mriedem | *no | 21:26 |
| *** smatzek has quit IRC | 21:27 | |
| mriedem | https://github.com/openstack/nova/blob/master/nova/tests/unit/servicegroup/test_zk_driver.py#L46 | 21:27 |
| *** ccarmack has quit IRC | 21:28 | |
| *** mdrabe has quit IRC | 21:28 | |
| mriedem | doffm: proof http://logs.openstack.org/53/232653/1/check/gate-nova-python27/21b00e4/console.html#_2015-10-14_17_04_13_849 | 21:28 |
| mriedem | doffm: i think at one point jaypipes tried getting zookeeper unit tests to run in nova and it was stupid | 21:29 |
| mriedem | https://bugs.launchpad.net/nova/+bug/1414517 | 21:29 |
| openstack | Launchpad bug 1414517 in OpenStack Compute (nova) "Zookeeper servicegroup API driver is not tested and apparently not usable either" [Medium,Confirmed] - Assigned to Hao Chen (chenh1987) | 21:29 |
| *** tpeoples has quit IRC | 21:29 | |
| *** smatzek has joined #openstack-nova | 21:29 | |
| *** erlarese has quit IRC | 21:30 | |
| *** penick has quit IRC | 21:31 | |
| *** penick has joined #openstack-nova | 21:31 | |
| *** penick has quit IRC | 21:32 | |
| doffm | mriedem: Thanks. | 21:33 |
| *** penick has joined #openstack-nova | 21:34 | |
| *** su_zhang has joined #openstack-nova | 21:34 | |
| *** rajesht_ has joined #openstack-nova | 21:34 | |
| *** thorst has quit IRC | 21:34 | |
| *** ctrath has quit IRC | 21:35 | |
| *** Marga_ has joined #openstack-nova | 21:36 | |
| *** annegentle has quit IRC | 21:37 | |
| *** rajesht has quit IRC | 21:37 | |
| *** rlrossit has left #openstack-nova | 21:40 | |
| mrkz | mriedem: thank you for your replies and sorry for being so insistent on this, but just want to be sure I got it right; so feature freeze would be in Mitaka-2 and code should be ready for review by that time, but the blueprint/spec freeze should be < Mitaka-2 milestone so does that mean that the time to iterate over spec revisions is from now until... Mitaka-1 milestone? | 21:40 |
| *** ccarmack has joined #openstack-nova | 21:42 | |
| *** penick has quit IRC | 21:43 | |
| mriedem | feature freeze is the end of the release | 21:45 |
| mriedem | spec freeze is mitaka 2 | 21:45 |
| mriedem | code and specs should be up for review as soon as posisble | 21:46 |
| mriedem | the firm dates will probably be posted in the ML after the tokyo summit | 21:46 |
| mriedem | feature freeze is the first week of march https://wiki.openstack.org/wiki/Mitaka_Release_Schedule | 21:47 |
| *** rotbeard has quit IRC | 21:47 | |
| *** rotbeard has joined #openstack-nova | 21:47 | |
| *** achanda has quit IRC | 21:48 | |
| *** ctrath has joined #openstack-nova | 21:50 | |
| mrkz | mriedem: ah, that link will help a lot, thanks for the clarification! | 21:51 |
| *** burgerk has quit IRC | 21:54 | |
| *** smatzek has quit IRC | 21:56 | |
| *** pm90_ has joined #openstack-nova | 21:57 | |
| *** nikeshm has quit IRC | 21:57 | |
| mriedem | mrkz: godspeed | 21:57 |
| *** mriedem has quit IRC | 21:58 | |
| *** pm90__ has joined #openstack-nova | 21:59 | |
| *** baoli_ has quit IRC | 21:59 | |
| *** baoli has joined #openstack-nova | 22:00 | |
| *** baoli has quit IRC | 22:00 | |
| *** baoli has joined #openstack-nova | 22:01 | |
| *** pm90_ has quit IRC | 22:02 | |
| openstackgerrit | Dan Smith proposed openstack/nova: Remove obj_relationships from objects https://review.openstack.org/231271 | 22:04 |
| *** sneti has quit IRC | 22:04 | |
| openstackgerrit | Dan Smith proposed openstack/nova: Clean up tests for dropping obj_relationships https://review.openstack.org/231270 | 22:04 |
| openstackgerrit | Dan Smith proposed openstack/nova: DNM: Canary test the backporting from a liberty compute https://review.openstack.org/231486 | 22:04 |
| openstackgerrit | Dan Smith proposed openstack/nova: Fix up Service object for manifest-based backports https://review.openstack.org/231269 | 22:04 |
| openstackgerrit | Dan Smith proposed openstack/nova: DNM: Drop Instance v1.x support https://review.openstack.org/220690 | 22:04 |
| *** smurke has quit IRC | 22:06 | |
| *** pumaranikar has quit IRC | 22:07 | |
| *** jamielennox|away is now known as jamielennox | 22:07 | |
| *** yassine has joined #openstack-nova | 22:09 | |
| *** yassine has quit IRC | 22:09 | |
| *** mc_nair has quit IRC | 22:11 | |
| *** reed_ has quit IRC | 22:14 | |
| *** burt has quit IRC | 22:15 | |
| *** vkmc-afk is now known as vkmc | 22:15 | |
| *** reed_ has joined #openstack-nova | 22:15 | |
| *** achanda has joined #openstack-nova | 22:15 | |
| *** reed_ has quit IRC | 22:15 | |
| *** mmmpork has joined #openstack-nova | 22:15 | |
| *** tonytan4ever has quit IRC | 22:20 | |
| *** kylek3h has quit IRC | 22:21 | |
| *** aysyd has quit IRC | 22:21 | |
| *** josecastroleon has quit IRC | 22:22 | |
| *** pm90__ has quit IRC | 22:23 | |
| *** pm90_ has joined #openstack-nova | 22:23 | |
| *** ssurana has quit IRC | 22:23 | |
| *** ijuwang has quit IRC | 22:28 | |
| *** otter768 has quit IRC | 22:29 | |
| *** otter768 has joined #openstack-nova | 22:29 | |
| *** markvoelker has quit IRC | 22:43 | |
| *** asettle has joined #openstack-nova | 22:44 | |
| *** sneti has joined #openstack-nova | 22:45 | |
| *** achanda has quit IRC | 22:45 | |
| *** openstackgerrit has quit IRC | 22:46 | |
| *** openstackgerrit has joined #openstack-nova | 22:47 | |
| *** achanda has joined #openstack-nova | 22:47 | |
| *** achanda has quit IRC | 22:47 | |
| *** rdo has quit IRC | 22:50 | |
| *** ctrath has quit IRC | 22:50 | |
| *** annegentle has joined #openstack-nova | 22:52 | |
| *** markvoelker has joined #openstack-nova | 22:53 | |
| *** baoli has quit IRC | 22:57 | |
| *** VW has joined #openstack-nova | 22:59 | |
| *** jhesketh has quit IRC | 23:00 | |
| *** jhesketh has joined #openstack-nova | 23:01 | |
| *** VW_ has quit IRC | 23:03 | |
| *** dave-mccowan has joined #openstack-nova | 23:04 | |
| *** VW has quit IRC | 23:04 | |
| *** su_zhang has quit IRC | 23:05 | |
| *** spandhe has joined #openstack-nova | 23:06 | |
| *** su_zhang has joined #openstack-nova | 23:06 | |
| *** ZZelle_ has quit IRC | 23:07 | |
| *** otter768 has quit IRC | 23:08 | |
| *** su_zhang has quit IRC | 23:08 | |
| *** otter768 has joined #openstack-nova | 23:08 | |
| *** achanda has joined #openstack-nova | 23:09 | |
| *** su_zhang has joined #openstack-nova | 23:09 | |
| *** su_zhang has quit IRC | 23:12 | |
| *** su_zhang has joined #openstack-nova | 23:12 | |
| *** pm90_ has quit IRC | 23:16 | |
| *** zhangjn has joined #openstack-nova | 23:17 | |
| *** dims_ has joined #openstack-nova | 23:19 | |
| *** tpeoples has joined #openstack-nova | 23:19 | |
| *** dims has quit IRC | 23:22 | |
| *** Piet has quit IRC | 23:23 | |
| *** sfinucan has joined #openstack-nova | 23:23 | |
| *** su_zhang has quit IRC | 23:24 | |
| *** thorst has joined #openstack-nova | 23:24 | |
| *** dewsday has joined #openstack-nova | 23:26 | |
| *** asettle has quit IRC | 23:26 | |
| *** apoorvad has quit IRC | 23:27 | |
| *** dewsday is now known as asettle | 23:27 | |
| *** apoorvad has joined #openstack-nova | 23:27 | |
| *** annegentle has quit IRC | 23:28 | |
| openstackgerrit | Mike Fedosin proposed openstack/nova: Use helper method to remove version hardcoding https://review.openstack.org/228578 | 23:28 |
| *** annegentle has joined #openstack-nova | 23:29 | |
| *** haomaiwang has joined #openstack-nova | 23:31 | |
| *** annegentle has quit IRC | 23:33 | |
| *** _amrith_ is now known as amrith | 23:34 | |
| *** electroc_ has quit IRC | 23:35 | |
| *** alex_klimov has quit IRC | 23:36 | |
| *** haomaiwang has quit IRC | 23:37 | |
| *** markvoelker has quit IRC | 23:38 | |
| *** oomichi has joined #openstack-nova | 23:38 | |
| *** erhudy has quit IRC | 23:39 | |
| oomichi | mtreinish: hi, are you still working? | 23:39 |
| oomichi | mtreinish: oops, wrong channel | 23:40 |
| mtreinish | yes | 23:40 |
| *** electrocucaracha has joined #openstack-nova | 23:42 | |
| *** apuimedo_ has quit IRC | 23:46 | |
| *** rfolco has joined #openstack-nova | 23:46 | |
| *** rfolco has quit IRC | 23:46 | |
| *** su_zhang has joined #openstack-nova | 23:46 | |
| *** kenji has joined #openstack-nova | 23:48 | |
| *** kenji is now known as Guest22679 | 23:48 | |
| *** Guest22679 has quit IRC | 23:48 | |
| *** agireud has quit IRC | 23:51 | |
| *** electrocucaracha has quit IRC | 23:54 | |
| *** penick has joined #openstack-nova | 23:56 | |
| *** agireud has joined #openstack-nova | 23:57 | |
| *** mgoddard_ has joined #openstack-nova | 23:57 | |
| *** mgoddard has quit IRC | 23:57 | |
| *** edmondsw has joined #openstack-nova | 23:58 | |
| *** asselin has quit IRC | 23:59 | |
| *** ccarmack has quit IRC | 23:59 | |
| *** asselin has joined #openstack-nova | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!