| *** dave-mccowan has joined #openstack-cinder | 00:03 | |
| *** markvoelker has quit IRC | 00:04 | |
| *** gouthamr has joined #openstack-cinder | 00:10 | |
| *** itlinux has quit IRC | 00:11 | |
| *** salv-orlando has joined #openstack-cinder | 00:15 | |
| *** gouthamr has quit IRC | 00:16 | |
| *** jwcroppe has joined #openstack-cinder | 00:17 | |
| *** salv-orlando has quit IRC | 00:19 | |
| *** diablo_rojo has quit IRC | 00:31 | |
| *** mriedem has quit IRC | 00:38 | |
| *** Julien-zte has joined #openstack-cinder | 00:46 | |
| *** zhengyin has joined #openstack-cinder | 00:47 | |
| *** markvoelker has joined #openstack-cinder | 00:47 | |
| *** wanghao has joined #openstack-cinder | 00:49 | |
| *** Julien-zte has quit IRC | 00:49 | |
| *** frontrunner has joined #openstack-cinder | 00:49 | |
| *** wanghao has quit IRC | 00:50 | |
| *** Julien-zte has joined #openstack-cinder | 00:50 | |
| *** wanghao has joined #openstack-cinder | 00:50 | |
| *** jwcroppe has quit IRC | 00:51 | |
| *** dikonoor has joined #openstack-cinder | 01:01 | |
| *** Shunli has joined #openstack-cinder | 01:04 | |
| *** wN has joined #openstack-cinder | 01:04 | |
| *** wN has quit IRC | 01:04 | |
| *** wN has joined #openstack-cinder | 01:04 | |
| *** wanghao_ has joined #openstack-cinder | 01:07 | |
| *** zhangguoqing has joined #openstack-cinder | 01:08 | |
| *** wanghao has quit IRC | 01:10 | |
| *** gouthamr has joined #openstack-cinder | 01:13 | |
| *** gcb has joined #openstack-cinder | 01:13 | |
| *** wanghao_ has quit IRC | 01:13 | |
| *** wanghao has joined #openstack-cinder | 01:14 | |
| openstackgerrit | Liu Qing proposed openstack/cinder master: Fix SSHPool current_size not work correctly https://review.openstack.org/468634 | 01:16 |
|---|---|---|
| *** markvoelker_ has joined #openstack-cinder | 01:16 | |
| *** markvoelker has quit IRC | 01:17 | |
| *** lucasxu has joined #openstack-cinder | 01:17 | |
| *** lucasxu has quit IRC | 01:17 | |
| *** Apoorva_ has joined #openstack-cinder | 01:18 | |
| *** jose-phillips has quit IRC | 01:20 | |
| *** lucasxu has joined #openstack-cinder | 01:20 | |
| *** markvoelker_ has quit IRC | 01:21 | |
| *** jose-phillips has joined #openstack-cinder | 01:22 | |
| *** Apoorva_ has quit IRC | 01:22 | |
| *** Apoorva has quit IRC | 01:22 | |
| *** lucasxu has quit IRC | 01:22 | |
| *** lucasxu has joined #openstack-cinder | 01:23 | |
| *** markvoelker has joined #openstack-cinder | 01:23 | |
| *** dikonoor has quit IRC | 01:24 | |
| *** dhinesh has quit IRC | 01:25 | |
| *** wanghao_ has joined #openstack-cinder | 01:25 | |
| *** wanghao has quit IRC | 01:28 | |
| *** arahal_ has quit IRC | 01:32 | |
| openstackgerrit | wangxiyuan proposed openstack/cinder master: Return metadata in volume summary https://review.openstack.org/459143 | 01:35 |
| *** wanghao_ has quit IRC | 01:36 | |
| *** wanghao has joined #openstack-cinder | 01:36 | |
| *** aselius has quit IRC | 01:45 | |
| *** zhaochao has joined #openstack-cinder | 01:45 | |
| openstackgerrit | wzyuliyang proposed openstack/cinder master: Add pigz support for cinder-backup chunked driver compression https://review.openstack.org/450291 | 01:51 |
| *** hongbin has quit IRC | 01:51 | |
| *** kaisers_ has joined #openstack-cinder | 01:54 | |
| openstackgerrit | wzyuliyang proposed openstack/cinder master: Add pigz support for cinder-backup chunked driver compression https://review.openstack.org/450291 | 01:56 |
| *** kaisers has quit IRC | 01:57 | |
| *** lucasxu has quit IRC | 02:01 | |
| *** lucasxu has joined #openstack-cinder | 02:11 | |
| *** lucasxu has quit IRC | 02:13 | |
| *** lucasxu has joined #openstack-cinder | 02:14 | |
| *** salv-orlando has joined #openstack-cinder | 02:15 | |
| *** dave-mccowan has quit IRC | 02:15 | |
| *** wanghao_ has joined #openstack-cinder | 02:17 | |
| *** wanghao_ has quit IRC | 02:17 | |
| *** wanghao has quit IRC | 02:17 | |
| *** wanghao has joined #openstack-cinder | 02:17 | |
| *** salv-orlando has quit IRC | 02:20 | |
| *** wanghao_ has joined #openstack-cinder | 02:35 | |
| *** wanghao has quit IRC | 02:38 | |
| *** lucasxu has quit IRC | 02:40 | |
| openstackgerrit | wangxiyuan proposed openstack/cinder master: Return metadata in volume summary https://review.openstack.org/459143 | 02:42 |
| *** zsli_ has joined #openstack-cinder | 02:44 | |
| *** wanghao has joined #openstack-cinder | 02:44 | |
| *** lucasxu has joined #openstack-cinder | 02:45 | |
| *** wanghao_ has quit IRC | 02:46 | |
| *** Shunli has quit IRC | 02:46 | |
| *** lucasxu has quit IRC | 02:47 | |
| *** nicolasbock has quit IRC | 02:49 | |
| *** markstur has quit IRC | 02:55 | |
| *** lucasxu has joined #openstack-cinder | 02:56 | |
| *** Julien-zte has quit IRC | 03:00 | |
| *** Julien-zte has joined #openstack-cinder | 03:01 | |
| *** lucasxu has quit IRC | 03:04 | |
| *** zsli_ has quit IRC | 03:08 | |
| *** Shunli has joined #openstack-cinder | 03:08 | |
| *** Julien-zte has quit IRC | 03:10 | |
| *** Julien-zte has joined #openstack-cinder | 03:12 | |
| *** Julien-zte has quit IRC | 03:19 | |
| *** nicolasbock has joined #openstack-cinder | 03:20 | |
| *** Julien-zte has joined #openstack-cinder | 03:20 | |
| *** esker has quit IRC | 03:22 | |
| *** Julien-zte has quit IRC | 03:23 | |
| *** Julien-zte has joined #openstack-cinder | 03:24 | |
| *** dhinesh has joined #openstack-cinder | 03:24 | |
| *** frontrunner has quit IRC | 03:26 | |
| *** Julien-zte has quit IRC | 03:27 | |
| *** Julien-zte has joined #openstack-cinder | 03:28 | |
| *** psachin has quit IRC | 03:29 | |
| *** nicolasbock has quit IRC | 03:33 | |
| *** david-lyle has joined #openstack-cinder | 03:35 | |
| *** Julien-z_ has joined #openstack-cinder | 03:35 | |
| *** Julien-zte has quit IRC | 03:37 | |
| *** Julien-z_ has quit IRC | 03:45 | |
| *** links has joined #openstack-cinder | 03:45 | |
| *** lucasxu has joined #openstack-cinder | 03:45 | |
| *** Julien-zte has joined #openstack-cinder | 03:46 | |
| *** psachin has joined #openstack-cinder | 03:46 | |
| *** Julien-zte has quit IRC | 03:50 | |
| *** salv-orlando has joined #openstack-cinder | 03:50 | |
| *** Julien-zte has joined #openstack-cinder | 03:50 | |
| *** Julien-zte has quit IRC | 03:53 | |
| *** salv-orlando has quit IRC | 03:54 | |
| *** Julien-zte has joined #openstack-cinder | 03:57 | |
| *** zhangguoqing has quit IRC | 04:00 | |
| *** kaisers_ has quit IRC | 04:06 | |
| *** arnewiebalck_ has joined #openstack-cinder | 04:07 | |
| *** kaisers has joined #openstack-cinder | 04:10 | |
| *** lucasxu has quit IRC | 04:11 | |
| *** kaisers has quit IRC | 04:15 | |
| *** zhurong has joined #openstack-cinder | 04:15 | |
| *** dikonoor has joined #openstack-cinder | 04:18 | |
| *** zhurong has quit IRC | 04:19 | |
| *** mtanino__ has joined #openstack-cinder | 04:24 | |
| *** zhengyin has quit IRC | 04:32 | |
| *** dikonoor has quit IRC | 04:37 | |
| *** salv-orlando has joined #openstack-cinder | 04:39 | |
| *** chhavi has joined #openstack-cinder | 04:40 | |
| *** mtanin___ has joined #openstack-cinder | 04:45 | |
| *** mtanino__ has quit IRC | 04:46 | |
| *** gkadam has joined #openstack-cinder | 04:46 | |
| *** gouthamr has quit IRC | 04:51 | |
| *** zhangguoqing has joined #openstack-cinder | 04:51 | |
| *** Julien-zte has quit IRC | 04:52 | |
| *** dikonoor has joined #openstack-cinder | 04:55 | |
| *** Julien-zte has joined #openstack-cinder | 04:56 | |
| *** pgadiya has joined #openstack-cinder | 04:59 | |
| *** pgadiya has quit IRC | 04:59 | |
| openstackgerrit | jiamin proposed openstack/cinder master: DS8K: support clone volume asynchroniously https://review.openstack.org/467136 | 05:00 |
| *** dikonoor has quit IRC | 05:01 | |
| *** adisky__ has joined #openstack-cinder | 05:03 | |
| *** mtanin___ has quit IRC | 05:07 | |
| *** zhengyin has joined #openstack-cinder | 05:11 | |
| *** kaisers has joined #openstack-cinder | 05:11 | |
| *** arnewiebalck_ has quit IRC | 05:14 | |
| *** Poornima has joined #openstack-cinder | 05:39 | |
| *** Julien-zte has quit IRC | 05:52 | |
| openstackgerrit | Cheng Yong Zhao proposed openstack/cinder master: Add gmcv support in SVC driver https://review.openstack.org/461191 | 05:56 |
| *** ChubYann has quit IRC | 05:58 | |
| *** xiangxinyong is now known as edison | 06:08 | |
| *** edison is now known as edisonxiang | 06:08 | |
| *** mvk has quit IRC | 06:09 | |
| *** rcernin has joined #openstack-cinder | 06:10 | |
| *** KeithMnemonic1 has joined #openstack-cinder | 06:16 | |
| *** karthikp has quit IRC | 06:16 | |
| *** karthikp_ is now known as karthikp | 06:16 | |
| *** dhinesh_ has joined #openstack-cinder | 06:17 | |
| *** dhinesh has quit IRC | 06:17 | |
| *** bluex has quit IRC | 06:18 | |
| *** karthikp_ has joined #openstack-cinder | 06:18 | |
| *** KeithMnemonic has quit IRC | 06:18 | |
| openstackgerrit | Yingzhe Zeng proposed openstack/cinder master: Use requests lib for Huawei array connection https://review.openstack.org/459042 | 06:24 |
| openstackgerrit | Cheng Yong Zhao proposed openstack/cinder master: Add gmcv support in SVC driver https://review.openstack.org/461191 | 06:27 |
| *** bluex has joined #openstack-cinder | 06:27 | |
| *** zhengyin has quit IRC | 06:33 | |
| *** zhengyin has joined #openstack-cinder | 06:34 | |
| *** e0ne has joined #openstack-cinder | 06:42 | |
| *** zhengyin has quit IRC | 06:44 | |
| *** e0ne has quit IRC | 06:50 | |
| *** tesseract has joined #openstack-cinder | 06:53 | |
| *** jsuchome has joined #openstack-cinder | 06:59 | |
| *** zhengyin has joined #openstack-cinder | 07:00 | |
| *** e0ne has joined #openstack-cinder | 07:04 | |
| *** mvk has joined #openstack-cinder | 07:08 | |
| *** wanghao_ has joined #openstack-cinder | 07:09 | |
| *** wanghao has quit IRC | 07:12 | |
| *** chhavi has quit IRC | 07:22 | |
| *** pcaruana has joined #openstack-cinder | 07:26 | |
| *** dhinesh_ has quit IRC | 07:26 | |
| *** salv-orl_ has joined #openstack-cinder | 07:27 | |
| *** salv-orlando has quit IRC | 07:30 | |
| *** yangyapeng has joined #openstack-cinder | 07:31 | |
| *** yangyapeng has quit IRC | 07:31 | |
| *** e0ne has quit IRC | 07:31 | |
| *** pjm6 has quit IRC | 07:34 | |
| *** pjm6 has joined #openstack-cinder | 07:37 | |
| *** Poornima has quit IRC | 07:43 | |
| *** aselius has joined #openstack-cinder | 07:45 | |
| *** arnewiebalck_ has joined #openstack-cinder | 07:46 | |
| *** aarefiev_afk is now known as aarefiev | 07:54 | |
| *** zzzeek has quit IRC | 08:00 | |
| *** mnestratov has joined #openstack-cinder | 08:01 | |
| *** zzzeek has joined #openstack-cinder | 08:01 | |
| *** chhavi has joined #openstack-cinder | 08:02 | |
| *** dikonoor has joined #openstack-cinder | 08:02 | |
| *** lpetrut has joined #openstack-cinder | 08:02 | |
| *** Shunli has quit IRC | 08:12 | |
| *** Shunli has joined #openstack-cinder | 08:13 | |
| *** arnewiebalck_ has quit IRC | 08:14 | |
| *** tomh_ has joined #openstack-cinder | 08:25 | |
| *** stakeda has quit IRC | 08:28 | |
| *** Poornima has joined #openstack-cinder | 08:29 | |
| *** dikonoor has quit IRC | 08:39 | |
| *** savvas_ has joined #openstack-cinder | 08:43 | |
| *** savvas has quit IRC | 08:45 | |
| *** cristicalin has joined #openstack-cinder | 08:48 | |
| *** cristicalin has quit IRC | 08:50 | |
| *** gkadam_ has joined #openstack-cinder | 08:52 | |
| *** jose-phi_ has joined #openstack-cinder | 08:53 | |
| *** wxy_ has joined #openstack-cinder | 08:53 | |
| *** dobson has joined #openstack-cinder | 08:54 | |
| *** fmccrthy_ has joined #openstack-cinder | 08:54 | |
| *** wanghao has joined #openstack-cinder | 08:54 | |
| *** ediardo_ has joined #openstack-cinder | 08:54 | |
| *** strigazi_ has joined #openstack-cinder | 08:56 | |
| *** dalvarezs has joined #openstack-cinder | 08:59 | |
| *** oomichi_ has joined #openstack-cinder | 08:59 | |
| *** zzzeek_ has joined #openstack-cinder | 09:00 | |
| *** mmedvede_ has joined #openstack-cinder | 09:00 | |
| *** johnthetubaguy_ has joined #openstack-cinder | 09:00 | |
| *** dmellado_ has joined #openstack-cinder | 09:00 | |
| *** zzzeek has quit IRC | 09:01 | |
| *** wanghao_ has quit IRC | 09:01 | |
| *** gkadam has quit IRC | 09:01 | |
| *** links has quit IRC | 09:01 | |
| *** jose-phillips has quit IRC | 09:01 | |
| *** fmccrthy has quit IRC | 09:01 | |
| *** ediardo has quit IRC | 09:01 | |
| *** kaisers1 has quit IRC | 09:01 | |
| *** mmedvede has quit IRC | 09:01 | |
| *** wxy has quit IRC | 09:01 | |
| *** rushiagr has quit IRC | 09:01 | |
| *** dobson` has quit IRC | 09:01 | |
| *** dmellado has quit IRC | 09:01 | |
| *** dalvarez has quit IRC | 09:01 | |
| *** stvnoyes has quit IRC | 09:01 | |
| *** oomichi has quit IRC | 09:01 | |
| *** strigazi has quit IRC | 09:01 | |
| *** johnthetubaguy has quit IRC | 09:01 | |
| *** jistr has quit IRC | 09:01 | |
| *** mmedvede_ is now known as mmedvede | 09:01 | |
| *** oomichi_ is now known as oomichi | 09:01 | |
| *** ediardo_ is now known as ediardo | 09:01 | |
| *** fmccrthy_ is now known as fmccrthy | 09:01 | |
| *** wxy_ is now known as wxy | 09:01 | |
| *** rarora has joined #openstack-cinder | 09:01 | |
| *** jistr has joined #openstack-cinder | 09:02 | |
| *** gkadam_ is now known as gkadam | 09:04 | |
| *** links has joined #openstack-cinder | 09:07 | |
| *** stvnoyes has joined #openstack-cinder | 09:08 | |
| *** kaisers1 has joined #openstack-cinder | 09:09 | |
| *** david-lyle has quit IRC | 09:11 | |
| *** david-lyle has joined #openstack-cinder | 09:12 | |
| *** rushiagr has joined #openstack-cinder | 09:14 | |
| *** cristicalin has joined #openstack-cinder | 09:22 | |
| openstackgerrit | xiaoqin proposed openstack/cinder master: [WIP] Add Storwize replication group support https://review.openstack.org/469394 | 09:26 |
| *** Shunli has quit IRC | 09:32 | |
| *** dalgaaf has quit IRC | 09:39 | |
| *** dalgaaf has joined #openstack-cinder | 09:44 | |
| *** dmellado_ is now known as dmellado | 09:46 | |
| openstackgerrit | Shunei Shiono proposed openstack/cinder master: Refactor NEC volume driver https://review.openstack.org/453022 | 09:50 |
| *** cristicalin has quit IRC | 09:52 | |
| *** aselius has quit IRC | 09:54 | |
| *** mvk has quit IRC | 09:55 | |
| *** dikonoor has joined #openstack-cinder | 09:56 | |
| *** nicolasbock has joined #openstack-cinder | 09:56 | |
| *** wanghao has quit IRC | 09:58 | |
| *** wanghao has joined #openstack-cinder | 10:03 | |
| *** wanghao has quit IRC | 10:04 | |
| *** Julien-zte has joined #openstack-cinder | 10:10 | |
| *** mvk has joined #openstack-cinder | 10:10 | |
| *** Julien-zte has quit IRC | 10:12 | |
| openstackgerrit | Gorka Eguileor proposed openstack/os-brick master: Add open-iscsi manual scan support https://review.openstack.org/455394 | 10:12 |
| openstackgerrit | Gorka Eguileor proposed openstack/os-brick master: Refactor iSCSI disconnect https://review.openstack.org/455392 | 10:12 |
| openstackgerrit | Gorka Eguileor proposed openstack/os-brick master: Refactor iSCSI connect https://review.openstack.org/455393 | 10:12 |
| *** Julien-zte has joined #openstack-cinder | 10:13 | |
| *** dikonoor has quit IRC | 10:13 | |
| *** dikonoor has joined #openstack-cinder | 10:14 | |
| *** gkadam has quit IRC | 10:14 | |
| *** links has quit IRC | 10:16 | |
| *** e0ne has joined #openstack-cinder | 10:22 | |
| *** Julien-zte has quit IRC | 10:23 | |
| *** Julien-zte has joined #openstack-cinder | 10:24 | |
| *** Julien-zte has quit IRC | 10:25 | |
| *** Julien-zte has joined #openstack-cinder | 10:26 | |
| *** Julien-zte has quit IRC | 10:26 | |
| *** nicolasbock has quit IRC | 10:27 | |
| *** nicolasbock has joined #openstack-cinder | 10:27 | |
| *** Julien-zte has joined #openstack-cinder | 10:27 | |
| *** nicolasbock has quit IRC | 10:28 | |
| *** nicolasbock has joined #openstack-cinder | 10:28 | |
| *** gkadam has joined #openstack-cinder | 10:34 | |
| *** links has joined #openstack-cinder | 10:36 | |
| *** chhavi has quit IRC | 10:37 | |
| *** Julien-zte has quit IRC | 10:39 | |
| *** Julien-zte has joined #openstack-cinder | 10:40 | |
| *** chhavi has joined #openstack-cinder | 10:40 | |
| *** Julien-zte has quit IRC | 10:44 | |
| *** Julien-zte has joined #openstack-cinder | 10:45 | |
| *** zhangguoqing has quit IRC | 10:50 | |
| *** gkadam_ has joined #openstack-cinder | 10:54 | |
| *** gkadam has quit IRC | 10:57 | |
| *** salv-orl_ has quit IRC | 10:59 | |
| openstackgerrit | Sean McGinnis proposed openstack/python-cinderclient master: Fix minor issues in shell help messages https://review.openstack.org/469428 | 11:02 |
| *** Julien-zte has quit IRC | 11:03 | |
| *** Julien-zte has joined #openstack-cinder | 11:04 | |
| *** cristicalin has joined #openstack-cinder | 11:06 | |
| *** Julien-zte has quit IRC | 11:07 | |
| *** Julien-zte has joined #openstack-cinder | 11:08 | |
| *** gkadam_ is now known as gkadam | 11:08 | |
| *** zhengyin has quit IRC | 11:08 | |
| *** dave-mccowan has joined #openstack-cinder | 11:12 | |
| *** liverpooler has quit IRC | 11:13 | |
| *** Julien-zte has quit IRC | 11:15 | |
| openstackgerrit | wzyuliyang proposed openstack/cinder master: Add pigz support for cinder-backup chunked driver compression https://review.openstack.org/450291 | 11:17 |
| *** chhavi has quit IRC | 11:33 | |
| openstackgerrit | TommyLike proposed openstack/python-cinderclient master: Support generalized resource filter in client https://review.openstack.org/465002 | 11:33 |
| *** savvas_ has quit IRC | 11:37 | |
| *** savvas has joined #openstack-cinder | 11:37 | |
| *** psachin has quit IRC | 11:49 | |
| *** psachin has joined #openstack-cinder | 11:49 | |
| openstackgerrit | Gorka Eguileor proposed openstack/cinder master: Do proper cleanup if connect volume fails https://review.openstack.org/459453 | 11:51 |
| openstackgerrit | Gorka Eguileor proposed openstack/cinder master: Add support for OS-Brick force disconnect https://review.openstack.org/459454 | 11:51 |
| openstackgerrit | Gorka Eguileor proposed openstack/cinder master: Multipath tempest scenarios https://review.openstack.org/469445 | 11:51 |
| *** esker has joined #openstack-cinder | 11:55 | |
| *** abishop has joined #openstack-cinder | 11:56 | |
| *** esker has quit IRC | 12:00 | |
| *** frontrunner has joined #openstack-cinder | 12:00 | |
| *** MVenesio has joined #openstack-cinder | 12:01 | |
| *** edmondsw has joined #openstack-cinder | 12:02 | |
| *** chlong has quit IRC | 12:03 | |
| *** bjornar_ has joined #openstack-cinder | 12:04 | |
| *** kbaegis has joined #openstack-cinder | 12:05 | |
| *** kbaegis has quit IRC | 12:05 | |
| *** kbaegis has joined #openstack-cinder | 12:06 | |
| *** kbaegis has quit IRC | 12:06 | |
| *** kbaegis has joined #openstack-cinder | 12:06 | |
| *** kbaegis has quit IRC | 12:06 | |
| *** kbaegis has joined #openstack-cinder | 12:07 | |
| *** kbaegis has quit IRC | 12:07 | |
| *** psachin has quit IRC | 12:07 | |
| *** felipemonteiro has joined #openstack-cinder | 12:10 | |
| *** links has quit IRC | 12:11 | |
| *** felipemonteiro_ has joined #openstack-cinder | 12:19 | |
| openstackgerrit | Gorka Eguileor proposed openstack/os-brick master: Add open-iscsi manual scan support https://review.openstack.org/455394 | 12:20 |
| openstackgerrit | Gorka Eguileor proposed openstack/os-brick master: Refactor iSCSI connect https://review.openstack.org/455393 | 12:20 |
| *** felipemonteiro has quit IRC | 12:22 | |
| *** kaisers has quit IRC | 12:30 | |
| openstackgerrit | Merged openstack/cinder master: Fix api-ref for POST backup for optional name https://review.openstack.org/452090 | 12:31 |
| *** gkadam_ has joined #openstack-cinder | 12:34 | |
| *** gkadam has quit IRC | 12:37 | |
| *** gkadam has joined #openstack-cinder | 12:40 | |
| *** gkadam_ has quit IRC | 12:40 | |
| *** Julien-zte has joined #openstack-cinder | 12:44 | |
| *** liverpooler has joined #openstack-cinder | 12:47 | |
| *** gkadam has quit IRC | 12:47 | |
| *** Julien-zte has quit IRC | 12:48 | |
| *** Julien-zte has joined #openstack-cinder | 12:48 | |
| *** zhangguoqing has joined #openstack-cinder | 13:01 | |
| *** Julien-zte has quit IRC | 13:01 | |
| *** Julien-zte has joined #openstack-cinder | 13:02 | |
| *** eharney has joined #openstack-cinder | 13:03 | |
| *** xyang1 has joined #openstack-cinder | 13:04 | |
| *** dpawlik has joined #openstack-cinder | 13:05 | |
| *** Poornima has quit IRC | 13:07 | |
| *** cristicalin has quit IRC | 13:07 | |
| *** psachin has joined #openstack-cinder | 13:08 | |
| *** danpawlik has quit IRC | 13:09 | |
| *** jwcroppe has joined #openstack-cinder | 13:10 | |
| *** lucasxu has joined #openstack-cinder | 13:11 | |
| *** strigazi_ is now known as strigazi | 13:15 | |
| *** lifeless has quit IRC | 13:15 | |
| *** lixiaoy1 has joined #openstack-cinder | 13:18 | |
| *** gouthamr has joined #openstack-cinder | 13:19 | |
| *** lixiaoy1 has quit IRC | 13:21 | |
| *** superdan is now known as dansmith | 13:26 | |
| *** kaisers has joined #openstack-cinder | 13:26 | |
| *** mugsie has quit IRC | 13:27 | |
| *** chlong has joined #openstack-cinder | 13:30 | |
| *** chhavi has joined #openstack-cinder | 13:32 | |
| *** lifeless has joined #openstack-cinder | 13:33 | |
| *** zhangguoqing has quit IRC | 13:34 | |
| *** crushil has joined #openstack-cinder | 13:35 | |
| *** dfisher has joined #openstack-cinder | 13:39 | |
| *** zhangguoqing has joined #openstack-cinder | 13:41 | |
| *** zhurong has joined #openstack-cinder | 13:42 | |
| *** mriedem has joined #openstack-cinder | 13:42 | |
| *** cristicalin has joined #openstack-cinder | 13:43 | |
| *** chlong has quit IRC | 13:43 | |
| *** chlong has joined #openstack-cinder | 13:46 | |
| *** bjornar_ has quit IRC | 13:49 | |
| *** psachin has quit IRC | 13:53 | |
| *** mtanino__ has joined #openstack-cinder | 13:53 | |
| *** kaisers has quit IRC | 13:55 | |
| *** wanghao has joined #openstack-cinder | 13:59 | |
| *** gluytium has quit IRC | 14:02 | |
| openstackgerrit | Merged openstack/python-cinderclient master: Support generalized resource filter in client https://review.openstack.org/465002 | 14:05 |
| openstackgerrit | Gorka Eguileor proposed openstack/cinder master: Remove unused _detach_snapshot method https://review.openstack.org/469499 | 14:05 |
| *** bjornar_ has joined #openstack-cinder | 14:05 | |
| *** felipemonteiro_ has quit IRC | 14:06 | |
| openstackgerrit | Gorka Eguileor proposed openstack/os-brick master: Add open-iscsi manual scan support https://review.openstack.org/455394 | 14:06 |
| openstackgerrit | Gorka Eguileor proposed openstack/os-brick master: Refactor iSCSI disconnect https://review.openstack.org/455392 | 14:06 |
| openstackgerrit | Gorka Eguileor proposed openstack/os-brick master: Refactor iSCSI connect https://review.openstack.org/455393 | 14:06 |
| openstackgerrit | Gorka Eguileor proposed openstack/cinder master: Multipath tempest scenarios https://review.openstack.org/469445 | 14:06 |
| openstackgerrit | Gorka Eguileor proposed openstack/cinder master: Allow logging in tempest tests https://review.openstack.org/469502 | 14:06 |
| *** felipemonteiro has joined #openstack-cinder | 14:09 | |
| smcginnis | geguileo: You're almost back to filling my whole IRC screen. :) | 14:09 |
| geguileo | lol | 14:09 |
| *** salv-orlando has joined #openstack-cinder | 14:10 | |
| *** aselius has joined #openstack-cinder | 14:13 | |
| mriedem | smcginnis: jungleboyj: scottda: jgriffith: ildikov: got a question about this cinderclient code that is doing version discovery https://github.com/openstack/python-cinderclient/blob/2.1.0/cinderclient/client.py#L88 | 14:24 |
| *** hongbin has joined #openstack-cinder | 14:25 | |
| mriedem | what happens if the server doesn't have v3? doesn't that for loop not error but also doesn't return anything? | 14:25 |
| mriedem | since '3.' isn't in there? | 14:25 |
| jgriffith | mriedem it would appear you're correct... scottda ? | 14:25 |
| mriedem | also, looking at this https://developer.openstack.org/api-ref/block-storage/v3/?expanded=list-all-api-versions-detail#response | 14:27 |
| mriedem | it looks like looking at 'version' is the wrong key | 14:27 |
| jgriffith | mriedem looks like that needs to set a deafault for the return | 14:27 |
| mriedem | "version": "", | 14:27 |
| mriedem | the id is v1.0, or v2.0 or v3.0 | 14:27 |
| *** dikonoor has quit IRC | 14:27 | |
| mriedem | but version should be set - is that doc just wrong? | 14:27 |
| ildikov | mriedem: good morning :) | 14:28 |
| mriedem | i was thinking about cinder version discovery at 4am | 14:29 |
| mriedem | so here i am | 14:29 |
| ildikov | that looks like nightmare-ish, I agree | 14:29 |
| mriedem | or maybe the version key didn't get set until the v3 API? | 14:30 |
| jgriffith | mriedem I believe that's accurate, but scottda should have all the answers around this stuff | 14:32 |
| mriedem | ok yeah i think it's this https://github.com/openstack/cinder/blob/master/cinder/api/versions.py#L36-L73 | 14:32 |
| ildikov | I believe most of this stuff got introduced with microversions or with that in mind | 14:32 |
| jgriffith | I'm not quite sure how that v3 response is even useful either though looking at the doc there | 14:32 |
| mriedem | jgriffith: right it's the same as the GET / | 14:32 |
| ildikov | but not many of us has the detailed knowledge about this | 14:33 |
| mriedem | but i'm pretty sure the client code is missing an else on the for loop to raise a ClientException to default to 2.0 | 14:33 |
| jgriffith | mriedem it's definitely missing that, I'm writing a patch now | 14:33 |
| mriedem | if i'm just passing the endpoint url from the service catalog for volumev2, or the server is not at ocata yet, then it won't have 3.x in the response and return None, which will lead to NoneType | 14:33 |
| mriedem | jgriffith: i'll report a bug | 14:34 |
| jgriffith | mriedem but the response thing I'm punting to somebody else because frankly I don't want to go anywhere near it | 14:34 |
| *** zhurong has quit IRC | 14:35 | |
| ildikov | jgriffith: you mean the 'version' key? | 14:35 |
| mriedem | jgriffith: unless cinder didn't support version discovery before v3, the v3 change actually changed the version discovery response :( https://review.openstack.org/#/c/224910/36/cinder/api/versions.py | 14:36 |
| jgriffith | ildikov ? | 14:36 |
| mriedem | by adding new keys | 14:36 |
| jgriffith | ildikov I mean this: https://github.com/openstack/python-cinderclient/blob/2.1.0/cinderclient/client.py#L88 | 14:36 |
| mriedem | how ironic | 14:36 |
| *** jonesn has joined #openstack-cinder | 14:37 | |
| jgriffith | mriedem ha! Yep!! | 14:37 |
| ildikov | jgriffith: I meant that one, yes | 14:37 |
| mriedem | jgriffith: https://bugs.launchpad.net/python-cinderclient/+bug/1694729 | 14:39 |
| openstack | Launchpad bug 1694729 in python-cinderclient "get_server_version does not handle older (non-v3) servers" [Undecided,New] | 14:39 |
| jgriffith | mriedem thanks | 14:39 |
| *** ericyoung has joined #openstack-cinder | 14:43 | |
| scottda | mriedem: Looking... | 14:46 |
| jgriffith | I'm sure my patch will be objectionable, but oh well. | 14:47 |
| openstackgerrit | John Griffith proposed openstack/python-cinderclient master: cinderclient might not return version for V2 API https://review.openstack.org/469520 | 14:47 |
| mriedem | i think all you need is for/else raise ClientException | 14:47 |
| jgriffith | mriedem I'd prefer to just make it less stupid looking and have it be somewhat clear | 14:47 |
| jgriffith | but most people don't like it when I do that... or my brain is just broken maybe | 14:48 |
| jgriffith | inner returns suck IMO | 14:48 |
| *** zhangguoqing has quit IRC | 14:48 | |
| jgriffith | anyway, get your -1's ready | 14:48 |
| mriedem | working on it | 14:49 |
| *** cristicalin has quit IRC | 14:49 | |
| jgriffith | aww crap | 14:50 |
| jgriffith | break | 14:50 |
| openstackgerrit | Merged openstack/cinder master: Add an instance-uuid check on attachment_reserve https://review.openstack.org/469252 | 14:50 |
| openstackgerrit | John Griffith proposed openstack/python-cinderclient master: cinderclient might not return version for V2 API https://review.openstack.org/469520 | 14:50 |
| scottda | Yes, that code lacked a default return. Patch looks fine to me jgriffith | 14:51 |
| jgriffith | scottda thanks! | 14:51 |
| scottda | And, yes, there was no version discovery before microversions. | 14:52 |
| *** kaisers has joined #openstack-cinder | 14:52 | |
| scottda | So only exists in V3 | 14:52 |
| *** bjornar_ has quit IRC | 14:52 | |
| mriedem | scottda: i'm not buying that | 14:52 |
| mriedem | v2 has a GET / response https://developer.openstack.org/api-ref/block-storage/v2/#list-api-versions | 14:52 |
| *** zhangguoqing has joined #openstack-cinder | 14:53 | |
| mriedem | and https://review.openstack.org/#/c/224910/36/cinder/api/versions.py was returning v1 and v2 in the response before v3 was added | 14:53 |
| mriedem | that's version discovery | 14:53 |
| *** zhangguoqing has quit IRC | 14:53 | |
| scottda | ahh..right. I should say, there was no generic way to get the current version. You had to hit the /v1 or v2 endpoint | 14:53 |
| scottda | But, it didn't tell you much, IIRC. I mean, you hit /v2 and it tells you that you are on v2. | 14:54 |
| mriedem | jgriffith: so there is also going to be a similar issue in this one https://github.com/openstack/python-cinderclient/blob/2.1.0/cinderclient/api_versions.py#L242 | 14:54 |
| mriedem | jgriffith: although that's probably a programming error if it returns None | 14:54 |
| *** wanghao has quit IRC | 14:55 | |
| jgriffith | si | 14:55 |
| jgriffith | but it probably wouldn't hurt to make it explicit and tag it onto the existing bug/review IMO | 14:55 |
| scottda | mriedem: I'll have to look at that one. I think the None, None is dealt with in the calling code. Not that that is the best way to do it... | 14:55 |
| *** markstur has joined #openstack-cinder | 14:56 | |
| jgriffith | I'll let scottda look at if he wishes to do so | 14:56 |
| *** markstur has quit IRC | 14:56 | |
| *** markstur has joined #openstack-cinder | 14:57 | |
| *** markstur has quit IRC | 14:57 | |
| mriedem | scottda: it won't return a tuple | 14:57 |
| mriedem | it will return none | 14:57 |
| mriedem | None | 14:57 |
| mriedem | which will break here https://github.com/openstack/python-cinderclient/blob/2.1.0/cinderclient/api_versions.py#L254 | 14:57 |
| mriedem | because it can't unpack None into a tuple | 14:57 |
| jgriffith | aka vomit | 14:57 |
| mriedem | i'm trying to figure out what the best way, as a client (nova), to determine if we can do a given microversion, and i'm lost between several different methods in the cinderclient code | 14:58 |
| mriedem | there is get_highest_version https://github.com/openstack/python-cinderclient/blob/2.1.0/cinderclient/api_versions.py#L247 | 14:58 |
| mriedem | discover_version https://github.com/openstack/python-cinderclient/blob/2.1.0/cinderclient/api_versions.py#L259 | 14:58 |
| mriedem | and get_server_version https://github.com/openstack/python-cinderclient/blob/2.1.0/cinderclient/client.py#L74 | 14:58 |
| mriedem | oh and get_highest_client_server_version | 14:58 |
| *** arahal_ has joined #openstack-cinder | 14:59 | |
| smcginnis | mriedem: Usually bswartz would have jumped in by now, so I'll play the part. | 14:59 |
| mriedem | we'll use that to set the global in nova's volume api client and key off that to know if we can do new attachment stuff | 14:59 |
| smcginnis | mriedem: What about just callingn the new API and handling the failure to know it's not supported? | 14:59 |
| scottda | jgriffith: Feel free to put up a patch if you'd like. I may not get to it until later... | 15:00 |
| scottda | Still trying to get some cycles for this. | 15:00 |
| scottda | mriedem: Sorry if the documentation isn't clear on the version calls. Depends on whether you have the keystone service catalog available... | 15:01 |
| mriedem | scottda: we do have the url | 15:01 |
| scottda | yup | 15:02 |
| mriedem | smcginnis: b/c i don't know how that's going to fail | 15:02 |
| mriedem | will it 404? and does that 404 mean the microversion doesn't exist or that the volume on which we tried to create the attachment doesn't exist? | 15:02 |
| *** markstur has joined #openstack-cinder | 15:03 | |
| smcginnis | scottda: What's the return if you specify a microversion that isn't supported? I didn't think it was a 404. | 15:03 |
| *** isaacb has joined #openstack-cinder | 15:04 | |
| scottda | If it's < v3, the microversion header is ignored. | 15:04 |
| mriedem | scottda: https://github.com/openstack/nova/blob/master/nova/volume/cinder.py#L81 | 15:04 |
| * mordred heads people are talking about consuming versions | 15:04 | |
| scottda | So depends on if the call exists in v2 or not | 15:04 |
| *** hoonetorg has quit IRC | 15:04 | |
| *** kfarr has joined #openstack-cinder | 15:04 | |
| mordred | scottda, smcginnis: SO - if y'all haven't read it yet, https://review.openstack.org/#/q/status:open+project:openstack/api-wg+branch:master+topic:version-discovery | 15:04 |
| scottda | if v3, It returns something meaningful, but I don't recall the details. | 15:05 |
| mordred | is the write up that attempts to describe how to consume versions and microversions consistently across all openstack services with both modern state and history accounted for | 15:05 |
| mordred | it would be great if y'all could verify it's not missing any corner cases (I'll also go read this scrollback) | 15:05 |
| mordred | AND | 15:05 |
| *** wxy| has joined #openstack-cinder | 15:05 | |
| mordred | I've got patches to ksa implemeting the logic up for review as well | 15:05 |
| scottda | Thanks. I'll have a look at those mordred | 15:06 |
| mriedem | mordred: so basically this http://docs-draft.openstack.org/10/459710/13/check/gate-api-wg-docs-ubuntu-xenial/7a8995b//doc/build/html/guidelines/consuming-version-discovery.html | 15:06 |
| smcginnis | mordred: Thanks, I was aware there was work being done, but haven't kept up with the latest. | 15:06 |
| mordred | so end-goal of that woul dbe that cinderclient ceases having any local logic handling either major version discovery or finding availble microversions | 15:06 |
| mriedem | right? | 15:06 |
| mordred | mriedem: yes | 15:06 |
| mordred | mriedem: also consuming-catalog.html ... becauseyou can't do discovery until you've got the endpoint from the catalog | 15:07 |
| mordred | but for this convo, consuming-version-discovery.html is the one | 15:07 |
| mriedem | mordred: nova, as the client, has the endpoint here https://github.com/openstack/nova/blob/master/nova/volume/cinder.py#L81 | 15:07 |
| mriedem | that's by default the volumev3 endpoint | 15:08 |
| mordred | mriedem: right. well, that changes too - which is why the consuming-catalog part is important to grok and get implemented | 15:08 |
| mriedem | is that related to what efried is working on in nova? | 15:08 |
| mordred | yes, it is | 15:08 |
| mordred | "I want block-storage v3" should properly find volumev3 if that's what's in the catalog once we're done | 15:09 |
| mordred | and should also find the v3 endpoint if the catalog has block-storage | 15:09 |
| openstackgerrit | Merged openstack/cinder master: Do proper cleanup if connect volume fails https://review.openstack.org/459453 | 15:09 |
| mordred | and in both cases should return microversion ranges if they exist | 15:09 |
| *** diablo_rojo has joined #openstack-cinder | 15:10 | |
| mordred | it's a long thread to pull on - but we're actually very close to actually having one impl that works for all the cases that we can all use - which should be really exciting | 15:10 |
| mordred | mriedem: with the ksa patches, instead of "session.get_endpoint" that efried code will call "session.get_endpoint_data" which will return a thing that has the endpoint url - but also the min/max microversions | 15:11 |
| mordred | mriedem: so from the nova side, you can do the normal microversion consume logic however you want - but you can know you will have the data | 15:11 |
| scottda | mriedem: So, I think get_highest_version is probably how you'd call cinder for the version info. | 15:12 |
| *** kaisers has quit IRC | 15:13 | |
| mriedem | scottda: that takes a client | 15:14 |
| mordred | yah - and once we have the ksa patch landed, "get_server_version" can call session.get_endpoint_data and get_highest_client_server_version can call that and grab the microversions - so the using thatshould be upwards compat | 15:14 |
| mriedem | i want to create the client based on the version i know i need (3.27) | 15:14 |
| bswartz | smcginnis: ? | 15:15 |
| mriedem | scottda: because i want to create the client with 3.27 so that all requests are made using 3.27, not 3.0 | 15:15 |
| scottda | mriedem: But you want to query cinder first to see if the version is >= 3.27 first? | 15:15 |
| mriedem | but, we could go either way | 15:15 |
| mriedem | scottda: i want to see if 3.27 is available | 15:15 |
| smcginnis | bswartz: I tried being you on how you usually argued microversions should be consumed, but I just don't have the same fire. :) | 15:15 |
| mriedem | which i think i can do with get_highest_client_server_version | 15:15 |
| scottda | mriedem: Yes, I think so. | 15:16 |
| mriedem | version discovery in cinder doesn't require a token does it? | 15:16 |
| mordred | yes. I agree | 15:16 |
| mordred | and no, it doesn't | 15:16 |
| scottda | mriedem: No | 15:16 |
| mriedem | ok | 15:16 |
| mriedem | whew | 15:16 |
| bswartz | it's okay to query for supported microversions, but it's faster to optimistically try the highest version you know about and negotiate down on failure | 15:16 |
| dfisher | good morning! could I hassle somebody to give https://review.openstack.org/#/c/466397/ a +A? I've got two +2s and the corresponding 3rd party ran successfully so things should be aligned to merge (I hope) | 15:16 |
| mriedem | bswartz: knowing what the failure is is the hard part | 15:16 |
| bswartz | microversion too high should be a pretty obvious error response | 15:16 |
| *** hoonetorg has joined #openstack-cinder | 15:17 | |
| mriedem | bswartz: it's not obvious to any of the cinder cores in channel right now :) | 15:17 |
| bswartz | if not then that should be fixed | 15:17 |
| mriedem | if it's a 404 or not | 15:17 |
| mriedem | and don't tell me to rely on parsing the error message from the 404 to know the difference :) | 15:17 |
| scottda | mriedem: Not obvious from memory. Sorry, have had to step back from the project. | 15:17 |
| mordred | so, there is ALSO work afoot to improve the efficiency of version querying - I'd suggest writing it to do the query and let the improvement to the process flow via the keystoneauth updates | 15:17 |
| mordred | because once that's in place, trying a version, failing and then querying will be less efficient | 15:18 |
| *** mvk has quit IRC | 15:19 | |
| mordred | mriedem: so I'd use get_highest_client_server_version for now. when ksa update 1 is out we can update cinderclient to use it. then when update 2 exists, it'll just magically get it and your flow will be clear and understandable | 15:19 |
| bswartz | mordred: it doesn't matter if it's less efficient if it's the uncommon case | 15:19 |
| bswartz | the common case is that the client versions matches the server version -- that's what should be optimized | 15:19 |
| mordred | bswartz: just saying - this is eing solved generally for all API consumers - there is no need to code around it specifically in nova | 15:19 |
| mriedem | i will hack a bit | 15:20 |
| mordred | and doing so makes nova's consumption of the api more brittle | 15:20 |
| mordred | because there are times when it actually won't match | 15:20 |
| bswartz | thing that can be done in 1 round trip should always be done in 1 round trip | 15:20 |
| mordred | yes | 15:20 |
| mordred | which is what the end result will be | 15:20 |
| mriedem | sure but again as a client i don't know how this is going to fail if 3.27 isn't supported on the server | 15:21 |
| mordred | which is why I'm saying flow it through ksa and you'll get there correctly for free and will degrade gracefully for non-standard cases | 15:21 |
| bswartz | I don't see a way to get 1 round trip without optimistically trying the highest version you know about | 15:21 |
| bswartz | not unless you have stateful clients (YUCK) | 15:22 |
| mordred | bswartz: I'll point you at the cloud profile spec as soon as I update it from latest feedback - but there was a session on this at the summit | 15:22 |
| bswartz | mordred: that would be appreciated | 15:22 |
| mordred | bswartz: tl;dr is "single global etag-supporting extremely cacheable profile document for a cloud that contians all of the contents of all of the version discovery documents" - client fetches profile once when they first interact with cloud, no more round trips needed. when it changes, etag in catalog will be updated so that a client can know just from the token that they need to refresh the cached | 15:24 |
| mordred | copy of the profile | 15:24 |
| mordred | bswartz: so after the very first api negotiation with the cloud, all version information can be used proactively without incurring additional round-trips | 15:24 |
| bswartz | mordred: yeah but your first API call is still 2 round trips... | 15:25 |
| mordred | bswartz: yes. your first API to the cloud ever | 15:25 |
| bswartz | mordred: how do you detect that the server has been upgraded? | 15:25 |
| scottda | https://www.irccloud.com/pastebin/qwW36dx3/ | 15:25 |
| scottda | mriedem: ^^^ | 15:25 |
| mordred | bswartz: etag in the token | 15:25 |
| scottda | That's for a cinder server with v3, but too high a version | 15:26 |
| mriedem | scottda: computeFault ? :) | 15:27 |
| mordred | bswartz: that will tell you if your cached copy of the profile is out of date - so you'll have an extra roundtrip once after a server version update - but the libs can handle that for you transparently | 15:27 |
| scottda | Hit v2 endpoint and it get's you the volume list (in the case of that command/curl) without error | 15:27 |
| scottda | mriedem: I'm not saying that's the correct return, I'm saying that's what the return is. | 15:27 |
| bswartz | mordred: so if the admin upgraded cinder since my last REST API call, the next API call will get a response that tells me that I'm using a version that not the newest? | 15:27 |
| mriedem | scottda: i know. 406 is the point. | 15:27 |
| scottda | mriedem: I don't recall the details, but I'm almost certain I copied the code from someone else. | 15:28 |
| scottda | :) | 15:28 |
| scottda | Manila or Nova | 15:28 |
| mordred | bswartz: the next time you get a token, you'll get an updated value and you can then go fetch a new doc | 15:28 |
| scottda | Not to shirk responsibilty... | 15:28 |
| bswartz | mordred: does that time have an upper bound? | 15:28 |
| bswartz | to limit the maximum time a client uses an old API version? | 15:29 |
| mordred | token age | 15:29 |
| mriedem | scottda: looks like nova is similar | 15:29 |
| mordred | which is usually an hour or less iirc | 15:29 |
| bswartz | k | 15:29 |
| mordred | bswartz: in any case, I'll get the profile spec updated today and send it your way for review | 15:30 |
| mordred | bswartz: this is an issue across all api consumers - and we've been hacking and avoiding all of our discovery info for ages because of the double-api hit fear | 15:30 |
| mordred | which has made all of the clients unusably brittle | 15:30 |
| mordred | which is sad - given there is excellent data availabe so they don't have to be brittle | 15:31 |
| *** zhaochao has quit IRC | 15:31 | |
| mordred | so it's important we solve this well | 15:31 |
| mriedem | found another issue in cinderclient which could be nasty https://github.com/openstack/python-cinderclient/blob/2.1.0/cinderclient/api_versions.py#L247 | 15:32 |
| mriedem | ^ calls _get_server_version_range | 15:32 |
| mriedem | which calls client.services.server_api_version() | 15:32 |
| mriedem | but server_api_version() is only implemented on the cinderclient.v3.client.services manager | 15:32 |
| mriedem | it'll be an AttributeError on v1 or v2 clients | 15:32 |
| jgriffith | house of cards | 15:32 |
| mriedem | i guess the docstring says "Queries the server version info and returns highest supported microversion" which implies it should be a v3 client | 15:33 |
| mriedem | i might push a patch for that though | 15:33 |
| jgriffith | maybe I can just load up an old Cinder later today and ferret some of this out | 15:34 |
| jgriffith | unless somebody else is on it | 15:34 |
| *** aarefiev is now known as aarefiev_afk | 15:34 | |
| bswartz | mordred: I agree that clients which are willing to store state should have a very robust way to know what version to use -- my only worry is that clients which want or need to be stateless should be allowed to optimistically try their request at a version of their choice before falling back to a more robust path using some kind of negotiation | 15:35 |
| mordred | bswartz: any client should absolutely be able to try that | 15:36 |
| scottda | mriedem: Yes, those client tools were written under the assumption nova would first call: | 15:37 |
| scottda | url = _SESSION.get_endpoint(auth, **service_parameters) | 15:37 |
| mordred | bswartz: I'm mostly just saying that keystoneauth, gophercloud, fog, openstack4j, jclouds and php-opencloud at least are all going to grow the ability to transparently handle this - and since nova is using cinderclient which uses keystoneauth - if nova just uses the methods to ask for the data up front, nova will get corect data now and will get that correct data more efficiently in the not too | 15:37 |
| mordred | distant future | 15:37 |
| scottda | And therefore know which version (v1,v2,v3) cinder was using | 15:37 |
| scottda | i.e. I went down this path: https://review.openstack.org/#/c/385682/3/nova/volume/cinder.py | 15:37 |
| scottda | mriedem: I get that you may not want to use the keystone session to get the endpoint version. That seemed like the way to go for me. | 15:40 |
| *** lpetrut has quit IRC | 15:47 | |
| openstackgerrit | wangxiyuan proposed openstack/cinder master: Return metadata in volume summary https://review.openstack.org/459143 | 15:50 |
| openstackgerrit | John Griffith proposed openstack/python-cinderclient master: cinderclient might not return version for V2 API https://review.openstack.org/469520 | 15:56 |
| *** ryanender has joined #openstack-cinder | 15:56 | |
| openstackgerrit | Matt Riedemann proposed openstack/python-cinderclient master: Handle AttributeError in _get_server_version_range https://review.openstack.org/469545 | 15:56 |
| *** cebruns has joined #openstack-cinder | 15:57 | |
| * jgriffith goes to put pants on before meeting | 16:00 | |
| jgriffith | too cold for shorts | 16:00 |
| mriedem | smcginnis: looks like you need to create an ocata series for python-cinderclient in LP https://bugs.launchpad.net/python-cinderclient/+bug/1694729/+nominate | 16:01 |
| openstack | Launchpad bug 1694729 in python-cinderclient "get_server_version does not handle older (non-v3) servers" [Medium,In progress] - Assigned to John Griffith (john-griffith) | 16:01 |
| *** Apoorva has joined #openstack-cinder | 16:03 | |
| *** Apoorva has quit IRC | 16:03 | |
| mriedem | jgriffith: ildikov: so i'll float 2 options by you for handling 3.27 when creating an attachment in nova, | 16:05 |
| *** winston-d_ has joined #openstack-cinder | 16:05 | |
| *** Apoorva has joined #openstack-cinder | 16:05 | |
| mriedem | 1) we discover the max version from the v3 client and key off that in code to see if we should call attachment_create, or | 16:06 |
| *** david-lyle has quit IRC | 16:06 | |
| mriedem | 2) when constructing cinderclient v3 client for the attachment_create call, we check if 3.27 is supported and if not, raise some specific error and handle that in the nova code to fallback to the old style | 16:06 |
| *** rcernin has quit IRC | 16:07 | |
| mriedem | once we create the attachment the new way, we are setting that on the bdm.attachment_id so everything after that just checks bdm.attachment_id, it doesn't need to worry about the microversion checking | 16:07 |
| mriedem | so i'm inclined to think #2 is cleaner | 16:07 |
| mriedem | and that way we don't need a global variable set for the max microversion | 16:07 |
| *** mnestratov has quit IRC | 16:07 | |
| *** tesseract has quit IRC | 16:08 | |
| jgriffith | mriedem I prefer option #2 personally | 16:08 |
| mriedem | me too | 16:08 |
| mriedem | i'll wip something on on that quick to see how it looks | 16:08 |
| mriedem | *up on | 16:08 |
| jgriffith | mriedem thanks | 16:08 |
| ildikov | mriedem: jgriffith: that one sounds good to me too | 16:08 |
| ildikov | mriedem: tnx! | 16:09 |
| *** kaisers has joined #openstack-cinder | 16:10 | |
| *** david-lyle has joined #openstack-cinder | 16:14 | |
| *** jwcroppe has quit IRC | 16:16 | |
| openstackgerrit | John Griffith proposed openstack/python-cinderclient master: cinderclient might not return version for V2 API https://review.openstack.org/469520 | 16:22 |
| openstackgerrit | wangxiyuan proposed openstack/cinder master: Ceph: Fix delete backup with non-existent pool https://review.openstack.org/461082 | 16:25 |
| openstackgerrit | Gorka Eguileor proposed openstack/cinder master: Allow logging in tempest tests https://review.openstack.org/469502 | 16:27 |
| openstackgerrit | Gorka Eguileor proposed openstack/cinder master: Add support for OS-Brick force disconnect https://review.openstack.org/459454 | 16:27 |
| openstackgerrit | Gorka Eguileor proposed openstack/cinder master: Multipath tempest scenarios https://review.openstack.org/469445 | 16:27 |
| *** mewald has joined #openstack-cinder | 16:29 | |
| *** eharney has quit IRC | 16:29 | |
| *** eharney has joined #openstack-cinder | 16:31 | |
| openstackgerrit | wangxiyuan proposed openstack/cinder-specs master: Support get volume metadata summary https://review.openstack.org/466691 | 16:33 |
| openstackgerrit | Merged openstack/cinder master: GPFS: Fix forceful delete for consistency group https://review.openstack.org/468760 | 16:35 |
| *** kaisers has quit IRC | 16:37 | |
| openstackgerrit | Merged openstack/cinder master: ZFSSA: Allow clones to differ in size from parent https://review.openstack.org/466397 | 16:39 |
| *** mewald has quit IRC | 16:45 | |
| *** Poornima has joined #openstack-cinder | 16:45 | |
| *** rcernin has joined #openstack-cinder | 16:46 | |
| *** mewald has joined #openstack-cinder | 16:48 | |
| *** mvk has joined #openstack-cinder | 16:49 | |
| *** pcaruana has quit IRC | 16:50 | |
| *** kaisers has joined #openstack-cinder | 16:54 | |
| jgriffith | dims btw, pushing the snapshot additions to gophercloud today, then possibly the V3 attachment stuff | 16:55 |
| dims | cool jgriffith | 16:55 |
| jgriffith | will probably be looking at the start you pointed out in the cloud-provider changes next | 16:55 |
| *** salv-orlando has quit IRC | 16:55 | |
| openstackgerrit | Gorka Eguileor proposed openstack/cinder master: Allow logging in tempest tests https://review.openstack.org/469502 | 16:56 |
| openstackgerrit | Gorka Eguileor proposed openstack/cinder master: Add support for OS-Brick force disconnect https://review.openstack.org/459454 | 16:56 |
| openstackgerrit | Gorka Eguileor proposed openstack/cinder master: Multipath tempest scenarios https://review.openstack.org/469445 | 16:56 |
| openstackgerrit | Walter A. Boring IV (hemna) proposed openstack/python-brick-cinderclient-ext master: WIP Add new v3 attach/detach workflow https://review.openstack.org/469594 | 16:56 |
| dims | ack jgriffith you know where to find me :) | 16:57 |
| jgriffith | dims indeed :) | 16:59 |
| *** wxy| has quit IRC | 16:59 | |
| *** gaurangt_ has joined #openstack-cinder | 16:59 | |
| *** mdovgal has joined #openstack-cinder | 17:00 | |
| mewald | Is it right that Cinder with Ceph does not support "cinder migrate"? | 17:00 |
| *** e0ne has quit IRC | 17:00 | |
| *** dikonoor has joined #openstack-cinder | 17:00 | |
| *** gaurangt- has joined #openstack-cinder | 17:01 | |
| winston-d | jgriffith: sorry I didn't chime in in the meeting just now. | 17:02 |
| *** gaurangt has quit IRC | 17:03 | |
| winston-d | I'm definitely interested in making Cinder standalone, working perfectly with k8s. | 17:03 |
| patrickeast | anyone have a tl;dr for the force detach connector thing with tempest? i'm trying to resurrect my ci and running into that test failing... trying to figure out where things stood | 17:03 |
| *** gaurangt_ has quit IRC | 17:04 | |
| *** jwcroppe has joined #openstack-cinder | 17:04 | |
| winston-d | We are moving fast to k8s, by the end of this year, a major part of our applications will be deployed w/ k8s + baremetals. | 17:05 |
| *** isaacb has quit IRC | 17:05 | |
| winston-d | We still need Cinder 'cos there are a lot of nice stuff we have done in it. | 17:06 |
| smcginnis | patrickeast: We discussed it in a meeting a few weeks back. | 17:07 |
| *** wN has quit IRC | 17:07 | |
| xyang1 | winston-d: just Cinder standalone without other openstack components? | 17:07 |
| smcginnis | patrickeast: The agreement at the time was if no connector is provided, the driver should remove all attachments. | 17:07 |
| patrickeast | smcginnis: yea i'm trying to find the eavesdrop | 17:07 |
| smcginnis | patrickeast: Can't remember even which month that was in now. | 17:08 |
| patrickeast | smcginnis: oh, interesting | 17:08 |
| xyang1 | winston-d: I mean your k8s deployment | 17:08 |
| patrickeast | smcginnis: so i guess multi-attach is SOL? | 17:08 |
| smcginnis | ? | 17:08 |
| patrickeast | if you pass in no connector to cleanup one host, it cleans up all the attachments | 17:08 |
| patrickeast | which would probably be the other host it was on? | 17:09 |
| * patrickeast should probably just keep looking for the meeting notes | 17:09 | |
| smcginnis | patrickeast: It shouldn't ever happen in a normal scenario like that. | 17:09 |
| patrickeast | smcginnis: once (if) we have multi-attach how does the driver know which initiator to clean up the connection for though? | 17:10 |
| winston-d | xyang1: our k8s? we still have keystone, and we use ironic to do BM provisioning | 17:11 |
| patrickeast | smcginnis: i guess we store more info with the new attach apis... | 17:11 |
| *** ChubYann has joined #openstack-cinder | 17:11 | |
| patrickeast | smcginnis: i'm just not gonna worry about it :D | 17:11 |
| xyang1 | winston-d: ok, just wondering if you still use openstack | 17:11 |
| smcginnis | patrickeast: In normal operation, a connector would be provided for the attachment that needs to be removed. | 17:11 |
| ildikov | patrickeast: detach is not sorted out with the new API either yet | 17:11 |
| ildikov | patrickeast: according to jgriffith it's easy though :) | 17:12 |
| winston-d | xyang1: we still have openstack, for sure. | 17:12 |
| xyang1 | winston-d: ok, good to know | 17:13 |
| winston-d | we even have ESX/vCenter running. long long tail. | 17:13 |
| jgriffith | patrickeast smcginnis are you guys mixing the new attach api's and the old? | 17:13 |
| jgriffith | sorry... was away getting coffee | 17:13 |
| patrickeast | jgriffith: probably | 17:13 |
| jgriffith | but the pot was empty :( | 17:13 |
| patrickeast | jgriffith: how does the new attach stuff work with force-detach? | 17:14 |
| patrickeast | or does it? | 17:14 |
| jgriffith | patrickeast there's no such thing as force detach really | 17:14 |
| jgriffith | You tell cinder "attachment_delete <attachment-id>" cinder deletes that attachment | 17:14 |
| jgriffith | including the call to the driver to do what it needs to do | 17:15 |
| patrickeast | yea, that sounds right | 17:15 |
| smcginnis | jgriffith: This is just the old stuff. | 17:15 |
| patrickeast | and cinder has the connector info for the attach so we always have the right params for the driver.... right? | 17:15 |
| jgriffith | smcginnis no, I'm describing the new stuff | 17:15 |
| jgriffith | patrickeast yes | 17:15 |
| jgriffith | smcginnis am I confused? | 17:15 |
| patrickeast | jgriffith: ok cool so thats good | 17:15 |
| smcginnis | jgriffith: Right. I'm answering you're earlier question. We were talking about the old stuff. | 17:15 |
| smcginnis | *your | 17:16 |
| patrickeast | so yea, for the old one the force-detach call doesn't do that | 17:16 |
| jgriffith | patrickeast so the goal was to access evertyhing by attachment_id so we don't have to worry about the multi-attach case and knowing what's what | 17:16 |
| patrickeast | its just for some volume, and maybe a connector | 17:16 |
| * jgriffith has moved on... old attach/detach code is dead to him | 17:16 | |
| jgriffith | now if I could just fast fwd 2 years so everybody has it :) | 17:16 |
| patrickeast | the old api calls will not work on new api attaches, right? | 17:16 |
| jgriffith | patrickeast not really no | 17:17 |
| ildikov | I might not bring up my confusion with the shared connections stuff with multi-attach at this point | 17:17 |
| patrickeast | like... if i have (2 years from now) a multi-attached new style attached volume and call force-delete | 17:17 |
| patrickeast | it breaks? | 17:17 |
| jgriffith | ildikov *might not* :) | 17:17 |
| ildikov | so that no one wants to kill me :) | 17:17 |
| patrickeast | lol | 17:17 |
| patrickeast | ildikov: its ok, all this is hypothetical | 17:17 |
| jgriffith | ildikov who would possibly want to kill you? | 17:17 |
| jgriffith | :) | 17:17 |
| jgriffith | patrickeast so do you mean like attach with the old and detach with the new? | 17:18 |
| patrickeast | more importantly the other way around | 17:18 |
| ildikov | jgriffith: it just felt like thin ice :) | 17:18 |
| jgriffith | patrickeast or some other weird frankenstien mix/match | 17:18 |
| jgriffith | attach with new, detach with old? | 17:18 |
| *** Poornima has quit IRC | 17:18 | |
| patrickeast | i guess what i'm trying to decide is if i should change my driver to allow terminate_connection with no connector and to detach *all* attachments | 17:18 |
| patrickeast | just so it passes the stupid tempest test | 17:18 |
| jgriffith | in theory that *would* work but you're going to be all screwed up | 17:19 |
| jgriffith | Ahhhh! | 17:19 |
| jgriffith | patrickeast if tempest: detach-all | 17:19 |
| jgriffith | :) | 17:19 |
| patrickeast | lol | 17:19 |
| jgriffith | who wrote the test? | 17:19 |
| patrickeast | dunno | 17:19 |
| patrickeast | but it works on lvm | 17:19 |
| patrickeast | so its correct, right? | 17:19 |
| jgriffith | of is it that same old force-detach one? | 17:19 |
| patrickeast | right?! | 17:19 |
| jgriffith | LOL | 17:19 |
| jgriffith | well YES | 17:19 |
| patrickeast | lol | 17:20 |
| jgriffith | no shared connection stuff so life is EZ/PZ | 17:20 |
| jgriffith | patrickeast you should change how Pure works and make it more like LVM... then all your problems will be solved :) | 17:20 |
| patrickeast | jgriffith: not the worst idea | 17:20 |
| *** gaurangt has joined #openstack-cinder | 17:21 | |
| patrickeast | ok, well... i guess i've caused enough trouble with attach apis | 17:21 |
| patrickeast | jgriffith: got a min to talk about gos-brick? | 17:22 |
| jgriffith | patrickeast so for the tempest test I wouldn't try and mix/match anything to get around it | 17:22 |
| jgriffith | it will likely come back to bight you | 17:22 |
| patrickeast | jgriffith: i'm thinking i'll just turn off the test... | 17:22 |
| jgriffith | although notice that the attachment calls are just wrappers to the old stuff | 17:22 |
| jgriffith | patrickeast +1 | 17:22 |
| jgriffith | that's what I'd do honeslty | 17:22 |
| patrickeast | jgriffith: and hope it doesn't show up on some vendors cert list | 17:22 |
| jgriffith | meh | 17:22 |
| jgriffith | only on SolidFires :) | 17:22 |
| jgriffith | err... NetApps | 17:22 |
| patrickeast | lol | 17:23 |
| jgriffith | anyway.. yeah... gos-brick? | 17:23 |
| patrickeast | ah yea | 17:23 |
| hemna | gos-brick | 17:23 |
| patrickeast | so, were you thinking of just a kind of port over of os-brick? | 17:23 |
| *** Julien-zte has quit IRC | 17:23 | |
| patrickeast | or an opportunity to make things better? | 17:23 |
| jgriffith | patrickeast maybe both? | 17:23 |
| *** gaurangt- has quit IRC | 17:23 | |
| hemna | is there a way to wrap python calls in go ? | 17:24 |
| jgriffith | Honestly my initial incling was the rest-server in front of os-brick, but it was pointed out that might not be such a hot idea | 17:24 |
| *** wN has joined #openstack-cinder | 17:24 | |
| *** wN has quit IRC | 17:24 | |
| *** wN has joined #openstack-cinder | 17:24 | |
| jgriffith | to then the idea of actually creating a go version / pkg came up and honestly I think that's a great idea | 17:24 |
| jgriffith | it's something I have always been amazed there isn't something out there for already | 17:24 |
| hemna | are the FC HBAs even available in a docker/k8s container ? | 17:24 |
| patrickeast | yea, so, i ask because the more time i spend trying to get the attach stuff to work on multiple os's and platforms the more i'm convinced that it isn't something a library with like synchronous attach/detach calls can really work | 17:25 |
| jgriffith | hemna you mean like the old swig style stuff? | 17:25 |
| jgriffith | Run away!!! | 17:25 |
| hemna | or even the /proc filesystem | 17:25 |
| patrickeast | hemna: yea they can be | 17:25 |
| hemna | yah swig sux0rs, but that's for python calling C | 17:25 |
| jgriffith | patrickeast yeah, so the idea isn't necessarily a direct port of os-brick, but something that does the same *stuff* | 17:25 |
| patrickeast | jgriffith: what i've been thinking about that would work much much better is a volume service | 17:25 |
| jgriffith | if there's room for improvement along the way then that's awesome | 17:25 |
| patrickeast | and will probably end up with something like that for the pure k8s and docker stuff longer term | 17:26 |
| jgriffith | patrickeast not sure I track... can you explain more? | 17:26 |
| patrickeast | more similar to an api call of "i expect this volume to be mounted here, use this info" and then the service makes sure that happens | 17:26 |
| hemna | part of the problem I see is overcoming the need to access hw interfaces inside the containers | 17:26 |
| patrickeast | the issue being that over time you need to add/remove paths, update things, etc | 17:26 |
| jgriffith | patrickeast ahh, and have that be it's own service.... | 17:27 |
| patrickeast | plus the workflow is a nightmare with retries and stuff | 17:27 |
| jgriffith | interesting | 17:27 |
| patrickeast | yea | 17:27 |
| *** Julien-zte has joined #openstack-cinder | 17:27 | |
| hemna | isn't that what you get with rest + os-brick ? | 17:27 |
| jgriffith | hemna meh, that's not so bad actually | 17:27 |
| jgriffith | --device /dev/* | 17:27 |
| jgriffith | :) | 17:27 |
| patrickeast | yea hw stuff is easy in containers | 17:27 |
| * jgriffith ducks | 17:27 | |
| xyang1 | hemna: jgriffith does this look like a FC driver for k8s? https://github.com/kubernetes/kubernetes/tree/master/pkg/volume/fc | 17:28 |
| patrickeast | make sure you pass through /sys:/sys | 17:28 |
| jgriffith | unless you want to follow the new craze of "don't run privileged" then your SOL :) | 17:28 |
| patrickeast | pff | 17:28 |
| patrickeast | thats crazy talk | 17:28 |
| hemna | heh | 17:28 |
| jgriffith | xyang1 yeah | 17:29 |
| jgriffith | xyang1 so K8's has iscsi and other *stuffs* built in FWIW | 17:29 |
| jgriffith | xyang1 it's just not nearly as extensive or feature rich as what we've got | 17:30 |
| xyang1 | jgriffith: ok | 17:30 |
| jgriffith | patrickeast crazies are us | 17:30 |
| patrickeast | jgriffith: haha, maybe | 17:30 |
| hemna | multipath support? | 17:31 |
| jgriffith | anyway, the services idea running in a container/pod etc might be an interesting angle | 17:31 |
| patrickeast | hemna: nope :( | 17:31 |
| hemna | pewp | 17:31 |
| patrickeast | jgriffith: you kinda need something there for like flexvols anyway | 17:31 |
| jgriffith | and probably going to be a requirement at some point | 17:31 |
| jgriffith | patrickeast yeah | 17:31 |
| hemna | we need a new multipath daemon..... | 17:32 |
| hemna | that doesn't suck | 17:32 |
| patrickeast | i guess what i would (very selfishly) like to see is gos-brick either be like a thin helper lib that can do the individual iscsiadm, sys calls, etc, that we can build that service out of, or the actual service | 17:33 |
| jgriffith | patrickeast so I think the service idea is the right way to go, just trying to wrap my head around what the interface would be and how you'd call it etc | 17:33 |
| patrickeast | where its at right now the one-time synchronous api calls to attach aren't as useful to make a robust attacher thingy | 17:33 |
| jgriffith | patrickeast just link to the pod/container | 17:33 |
| hemna | so if we want a service, then why not use os-brick behind the rest API ? | 17:33 |
| jgriffith | call it as an executable... 'attach my-shit here' | 17:33 |
| hemna | all of those connectors are already there. | 17:33 |
| patrickeast | yea, i mean, it could be pretty flexible and use a REST-like interface or something on a local socket | 17:34 |
| patrickeast | for cli, or other bins | 17:34 |
| patrickeast | or even a tcp interface | 17:34 |
| jgriffith | smcginnis pointed out there are some security issues/concerns with that sort of thing | 17:34 |
| patrickeast | ala rest+os-brick | 17:34 |
| jgriffith | the Rest piece that is | 17:34 |
| patrickeast | ah, yea, thats good to know | 17:35 |
| jgriffith | TBH the only real problem I have with that approach is it's SOOO HEAVY | 17:35 |
| patrickeast | yea :( | 17:35 |
| patrickeast | the shitty part is that it seems like it needs to be | 17:35 |
| jgriffith | patrickeast yeah, that may be the sad reality | 17:35 |
| patrickeast | to hemna's point, we need to like rewrite multipathd to make this junk work | 17:35 |
| jgriffith | wasn't geguileo doing that already :) | 17:36 |
| patrickeast | yea, but for being able to work with a wide variety of systems it needs to work with other versions | 17:36 |
| jgriffith | or stop using multipath :) | 17:36 |
| patrickeast | well, its not just multipath | 17:36 |
| jgriffith | lalalalalaallallaallaala | 17:36 |
| patrickeast | from what i've seen udevd is the real killer | 17:36 |
| * jgriffith can't hear you | 17:36 | |
| jgriffith | damn it | 17:36 |
| jgriffith | I KNEW you were going there!!! | 17:36 |
| patrickeast | lol | 17:36 |
| geguileo | jgriffith: Yeah, I just sent an email to the ML discussing it and with tempest tests to test multipath with error injection | 17:37 |
| hemna | bah! good luck with not using mpath :P | 17:37 |
| jgriffith | hemna I dunno, I seem to be having better luck not using it than you suckers do using it :) | 17:39 |
| jgriffith | although my days are numbered | 17:39 |
| jgriffith | patrickeast hemna anyway... it would be good to noodle on the gos-brick/rest-brick thing a bit and get some more input from people | 17:40 |
| jgriffith | what I meant by heavy earlier was more against the python pieces than anything else | 17:40 |
| xyang1 | hemna: a related question. I somehow remembered FC multipath is auto detected and cannot be turned off by the multipath flag, but it looks like I am wrong. In the code, it does use the multipath flag to determine whether to use it or not | 17:40 |
| *** dhinesh has joined #openstack-cinder | 17:41 | |
| xyang1 | hemna: was this changed when os-brick was first created? | 17:42 |
| patrickeast | jgriffith: sounds good | 17:42 |
| hemna | xyang1, I think the default was to use mpath if it was on | 17:44 |
| hemna | or use the iscsi_use_multipath flag originally | 17:44 |
| xyang1 | hemna: that is the nova side flag. It doesn't look like that flag is doing anything though. It all depends on what is returned from connection info | 17:48 |
| *** xyang has joined #openstack-cinder | 17:49 | |
| hemna | xyang, the FC driver doesn't try to use mpath if the connector was created with the flag being false | 17:49 |
| hemna | https://github.com/openstack/os-brick/blob/master/os_brick/initiator/connectors/fibre_channel.py#L185 | 17:49 |
| xyang | hemna: I see. So mpath can be turned off if both flags (in cinder and nova) are set to false | 17:54 |
| *** cebruns has quit IRC | 17:56 | |
| *** eharney has quit IRC | 17:56 | |
| *** nicolasbock has quit IRC | 17:58 | |
| *** kaisers has quit IRC | 17:59 | |
| *** chlong has quit IRC | 18:05 | |
| *** jsuchome has quit IRC | 18:09 | |
| *** eharney has joined #openstack-cinder | 18:09 | |
| *** cebruns has joined #openstack-cinder | 18:10 | |
| *** chhavi has quit IRC | 18:17 | |
| *** cebruns has quit IRC | 18:21 | |
| *** chlong has joined #openstack-cinder | 18:22 | |
| *** cebruns has joined #openstack-cinder | 18:36 | |
| *** kaisers has joined #openstack-cinder | 18:37 | |
| *** cebruns has joined #openstack-cinder | 18:38 | |
| *** cebruns has quit IRC | 18:39 | |
| *** cebruns has joined #openstack-cinder | 18:41 | |
| *** cebruns has quit IRC | 18:41 | |
| *** cebruns has joined #openstack-cinder | 18:42 | |
| *** cebruns has joined #openstack-cinder | 18:46 | |
| *** dikonoor has quit IRC | 18:50 | |
| *** arnewiebalck_ has joined #openstack-cinder | 18:59 | |
| *** adisky__ has quit IRC | 19:01 | |
| *** gkadam has joined #openstack-cinder | 19:02 | |
| *** arnewiebalck_ has quit IRC | 19:05 | |
| *** arnewiebalck_ has joined #openstack-cinder | 19:12 | |
| *** lucasxu has quit IRC | 19:13 | |
| *** crushil has quit IRC | 19:14 | |
| *** crushil has joined #openstack-cinder | 19:17 | |
| *** bjornar_ has joined #openstack-cinder | 19:22 | |
| *** lpetrut has joined #openstack-cinder | 19:23 | |
| *** pcaruana has joined #openstack-cinder | 19:45 | |
| *** arnewiebalck_ has quit IRC | 19:47 | |
| *** mtanino__ has quit IRC | 19:52 | |
| *** salv-orlando has joined #openstack-cinder | 20:00 | |
| *** pcaruana has quit IRC | 20:02 | |
| *** crushil has quit IRC | 20:02 | |
| *** mnestratov has joined #openstack-cinder | 20:05 | |
| *** liverpooler has quit IRC | 20:06 | |
| *** gkadam has quit IRC | 20:07 | |
| *** gkadam has joined #openstack-cinder | 20:08 | |
| *** e0ne has joined #openstack-cinder | 20:09 | |
| *** salv-orlando has quit IRC | 20:09 | |
| *** salv-orlando has joined #openstack-cinder | 20:11 | |
| e0ne | mewald: hi. no, migration with ceph backend should work | 20:13 |
| *** crushil has joined #openstack-cinder | 20:17 | |
| *** crushil has quit IRC | 20:18 | |
| *** crushil has joined #openstack-cinder | 20:19 | |
| *** lucasxu has joined #openstack-cinder | 20:20 | |
| *** rcernin has quit IRC | 20:21 | |
| *** lucasxu has quit IRC | 20:21 | |
| *** rcernin has joined #openstack-cinder | 20:22 | |
| *** crushil has quit IRC | 20:23 | |
| *** crushil has joined #openstack-cinder | 20:24 | |
| *** crushil has quit IRC | 20:26 | |
| *** crushil has joined #openstack-cinder | 20:27 | |
| *** valeska has joined #openstack-cinder | 20:29 | |
| *** crushil has quit IRC | 20:32 | |
| *** crushil has joined #openstack-cinder | 20:33 | |
| valeska | hola?? | 20:33 |
| *** kfarr has quit IRC | 20:35 | |
| *** valeska has left #openstack-cinder | 20:35 | |
| *** rcernin has quit IRC | 20:42 | |
| openstackgerrit | Merged openstack/cinder master: Remove unused _detach_snapshot method https://review.openstack.org/469499 | 20:49 |
| *** chlong has quit IRC | 20:53 | |
| *** ericyoung has quit IRC | 20:58 | |
| *** jonesn has quit IRC | 21:00 | |
| *** winston-d_ has quit IRC | 21:00 | |
| openstackgerrit | Walter A. Boring IV (hemna) proposed openstack/python-brick-cinderclient-ext master: Add new v3 attach/detach workflow https://review.openstack.org/469594 | 21:05 |
| *** Apoorva_ has joined #openstack-cinder | 21:05 | |
| *** chlong has joined #openstack-cinder | 21:05 | |
| *** e0ne has quit IRC | 21:07 | |
| *** Apoorva has quit IRC | 21:07 | |
| openstackgerrit | xing-yang proposed openstack/python-cinderclient master: Tiramisu: replication group support https://review.openstack.org/352229 | 21:11 |
| *** pcaruana has joined #openstack-cinder | 21:12 | |
| *** kaisers has quit IRC | 21:15 | |
| *** salv-orlando has quit IRC | 21:16 | |
| *** harlowja has quit IRC | 21:21 | |
| *** pcaruana has quit IRC | 21:29 | |
| *** lpetrut has quit IRC | 21:30 | |
| *** MVenesio has quit IRC | 21:31 | |
| *** MVenesio has joined #openstack-cinder | 21:31 | |
| *** MVenesio has quit IRC | 21:32 | |
| *** abishop has quit IRC | 21:33 | |
| *** ryanender has quit IRC | 21:34 | |
| *** crushil has quit IRC | 21:39 | |
| *** bjornar_ has quit IRC | 21:40 | |
| *** eharney has quit IRC | 21:42 | |
| *** kaisers has joined #openstack-cinder | 21:45 | |
| *** jbernard has quit IRC | 21:47 | |
| *** mewald has quit IRC | 21:49 | |
| stvnoyes | i'm using devstack and just updated nova/cinder/etc to latest master. I am hitting this exception a lot- AttributeError: 'RequestContext' object has no attribute 'global_id' any idea why my context obj now needs an id? | 21:50 |
| stvnoyes | this is being thrown by the cinder client | 21:51 |
| *** jbernard has joined #openstack-cinder | 21:53 | |
| *** felipemonteiro has quit IRC | 21:53 | |
| *** Apoorva_ has quit IRC | 21:54 | |
| *** Apoorva has joined #openstack-cinder | 21:54 | |
| *** Apoorva has quit IRC | 21:54 | |
| ildikov | hmm, I don't recall that one | 21:54 |
| *** Apoorva has joined #openstack-cinder | 21:55 | |
| *** xyang has quit IRC | 21:56 | |
| *** Apoorva_ has joined #openstack-cinder | 21:57 | |
| *** Apoorva has quit IRC | 21:57 | |
| ildikov | although I had issues with openrc at some point, repeating the command seemed to solve it, no idea what happened there :/ | 21:58 |
| *** felipemonteiro has joined #openstack-cinder | 21:58 | |
| *** xyang has joined #openstack-cinder | 21:59 | |
| *** felipemonteiro__ has joined #openstack-cinder | 21:59 | |
| *** edmondsw has quit IRC | 22:01 | |
| *** edmondsw has joined #openstack-cinder | 22:01 | |
| *** felipemonteiro has quit IRC | 22:03 | |
| *** edmondsw_ has joined #openstack-cinder | 22:03 | |
| hemna | unset OS_AUTH_TYPE | 22:04 |
| *** edmondsw has quit IRC | 22:05 | |
| *** harlowja has joined #openstack-cinder | 22:06 | |
| *** edmondsw_ has quit IRC | 22:07 | |
| *** felipemonteiro__ has quit IRC | 22:08 | |
| ildikov | hemna: I did that, had to look up the channel logs though, but I finally found it :) | 22:08 |
| stvnoyes | hemna: OS_AUTH_TYPE wasn't set in my env. | 22:12 |
| *** gouthamr has quit IRC | 22:20 | |
| *** dfisher has quit IRC | 22:23 | |
| *** xyang1 has quit IRC | 22:28 | |
| *** edmondsw has joined #openstack-cinder | 22:31 | |
| *** edmondsw has quit IRC | 22:35 | |
| *** jose-phi_ has quit IRC | 22:42 | |
| *** jwcroppe has quit IRC | 22:48 | |
| *** jose-phillips has joined #openstack-cinder | 22:50 | |
| *** xyang has quit IRC | 22:57 | |
| *** ekoorjr has joined #openstack-cinder | 22:59 | |
| *** chlong has quit IRC | 23:00 | |
| *** xyang has joined #openstack-cinder | 23:02 | |
| *** mnestratov has quit IRC | 23:02 | |
| *** gouthamr has joined #openstack-cinder | 23:11 | |
| *** nicolasbock has joined #openstack-cinder | 23:16 | |
| *** liverpooler has joined #openstack-cinder | 23:36 | |
| *** Julien-zte has quit IRC | 23:37 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!