| *** jwcroppe has quit IRC | 00:01 | |
| *** apoorvad has quit IRC | 00:02 | |
| *** EinstCrazy has quit IRC | 00:04 | |
| *** apoorvad has joined #openstack-cinder | 00:05 | |
| *** alexpilotti has joined #openstack-cinder | 00:06 | |
| *** rhagarty_ has quit IRC | 00:09 | |
| *** rhagarty has joined #openstack-cinder | 00:11 | |
| *** apoorvad has quit IRC | 00:11 | |
| *** frakt has quit IRC | 00:15 | |
| *** frakt has joined #openstack-cinder | 00:16 | |
| *** rhagarty has quit IRC | 00:17 | |
| *** rhagarty has joined #openstack-cinder | 00:17 | |
| *** markvoelker has quit IRC | 00:26 | |
| *** apoorvad has joined #openstack-cinder | 00:27 | |
| *** dims has joined #openstack-cinder | 00:30 | |
| *** dims_ has quit IRC | 00:31 | |
| *** akshai_ has quit IRC | 00:31 | |
| *** jwcroppe has joined #openstack-cinder | 00:39 | |
| *** rhagarty_ has joined #openstack-cinder | 00:45 | |
| *** jerrygb has quit IRC | 00:45 | |
| *** jerrygb has joined #openstack-cinder | 00:46 | |
| *** rhagarty has quit IRC | 00:47 | |
| *** jerrygb has quit IRC | 00:50 | |
| *** bardia has quit IRC | 00:56 | |
| *** alexpilotti has quit IRC | 00:59 | |
| *** apoorvad has quit IRC | 00:59 | |
| *** alexpilotti has joined #openstack-cinder | 00:59 | |
| *** salv-orl_ has quit IRC | 01:01 | |
| *** EinstCrazy has joined #openstack-cinder | 01:05 | |
| *** zhangjn has joined #openstack-cinder | 01:05 | |
| *** swebb has quit IRC | 01:14 | |
| *** Lee1092 has joined #openstack-cinder | 01:21 | |
| *** swebb has joined #openstack-cinder | 01:21 | |
| *** markvoelker has joined #openstack-cinder | 01:27 | |
| *** alexpilotti has quit IRC | 01:31 | |
| *** markvoelker has quit IRC | 01:31 | |
| *** pots2 has quit IRC | 01:34 | |
| *** leeantho has quit IRC | 01:34 | |
| *** zhenguo has joined #openstack-cinder | 01:37 | |
| *** mudassirlatif has joined #openstack-cinder | 01:47 | |
| *** zhangjn has quit IRC | 01:49 | |
| *** dustins has quit IRC | 01:50 | |
| *** mriedem has joined #openstack-cinder | 01:52 | |
| *** zhangjn has joined #openstack-cinder | 01:52 | |
| *** dustins has joined #openstack-cinder | 01:56 | |
| *** jerrygb has joined #openstack-cinder | 01:59 | |
| *** haomaiwa_ has joined #openstack-cinder | 02:01 | |
| *** dustins has quit IRC | 02:02 | |
| *** smoriya_ has joined #openstack-cinder | 02:05 | |
| openstackgerrit | wuyuting proposed openstack/cinder: Declare multiattach is True in Dell Eqlx driver https://review.openstack.org/249516 | 02:07 |
|---|---|---|
| openstackgerrit | John Griffith proposed openstack/cinder: Add cluster Pairing setup to SolidFire driver https://review.openstack.org/238235 | 02:08 |
| *** mtanino has quit IRC | 02:13 | |
| *** dustins has joined #openstack-cinder | 02:15 | |
| openstackgerrit | wuyuting proposed openstack/cinder: Declare multiattach is True in Dell Eqlx driver https://review.openstack.org/249516 | 02:17 |
| openstackgerrit | John Griffith proposed openstack/cinder: Add replicate by volume-type to SolidFire driver https://review.openstack.org/238276 | 02:24 |
| openstackgerrit | John Griffith proposed openstack/cinder: Add replicate by volume-type to SolidFire driver https://review.openstack.org/238276 | 02:25 |
| *** markvoelker has joined #openstack-cinder | 02:28 | |
| *** mudassirlatif has quit IRC | 02:29 | |
| *** jerrygb has quit IRC | 02:32 | |
| *** markvoelker has quit IRC | 02:33 | |
| *** dustins has quit IRC | 02:33 | |
| *** jerrygb has joined #openstack-cinder | 02:33 | |
| openstackgerrit | Thang Pham proposed openstack/cinder: Implement refresh() for cinder objects https://review.openstack.org/248998 | 02:34 |
| *** dustins has joined #openstack-cinder | 02:35 | |
| *** dustins has quit IRC | 02:36 | |
| *** jerrygb has quit IRC | 02:37 | |
| *** arnewiebalck__ has joined #openstack-cinder | 02:38 | |
| *** dims has quit IRC | 02:39 | |
| *** jerrygb has joined #openstack-cinder | 02:40 | |
| *** arnewiebalck_ has quit IRC | 02:40 | |
| *** jwcroppe has quit IRC | 02:47 | |
| *** dustins has joined #openstack-cinder | 02:47 | |
| *** jwcroppe has joined #openstack-cinder | 02:48 | |
| *** dustins has quit IRC | 02:49 | |
| openstackgerrit | wanghao proposed openstack/python-cinderclient: support for snapshot management https://review.openstack.org/249090 | 02:49 |
| *** jwcroppe has quit IRC | 02:52 | |
| *** dustins has joined #openstack-cinder | 02:55 | |
| *** haomaiwa_ has quit IRC | 03:01 | |
| *** 6JTACJ1YS has joined #openstack-cinder | 03:01 | |
| *** salv-orlando has joined #openstack-cinder | 03:05 | |
| *** sghanekar__ has quit IRC | 03:07 | |
| *** breitz has quit IRC | 03:09 | |
| *** breitz has joined #openstack-cinder | 03:09 | |
| openstackgerrit | Thang Pham proposed openstack/cinder: Implement refresh() for cinder objects https://review.openstack.org/248998 | 03:13 |
| *** mriedem has quit IRC | 03:13 | |
| *** RA has joined #openstack-cinder | 03:14 | |
| *** RA is now known as Guest30762 | 03:14 | |
| *** Guest30762 is now known as RA_ | 03:16 | |
| *** jerrygb has quit IRC | 03:16 | |
| *** jerrygb has joined #openstack-cinder | 03:16 | |
| *** terryyao has joined #openstack-cinder | 03:20 | |
| openstackgerrit | wuyuting proposed openstack/cinder: Declare multiattach is True in LVMVolume driver https://review.openstack.org/249532 | 03:20 |
| *** mtanino has joined #openstack-cinder | 03:21 | |
| *** jerrygb has quit IRC | 03:21 | |
| *** markvoelker has joined #openstack-cinder | 03:29 | |
| openstackgerrit | wuyuting proposed openstack/cinder: Declare multiattach is True in LVMVolume driver https://review.openstack.org/249532 | 03:34 |
| *** changbl has quit IRC | 03:34 | |
| *** markvoelker has quit IRC | 03:34 | |
| *** markvoelker has joined #openstack-cinder | 03:35 | |
| *** martyturner has joined #openstack-cinder | 03:37 | |
| *** salv-orlando has quit IRC | 03:38 | |
| *** boris-42 has quit IRC | 03:38 | |
| *** martyturner1 has joined #openstack-cinder | 03:39 | |
| *** martyturner has quit IRC | 03:41 | |
| *** john5223 is now known as zz_john5223 | 03:41 | |
| *** martyturner1 has left #openstack-cinder | 03:44 | |
| *** changbl has joined #openstack-cinder | 03:47 | |
| *** jwcroppe has joined #openstack-cinder | 03:49 | |
| openstackgerrit | Edwin Wang proposed openstack/cinder: FlashSystem reports error in _find_host_exhaustive() https://review.openstack.org/247546 | 03:49 |
| *** zhangjn has quit IRC | 03:51 | |
| *** salv-orlando has joined #openstack-cinder | 03:51 | |
| *** dustins has quit IRC | 03:53 | |
| *** salv-orlando has quit IRC | 03:56 | |
| *** dustins has joined #openstack-cinder | 03:57 | |
| openstackgerrit | zhangsong proposed openstack/os-brick: Improve the remove_multipath_device() method for linuxscsi https://review.openstack.org/249308 | 03:57 |
| *** 6JTACJ1YS has quit IRC | 04:01 | |
| *** haomaiwang has joined #openstack-cinder | 04:01 | |
| openstackgerrit | Vincent Hou proposed openstack/cinder: Storwize: Implement v2 replication (global) https://review.openstack.org/249042 | 04:01 |
| openstackgerrit | Vincent Hou proposed openstack/cinder: Storwize: Implement v2 replication (split IO) https://review.openstack.org/237387 | 04:01 |
| openstackgerrit | LisaLi proposed openstack/cinder-specs: Retype encrypted volumes https://review.openstack.org/248593 | 04:02 |
| *** jwcroppe has quit IRC | 04:02 | |
| *** strictlyb is now known as sb | 04:05 | |
| *** Thelo has quit IRC | 04:07 | |
| *** sgundur has joined #openstack-cinder | 04:16 | |
| *** sgundur has left #openstack-cinder | 04:17 | |
| *** Thelo has joined #openstack-cinder | 04:17 | |
| *** zz_john5223 is now known as john5223 | 04:20 | |
| openstackgerrit | xing-yang proposed openstack/cinder: Backup snapshots https://review.openstack.org/243406 | 04:21 |
| *** zao has joined #openstack-cinder | 04:44 | |
| *** zao has left #openstack-cinder | 04:46 | |
| *** houming has joined #openstack-cinder | 04:47 | |
| *** gcb has joined #openstack-cinder | 04:48 | |
| *** mtanino has quit IRC | 04:59 | |
| *** jwcroppe has joined #openstack-cinder | 05:00 | |
| *** haomaiwang has quit IRC | 05:01 | |
| *** haomaiwang has joined #openstack-cinder | 05:01 | |
| openstackgerrit | xing-yang proposed openstack/cinder: Fix ScaleIO driver provisioning key https://review.openstack.org/249552 | 05:09 |
| openstackgerrit | LisaLi proposed openstack/cinder-specs: Retype encrypted volumes https://review.openstack.org/248593 | 05:10 |
| openstackgerrit | xing-yang proposed openstack/cinder: Fix ScaleIO driver provisioning key https://review.openstack.org/249552 | 05:15 |
| openstackgerrit | Merged openstack/cinder: Remove stubs and add resource cleanup https://review.openstack.org/249218 | 05:16 |
| *** salv-orlando has joined #openstack-cinder | 05:18 | |
| *** shausy has joined #openstack-cinder | 05:21 | |
| openstackgerrit | LisaLi proposed openstack/cinder: Delete unused codes in rbd.retype https://review.openstack.org/244460 | 05:22 |
| openstackgerrit | xing-yang proposed openstack/cinder: ScaleIO extend volume round up capacity https://review.openstack.org/249556 | 05:24 |
| *** salv-orlando has quit IRC | 05:24 | |
| *** zhangjn has joined #openstack-cinder | 05:28 | |
| *** haomaiwang has quit IRC | 06:01 | |
| *** haomaiwang has joined #openstack-cinder | 06:01 | |
| *** asselin has quit IRC | 06:04 | |
| *** asselin has joined #openstack-cinder | 06:04 | |
| *** houming has quit IRC | 06:07 | |
| openstackgerrit | Vincent Hou proposed openstack/cinder: Storwize: Implement v2 replication (global) https://review.openstack.org/249042 | 06:08 |
| openstackgerrit | Vincent Hou proposed openstack/cinder: Storwize: Implement v2 replication (split IO) https://review.openstack.org/237387 | 06:08 |
| *** nikeshm has quit IRC | 06:10 | |
| *** jiangliucheng has joined #openstack-cinder | 06:16 | |
| openstackgerrit | zhangsong proposed openstack/os-brick: Improve the remove_multipath_device() method for linuxscsi https://review.openstack.org/249308 | 06:17 |
| *** rcernin has joined #openstack-cinder | 06:20 | |
| *** nikeshm has joined #openstack-cinder | 06:23 | |
| openstackgerrit | Vipin Balachandran proposed openstack/cinder: VMware: Unit test refactoring (image to vol - 1/2) https://review.openstack.org/249174 | 06:23 |
| *** sgotliv has joined #openstack-cinder | 06:25 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/cinder: Imported Translations from Zanata https://review.openstack.org/249029 | 06:26 |
| *** nkrinner has joined #openstack-cinder | 06:26 | |
| *** houming has joined #openstack-cinder | 06:26 | |
| *** lpetrut has joined #openstack-cinder | 06:31 | |
| *** zhangjn has quit IRC | 06:42 | |
| *** jwcroppe has quit IRC | 06:47 | |
| *** ankit_ag has joined #openstack-cinder | 06:47 | |
| *** vgridnev has joined #openstack-cinder | 06:54 | |
| *** haomaiwang has quit IRC | 07:01 | |
| *** haomaiwang has joined #openstack-cinder | 07:01 | |
| *** jwcroppe has joined #openstack-cinder | 07:05 | |
| openstackgerrit | Vipin Balachandran proposed openstack/cinder: VMware: Unit test refactoring (image to vol - 2/2) https://review.openstack.org/249175 | 07:12 |
| *** lpetrut has quit IRC | 07:14 | |
| *** lpetrut has joined #openstack-cinder | 07:15 | |
| *** lpetrut has quit IRC | 07:24 | |
| *** sgotliv has quit IRC | 07:27 | |
| openstackgerrit | wanghao proposed openstack/python-cinderclient: support for snapshot management https://review.openstack.org/249090 | 07:28 |
| *** mudassirlatif has joined #openstack-cinder | 07:30 | |
| *** dims has joined #openstack-cinder | 07:35 | |
| *** anshul has joined #openstack-cinder | 07:37 | |
| *** salv-orlando has joined #openstack-cinder | 07:38 | |
| *** asselin has quit IRC | 07:39 | |
| *** dims_ has joined #openstack-cinder | 07:39 | |
| *** asselin has joined #openstack-cinder | 07:40 | |
| *** alexschm has joined #openstack-cinder | 07:40 | |
| *** dims has quit IRC | 07:40 | |
| *** salv-orlando has quit IRC | 07:41 | |
| *** rcernin_ has joined #openstack-cinder | 07:47 | |
| openstackgerrit | wanghao proposed openstack/cinder-specs: Get the replicated volume ref for DR https://review.openstack.org/182150 | 07:49 |
| *** zhangjn has joined #openstack-cinder | 07:59 | |
| *** zhangjn has quit IRC | 08:00 | |
| *** haomaiwang has quit IRC | 08:01 | |
| *** haomaiwang has joined #openstack-cinder | 08:01 | |
| *** zhangjn has joined #openstack-cinder | 08:01 | |
| *** RA_ has quit IRC | 08:06 | |
| *** e0ne has joined #openstack-cinder | 08:07 | |
| *** haomaiwang has quit IRC | 08:09 | |
| *** e0ne has quit IRC | 08:12 | |
| *** haomaiwang has joined #openstack-cinder | 08:14 | |
| *** markus_z has joined #openstack-cinder | 08:15 | |
| *** zhangjn has quit IRC | 08:16 | |
| *** e0ne has joined #openstack-cinder | 08:16 | |
| *** gouthamr has joined #openstack-cinder | 08:19 | |
| *** gouthamr_ has joined #openstack-cinder | 08:21 | |
| *** IlyaG has quit IRC | 08:21 | |
| *** IlyaG has joined #openstack-cinder | 08:22 | |
| *** gouthamr has quit IRC | 08:25 | |
| *** IlyaG has quit IRC | 08:26 | |
| *** vgridnev has quit IRC | 08:26 | |
| *** e0ne has quit IRC | 08:28 | |
| *** zhangjn has joined #openstack-cinder | 08:29 | |
| *** e0ne has joined #openstack-cinder | 08:33 | |
| *** mudassirlatif has quit IRC | 08:38 | |
| *** subscope has joined #openstack-cinder | 08:44 | |
| *** yhayashi has quit IRC | 08:44 | |
| *** e0ne has quit IRC | 08:44 | |
| *** anshul has quit IRC | 08:47 | |
| *** jordanP has joined #openstack-cinder | 08:51 | |
| *** anshul has joined #openstack-cinder | 08:51 | |
| *** haomaiwang has quit IRC | 08:53 | |
| *** liverpooler has joined #openstack-cinder | 08:54 | |
| *** haomaiwang has joined #openstack-cinder | 08:55 | |
| *** gouthamr_ has quit IRC | 09:00 | |
| *** haomaiwang has quit IRC | 09:01 | |
| *** haomaiwang has joined #openstack-cinder | 09:01 | |
| *** gouthamr has joined #openstack-cinder | 09:03 | |
| *** ndipanov has quit IRC | 09:03 | |
| *** mkoderer_ is now known as mkoderer | 09:11 | |
| openstackgerrit | Zhang Ni proposed openstack/cinder-specs: Add local tgt to lvm https://review.openstack.org/247880 | 09:11 |
| *** genunix has quit IRC | 09:16 | |
| *** bluex has joined #openstack-cinder | 09:19 | |
| *** lpetrut has joined #openstack-cinder | 09:22 | |
| *** skraynev has quit IRC | 09:23 | |
| *** isaacb has joined #openstack-cinder | 09:27 | |
| *** salv-orlando has joined #openstack-cinder | 09:27 | |
| *** gouthamr has quit IRC | 09:28 | |
| *** salv-orlando has quit IRC | 09:28 | |
| *** salv-orlando has joined #openstack-cinder | 09:29 | |
| *** vgridnev has joined #openstack-cinder | 09:31 | |
| *** dims_ has quit IRC | 09:34 | |
| *** jistr has joined #openstack-cinder | 09:35 | |
| *** terryyao has quit IRC | 09:37 | |
| swamireddy | HI | 09:41 |
| swamireddy | Using pacemaker, cinder-volume can be configured as active/passive using a shared storage backend (ie ceph cluster). Is this configuration works as expected? | 09:42 |
| *** ndipanov has joined #openstack-cinder | 09:42 | |
| swamireddy | or use only active-active mode with know issues/problem?? | 09:42 |
| *** kbyrne has joined #openstack-cinder | 09:42 | |
| *** dansmith has quit IRC | 09:42 | |
| *** dansmith has joined #openstack-cinder | 09:45 | |
| *** dansmith is now known as Guest11153 | 09:45 | |
| *** subscope has quit IRC | 09:46 | |
| DuncanT | swamireddy: It works well using active/passive | 09:47 |
| *** alonma has joined #openstack-cinder | 09:49 | |
| *** salv-orlando has quit IRC | 09:50 | |
| *** kbyrne has quit IRC | 09:50 | |
| *** kbyrne has joined #openstack-cinder | 09:50 | |
| *** alonma is now known as alonmarx | 09:51 | |
| *** IlyaG has joined #openstack-cinder | 09:52 | |
| swamireddy | DuncanT: So current cinder support A/P mode? | 09:53 |
| DuncanT | swamireddy: No cinder support is needed, it will just work | 09:53 |
| *** bapalm has quit IRC | 09:53 | |
| swamireddy | DuncanT: OK, do we have document ofr some steps for the active/passive mode setup (currently Iam using the A-A setup and need to go with A-P)? | 09:55 |
| *** bapalm has joined #openstack-cinder | 09:55 | |
| *** martyturner has joined #openstack-cinder | 09:55 | |
| DuncanT | swamireddy: Any attempts I've made to document pacemaker setup has led to me spending far too long supporting it. You just need to use pacemaker to ensure there is one and only one instance of cinder-volume running | 09:56 |
| *** IlyaG has quit IRC | 09:57 | |
| *** salv-orlando has joined #openstack-cinder | 09:57 | |
| *** gouthamr has joined #openstack-cinder | 09:58 | |
| *** martyturner has quit IRC | 10:00 | |
| *** e0ne has joined #openstack-cinder | 10:00 | |
| *** haomaiwang has quit IRC | 10:01 | |
| *** haomaiwang has joined #openstack-cinder | 10:01 | |
| *** zhenguo has quit IRC | 10:01 | |
| *** gouthamr has quit IRC | 10:03 | |
| *** gouthamr has joined #openstack-cinder | 10:03 | |
| *** rcernin_ has quit IRC | 10:06 | |
| *** RA_ has joined #openstack-cinder | 10:07 | |
| *** sgotliv has joined #openstack-cinder | 10:09 | |
| *** zhangjn has quit IRC | 10:09 | |
| swamireddy | DuncanT: sure, let me try it. Thanks | 10:13 |
| *** EinstCrazy has quit IRC | 10:13 | |
| DuncanT | swamireddy: If you do end up with a clean, documented process, I'm sure the setup guide would welcome the contribution | 10:14 |
| swamireddy | DuncanT: Sure. Let me first setup and make it work...then surely will add a doc on the same. | 10:16 |
| openstackgerrit | Vincent Hou proposed openstack/cinder: Storwize: Implement v2 replication (global) https://review.openstack.org/249042 | 10:16 |
| openstackgerrit | Vincent Hou proposed openstack/cinder: Storwize: Implement v2 replication (split IO) https://review.openstack.org/237387 | 10:16 |
| *** subscope has joined #openstack-cinder | 10:16 | |
| *** gouthamr has quit IRC | 10:17 | |
| *** alonmarx has quit IRC | 10:17 | |
| *** alonmarx has joined #openstack-cinder | 10:18 | |
| *** alonmarx has quit IRC | 10:22 | |
| *** anshul has quit IRC | 10:22 | |
| *** alonmarx has joined #openstack-cinder | 10:24 | |
| openstackgerrit | Vincent Hou proposed openstack/cinder: Storwize: Implement v2 replication (Mirror) https://review.openstack.org/249042 | 10:24 |
| *** alonmarx has quit IRC | 10:29 | |
| *** houming has quit IRC | 10:29 | |
| *** anshul has joined #openstack-cinder | 10:30 | |
| *** alonmarx has joined #openstack-cinder | 10:30 | |
| *** RA_ has quit IRC | 10:34 | |
| *** alonmarx has quit IRC | 10:34 | |
| *** alonmarx has joined #openstack-cinder | 10:36 | |
| *** haomaiwang has quit IRC | 10:39 | |
| *** alonmarx has quit IRC | 10:41 | |
| *** alonmarx has joined #openstack-cinder | 10:43 | |
| *** haomaiwa_ has joined #openstack-cinder | 10:44 | |
| *** subscope has quit IRC | 10:47 | |
| *** alonmarx has quit IRC | 10:47 | |
| *** RA_ has joined #openstack-cinder | 10:47 | |
| openstackgerrit | Szymon Wróblewski proposed openstack/cinder: Update migrate_volume API to use versionedobjects https://review.openstack.org/228916 | 10:48 |
| *** alonmarx has joined #openstack-cinder | 10:49 | |
| *** haomaiwa_ has quit IRC | 11:00 | |
| *** RA_ has quit IRC | 11:00 | |
| *** jwcroppe has quit IRC | 11:07 | |
| *** smoriya_ has quit IRC | 11:07 | |
| *** jwcroppe has joined #openstack-cinder | 11:07 | |
| *** alonmarx has quit IRC | 11:08 | |
| *** alonmarx has joined #openstack-cinder | 11:08 | |
| *** jwcroppe has quit IRC | 11:12 | |
| *** Guest11153 has quit IRC | 11:12 | |
| *** dansmith has joined #openstack-cinder | 11:15 | |
| *** dansmith is now known as Guest10504 | 11:15 | |
| *** RA_ has joined #openstack-cinder | 11:15 | |
| *** Guest10504 has quit IRC | 11:29 | |
| *** dansmith_ has joined #openstack-cinder | 11:31 | |
| *** jamielennox is now known as jamielennox|away | 11:42 | |
| openstackgerrit | Szymon Wróblewski proposed openstack/cinder: Update replication to use versionedobjects https://review.openstack.org/238906 | 11:47 |
| openstackgerrit | Szymon Wróblewski proposed openstack/cinder: Update initialize_connection to use versionedobjects https://review.openstack.org/238894 | 11:47 |
| *** marcusvrn_ has joined #openstack-cinder | 11:47 | |
| *** EinstCrazy has joined #openstack-cinder | 11:47 | |
| *** zhangjn has joined #openstack-cinder | 11:48 | |
| *** zhangjn has quit IRC | 11:49 | |
| *** haomaiwang has joined #openstack-cinder | 11:49 | |
| *** zhangjn has joined #openstack-cinder | 11:49 | |
| *** EinstCrazy has quit IRC | 11:49 | |
| *** EinstCrazy has joined #openstack-cinder | 11:50 | |
| *** dansmith_ has quit IRC | 11:52 | |
| *** dansmith_ has joined #openstack-cinder | 11:52 | |
| openstackgerrit | Szymon Wróblewski proposed openstack/cinder: Tooz locks https://review.openstack.org/183537 | 11:54 |
| *** haomaiwang has quit IRC | 12:01 | |
| *** jamielennox|away is now known as jamielennox | 12:01 | |
| *** bluex has quit IRC | 12:01 | |
| *** haomaiwang has joined #openstack-cinder | 12:01 | |
| *** IlyaG has joined #openstack-cinder | 12:02 | |
| *** bluex has joined #openstack-cinder | 12:02 | |
| *** subscope has joined #openstack-cinder | 12:02 | |
| *** dansmith_ is now known as dansmith | 12:04 | |
| *** dansmith is now known as Guest87843 | 12:05 | |
| *** jaypipes has joined #openstack-cinder | 12:05 | |
| *** IlyaG has quit IRC | 12:07 | |
| *** jwcroppe has joined #openstack-cinder | 12:09 | |
| *** houming has joined #openstack-cinder | 12:10 | |
| *** alonmarx has quit IRC | 12:19 | |
| *** alonmarx has joined #openstack-cinder | 12:19 | |
| openstackgerrit | Vipin Balachandran proposed openstack/cinder: VMware: Validate extra spec opt vmware:clone_type https://review.openstack.org/249735 | 12:20 |
| *** raildo-afk is now known as raildo | 12:23 | |
| *** alonmarx has quit IRC | 12:24 | |
| *** alonmarx has joined #openstack-cinder | 12:25 | |
| *** alonmarx has quit IRC | 12:31 | |
| *** jwcroppe has quit IRC | 12:31 | |
| *** alonmarx has joined #openstack-cinder | 12:32 | |
| *** alonmarx has quit IRC | 12:36 | |
| *** RA_ has quit IRC | 12:36 | |
| openstackgerrit | Shay Halsband proposed openstack/cinder: XtremIO: fix iscsi chap discovery bug https://review.openstack.org/249051 | 12:37 |
| *** alonmarx has joined #openstack-cinder | 12:38 | |
| *** takedakn has joined #openstack-cinder | 12:39 | |
| *** dims has joined #openstack-cinder | 12:42 | |
| *** alonmarx has quit IRC | 12:42 | |
| *** ociuhandu has quit IRC | 12:43 | |
| openstackgerrit | Michal Dulko proposed openstack/cinder: Add metadata aliases to Volume object https://review.openstack.org/247763 | 12:43 |
| openstackgerrit | Michal Dulko proposed openstack/cinder: Recognize extra fields in CinderObjectDictCompat https://review.openstack.org/247762 | 12:43 |
| openstackgerrit | Michal Dulko proposed openstack/cinder: DO NOT MERGE: Revert "VNX: Fix metadata get overriden issue" https://review.openstack.org/249742 | 12:44 |
| *** alonmarx has joined #openstack-cinder | 12:44 | |
| *** jerrygb has joined #openstack-cinder | 12:48 | |
| *** alonmarx has quit IRC | 12:48 | |
| *** e0ne_ has joined #openstack-cinder | 12:48 | |
| *** terryyao has joined #openstack-cinder | 12:49 | |
| *** boris-42 has joined #openstack-cinder | 12:49 | |
| *** alonmarx has joined #openstack-cinder | 12:50 | |
| *** jaypipes_ has joined #openstack-cinder | 12:51 | |
| *** e0ne has quit IRC | 12:51 | |
| *** jaypipes has quit IRC | 12:51 | |
| *** akerr has joined #openstack-cinder | 12:52 | |
| *** jaypipes_ is now known as jaypipes | 12:53 | |
| openstackgerrit | Merged openstack/cinder: Port key manager to Python 3 https://review.openstack.org/249397 | 12:53 |
| *** anshul has quit IRC | 12:54 | |
| *** akerr_ has joined #openstack-cinder | 12:54 | |
| *** alonmarx has quit IRC | 12:55 | |
| *** gcb has quit IRC | 12:56 | |
| dulek | geguileo: Hi. Is there a reason to implement get_by_id in CinderObject instead of CinderPersistentObject mixin? | 12:56 |
| *** alonmarx has joined #openstack-cinder | 12:56 | |
| dulek | geguileo: I'm working on RequestSpec object and just noticed that I'm getting a method which won't actually work. | 12:57 |
| *** akerr has quit IRC | 12:57 | |
| geguileo | dulek: By mix-in you mean the one that provides dictionary key access? | 12:58 |
| geguileo | dulek: Oh, sorry, you said Persistent | 12:58 |
| dulek | geguileo: Yup. :) | 12:58 |
| *** takedakn has quit IRC | 12:59 | |
| geguileo | dulek: I thought that VOs where mostly for DB objects and since we always have an ID on those... | 12:59 |
| geguileo | dulek: I didn't think of putting it in the persistent object to be frank | 13:00 |
| dulek | geguileo: Mostly DB objects is totally true. RequestSpec and FilterProperties are only two exceptions in Cinder I think. | 13:00 |
| geguileo | dulek: And why is FilterProperties a VO? | 13:00 |
| *** haomaiwang has quit IRC | 13:01 | |
| dulek | geguileo: It isn't now, I'm working on it. So thing is - these are dicts passed over RPC> | 13:01 |
| *** alonmarx has quit IRC | 13:01 | |
| geguileo | dulek: But do we need it to be VO? | 13:01 |
| geguileo | Because I don't think everything needs to be a VO | 13:01 |
| *** takedakn has joined #openstack-cinder | 13:02 | |
| *** zhipeng has joined #openstack-cinder | 13:02 | |
| dulek | If we want be able to change their structure (remove, rename things) and keep backward compatibility - we need to downgrade their version before sending over RPC to older services. | 13:02 |
| *** alonmarx has joined #openstack-cinder | 13:02 | |
| dulek | Live upgrades mean that we can have M's c-api and L's c-sch or c-vol. | 13:03 |
| *** chlong has joined #openstack-cinder | 13:03 | |
| dulek | Older versions may not understand newer versions - that's why we need OVO as translation layer. | 13:04 |
| *** asselin has quit IRC | 13:04 | |
| geguileo | But that would be solved with microversioning, right? | 13:04 |
| *** asselin has joined #openstack-cinder | 13:04 | |
| dulek | (or facade, like dansmith likes to say :>) | 13:04 |
| *** jaypipes has quit IRC | 13:04 | |
| dulek | With some complexity added it probably could, but what about c-sch -> c-vol communication? | 13:04 |
| dulek | During upgrade we can have M's c-sch and L's c-vol. | 13:05 |
| *** bluex has quit IRC | 13:05 | |
| dulek | And besides even in the API - I think microversions compatibility layer is in cinder.api, and we may have logic in cinder.volume.api also. | 13:06 |
| *** lixiaoy1 has quit IRC | 13:07 | |
| geguileo | cinder.volume.api should only be called from cinder.api, so that shouldn't be a problem, right? | 13:07 |
| *** alonmarx has quit IRC | 13:07 | |
| geguileo | And I don't like this idea of us moving everything to VOs so we can just remove, rename things | 13:07 |
| geguileo | Because then it's going to be crazy to follow what we have | 13:07 |
| geguileo | But I do see the benefit XD | 13:08 |
| dulek | Like I've said "with some complexity added it probably could". ;) We would need to pass microversions to cinder.volume.api methods. It's doable… | 13:08 |
| dulek | But not too pretty. | 13:09 |
| *** alonmarx has joined #openstack-cinder | 13:09 | |
| dulek | Yeah, noone ever said that rolling upgrades are easy. | 13:09 |
| *** david-lyle has quit IRC | 13:09 | |
| dulek | Even this release I've noticed some patches proposing renaming members of request_spec dict passed around, so I think there is a need for that. | 13:10 |
| *** alonmarx has quit IRC | 13:10 | |
| *** alonmarx has joined #openstack-cinder | 13:10 | |
| geguileo | So you want to move get_by_id from the base to persistent | 13:10 |
| *** pots has joined #openstack-cinder | 13:11 | |
| geguileo | Because if you call it in your new VO it will raise a NotImplemented exception? | 13:11 |
| dulek | Well, now to be honest I don't care much… I wouldn't simply call get_by_id on RequestSpec - why would I? | 13:12 |
| *** salv-orlando has quit IRC | 13:12 | |
| dulek | I've just noticed an inconsistency, so raised a question. | 13:12 |
| geguileo | Yeah, that's what I was thinking... | 13:12 |
| *** haomaiwang has joined #openstack-cinder | 13:12 | |
| geguileo | But I do see your point in moving it | 13:12 |
| *** houming has quit IRC | 13:12 | |
| dulek | It would probably fit better in Persistent mixin, so maybe in the future. :) | 13:13 |
| *** jerrygb has quit IRC | 13:13 | |
| kaisers | Hi Cinder! I'm looking for a way to react on updated volume metadata in a backend driver. My first idea is to use the _ensure_shares_mounted method. Does this make sense or does anybody have better ideas? | 13:14 |
| geguileo | dulek: I've added it to my list of "things to do in-between frustrating patches" ;-) | 13:15 |
| dulek | geguileo: I need my own one! ;) | 13:15 |
| dulek | kaisers: Why would you want to react? | 13:15 |
| *** lixiaoy1 has joined #openstack-cinder | 13:16 | |
| kaisers | dulek: In order to allow an admin to change volume specific settings in the backend at runtime | 13:16 |
| dulek | kaisers: Isn't it more like QoS specs and retype operation? | 13:16 |
| kaisers | dulek: e.g. if for whatever reason the backend replication factor of a specific volume should be increased or the like | 13:16 |
| kaisers | dulek: hmm, not sure. But i'll look into that right away, thanks for the hint | 13:17 |
| *** skraynev has joined #openstack-cinder | 13:17 | |
| openstackgerrit | zhangsong proposed openstack/os-brick: Improve the remove_multipath_device() method for linuxscsi https://review.openstack.org/249308 | 13:18 |
| dulek | kaisers: Hey, I'm not sure either. ;) Most drivers maintainers that are more experienced on that will be there around 15:00-16:00 UTC. | 13:18 |
| dulek | geguileo: So if you're here… Have you thought on autofencing case? | 13:19 |
| kaisers | dulek: Seems like an idea that should be checked out anyways an then i can coninue asking more profoundly later on... ;-) | 13:20 |
| geguileo | dulek: I have mostly cried about it... | 13:20 |
| dulek | geguileo: :D | 13:20 |
| geguileo | dulek: I have to give it another go, but the only option I see is to shoot ourselves in the head :-( | 13:21 |
| dulek | geguileo: Do we have a defined case on which we can experience data loss without it? | 13:21 |
| geguileo | dulek: Because of the problem with stoping the threads | 13:21 |
| dulek | geguileo: I mean - in most cases the resource we're working on is protected by it's state… | 13:21 |
| dulek | The only operation when it isn't would be clone, create from snapshot, et.c | 13:22 |
| geguileo | dulek: A problem would be with attaching/detaching | 13:23 |
| dulek | Okay, so a result of a clone operation that was interrupted with source deletion would be undefined… | 13:23 |
| geguileo | dulek: When we lose the DLM | 13:23 |
| dulek | Ah, multiple attaches/detaches? | 13:23 |
| *** salv-orlando has joined #openstack-cinder | 13:24 | |
| geguileo | dulek: Yep | 13:24 |
| *** salv-orlando has quit IRC | 13:24 | |
| *** salv-orlando has joined #openstack-cinder | 13:25 | |
| geguileo | dulek: And we are also only allowing to create 1 volume from the same source | 13:25 |
| *** zhipeng has quit IRC | 13:25 | |
| geguileo | dulek: But if the connection to the DLM was lost we would be allowing 2 (Although this doesn't look like a problem) | 13:25 |
| dulek | geguileo: Riiiight… | 13:25 |
| dulek | So that would need to be checked. | 13:25 |
| geguileo | dulek: The most likely problem with locks from creation would be the one you described | 13:25 |
| *** amrith is now known as _amrith_ | 13:26 | |
| dulek | And on multiple attach/detach - cannot we protect them by state even in internal calls? | 13:26 |
| geguileo | dulek: Trying to delete a resource that is being used by a stray node | 13:26 |
| dulek | Ah, right now internal calls are queuing, we won't have such possibility with state-based-locking. | 13:26 |
| geguileo | dulek: That's why we are not removing the locks from the manager | 13:27 |
| geguileo | Implicit contract | 13:27 |
| ankit_ag | eharney: hi, you around? | 13:28 |
| dulek | I mean - for attach/detach calls from inside Cinder we can change the contract. | 13:28 |
| *** jordanP has quit IRC | 13:30 | |
| dulek | So basically I'm trying to determine what will be happening if we won't have autofencing. Maybe we can get over it somehow. | 13:30 |
| *** ociuhandu has joined #openstack-cinder | 13:30 | |
| dulek | There are 2 dangerous situations - clone when source is getting deleted and multiple attach/detach | 13:31 |
| dulek | Clone one seems like not a very big deal IMO. | 13:32 |
| geguileo | dulek: Depends on how the back-end behaves | 13:32 |
| dulek | ? | 13:32 |
| *** takedakn has quit IRC | 13:33 | |
| *** ccesario has quit IRC | 13:33 | |
| *** bluex has joined #openstack-cinder | 13:34 | |
| *** jerrygb has joined #openstack-cinder | 13:34 | |
| geguileo | When you try to delete a resource that is being used | 13:35 |
| geguileo | Though I guess it should be ok | 13:35 |
| dulek | The resulting volume will be in undefined state. We can even check at the end of creation (and before releasing the lock) if source exists in the DB. If not - get the volume in ERROR state. | 13:36 |
| dulek | And if Galera will also get partitioned… We won't be able to update the volume - Galera will be read-only. | 13:37 |
| geguileo | dulek: Most likely if another node deletes the resource while creating a volume | 13:37 |
| geguileo | It will either fail or if it succeeds then the one creating will fail and set the volume to error | 13:38 |
| *** ankit_ag has quit IRC | 13:38 | |
| *** jerrygb has quit IRC | 13:39 | |
| dulek | I agree. So the point seems to be - easier data corruption case seems less problematic. | 13:39 |
| geguileo | The DB side is another matter... | 13:39 |
| geguileo | dulek: The problem is that we would actually need to test this before deciding we don't really need it | 13:39 |
| dulek | Oh definitely! | 13:40 |
| geguileo | dulek: And saying now we don't need and once in production see that it is needed would be bad | 13:40 |
| dulek | And we don't have a second data corruption case solved. | 13:40 |
| *** dustins has joined #openstack-cinder | 13:40 | |
| openstackgerrit | Merged openstack/cinder: Port xio driver to Python 3 https://review.openstack.org/242139 | 13:41 |
| eikke | are Cinder patches supposed to carry (unrelated) changes to cinder/opts.py? | 13:41 |
| openstackgerrit | Merged openstack/cinder: Remove kombu as a dependency for Cinder https://review.openstack.org/245188 | 13:41 |
| dulek | eikke: This file seems to be autogenerated by some test. I always remove it when committing. | 13:42 |
| eikke | but then a (non-voting) Jenkins job fails ;-) | 13:42 |
| *** bswartz has quit IRC | 13:42 | |
| eikke | so it's a bit unclear to me what the intent is | 13:43 |
| *** skraynev has quit IRC | 13:44 | |
| dulek | eikke: I'm not sure on the status of config generator (I believe it's that)… You would need to ask diablo_rojo once she'll be online. | 13:44 |
| dulek | geguileo: I'll give attach/detach more thought and see. Maybe we can find a way to be fairly safe for Mitaka. Or even mark A/A as experimental. | 13:45 |
| dulek | geguileo: I mean - I don't want to get stuck with the work. | 13:45 |
| geguileo | dulek: Me neither!!! | 13:45 |
| dulek | geguileo: And speaking of that… ;) I've seen you've mentioned that you've got 30% of API races patches ready. Need help? | 13:46 |
| geguileo | dulek: I'm changing all the patches that I already had to the generic error | 13:46 |
| dulek | geguileo: I'm a little swamped with upgrades stuff right now, but bluex resources seems to be more available. | 13:46 |
| * dulek is going to show bluex the IRC log. ;) | 13:47 | |
| geguileo | dulek: Once I'm finished with that I'll see where I'm really at | 13:47 |
| geguileo | dulek: And see what APIs are remaining and get back to you | 13:48 |
| geguileo | dulek: Because we need to get this first step out of the way ASASP | 13:48 |
| *** jgregor has joined #openstack-cinder | 13:48 | |
| *** houming has joined #openstack-cinder | 13:48 | |
| dulek | geguileo: Okay, then please ping me and bluex as soon you define where you need help. | 13:49 |
| geguileo | dulek: I definitely will, thanks!! | 13:49 |
| openstackgerrit | Nicolas Trangez proposed openstack/cinder: Revert "Add Scality SRB driver" https://review.openstack.org/248810 | 13:52 |
| *** merooney has joined #openstack-cinder | 13:55 | |
| *** jaypipes has joined #openstack-cinder | 13:57 | |
| *** xyang1 has joined #openstack-cinder | 13:57 | |
| *** merooeny_ has joined #openstack-cinder | 13:59 | |
| *** zhenguo has joined #openstack-cinder | 13:59 | |
| *** jwcroppe has joined #openstack-cinder | 14:00 | |
| openstackgerrit | Vipin Balachandran proposed openstack/cinder: VMware: Replace mox with mock https://review.openstack.org/249797 | 14:00 |
| *** skraynev has joined #openstack-cinder | 14:01 | |
| *** merooney has quit IRC | 14:02 | |
| *** shausy has quit IRC | 14:03 | |
| *** [1]Thelo has joined #openstack-cinder | 14:04 | |
| *** openstackgerrit has quit IRC | 14:06 | |
| *** Thelo has quit IRC | 14:07 | |
| *** [1]Thelo is now known as Thelo | 14:07 | |
| *** openstackgerrit has joined #openstack-cinder | 14:07 | |
| *** ccesario has joined #openstack-cinder | 14:11 | |
| *** jerrygb has joined #openstack-cinder | 14:13 | |
| *** thangp has joined #openstack-cinder | 14:14 | |
| *** kaisers1 has quit IRC | 14:18 | |
| *** rcernin has quit IRC | 14:19 | |
| openstackgerrit | zhangsong proposed openstack/cinder: Sheepdog: Improve copy_volume_to_image method https://review.openstack.org/249804 | 14:19 |
| *** bluex has quit IRC | 14:21 | |
| *** bluex has joined #openstack-cinder | 14:22 | |
| *** bluex has quit IRC | 14:22 | |
| openstackgerrit | Shay Halsband proposed openstack/cinder: XtremIO: fix generic glance cache with XtremIO https://review.openstack.org/249814 | 14:27 |
| *** baumann has joined #openstack-cinder | 14:27 | |
| *** Guest87843 is now known as dansmith | 14:32 | |
| *** lprice has quit IRC | 14:33 | |
| *** lprice has joined #openstack-cinder | 14:34 | |
| *** bluex has joined #openstack-cinder | 14:38 | |
| openstackgerrit | xing-yang proposed openstack/cinder: Backup snapshots https://review.openstack.org/243406 | 14:38 |
| *** terryyao has quit IRC | 14:39 | |
| *** bluex has quit IRC | 14:42 | |
| *** lprice has quit IRC | 14:43 | |
| *** bluex has joined #openstack-cinder | 14:43 | |
| *** diablo_rojo has joined #openstack-cinder | 14:44 | |
| *** lprice has joined #openstack-cinder | 14:44 | |
| *** akshai has joined #openstack-cinder | 14:47 | |
| *** jordanP has joined #openstack-cinder | 14:47 | |
| openstackgerrit | Masaki Kanno proposed openstack/cinder: Don't create snapshots from error volumes https://review.openstack.org/249826 | 14:48 |
| *** lprice1 has joined #openstack-cinder | 14:49 | |
| *** lprice has quit IRC | 14:50 | |
| *** liverpooler has quit IRC | 14:51 | |
| *** bluex has quit IRC | 14:53 | |
| *** bluex has joined #openstack-cinder | 14:53 | |
| *** thangp_ has joined #openstack-cinder | 14:55 | |
| *** sgotliv has quit IRC | 14:56 | |
| dulek | thangp: Hi. I'm looking for a way to easily implement loading a request_spec into object from dict (for compatibility in managers when receiving request_spec from older clients). | 14:57 |
| dulek | thangp: Any idea how to implement that easily? Adding from_dict to all the nested objects seems error-prone. | 14:58 |
| *** sgundur has joined #openstack-cinder | 14:58 | |
| *** thangp has quit IRC | 14:58 | |
| thangp_ | dulek: do you mean before we serialize it in rpcapi.py? | 15:01 |
| dulek | thangp_: Not exactly. | 15:02 |
| dulek | thangp_: L's c-vol will send request_spec as dict when rescheduling. It may happen that M's c-sch will receive it. I need a way to convert it from dict to RequestSpec object. | 15:03 |
| thangp_ | dulek: i thought that the order of upgrade is c-api, c-sch, then c-vol? | 15:03 |
| openstackgerrit | Eric Harney proposed openstack/cinder: NFS snapshots https://review.openstack.org/147186 | 15:04 |
| dulek | thangp_: Yeah, but rescheduling is triggered by c-vol. | 15:04 |
| dulek | So older c-vol can call newer c-sch. | 15:04 |
| thangp_ | dulek: ok | 15:04 |
| thangp_ | dulek: you could check in the c-sch, if what you get is a dict and convert it to a vo | 15:05 |
| thangp_ | dulek: that's the only way i know of right now | 15:06 |
| thangp_ | dulek: if c-sch gets a vo, then just continue | 15:06 |
| thangp_ | dulek: no conversion necessary | 15:07 |
| dulek | thangp_: Right, that's what I'm going to do. The problem is how to do the dict->vo conversion? RequestSpec has object nested (and this objects has anothers…). | 15:07 |
| dulek | thangp_: I can write several _from_dict methods to all this objects, but this seems like a lot of writing. I wonder if there isn't a better way. | 15:08 |
| thangp_ | dulek: eh...that's messy. maybe you should just forward the request_spec as-in back the c-vol and do the conversion in c-vol | 15:08 |
| *** sgotliv has joined #openstack-cinder | 15:09 | |
| *** e0ne_ has quit IRC | 15:09 | |
| *** _amrith_ is now known as amrith | 15:10 | |
| dulek | thangp_: That's even more messy - I would need to base some logic in the scheduler on the fact if I'm processing a dict or ovo. | 15:10 |
| dulek | thangp_: I see that Nova does the conversion manually, but all the operations inside request_spec object. I would need to write something similar probably. | 15:11 |
| thangp_ | dulek: not really, if you continue to support the old kwargs, then it should send request_spec as it had before | 15:11 |
| *** e0ne has joined #openstack-cinder | 15:11 | |
| thangp_ | dulek: then on the manager side, you can detect if you got and obj or dict | 15:12 |
| thangp_ | dulek: right, have some method in the request_spec to do the conversion all at once | 15:12 |
| thangp_ | dulek: then send it back to the c-vol | 15:13 |
| dulek | thangp_: Yup, that would be better than adding a method to almost every object we have there. | 15:14 |
| dulek | thangp_: I'll try and see. Thanks! | 15:14 |
| thangp_ | np | 15:14 |
| lixiaoy1 | dulek: may I talk with you about your commetns in Scaling backup service spec? | 15:16 |
| dulek | lixiaoy1: Can you give me a minute to read your answers there? | 15:16 |
| lixiaoy1 | dulek: ok | 15:16 |
| *** aix has quit IRC | 15:16 | |
| *** ntpttr has joined #openstack-cinder | 15:17 | |
| dulek | lixiaoy1: Okay, I've answered your comments there. So what's up? :) | 15:22 |
| lixiaoy1 | dulek: and then for upgrading, I think no problem | 15:24 |
| openstackgerrit | Nicolas Trangez proposed openstack/cinder: Revert "Add Scality SRB driver" https://review.openstack.org/248810 | 15:24 |
| lixiaoy1 | dulek: "How do we upgrade from previous scheme?" I can't find out any special | 15:25 |
| lixiaoy1 | dulek: from your original comments in patch 4 | 15:25 |
| dulek | lixiaoy1: The main change is that now a c-bak service will clear host field in the DB. | 15:26 |
| lixiaoy1 | dulek: as nothing will change except adding a volume id for the restoring backups | 15:26 |
| dulek | lixiaoy1: That's a major logic change - I can start working on a backup only if host field is NULL. | 15:26 |
| dulek | lixiaoy1: (with "I" I mean c-bak service) | 15:27 |
| lixiaoy1 | dulek: I remember only when backup completes restore/backup, the host field is cleared. let me check | 15:27 |
| dulek | lixiaoy1: This is my concern - we need a way to clean up host fields when upgrading. I think my proposition is safe. | 15:27 |
| *** mtanino has joined #openstack-cinder | 15:28 | |
| dulek | lixiaoy1: That's true, so normally a backup will have host field set to NULL and this will change only when some instance of c-bak is doing restore/backup on that backup resource. | 15:29 |
| lixiaoy1 | dulek: I am wondering why we need to clear the host when operation completes. it is too late, and I am sleepy :) | 15:30 |
| *** vgridnev has quit IRC | 15:30 | |
| *** chlong has quit IRC | 15:31 | |
| lixiaoy1 | dulek: as when cleaning up backups, only the backing/restoring backups are cleaned | 15:31 |
| dulek | lixiaoy1: It's to know who needs to clean up. | 15:31 |
| *** dulek has left #openstack-cinder | 15:31 | |
| *** dulek has joined #openstack-cinder | 15:31 | |
| dulek | lixiaoy1: There's no way to tell if a backup is stalled and needs cleaning or some other service is working on it. | 15:32 |
| dulek | lixiaoy1: And we don't want a c-bak to clean up a backup while some other c-bak is actually restoring it. | 15:32 |
| dulek | lixiaoy1: Right now we have explicit matching - a backup for a volume sits on a node where the volume is located. So only one c-bak owns a backup. | 15:33 |
| *** davechen has joined #openstack-cinder | 15:34 | |
| dulek | lixiaoy1: But with this spec we decouple that relationship. That's why we need to mark ownership when doing some work. | 15:34 |
| lixiaoy1 | dulek: if backups are not in backing/restoring status, they don't need to clean. the host field is useless | 15:34 |
| *** bswartz has joined #openstack-cinder | 15:34 | |
| dulek | lixiaoy1: I mean - when doing some work on a backup. | 15:34 |
| lixiaoy1 | dulek: if backups are in backing/restoring status, the host is set as the operator host. | 15:34 |
| *** xiaohui has joined #openstack-cinder | 15:34 | |
| dulek | lixiaoy1: Host field tells only who needs to clean up. | 15:35 |
| dulek | lixiaoy1: Okay, imagine a situation: | 15:35 |
| dulek | lixiaoy1: We're starting to backup a volume. | 15:35 |
| dulek | lixiaoy1: c-bak A got a request. | 15:35 |
| dulek | lixiaoy1: And it starts working | 15:35 |
| *** chenying__ has joined #openstack-cinder | 15:36 | |
| dulek | lixiaoy1: Now c-bak B don't know if c-bak A died or is still creating a backup. | 15:36 |
| lixiaoy1 | dulek: before that the host is set as the c-bak host | 15:36 |
| dulek | lixiaoy1: Now if c-bak A dies - we know that it was working on that resource. | 15:36 |
| dulek | lixiaoy1: And on the restart of c-bak A it can identify backups he was working on and clean them up. | 15:37 |
| lixiaoy1 | dulek: yes | 15:37 |
| openstackgerrit | zhangsong proposed openstack/os-brick: Improve connect_volume method of ISCSIConnector https://review.openstack.org/247390 | 15:38 |
| lixiaoy1 | dulek: I mean we don't need to clear host field when an operation finishes. We can set it every time an operation starts | 15:38 |
| dulek | lixiaoy1: So that's why we need host field - to mark that temporary ownership. | 15:38 |
| lixiaoy1 | dulek: yeah. we need host | 15:38 |
| *** alonmarx has quit IRC | 15:39 | |
| dulek | lixiaoy1: Oh… | 15:39 |
| dulek | lixiaoy1: This makes sense, let me think for a moment… | 15:39 |
| lixiaoy1 | dulek: as a result, when upgrading we don't need to clear host | 15:39 |
| *** alonmarx has joined #openstack-cinder | 15:39 | |
| *** chenying_ has quit IRC | 15:39 | |
| lixiaoy1 | dulek: also let me have a think :) | 15:39 |
| *** jgregor has quit IRC | 15:42 | |
| xiaohui | geguileo:Hi, I replied a few comments on the https://review.openstack.org/#/c/206923/. How do you think about them? | 15:43 |
| lixiaoy1 | xiaohui: hi | 15:43 |
| *** jgregor has joined #openstack-cinder | 15:43 | |
| *** alonmarx has quit IRC | 15:43 | |
| xiaohui | lixiaoy1: you are still online? :-) | 15:44 |
| lixiaoy1 | xiaohui: yes, I have something to discuss | 15:44 |
| dulek | lixiaoy1: I only have doubts on reset_status operation. We're not protecting the resource here. | 15:44 |
| dulek | lixiaoy1: I mean protecting using backup's state. | 15:45 |
| lixiaoy1 | dulek: do you mean the api to reset status of bacup? | 15:45 |
| *** alonmarx has joined #openstack-cinder | 15:45 | |
| dulek | lixiaoy1: Yup. | 15:45 |
| dulek | lixiaoy1: I can imagine a situation - someone is using reset_status and meanwhile someone starts to restore a volume. We have a race condition here. | 15:45 |
| dulek | lixiaoy1: Oh, but this race condition is present even in current state. So nevermind. ;) | 15:46 |
| xiaohui | geguileo:Hi, are you available now? | 15:46 |
| lixiaoy1 | dulek: yes. | 15:46 |
| geguileo | xiaohui: On a meeting, should be available in 10-15 minutes | 15:47 |
| tbarron | dulek: lixiaoy1: yeah, we should fix that api race using compare and swap I think. | 15:47 |
| openstackgerrit | Nate Potter proposed openstack/cinder: Show qos_specs_id based on policy https://review.openstack.org/241659 | 15:47 |
| lixiaoy1 | dulek: Tom raised the request I think when reviewing A/A spec | 15:47 |
| *** jseiler has joined #openstack-cinder | 15:47 | |
| tbarron | dulek: lixiaoy1: I'm glad you two are moving ahead on this. I'd like to get back to it, but am tangled up with other duties for a while. | 15:47 |
| lixiaoy1 | dulek: It needs to do | 15:47 |
| lixiaoy1 | tbarron: hi Tom, glad to see you:) | 15:48 |
| tbarron | lixiaoy1: Glad to see you as well, it must be almost midnight for you. | 15:48 |
| dulek | tbarron: Right, compare and swap stuff would probably fit there, I wold need to dig more to decide. ;) | 15:48 |
| xiaohui | geguileo: that's ok. sorry. | 15:48 |
| lixiaoy1 | tbarron: yes, 11:48 pm :-( | 15:49 |
| tbarron | dulek: lixiaoy1: I agree we have that race already today. So the spec doesn't need to address it, or perhaps | 15:49 |
| tbarron | it should mention it as an example | 15:49 |
| dulek | lixiaoy1: I can imagine just one problem with not-cleaning host column. | 15:49 |
| tbarron | of races we have today that will be exacerbated if we are running with even more concurrency. | 15:49 |
| *** aix has joined #openstack-cinder | 15:49 | |
| *** alonmarx has quit IRC | 15:49 | |
| dulek | lixiaoy1: Let's assume that during upgrade we have some c-bak's in L version and some in M. | 15:49 |
| dulek | Then L's will expect that they are the only one working on resources with their host field. | 15:50 |
| dulek | But that's not true for M's - they will… | 15:50 |
| dulek | Awww, nevermind, that's also solved. | 15:51 |
| dulek | lixiaoy1: Case closed for me, you're right. :) | 15:51 |
| *** alonmarx has joined #openstack-cinder | 15:51 | |
| lixiaoy1 | dulek: thank you. :) | 15:52 |
| *** houming has quit IRC | 15:52 | |
| dulek | I would only insist on changing host field with compare-and-swap, but I guess that's addressed in the spec. :) | 15:52 |
| tbarron | lixiaoy1: dulek: we need to put that paragraph about compare and swap back in I think | 15:52 |
| tbarron | lixiaoy1: dulek: I dropped it by accident a couple patches back. | 15:53 |
| *** eduardo__ has joined #openstack-cinder | 15:53 | |
| openstackgerrit | Bertrand Lallau proposed openstack/cinder: Remove netaddr useless requirement https://review.openstack.org/249866 | 15:53 |
| ntpttr | dulek: would it be possible to get a link to the spec you've been talking about to look it over :) | 15:53 |
| dulek | ntpttr: https://review.openstack.org/#/c/240978 | 15:53 |
| ntpttr | dulek: thanks | 15:53 |
| kaisers | dulek: FYI: qos is tied to volume_type not volume, thus not exactly what i'm aiming at. But it was a good point mentioning it! :) | 15:54 |
| tbarron | ntpttr: it would be good to get your eyes on it too! | 15:54 |
| ntpttr | tbarron: giving it a read through right now :) | 15:55 |
| *** houming has joined #openstack-cinder | 15:55 | |
| lixiaoy1 | dulek: changing host field with compare-and-swap? I can image two places to update the host. One is to update the status of backup and host in a transaction, and then when finishing, also update the status of backup and host in a transation | 15:55 |
| lixiaoy1 | tbarron: please see above setence | 15:55 |
| dulek | lixiaoy1: I thought that we've decided not to clear host column after finishing? | 15:55 |
| *** alonmarx has quit IRC | 15:56 | |
| lixiaoy1 | dulek: yes | 15:56 |
| lixiaoy1 | dulek: sorry, ignore the finishing operation | 15:56 |
| dulek | ntpttr: That's a fun one - much simpler than c-vol A/A, but dealing with similar issues - all related to distributed systems. :) | 15:56 |
| tbarron | dulek: +1 | 15:56 |
| dulek | lixiaoy1: Yeah, so that's the place + changing status from ACTIVE to RESTORING, etc. | 15:57 |
| *** alonmarx has joined #openstack-cinder | 15:57 | |
| tbarron | lixiaoy1: dulek: I agree | 15:58 |
| lixiaoy1 | dulek: tbarron: so done for backup. and wait for the meeting then | 15:58 |
| *** boris-42 has quit IRC | 15:58 | |
| smcginnis | T-1 minute to meeting | 15:59 |
| lixiaoy1 | smcginnis: Sorry for first time to join in cinder meeting. But it is too late for Chinese timezone | 16:00 |
| smcginnis | lixiaoy1: Yeah, sorry the timing is so bad for you. | 16:00 |
| *** zhangjn has quit IRC | 16:00 | |
| *** mudassirlatif has joined #openstack-cinder | 16:00 | |
| *** alonmarx has quit IRC | 16:02 | |
| *** alonmarx has joined #openstack-cinder | 16:04 | |
| *** dustins has quit IRC | 16:05 | |
| *** dustins has joined #openstack-cinder | 16:07 | |
| *** alonmarx has quit IRC | 16:08 | |
| *** daneyon_ has quit IRC | 16:08 | |
| *** dims has quit IRC | 16:09 | |
| *** daneyon has joined #openstack-cinder | 16:09 | |
| *** dims has joined #openstack-cinder | 16:11 | |
| *** alonmarx has joined #openstack-cinder | 16:11 | |
| *** jdurgin1 has joined #openstack-cinder | 16:12 | |
| *** crose has joined #openstack-cinder | 16:13 | |
| *** EinstCrazy has quit IRC | 16:14 | |
| *** haomaiwang has quit IRC | 16:15 | |
| *** alonmarx has quit IRC | 16:15 | |
| *** haomaiwa_ has joined #openstack-cinder | 16:15 | |
| *** pots has quit IRC | 16:17 | |
| *** amrith is now known as _amrith_ | 16:17 | |
| *** IlyaG has joined #openstack-cinder | 16:18 | |
| *** alexschm has quit IRC | 16:22 | |
| *** IlyaG has quit IRC | 16:23 | |
| *** xiaohui has quit IRC | 16:23 | |
| *** xiaohui_ has joined #openstack-cinder | 16:23 | |
| *** zhenguo has quit IRC | 16:25 | |
| *** lixiaoy1 has left #openstack-cinder | 16:26 | |
| *** crose has quit IRC | 16:26 | |
| *** lixiaoy1_ has joined #openstack-cinder | 16:27 | |
| *** vgridnev has joined #openstack-cinder | 16:28 | |
| *** subscope has quit IRC | 16:29 | |
| openstackgerrit | Merged openstack/cinder: VMware: Validate extra spec opt vmware:clone_type https://review.openstack.org/249735 | 16:34 |
| xiaohui_ | geguileo:I'm not very sure about a few of the comments on https://review.openstack.org/#/c/206923/. Would you like to reply the comments or to discuss it here? | 16:34 |
| geguileo | xiaohui_: I was reviewing it and called to another meeting/call | 16:35 |
| xiaohui_ | geguileo:Ok. Sorry for that. | 16:36 |
| *** houming has quit IRC | 16:37 | |
| *** jseiler has quit IRC | 16:45 | |
| openstackgerrit | Merged openstack/cinder: Fix ScaleIO driver provisioning key https://review.openstack.org/249552 | 16:45 |
| kaisers | smcginnis: Quick Q: i've a minor vendor driver related bp/change already aimed at Mitaka. Should this go somewhere into the milestone lists or the etherpad for Mitaka or should i simply hold still & wait? | 16:46 |
| smcginnis | kaisers: The BP is filed? | 16:46 |
| kaisers | smcginnis: yep (https://blueprints.launchpad.net/cinder/+spec/allow-insecure-conf-in-quobyte) | 16:47 |
| smcginnis | kaisers: Looks like I've already approved it. | 16:47 |
| smcginnis | kaisers: Just go ahead with it then. | 16:47 |
| thingee | bummer doesn't look like I'll be able to attend the sprint in january | 16:47 |
| smcginnis | kaisers: It's OK if it's not targeted to a specific milestone. | 16:47 |
| smcginnis | kaisers: But sooner is better. | 16:47 |
| smcginnis | thingee: :( | 16:47 |
| smcginnis | thingee: Attending a different one? | 16:48 |
| thingee | smcginnis: the foundation has an offsite then | 16:48 |
| *** arnewiebalck__ has quit IRC | 16:48 | |
| smcginnis | thingee: Seems to be a popular week. | 16:48 |
| kaisers | smcginnis: it's implemented, it just does not turn up on the launchpad milestone lists or the etherpad and i'm unsure if it should be added somewhere there or not. | 16:49 |
| smcginnis | kaisers: Did it merge? Or it's being reviewed? | 16:49 |
| * smcginnis sees link | 16:49 | |
| kaisers | smcginnis: in review, couple of +1 | 16:49 |
| smcginnis | kaisers: That's fine. When it merges it will get picked up as part of that milestone. | 16:50 |
| kaisers | smcginnis: ok, thanks! | 16:50 |
| smcginnis | kaisers: np | 16:50 |
| smcginnis | kaisers: Thanks for checking and staying on top if it. | 16:50 |
| *** arnewiebalck_ has joined #openstack-cinder | 16:50 | |
| kaisers | smcginnis: np :) | 16:51 |
| *** IlyaG has joined #openstack-cinder | 16:52 | |
| *** jwcroppe has quit IRC | 16:56 | |
| *** jwcroppe has joined #openstack-cinder | 16:57 | |
| *** jordanP has quit IRC | 16:57 | |
| *** jordanP has joined #openstack-cinder | 16:57 | |
| *** asselin has quit IRC | 17:00 | |
| *** asselin has joined #openstack-cinder | 17:00 | |
| *** haomaiwa_ has quit IRC | 17:01 | |
| *** haomaiwang has joined #openstack-cinder | 17:01 | |
| *** jwcroppe has quit IRC | 17:01 | |
| *** jistr has quit IRC | 17:02 | |
| *** markvoelker has quit IRC | 17:03 | |
| *** mudassirlatif_ has joined #openstack-cinder | 17:03 | |
| *** mudassirlatif has quit IRC | 17:05 | |
| *** mudassirlatif_ is now known as mudassirlatif | 17:05 | |
| *** yuriy_n17 has quit IRC | 17:06 | |
| *** e0ne has quit IRC | 17:09 | |
| *** houming has joined #openstack-cinder | 17:10 | |
| *** cebruns has quit IRC | 17:11 | |
| *** davechen1 has joined #openstack-cinder | 17:12 | |
| *** alonmarx has joined #openstack-cinder | 17:12 | |
| *** cebruns has joined #openstack-cinder | 17:12 | |
| *** davechen has quit IRC | 17:12 | |
| *** EinstCrazy has joined #openstack-cinder | 17:14 | |
| eikke | smcginnis: thanks for +2 | 17:16 |
| *** alonmarx has quit IRC | 17:16 | |
| *** alonmarx has joined #openstack-cinder | 17:18 | |
| *** dzamboni has joined #openstack-cinder | 17:18 | |
| *** leeantho has joined #openstack-cinder | 17:19 | |
| *** mudassirlatif has quit IRC | 17:21 | |
| *** alonmarx has quit IRC | 17:23 | |
| openstackgerrit | XinXiaohui proposed openstack/cinder-specs: Moving capacity-headroom spec to mitaka cycle https://review.openstack.org/249932 | 17:24 |
| *** alonmarx has joined #openstack-cinder | 17:26 | |
| openstackgerrit | Kendall Nelson proposed openstack/cinder: Hacking Checks for assertTrue/IsNone() https://review.openstack.org/246554 | 17:27 |
| *** jerrygb has quit IRC | 17:27 | |
| openstackgerrit | Jacob Gregor proposed openstack/cinder: Storwize: Split up __init__ into separate files https://review.openstack.org/240329 | 17:28 |
| *** jerrygb has joined #openstack-cinder | 17:28 | |
| *** jerrygb has quit IRC | 17:28 | |
| *** jerrygb has joined #openstack-cinder | 17:28 | |
| *** EinstCrazy has quit IRC | 17:29 | |
| *** mudassirlatif has joined #openstack-cinder | 17:29 | |
| *** david-lyle has joined #openstack-cinder | 17:30 | |
| *** nkrinner has quit IRC | 17:30 | |
| *** alonmarx has quit IRC | 17:31 | |
| *** mudassirlatif has quit IRC | 17:37 | |
| *** raildo is now known as raildo-afk | 17:38 | |
| *** raildo-afk is now known as raildo | 17:39 | |
| *** jerrygb has quit IRC | 17:43 | |
| *** davechen1 has quit IRC | 17:44 | |
| *** xiaohui_ has quit IRC | 17:45 | |
| *** changbl has quit IRC | 17:46 | |
| *** lpetrut has quit IRC | 17:48 | |
| *** mudassirlatif has joined #openstack-cinder | 17:51 | |
| *** david-lyle has quit IRC | 18:00 | |
| *** haomaiwang has quit IRC | 18:01 | |
| *** haomaiwang has joined #openstack-cinder | 18:01 | |
| *** eduardo__ has quit IRC | 18:02 | |
| *** jordanP has quit IRC | 18:03 | |
| *** e0ne has joined #openstack-cinder | 18:07 | |
| *** _amrith_ is now known as amrith | 18:08 | |
| *** sghanekar__ has joined #openstack-cinder | 18:09 | |
| *** isaacb has quit IRC | 18:09 | |
| *** edtubill has joined #openstack-cinder | 18:10 | |
| openstackgerrit | xing-yang proposed openstack/cinder: ScaleIO extend volume round up capacity https://review.openstack.org/249556 | 18:15 |
| *** ociuhandu has quit IRC | 18:15 | |
| *** markus_z has quit IRC | 18:19 | |
| patrickeast | smcginnis: hey for the ThirdPartySystems wiki pages should we modify the template so the recheck syntax goes in the table? | 18:19 |
| *** diablo_rojo has quit IRC | 18:20 | |
| smcginnis | patrickeast: Hah! Just took a beating in infra because someone who I won't name tried that. | 18:20 |
| *** edtubill has quit IRC | 18:20 | |
| *** david-lyle has joined #openstack-cinder | 18:20 | |
| patrickeast | oh? they didn't like it? | 18:20 |
| smcginnis | patrickeast: I think it should go there, but for now easier putting it outside the template. | 18:20 |
| smcginnis | patrickeast: Yeah, not too happy about it. | 18:20 |
| smcginnis | patrickeast: They did invite someone to put it on the infra agenda to be considered, but they didn't like not having notice of it. | 18:20 |
| smcginnis | patrickeast: So I don't think it was that they were entirely opposed to the idea. | 18:21 |
| patrickeast | notice of it?... its a wiki... | 18:21 |
| patrickeast | oh well | 18:21 |
| smcginnis | patrickeast: Just that no one was consulted first. | 18:21 |
| smcginnis | patrickeast: That was kind of my thought, but I will say no more. :) | 18:21 |
| patrickeast | yea we should be careful with groundbreaking table changes in the future | 18:21 |
| smcginnis | LOL | 18:21 |
| patrickeast | i'll just update the footer and call it good | 18:21 |
| smcginnis | patrickeast: Thanks! | 18:21 |
| *** sghanekar__ has quit IRC | 18:23 | |
| rhedlind | I just had the same idea about the wiki and starting looking at the table template. :) | 18:29 |
| *** vgridnev has quit IRC | 18:29 | |
| rhedlind | but I will leave it alone. Not up for any beatings. | 18:30 |
| mtanino | dulek: Thank you for the confiamation of quota bugs :) | 18:34 |
| openstackgerrit | Merged openstack/python-cinderclient: Fix v2 qos-key command https://review.openstack.org/248944 | 18:38 |
| *** houming has quit IRC | 18:46 | |
| *** sgotliv has quit IRC | 18:49 | |
| ntpttr | eharney: ping | 18:54 |
| *** vgridnev has joined #openstack-cinder | 18:56 | |
| *** ociuhandu has joined #openstack-cinder | 18:57 | |
| openstackgerrit | Gorka Eguileor proposed openstack/cinder: Remove API races from delete methods https://review.openstack.org/205835 | 18:57 |
| openstackgerrit | Gorka Eguileor proposed openstack/cinder: Add atomic conditional updates to objects https://review.openstack.org/205834 | 18:57 |
| openstackgerrit | Gorka Eguileor proposed openstack/cinder: Remove API races on extend and volume_upload_image https://review.openstack.org/216378 | 18:57 |
| openstackgerrit | Gorka Eguileor proposed openstack/cinder: Remove API races from migrate and retype https://review.openstack.org/221442 | 18:57 |
| openstackgerrit | Gorka Eguileor proposed openstack/cinder: Remove API races on replication methods https://review.openstack.org/221441 | 18:57 |
| openstackgerrit | Gorka Eguileor proposed openstack/cinder: Improve metadata update operations https://review.openstack.org/216376 | 18:57 |
| openstackgerrit | Gorka Eguileor proposed openstack/cinder: Remove API races from attach and detach methods https://review.openstack.org/216377 | 18:57 |
| openstackgerrit | Gorka Eguileor proposed openstack/cinder: Add ordering possibilities to conditional update https://review.openstack.org/231936 | 18:57 |
| *** e0ne has quit IRC | 18:58 | |
| tbarron | smcginnis: patrickeast: just got back from lunch but I see that Sean was coveriing for me. Thanks Sean. | 18:59 |
| tbarron | smcginnis: patrickeast: There shouldn't be an edit me link available to everyone if we're not supposed to edit it. | 18:59 |
| patrickeast | tbarron: haha yea, exactly | 19:00 |
| tbarron | oh well, I've been trying to increase my visibility with infra! | 19:00 |
| patrickeast | :D great way to get your name out there | 19:00 |
| *** haomaiwang has quit IRC | 19:01 | |
| * tbarron goes looking for other stuff to edit | 19:01 | |
| *** haomaiwang has joined #openstack-cinder | 19:01 | |
| *** jwcroppe has joined #openstack-cinder | 19:09 | |
| *** aix has quit IRC | 19:11 | |
| *** alonmarx has joined #openstack-cinder | 19:11 | |
| *** barra204 is now known as shakamunyi | 19:12 | |
| *** davechen has joined #openstack-cinder | 19:12 | |
| *** alonmarx has quit IRC | 19:16 | |
| *** alonmarx has joined #openstack-cinder | 19:18 | |
| *** salv-orl_ has joined #openstack-cinder | 19:19 | |
| *** salv-orlando has quit IRC | 19:22 | |
| *** alonmarx has quit IRC | 19:22 | |
| *** alonmarx has joined #openstack-cinder | 19:26 | |
| *** alonmarx has quit IRC | 19:31 | |
| *** alonmarx has joined #openstack-cinder | 19:33 | |
| *** alonmarx has quit IRC | 19:37 | |
| *** eharney has quit IRC | 19:38 | |
| *** alonmarx has joined #openstack-cinder | 19:41 | |
| *** zohar has joined #openstack-cinder | 19:41 | |
| zohar | Hi all | 19:42 |
| zohar | this is Zohar from Nexenta team | 19:42 |
| zohar | I would like to ask for some help with writing mock unit tests and how to verify that params are being passed correctly | 19:42 |
| zohar | currently we have these unit tests: https://review.openstack.org/#/c/158745/32/cinder/tests/unit/test_nexenta_edge.py | 19:43 |
| zohar | im looking for how to extend these | 19:43 |
| smcginnis | zohar: Hey! | 19:44 |
| *** changbl has joined #openstack-cinder | 19:45 | |
| zohar | Hi! :) | 19:45 |
| smcginnis | zohar: Right now you are only covering the driver class. | 19:45 |
| smcginnis | zohar: If I remember right, you have other helper classes in your driver code as well. | 19:45 |
| zohar | yes | 19:45 |
| *** alonmarx has quit IRC | 19:45 | |
| smcginnis | zohar: A good thing would be to add unit tests for the methods in those classes too. | 19:45 |
| smcginnis | zohar: In general, a unit test should cover each method where there is enough internal logic to make sense. | 19:46 |
| zohar | ok, good point, i will add unit test for the rest api helper class | 19:46 |
| zohar | besides that though, what about the current driver class tests? | 19:46 |
| zohar | based on comments, they are lacking? | 19:46 |
| smcginnis | zohar: In other words, if the method is just a wrapper that has a single line calling another class, maybe it's not significant enough. | 19:46 |
| *** IlyaG has quit IRC | 19:46 | |
| smcginnis | zohar: But more than that, it makes sense to have a test. | 19:46 |
| zohar | yes that is what i was thinking... | 19:46 |
| smcginnis | zohar: Then look at what the valid and invalid input would be to each method. | 19:47 |
| zohar | so the driver class is really all it does, just a method wrapper for a single call | 19:47 |
| smcginnis | zohar: And validate the expected result from each. | 19:47 |
| zohar | ok | 19:47 |
| zohar | how do i do that? do you have any pointers? | 19:47 |
| smcginnis | zohar: Yeah, I think more of your logic was in the other classes. | 19:47 |
| zohar | most of the logic is actually on the backend :) | 19:47 |
| smcginnis | zohar: Not really other than going through and looking at each one and thinking about what could possibly be the input and what would be the expected output. | 19:48 |
| zohar | the REST api/rpc class just takes care of doing rest requests and figuring out if they are errors | 19:48 |
| smcginnis | zohar: That does make it more challenging to cover with unit tests when the logic is external to the code base. | 19:48 |
| zohar | yes definitely, i just mean code-wise, how do i use mock to check input? | 19:48 |
| smcginnis | zohar: But you can still maybe do things like validating the data sent to the external commands has the expected content. | 19:48 |
| zohar | so i created a driver using a mock.Mock() as the execute field | 19:49 |
| smcginnis | zohar: That error conditions (if any) contain the right error codes,e tc. | 19:49 |
| zohar | and i make a mock patcher of the rest api class | 19:49 |
| zohar | so for example, currently what i do in " | 19:49 |
| zohar | test_create_volume" is | 19:49 |
| zohar | self.assertIsNone(self.driver.create_volume(MOCK_VOL)) | 19:49 |
| zohar | how can i check that the mock driver got the right params? | 19:50 |
| smcginnis | zohar: If it's a mock you can do things like self.assert_called_once_with(xxx) and things like that. | 19:51 |
| *** diablo_rojo has joined #openstack-cinder | 19:51 | |
| zohar | nice! is there a way to print out what it was called for for debuging? | 19:51 |
| smcginnis | zohar: I guess just don't do things like asserting that what you've set for the mock to return gets returned. | 19:51 |
| zohar | what it was called with* | 19:51 |
| smcginnis | zohar: That just tests that mock works. ;) | 19:52 |
| zohar | yes haha | 19:52 |
| smcginnis | zohar: You can set breakpoints. | 19:52 |
| smcginnis | zohar: There's some info in our devref docs about debugging with unit tests. | 19:52 |
| smcginnis | zohar: Sorry, gotta run. | 19:52 |
| zohar | any chance i can get a link? | 19:52 |
| zohar | its ok, thank you for your help! | 19:52 |
| smcginnis | zohar: Maybe someone else here can give you a pointer. bbl | 19:53 |
| *** alonmarx has joined #openstack-cinder | 19:53 | |
| openstackgerrit | Nate Potter proposed openstack/cinder: Show qos_specs_id based on policy https://review.openstack.org/241659 | 19:54 |
| *** alonmarx has quit IRC | 19:57 | |
| *** haomaiwang has quit IRC | 20:01 | |
| *** haomaiwang has joined #openstack-cinder | 20:01 | |
| *** salv-orl_ has quit IRC | 20:07 | |
| *** changbl has quit IRC | 20:09 | |
| *** Lee1092 has quit IRC | 20:09 | |
| *** edtubill has joined #openstack-cinder | 20:12 | |
| *** akerr_ is now known as akerr_away | 20:22 | |
| *** changbl has joined #openstack-cinder | 20:22 | |
| *** dustins has quit IRC | 20:27 | |
| *** john5223 is now known as zz_john5223 | 20:27 | |
| *** akerr_away is now known as akerr_ | 20:35 | |
| *** angela-s has joined #openstack-cinder | 20:35 | |
| openstackgerrit | Zohar Mamedov proposed openstack/cinder: Nexenta Edge iSCSI backend driver https://review.openstack.org/158745 | 20:37 |
| raildo | hey guys, I'm trying fix this bug: https://bugs.launchpad.net/cinder/+bug/1505801 | 20:41 |
| openstack | Launchpad bug 1505801 in Cinder "Create volume ignores allocated quota values of the parent project" [Undecided,In progress] - Assigned to Raildo Mascena de Sousa Filho (raildo) | 20:41 |
| raildo | and I thinkk taht this should be backported for liberty | 20:41 |
| raildo | is there something different (like send for the liberty-stable) to do? | 20:41 |
| raildo | ou should I wait to this patch get merged and later I have to do a cherry-pick for liberty? | 20:42 |
| *** dzamboni has quit IRC | 20:43 | |
| *** salv-orlando has joined #openstack-cinder | 20:44 | |
| *** willsama has joined #openstack-cinder | 20:45 | |
| *** willsama has quit IRC | 20:45 | |
| smcginnis | raildo: Best would be to wait for it to merge, then cherry-pick for liberty. | 20:46 |
| smcginnis | raildo: Thanks for working on that! | 20:46 |
| smcginnis | zohar: Saw your patch. Any luck tracking down my vague references? :) | 20:47 |
| smcginnis | zohar: I actually can't find it now, though I remember looking at it not long ago. :/ | 20:51 |
| zohar | smcginnis: your tip about "assert_called_with" was exactly what i needed | 20:52 |
| smcginnis | zohar: Awesome! | 20:52 |
| *** vgridnev has quit IRC | 20:52 | |
| zohar | thank you very much, lets hope all reviewers will be happy with it this time :) | 20:52 |
| raildo | smcginnis: no problem, thanks for the advice, I'll do that | 20:53 |
| *** alonmarx has joined #openstack-cinder | 20:53 | |
| smcginnis | zohar: I'll take a look soon. I appreciate the extra work on unit tests. They can be a pain, but definitely help. | 20:53 |
| smcginnis | raildo: +1 | 20:53 |
| *** grumpycatt has joined #openstack-cinder | 20:57 | |
| *** e0ne has joined #openstack-cinder | 20:58 | |
| *** alonmarx has quit IRC | 20:58 | |
| *** haomaiwang has quit IRC | 21:01 | |
| *** haomaiwang has joined #openstack-cinder | 21:01 | |
| *** thangp_ has quit IRC | 21:02 | |
| *** tpsilva has quit IRC | 21:02 | |
| *** baumann has left #openstack-cinder | 21:02 | |
| *** e0ne has quit IRC | 21:03 | |
| *** vgridnev has joined #openstack-cinder | 21:04 | |
| *** ntpttr has quit IRC | 21:06 | |
| *** mtanino has quit IRC | 21:06 | |
| *** alonmarx has joined #openstack-cinder | 21:08 | |
| *** raildo is now known as raildo-afk | 21:11 | |
| *** alonmarx has quit IRC | 21:13 | |
| *** zz_john5223 is now known as john5223 | 21:13 | |
| *** sgotliv has joined #openstack-cinder | 21:16 | |
| *** diablo_rojo has quit IRC | 21:17 | |
| *** jgregor has quit IRC | 21:20 | |
| *** diablo_rojo has joined #openstack-cinder | 21:20 | |
| *** dims_ has joined #openstack-cinder | 21:21 | |
| *** dims has quit IRC | 21:22 | |
| openstackgerrit | Mudassir Latif proposed openstack/cinder: Allow replicated volumes to be recoverable https://review.openstack.org/219900 | 21:24 |
| *** dims has joined #openstack-cinder | 21:26 | |
| *** dims_ has quit IRC | 21:28 | |
| *** ntpttr has joined #openstack-cinder | 21:29 | |
| *** openstackgerrit has quit IRC | 21:36 | |
| *** openstackgerrit has joined #openstack-cinder | 21:37 | |
| *** bswartz has quit IRC | 21:40 | |
| *** ndipanov has quit IRC | 21:43 | |
| *** ntpttr has left #openstack-cinder | 21:44 | |
| *** alonmarx has joined #openstack-cinder | 21:46 | |
| *** merooeny_ has quit IRC | 21:47 | |
| *** alonmarx has quit IRC | 21:50 | |
| *** takedakn has joined #openstack-cinder | 21:50 | |
| *** takedakn has quit IRC | 21:51 | |
| *** vgridnev has quit IRC | 21:51 | |
| *** edtubill has quit IRC | 21:52 | |
| *** cbader has quit IRC | 21:52 | |
| zohar | hi guys | 21:53 |
| zohar | i got this jenkins error: | 21:53 |
| zohar | http://logs.openstack.org/45/158745/33/check/gate-cinder-pep8/d01937c/console.html | 21:53 |
| *** dims_ has joined #openstack-cinder | 21:53 | |
| zohar | pep8 error, but looks like not a lint issue | 21:53 |
| zohar | cant really make sense what failed | 21:54 |
| smcginnis | zohar: Ah, you hit our protection for config options changing without updating the config file. | 21:55 |
| *** takedakn has joined #openstack-cinder | 21:55 | |
| smcginnis | zohar: You should be able to follow the instructions there to resolve. | 21:55 |
| smcginnis | zohar: http://logs.openstack.org/45/158745/33/check/gate-cinder-pep8/d01937c/console.html#_2015-11-25_20_49_41_377 | 21:55 |
| *** dims has quit IRC | 21:56 | |
| *** alonmarx has joined #openstack-cinder | 21:58 | |
| zohar | so just add opts.py file to the commit? | 21:59 |
| zohar | i ran it, it generated | 21:59 |
| zohar | on a devstack cloned from master a few days ago | 22:00 |
| zohar | that should be fine right? | 22:00 |
| *** haomaiwang has quit IRC | 22:01 | |
| smcginnis | zohar: Yep. That opts.py file should already exist. | 22:01 |
| smcginnis | zohar: Running that command should regenerate it with the updated config opts. | 22:01 |
| smcginnis | zohar: Then you just add that file to your commit and resubmit. | 22:01 |
| *** haomaiwa_ has joined #openstack-cinder | 22:01 | |
| smcginnis | zohar: git add cinder/opts.py; git commit --amend; git review | 22:01 |
| zohar | in my review checkout, where devstack never ran, i don't have that file | 22:02 |
| zohar | is that fine? | 22:02 |
| *** alonmarx has quit IRC | 22:03 | |
| * smcginnis is checking something | 22:03 | |
| smcginnis | zohar: You might need to rebase on master for it. | 22:03 |
| smcginnis | zohar: It's here: https://github.com/openstack/cinder/blob/master/cinder/opts.py | 22:03 |
| smcginnis | zohar: Wouldn't hurt to do anyway. | 22:03 |
| smcginnis | zohar: git checkout master; git pull; git checkout yourbranch; git rebase master; rest of the commands from aboe. | 22:04 |
| smcginnis | s/aboe/above/ | 22:04 |
| *** IlyaG has joined #openstack-cinder | 22:05 | |
| openstackgerrit | Zohar Mamedov proposed openstack/cinder: Nexenta Edge iSCSI backend driver https://review.openstack.org/158745 | 22:17 |
| *** JoseMello has joined #openstack-cinder | 22:18 | |
| openstackgerrit | Sean McGinnis proposed openstack/cinder: RBD: Make snapshot_delete more robust https://review.openstack.org/151257 | 22:20 |
| *** diablo_rojo has quit IRC | 22:21 | |
| *** angela-s has quit IRC | 22:24 | |
| *** davechen has left #openstack-cinder | 22:24 | |
| *** xyang has joined #openstack-cinder | 22:25 | |
| xyang | smcginnis: hi, I submitted a fix to stable/liberty: https://review.openstack.org/#/c/250039/ | 22:26 |
| xyang | smcginnis: stable/liberty unit test is broken. https://review.openstack.org/#/c/249358/ | 22:27 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-cinderclient: Updated from global requirements https://review.openstack.org/247639 | 22:28 |
| xyang | smcginnis: I see the same failure in all stable/liberty patches. the problem was already fixed in the trunk as part of porting to py3. that's why I submitted a separate fix | 22:28 |
| xyang | smcginnis: let's see if my patch passes Jenkins:). I reproduced the problem on my setup and now it passes with the fix | 22:29 |
| *** sgundur has quit IRC | 22:30 | |
| *** jaypipes has quit IRC | 22:30 | |
| *** takedakn has quit IRC | 22:33 | |
| *** david-lyle has quit IRC | 22:34 | |
| *** jgregor has joined #openstack-cinder | 22:39 | |
| smcginnis | xyang: Thanks, I saw one of those failures late yesterday and didn't have a chance to dig into it. | 22:40 |
| smcginnis | xyang: If that passes I'll push it through. | 22:41 |
| xyang | smcginnis: thanks! | 22:41 |
| *** alonmarx has joined #openstack-cinder | 22:41 | |
| *** jgregor has quit IRC | 22:43 | |
| smcginnis | xyang: You're working pretty late on a virtual Friday. :) | 22:44 |
| *** alonmarx has quit IRC | 22:46 | |
| openstackgerrit | Sean McGinnis proposed openstack/cinder: Fix useless-else-on-loop pylint issues https://review.openstack.org/192655 | 22:46 |
| *** alonmarx has joined #openstack-cinder | 22:47 | |
| *** zohar has quit IRC | 22:51 | |
| *** alonmarx has quit IRC | 22:52 | |
| *** bardia has joined #openstack-cinder | 22:53 | |
| xyang | smcginnis: you too:) | 22:55 |
| smcginnis | xyang: I'm at least an hour later than you. :) | 22:56 |
| xyang | smcginnis: that's right:) | 22:56 |
| *** haomaiwa_ has quit IRC | 23:01 | |
| *** 7YUAAEOGN has joined #openstack-cinder | 23:01 | |
| *** david-lyle has joined #openstack-cinder | 23:07 | |
| *** erlon has quit IRC | 23:09 | |
| *** jwcroppe has quit IRC | 23:12 | |
| *** med_ has quit IRC | 23:17 | |
| *** akshai has quit IRC | 23:17 | |
| *** RA has joined #openstack-cinder | 23:20 | |
| *** RA is now known as Guest43529 | 23:20 | |
| *** akshai has joined #openstack-cinder | 23:20 | |
| *** david-lyle has quit IRC | 23:24 | |
| *** akshai has quit IRC | 23:25 | |
| *** med_ has joined #openstack-cinder | 23:25 | |
| *** med_ has quit IRC | 23:25 | |
| *** med_ has joined #openstack-cinder | 23:25 | |
| *** akerr_ is now known as akerr_away | 23:29 | |
| *** edtubill has joined #openstack-cinder | 23:34 | |
| openstackgerrit | Sean McGinnis proposed openstack/cinder: Nexenta Edge iSCSI backend driver https://review.openstack.org/158745 | 23:37 |
| *** edtubill has quit IRC | 23:39 | |
| *** EinstCrazy has joined #openstack-cinder | 23:48 | |
| *** alonmarx has joined #openstack-cinder | 23:48 | |
| *** alonmarx has quit IRC | 23:52 | |
| *** sgotliv has quit IRC | 23:52 | |
| *** yhayashi has joined #openstack-cinder | 23:52 | |
| *** EinstCrazy has quit IRC | 23:52 | |
| *** alonmarx has joined #openstack-cinder | 23:54 | |
| *** IlyaG_ has joined #openstack-cinder | 23:55 | |
| *** IlyaG has quit IRC | 23:57 | |
| *** alonmarx has quit IRC | 23:58 | |
| *** xyang1 has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!