| *** ddixit has quit IRC | 00:00 | |
| *** ddixit has joined #openstack-cinder | 00:00 | |
| *** harlowja has joined #openstack-cinder | 00:03 | |
| *** yangyapeng has quit IRC | 00:06 | |
| *** yangyapeng has joined #openstack-cinder | 00:07 | |
| *** harlowja has quit IRC | 00:09 | |
| *** yangyapeng has quit IRC | 00:11 | |
| *** s-shiono has joined #openstack-cinder | 00:12 | |
| *** harlowja has joined #openstack-cinder | 00:14 | |
| *** yangyapeng has joined #openstack-cinder | 00:18 | |
| *** edmondsw has joined #openstack-cinder | 00:19 | |
| *** yangyapeng has quit IRC | 00:22 | |
| *** edmondsw has quit IRC | 00:24 | |
| *** wanghao has quit IRC | 00:27 | |
| *** wanghao has joined #openstack-cinder | 00:27 | |
| *** gouthamr has quit IRC | 00:45 | |
| openstackgerrit | Sean McGinnis proposed openstack/cinder master: Remove API v1 https://review.openstack.org/499342 | 00:46 |
|---|---|---|
| *** dave-mccowan has quit IRC | 00:49 | |
| *** gouthamr has joined #openstack-cinder | 00:49 | |
| *** zhurong has joined #openstack-cinder | 00:53 | |
| *** mriedem has quit IRC | 00:55 | |
| openstackgerrit | TommyLike proposed openstack/cinder-specs master: Show resource's total count info in list APIs https://review.openstack.org/500665 | 00:55 |
| *** harlowja has quit IRC | 00:57 | |
| openstackgerrit | Yingzhe Zeng proposed openstack/cinder master: Fix an compatibility issue after Huawei driver upgrade https://review.openstack.org/500746 | 01:00 |
| *** harlowja has joined #openstack-cinder | 01:01 | |
| *** ddixit has quit IRC | 01:03 | |
| *** ddixit has joined #openstack-cinder | 01:03 | |
| *** Apoorva_ has joined #openstack-cinder | 01:05 | |
| *** nicolasbock has quit IRC | 01:06 | |
| openstackgerrit | Merged openstack/cinder master: RBD: Fix stats reporting https://review.openstack.org/486734 | 01:09 |
| *** Apoorva has quit IRC | 01:09 | |
| *** Apoorva_ has quit IRC | 01:10 | |
| *** yangyapeng has joined #openstack-cinder | 01:15 | |
| *** nicolasbock has joined #openstack-cinder | 01:21 | |
| *** scottda_ has quit IRC | 01:24 | |
| *** markvoelker has joined #openstack-cinder | 01:43 | |
| *** itlinux has joined #openstack-cinder | 01:54 | |
| *** markstur has quit IRC | 01:57 | |
| *** edmondsw has joined #openstack-cinder | 01:58 | |
| *** lhx__ has joined #openstack-cinder | 02:00 | |
| *** edmondsw has quit IRC | 02:02 | |
| *** winston-d has joined #openstack-cinder | 02:13 | |
| openstackgerrit | Vipin Balachandran proposed openstack/cinder master: VMware: Use vSphere template as snapshot format https://review.openstack.org/501070 | 02:15 |
| openstackgerrit | Vipin Balachandran proposed openstack/cinder master: VMware: Use vSphere template as snapshot format https://review.openstack.org/501070 | 02:16 |
| *** markvoelker has quit IRC | 02:18 | |
| *** lhx__ has quit IRC | 02:51 | |
| *** lhx__ has joined #openstack-cinder | 02:51 | |
| *** wanghao has quit IRC | 02:54 | |
| *** wanghao has joined #openstack-cinder | 02:55 | |
| *** markstur has joined #openstack-cinder | 02:56 | |
| *** zhurong has quit IRC | 02:58 | |
| *** yangyape_ has joined #openstack-cinder | 03:11 | |
| *** yangyap__ has joined #openstack-cinder | 03:14 | |
| *** yangyape_ has quit IRC | 03:14 | |
| *** yangyapeng has quit IRC | 03:14 | |
| *** markvoelker has joined #openstack-cinder | 03:15 | |
| *** gkadam has joined #openstack-cinder | 03:16 | |
| *** yangyap__ has quit IRC | 03:19 | |
| *** yangyapeng has joined #openstack-cinder | 03:19 | |
| *** nicolasbock has quit IRC | 03:20 | |
| openstackgerrit | zhangbailin proposed openstack/cinder-specs master: Update url of the cinder-specs's document https://review.openstack.org/501098 | 03:23 |
| *** links has joined #openstack-cinder | 03:28 | |
| *** yangyape_ has joined #openstack-cinder | 03:30 | |
| *** yangyapeng has quit IRC | 03:30 | |
| *** gouthamr has quit IRC | 03:46 | |
| *** edmondsw has joined #openstack-cinder | 03:46 | |
| *** markvoelker has quit IRC | 03:48 | |
| *** edmondsw has quit IRC | 03:51 | |
| *** adisky__ has quit IRC | 03:55 | |
| *** zhurong has joined #openstack-cinder | 04:09 | |
| *** gcb has quit IRC | 04:19 | |
| openstackgerrit | lihaijing proposed openstack/cinder master: Api-ref: fix v2/v3 hosts extension api doc https://review.openstack.org/477798 | 04:20 |
| *** gcb has joined #openstack-cinder | 04:21 | |
| *** psachin has joined #openstack-cinder | 04:31 | |
| *** armax has quit IRC | 04:44 | |
| *** markvoelker has joined #openstack-cinder | 04:45 | |
| *** markstur_ has joined #openstack-cinder | 05:08 | |
| *** markstur has quit IRC | 05:08 | |
| *** markvoelker has quit IRC | 05:18 | |
| *** lhx__ has quit IRC | 05:20 | |
| *** lhx__ has joined #openstack-cinder | 05:20 | |
| *** pcaruana has joined #openstack-cinder | 05:27 | |
| *** markstur has joined #openstack-cinder | 05:30 | |
| *** lkuchlan_wfh has joined #openstack-cinder | 05:31 | |
| *** markstur_ has quit IRC | 05:33 | |
| *** lhx__ has quit IRC | 05:35 | |
| *** sticker has quit IRC | 05:37 | |
| *** liverpooler has joined #openstack-cinder | 05:46 | |
| *** yangyapeng has joined #openstack-cinder | 05:57 | |
| *** yangyape_ has quit IRC | 05:57 | |
| *** markstur has quit IRC | 06:00 | |
| *** zhaochao has quit IRC | 06:03 | |
| openstackgerrit | zhangbailin proposed openstack/python-cinderclient master: Update old url for cinderclient's document https://review.openstack.org/501113 | 06:06 |
| *** zhaochao has joined #openstack-cinder | 06:08 | |
| *** markstur has joined #openstack-cinder | 06:21 | |
| *** markstur has quit IRC | 06:25 | |
| *** e0ne has joined #openstack-cinder | 06:28 | |
| *** links has quit IRC | 06:29 | |
| *** psachin has quit IRC | 06:29 | |
| *** lhx__ has joined #openstack-cinder | 06:30 | |
| *** yangyapeng has quit IRC | 06:35 | |
| *** yangyapeng has joined #openstack-cinder | 06:35 | |
| *** adisky__ has joined #openstack-cinder | 06:38 | |
| *** dulek has joined #openstack-cinder | 06:51 | |
| *** links has joined #openstack-cinder | 06:55 | |
| *** rcernin has joined #openstack-cinder | 06:57 | |
| *** e0ne has quit IRC | 06:57 | |
| *** abhi89 has joined #openstack-cinder | 06:57 | |
| *** psachin has joined #openstack-cinder | 06:59 | |
| *** pgadiya has joined #openstack-cinder | 07:12 | |
| *** markvoelker has joined #openstack-cinder | 07:15 | |
| openstackgerrit | yixuan zhang proposed openstack/cinder master: Storwize: fix the wwpns error after NPIV enable. https://review.openstack.org/501128 | 07:16 |
| openstackgerrit | yanghuichan proposed openstack/cinder master: Fix wrong links in Cinder https://review.openstack.org/501129 | 07:18 |
| *** edmondsw has joined #openstack-cinder | 07:22 | |
| *** edmondsw has quit IRC | 07:26 | |
| *** tesseract has joined #openstack-cinder | 07:30 | |
| *** lhx__ has quit IRC | 07:31 | |
| *** lhx__ has joined #openstack-cinder | 07:32 | |
| *** markvoelker has quit IRC | 07:49 | |
| *** hoonetorg has quit IRC | 07:58 | |
| *** edmondsw has joined #openstack-cinder | 08:03 | |
| openstackgerrit | Jesse Wu proposed openstack/cinder master: Synology: Driver unable to be initialized https://review.openstack.org/474917 | 08:10 |
| *** lhx__ has quit IRC | 08:12 | |
| *** ociuhandu_ has quit IRC | 08:14 | |
| *** hoonetorg has joined #openstack-cinder | 08:15 | |
| *** yangyapeng has quit IRC | 08:19 | |
| *** yangyapeng has joined #openstack-cinder | 08:20 | |
| *** gcb has quit IRC | 08:23 | |
| *** liverpooler has quit IRC | 08:25 | |
| *** lhx__ has joined #openstack-cinder | 08:25 | |
| *** s-shiono has quit IRC | 08:34 | |
| openstackgerrit | Arnon Yaari proposed openstack/cinder master: INFINIDAT: add support for overprovisioning https://review.openstack.org/500783 | 08:44 |
| *** markvoelker has joined #openstack-cinder | 08:46 | |
| *** alexchadin has joined #openstack-cinder | 08:47 | |
| *** e0ne has joined #openstack-cinder | 09:03 | |
| *** dalgaaf has quit IRC | 09:07 | |
| *** ociuhandu has joined #openstack-cinder | 09:08 | |
| *** dalgaaf has joined #openstack-cinder | 09:10 | |
| *** lpetrut has joined #openstack-cinder | 09:13 | |
| *** yangyape_ has joined #openstack-cinder | 09:15 | |
| *** ociuhandu has quit IRC | 09:18 | |
| *** openstackgerrit has quit IRC | 09:18 | |
| *** yangyapeng has quit IRC | 09:18 | |
| *** yangyape_ has quit IRC | 09:19 | |
| *** yangyapeng has joined #openstack-cinder | 09:19 | |
| *** markvoelker has quit IRC | 09:19 | |
| *** ociuhandu has joined #openstack-cinder | 09:20 | |
| *** zengyingzhe_ has joined #openstack-cinder | 09:23 | |
| *** zengyingzhe__ has quit IRC | 09:26 | |
| *** openstackgerrit has joined #openstack-cinder | 09:29 | |
| openstackgerrit | yixuan zhang proposed openstack/cinder master: Storwize: add NPIV support https://review.openstack.org/501128 | 09:29 |
| *** adisky__ has quit IRC | 09:32 | |
| *** stakeda has quit IRC | 09:35 | |
| *** HamdyK has joined #openstack-cinder | 09:37 | |
| *** alexchadin has quit IRC | 09:42 | |
| *** ociuhandu_ has joined #openstack-cinder | 09:52 | |
| *** sdague has joined #openstack-cinder | 09:53 | |
| *** ociuhandu has quit IRC | 09:54 | |
| *** yangyapeng has quit IRC | 09:56 | |
| *** yangyapeng has joined #openstack-cinder | 09:56 | |
| openstackgerrit | Dai Dang Van proposed openstack/cinder master: Remove oslo.config from cinder-config-generator https://review.openstack.org/487663 | 09:59 |
| *** nicolasbock has joined #openstack-cinder | 10:02 | |
| *** openstackgerrit has quit IRC | 10:03 | |
| *** nicolasbock has quit IRC | 10:07 | |
| *** abhi89 has quit IRC | 10:11 | |
| *** openstackgerrit has joined #openstack-cinder | 10:12 | |
| openstackgerrit | Anton Arefiev proposed openstack/cinder master: RBD: get provisioned capacity using same connection https://review.openstack.org/500047 | 10:12 |
| *** edmondsw has quit IRC | 10:13 | |
| *** markvoelker has joined #openstack-cinder | 10:17 | |
| *** HamdyK has quit IRC | 10:17 | |
| *** ociuhandu_ has quit IRC | 10:18 | |
| *** yangyapeng has quit IRC | 10:25 | |
| *** yangyapeng has joined #openstack-cinder | 10:25 | |
| *** yangyapeng has quit IRC | 10:29 | |
| *** nicolasbock has joined #openstack-cinder | 10:36 | |
| openstackgerrit | Ivan Kolodyazhny proposed openstack/cinder master: Deprecate service-to-driver mapping for backups https://review.openstack.org/496816 | 10:41 |
| *** gaurangt has quit IRC | 10:48 | |
| *** markvoelker has quit IRC | 10:49 | |
| *** ganso has joined #openstack-cinder | 10:49 | |
| *** lhx_ has joined #openstack-cinder | 10:50 | |
| *** gaurangt has joined #openstack-cinder | 10:52 | |
| *** liverpooler has joined #openstack-cinder | 10:52 | |
| *** HamdyK has joined #openstack-cinder | 10:52 | |
| *** lhx__ has quit IRC | 10:53 | |
| *** zhurong has quit IRC | 10:54 | |
| *** alexchadin has joined #openstack-cinder | 10:58 | |
| *** alexchadin has joined #openstack-cinder | 10:58 | |
| *** nicolasbock has quit IRC | 11:07 | |
| *** nicolasbock has joined #openstack-cinder | 11:09 | |
| *** psachin has quit IRC | 11:19 | |
| *** pgadiya has quit IRC | 11:19 | |
| *** links has quit IRC | 11:20 | |
| openstackgerrit | Arnon Yaari proposed openstack/cinder master: INFINIDAT: add support for overprovisioning https://review.openstack.org/500783 | 11:25 |
| *** abhi89 has joined #openstack-cinder | 11:27 | |
| *** sumitshatwara has quit IRC | 11:28 | |
| *** yangyapeng has joined #openstack-cinder | 11:29 | |
| *** psachin has joined #openstack-cinder | 11:32 | |
| *** links has joined #openstack-cinder | 11:32 | |
| *** pgadiya has joined #openstack-cinder | 11:33 | |
| *** dave-mccowan has joined #openstack-cinder | 11:39 | |
| *** nicolasbock has quit IRC | 11:40 | |
| *** markvoelker has joined #openstack-cinder | 11:46 | |
| *** pgadiya has quit IRC | 11:48 | |
| *** lpetrut has quit IRC | 11:51 | |
| *** nicolasbock has joined #openstack-cinder | 11:52 | |
| *** alexchadin has quit IRC | 11:57 | |
| abhi89 | geguileo: hi.. reg https://review.openstack.org/#/c/500917/.. i had thought of covering TC for this change but couldn't find any UT related to that method.. It would be helpful if you can point me where/which TC i need to change for my changes.. | 12:03 |
| *** yangyapeng has quit IRC | 12:05 | |
| *** yangyapeng has joined #openstack-cinder | 12:05 | |
| *** gcb has joined #openstack-cinder | 12:05 | |
| geguileo | abhi89: then you just need to add one | 12:06 |
| geguileo | abhi89: The code in there is probably covered by tests that are not following the correct unit tests procedures | 12:06 |
| geguileo | abhi89: and are indirectly testing it via the method that actually calls that one | 12:06 |
| geguileo | abhi89: check cinder/tests/unit/group/test_groups_api.py | 12:06 |
| *** adrianofr has joined #openstack-cinder | 12:07 | |
| *** markstur has joined #openstack-cinder | 12:09 | |
| smcginnis | geguileo: Hey, if you have a moment, I'd apprecaite your eyes on https://review.openstack.org/#/c/500194/ | 12:09 |
| *** yangyapeng has quit IRC | 12:09 | |
| geguileo | smcginnis: will do | 12:10 |
| smcginnis | geguileo: Thanks | 12:10 |
| *** edmondsw has joined #openstack-cinder | 12:10 | |
| *** ociuhandu has joined #openstack-cinder | 12:10 | |
| *** edmondsw has quit IRC | 12:11 | |
| *** edmondsw has joined #openstack-cinder | 12:11 | |
| *** markstur has quit IRC | 12:13 | |
| *** markvoelker has quit IRC | 12:20 | |
| abhi89 | geguileo: I had checked cinder/tests/unit/group/test_groups_api.py.. _validate_add_volumes() where I have my change is called by group update api call.. we have TC covered for update().. but nowhere have we referenced _validate_add_volumes() in that TC not have we written a separate TC for _validate_add_volumes.. | 12:21 |
| *** lhx_ has quit IRC | 12:21 | |
| geguileo | abhi89: that's why I say that the code in there is probably covered by the test of the method that calls it | 12:21 |
| *** lhx_ has joined #openstack-cinder | 12:21 | |
| geguileo | even though that goes agains the unit test way of doing things | 12:22 |
| geguileo | you can just add a test that tests _validate_add_volumes and just covers the case you have added | 12:22 |
| geguileo | I'm OK with that, I won't require you to add all other tests | 12:22 |
| geguileo | but you should have a test for what you've fixed | 12:23 |
| abhi89 | geguileo: sure.. i will add the TC for it.. | 12:23 |
| geguileo | abhi89: thanks | 12:23 |
| *** markvoelker has joined #openstack-cinder | 12:28 | |
| *** lpetrut has joined #openstack-cinder | 12:28 | |
| *** evrardjp_ has quit IRC | 12:35 | |
| *** evrardjp has joined #openstack-cinder | 12:35 | |
| openstackgerrit | Sean McGinnis proposed openstack/cinder master: Remove API v1 https://review.openstack.org/499342 | 12:39 |
| *** alexchadin has joined #openstack-cinder | 12:40 | |
| *** scottda_ has joined #openstack-cinder | 12:49 | |
| *** abishop has joined #openstack-cinder | 12:50 | |
| *** gkadam has quit IRC | 12:59 | |
| *** lhx_ has quit IRC | 13:02 | |
| *** mikal has quit IRC | 13:05 | |
| *** xyang1 has joined #openstack-cinder | 13:05 | |
| *** adisky__ has joined #openstack-cinder | 13:07 | |
| *** david-lyle has quit IRC | 13:09 | |
| *** david-lyle has joined #openstack-cinder | 13:09 | |
| *** takedakn has joined #openstack-cinder | 13:10 | |
| *** lucasxu has joined #openstack-cinder | 13:10 | |
| *** mikal has joined #openstack-cinder | 13:12 | |
| *** dulek has quit IRC | 13:12 | |
| openstackgerrit | Chhavi Agarwal proposed openstack/os-brick master: FC PPC64 device discovery issue https://review.openstack.org/501253 | 13:13 |
| *** mriedem has joined #openstack-cinder | 13:13 | |
| *** gouthamr has joined #openstack-cinder | 13:14 | |
| *** dulek has joined #openstack-cinder | 13:15 | |
| *** liverpooler has quit IRC | 13:16 | |
| openstackgerrit | Steve Noyes proposed openstack/cinder master: Fix attachments on volume show when volume is attaching https://review.openstack.org/500102 | 13:16 |
| *** yangyapeng has joined #openstack-cinder | 13:16 | |
| *** liverpooler has joined #openstack-cinder | 13:16 | |
| *** zengyingzhe_ has quit IRC | 13:16 | |
| *** mchlumsky has joined #openstack-cinder | 13:17 | |
| *** zengyingzhe_ has joined #openstack-cinder | 13:17 | |
| *** yangyapeng has quit IRC | 13:20 | |
| openstackgerrit | Chuck Fouts proposed openstack/cinder master: NetApp: Adds logging of ONTAP API version https://review.openstack.org/500965 | 13:28 |
| *** catintheroof has joined #openstack-cinder | 13:29 | |
| *** thegreenhundred has joined #openstack-cinder | 13:30 | |
| *** crose has joined #openstack-cinder | 13:32 | |
| *** alexchadin has quit IRC | 13:43 | |
| *** yangyapeng has joined #openstack-cinder | 13:43 | |
| kwathore | Hi Team, Please review below link: https://review.openstack.org/#/c/499970/ | 13:45 |
| smcginnis | kwathore: Done. | 13:45 |
| *** alexchadin has joined #openstack-cinder | 13:45 | |
| kwathore | Sean: Thanks | 13:47 |
| *** yangyapeng has quit IRC | 13:48 | |
| *** crushil has quit IRC | 13:50 | |
| *** liverpooler has quit IRC | 13:50 | |
| *** Yogi1 has joined #openstack-cinder | 13:52 | |
| *** liverpooler has joined #openstack-cinder | 13:52 | |
| *** crushil has joined #openstack-cinder | 13:58 | |
| *** cknight has joined #openstack-cinder | 14:05 | |
| *** armax has joined #openstack-cinder | 14:10 | |
| *** takedakn has quit IRC | 14:17 | |
| *** ociuhandu has quit IRC | 14:17 | |
| *** yangyapeng has joined #openstack-cinder | 14:18 | |
| *** dustins has joined #openstack-cinder | 14:19 | |
| *** aspiers has quit IRC | 14:21 | |
| *** yangyapeng has quit IRC | 14:22 | |
| *** ntpttr_laptop has joined #openstack-cinder | 14:23 | |
| *** eharney has joined #openstack-cinder | 14:23 | |
| *** lkuchlan_wfh has quit IRC | 14:24 | |
| *** aspiers has joined #openstack-cinder | 14:27 | |
| *** links has quit IRC | 14:31 | |
| *** markstur has joined #openstack-cinder | 14:32 | |
| *** juzuluag_ has joined #openstack-cinder | 14:34 | |
| *** markstur has quit IRC | 14:37 | |
| openstackgerrit | Evgeny Antyshev proposed openstack/cinder master: vzstorage: return from _create_cloned_volume https://review.openstack.org/494539 | 14:37 |
| *** lhx_ has joined #openstack-cinder | 14:43 | |
| *** wN has quit IRC | 14:47 | |
| HamdyK | Hi guys, Can you please review this commit https://review.openstack.org/#/c/482562/ | 14:48 |
| *** alexchadin has quit IRC | 14:51 | |
| *** markstur has joined #openstack-cinder | 14:52 | |
| *** wN has joined #openstack-cinder | 14:53 | |
| *** wN has quit IRC | 14:53 | |
| *** wN has joined #openstack-cinder | 14:53 | |
| openstackgerrit | Eric Harney proposed openstack/cinder master: Enable hash randomization in unit tests https://review.openstack.org/465636 | 14:54 |
| *** markstur_ has joined #openstack-cinder | 14:54 | |
| *** HamdyK has quit IRC | 14:55 | |
| *** esker has joined #openstack-cinder | 14:57 | |
| *** esker has quit IRC | 14:57 | |
| *** markstur has quit IRC | 14:58 | |
| *** esker has joined #openstack-cinder | 14:59 | |
| *** juzuluag_ has quit IRC | 15:01 | |
| openstackgerrit | Evgeny Antyshev proposed openstack/cinder master: vzstorage: return from _create_cloned_volume https://review.openstack.org/494539 | 15:02 |
| *** felipemonteiro has joined #openstack-cinder | 15:03 | |
| *** felipemonteiro_ has joined #openstack-cinder | 15:05 | |
| *** adrianofr has quit IRC | 15:07 | |
| *** felipemonteiro has quit IRC | 15:08 | |
| *** david-lyle has quit IRC | 15:10 | |
| *** itlinux has quit IRC | 15:11 | |
| *** david-lyle has joined #openstack-cinder | 15:15 | |
| openstackgerrit | Eric Harney proposed openstack/cinder master: Tests: Don't write files to /tmp/nec/ https://review.openstack.org/501309 | 15:17 |
| *** adrianofr has joined #openstack-cinder | 15:35 | |
| *** felipemonteiro__ has joined #openstack-cinder | 15:38 | |
| *** lkuchlan_wfh has joined #openstack-cinder | 15:39 | |
| eharney | xyang1: have you seen this? https://bugs.launchpad.net/cinder/+bug/1711455 | 15:41 |
| openstack | Launchpad bug 1711455 in Cinder "GroupsTest.test_group_update failed with MismatchError: 2 != 1" [High,New] | 15:41 |
| *** felipemonteiro_ has quit IRC | 15:41 | |
| xyang1 | eharney: not yet. Will take a look | 15:41 |
| eharney | xyang1: thanks | 15:42 |
| *** felipemonteiro_ has joined #openstack-cinder | 15:43 | |
| hemna | wait, 2 doesn't equal 1 ? | 15:45 |
| *** felipemonteiro__ has quit IRC | 15:46 | |
| eharney | only for very large values of 1 :) | 15:46 |
| *** kaisers_ has joined #openstack-cinder | 15:47 | |
| e0ne | :) | 15:47 |
| *** ntpttr_laptop has quit IRC | 15:57 | |
| xyang1 | eharney: do you how to get cinder logs from this? I couldn't find it | 15:59 |
| eharney | xyang1: they're up in the directory tree from console.html -- http://logs.openstack.org/53/494553/1/check/gate-tempest-dsvm-neutron-full-ubuntu-xenial/1793ab8/logs/ | 16:00 |
| xyang1 | eharney: thanks | 16:00 |
| *** lucasxu has quit IRC | 16:00 | |
| *** lpetrut_ has joined #openstack-cinder | 16:00 | |
| *** lpetrut has quit IRC | 16:01 | |
| *** lpetrut__ has joined #openstack-cinder | 16:03 | |
| openstackgerrit | Sean McGinnis proposed openstack/cinder master: Remove Coho volume driver https://review.openstack.org/501327 | 16:03 |
| openstackgerrit | Sean McGinnis proposed openstack/cinder master: Remove Blockbridge volume driver https://review.openstack.org/501328 | 16:03 |
| openstackgerrit | Sean McGinnis proposed openstack/cinder master: Remove FalconStor volume drivers https://review.openstack.org/501329 | 16:03 |
| openstackgerrit | Sean McGinnis proposed openstack/cinder master: Remove Infortrend drivers https://review.openstack.org/501330 | 16:03 |
| openstackgerrit | Sean McGinnis proposed openstack/cinder master: Remove QNAP volume driver https://review.openstack.org/501331 | 16:03 |
| openstackgerrit | Sean McGinnis proposed openstack/cinder master: Remove Reduxio volume driver https://review.openstack.org/501332 | 16:03 |
| openstackgerrit | Sean McGinnis proposed openstack/cinder master: Remove Tegile volume driver https://review.openstack.org/501333 | 16:03 |
| openstackgerrit | Sean McGinnis proposed openstack/cinder master: Remove Violin volume drivers https://review.openstack.org/501334 | 16:03 |
| openstackgerrit | Sean McGinnis proposed openstack/cinder master: Remove X-IO volume driver https://review.openstack.org/501335 | 16:03 |
| openstackgerrit | Sean McGinnis proposed openstack/cinder master: Remove ZTE volume driver https://review.openstack.org/501336 | 16:03 |
| *** david-lyle has quit IRC | 16:04 | |
| *** dklyle has joined #openstack-cinder | 16:04 | |
| *** ntpttr_laptop has joined #openstack-cinder | 16:04 | |
| *** lpetrut_ has quit IRC | 16:05 | |
| e0ne | smcginnis: good clean up! | 16:09 |
| smcginnis | e0ne: Going for the most lines of code removed for Queens. ;) | 16:13 |
| e0ne | smcginnis: chellange accepted! | 16:13 |
| smcginnis | Hehe | 16:13 |
| *** itlinux has joined #openstack-cinder | 16:14 | |
| *** dklyle has quit IRC | 16:15 | |
| *** dave-mccowan has quit IRC | 16:17 | |
| *** david-lyle has joined #openstack-cinder | 16:18 | |
| *** lkuchlan_wfh has quit IRC | 16:23 | |
| *** Apoorva has joined #openstack-cinder | 16:29 | |
| *** felipemonteiro__ has joined #openstack-cinder | 16:30 | |
| *** itlinux_ has joined #openstack-cinder | 16:31 | |
| *** itlinux has quit IRC | 16:32 | |
| *** felipemonteiro__ has quit IRC | 16:33 | |
| *** felipemonteiro__ has joined #openstack-cinder | 16:33 | |
| *** felipemonteiro_ has quit IRC | 16:34 | |
| *** dave-mccowan has joined #openstack-cinder | 16:38 | |
| *** lucasxu has joined #openstack-cinder | 16:47 | |
| geguileo | tommylikehu: in my defense according to our docs we should have the microversion, that's why I said yes ;-) | 16:47 |
| openstackgerrit | Merged openstack/cinder master: Fix an compatibility issue after Huawei driver upgrade https://review.openstack.org/500746 | 16:48 |
| tommylikehu | geguileo: ok.. | 16:48 |
| geguileo | because if I'm not mistaken this changes the query parameter values of a non free form field | 16:48 |
| geguileo | but I agree that it's crazy to bump a microversion for these kind of things if there's a discovery mechanism for possible values | 16:48 |
| jgriffith | smcginnis tommylikehu sorry... was tied up; So my proposal was to provide an api call that: | 16:48 |
| jgriffith | a. Lists available/valid filters | 16:49 |
| jgriffith | b. Lists configured filters | 16:49 |
| jgriffith | I guess that's 2 api calls :) | 16:49 |
| smcginnis | Hah | 16:49 |
| tommylikehu | jgriffith: come on man | 16:49 |
| smcginnis | There wasn't actually a comment earlier about 1 != 2 | 16:49 |
| jgriffith | sue me :) | 16:50 |
| smcginnis | :) | 16:50 |
| *** felipemonteiro_ has joined #openstack-cinder | 16:50 | |
| *** felipemonteiro__ has quit IRC | 16:54 | |
| tommylikehu | jgriffith, smcginnis: if most of our users use the release version, then the doc which describes the valid filters for this release is good enough I guess. | 16:54 |
| *** psachin has quit IRC | 16:55 | |
| *** dave-mccowan has quit IRC | 16:56 | |
| *** dave-mccowan has joined #openstack-cinder | 16:56 | |
| smcginnis | tommylikehu: We should see if there is somewhere in the configuration reference that we can note this. | 16:57 |
| smcginnis | Then if they use the config reference for whichever release they are using, they will have accurate information. | 16:57 |
| tommylikehu | smcginnis: yes, and I already added one: https://github.com/openstack/cinder/blob/master/doc/source/man/generalized_filters.rst | 16:59 |
| *** lpetrut_ has joined #openstack-cinder | 16:59 | |
| bswartz | geguileo: did you want to talk here? | 17:00 |
| geguileo | bswartz: hi | 17:00 |
| bswartz | gouthamr: ^ | 17:00 |
| gouthamr | pong | 17:00 |
| geguileo | bswartz: any idea when over provisioning will be fixed? | 17:00 |
| *** e0ne has quit IRC | 17:01 | |
| *** lpetrut__ has quit IRC | 17:01 | |
| bswartz | let me find out who is working on that specific bug -- we definitely know about it | 17:01 |
| geguileo | bswartz: yeah, sorry about it reaching you by 2 different ways... | 17:02 |
| *** rcernin has quit IRC | 17:02 | |
| *** yangyapeng has joined #openstack-cinder | 17:06 | |
| *** harlowja has quit IRC | 17:07 | |
| *** harlowja has joined #openstack-cinder | 17:07 | |
| *** felipemonteiro_ has quit IRC | 17:08 | |
| *** lpetrut_ has quit IRC | 17:09 | |
| *** felipemonteiro_ has joined #openstack-cinder | 17:09 | |
| *** yangyapeng has quit IRC | 17:10 | |
| *** felipemonteiro__ has joined #openstack-cinder | 17:11 | |
| * gouthamr #end hallway conversation | 17:12 | |
| bswartz | geguileo: okay it appears that bug is on the backlog but not being actively worked on | 17:12 |
| geguileo | :'-( | 17:12 |
| bswartz | the fix is expected to be quite simple but testing it will require some effort | 17:12 |
| *** Aditya has joined #openstack-cinder | 17:13 | |
| bswartz | I think the bug stems from a misunderstanding about the thin provisioning design going all the way back to paris | 17:13 |
| geguileo | yeah, that's why I started working on the stats spec | 17:13 |
| geguileo | because we are all understanding different things regarding what the terms mean | 17:13 |
| *** Aditya has quit IRC | 17:13 | |
| bswartz | We believe that we simply shouldn't be reporting a value for provisioned capacity, because the value we're computing has nothing to do with what the scheduler expects | 17:13 |
| geguileo | so drivers reports something that later on doesn't match with how cinder core code does the calculations | 17:13 |
| *** Aditya has joined #openstack-cinder | 17:14 | |
| *** felipemonteiro_ has quit IRC | 17:14 | |
| geguileo | bswartz: my initial idea was to do that, just remove it | 17:14 |
| bswartz | and the scheduler is able to compute its own value which will be correct | 17:14 |
| bswartz | so what needs to happens is that we need to remove the calculation and then try it out in some real world scenarios | 17:15 |
| geguileo | bswartz: but it will only be correct if sum(pool_volumes.sizes) = sum(backend_volumes.sizes) | 17:15 |
| bswartz | it's unfortunate that automated test coverage doesn't exist in this area | 17:15 |
| bswartz | so the testing will be manual | 17:15 |
| geguileo | bswartz: does the netapp driver support multiple pools on each backend? | 17:15 |
| bswartz | if we had automated test coverage it would probably have been fixed and tested already | 17:15 |
| bswartz | geguileo: yes | 17:16 |
| bswartz | but cinder knows which volumes are in which pools, so it's able to compute a per-pool sum if it wishes | 17:16 |
| geguileo | bswartz: then I believe you cannot just remove the field | 17:16 |
| *** JoseMello has joined #openstack-cinder | 17:16 | |
| geguileo | bswartz: It doesn't work like that right now | 17:16 |
| geguileo | afaik | 17:16 |
| bswartz | I see | 17:17 |
| bswartz | well the information the cinder scheduler wants seems to exist in the database and not in the storage controller | 17:17 |
| bswartz | so it would be weird to put a fix in the driver | 17:17 |
| bswartz | we can stop reporting the incorrect value, but the calculation of the correct value probably needs to come from the manager or the scheduler itself | 17:18 |
| bswartz | we can help write that code if needed | 17:18 |
| *** ntpttr_laptop has quit IRC | 17:18 | |
| geguileo | bswartz: here you can see what happens when you don't repport it https://github.com/openstack/cinder/blob/master/cinder/scheduler/host_manager.py#L345 | 17:18 |
| bswartz | but that's more effort than just removing some bad code and testing the change | 17:19 |
| geguileo | bswartz: the thing is that you could have other pools and volumes not used by cinder, so Cinder wouldn't be able to calculate the actual provisioned capacity | 17:19 |
| bswartz | okay the comments don't clear anything up | 17:20 |
| jgriffith | smcginnis tommylikehu just keep in mind you need a way for automation (ie Ansible) to query things like this | 17:20 |
| bswartz | I'm as confused today as I was in Paris | 17:20 |
| geguileo | bswartz: lol | 17:20 |
| geguileo | bswartz: I tried to improve it a little bit: https://review.openstack.org/#/c/490116/2/specs/queens/provisioning-improvements.rst | 17:21 |
| bswartz | geguileo: perhaps you could explain in your own words what provisioned_capacity_gb and allocated_capacity_gb should be | 17:21 |
| geguileo | bswartz: I explained it in that doc, but basically | 17:21 |
| bswartz | oh good | 17:21 |
| geguileo | allocated = volumes cinder know about, should not be reported by the driver | 17:21 |
| bswartz | I'm reading it now | 17:22 |
| geguileo | provisioned = every volume in the backend | 17:22 |
| geguileo | and both are "virtual sizes" not real disk usage | 17:22 |
| geguileo | if the spec is not clear please add a comment, as I'd like to ensure that there are no more confussion | 17:22 |
| bswartz | okay how is provisioned different from total space (minus) free space? | 17:22 |
| geguileo | 2 lines above is the answer | 17:23 |
| geguileo | they are "virtual sizes", which means the maximum size the volume can be | 17:23 |
| geguileo | if it's full | 17:23 |
| bswartz | okay what about stuff other than volumes that consumes space out of the pool? | 17:24 |
| geguileo | provisioned is not regarding the pool is in the backend | 17:24 |
| geguileo | the whole backend | 17:24 |
| geguileo | and it's not about physical usage | 17:24 |
| geguileo | so if snapshots take up space, it doesn't matter | 17:25 |
| geguileo | they are not reported there | 17:25 |
| *** lpetrut has joined #openstack-cinder | 17:25 | |
| geguileo | at least that's what I understand after reading the various specs, reading the code, etc. | 17:25 |
| bswartz | has xyang1 reviewed your spec yet? does she agree with your definitions? | 17:26 |
| geguileo | bswartz: she has taken a look, on the first patch she pointed out a mistake | 17:26 |
| geguileo | but I'm not sure if she has given it a real good look or just a quick one | 17:27 |
| xyang1 | bswartz: I did a quick look of the first patch set. Will take a closer look later | 17:27 |
| bswartz | okay give me a few moments to absorb this whole document | 17:27 |
| geguileo | bswartz: and please, please, point out everything that is too convoluted, or just not clear | 17:27 |
| xyang1 | bswartz: geguileo I'll try to get to it this week | 17:27 |
| geguileo | xyang1: thanks!! | 17:28 |
| xyang1 | geguileo: np | 17:28 |
| bswartz | so my problem is that concepts around thin provisioning and space accounting are inherently complex and hard to understand -- you're forced to make some simplifying assumptions if you want to maintain sanity | 17:28 |
| *** ntpttr_laptop has joined #openstack-cinder | 17:29 | |
| *** diablo_rojo has quit IRC | 17:29 | |
| geguileo | and you don't think Cinder's approach is right? | 17:31 |
| geguileo | or what's the problem you se/ | 17:31 |
| geguileo | see? | 17:31 |
| bswartz | cinder makes one set of simplifying assumptions -- which is a good thing -- as long as we agree on them and understand them | 17:32 |
| mriedem | did something change recently in cinder/castellan related to the key manager and how it's instantiated? http://logs.openstack.org/70/500770/7/check/gate-tempest-dsvm-neutron-full-ubuntu-xenial/b813494/logs/screen-c-api.txt.gz?level=TRACE#_Sep_06_17_25_08_182255 | 17:32 |
| geguileo | bswartz: I completely agree!!! | 17:32 |
| *** adisky__ has quit IRC | 17:32 | |
| geguileo | that's why I'm making the effort on this area | 17:32 |
| geguileo | because I think everyone understand something different | 17:33 |
| geguileo | and currect docs don't help | 17:33 |
| bswartz | the key is to make sure that we're on the same page about which complexity we're ignoring -- the complexity won't go away, but can pretend it's not there to make the problem tractable | 17:33 |
| bswartz | in the case of your document it appears we're chosing to ignore phantom space effects like free space created by the magic of deduplication | 17:34 |
| geguileo | correct, that's how Cinder is working | 17:35 |
| geguileo | because if you have deduplication you know you can set a higher over provisioning ratio | 17:35 |
| bswartz | so given the choice between 2 backends, each with the same amount of volumes and total size, cinder won't prefer the one that has better deduplication | 17:36 |
| bswartz | deduplication is an unpredictable phenomenon -- it's hard to know how it will perform without trying it, and its performance can change over time -- simply writing to an existing full volume can cause your free space to go down without warning | 17:38 |
| jgriffith | oh lord, this whole thing again :( | 17:39 |
| geguileo | bswartz: that should be decided on the weighing | 17:39 |
| bswartz | and in the worst case you can hit the dreaded ENOSPC while deleting a file | 17:39 |
| jgriffith | bswartz yes, it's somewhat unpredictable; however as the driver continuously updates it's effeciency calculations it's fairly stable | 17:40 |
| jgriffith | if you have 1K volumes deployed in your cloud and a TB of data written and you're getting 4x it's safe to say that adding another volume isn't going to alter that ratio significantly | 17:40 |
| mriedem | this breaks cinder with castellan 0.13.0 https://github.com/openstack/castellan/commit/1a13c2b2030390e3c0a5d498da486d92ddd1152c#diff-db470a49644fe6fb19e2a281019745c8 | 17:40 |
| mriedem | fyi | 17:40 |
| mriedem | missing ABC method | 17:41 |
| jgriffith | bswartz and you won't hit ENOSPC when the driver has enough sense to stop provisioning after a safe threshold is reached | 17:41 |
| bswartz | jgriffith: but according to geguileo cinder doesn't pay attention to anything other than total space, free space, allocated space, and privisioned space, and the only one of those that reflects the effects of deduplication is free space | 17:42 |
| jgriffith | bswartz and it shouldn't | 17:42 |
| hemna | overprovisioning ratio as well | 17:42 |
| jgriffith | but drivers can and should be smart enough in what they report back | 17:42 |
| hemna | jgriffith, +1 | 17:43 |
| jgriffith | anyway | 17:43 |
| jgriffith | Nothings perfect | 17:43 |
| bswartz | jgriffith: when I say worst case I mean gremlins writing data to all of your volumes after you've overprovisioned | 17:43 |
| jgriffith | but shooting holes in every solution that's been proposed for the last 5 years isn't helping either :) | 17:43 |
| jgriffith | bswartz sure | 17:43 |
| jgriffith | bswartz there's certainly risky things like that | 17:43 |
| jgriffith | but that's sort of life with modern day storage IMO | 17:44 |
| jgriffith | for those that don't like that, we offer things like excluding efficiency calculations in reporting | 17:44 |
| jgriffith | I actually have customers that do exactly that | 17:44 |
| jgriffith | fine by me... they keep buying more nodes even though they're not using 1/2 of the ones they have :) | 17:44 |
| jgriffith | bswartz so if you want 1:1 then do 1:1 | 17:45 |
| jgriffith | problem solved | 17:45 |
| bswartz | yeah thick provisioning exists for a reason | 17:45 |
| stvnoyes | hi jgriffith, would you take a look at https://review.openstack.org/#/c/500102/ when you get a chance? Thanks. | 17:46 |
| bswartz | anyways I'm not trying to shoot holes the current solution I'm trying to make sure I understand which details the current solution ignores so we can provide the correct numbers | 17:46 |
| jgriffith | stvnoyes sure, I'll take a look right now | 17:46 |
| jgriffith | bswartz yeah... sorry; didn't mean it that way | 17:46 |
| jgriffith | bswartz I gave up on fully understanding these things a long time ago :) | 17:47 |
| jgriffith | stvnoyes oh cool! | 17:48 |
| jgriffith | stvnoyes so believe it or not that weirdness I had on L#120 was there for a reason | 17:49 |
| jgriffith | stvnoyes there was something missing in the object and I was far too lazy to figure it out or fix it so I hacked around it by sucking the data out of the volume | 17:49 |
| jgriffith | stvnoyes which then resulted in the bug you logged here | 17:49 |
| jgriffith | because the volume isn't updated yet | 17:50 |
| bswartz | geguileo: based on my understanding of your document, on NetApp pools provisioned capacity will always equal allocated capacity, because there is never a case when we attempt to share underlying storage devices across pools or backends | 17:51 |
| geguileo | bswartz: it does sound like in your case you can just remove the key from the stats then | 17:53 |
| bswartz | and when you do something like unmanage a volume the resulting thing becomes indistinguishable from all the other stuff lying around which we're explicitly not counting under either allocated or provisioned capacity | 17:53 |
| ildikov | smcginnis: hi | 17:53 |
| *** nicolasbock has quit IRC | 17:53 | |
| ildikov | smcginnis: quick question, what's the story with this patch now: https://review.openstack.org/#/c/489715/ ? | 17:54 |
| *** ddixit has joined #openstack-cinder | 17:55 | |
| ddixit | Tintri-CI has been failing to build devstack since this update: https://review.openstack.org/#/c/499984/3 was merged. Here are the logs for the failure: http://openstack-ci.tintri.com/tintri/refs-changes-84-499984-3/stack.sh.log.out. | 17:55 |
| ddixit | We tried upgrading/reinstalling pyasn1 and pyasn-modules. It still fails with ContextualVersionConflict: (pyasn1 0.3.3 (/usr/local/lib/python2.7/dist-packages), Requirement.parse(‘pyasn1==0.3.2’), set([‘pyasn1-modules’])). Can someone have a look and suggest ways to resolve this? Thanks! | 17:55 |
| geguileo | bswartz: yeah, we can consider those as corner cases | 17:55 |
| smcginnis | ddixit: Wrong channel? | 17:55 |
| smcginnis | ildikov: Not sure now. I think Jens was working on a better overall fix that would get rid of public_endpoint. | 17:56 |
| ildikov | smcginnis: hmm, ok | 17:56 |
| *** wanghao_ has joined #openstack-cinder | 17:57 | |
| geguileo | bswartz: and you can always have a not that says that if they use the pool for something other than cinder then the provisioning will not be as accurate | 17:57 |
| ildikov | smcginnis: I haven't checked, but I think we will still need a fix in Devstack to have the microversion detection working correctly with the client | 17:57 |
| ildikov | smcginnis: so I guess I'll need to ping Jens then to figure this out | 17:57 |
| *** lhx_ has quit IRC | 17:57 | |
| ddixit | smcginnis this is about Tintri Cinder CI failure | 17:58 |
| smcginnis | ildikov: That might be best. | 17:59 |
| *** wanghao has quit IRC | 17:59 | |
| bswartz | geguileo: it's unclear why the "provisioned capacity" number of useful to cinder if it doesn't account for miscellaneous space consumers and only tries to account for non-cinder-owned "volumes" on the pool in question | 17:59 |
| geguileo | bswartz: because that's how over provisioning was designed | 17:59 |
| ildikov | smcginnis: ok, will do, thanks | 18:00 |
| geguileo | or you could say that's the simplification Cinder went with | 18:00 |
| geguileo | you define max over provisioning ratio | 18:00 |
| geguileo | and the backends report the current provisioning value | 18:00 |
| bswartz | I think the key here is the calculations section that explains what cinder will actually do with these values | 18:00 |
| smcginnis | ildikov: Might be interesting to change your Depends-on to point to that other patch and see if it makes a difference. | 18:01 |
| bswartz | so that driver authors can understand whether it makes sense to report other "stuff" or not in the number based on how cinder will behave if they do or don't | 18:01 |
| geguileo | and then you use the max ratio multiplied by the current reported value to know if the backend is elligible | 18:01 |
| openstackgerrit | Merged openstack/cinder master: HPE Lefthand: fix volume manage operation https://review.openstack.org/499970 | 18:01 |
| ildikov | smcginnis: that's not a bad idea, I'll do that | 18:01 |
| *** itlinux_ has quit IRC | 18:01 | |
| geguileo | bswartz: https://review.openstack.org/#/c/490116/2/specs/queens/provisioning-improvements.rst@252 | 18:02 |
| *** harlowja has quit IRC | 18:02 | |
| bswartz | basically we're trying to represent a risk factor here by telling cinder how much unwritten data could get written in the future, but the actual knob we give to administrator is an overprovisioning ratio, which is a really clumsy tool to manage space with | 18:03 |
| openstackgerrit | Merged openstack/cinder master: Remove unsupported BlockDeviceDriver https://review.openstack.org/500831 | 18:03 |
| bswartz | managing space with an overprovisioning ratio is only possible with the benefit of historical data to make you feel safe about any given value | 18:04 |
| geguileo | bswartz: the only "tool" you have right now to do so is free space | 18:04 |
| geguileo | bswartz: if you want to report the real physical usage: https://review.openstack.org/#/c/490116/2/specs/queens/provisioning-improvements.rst@297 | 18:05 |
| openstackgerrit | Merged openstack/cinder master: VMware: Remove unused variables https://review.openstack.org/484085 | 18:05 |
| bswartz | geguileo: how is "total used" different from "total capacity" (minus) "free capacity" | 18:06 |
| geguileo | bswartz: lol, it's not (I really need to sleep more) | 18:07 |
| geguileo | bswartz: is there anything you think should be reported that would help admins? | 18:08 |
| *** Aditya has quit IRC | 18:08 | |
| bswartz | geguileo: if there was a way to predict the future, that would be helpful | 18:12 |
| geguileo | rofl | 18:13 |
| *** xyang has joined #openstack-cinder | 18:13 | |
| bswartz | for example, if we know that, on average, each volume will get filled up 25% (based on observations about the past) then it would be reasonable to assume that newly created volumes will probably also used up 25% of their apparent size | 18:13 |
| bswartz | unfortunately we can only guess about such things | 18:14 |
| geguileo | we could implement some kind of automatic over provisioning ration | 18:15 |
| geguileo | s/ration/ratio | 18:15 |
| geguileo | so max over provisioning ratio is set to "auto" | 18:15 |
| bswartz | the biggest problem I see with scheduling is how we handle large bursts of provisioning followed by large bursts of writing, rather than a simpler, provision/write/provision/write cycle | 18:15 |
| bswartz | when you create a volume (not from an image or snapshot) it initially consumes zero bytes if its thin provisioned | 18:16 |
| geguileo | I think I know what you mean... | 18:16 |
| bswartz | and oversubscription is insane unless you have good reason to believe the request won't actually use the whole thing | 18:17 |
| bswartz | so if you can only come up with a good guess about the future usage of that volume, then you can make a better decision about where to put it | 18:17 |
| bswartz | s/request/requester/ | 18:17 |
| geguileo | I see, that's something different from automatic over provisioning ratio | 18:18 |
| geguileo | or it could be that, but at the scheduler level | 18:18 |
| bswartz | perhaps | 18:18 |
| geguileo | I'll give it a thought and see if I can come up with something that helps | 18:18 |
| bswartz | if I was a storage admin, the only thing that could help me sleep at night would be the ability to automatically migrate volumes to deal with low-space conditions | 18:19 |
| xyang | eharney: I updated the bug report: https://bugs.launchpad.net/cinder/+bug/1711455 | 18:20 |
| openstack | Launchpad bug 1711455 in Cinder "GroupsTest.test_group_update failed with MismatchError: 2 != 1" [High,New] | 18:20 |
| bswartz | and if I had that, then I could set my overprovisioning ratios to infinity and not worry about it | 18:20 |
| xyang | eharney: we need to backport https://review.openstack.org/#/c/484246/ to stable/ocata and stable/newton | 18:20 |
| xyang | eharney: but they are now closed | 18:20 |
| eharney | xyang: great, thanks for analyzing that | 18:20 |
| xyang | eharney: alternatively I could modify the tempest test to accommodate this problem | 18:21 |
| geguileo | bswartz: that's an interesting idea indeed... | 18:21 |
| eharney | xyang: landing that fix in stable/ocata seems doable? | 18:22 |
| openstackgerrit | Chhavi Agarwal proposed openstack/os-brick master: FC PPC64 device discovery issue https://review.openstack.org/501253 | 18:22 |
| geguileo | bswartz: migrating volumes effectively would be tricky if you have deduplication, as you could be moving volumes with barely any gain... | 18:24 |
| xyang | eharney: it's doable. I thought it's closed? | 18:24 |
| xyang | eharney: but this is a critical issue so we can still backport? | 18:24 |
| geguileo | bswartz: do your backends have a way of knowing how much data will be freed if a volume is removed? | 18:25 |
| bswartz | geguileo: yeah the hardest problem I've ever encountered in storage is this question: If I delete this thing, how much space will I get back? | 18:25 |
| geguileo | bswartz: I know how to get that info on the RBD backend, that's why I was asking if it was possible in yours | 18:26 |
| bswartz | geguileo: yes it's possible to do, but it requires time+cpu linear to the volume size | 18:26 |
| eharney | xyang: stable/ocata should be in "critical bugfixes" mode now -- this seems like it would still qualify since it's a gate-failure | 18:26 |
| geguileo | bswartz: an example of size & time? | 18:26 |
| bswartz | geguileo: I just mean it's not a constant time operation | 18:27 |
| xyang | eharney: alright. I'll submit a backport then | 18:27 |
| xyang | eharney: what about newton? they say it failed in newton as well | 18:27 |
| bswartz | you can to scan the block tree for all the blocks that would be freed by deletion of the object, which takes time proportional to the object | 18:27 |
| geguileo | bswartz: yeah, I understood, but even if it's linear if we are talking 10ms for 1GB it may not a problem, that's why I was asking about an example ;-) | 18:28 |
| bswartz | proportional to the size of the object | 18:28 |
| bswartz | oh, I'm not sure but I think it's worse than that | 18:28 |
| eharney | xyang: https://releases.openstack.org/ says EOL in October, not sure what the best plan is there | 18:28 |
| bswartz | more importantly, other I/O would be impacted by such queries so it would be unwise to his the controller with those kinds of queries all the time | 18:29 |
| bswartz | s/his/hit/ | 18:29 |
| geguileo | bswartz: I think automatic over provisioning would help alleviate that problem, but I'll try to give that idea some thought as well... | 18:29 |
| xyang | eharney: ok, I'll backport to stable/ocata first. maybe we'll have to skip the test in stable/newton if we can't backport to there | 18:30 |
| *** tesseract has quit IRC | 18:31 | |
| *** xyang has quit IRC | 18:32 | |
| openstackgerrit | Chhavi Agarwal proposed openstack/os-brick master: FC PPC64 device discovery issue https://review.openstack.org/501253 | 18:42 |
| *** chhavi has joined #openstack-cinder | 18:43 | |
| chhavi | jungleboyj, eharney: review request https://review.openstack.org/#/c/501253/ | 18:44 |
| *** adrianofr has quit IRC | 18:50 | |
| *** crose has quit IRC | 18:53 | |
| *** adrianofr has joined #openstack-cinder | 18:54 | |
| xyang1 | eharney: https://review.openstack.org/#/c/501377/ | 18:55 |
| *** xyang has joined #openstack-cinder | 18:57 | |
| *** catinthe_ has joined #openstack-cinder | 19:04 | |
| *** catintheroof has quit IRC | 19:06 | |
| *** catintheroof has joined #openstack-cinder | 19:06 | |
| *** e0ne has joined #openstack-cinder | 19:08 | |
| *** catinthe_ has quit IRC | 19:10 | |
| openstackgerrit | Matt Riedemann proposed openstack/cinder master: Fix test_rpc_consumer_isolation for oslo.messaging 5.31.0 https://review.openstack.org/501385 | 19:14 |
| mriedem | smcginnis: ^ fixes that one test | 19:18 |
| smcginnis | mriedem: Thanks | 19:19 |
| *** chhavi has quit IRC | 19:20 | |
| mriedem | nova has the same test :) | 19:20 |
| mriedem | surprise surprise | 19:20 |
| smcginnis | Gee, how'd that happen. | 19:20 |
| *** ddixit has quit IRC | 19:22 | |
| *** abhi89 has quit IRC | 19:23 | |
| *** itlinux has joined #openstack-cinder | 19:29 | |
| *** crushil has quit IRC | 19:31 | |
| *** itlinux has quit IRC | 19:33 | |
| *** pcaruana has quit IRC | 19:34 | |
| *** ntpttr_laptop has quit IRC | 19:41 | |
| *** lbragstad has quit IRC | 19:43 | |
| *** lbragstad has joined #openstack-cinder | 19:45 | |
| *** ddixit has joined #openstack-cinder | 19:45 | |
| *** xyang has quit IRC | 19:47 | |
| openstackgerrit | Sean McGinnis proposed openstack/cinder master: Implement keymgr list() method https://review.openstack.org/501398 | 19:47 |
| *** ddixit has quit IRC | 19:50 | |
| *** xyang has joined #openstack-cinder | 19:50 | |
| *** ntpttr_laptop has joined #openstack-cinder | 19:53 | |
| *** Apoorva has quit IRC | 20:00 | |
| *** eharney has quit IRC | 20:04 | |
| *** felipemonteiro_ has joined #openstack-cinder | 20:09 | |
| *** crushil has joined #openstack-cinder | 20:09 | |
| *** kaisers_ has quit IRC | 20:09 | |
| *** prometheanfire has joined #openstack-cinder | 20:11 | |
| prometheanfire | looks like the new oslo-messaging may break cinder http://logs.openstack.org/57/501157/1/check/gate-cross-cinder-python27-ubuntu-xenial/cb5afa5/ | 20:11 |
| *** dave-mccowan has quit IRC | 20:11 | |
| *** liverpooler has quit IRC | 20:11 | |
| smcginnis | prometheanfire: I think mriedem just pushed up a patch to fix that. | 20:11 |
| prometheanfire | ah, cool beans | 20:12 |
| smcginnis | But like it says, that should never get called. :D | 20:12 |
| prometheanfire | :D | 20:12 |
| *** felipemonteiro__ has quit IRC | 20:12 | |
| smcginnis | prometheanfire: If you're interested: https://review.openstack.org/501385 | 20:13 |
| smcginnis | jungleboyj, xyang1, e0ne, geguileo: Might be good to get that through. ^^ | 20:13 |
| *** Apoorva has joined #openstack-cinder | 20:13 | |
| smcginnis | I think that and https://review.openstack.org/501398 are holding up a requirements update. | 20:14 |
| smcginnis | prometheanfire: Think it's worth the time to put Depends-on tags on the u-c patch? | 20:14 |
| prometheanfire | you know where the review for https://bugs.launchpad.net/openstack-requirements/+bug/1715451 is (if it's in progress)? | 20:15 |
| openstack | Launchpad bug 1715451 in Cinder "Castellan 0.13.0 doesn't work with ConfKeyManager due to missing list() abstract method" [Low,In progress] - Assigned to Sean McGinnis (sean-mcginnis) | 20:15 |
| prometheanfire | ya, probably | 20:15 |
| smcginnis | prometheanfire: Yep, that's https://review.openstack.org/501398 | 20:16 |
| prometheanfire | added deps-on to the oslo-messaging update | 20:17 |
| prometheanfire | that fixes castellan as well? | 20:17 |
| prometheanfire | oh, wrong tab | 20:17 |
| smcginnis | Castellan: https://review.openstack.org/501398, oslo-messaging: https://review.openstack.org/501385 | 20:18 |
| prometheanfire | yep | 20:18 |
| *** ddixit has joined #openstack-cinder | 20:20 | |
| *** ddixit has quit IRC | 20:21 | |
| prometheanfire | ok, deps on for ALLTHETHINGS | 20:21 |
| *** jmlowe_ has quit IRC | 20:33 | |
| *** jmlowe has joined #openstack-cinder | 20:34 | |
| *** Apoorva_ has joined #openstack-cinder | 20:36 | |
| *** e0ne has quit IRC | 20:38 | |
| *** dave-mccowan has joined #openstack-cinder | 20:40 | |
| *** Apoorva has quit IRC | 20:40 | |
| *** ddixit has joined #openstack-cinder | 20:41 | |
| *** lucasxu has quit IRC | 20:42 | |
| *** ddixit has quit IRC | 20:43 | |
| *** ddixit has joined #openstack-cinder | 20:44 | |
| *** cknight has quit IRC | 20:51 | |
| *** lpetrut has quit IRC | 20:52 | |
| *** eharney has joined #openstack-cinder | 21:02 | |
| *** mriedem has quit IRC | 21:02 | |
| *** dave-mcc_ has joined #openstack-cinder | 21:02 | |
| *** mriedem has joined #openstack-cinder | 21:03 | |
| *** dave-mccowan has quit IRC | 21:05 | |
| *** adrianofr has quit IRC | 21:10 | |
| *** itlinux has joined #openstack-cinder | 21:12 | |
| *** Apoorva_ has quit IRC | 21:14 | |
| *** Yogi1 has quit IRC | 21:14 | |
| *** Apoorva has joined #openstack-cinder | 21:14 | |
| *** gouthamr has quit IRC | 21:15 | |
| ildikov | smcginnis: FYI, the Nova change works well with Jens's patch | 21:19 |
| smcginnis | ildikov: Excellent! | 21:19 |
| mriedem | smcginnis: i think i know about the iso8601 thing | 21:21 |
| mriedem | with 0.1.12 | 21:21 |
| mriedem | someone recently fixed a problem with that in nova | 21:21 |
| smcginnis | mriedem: Just got confused by that. | 21:21 |
| mriedem | https://github.com/openstack/nova/commit/3d86ae664a784e3f261943bd014dc88c82ea8b5d | 21:22 |
| smcginnis | mriedem: Tested in one venv and I see the new behavior. Ran tests on my change and it failed. | 21:22 |
| mriedem | copy ^ | 21:22 |
| *** yangyapeng has joined #openstack-cinder | 21:22 | |
| smcginnis | mriedem: What's the trick? http://paste.openstack.org/show/620564/ | 21:23 |
| mriedem | https://review.openstack.org/#/c/498287/ | 21:23 |
| smcginnis | Oooh! Utc>UTC | 21:24 |
| mriedem | smcginnis: it's different on py3 | 21:24 |
| smcginnis | Some library maintainer needs a swift kick to somewhere soft. | 21:24 |
| smcginnis | OK, that difference works on all my venvs. | 21:25 |
| *** yangyapeng has quit IRC | 21:27 | |
| *** crushil has quit IRC | 21:29 | |
| *** crushil has joined #openstack-cinder | 21:30 | |
| *** Apoorva_ has joined #openstack-cinder | 21:33 | |
| *** Apoorva has quit IRC | 21:37 | |
| *** ddixit has quit IRC | 21:41 | |
| *** ddixit has joined #openstack-cinder | 21:42 | |
| openstackgerrit | Sean McGinnis proposed openstack/cinder master: Implement keymgr list() method https://review.openstack.org/501398 | 21:45 |
| *** harlowja has joined #openstack-cinder | 21:45 | |
| *** xyang has quit IRC | 21:49 | |
| openstackgerrit | Sean McGinnis proposed openstack/cinder master: Use newer location for iso8601 UTC https://review.openstack.org/501472 | 21:49 |
| *** xyang has joined #openstack-cinder | 21:50 | |
| *** xyang has quit IRC | 21:51 | |
| *** edmondsw has quit IRC | 21:52 | |
| openstackgerrit | Chuck Fouts proposed openstack/cinder master: NetApp: Remove redundant QoS cleanup task https://review.openstack.org/501478 | 21:56 |
| *** catintheroof has quit IRC | 21:56 | |
| *** dave-mcc_ has quit IRC | 21:57 | |
| *** ddixit has quit IRC | 21:59 | |
| openstackgerrit | Sean McGinnis proposed openstack/cinder master: Remove Blockbridge volume driver https://review.openstack.org/501328 | 22:07 |
| openstackgerrit | Sean McGinnis proposed openstack/cinder master: Remove FalconStor volume drivers https://review.openstack.org/501329 | 22:07 |
| openstackgerrit | Sean McGinnis proposed openstack/cinder master: Remove Coho volume driver https://review.openstack.org/501327 | 22:07 |
| *** xyang1 has quit IRC | 22:09 | |
| openstackgerrit | Merged openstack/cinder master: Fix test_rpc_consumer_isolation for oslo.messaging 5.31.0 https://review.openstack.org/501385 | 22:10 |
| *** erlon has quit IRC | 22:11 | |
| *** mchlumsky has quit IRC | 22:15 | |
| *** ddixit has joined #openstack-cinder | 22:19 | |
| *** edmondsw has joined #openstack-cinder | 22:23 | |
| *** ddixit has quit IRC | 22:24 | |
| *** edmondsw has quit IRC | 22:28 | |
| *** JoseMello has quit IRC | 22:35 | |
| *** dustins has quit IRC | 22:39 | |
| *** Apoorva_ has quit IRC | 22:40 | |
| *** Apoorva has joined #openstack-cinder | 22:41 | |
| *** gouthamr has joined #openstack-cinder | 22:53 | |
| *** thegreenhundred has quit IRC | 22:55 | |
| *** scottda_ has quit IRC | 23:00 | |
| *** felipemonteiro_ has quit IRC | 23:01 | |
| *** Apoorva_ has joined #openstack-cinder | 23:01 | |
| *** Apoorva has quit IRC | 23:05 | |
| *** felipemonteiro has joined #openstack-cinder | 23:09 | |
| *** itlinux has quit IRC | 23:16 | |
| *** ddixit has joined #openstack-cinder | 23:22 | |
| *** ddixit has quit IRC | 23:24 | |
| *** alkhodos has quit IRC | 23:35 | |
| *** ntpttr_laptop has quit IRC | 23:36 | |
| *** yangyapeng has joined #openstack-cinder | 23:38 | |
| *** yangyapeng has quit IRC | 23:43 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!