| *** thumpba_ has joined #openstack-cinder | 00:03 | |
| patrickeast | wtf is all this neutron crap breaking my devstack :/ | 00:03 |
|---|---|---|
| *** julim has quit IRC | 00:03 | |
| *** thumpba has quit IRC | 00:05 | |
| *** yangyapeng has quit IRC | 00:06 | |
| *** Julien-zte has joined #openstack-cinder | 00:08 | |
| *** leeantho has quit IRC | 00:09 | |
| bswartz | patrickeast: if neutron didn't break devstack then who else would? | 00:14 |
| patrickeast | hah | 00:14 |
| patrickeast | bswartz: me, if it would get far enough :p | 00:14 |
| *** bswartz has quit IRC | 00:17 | |
| *** raunak has quit IRC | 00:18 | |
| *** mriedem has joined #openstack-cinder | 00:18 | |
| *** laughterwym has joined #openstack-cinder | 00:18 | |
| *** laughter_ has joined #openstack-cinder | 00:22 | |
| *** laughte__ has joined #openstack-cinder | 00:22 | |
| *** Julien-zte has quit IRC | 00:22 | |
| *** laughterwym has quit IRC | 00:25 | |
| *** bwallis has quit IRC | 00:25 | |
| *** laughter_ has quit IRC | 00:26 | |
| *** bswartz has joined #openstack-cinder | 00:32 | |
| *** sdake has joined #openstack-cinder | 00:35 | |
| *** Suyi has quit IRC | 00:44 | |
| *** lixiaoy1 has joined #openstack-cinder | 00:47 | |
| *** jamielennox is now known as jamielennox|away | 00:48 | |
| *** gus has joined #openstack-cinder | 00:48 | |
| *** raunak has joined #openstack-cinder | 00:53 | |
| *** cheneydc has joined #openstack-cinder | 00:55 | |
| *** cheneydc has quit IRC | 00:56 | |
| *** xyang has joined #openstack-cinder | 00:58 | |
| *** sdake has quit IRC | 00:59 | |
| openstackgerrit | zhangyanxian proposed openstack/cinder: Fix typo in the huawei_driver.py https://review.openstack.org/340276 | 00:59 |
| *** xyang has quit IRC | 00:59 | |
| *** bwallis has joined #openstack-cinder | 01:00 | |
| *** bardia has quit IRC | 01:01 | |
| *** bardia has joined #openstack-cinder | 01:02 | |
| *** sdake has joined #openstack-cinder | 01:02 | |
| *** bardia has quit IRC | 01:03 | |
| *** jamielennox|away is now known as jamielennox | 01:04 | |
| *** cdelatte has joined #openstack-cinder | 01:06 | |
| *** sdake has quit IRC | 01:11 | |
| *** chrichip has joined #openstack-cinder | 01:19 | |
| *** yangyapeng has joined #openstack-cinder | 01:21 | |
| *** ducttape_ has joined #openstack-cinder | 01:23 | |
| *** ducttape_ has quit IRC | 01:23 | |
| *** ducttape_ has joined #openstack-cinder | 01:23 | |
| *** gouthamr_ has joined #openstack-cinder | 01:28 | |
| *** campee has joined #openstack-cinder | 01:29 | |
| *** EinstCrazy has joined #openstack-cinder | 01:32 | |
| *** gouthamr has quit IRC | 01:32 | |
| *** EinstCrazy has quit IRC | 01:32 | |
| *** EinstCra_ has joined #openstack-cinder | 01:33 | |
| *** bardia has joined #openstack-cinder | 01:37 | |
| *** Lee1092 has joined #openstack-cinder | 01:37 | |
| *** alonma has joined #openstack-cinder | 01:38 | |
| *** edmondsw has quit IRC | 01:40 | |
| campee | does anyone know of a good document that describes what the architecture of a SAN used with Cinder should look like? | 01:41 |
| campee | i'm a little confused as to whether or not each compute host would need to access the storage through a cinder host or whether they can just talk to it directly. i assume they can talk to it directly, but i'm not sure what that would look like | 01:42 |
| *** alonma has quit IRC | 01:43 | |
| *** Krenair has left #openstack-cinder | 01:43 | |
| *** takedakn has joined #openstack-cinder | 01:44 | |
| *** Asaithambi has joined #openstack-cinder | 01:44 | |
| *** alonma has joined #openstack-cinder | 01:46 | |
| *** delattec has joined #openstack-cinder | 01:47 | |
| openstackgerrit | Sean McGinnis proposed openstack/python-cinderclient: Update the home-page with developer documentation https://review.openstack.org/317232 | 01:47 |
| *** raunak has quit IRC | 01:48 | |
| *** bardia has quit IRC | 01:49 | |
| *** cdelatte has quit IRC | 01:49 | |
| *** haplo37__ has joined #openstack-cinder | 01:50 | |
| *** alonma has quit IRC | 01:51 | |
| *** takedakn has quit IRC | 01:51 | |
| *** thumpba_ has quit IRC | 01:52 | |
| openstackgerrit | xianming.mao proposed openstack/python-cinderclient: Update the homepage with developer documentation page https://review.openstack.org/354698 | 01:53 |
| *** alonma has joined #openstack-cinder | 01:53 | |
| openstackgerrit | xianming.mao proposed openstack/cinder-specs: Replace the homepage with developer documentation page https://review.openstack.org/354675 | 01:56 |
| *** cknight has joined #openstack-cinder | 01:57 | |
| *** Apoorva has quit IRC | 01:57 | |
| *** alonma has quit IRC | 01:58 | |
| *** bill_az has quit IRC | 01:58 | |
| *** alonma has joined #openstack-cinder | 02:00 | |
| *** tbarron is now known as tbarron|afk | 02:00 | |
| *** EinstCrazy has joined #openstack-cinder | 02:03 | |
| *** ducttape_ has quit IRC | 02:04 | |
| *** alonma has quit IRC | 02:05 | |
| *** EinstCra_ has quit IRC | 02:07 | |
| *** alonma has joined #openstack-cinder | 02:08 | |
| *** delattec has quit IRC | 02:09 | |
| *** neophy has joined #openstack-cinder | 02:13 | |
| *** chrichip has quit IRC | 02:13 | |
| *** alonma has quit IRC | 02:13 | |
| *** delattec has joined #openstack-cinder | 02:14 | |
| *** chrichip has joined #openstack-cinder | 02:15 | |
| *** diogogmt has joined #openstack-cinder | 02:17 | |
| *** mriedem has quit IRC | 02:18 | |
| *** diogogmt has quit IRC | 02:18 | |
| *** bardia has joined #openstack-cinder | 02:20 | |
| *** laughte__ has quit IRC | 02:20 | |
| *** laughterwym has joined #openstack-cinder | 02:20 | |
| *** haplo37__ has quit IRC | 02:20 | |
| *** ebalduf has quit IRC | 02:21 | |
| *** thumpba has joined #openstack-cinder | 02:23 | |
| *** ducttape_ has joined #openstack-cinder | 02:23 | |
| *** EinstCra_ has joined #openstack-cinder | 02:27 | |
| *** yangyape_ has joined #openstack-cinder | 02:27 | |
| *** Asaithambi has quit IRC | 02:28 | |
| *** yangyapeng has quit IRC | 02:28 | |
| *** EinstCrazy has quit IRC | 02:29 | |
| *** Asaithambi has joined #openstack-cinder | 02:38 | |
| *** zhangguoqing has joined #openstack-cinder | 02:41 | |
| *** jamielennox is now known as jamielennox|away | 02:42 | |
| *** coolsvap has joined #openstack-cinder | 02:45 | |
| *** chrichip has quit IRC | 02:45 | |
| *** Asaithambi has quit IRC | 02:46 | |
| *** takedakn has joined #openstack-cinder | 02:46 | |
| *** chrichip has joined #openstack-cinder | 02:47 | |
| *** ducttape_ has quit IRC | 02:48 | |
| openstackgerrit | Dao Cong Tien proposed openstack/cinder: Stop using mox in unit/api/v3/ tests https://review.openstack.org/356841 | 02:50 |
| *** ducttape_ has joined #openstack-cinder | 02:50 | |
| *** ducttape_ has quit IRC | 02:58 | |
| *** julim has joined #openstack-cinder | 03:05 | |
| *** delattec has quit IRC | 03:06 | |
| *** jamielennox|away is now known as jamielennox | 03:07 | |
| *** thumpba has quit IRC | 03:15 | |
| *** raunak has joined #openstack-cinder | 03:16 | |
| *** julim has quit IRC | 03:19 | |
| *** coolsvap is now known as _coolsvap_ | 03:21 | |
| *** bardia has quit IRC | 03:24 | |
| *** raunak has quit IRC | 03:25 | |
| *** raunak has joined #openstack-cinder | 03:30 | |
| openstackgerrit | Soffie Huang proposed openstack/cinder: Add a missing option for iSCSI multipath in FalconStor FSS driver https://review.openstack.org/355836 | 03:33 |
| *** gouthamr_ has quit IRC | 03:34 | |
| openstackgerrit | Merged openstack/cinder: Imported Translations from Zanata https://review.openstack.org/356248 | 03:34 |
| openstackgerrit | Merged openstack/cinder: Revise Synology DSM storage driver https://review.openstack.org/341345 | 03:35 |
| *** ArsenC has quit IRC | 03:36 | |
| openstackgerrit | LisaLi proposed openstack/cinder: Fix the interface to call update_server_volume https://review.openstack.org/356850 | 03:40 |
| *** yangyapeng has joined #openstack-cinder | 03:40 | |
| *** raunak has quit IRC | 03:41 | |
| *** yangyape_ has quit IRC | 03:44 | |
| *** links has joined #openstack-cinder | 03:45 | |
| *** raunak has joined #openstack-cinder | 03:45 | |
| *** takedakn has quit IRC | 03:46 | |
| *** pbandark has joined #openstack-cinder | 03:51 | |
| *** shausy has joined #openstack-cinder | 03:56 | |
| *** Asaithambi has joined #openstack-cinder | 03:57 | |
| *** ducttape_ has joined #openstack-cinder | 03:58 | |
| *** pbandark is now known as pbandark`brb | 04:02 | |
| *** bill_az has joined #openstack-cinder | 04:03 | |
| *** ducttape_ has quit IRC | 04:03 | |
| *** neophy has quit IRC | 04:19 | |
| *** neophy has joined #openstack-cinder | 04:26 | |
| *** bardia has joined #openstack-cinder | 04:28 | |
| openstackgerrit | Dao Cong Tien proposed openstack/cinder: Stop using mox in unit/api/contrib/ tests (1) https://review.openstack.org/356856 | 04:28 |
| *** lixiaoy1 has quit IRC | 04:33 | |
| *** lixiaoy1 has joined #openstack-cinder | 04:36 | |
| openstackgerrit | Dao Cong Tien proposed openstack/cinder: Stop using mox in unit/api/contrib/ tests (end) https://review.openstack.org/356859 | 04:41 |
| openstackgerrit | Dao Cong Tien proposed openstack/cinder: Stop using mox in unit/api/contrib/ tests (1) https://review.openstack.org/356856 | 04:41 |
| openstackgerrit | Dao Cong Tien proposed openstack/cinder: Stop using mox in unit/api/v2/ tests https://review.openstack.org/356347 | 04:43 |
| *** hemna__ has joined #openstack-cinder | 04:53 | |
| *** chrichip has quit IRC | 04:53 | |
| *** chrichip has joined #openstack-cinder | 04:53 | |
| *** hemna__ has quit IRC | 04:54 | |
| *** hemna__ has joined #openstack-cinder | 04:54 | |
| openstackgerrit | Merged openstack/cinder: Enhance api_microversion doc https://review.openstack.org/331447 | 04:56 |
| *** narayrak has joined #openstack-cinder | 04:59 | |
| *** ducttape_ has joined #openstack-cinder | 05:00 | |
| *** narayrak has quit IRC | 05:02 | |
| *** ducttape_ has quit IRC | 05:04 | |
| *** laughter_ has joined #openstack-cinder | 05:07 | |
| *** cknight has quit IRC | 05:08 | |
| *** laughterwym has quit IRC | 05:10 | |
| *** raunak has quit IRC | 05:12 | |
| *** raunak has joined #openstack-cinder | 05:16 | |
| *** raunak has quit IRC | 05:22 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/cinder: Updated from global requirements https://review.openstack.org/356867 | 05:25 |
| *** rajinir has quit IRC | 05:35 | |
| *** alonma has joined #openstack-cinder | 05:42 | |
| *** laughter_ has quit IRC | 05:44 | |
| *** laughterwym has joined #openstack-cinder | 05:45 | |
| openstackgerrit | wanghao proposed openstack/cinder: Backup tenant attribute support https://review.openstack.org/258374 | 05:45 |
| *** alonma has quit IRC | 05:47 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/cinder: Updated from global requirements https://review.openstack.org/356867 | 05:49 |
| *** alonma has joined #openstack-cinder | 05:50 | |
| openstackgerrit | Karthik Prabhu Vinod proposed openstack/cinder: WIP, DNM: multinode partial grenade testing https://review.openstack.org/353014 | 05:52 |
| openstackgerrit | wanghao proposed openstack/cinder: Support to query volume filter by change since/before https://review.openstack.org/352688 | 05:53 |
| *** lixiaoy1 has quit IRC | 05:53 | |
| *** lixiaoy1 has joined #openstack-cinder | 05:54 | |
| *** alonma has quit IRC | 05:55 | |
| *** alonma has joined #openstack-cinder | 05:58 | |
| *** bardia has quit IRC | 05:59 | |
| *** lkuchlan has joined #openstack-cinder | 05:59 | |
| *** bardia has joined #openstack-cinder | 05:59 | |
| *** ducttape_ has joined #openstack-cinder | 06:01 | |
| *** alonma has quit IRC | 06:02 | |
| *** hemna__ has quit IRC | 06:02 | |
| *** lkuchlan_ has joined #openstack-cinder | 06:04 | |
| *** ducttape_ has quit IRC | 06:05 | |
| openstackgerrit | wanghao proposed openstack/cinder-specs: Transfer snapshots with volumes https://review.openstack.org/319831 | 06:07 |
| *** lkuchlan has quit IRC | 06:07 | |
| *** lkuchlan_ has quit IRC | 06:08 | |
| *** neophy has quit IRC | 06:08 | |
| *** lkuchlan has joined #openstack-cinder | 06:08 | |
| *** lkuchlan is now known as lkuchlan_wfh | 06:08 | |
| *** bardia has quit IRC | 06:12 | |
| *** rcernin has joined #openstack-cinder | 06:12 | |
| *** alonma has joined #openstack-cinder | 06:14 | |
| *** alonma has quit IRC | 06:18 | |
| *** lpetrut has joined #openstack-cinder | 06:18 | |
| openstackgerrit | Soffie Huang proposed openstack/cinder: Refactoring FalconStor FSS storage driver https://review.openstack.org/355836 | 06:19 |
| *** alonma has joined #openstack-cinder | 06:21 | |
| *** alonma has quit IRC | 06:26 | |
| *** alonma has joined #openstack-cinder | 06:28 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/cinder: Imported Translations from Zanata https://review.openstack.org/356956 | 06:30 |
| *** arecknag has joined #openstack-cinder | 06:30 | |
| *** alonma has quit IRC | 06:33 | |
| *** alonma has joined #openstack-cinder | 06:44 | |
| *** takedakn has joined #openstack-cinder | 06:44 | |
| *** alonma has quit IRC | 06:49 | |
| *** takedakn has quit IRC | 06:49 | |
| *** bill_az has quit IRC | 06:49 | |
| openstackgerrit | Merged openstack/cinder: Specify key algorithm and size for create_key https://review.openstack.org/251503 | 06:49 |
| *** pcaruana has joined #openstack-cinder | 06:52 | |
| openstackgerrit | Guy Kim proposed openstack/cinder: ITRI DISCO cinder volume driver https://review.openstack.org/347100 | 06:54 |
| *** gaurangt has joined #openstack-cinder | 06:55 | |
| *** takedakn has joined #openstack-cinder | 06:56 | |
| *** ducttape_ has joined #openstack-cinder | 07:02 | |
| openstackgerrit | Guy Kim proposed openstack/cinder: ITRI DISCO cinder backup driver https://review.openstack.org/349318 | 07:06 |
| *** ducttape_ has quit IRC | 07:07 | |
| *** xek__ has quit IRC | 07:08 | |
| *** lpetrut has quit IRC | 07:11 | |
| openstackgerrit | Ellen Leahy proposed openstack/python-cinderclient: Changed backup-restore to accept backup name https://review.openstack.org/352778 | 07:14 |
| *** tesseract- has joined #openstack-cinder | 07:17 | |
| *** e0ne has joined #openstack-cinder | 07:18 | |
| *** EinstCra_ has quit IRC | 07:18 | |
| *** lkuchlan_wfh has quit IRC | 07:20 | |
| *** ManishD has joined #openstack-cinder | 07:22 | |
| *** EinstCrazy has joined #openstack-cinder | 07:34 | |
| openstackgerrit | Arnon Yaari proposed openstack/cinder: New cinder driver to support INFINIDAT InfiniBox https://review.openstack.org/356986 | 07:36 |
| *** nkrinner_afk has quit IRC | 07:39 | |
| openstackgerrit | Matan Sabag proposed openstack/cinder: Changes stats report for ScaleIO driver https://review.openstack.org/356120 | 07:41 |
| openstackgerrit | LisaLi proposed openstack/cinder: Attachment lost after migrating in-use volume https://review.openstack.org/356992 | 07:45 |
| openstackgerrit | LisaLi proposed openstack/cinder: Attachment lost after migrating in-use volume https://review.openstack.org/356992 | 07:46 |
| *** liverpooler has joined #openstack-cinder | 07:48 | |
| *** Asaithambi has quit IRC | 07:49 | |
| *** Asaithambi has joined #openstack-cinder | 07:49 | |
| *** nkrinner_afk has joined #openstack-cinder | 07:52 | |
| *** vladani has joined #openstack-cinder | 07:54 | |
| *** alonma has joined #openstack-cinder | 07:57 | |
| *** akapil has joined #openstack-cinder | 07:59 | |
| *** zzzeek has quit IRC | 08:00 | |
| *** zzzeek has joined #openstack-cinder | 08:01 | |
| *** Asaithambi has quit IRC | 08:07 | |
| openstackgerrit | xianming.mao proposed openstack/cinder-specs: Update homepage with developer documentation page https://review.openstack.org/357003 | 08:09 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/cinder: [WIP] Testing latest u-c https://review.openstack.org/303766 | 08:10 |
| *** lennyb has quit IRC | 08:13 | |
| lixiaoy1 | geguileo: dulek: ping | 08:14 |
| geguileo | lixiaoy1: pong | 08:14 |
| lixiaoy1 | lixiaoy1: I run into a problem, https://review.openstack.org/#/c/356992/ | 08:15 |
| lixiaoy1 | geguileo: https://review.openstack.org/#/c/356992/2 | 08:15 |
| lixiaoy1 | geguileo: this is about when to refresh objects etc. I have difficulities to create unit test for such changes | 08:16 |
| lixiaoy1 | geguileo: as we don't use real db data in current unit test, so whether refresh or not, can't be checked by the ut | 08:17 |
| geguileo | lixiaoy1: We can check the refresh | 08:17 |
| geguileo | lixiaoy1: You can mock the method and confirm that it's being called | 08:17 |
| lixiaoy1 | geguileo: but may sometime the refresh is not necessary | 08:17 |
| geguileo | lixiaoy1: Or if you have mocked the get_by_id method | 08:17 |
| geguileo | lixiaoy1: You can return different value for the N call | 08:18 |
| geguileo | lixiaoy1: But in unit tests you either need the refresh or you don't | 08:18 |
| geguileo | lixiaoy1: And you can test for both cases | 08:18 |
| geguileo | lixiaoy1: If it's only called sometimes | 08:18 |
| geguileo | lixiaoy1: Why do we need a refresh there? | 08:19 |
| lixiaoy1 | geguileo: which line | 08:19 |
| geguileo | lixiaoy1: https://review.openstack.org/#/c/356992/2/cinder/volume/manager.py@1830 | 08:20 |
| lixiaoy1 | geguileo: as line 1824 changes volume_attachment of the volume | 08:20 |
| *** EinstCra_ has joined #openstack-cinder | 08:21 | |
| geguileo | lixiaoy1: I would have to review the whole code path | 08:21 |
| geguileo | lixiaoy1: But I have the impression this is not the right approach | 08:21 |
| geguileo | lixiaoy1: You shouldn't need to refresh that volume there | 08:21 |
| geguileo | lixiaoy1: Or save it in the OVO class | 08:22 |
| lixiaoy1 | lixiaoy1: I did the test, if no refresh here, volume_attachment will be lost in the volume | 08:22 |
| *** dosaboy_ has joined #openstack-cinder | 08:23 | |
| *** dosaboy_ has quit IRC | 08:24 | |
| *** dosaboy has quit IRC | 08:24 | |
| geguileo | lixiaoy1: But that's because we are doing something wrong | 08:24 |
| geguileo | lixiaoy1: I'm not saying that the bug doesn't exist | 08:24 |
| lixiaoy1 | geguileo: when calling save() of an object, only the obj_get_changes() fields will be updated to db. Correct? | 08:24 |
| geguileo | lixiaoy1: Correct | 08:24 |
| geguileo | lixiaoy1: You are introducing 2 new DB queries, and one of them is not cheap (the load of a volume) | 08:25 |
| *** EinstCrazy has quit IRC | 08:25 | |
| geguileo | lixiaoy1: And I believe that's not necessary | 08:25 |
| lixiaoy1 | lixiaoy1: yes, I know. this refresh should not be called. I haven't figured out why it can fix the problem | 08:26 |
| geguileo | lixiaoy1: Do you want to keep trying to figure out or do you want me to see if I can fix it? | 08:26 |
| geguileo | And propose and alternative? | 08:27 |
| *** lennyb has joined #openstack-cinder | 08:28 | |
| lixiaoy1 | lixiaoy1: let me try at first. And if I can't, talk to you tomorrow? | 08:28 |
| lixiaoy1 | geguileo: thank you for your help | 08:28 |
| geguileo | lixiaoy1: Great! | 08:28 |
| *** dosaboy has joined #openstack-cinder | 08:29 | |
| *** laughterwym has quit IRC | 08:34 | |
| *** laughterwym has joined #openstack-cinder | 08:35 | |
| *** laughter_ has joined #openstack-cinder | 08:36 | |
| *** takedakn has quit IRC | 08:37 | |
| *** Asaithambi has joined #openstack-cinder | 08:38 | |
| *** laughterwym has quit IRC | 08:39 | |
| *** lixiaoy1 has quit IRC | 08:39 | |
| *** laughter_ has quit IRC | 08:40 | |
| *** lixiaoy1 has joined #openstack-cinder | 08:42 | |
| *** vbala has joined #openstack-cinder | 08:49 | |
| openstackgerrit | Ivan Kolodyazhny proposed openstack/cinder: Merge two fake drivers into the one FakeLoggingVolumeDriver https://review.openstack.org/348428 | 08:59 |
| *** dkehn_ has quit IRC | 09:00 | |
| *** ducttape_ has joined #openstack-cinder | 09:03 | |
| openstackgerrit | LisaLi proposed openstack/cinder: Attachment lost after migrating in-use volume https://review.openstack.org/356992 | 09:06 |
| *** _coolsvap_ is now known as coolsvap | 09:06 | |
| *** e0ne has quit IRC | 09:07 | |
| *** ducttape_ has quit IRC | 09:08 | |
| openstackgerrit | LisaLi proposed openstack/cinder: Attachment lost after migrating in-use volume https://review.openstack.org/356992 | 09:10 |
| *** dkehn_ has joined #openstack-cinder | 09:19 | |
| *** EinstCrazy has joined #openstack-cinder | 09:28 | |
| *** mvk has quit IRC | 09:29 | |
| *** lpetrut has joined #openstack-cinder | 09:29 | |
| *** EinstCra_ has quit IRC | 09:32 | |
| lixiaoy1 | geguileo: Gorka, thank you for your review. https://review.openstack.org/#/c/356992/4/cinder/objects/volume.py@466 , why not need save after some fields are changed https://review.openstack.org/#/c/356992/4/cinder/objects/volume.py@463 | 09:33 |
| *** alonma has quit IRC | 09:36 | |
| *** laughterwym has joined #openstack-cinder | 09:37 | |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Use the ClientHelper class to reduce amount of code https://review.openstack.org/347270 | 09:41 |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Avoid an extra resize later on https://review.openstack.org/347271 | 09:41 |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Do batch delivery https://review.openstack.org/347272 | 09:41 |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Report an error if a snapshot has vanished https://review.openstack.org/347273 | 09:41 |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Avoid duplicated volumes from the error path https://review.openstack.org/347274 | 09:41 |
| *** laughterwym has quit IRC | 09:42 | |
| *** alonma has joined #openstack-cinder | 09:44 | |
| *** jordanP has joined #openstack-cinder | 09:47 | |
| ManishD | @Lisa, does https://review.openstack.org/#/c/356992 is on similar lines on what Wlat is trying here https://review.openstack.org/#/c/312773/ | 09:48 |
| *** alonma has quit IRC | 09:49 | |
| ManishD | *Wlat=Walt | 09:49 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/cinder: Updated from global requirements https://review.openstack.org/356867 | 09:50 |
| lixiaoy1 | ManishD: let me check | 09:50 |
| geguileo | lixiaoy1: Because we are already saving it here: https://review.openstack.org/#/c/356992/4/cinder/volume/manager.py@1832 | 09:50 |
| lixiaoy1 | ManishD: they are different | 09:51 |
| ManishD | ok | 09:51 |
| *** alonma has joined #openstack-cinder | 09:51 | |
| lixiaoy1 | geguileo: I see. But I think it is not to safe. As it calls attach_volume, and which may not save volume and refresh. | 09:53 |
| geguileo | lixiaoy1: If you think something's not safe around there, that's for another patch ;-) | 09:54 |
| geguileo | lixiaoy1: In this one you are fixing the attachment info part | 09:54 |
| lixiaoy1 | ManishD: hemna's patch is to save connection_info during live migration, my patch is to fix attachment lost during migration in-use volume | 09:54 |
| lixiaoy1 | geguileo: ok | 09:54 |
| lixiaoy1 | geguileo: may I ask you a question about how to put several patches together without rebase? | 09:55 |
| *** alonma has quit IRC | 09:55 | |
| geguileo | lixiaoy1: Sure, but I don't know what do you mean without rebase | 09:56 |
| lixiaoy1 | geguileo: https://review.openstack.org/#/c/356226/ here is my tempest test | 09:56 |
| lixiaoy1 | geguileo: it needs three cinder patches | 09:56 |
| geguileo | Depends-On is for out of the project patches | 09:57 |
| lixiaoy1 | geguileo: but these three patches don't depend on each other, so I don't want to rebase them one by the other | 09:57 |
| geguileo | lixiaoy1: afaik there's no solution for that | 09:57 |
| *** akapil has quit IRC | 09:57 | |
| lixiaoy1 | geguileo: ok, thank you anyway | 09:57 |
| geguileo | Yeah, sorry about that :-( | 09:58 |
| *** alonma has joined #openstack-cinder | 09:58 | |
| lixiaoy1 | geguileo: :) | 09:58 |
| geguileo | It's not a very common problem, though I have sometimes wondered what I would do if it happened to me XD | 09:58 |
| *** mvk has joined #openstack-cinder | 09:59 | |
| lixiaoy1 | geguileo: yes it is. as from this tempest such problems are found. And I would like to use tempest test to prove the changes are correct | 10:00 |
| openstackgerrit | Clenimar Filemon proposed openstack/cinder: Read domain info from context when contacting nova https://review.openstack.org/337322 | 10:01 |
| geguileo | lixiaoy1: Oh, wait, your patch was for tempest, not for in-tree tempest tests | 10:01 |
| *** gcb_ has quit IRC | 10:02 | |
| geguileo | lixiaoy1: So the Depends-On should work, I think, but only if the patches can be merged without conflict I guess | 10:02 |
| * geguileo is full of certainty today | 10:02 | |
| *** alonma has quit IRC | 10:03 | |
| lixiaoy1 | geguileo: yes Depdens-On works. Only I adds three cinder patches in Depends-on, it gave error message. | 10:03 |
| geguileo | lixiaoy1: Probably because those patches are in conflict | 10:03 |
| *** ducttape_ has joined #openstack-cinder | 10:04 | |
| lixiaoy1 | lixiaoy1: two patches update same file, this may be the reason | 10:05 |
| *** mvk has quit IRC | 10:05 | |
| lixiaoy1 | geguileo: thank you. And I am going to leave office. Have a good day | 10:05 |
| *** sdague has joined #openstack-cinder | 10:06 | |
| geguileo | lixiaoy1: Thanks, you too | 10:06 |
| *** lixiaoy1 has quit IRC | 10:06 | |
| *** lixiaoy1 has joined #openstack-cinder | 10:07 | |
| *** ManishD has quit IRC | 10:08 | |
| *** akapil has joined #openstack-cinder | 10:08 | |
| *** ducttape_ has quit IRC | 10:09 | |
| *** laughterwym has joined #openstack-cinder | 10:11 | |
| *** alonma has joined #openstack-cinder | 10:14 | |
| *** alonma has quit IRC | 10:15 | |
| *** alonma has joined #openstack-cinder | 10:15 | |
| *** yangyapeng has quit IRC | 10:16 | |
| *** mvk has joined #openstack-cinder | 10:20 | |
| *** Asaithambi has quit IRC | 10:28 | |
| *** akapil has quit IRC | 10:29 | |
| *** Asaithambi has joined #openstack-cinder | 10:29 | |
| *** EinstCrazy has quit IRC | 10:32 | |
| *** ociuhandu has quit IRC | 10:33 | |
| *** laughterwym has quit IRC | 10:33 | |
| *** laughterwym has joined #openstack-cinder | 10:34 | |
| *** laughterwym has quit IRC | 10:36 | |
| *** laughterwym has joined #openstack-cinder | 10:37 | |
| *** Asaithambi has quit IRC | 10:37 | |
| *** laughterwym has quit IRC | 10:38 | |
| *** laughterwym has joined #openstack-cinder | 10:38 | |
| *** Asaithambi has joined #openstack-cinder | 10:38 | |
| *** laughterwym has quit IRC | 10:43 | |
| *** lpetrut1 has joined #openstack-cinder | 10:43 | |
| *** lpetrut has quit IRC | 10:45 | |
| *** lpetrut1 is now known as lpetrut | 10:45 | |
| *** akapil has joined #openstack-cinder | 10:48 | |
| *** coolsvap is now known as _coolsvap_ | 10:55 | |
| *** Asaithambi has quit IRC | 10:59 | |
| *** ducttape_ has joined #openstack-cinder | 11:05 | |
| *** bhagya has quit IRC | 11:08 | |
| *** gaurangt has left #openstack-cinder | 11:08 | |
| *** neophy has joined #openstack-cinder | 11:08 | |
| *** neophy has quit IRC | 11:09 | |
| *** JoseMello has joined #openstack-cinder | 11:09 | |
| *** ducttape_ has quit IRC | 11:09 | |
| *** Linkmark has joined #openstack-cinder | 11:15 | |
| *** GB21 has joined #openstack-cinder | 11:28 | |
| openstackgerrit | Satish Venkatasubramanian proposed openstack/cinder: Add volumes/summary API support https://review.openstack.org/335067 | 11:33 |
| *** takedakn has joined #openstack-cinder | 11:34 | |
| vladani | https://review.openstack.org/#/c/342722/ | 11:36 |
| *** yangyapeng has joined #openstack-cinder | 11:38 | |
| openstackgerrit | naga venkata proposed openstack/os-brick: Fixes backup with customized ceph cluster name https://review.openstack.org/351005 | 11:38 |
| *** takedakn has quit IRC | 11:39 | |
| *** ociuhandu has joined #openstack-cinder | 11:39 | |
| *** laughterwym has joined #openstack-cinder | 11:40 | |
| *** haplo37__ has joined #openstack-cinder | 11:42 | |
| swamireddy | vbala: Thanks for review...updated the patch with your comments...https://review.openstack.org/335067 | 11:45 |
| *** laughterwym has quit IRC | 11:53 | |
| *** lprice1 has quit IRC | 12:03 | |
| openstackgerrit | xianming.mao proposed openstack/cinder: Drop unused logging import https://review.openstack.org/357150 | 12:03 |
| *** lprice has joined #openstack-cinder | 12:04 | |
| *** kwathore_ has joined #openstack-cinder | 12:05 | |
| *** ducttape_ has joined #openstack-cinder | 12:06 | |
| *** ducttape_ has quit IRC | 12:08 | |
| *** ducttape_ has joined #openstack-cinder | 12:08 | |
| kwathore_ | HI All, | 12:09 |
| kwathore_ | I am getting an error while detaching any iscsi or fc volume from an instance: | 12:09 |
| kwathore_ | “DeviceDetachFailed: Device detach failed for vdb: Unable to detach from guest transient domain on” | 12:09 |
| kwathore_ | Libvert version:libvirt-daemon-driver-qemu-1.2.17-13.el7_2.5.x86_64 | 12:09 |
| kwathore_ | Could anyone help me in this? | 12:09 |
| *** e0ne has joined #openstack-cinder | 12:09 | |
| *** alyson_ has joined #openstack-cinder | 12:11 | |
| *** edmondsw has joined #openstack-cinder | 12:15 | |
| *** laughterwym has joined #openstack-cinder | 12:15 | |
| *** Julien-zte has joined #openstack-cinder | 12:19 | |
| *** haplo37__ has quit IRC | 12:21 | |
| *** mvk has quit IRC | 12:21 | |
| *** tpsilva has joined #openstack-cinder | 12:22 | |
| openstackgerrit | zheng yin proposed openstack/cinder: Add command information to cinder-manage.rst https://review.openstack.org/357162 | 12:24 |
| *** pauloewerton has joined #openstack-cinder | 12:25 | |
| *** ducttape_ has quit IRC | 12:26 | |
| *** gouthamr has joined #openstack-cinder | 12:26 | |
| openstackgerrit | zheng yin proposed openstack/cinder: Add command information to cinder-manage.rst https://review.openstack.org/357162 | 12:27 |
| *** laughterwym has quit IRC | 12:46 | |
| *** laughterwym has joined #openstack-cinder | 12:47 | |
| lixiaoy1 | jgriffith: ping | 12:50 |
| *** gouthamr is now known as gouthamr_afk | 12:51 | |
| *** e0ne has quit IRC | 12:52 | |
| *** julim has joined #openstack-cinder | 12:53 | |
| *** e0ne has joined #openstack-cinder | 12:55 | |
| *** ducttape_ has joined #openstack-cinder | 12:56 | |
| *** rlrossit has joined #openstack-cinder | 12:59 | |
| *** ducttape_ has quit IRC | 13:03 | |
| *** shausy has quit IRC | 13:04 | |
| openstackgerrit | Ivan Kolodyazhny proposed openstack/cinder: Merge two fake drivers into the one FakeLoggingVolumeDriver https://review.openstack.org/348428 | 13:04 |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Use the ClientHelper class to reduce amount of code https://review.openstack.org/347270 | 13:09 |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Avoid an extra resize later on https://review.openstack.org/347271 | 13:09 |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Do batch delivery https://review.openstack.org/347272 | 13:09 |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Report an error if a snapshot has vanished https://review.openstack.org/347273 | 13:09 |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Avoid duplicated volumes from the error path https://review.openstack.org/347274 | 13:09 |
| *** xyang1 has joined #openstack-cinder | 13:12 | |
| *** xyang has joined #openstack-cinder | 13:13 | |
| *** links has quit IRC | 13:16 | |
| *** thumpba has joined #openstack-cinder | 13:17 | |
| * cFouts gently reminds DuncanT of https://review.openstack.org/#/c/346135 | 13:18 | |
| *** Julien-zte has quit IRC | 13:18 | |
| DuncanT | cFouts: Hey, I got to that one a few minutes ago! | 13:19 |
| cFouts | hah | 13:19 |
| cFouts | thanks! sorry for the nag | 13:19 |
| *** ametts has joined #openstack-cinder | 13:20 | |
| DuncanT | cFouts: I'm really starting to think about getting theory-of-operations docs for drivers like netapp... they rely so much on vendor features that I need to keep notes to do the reviews. | 13:20 |
| *** dustins has joined #openstack-cinder | 13:20 | |
| *** liverpooler has quit IRC | 13:20 | |
| DuncanT | cFouts: A short doc on concepts like flexvols and such might be useful? | 13:21 |
| *** vgridnev has joined #openstack-cinder | 13:21 | |
| vgridnev | hello cinder team, do you know something about DSS Cinder CI ? | 13:22 |
| cFouts | DuncanT: we have the DOG that we generate for each release: http://netapp.github.io/openstack-deploy-ops-guide/mitaka/content/ch_cinder.html That isn't exactly what you are asking for though. I'll see what I can dig up | 13:24 |
| openstackgerrit | LisaLi proposed openstack/cinder: Attachment lost after migrating in-use volume https://review.openstack.org/356992 | 13:24 |
| *** mriedem has joined #openstack-cinder | 13:25 | |
| *** porrua has joined #openstack-cinder | 13:26 | |
| *** Julien-zte has joined #openstack-cinder | 13:26 | |
| DuncanT | cFouts: Some interesting stuff there though, thanks for the link | 13:26 |
| openstackgerrit | Hussain Chachuliya proposed openstack/cinder: Use min attribute from oslo_config https://review.openstack.org/356388 | 13:29 |
| *** cknight has joined #openstack-cinder | 13:29 | |
| e0ne | DuncanT: hi. do you have few minutes to talk about functional tests? | 13:33 |
| xyang | e0ne: If we add any new functional tests, they should be dependent on your fake driver patch, right? | 13:35 |
| *** vgridnev has quit IRC | 13:36 | |
| e0ne | xyang: yes. we're going to do so | 13:36 |
| xyang | e0ne: ok. I have someone look at the snapshots tests. I'll let you know if there's any progress on it | 13:37 |
| e0ne | xyang: here is patch to devstack https://review.openstack.org/#/c/348449/ to enable fake drivers | 13:37 |
| xyang | e0ne: thanks | 13:37 |
| e0ne | xyang: I want go get everything in infra before writing tests | 13:37 |
| xyang | e0ne: so we should hold off then? | 13:38 |
| e0ne | xyang: e.g. I'm working on fixing noauth mode | 13:38 |
| DuncanT | e0ne: Yes | 13:38 |
| *** akerr has joined #openstack-cinder | 13:38 | |
| e0ne | xyang: are you talking about cinder or cinderclient tests? | 13:38 |
| xyang | e0ne: cinder functional tests. right now there are some for volume, I asked someone to add tests for snapshot | 13:38 |
| xyang | e0ne: if you think we should wait for your patches to land, we can wait | 13:39 |
| openstackgerrit | Yuriy Nesenenko proposed openstack/cinder: Separate create and update rules for volume metadata https://review.openstack.org/356624 | 13:39 |
| *** BigWillie has joined #openstack-cinder | 13:39 | |
| e0ne | xyang: now, functional tests for cinder doen't run cinder services on gates | 13:39 |
| e0ne | DuncanT: according to noauth support for functional test: do we need cinderclient support too? | 13:40 |
| *** pbandark`brb has quit IRC | 13:40 | |
| e0ne | DuncanT: server-side is broken, but looks like it's easy to fixi t | 13:41 |
| DuncanT | e0ne: If the noauth is done as a fake middleware that has a fixed tenant / password hard coded (or two, one for admin) then the client needs no changes | 13:41 |
| DuncanT | e0ne: We reach into the context in too many places to make the old noauth approach sane, in my estimation | 13:41 |
| e0ne | DuncanT: but client uses keystone before any API call | 13:41 |
| *** ducttape_ has joined #openstack-cinder | 13:42 | |
| *** _coolsvap_ is now known as coolsvap | 13:42 | |
| DuncanT | e0ne: Oh, of course it does. Bugger | 13:43 |
| DuncanT | e0ne: Suggestions? You could put a fake keystone client in, since all the keystone stuff goes through that now, but it might be a pain | 13:44 |
| e0ne | DuncanT: in current implementation, we have to fix https://github.com/openstack/cinder/blob/master/cinder/api/middleware/auth.py#L129 to pass https://github.com/openstack/cinder/blob/master/cinder/api/openstack/wsgi.py#L891 | 13:45 |
| *** Asaithambi has joined #openstack-cinder | 13:45 | |
| DuncanT | e0ne: There's a bunch of reaching into the context object for quota stuff too | 13:46 |
| e0ne | DuncanT: yea, I didn't test qoutas with noauth yet | 13:46 |
| DuncanT | e0ne: If noauth can be fixed as it stands, how does that tie in with the client? | 13:47 |
| *** xyang has quit IRC | 13:47 | |
| *** dustins has quit IRC | 13:47 | |
| *** xyang has joined #openstack-cinder | 13:49 | |
| e0ne | DuncanT: I'm going to try implement NoAuthPlugin | 13:50 |
| DuncanT | e0ne: I've not looked at the framework there very much, when I was experimenting I just deleted a lot of code from the client until it did what I wanted | 13:52 |
| e0ne | DuncanT: :) | 13:52 |
| DuncanT | e0ne: It would be nice if as few codepaths as possible change between noauth and keystone auth | 13:52 |
| e0ne | DuncanT: we've got auth_plugin.py module. TBH, I'm not sure if it is supported | 13:53 |
| DuncanT | I'm not sure how practical that is though | 13:53 |
| DuncanT | e0ne: It's not tested, so I'd bet buttons to dollars that it is broken | 13:53 |
| e0ne | DuncanT: OK, so in general, getting NoAuthPlugin for client sounds good | 13:54 |
| *** asselin_ has joined #openstack-cinder | 13:54 | |
| DuncanT | e0ne: I guess so, yes. If it ends up requiring substantial work in the main cinderclient code, it might be worth making a fake keystoneclient instead | 13:54 |
| *** baojg has joined #openstack-cinder | 13:55 | |
| *** asselin has quit IRC | 13:56 | |
| e0ne | DuncanT: got it. I'll try to get working PoC and submit it to the review | 13:56 |
| DuncanT | e0ne: Awesome. Thanks. | 13:56 |
| e0ne | DuncanT: thanks for the help | 13:56 |
| DuncanT | e0ne: You're more then welcome. | 13:56 |
| *** baojg has quit IRC | 13:56 | |
| DuncanT | e0ne: I'll sit here, swear at ansible some more and question my life choices | 13:56 |
| *** jgregor has joined #openstack-cinder | 13:57 | |
| *** yangyape_ has joined #openstack-cinder | 13:57 | |
| e0ne | DuncanT: I like the idea of auth plugin, because it could be usable for standalone cinder mode too | 13:57 |
| DuncanT | e0ne: Yup, if it works then great, for sure. In the end, a fake keystoneclient is just a plugin by another name... | 13:58 |
| e0ne | DuncanT: yea, but I don't want to fix it each time, when we'll add something to keystoneclient | 13:59 |
| e0ne | DuncanT: less dependencies sounds better | 13:59 |
| DuncanT | e0ne: If it's something we rely on from keystoneclient in cinderclient, you'll end up having to do something, even if it is a proxy through the auth plugin | 14:00 |
| *** yangyapeng has quit IRC | 14:01 | |
| *** EinstCrazy has joined #openstack-cinder | 14:04 | |
| *** arecknag has quit IRC | 14:04 | |
| *** e0ne has quit IRC | 14:10 | |
| *** e0ne has joined #openstack-cinder | 14:14 | |
| hoobaman | hi | 14:15 |
| hoobaman | why is the qemu-img convert image validation not honouring the volume_dd_blocksize? | 14:16 |
| *** Julien-zte has quit IRC | 14:18 | |
| *** alonma has quit IRC | 14:19 | |
| *** xinli has joined #openstack-cinder | 14:24 | |
| *** xinli has quit IRC | 14:24 | |
| *** lprice has quit IRC | 14:25 | |
| *** alonma has joined #openstack-cinder | 14:26 | |
| *** haplo37__ has joined #openstack-cinder | 14:28 | |
| *** raunak has joined #openstack-cinder | 14:28 | |
| *** Asaithambi has quit IRC | 14:28 | |
| *** _ducttape_ has joined #openstack-cinder | 14:28 | |
| xyang | hoobaman: probably because it needs to convert to raw. It's the virtual size of the image that matters | 14:30 |
| *** zhangguoqing has quit IRC | 14:31 | |
| *** ducttape_ has quit IRC | 14:31 | |
| xyang | hoobaman: by the way, regarding your question yesterday, I think it has to do with the virtual size of your image | 14:31 |
| *** alonma has quit IRC | 14:31 | |
| xyang | hoobaman: The slowest part is to download the image to a temp location. It has not reached the backend yet | 14:32 |
| *** alonma has joined #openstack-cinder | 14:34 | |
| openstackgerrit | Justin A Wilson proposed openstack/cinder: Added config option to enable SSL https://review.openstack.org/352958 | 14:34 |
| *** enriquetaso has joined #openstack-cinder | 14:36 | |
| *** jschwarz has joined #openstack-cinder | 14:38 | |
| jschwarz | geguileo, heya :) | 14:38 |
| *** alonma has quit IRC | 14:38 | |
| *** Julien-zte has joined #openstack-cinder | 14:39 | |
| *** raunak has quit IRC | 14:41 | |
| *** alonma has joined #openstack-cinder | 14:41 | |
| *** Lee1092 has quit IRC | 14:43 | |
| *** Julien-zte has quit IRC | 14:44 | |
| *** alonma has quit IRC | 14:46 | |
| geguileo | jschwarz: Hi | 14:46 |
| * geguileo is on a meeting :-( | 14:47 | |
| jschwarz | geguileo, then ping me later :) | 14:47 |
| geguileo | jschwarz: Ok | 14:47 |
| *** xenogear has quit IRC | 14:47 | |
| *** alonma has joined #openstack-cinder | 14:48 | |
| openstackgerrit | Matan Sabag proposed openstack/cinder: Changes stats report for ScaleIO driver https://review.openstack.org/356120 | 14:49 |
| *** yangyapeng has joined #openstack-cinder | 14:49 | |
| *** rajinir has joined #openstack-cinder | 14:50 | |
| *** yangyape_ has quit IRC | 14:50 | |
| *** _ducttape_ has quit IRC | 14:51 | |
| *** ducttape_ has joined #openstack-cinder | 14:52 | |
| *** dmellado|off is now known as dmellado | 14:52 | |
| openstackgerrit | Yuriy Nesenenko proposed openstack/cinder: Separate create and update rules for volume metadata https://review.openstack.org/356624 | 14:53 |
| *** alonma has quit IRC | 14:53 | |
| *** bill_az has joined #openstack-cinder | 14:54 | |
| *** xyang has quit IRC | 14:55 | |
| *** alonma has joined #openstack-cinder | 14:55 | |
| *** xyang has joined #openstack-cinder | 14:55 | |
| *** jgregor has quit IRC | 14:55 | |
| openstackgerrit | Merged openstack/python-cinderclient: Add "start_version" and "end_version" support to argparse https://review.openstack.org/340129 | 14:56 |
| *** jgregor has joined #openstack-cinder | 14:57 | |
| *** gouthamr_afk is now known as gouthamr | 14:59 | |
| *** alonma has quit IRC | 14:59 | |
| *** tbarron|afk is now known as tbarron | 15:00 | |
| *** diablo_rojo has joined #openstack-cinder | 15:00 | |
| *** salv-orlando has joined #openstack-cinder | 15:00 | |
| *** bill_az has quit IRC | 15:01 | |
| *** jungleboyj has joined #openstack-cinder | 15:01 | |
| *** lprice has joined #openstack-cinder | 15:01 | |
| *** xenogear has joined #openstack-cinder | 15:02 | |
| *** Linkmark has quit IRC | 15:05 | |
| *** EinstCrazy has quit IRC | 15:05 | |
| *** jdurgin1 has joined #openstack-cinder | 15:09 | |
| *** dustins has joined #openstack-cinder | 15:09 | |
| *** xinli has joined #openstack-cinder | 15:10 | |
| *** coolsvap is now known as _coolsvap_ | 15:16 | |
| smcginnis | Pike and Queens. Could have been worse. :) | 15:17 |
| hemna | mep | 15:17 |
| *** IanGovett has joined #openstack-cinder | 15:19 | |
| *** mvk has joined #openstack-cinder | 15:19 | |
| *** mmedvede has quit IRC | 15:21 | |
| *** rcernin has quit IRC | 15:22 | |
| *** laughterwym has quit IRC | 15:25 | |
| Swanson | "Pike and Queens. Could have been worse. :)" | 15:25 |
| Swanson | How far back do I have to scroll for context? | 15:25 |
| *** laughterwym has joined #openstack-cinder | 15:25 | |
| smcginnis | No context. Just random spewage. | 15:25 |
| smcginnis | It's like a treasure hunt. :) | 15:25 |
| *** laughter_ has joined #openstack-cinder | 15:26 | |
| *** laughterwym has quit IRC | 15:26 | |
| *** sdake_ has joined #openstack-cinder | 15:29 | |
| *** IanGovett has quit IRC | 15:30 | |
| *** e0ne has quit IRC | 15:30 | |
| *** mmedvede has joined #openstack-cinder | 15:34 | |
| *** baumann has joined #openstack-cinder | 15:34 | |
| *** dustins has quit IRC | 15:34 | |
| *** leeantho has joined #openstack-cinder | 15:35 | |
| *** diablo_rojo has quit IRC | 15:35 | |
| _alastor_ | If anyone has a spare moment: https://review.openstack.org/#/c/352533/ | 15:37 |
| hemna | _alastor_, looking | 15:38 |
| _alastor_ | hemna: Thanks! | 15:38 |
| *** freerunner has joined #openstack-cinder | 15:39 | |
| *** Apoorva has joined #openstack-cinder | 15:39 | |
| *** diablo_rojo has joined #openstack-cinder | 15:42 | |
| hemna | _alastor_, looks good man | 15:42 |
| hemna | _alastor_, one possible addition is to add the @utils.trace decorator to some select functions | 15:42 |
| hemna | it has helped me debug some issues on our driver fwiw | 15:43 |
| hemna | like initialize_connection, terminate_connection | 15:43 |
| hemna | for attaches | 15:43 |
| smcginnis | _alastor_: A couple comments there. | 15:44 |
| _alastor_ | hemna: I set utils.setup_tracing at the method level with the datera_debug option. Is that not the same? I get method input/output in the logs with that. | 15:44 |
| *** dustins has joined #openstack-cinder | 15:44 | |
| hemna | ah ok, so that works automagically then. | 15:45 |
| hemna | cool | 15:45 |
| hemna | ah I see that on 171 | 15:45 |
| _alastor_ | hemna: Yeah, I think so. Not sure what black magic they're using for it though ;) | 15:45 |
| *** mmedvede has quit IRC | 15:45 | |
| hemna | that might be a bit much for the get_manageable_volumes though | 15:45 |
| hemna | but, that's up to you | 15:46 |
| _alastor_ | hemna: True, but it has to be set explicitly and I think it's better to have the option | 15:46 |
| hemna | yah works for me. | 15:47 |
| *** Asaithambi has joined #openstack-cinder | 15:47 | |
| *** mvk has quit IRC | 15:50 | |
| _alastor_ | smcginnis: I'm using the deprecated_for_removal keyword arg in those config options, I just additonally added the (DEPRECATED) string to the help to make it easy for folks reading the source to see | 15:51 |
| _alastor_ | smcginnis: Should I remove that? | 15:51 |
| smcginnis | _alastor_: Oops. Nevermind. :) | 15:52 |
| smcginnis | Maybe overkill for the DEPRECATED text in the string, but no harm. | 15:52 |
| hemna | doesn't oslo config automatically label it as that? | 15:52 |
| smcginnis | Yep | 15:53 |
| *** vbala has quit IRC | 15:53 | |
| _alastor_ | I can remove it, nbd | 15:53 |
| openstackgerrit | Matt Smith (_alastor_) proposed openstack/cinder: Datera driver 2.2 update https://review.openstack.org/352533 | 15:54 |
| *** Suyi has joined #openstack-cinder | 15:54 | |
| smcginnis | You can see some examples in here: http://docs.openstack.org/developer/cinder/sample_config.html | 15:54 |
| *** Asaithambi has quit IRC | 15:57 | |
| *** dustins has quit IRC | 15:57 | |
| *** ftersin has joined #openstack-cinder | 15:59 | |
| Swanson | What were the problem names? Pee and Queue would have been my choices. | 15:59 |
| *** aleksey8 has joined #openstack-cinder | 16:02 | |
| *** aleksey8 is now known as alekseyruban | 16:02 | |
| openstackgerrit | Arnon Yaari proposed openstack/cinder: New cinder driver to support INFINIDAT InfiniBox https://review.openstack.org/356986 | 16:02 |
| openstackgerrit | Walter A. Boring IV (hemna) proposed openstack/cinder: 3PAR use OSLO versioned Object for volume https://review.openstack.org/339147 | 16:05 |
| *** xyang has quit IRC | 16:06 | |
| openstackgerrit | Merged openstack/cinder: Handle snapshot delete conflict in 3par driver https://review.openstack.org/342722 | 16:08 |
| *** laughter_ has quit IRC | 16:09 | |
| *** laughterwym has joined #openstack-cinder | 16:09 | |
| openstackgerrit | Merged openstack/cinder: NetApp: Add Consistency Group Support for NFS https://review.openstack.org/346135 | 16:11 |
| *** arch-nemesis has joined #openstack-cinder | 16:12 | |
| *** xyang has joined #openstack-cinder | 16:13 | |
| *** laughterwym has quit IRC | 16:14 | |
| *** sdake_ has quit IRC | 16:16 | |
| *** ametts has quit IRC | 16:16 | |
| *** diablo_rojo has quit IRC | 16:19 | |
| *** jgregor has quit IRC | 16:19 | |
| *** mtanino has joined #openstack-cinder | 16:20 | |
| *** jgregor has joined #openstack-cinder | 16:20 | |
| *** baumann has quit IRC | 16:21 | |
| *** ametts has joined #openstack-cinder | 16:21 | |
| *** jungleboyj has quit IRC | 16:24 | |
| *** tesseract- has quit IRC | 16:26 | |
| *** xyang has quit IRC | 16:27 | |
| *** xyang has joined #openstack-cinder | 16:29 | |
| *** Apoorva has quit IRC | 16:30 | |
| *** bwallis has quit IRC | 16:34 | |
| *** liverpooler has joined #openstack-cinder | 16:37 | |
| *** jungleboyj has joined #openstack-cinder | 16:37 | |
| *** thumpba_ has joined #openstack-cinder | 16:37 | |
| *** thumpba has quit IRC | 16:39 | |
| *** liverpooler has quit IRC | 16:40 | |
| *** jordanP has quit IRC | 16:43 | |
| *** sdake has joined #openstack-cinder | 16:43 | |
| *** sdague has quit IRC | 16:49 | |
| geguileo | jschwarz: Sorry, I totally forgot to ping you back :-( | 16:49 |
| jschwarz | geguileo, oh that's ok, I was in a meeting until just now XD | 16:50 |
| geguileo | jschwarz: Great :-) | 16:50 |
| *** raunak has joined #openstack-cinder | 16:50 | |
| jschwarz | geguileo, so tooz | 16:50 |
| jschwarz | geguileo, I've started looking at things and it looks like you guys implemented a few stuff that the tooz guys also implemented concurrently | 16:51 |
| jschwarz | geguileo, mostly the heartbeat thread, etc | 16:51 |
| jschwarz | geguileo, so I'm gonna submit patches to join them together in tooz | 16:52 |
| jschwarz | geguileo, the idea is that the "useful" coordination.py code in cinder should be shared in tooz instead, and cinder's code should be as small as possible | 16:52 |
| jschwarz | geguileo, then we can re-use as much as possible in neutron | 16:52 |
| jschwarz | geguileo, thoughts? :) | 16:52 |
| geguileo | jschwarz: I'm going to be redoing some of that stuff | 16:53 |
| geguileo | jschwarz: Because I'll be needing a supervisor | 16:53 |
| jschwarz | geguileo, redoing how? | 16:54 |
| geguileo | jschwarz: That can auto fence the node | 16:54 |
| jschwarz | geguileo, ahh.. details on how that's going to work? | 16:54 |
| jschwarz | geguileo, alas, the principle persists - I think the fencer should implemented in tooz to begin with | 16:54 |
| geguileo | jschwarz: I'll need to keep count of the number of locks in use in the process | 16:54 |
| jschwarz | geguileo, and how do you notify a thread that its lock has been lost due to fencing? | 16:55 |
| geguileo | jschwarz: I'm not sure I'll be able to do it in tooz, it needs to have some knowledge of how to stop and properly restart the backend... | 16:55 |
| *** bwallis has joined #openstack-cinder | 16:55 | |
| jschwarz | geguileo, blah. | 16:55 |
| jschwarz | geguileo, callback? | 16:56 |
| geguileo | jschwarz: I have to look if a send a signal or if I just kill it | 16:56 |
| jschwarz | geguileo, think of it this way: if you implement it in cinder, we'll have to copy-paste it in neutron. practically a fork | 16:56 |
| jschwarz | which is not optimal at all | 16:56 |
| geguileo | jschwarz: Maybe Neutron doesn't agree with our view | 16:57 |
| geguileo | jschwarz: I think it's easier to get a specific solution | 16:57 |
| geguileo | jschwarz: And work from there to get the generic version | 16:57 |
| *** akapil has quit IRC | 16:57 | |
| geguileo | jschwarz: Trying to get consensus between Neutron, Cinder, Tooz, and anybody else would simply kill me | 16:57 |
| jschwarz | geguileo, if Neutron doesn't want it *right now*, it will soon enough once I mention this to them :) | 16:58 |
| geguileo | jschwarz: I mean, I've been working on this A/A stuff for over a year now | 16:58 |
| geguileo | jschwarz: And I would like to finish it in the next release | 16:58 |
| scottda | noone is allowed to bother geguileo , much less kill him :) | 16:58 |
| jschwarz | geguileo, so once you know you want to fench a process, you want to either call a function that initiates it (a callback) or send a signal | 16:58 |
| jschwarz | geguileo, both are generics | 16:59 |
| geguileo | scottda: XD XD XD | 16:59 |
| geguileo | scottda: Thanks :-) | 16:59 |
| jschwarz | geguileo, and I assure you, I neither want to kill you, nor not-finish it by next release. | 16:59 |
| geguileo | jschwarz: Or just kill the thread, whichever is more convenient or easier to implement ;-) | 16:59 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/cinder: Updated from global requirements https://review.openstack.org/356867 | 16:59 |
| jschwarz | geguileo, agreed | 16:59 |
| geguileo | jschwarz: I know that's not your intention | 17:00 |
| jschwarz | geguileo, either way, the point is that the action part is generic enough to be implemented in tooz | 17:00 |
| geguileo | jschwarz: But then this work would depend on moving our current code to Tooz | 17:00 |
| geguileo | jschwarz: I hope I can get to work on the specs for the auto fencing in 2 weeks | 17:00 |
| jschwarz | geguileo, I'll be willing to take on the migration-to-tooz part of the job myself | 17:00 |
| *** _ducttape_ has joined #openstack-cinder | 17:00 | |
| geguileo | jschwarz: Once I write some stuff and create a PoC I'll ping you so we can discuss it | 17:01 |
| geguileo | jschwarz: Sounds good? | 17:01 |
| jschwarz | geguileo, ok | 17:01 |
| jschwarz | geguileo, I didn't know you were writing a spec | 17:01 |
| jschwarz | that does complicate things :P | 17:02 |
| geguileo | jschwarz: I have to write one | 17:02 |
| geguileo | jschwarz: I wrote something, but it was when I was thinking something completely different | 17:02 |
| *** jordanP has joined #openstack-cinder | 17:02 | |
| erlon | smcginnis: ping | 17:03 |
| patrickeast | jschwarz: one thing to keep in mind is that there is a strong aversion by some in cinder to making tooz a hard dependency, not a deal breaker IMO, but worth trying to have a plan to deal with negative feedback on a spec or changes to move that stuff out of cinder | 17:03 |
| geguileo | jschwarz: https://review.openstack.org/#/c/237076/ | 17:03 |
| geguileo | patrickeast: The aversion should be against a DLM, not Tooz itself | 17:03 |
| scottda | yes, we want deployers who don't want AA/HA to be able to deploy without tooz if they want to. | 17:04 |
| geguileo | scottda: But tooz is a very small library | 17:04 |
| scottda | right, that's DLM in general, not tooz. | 17:04 |
| patrickeast | geguileo: yea, but they are kinda seen as one in the same :( | 17:04 |
| geguileo | Ok | 17:04 |
| *** ducttape_ has quit IRC | 17:04 | |
| scottda | I think we're DLM library agnostic. | 17:04 |
| scottda | or should be. | 17:04 |
| jschwarz | I'll be right back guys | 17:05 |
| geguileo | Well, I think it makes sense to move stuff to TooZ as long as everything can be optionally used or not | 17:05 |
| *** Apoorva has joined #openstack-cinder | 17:05 | |
| scottda | sure. Just as long as people that want a simple deployment with local file locks instead of DLM can still do that. | 17:05 |
| *** sdague has joined #openstack-cinder | 17:06 | |
| geguileo | scottda: Oh, that's for sure | 17:06 |
| geguileo | scottda: If we lose that we are doing to wrong, terribly wrong!! | 17:07 |
| scottda | I'm not sure how strongly people would feel about a lack of choice of DLM, if tooz started to contain the necessary logic for things like fencing, and some alternate DLM did not.. | 17:07 |
| geguileo | scottda: Well, the fencing should be DLM agnostic | 17:07 |
| geguileo | scottda: Otherwise it's useless | 17:07 |
| scottda | cool. we're in agreement, I think | 17:07 |
| geguileo | scottda: I'll ping you too once I have something concrete | 17:08 |
| smcginnis | erlon: pong | 17:08 |
| scottda | geguileo: great, thanks, and thanks for all your work on this. You are greatly appreciated. | 17:08 |
| smcginnis | scottda, geguileo: +1 | 17:09 |
| * scottda thinks maybe we should re-name Active-Active High Availability -> Gorkafication | 17:09 | |
| scottda | or at least geguileo 's favorite food.. | 17:09 |
| *** pbandark has joined #openstack-cinder | 17:10 | |
| jschwarz | I'll vote for that, scottda | 17:10 |
| erlon | smcginnis: Hey Sean, | 17:10 |
| smcginnis | erlon: Hey | 17:10 |
| erlon | smcginnis: while reviewing this driver: https://review.openstack.org/#/c/356986/1/cinder/volume/drivers/infinidat.py | 17:11 |
| geguileo | XD XD XD | 17:11 |
| jschwarz | geguileo, so that spec you mentioned is specifically for the auto-fencing? | 17:11 |
| erlon | smcginnis: I hit something I think that I think is not usual/right from the Cinder perpective | 17:12 |
| geguileo | jschwarz: If I didn't make a mistake it should | 17:12 |
| smcginnis | erlon: Oh, what was that? | 17:12 |
| geguileo | jschwarz: Because I created multiple specs for the A/A work | 17:12 |
| geguileo | jschwarz: And that's the last one that I will have to implement | 17:12 |
| jschwarz | geguileo, so that complicates things in the way that if a design was made according to a spec in cinder, I see tooz devs arguing against copying that to tooz | 17:13 |
| erlon | smcginnis: for what I understood, the driver clones volumes, using some kind of internal snapshot feature that can't decouple the clones from the original | 17:13 |
| geguileo | jschwarz: Oh, but it's not a problem | 17:13 |
| geguileo | jschwarz: That spec is useless now | 17:13 |
| geguileo | jschwarz: I'll work on a new one and see if it can be done project agnostic | 17:13 |
| jschwarz | geguileo, that sounds good | 17:13 |
| erlon | smcginnis: so, you clone/snapshot a volume and you are stuck with the original until you delete all copies | 17:14 |
| geguileo | jschwarz: Create a Poc, and then I may propose the spec in Tooz | 17:14 |
| geguileo | jschwarz: And change the one in Cinder to explain that it's going to use that feature, and blah, blah, blah | 17:14 |
| smcginnis | erlon: Hmm, they call out HPE and LVM as being the same. Need to look at that. | 17:14 |
| jschwarz | geguileo, sounds excellent | 17:15 |
| jschwarz | geguileo, so I'll wait for your POC before I start porting things | 17:15 |
| erlon | smcginnis: LVM does not do for shure, LVM only checks if the volume has snapshots, what IMO shouldn't be needed as the API already does that, an in that case it throuhs the exception | 17:15 |
| openstackgerrit | Alexey Khodos proposed openstack/cinder: NexentaStor5: add extend method to NFS driver https://review.openstack.org/297880 | 17:15 |
| smcginnis | erlon: They may be right. Looks like LVM "clones" by creating a snapshot, and in delete_volume it checks if there are existing snapshots. | 17:16 |
| erlon | smcginnis: so, if you create a clone in LVM driver you can't delete the source? | 17:17 |
| erlon | smcginnis: that seems sooo wrong | 17:17 |
| smcginnis | erlon: That's what it looks like to me. Maybe jgriffith or someone that has been good enough to maintain the LVM driver can confirm that. | 17:17 |
| bswartz | lvm snapshots should be independent | 17:18 |
| smcginnis | I just haven't used it in enough different scenarios to really know how it all works. | 17:18 |
| bswartz | it's cinder that prevents you from deleting a volume which has snapshots | 17:18 |
| bswartz | no driver is allowed to delete volumes with snapshots | 17:18 |
| smcginnis | bswartz: Not snapshots. Clones. But looks like LVM creates clones by creating a snapshot. | 17:18 |
| bswartz | yes that should be valid | 17:19 |
| bswartz | what's the problem? | 17:19 |
| smcginnis | LVM clones by creating a snapshot here: https://github.com/openstack/cinder/blob/master/cinder/volume/drivers/lvm.py#L482 | 17:19 |
| bswartz | an LVM "snapshot" is just another LV | 17:19 |
| *** cdelatte has joined #openstack-cinder | 17:19 | |
| smcginnis | And fails a delete here: https://github.com/openstack/cinder/blob/master/cinder/volume/drivers/lvm.py#L427 | 17:19 |
| *** _ducttape_ has quit IRC | 17:19 | |
| erlon | bswartz: the problem is if is possible/or not to delete the source after you have cloned a volume | 17:19 |
| smcginnis | bswartz: Right. So the philosophical question is = is a snapshot a clone? | 17:20 |
| bswartz | yeah... it should be possible to delete the souce | 17:20 |
| smcginnis | bswartz: Not if it's a snapshot here. | 17:20 |
| *** ducttape_ has joined #openstack-cinder | 17:20 | |
| bswartz | smcginnis -- no need to as philisophical questions -- what matters is the API contract | 17:20 |
| smcginnis | :) | 17:20 |
| bswartz | create volume, clone volume, delete (first) volume should succeed | 17:20 |
| erlon | bswartz: mhm, which makes total sense | 17:21 |
| smcginnis | bswartz: So basically at least LVM breaks that contract. And possibly others. | 17:21 |
| bswartz | netapp actually performs quite a big of magic to ensure the above is true | 17:21 |
| smcginnis | bswartz: I agree that's how it should work. | 17:21 |
| smcginnis | But it appears that's not always the case. | 17:21 |
| bswartz | smcginnis: this could simply be a small bug in LVM though | 17:21 |
| smcginnis | hemna: ping | 17:22 |
| erlon | bswartz: not that small once that to fix would require a lot of copy to release the source | 17:22 |
| *** alkhodos has joined #openstack-cinder | 17:23 | |
| scottda | maybe not small, but sounds like a bug to me. If it's a clone, you can delete the source. That's the point of clone. | 17:24 |
| hemna | smcginnis, sup | 17:24 |
| alkhodos | Hi everyone, I have this review https://review.openstack.org/#/c/297880/ which says "Patch in Merge Conflict". I have rebased it before pushing latest patch set 5 minutes ago, but it did not help. How can I fix this? | 17:25 |
| bswartz | erlon: I find it very hard to believe that this never worked correctly in the past | 17:25 |
| erlon | bswartz: me too, looks odd that nobody wouldn't already hit that | 17:25 |
| smcginnis | hemna: Do you know if any of the HPE arrays treat clones as snapshots and prevent deleting the source volume if clones exist? | 17:26 |
| smcginnis | hemna: Called out here: https://review.openstack.org/#/c/356986/1/cinder/volume/drivers/infinidat.py | 17:26 |
| erlon | bswartz: should be easy to test that in tempest | 17:27 |
| smcginnis | bswartz: Or just no one clones when using LVM. | 17:27 |
| smcginnis | erlon: I think that's a good idea. | 17:27 |
| hemna | the 3PAR clone operation creates a separate volume on the array that has no association with the original after the operation is completed. | 17:28 |
| *** Apoorva has quit IRC | 17:28 | |
| smcginnis | hemna: OK, good. | 17:28 |
| hemna | during the clone operation there is a temporary association | 17:28 |
| erlon | hemna: thanks | 17:28 |
| hemna | that prevents deletion, or requires some special work | 17:28 |
| erlon | hemna: after that they are split | 17:29 |
| hemna | yah, once it's completed, it's split. | 17:29 |
| smcginnis | So this might just be a bug/restriction with the LVM driver. | 17:30 |
| hemna | https://github.com/openstack/cinder/blob/master/cinder/volume/drivers/hpe/hpe_3par_common.py#L1991-L2019 | 17:30 |
| *** lpetrut has quit IRC | 17:31 | |
| hemna | line 2000 there explains it a little bit | 17:31 |
| smcginnis | hemna: Which I think is totally legit when you're in the middle of a clone anyway. | 17:31 |
| hemna | yah I would presume other arrays are similar | 17:32 |
| smcginnis | "Using non-online copy" - umm, like offline? :) | 17:32 |
| hemna | yah those are 3par terms | 17:32 |
| hemna | online copy means it's a background process that makes the volume available/online immediately for attaching | 17:32 |
| *** dustins has joined #openstack-cinder | 17:34 | |
| erlon | hemna: shouldn' t have a lock on the volume there? what happens if there's a clone + delete operation in the sequence? | 17:35 |
| hemna | we have code in delete that acounts for the online copy taking place and we stop the copy then delete and cleanup | 17:35 |
| hemna | https://github.com/openstack/cinder/blob/master/cinder/volume/drivers/hpe/hpe_3par_common.py#L2071 | 17:36 |
| erlon | hemna: hmm? it fails the copy already in place? | 17:36 |
| hemna | it stops it and nukes it. | 17:37 |
| smcginnis | Seems like it would be better to wait for the copy to complete, then delete the source. But as long as that's the expected behavior, they really should know better than to delete a volume being cloned anyway. | 17:39 |
| erlon | hemna: hmm, that sounds bad, the opposite looks like a better way to handle | 17:39 |
| hemna | that copy can take a very very long time | 17:40 |
| hemna | depending on how large the volume is | 17:40 |
| erlon | hemna: smcginnis: yes, the user must know what he is doing | 17:40 |
| hemna | if the user asks to delete, we assume they mean it, and stop and kill everything. | 17:40 |
| smcginnis | The customer is always right. ;) | 17:41 |
| hemna | even when they aren't | 17:42 |
| *** gouthamr is now known as gouthamr_afk | 17:42 | |
| akerr | ya but here you might have more than 1 user of the same tenant who aren't talking to each other :( | 17:42 |
| *** gouthamr_afk is now known as gouthamr-afk | 17:42 | |
| erlon | akerr: worst, you can have a imaga cache that is being used by many tenants | 17:43 |
| akerr | but images can be "protected" and we're talking about create from volume here aren't we? | 17:44 |
| *** bardia has joined #openstack-cinder | 17:45 | |
| akerr | not a great situation in general but I'd agree with the idea of finishing the clone in progress before nuking the source. just my $.02 | 17:46 |
| erlon | akerr: it's an image storage inside the storage | 17:46 |
| hemna | there is no mechanism built in that can allow drivers to take an action at some time in the future | 17:47 |
| hemna | we are talking possibly hours | 17:47 |
| hemna | depending on how many of these are queued up and ongoing. | 17:47 |
| hemna | and the size. | 17:47 |
| hemna | it's completely unrealistic to expect the clone to finish in this case before deletion. | 17:48 |
| * erlon needs to start working with volumes bigger than 1G | 17:49 | |
| *** bardia has quit IRC | 17:49 | |
| hemna | heh yup :) | 17:50 |
| *** bardia has joined #openstack-cinder | 17:50 | |
| akerr | ya, reality sucks | 17:50 |
| akerr | i much prefer my idealistic mind-world | 17:50 |
| *** dustins has quit IRC | 17:50 | |
| *** rcernin has joined #openstack-cinder | 17:51 | |
| bswartz | erlon: do you have LVM set to thick or thin | 17:51 |
| erlon | bswartz: haven't tested that yet, Im fixing the LVM BE | 17:52 |
| erlon | bswartz: Ill try both | 17:52 |
| bswartz | smcginnis, erlon: my reading of the code is that if it's thick, the clone method does nothing | 17:53 |
| bswartz | and if it's thin, it seems that LVM lets you delete the parent LV of a snap | 17:53 |
| erlon | bswartz: if is thick it should use dd | 17:54 |
| bswartz | erlon: I agree | 17:54 |
| bswartz | nobody thinks thick LVM is a good idea though | 17:54 |
| bswartz | thin LVM is the only way to go | 17:55 |
| bswartz | thick is supported for backwards compat only | 17:55 |
| *** Apoorva has joined #openstack-cinder | 17:56 | |
| *** ftersin has left #openstack-cinder | 18:02 | |
| *** mmedvede- has joined #openstack-cinder | 18:08 | |
| *** dustins has joined #openstack-cinder | 18:08 | |
| hemna | https://bugs.launchpad.net/oslo.privsep/+bug/1611061 | 18:10 |
| openstack | Launchpad bug 1611061 in oslo.privsep "redirecting stderr truncates output" [Undecided,New] | 18:10 |
| hemna | sadness | 18:10 |
| hemna | not a single comment since it was filed | 18:10 |
| hemna | nothing | 18:10 |
| hemna | :( | 18:11 |
| openstackgerrit | Matt Smith (_alastor_) proposed openstack/cinder: Datera driver 2.2 update https://review.openstack.org/352533 | 18:13 |
| *** zul has quit IRC | 18:16 | |
| *** baumann has joined #openstack-cinder | 18:17 | |
| *** baumann has quit IRC | 18:17 | |
| *** baumann has joined #openstack-cinder | 18:18 | |
| patrickeast | hemna: my guess is as soon as people start deploying with it and complaining about all the weird logging it will gain some traction | 18:22 |
| *** zul has joined #openstack-cinder | 18:23 | |
| *** xyang has quit IRC | 18:26 | |
| erlon | bswartz: smcginnis: the deletion works with LVM | 18:27 |
| smcginnis | erlon: Oh, nice. Thanks for checking on it. | 18:27 |
| openstackgerrit | Yucong Feng proposed openstack/cinder: Fix Cisco Initiator zoning updates https://review.openstack.org/356121 | 18:28 |
| erlon | smcginnis: both thin and thick | 18:28 |
| smcginnis | erlon: So the unfortunate thing will be if they really have no way of supporting that with their product. | 18:30 |
| smcginnis | Which would seem really odd. | 18:30 |
| smcginnis | erlon: Good eye catching that! | 18:30 |
| *** xyang has joined #openstack-cinder | 18:30 | |
| erlon | smcginnis: thanks, so, not sure how they should fix that, may be using dd like LVM | 18:31 |
| openstackgerrit | Jay Bryant proposed openstack/cinder: Change NFS driver to not throw exception without nfs shares file https://review.openstack.org/341282 | 18:31 |
| erlon | smcginnis: but Its hard to imagine how a storage wouldn't support the split | 18:32 |
| smcginnis | erlon: Yeah, I would think there's got to be a way for them to handle that. | 18:32 |
| *** Lee1092 has joined #openstack-cinder | 18:33 | |
| smcginnis | erlon: Commented on there. | 18:35 |
| erlon | smcginnis: mhm | 18:37 |
| *** bardia_ has joined #openstack-cinder | 18:41 | |
| *** bardia has quit IRC | 18:41 | |
| *** dustins has quit IRC | 18:47 | |
| *** lpetrut has joined #openstack-cinder | 18:47 | |
| *** ociuhandu has quit IRC | 18:47 | |
| *** xyang has quit IRC | 18:48 | |
| *** catintheroof has joined #openstack-cinder | 18:49 | |
| *** xyang has joined #openstack-cinder | 18:50 | |
| *** rlrossit has quit IRC | 18:52 | |
| openstackgerrit | venkata krishna proposed openstack/cinder: Add replication failback in Kaminario K2 drivers https://review.openstack.org/356475 | 18:53 |
| *** rlrossi__ has joined #openstack-cinder | 18:55 | |
| *** fifieldt has quit IRC | 19:06 | |
| *** pbandark has quit IRC | 19:06 | |
| *** julim has quit IRC | 19:13 | |
| *** xyang has quit IRC | 19:13 | |
| openstackgerrit | Merged openstack/cinder: EMC VMAX - iSCSI Multipath support https://review.openstack.org/317060 | 19:13 |
| *** julim has joined #openstack-cinder | 19:14 | |
| *** xyang has joined #openstack-cinder | 19:15 | |
| *** salv-orl_ has joined #openstack-cinder | 19:15 | |
| *** fifieldt has joined #openstack-cinder | 19:18 | |
| *** salv-orlando has quit IRC | 19:18 | |
| patrickeast | oo thats a new one, zuul-merger's cinder repo got a corrupt git index so all my ci jobs have been silently failing this morning | 19:23 |
| bswartz | erlon: I'm glad it worked out | 19:24 |
| bswartz | erlon: do you know what code path is followed when you clone an LVM thick volume? | 19:24 |
| erlon | bswartz: yep, but just don't know how it can clone that fast and delete the source without coping | 19:25 |
| erlon | bswartz: may be it does take time if you change a lot in the desination | 19:26 |
| bswartz | erlon: 1g takes less time to copy then you'd think | 19:26 |
| bswartz | if you create a 100g volume and it's still fast then I'd be suspicious | 19:26 |
| erlon | bswartz: hmm, another point is that I didn't write to the volume, so, it was less that 1G | 19:27 |
| *** thumpba_ has quit IRC | 19:29 | |
| bswartz | erlon: thick volumes are thick | 19:30 |
| bswartz | dd doesn't know it's full of zeros | 19:30 |
| erlon | bswartz: I tested thin with 20GB volumes, and thick with 2GB | 19:31 |
| erlon | bswartz: it toke 5~10 s to create | 19:31 |
| erlon | the thick ones, thin are immediately | 19:32 |
| openstackgerrit | Patrick East proposed openstack/cinder: Add release note to warn about os-brick lock dir https://review.openstack.org/354501 | 19:32 |
| *** dgonzalez has quit IRC | 19:34 | |
| *** e0ne has joined #openstack-cinder | 19:34 | |
| openstackgerrit | Alyson proposed openstack/cinder: HNAS: Add support for manage/unmanage snapshots in NFS driver https://review.openstack.org/333992 | 19:34 |
| openstackgerrit | Nate Potter proposed openstack/cinder: Public core API to check backup capability at runtime. https://review.openstack.org/350310 | 19:38 |
| *** thumpba has joined #openstack-cinder | 19:39 | |
| openstackgerrit | Michael Price proposed openstack/cinder: NetApp E-Series: Add volume tagging for back-end https://review.openstack.org/348958 | 19:40 |
| *** leeantho_ has joined #openstack-cinder | 19:44 | |
| *** ociuhandu has joined #openstack-cinder | 19:45 | |
| *** dgonzalez has joined #openstack-cinder | 19:45 | |
| *** leeantho has quit IRC | 19:46 | |
| *** bill_az has joined #openstack-cinder | 19:48 | |
| openstackgerrit | Ivan Kolodyazhny proposed openstack/cinder: Merge two fake drivers into the one FakeLoggingVolumeDriver https://review.openstack.org/348428 | 19:53 |
| *** Apoorva_ has joined #openstack-cinder | 19:59 | |
| *** Apoorva has quit IRC | 20:03 | |
| *** ducttape_ has quit IRC | 20:04 | |
| *** xyang has quit IRC | 20:04 | |
| *** thumpba has quit IRC | 20:04 | |
| *** ducttape_ has joined #openstack-cinder | 20:04 | |
| *** GB21 has quit IRC | 20:04 | |
| *** enriquetaso has quit IRC | 20:05 | |
| *** xyang has joined #openstack-cinder | 20:06 | |
| *** porrua has quit IRC | 20:07 | |
| *** JoseMello has quit IRC | 20:10 | |
| openstackgerrit | Raunak Kumar proposed openstack/cinder: Add support for force backup for Nimble Storage https://review.openstack.org/349086 | 20:13 |
| *** xyang has quit IRC | 20:14 | |
| *** diablo_rojo has joined #openstack-cinder | 20:20 | |
| *** _coolsvap_ has quit IRC | 20:21 | |
| openstackgerrit | Aleksey Ruban proposed openstack/cinder: Nexenta: NexentaStor4 NFS driver cache capacity and use sessions https://review.openstack.org/356142 | 20:22 |
| *** BigWillie has quit IRC | 20:24 | |
| *** nikeshm has joined #openstack-cinder | 20:27 | |
| *** diogogmt has joined #openstack-cinder | 20:27 | |
| nikeshm | hi | 20:27 |
| nikeshm | if any one get time please review https://review.openstack.org/#/c/356475/ | 20:28 |
| nikeshm | thanks | 20:28 |
| *** Lee1092 has quit IRC | 20:33 | |
| *** akapil has joined #openstack-cinder | 20:37 | |
| tbarron | ping patrickeast when you get a chance, have a query about how to verify volume limits on purestorage side? | 20:40 |
| patrickeast | tbarron: there isn't an API to get the limits (i've got an RFE open for that), but its pretty easy right now | 20:41 |
| *** akapil has quit IRC | 20:41 | |
| patrickeast | tbarron: i think all arrays in the field that customers have its 500 as the limit | 20:41 |
| patrickeast | tbarron: relatively soon with a Purity update it goes to 5k | 20:42 |
| patrickeast | tbarron: more specifically 500 vols for Purity <= v4.7.x | 20:43 |
| *** rlrossi__ has quit IRC | 20:44 | |
| tbarron | patrickeast: is there a way to see if he or the customer can be storage admin on the array? | 20:46 |
| tbarron | patrickeast: or is it best just to look at the version of Purity and figure 500 or 5k? | 20:46 |
| tbarron | then see how many vols cinder has allocated and find out if anyone else could be consuming them? | 20:47 |
| tbarron | patrickeast: oh, i see: all in the field right now have 500 vols, that's pretty easy then | 20:47 |
| patrickeast | tbarron: easiest is to look at the version, but yea unless they're very special its 500 | 20:48 |
| patrickeast | tbarron: typically the issue is that there are volumes in the "pending eradication" state | 20:48 |
| patrickeast | tbarron: i added a config option to mitaka to eradicate immediately, but as a "feature" of the array deleted volumes are available to be restored for 24 hours | 20:49 |
| patrickeast | tbarron: in that pending state they still count as one of the volumes | 20:49 |
| *** alkhodos has quit IRC | 20:50 | |
| tbarron | patrickeast: ok, good info, i'll have him find out if they very recently deleted a bunch and don't see them freed yet | 20:51 |
| patrickeast | https://www.irccloud.com/pastebin/RTImO15V/ | 20:52 |
| patrickeast | tbarron: they can use something like that to see ^ | 20:52 |
| patrickeast | tbarron: the san api and api token can be grabbed from cinder.conf | 20:52 |
| tbarron | patrickeast: ++, ty | 20:53 |
| smcginnis | Forgot to mention in the IRC meeting this week - I'm travelling tomorrow through Monday. | 20:55 |
| smcginnis | So I might not be around as much, but I should be able to get back to any pings, just delayed. | 20:55 |
| smcginnis | Just an FYI | 20:55 |
| *** rlrossit has joined #openstack-cinder | 20:55 | |
| *** crose has quit IRC | 20:56 | |
| *** baumann has quit IRC | 20:58 | |
| *** gouthamr-afk is now known as gouthamr | 20:58 | |
| jungleboyj | smcginnis: You can't leave us! | 20:59 |
| smcginnis | jungleboyj: Hah! | 21:00 |
| smcginnis | jungleboyj: How was your "lunch". Your typing seems to be OK. ;) | 21:00 |
| *** baumann has joined #openstack-cinder | 21:00 | |
| *** xyang has joined #openstack-cinder | 21:00 | |
| jungleboyj | smcginnis: I have actually solved a number of problem that have been outstanding this afternoon. | 21:00 |
| jungleboyj | Need to do 'lunch' more often. | 21:01 |
| *** baumann has quit IRC | 21:01 | |
| smcginnis | jungleboyj: Nice! | 21:01 |
| *** baumann has joined #openstack-cinder | 21:01 | |
| jungleboyj | smcginnis: Hoping to have a rebased snapshot patch for NFS pushed up yet this week. | 21:02 |
| *** diablo_rojo has quit IRC | 21:02 | |
| smcginnis | jungleboyj: Ooh, nice. Glad to see progress there. | 21:03 |
| jungleboyj | smcginnis: Yeah, hoping no one else has started working on that. | 21:03 |
| *** julim has quit IRC | 21:04 | |
| *** jgregor has quit IRC | 21:04 | |
| smcginnis | jungleboyj: Unfortunately for you - I doubt it. :) | 21:04 |
| nikeshm | xyang1: if you get time can you review this replication failback support patch for kaminario driver https://review.openstack.org/#/c/356475/ | 21:04 |
| *** e0ne has quit IRC | 21:04 | |
| jungleboyj | smcginnis: I am learning a lot in the process, so it is all good. | 21:05 |
| xyang | nikeshm: ok | 21:05 |
| *** Apoorva_ has quit IRC | 21:08 | |
| *** Apoorva has joined #openstack-cinder | 21:08 | |
| *** leeantho_ has quit IRC | 21:10 | |
| *** crose has joined #openstack-cinder | 21:11 | |
| *** cdelatte has quit IRC | 21:16 | |
| *** catintheroof has quit IRC | 21:21 | |
| *** baumann has quit IRC | 21:23 | |
| *** crose has quit IRC | 21:23 | |
| *** edmondsw has quit IRC | 21:25 | |
| *** mriedem has left #openstack-cinder | 21:25 | |
| *** mriedem has joined #openstack-cinder | 21:26 | |
| *** leeantho has joined #openstack-cinder | 21:27 | |
| Swanson | Simple change here (all dell internal stuff): https://review.openstack.org/#/c/350262/ | 21:27 |
| *** rlrossit has quit IRC | 21:31 | |
| Swanson | This one is pretty basic, too.... https://review.openstack.org/356687 | 21:33 |
| *** pauloewerton has quit IRC | 21:34 | |
| Swanson | scottda, Thanks! | 21:37 |
| scottda | np. there's the other as well. | 21:37 |
| smcginnis | scottda: Don't be too nice to him. He might start to expect it. | 21:38 |
| smcginnis | :P | 21:38 |
| Swanson | xyang, also thanks! | 21:39 |
| xyang | Swanson: np | 21:39 |
| *** jungleboyj has quit IRC | 21:44 | |
| *** lprice has quit IRC | 21:45 | |
| *** lpetrut has quit IRC | 21:55 | |
| *** mriedem has quit IRC | 21:58 | |
| *** tbarron is now known as tbarron|afk | 22:02 | |
| *** xyang1 has quit IRC | 22:14 | |
| *** sumit7990 has quit IRC | 22:15 | |
| *** kwathore_ has quit IRC | 22:17 | |
| *** tpsilva has quit IRC | 22:18 | |
| *** julim has joined #openstack-cinder | 22:19 | |
| *** xyang has quit IRC | 22:20 | |
| *** alyson_ has quit IRC | 22:23 | |
| *** erlon has quit IRC | 22:27 | |
| *** xinli has quit IRC | 22:30 | |
| *** ntpttr has quit IRC | 22:31 | |
| *** ducttape_ has quit IRC | 22:32 | |
| *** leeantho has quit IRC | 22:34 | |
| *** ntpttr has joined #openstack-cinder | 22:36 | |
| *** ametts has quit IRC | 22:51 | |
| *** wxy__ has quit IRC | 22:53 | |
| *** hemna is now known as hemnafk | 22:54 | |
| openstackgerrit | Walter A. Boring IV (hemna) proposed openstack/cinder: Add warning message for None volume type set https://review.openstack.org/348610 | 22:54 |
| *** Apoorva_ has joined #openstack-cinder | 23:00 | |
| *** haplo37__ has quit IRC | 23:03 | |
| *** Apoorva has quit IRC | 23:03 | |
| *** sdake_ has joined #openstack-cinder | 23:12 | |
| *** sdake has quit IRC | 23:13 | |
| *** sdague has quit IRC | 23:14 | |
| *** jordanP has quit IRC | 23:16 | |
| *** chlong has quit IRC | 23:20 | |
| openstackgerrit | Merged openstack/cinder: Use min attribute from oslo_config https://review.openstack.org/356388 | 23:24 |
| openstackgerrit | Merged openstack/cinder: Updated from global requirements https://review.openstack.org/356867 | 23:25 |
| *** rcernin has quit IRC | 23:32 | |
| *** xenogear has quit IRC | 23:40 | |
| patrickeast | ew just had a c-sch segfault in the middle of tempest... i feel like that isn't supposed to happen with python | 23:48 |
| *** mriedem has joined #openstack-cinder | 23:55 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!