| *** ociuhandu has joined #openstack-cinder | 00:04 | |
| *** yangyapeng has quit IRC | 00:06 | |
| *** ducttape_ has joined #openstack-cinder | 00:08 | |
| *** alonma has joined #openstack-cinder | 00:14 | |
| *** baojg has joined #openstack-cinder | 00:16 | |
| *** alonma has quit IRC | 00:18 | |
| *** ducttape_ has quit IRC | 00:19 | |
| *** ducttape_ has joined #openstack-cinder | 00:22 | |
| *** cknight has joined #openstack-cinder | 00:22 | |
| *** lixiaoy1 has joined #openstack-cinder | 00:24 | |
| *** enriquetaso has quit IRC | 00:24 | |
| *** ducttape_ has quit IRC | 00:32 | |
| *** kaisers1 has joined #openstack-cinder | 00:35 | |
| *** bswartz has joined #openstack-cinder | 00:36 | |
| *** kaisers has quit IRC | 00:38 | |
| *** enriquetaso has joined #openstack-cinder | 00:39 | |
| *** baojg has quit IRC | 00:42 | |
| *** laughterwym has joined #openstack-cinder | 00:44 | |
| *** aabashkin_ has joined #openstack-cinder | 00:47 | |
| *** GB21 has joined #openstack-cinder | 00:47 | |
| *** laughterwym has quit IRC | 00:48 | |
| *** AlexeyAbashkin has quit IRC | 00:48 | |
| *** laughterwym has joined #openstack-cinder | 00:48 | |
| *** alonma has joined #openstack-cinder | 00:55 | |
| *** alonma has quit IRC | 01:00 | |
| *** cheneydc has joined #openstack-cinder | 01:01 | |
| *** fernnest_ has joined #openstack-cinder | 01:03 | |
| *** zhangguoqing has joined #openstack-cinder | 01:05 | |
| *** fernnest__ has quit IRC | 01:07 | |
| *** fernnest__ has joined #openstack-cinder | 01:07 | |
| *** fernnest_ has quit IRC | 01:07 | |
| *** chlong_ has joined #openstack-cinder | 01:12 | |
| *** chlong has joined #openstack-cinder | 01:13 | |
| *** yangyapeng has joined #openstack-cinder | 01:20 | |
| *** cheneydc has quit IRC | 01:20 | |
| *** ducttape_ has joined #openstack-cinder | 01:33 | |
| *** mriedem has quit IRC | 01:36 | |
| *** alonma has joined #openstack-cinder | 01:37 | |
| *** ducttape_ has quit IRC | 01:38 | |
| *** alonma has quit IRC | 01:41 | |
| *** enriquetaso has quit IRC | 01:42 | |
| openstackgerrit | Yingzhe Zeng proposed openstack/cinder: Huawei driver supports specifying copy speed https://review.openstack.org/368516 | 01:42 |
|---|---|---|
| openstackgerrit | xianming.mao proposed openstack/cinder: Update the oslo.db version in requirements.txt. https://review.openstack.org/368517 | 01:52 |
| *** EinstCrazy has joined #openstack-cinder | 01:57 | |
| *** laughterwym has quit IRC | 01:59 | |
| *** jamielennox is now known as jamielennox|away | 02:00 | |
| *** laughterwym has joined #openstack-cinder | 02:02 | |
| *** cknight has quit IRC | 02:03 | |
| *** jamielennox|away is now known as jamielennox | 02:04 | |
| *** akshai_ has quit IRC | 02:09 | |
| *** lixiaoy1 has quit IRC | 02:14 | |
| *** lixiaoy1 has joined #openstack-cinder | 02:14 | |
| *** alonma has joined #openstack-cinder | 02:18 | |
| *** lifeless_ has quit IRC | 02:19 | |
| *** alonma has quit IRC | 02:23 | |
| *** laughterwym has quit IRC | 02:26 | |
| *** laughterwym has joined #openstack-cinder | 02:27 | |
| *** laughterwym has quit IRC | 02:32 | |
| *** laughterwym has joined #openstack-cinder | 02:32 | |
| *** laughterwym has quit IRC | 02:32 | |
| *** laughterwym has joined #openstack-cinder | 02:33 | |
| *** lifeless has joined #openstack-cinder | 02:34 | |
| *** ducttape_ has joined #openstack-cinder | 02:34 | |
| *** laughterwym has quit IRC | 02:37 | |
| *** ducttape_ has quit IRC | 02:39 | |
| *** cknight has joined #openstack-cinder | 02:46 | |
| *** laughterwym has joined #openstack-cinder | 02:50 | |
| *** laughterwym has quit IRC | 02:51 | |
| *** laughterwym has joined #openstack-cinder | 02:52 | |
| *** bardia has joined #openstack-cinder | 02:55 | |
| *** alonma has joined #openstack-cinder | 03:00 | |
| *** alonma has quit IRC | 03:05 | |
| *** ducttape_ has joined #openstack-cinder | 03:22 | |
| *** ducttape_ has quit IRC | 03:22 | |
| *** ociuhandu has quit IRC | 03:25 | |
| *** lixiaoy1 has quit IRC | 03:35 | |
| *** Dinesh_Bhor has joined #openstack-cinder | 03:39 | |
| *** watanabe_isao has joined #openstack-cinder | 03:40 | |
| *** lixiaoy1 has joined #openstack-cinder | 03:42 | |
| *** alonma has joined #openstack-cinder | 03:42 | |
| *** baojg has joined #openstack-cinder | 03:43 | |
| *** cknight has quit IRC | 03:45 | |
| *** alonma has quit IRC | 03:47 | |
| *** baojg has quit IRC | 03:47 | |
| *** vladani_ has joined #openstack-cinder | 03:49 | |
| *** vladani has quit IRC | 03:52 | |
| *** bardia has quit IRC | 03:54 | |
| *** bhagyashri has joined #openstack-cinder | 04:02 | |
| *** bhagyashri has quit IRC | 04:16 | |
| *** gouthamr has joined #openstack-cinder | 04:19 | |
| *** jdurgin has joined #openstack-cinder | 04:20 | |
| *** ducttape_ has joined #openstack-cinder | 04:23 | |
| *** alonma has joined #openstack-cinder | 04:23 | |
| *** alonma has quit IRC | 04:28 | |
| *** ducttape_ has quit IRC | 04:28 | |
| *** martyt has joined #openstack-cinder | 04:30 | |
| *** lixiaoy1 has quit IRC | 04:32 | |
| *** liuyulong_ is now known as liuyulong | 04:34 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/cinder: Adjust doc about threading https://review.openstack.org/367768 | 04:36 |
| *** pdeore has joined #openstack-cinder | 04:37 | |
| *** kaisers has joined #openstack-cinder | 04:39 | |
| *** pbandark has joined #openstack-cinder | 04:41 | |
| *** pgadiya has joined #openstack-cinder | 04:50 | |
| *** adisky has joined #openstack-cinder | 04:51 | |
| *** salv-orlando has joined #openstack-cinder | 04:52 | |
| *** martyt has quit IRC | 04:57 | |
| openstackgerrit | Karthik Prabhu Vinod proposed openstack/python-cinderclient: Showing the metadata readonly value as a separate field https://review.openstack.org/332480 | 04:57 |
| *** lpetrut has joined #openstack-cinder | 05:02 | |
| *** chlong_ has quit IRC | 05:03 | |
| *** chlong has quit IRC | 05:03 | |
| *** GB21 has quit IRC | 05:04 | |
| *** alonma has joined #openstack-cinder | 05:05 | |
| *** gcb has joined #openstack-cinder | 05:07 | |
| *** laughte__ has joined #openstack-cinder | 05:07 | |
| *** alonma has quit IRC | 05:09 | |
| *** laughterwym has quit IRC | 05:11 | |
| *** lkuchlan has joined #openstack-cinder | 05:13 | |
| *** lixiaoy1 has joined #openstack-cinder | 05:14 | |
| *** chlong_ has joined #openstack-cinder | 05:17 | |
| *** GB21 has joined #openstack-cinder | 05:18 | |
| *** chlong has joined #openstack-cinder | 05:21 | |
| *** laughte__ has quit IRC | 05:22 | |
| *** laughterwym has joined #openstack-cinder | 05:22 | |
| *** ducttape_ has joined #openstack-cinder | 05:24 | |
| *** chlong has quit IRC | 05:25 | |
| *** ducttape_ has quit IRC | 05:29 | |
| *** Lee1092 has joined #openstack-cinder | 05:32 | |
| *** rcernin has joined #openstack-cinder | 05:33 | |
| *** laughter_ has joined #openstack-cinder | 05:34 | |
| *** laughter_ has joined #openstack-cinder | 05:35 | |
| *** laughterwym has quit IRC | 05:38 | |
| *** alonma has joined #openstack-cinder | 05:47 | |
| *** lpetrut has quit IRC | 05:50 | |
| *** alonma has quit IRC | 05:52 | |
| *** snarwade has quit IRC | 05:54 | |
| *** ChubYann has quit IRC | 05:55 | |
| *** houming has joined #openstack-cinder | 05:57 | |
| *** snarwade has joined #openstack-cinder | 06:00 | |
| *** lpetrut has joined #openstack-cinder | 06:03 | |
| *** chlong_ has quit IRC | 06:05 | |
| *** nherciu has joined #openstack-cinder | 06:19 | |
| *** lkuchlan has quit IRC | 06:20 | |
| *** mjura has joined #openstack-cinder | 06:21 | |
| *** coolsvap has joined #openstack-cinder | 06:21 | |
| *** ducttape_ has joined #openstack-cinder | 06:25 | |
| *** mjura has quit IRC | 06:26 | |
| *** lpetrut has quit IRC | 06:27 | |
| *** mjura has joined #openstack-cinder | 06:27 | |
| *** alonma has joined #openstack-cinder | 06:28 | |
| *** ducttape_ has quit IRC | 06:29 | |
| *** lpetrut has joined #openstack-cinder | 06:31 | |
| *** alonma has quit IRC | 06:33 | |
| *** ircuser-1 has quit IRC | 06:41 | |
| *** gaurangt has joined #openstack-cinder | 06:44 | |
| *** nkrinner_afk is now known as nkrinner | 06:45 | |
| *** pcaruana has joined #openstack-cinder | 06:49 | |
| openstackgerrit | xiaoqin proposed openstack/cinder: Storwize: remove SCSI LUN ID for volume attaching https://review.openstack.org/360394 | 06:54 |
| *** tesseract- has joined #openstack-cinder | 07:00 | |
| *** arecknag has joined #openstack-cinder | 07:01 | |
| *** gouthamr has quit IRC | 07:01 | |
| *** gouthamr has joined #openstack-cinder | 07:02 | |
| *** gouthamr has quit IRC | 07:03 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/cinder: Imported Translations from Zanata https://review.openstack.org/368584 | 07:04 |
| *** lpetrut has quit IRC | 07:07 | |
| *** alonma has joined #openstack-cinder | 07:10 | |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Don't put clients everywhere https://review.openstack.org/347268 | 07:11 |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Make calculation use more bits https://review.openstack.org/347269 | 07:11 |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Use the ClientHelper class to reduce amount of code https://review.openstack.org/347270 | 07:11 |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Avoid an extra resize later on https://review.openstack.org/347271 | 07:11 |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Do batch delivery https://review.openstack.org/347272 | 07:11 |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Report an error if a snapshot has vanished https://review.openstack.org/347273 | 07:11 |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Avoid duplicated volumes from the error path https://review.openstack.org/347274 | 07:11 |
| *** alonma has quit IRC | 07:15 | |
| *** ducttape_ has joined #openstack-cinder | 07:26 | |
| *** liuyulong has quit IRC | 07:29 | |
| *** ducttape_ has quit IRC | 07:31 | |
| *** liuyulong has joined #openstack-cinder | 07:31 | |
| *** akapil has joined #openstack-cinder | 07:38 | |
| *** akapil has quit IRC | 07:38 | |
| *** akapil has joined #openstack-cinder | 07:39 | |
| *** akapil has quit IRC | 07:43 | |
| *** akapil has joined #openstack-cinder | 07:43 | |
| openstackgerrit | Cao ShuFeng proposed openstack/cinder: Save volume_type/group_type uuid into db when creating group https://review.openstack.org/368610 | 07:49 |
| openstackgerrit | xiaoqin proposed openstack/cinder: Storwize: remove SCSI LUN ID for volume attaching https://review.openstack.org/360394 | 07:50 |
| *** alonma has joined #openstack-cinder | 07:51 | |
| *** alonma has quit IRC | 07:56 | |
| openstackgerrit | Dao Cong Tien proposed openstack/cinder: Stop using mox in unit/api/v2/ tests https://review.openstack.org/356347 | 07:57 |
| *** martyt has joined #openstack-cinder | 07:58 | |
| *** akapil_ has joined #openstack-cinder | 07:59 | |
| *** akapil_ has quit IRC | 07:59 | |
| *** akapil_ has joined #openstack-cinder | 07:59 | |
| *** zzzeek has quit IRC | 08:00 | |
| *** zzzeek has joined #openstack-cinder | 08:00 | |
| *** akapil has quit IRC | 08:02 | |
| *** martyt has quit IRC | 08:02 | |
| *** andymaier has joined #openstack-cinder | 08:03 | |
| *** andymaier has quit IRC | 08:04 | |
| *** aspiers_ is now known as aspiers | 08:06 | |
| *** aabashkin_ is now known as AlexeyAbashkin | 08:20 | |
| *** AlexeyAbashkin has quit IRC | 08:22 | |
| *** ducttape_ has joined #openstack-cinder | 08:27 | |
| *** salv-orl_ has joined #openstack-cinder | 08:31 | |
| *** ducttape_ has quit IRC | 08:31 | |
| *** alonma has joined #openstack-cinder | 08:33 | |
| *** salv-orlando has quit IRC | 08:34 | |
| flip214 | geguileo_PTO: ping about https://review.openstack.org/#/c/347271, please? | 08:35 |
| *** bhagyashri has joined #openstack-cinder | 08:36 | |
| *** alonma has quit IRC | 08:38 | |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Avoid an extra resize later on https://review.openstack.org/347271 | 08:39 |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Do batch delivery https://review.openstack.org/347272 | 08:39 |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Report an error if a snapshot has vanished https://review.openstack.org/347273 | 08:39 |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Avoid duplicated volumes from the error path https://review.openstack.org/347274 | 08:39 |
| *** zengyingzhe has joined #openstack-cinder | 08:41 | |
| *** AlexeyAbashkin has joined #openstack-cinder | 08:42 | |
| *** lpetrut has joined #openstack-cinder | 08:42 | |
| openstackgerrit | Cao ShuFeng proposed openstack/python-cinderclient: Parse filter item "name" correctly for snapshot-list https://review.openstack.org/357594 | 08:47 |
| openstackgerrit | Cao ShuFeng proposed openstack/python-cinderclient: Revmove extra 'u' from cli output https://review.openstack.org/347804 | 08:47 |
| *** lixiaoy1 has quit IRC | 08:56 | |
| *** watanabe_isao has quit IRC | 08:57 | |
| *** akapil_ has quit IRC | 09:00 | |
| *** laughterwym has joined #openstack-cinder | 09:03 | |
| *** openstackgerrit has quit IRC | 09:04 | |
| *** hoonetorg has joined #openstack-cinder | 09:05 | |
| *** openstackgerrit has joined #openstack-cinder | 09:05 | |
| *** laughter_ has quit IRC | 09:07 | |
| *** laughterwym has quit IRC | 09:07 | |
| *** andymaier has joined #openstack-cinder | 09:12 | |
| *** andymaier has quit IRC | 09:13 | |
| *** alonma has joined #openstack-cinder | 09:15 | |
| *** zhugaoxiao has quit IRC | 09:16 | |
| *** zhugaoxiao has joined #openstack-cinder | 09:17 | |
| *** akapil has joined #openstack-cinder | 09:18 | |
| *** alonma has quit IRC | 09:19 | |
| *** ducttape_ has joined #openstack-cinder | 09:27 | |
| *** houming has quit IRC | 09:31 | |
| *** ducttape_ has quit IRC | 09:32 | |
| *** jdurgin has quit IRC | 09:37 | |
| *** alonma has joined #openstack-cinder | 09:50 | |
| openstackgerrit | Yingzhe Zeng proposed openstack/cinder: Huawei driver supports specifying copy speed https://review.openstack.org/368516 | 09:50 |
| *** akapil has quit IRC | 09:52 | |
| *** vladani__ has joined #openstack-cinder | 10:05 | |
| *** vladani_ has quit IRC | 10:05 | |
| *** laughterwym has joined #openstack-cinder | 10:05 | |
| *** laughterwym has quit IRC | 10:10 | |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Avoid an extra resize later on https://review.openstack.org/347271 | 10:13 |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Do batch delivery https://review.openstack.org/347272 | 10:13 |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Report an error if a snapshot has vanished https://review.openstack.org/347273 | 10:13 |
| openstackgerrit | Philipp Marek proposed openstack/cinder: DRBD: Avoid duplicated volumes from the error path https://review.openstack.org/347274 | 10:13 |
| *** akapil has joined #openstack-cinder | 10:13 | |
| *** akapil_ has joined #openstack-cinder | 10:15 | |
| *** akapil has quit IRC | 10:19 | |
| *** kaisers has quit IRC | 10:20 | |
| *** markvoelker has joined #openstack-cinder | 10:22 | |
| *** markvoelker has quit IRC | 10:26 | |
| *** e0ne has joined #openstack-cinder | 10:27 | |
| *** kaisers has joined #openstack-cinder | 10:33 | |
| *** GB21 has quit IRC | 10:40 | |
| *** yangyapeng has quit IRC | 10:40 | |
| *** EinstCrazy has quit IRC | 10:42 | |
| *** diogogmt_ has joined #openstack-cinder | 10:52 | |
| *** diogogmt has quit IRC | 10:53 | |
| *** diogogmt_ is now known as diogogmt | 10:53 | |
| *** vladani_ has joined #openstack-cinder | 11:02 | |
| *** vladani__ has quit IRC | 11:02 | |
| *** hoonetorg has quit IRC | 11:03 | |
| *** kaisers has quit IRC | 11:06 | |
| openstackgerrit | xiaoqin proposed openstack/cinder: Storwize: do the correct update after failover https://review.openstack.org/292570 | 11:07 |
| *** cdelatte has joined #openstack-cinder | 11:10 | |
| *** GB21 has joined #openstack-cinder | 11:10 | |
| openstackgerrit | Szymon Borkowski proposed openstack/cinder: Convert backup_device to OVO https://review.openstack.org/329332 | 11:16 |
| *** alonma has quit IRC | 11:27 | |
| *** ducttape_ has joined #openstack-cinder | 11:29 | |
| *** ducttape_ has quit IRC | 11:33 | |
| *** sdague has joined #openstack-cinder | 11:34 | |
| *** ygl has joined #openstack-cinder | 11:35 | |
| ygl | hi all | 11:36 |
| ygl | i am unable to create volumes in cinder | 11:36 |
| ygl | anyone here ? | 11:40 |
| *** sandanar has joined #openstack-cinder | 11:42 | |
| *** ccesario has joined #openstack-cinder | 11:45 | |
| *** alonma has joined #openstack-cinder | 11:48 | |
| *** alonma has quit IRC | 11:53 | |
| *** ygl has quit IRC | 11:54 | |
| *** gaurangt has left #openstack-cinder | 11:55 | |
| *** rooneym has joined #openstack-cinder | 11:55 | |
| *** eharney has quit IRC | 11:57 | |
| openstackgerrit | boden proposed openstack/cinder: Replace retrying with tenacity https://review.openstack.org/368235 | 12:01 |
| *** sandanar has quit IRC | 12:01 | |
| *** markvoelker has joined #openstack-cinder | 12:02 | |
| *** pdeore has quit IRC | 12:05 | |
| *** cknight has joined #openstack-cinder | 12:05 | |
| e0ne | ygl: hi. do you have any errors in cinder-scheduler or cinder-volume logs? | 12:05 |
| *** alonma has joined #openstack-cinder | 12:05 | |
| *** avishay has joined #openstack-cinder | 12:07 | |
| *** alyson_ has joined #openstack-cinder | 12:09 | |
| openstackgerrit | Matan Sabag proposed openstack/cinder: Added default for ScaleIO driver max_over_subscription https://review.openstack.org/368358 | 12:10 |
| openstackgerrit | Matan Sabag proposed openstack/cinder: Added default for ScaleIO driver max_over_subscription https://review.openstack.org/368358 | 12:12 |
| *** adrianofr has joined #openstack-cinder | 12:15 | |
| *** ducttape_ has joined #openstack-cinder | 12:19 | |
| *** pbandark has quit IRC | 12:21 | |
| *** coolsvap is now known as _coolsvap_ | 12:22 | |
| *** akapil_ has quit IRC | 12:24 | |
| *** akapil has joined #openstack-cinder | 12:25 | |
| *** porrua has joined #openstack-cinder | 12:26 | |
| *** kvidvans has joined #openstack-cinder | 12:27 | |
| *** porrua_ has joined #openstack-cinder | 12:28 | |
| *** porrua has quit IRC | 12:31 | |
| *** ducttape_ has quit IRC | 12:44 | |
| *** pbandark has joined #openstack-cinder | 12:49 | |
| openstackgerrit | Merged openstack/cinder: Imported Translations from Zanata https://review.openstack.org/368584 | 12:51 |
| *** pauloewerton has joined #openstack-cinder | 12:54 | |
| *** pgadiya has quit IRC | 12:56 | |
| *** zhangguoqing has quit IRC | 12:57 | |
| *** jungleboyj has quit IRC | 13:01 | |
| *** rlrossit has joined #openstack-cinder | 13:08 | |
| *** eharney has joined #openstack-cinder | 13:15 | |
| *** akerr has joined #openstack-cinder | 13:16 | |
| *** openstackgerrit has quit IRC | 13:19 | |
| *** openstackgerrit has joined #openstack-cinder | 13:20 | |
| *** crose has quit IRC | 13:21 | |
| *** dustins has joined #openstack-cinder | 13:22 | |
| *** kaisers has joined #openstack-cinder | 13:25 | |
| bhagyashri | scottda: Hi, do you have free time? to discuss regarding "os-force_detach" api. | 13:26 |
| *** ducttape_ has joined #openstack-cinder | 13:26 | |
| *** akshai has joined #openstack-cinder | 13:27 | |
| *** lprice has joined #openstack-cinder | 13:27 | |
| *** cknight has quit IRC | 13:30 | |
| scottda | bhagyashri: No, but I will in a few hours. | 13:31 |
| *** EinstCrazy has joined #openstack-cinder | 13:31 | |
| *** cknight has joined #openstack-cinder | 13:31 | |
| *** lprice has quit IRC | 13:32 | |
| bhagyashri | scottda: ok. | 13:32 |
| *** enriquetaso has joined #openstack-cinder | 13:36 | |
| *** takedakn has joined #openstack-cinder | 13:36 | |
| *** porrua_ has quit IRC | 13:38 | |
| *** wN has quit IRC | 13:39 | |
| *** yangyapeng has joined #openstack-cinder | 13:39 | |
| *** ducttape_ has quit IRC | 13:40 | |
| *** lprice has joined #openstack-cinder | 13:40 | |
| *** wN has joined #openstack-cinder | 13:44 | |
| *** wN has quit IRC | 13:44 | |
| *** wN has joined #openstack-cinder | 13:44 | |
| *** rlrossit has quit IRC | 13:45 | |
| *** ociuhandu has joined #openstack-cinder | 13:46 | |
| e0ne | bhagyashri, scottda: please, ping me roo about force detach. I'm interesting in it too | 13:46 |
| *** lprice has quit IRC | 13:53 | |
| *** mriedem has joined #openstack-cinder | 13:53 | |
| *** ametts has joined #openstack-cinder | 13:54 | |
| *** jgregor has joined #openstack-cinder | 13:55 | |
| *** diablo_rojo has joined #openstack-cinder | 13:59 | |
| *** lcurtis has quit IRC | 14:02 | |
| *** tongli has joined #openstack-cinder | 14:03 | |
| bhagyashri | e0ne: Hi, I want to know that is the "connector" parameter is mandatory in request body for "os-force_detach" api? | 14:04 |
| *** jungleboyj has joined #openstack-cinder | 14:04 | |
| bhagyashri | e0ne: because I can see that in api-ref it is shown as mandatory but in case of "lvm" driver it is not mandatory and not used anywhere in the code. | 14:08 |
| *** ducttape_ has joined #openstack-cinder | 14:12 | |
| *** porrua has joined #openstack-cinder | 14:14 | |
| *** laughterwym has joined #openstack-cinder | 14:20 | |
| *** pbandark has quit IRC | 14:21 | |
| *** nherciu has quit IRC | 14:23 | |
| *** bhagyashri has quit IRC | 14:25 | |
| *** bhagyashri has joined #openstack-cinder | 14:25 | |
| *** kaisers has quit IRC | 14:25 | |
| *** bhagyashri has quit IRC | 14:25 | |
| *** bhagyashri has joined #openstack-cinder | 14:26 | |
| *** bhagyashri has quit IRC | 14:26 | |
| *** baojg has joined #openstack-cinder | 14:27 | |
| *** lprice has joined #openstack-cinder | 14:29 | |
| e0ne | bhagyashri: afair, we were going to make this param optionally | 14:29 |
| *** salv-orlando has joined #openstack-cinder | 14:30 | |
| *** salv-orl_ has quit IRC | 14:33 | |
| *** smile-luobin has quit IRC | 14:33 | |
| *** smile-luobin has joined #openstack-cinder | 14:35 | |
| *** akshai has quit IRC | 14:37 | |
| *** laughter_ has joined #openstack-cinder | 14:37 | |
| *** mvk has quit IRC | 14:37 | |
| *** akshai has joined #openstack-cinder | 14:38 | |
| *** hoonetorg has joined #openstack-cinder | 14:38 | |
| *** laughterwym has quit IRC | 14:40 | |
| *** laughter_ has quit IRC | 14:41 | |
| *** laughterwym has joined #openstack-cinder | 14:42 | |
| *** rlrossit has joined #openstack-cinder | 14:42 | |
| *** arecknag has quit IRC | 14:43 | |
| *** pbandark has joined #openstack-cinder | 14:46 | |
| *** _ducttape_ has joined #openstack-cinder | 14:56 | |
| openstackgerrit | Merged openstack/cinder: fix the %{} when string formating https://review.openstack.org/367131 | 14:58 |
| *** ducttape_ has quit IRC | 15:00 | |
| *** fernnest__ has quit IRC | 15:00 | |
| *** mjura has quit IRC | 15:03 | |
| *** jgriffith has quit IRC | 15:03 | |
| *** gouthamr has joined #openstack-cinder | 15:09 | |
| *** zul has joined #openstack-cinder | 15:12 | |
| *** pbandark has quit IRC | 15:15 | |
| *** TommyLikeHu has quit IRC | 15:15 | |
| *** TommyLikeHu has joined #openstack-cinder | 15:15 | |
| *** laughter_ has joined #openstack-cinder | 15:17 | |
| *** jgriffith has joined #openstack-cinder | 15:17 | |
| *** hemnafk has quit IRC | 15:18 | |
| *** nkrinner is now known as nkrinner_afk | 15:20 | |
| *** laughterwym has quit IRC | 15:20 | |
| *** baojg has quit IRC | 15:20 | |
| *** laughter_ has quit IRC | 15:21 | |
| *** gaurangt has joined #openstack-cinder | 15:22 | |
| *** xinli has joined #openstack-cinder | 15:23 | |
| *** harlowja_at_home has joined #openstack-cinder | 15:27 | |
| gaurangt | hi All, I'm trying out a PoC for replication with GPFS driver. One of the challenge I'm facing is - After primary fails and we fail over to secondary - the volumes created from primary are no more manageable from secondary unless we make changes in the database. Can we do something in the failover_host function in the driver so as to change the host attribute for the volumes are changed and we would be able to manage those volumes from secondary. | 15:28 |
| gaurangt | Thoughts, suggestions? | 15:28 |
| scottda | bhagyashri You will get an exception if you do not pass in a connector: https://github.com/openstack/cinder/blob/master/cinder/api/contrib/admin_actions.py#L194 | 15:29 |
| *** hemna has joined #openstack-cinder | 15:29 | |
| *** laughterwym has joined #openstack-cinder | 15:29 | |
| *** harlowja_still_a has joined #openstack-cinder | 15:29 | |
| *** laughterwym has quit IRC | 15:29 | |
| *** harlowja_still_a has quit IRC | 15:30 | |
| *** laughterwym has joined #openstack-cinder | 15:30 | |
| *** laughterwym has quit IRC | 15:30 | |
| *** adisky has quit IRC | 15:31 | |
| *** EinstCrazy has quit IRC | 15:33 | |
| *** ircuser-1 has joined #openstack-cinder | 15:35 | |
| *** chris_morrell has joined #openstack-cinder | 15:39 | |
| *** kaisers has joined #openstack-cinder | 15:40 | |
| *** e0ne has quit IRC | 15:41 | |
| jgriffith | gaurangt: are the volumes in question "replication enabled"? | 15:45 |
| jgriffith | gaurangt: if they are it's a bug, if they're not, it's an expected behavior | 15:45 |
| *** avishay has quit IRC | 15:45 | |
| *** chris_morrell has quit IRC | 15:46 | |
| gaurangt | jgriffith: yes, the volumes here are "replication enabled" | 15:47 |
| *** chris_morrell has joined #openstack-cinder | 15:48 | |
| *** nherciu has joined #openstack-cinder | 15:48 | |
| gaurangt | jgriffith: I've tried to change the host from model_update of the volumes and it seems to have working fine for me. | 15:49 |
| gaurangt | Is this a good way to change the host attribute? | 15:49 |
| jgriffith | gaurangt: I might not be understanding completely.... | 15:50 |
| jgriffith | gaurangt: so the way it *should* work.. is that all the volumes that are | 15:50 |
| jgriffith | "replication enabled" should auto failover to the secondary | 15:50 |
| jgriffith | gaurangt: when that happens, that also means that the volumes are now on the | 15:50 |
| jgriffith | secondary and manageable from there | 15:50 |
| patrickeast | jgriffith: +1 | 15:51 |
| patrickeast | when it fails over the host doesn't change in cinder, just the like active target id or whatever new field we added | 15:51 |
| patrickeast | the driver needs to know how to access both backends and do the right thing | 15:51 |
| *** rcernin has quit IRC | 15:53 | |
| *** xinli has quit IRC | 15:53 | |
| *** xinli has joined #openstack-cinder | 15:54 | |
| gaurangt | the condition in case of shared filesystem like GPFS is - we're running cinder volume service on two nodes each being the replication target of one another. When one of the host (node) fails, we can still access the volumes (sparse file on the filesystem) from secondary. But since the host attribute of that volume is not changed, the management of the volume is impacted. | 15:54 |
| gaurangt | I hope this clears what I am saying | 15:54 |
| *** pcaruana has quit IRC | 15:56 | |
| *** diogogmt has quit IRC | 15:57 | |
| patrickeast | gaurangt: yea, i think i understand | 15:59 |
| *** avishay has joined #openstack-cinder | 15:59 | |
| patrickeast | gaurangt: so in that setup you have two different cinder backends? or running one with like A-A or A-P HA? | 16:00 |
| *** tongli has quit IRC | 16:01 | |
| gaurangt | the filesystem is mounted on both the nodes and cinder volume service is running on both the hosts in A-A model | 16:01 |
| *** Apoorva has joined #openstack-cinder | 16:02 | |
| patrickeast | i'm not sure i understand the issue then, if they are both mounting the same shared filesystem why does it matter which one handles there request... or if they failover? | 16:02 |
| patrickeast | i don't think thats really the use-case that the cinder v2.1 replication is aiming for | 16:03 |
| patrickeast | its more for like one or more distinct backends that can't otherwise transparently handle a failure of the c-vol backend, does that make sense? | 16:04 |
| *** zul has quit IRC | 16:04 | |
| *** lpetrut has quit IRC | 16:04 | |
| *** thumpba has joined #openstack-cinder | 16:06 | |
| *** chris_morrell has quit IRC | 16:07 | |
| hemna | morning | 16:08 |
| hemna | for the record, the flu still sucks. | 16:08 |
| patrickeast | smcginnis: Swanson: was that failure on the dell sc iscsi job related to the change https://review.openstack.org/#/c/365129/ ? doesn't really look like it... but figured I'd check with the experts | 16:08 |
| *** gaurangt has quit IRC | 16:08 | |
| *** ChubYann has joined #openstack-cinder | 16:08 | |
| patrickeast | hemna: ouch | 16:08 |
| patrickeast | guess its time to re-up on flu shots | 16:09 |
| hemna | I'd been out since Friday. | 16:11 |
| hemna | I'm kinda back to human status today. | 16:11 |
| *** chris_morrell has joined #openstack-cinder | 16:11 | |
| smcginnis | hemna: Hope you feel better soon. What a pain. | 16:11 |
| Swanson | hemna, could be zika. | 16:11 |
| smcginnis | patrickeast: Doesn't look like it's related. I'll kick off a recheck of our CI. | 16:12 |
| hemna | yah, I have to cut down on the mosquitos | 16:12 |
| openstackgerrit | Walter A. Boring IV (hemna) proposed openstack/cinder: 3PAR use OSLO versioned Object for volume https://review.openstack.org/339147 | 16:13 |
| *** chris_morrell has quit IRC | 16:13 | |
| *** gaurangt has joined #openstack-cinder | 16:16 | |
| *** chris_morrell has joined #openstack-cinder | 16:17 | |
| gaurangt | patrickeast: Sorry, I got disconnected. Just re-joined. Did you say anything? | 16:17 |
| patrickeast | gaurangt: http://eavesdrop.openstack.org/irclogs/%23openstack-cinder/latest.log.html#t2016-09-12T16:02:56 | 16:18 |
| patrickeast | gaurangt: that was the last part | 16:18 |
| *** diogogmt has joined #openstack-cinder | 16:19 | |
| patrickeast | gaurangt: mainly, i'm not sure why with that kind of setup you would really even want to use the cinder replication if its a single backend deployed A-A | 16:19 |
| *** kvidvans has quit IRC | 16:20 | |
| *** avishay has quit IRC | 16:20 | |
| *** leeantho has joined #openstack-cinder | 16:21 | |
| openstackgerrit | xing-yang proposed openstack/cinder: Add functional tests for groups https://review.openstack.org/362584 | 16:22 |
| gaurangt | yeah.. but the backend is enabled with GPFS replication and we want to just expose it to the end user. | 16:23 |
| *** jgriffith_ has joined #openstack-cinder | 16:23 | |
| *** jgriffith has quit IRC | 16:25 | |
| patrickeast | gaurangt: expose what though? like, if you lose one of the sites of the clustered thingy you don't need to do anything as a cloud admin to make the "volumes" still accessable, right? | 16:27 |
| patrickeast | the mounted nfs share is still there, and the files are available, if i understand what gpfs replication does | 16:28 |
| *** lprice1 has joined #openstack-cinder | 16:28 | |
| gaurangt | right.. the volumes are accessible as files but not manageable. | 16:28 |
| patrickeast | ah ok, so there is something that needs to happen? | 16:29 |
| *** jgriffith_ has quit IRC | 16:29 | |
| *** rlrossit has quit IRC | 16:29 | |
| *** lprice has quit IRC | 16:29 | |
| patrickeast | how does that connect with the host attribute of the volume? | 16:29 |
| *** diablo_rojo has quit IRC | 16:29 | |
| * patrickeast should be taking notes | 16:29 | |
| *** rlrossit has joined #openstack-cinder | 16:30 | |
| *** porrua has quit IRC | 16:30 | |
| *** sdake_ has joined #openstack-cinder | 16:30 | |
| *** chris_morrell has quit IRC | 16:31 | |
| gaurangt | the node at one site which is running c-vol service is the host for a particular volume, right. If that fails, then the volume file is still accessible from other site, but we cannot do volume management as the host or c-vol service on that site is down. | 16:32 |
| *** krtaylor has quit IRC | 16:32 | |
| patrickeast | i thought you were running multiple c-vol services for the backend though? | 16:32 |
| *** kaisers has quit IRC | 16:33 | |
| patrickeast | "cinder volume service is running on both the hosts in A-A model" | 16:33 |
| patrickeast | that should keep the backend from going down, right? | 16:33 |
| patrickeast | or if not, i guess thats the bug | 16:33 |
| *** rlrossit has quit IRC | 16:34 | |
| *** rlrossit has joined #openstack-cinder | 16:35 | |
| *** rlrossit has quit IRC | 16:35 | |
| *** alonma has quit IRC | 16:35 | |
| *** rlrossit has joined #openstack-cinder | 16:35 | |
| *** harlowja_at_home has quit IRC | 16:35 | |
| *** jgriffith has joined #openstack-cinder | 16:35 | |
| openstackgerrit | Erlon R. Cruz proposed openstack/cinder: NFS snapshots https://review.openstack.org/147186 | 16:37 |
| *** takedakn has quit IRC | 16:37 | |
| gaurangt | cinder volume service is running on both the hosts and access the same filesystem. Each host will create it's own volumes. The point here I'm making is - when one of the host from one site goes down, the volume files are still accessible from other site ( because of gpfs replication and clustered filesystem ), but the management is not possible until we change the host attribute for that volume. So the nova I/Os should not ideally break, but the | 16:39 |
| *** jgregor has quit IRC | 16:41 | |
| *** jgregor has joined #openstack-cinder | 16:45 | |
| *** ametts has quit IRC | 16:45 | |
| *** avishay has joined #openstack-cinder | 16:45 | |
| erlon | jungleboyj: ping | 16:45 |
| patrickeast | gaurangt: ok, so maybe theres some confusion with the terminology we're throwing around. that sounds like you just have two cinder backends, not A-A c-vol's, but its an A-A storage backend | 16:46 |
| gaurangt | right | 16:46 |
| patrickeast | gaurangt: so like if you do 'cinder service-list' you see an entry for each one of those cinder volume nodes | 16:46 |
| patrickeast | gaurangt: so in that case when you failover you need to update the host entry for the volumes, but to your point you can't do that if the host is down | 16:46 |
| gaurangt | yes, I see two different entries for each host. | 16:46 |
| patrickeast | gaurangt: which means you do need to run A-A c-vol nodes | 16:47 |
| patrickeast | or A-P | 16:47 |
| patrickeast | for now | 16:47 |
| patrickeast | gaurangt: that way either site can handle requests for either backend | 16:47 |
| *** diogogmt has quit IRC | 16:47 | |
| patrickeast | gaurangt: does that make sense? | 16:47 |
| gaurangt | you mean, I need to have same host attribute in cinder.conf for both the c-vol services? | 16:48 |
| gaurangt | or just run one and not run another until we fail over to the other site? | 16:48 |
| patrickeast | gaurangt: for now the safe way is to only run one at a time with like an active-passive setup | 16:49 |
| patrickeast | gaurangt: soon we'll be able to just run them all in a cluster, but probably not for another release | 16:50 |
| gaurangt | >> "so in that case when you failover you need to update the host entry for the volumes, but to your point you can't do that if the host is down" | 16:50 |
| gaurangt | I can do it in failover_host function which is run from secondary. | 16:50 |
| jungleboyj | erlon: Pong. | 16:51 |
| patrickeast | yea as long as you have a node that can pickup the failover_host call for that cinder backend (regardless of where the c-vol is running) you can modify the host entry in the updates | 16:51 |
| erlon | jungleboyj: https://review.openstack.org/#/c/147186/32..33/cinder/volume/drivers/nfs.py | 16:52 |
| *** zhongjun_ has quit IRC | 16:52 | |
| gaurangt | ok, sounds good. I will post the patch soon and you can take a look to see if that makes sense. | 16:52 |
| erlon | jungleboyj: fix for the cloning of the NFS :) | 16:52 |
| *** zhongjun_ has joined #openstack-cinder | 16:53 | |
| erlon | jungleboyj: it was implemented already, just these small OVO errors | 16:53 |
| *** porrua has joined #openstack-cinder | 16:53 | |
| *** baumann has joined #openstack-cinder | 16:54 | |
| *** kfarr has joined #openstack-cinder | 16:54 | |
| erlon | jungleboyj: I'm trying to understand how the actual nfs job is passing as the host is a trusty wuth libvirt 1.2.2 | 16:55 |
| *** alonma has joined #openstack-cinder | 16:56 | |
| *** tesseract- has quit IRC | 16:56 | |
| *** akshai has quit IRC | 16:57 | |
| *** akapil_ has joined #openstack-cinder | 16:57 | |
| *** akshai has joined #openstack-cinder | 16:58 | |
| jungleboyj | erlon: Are the snapshot test cases disabled out there now? | 16:59 |
| *** alonma has quit IRC | 17:00 | |
| *** akapil has quit IRC | 17:01 | |
| *** Suyi_ has joined #openstack-cinder | 17:02 | |
| *** akapil_ has quit IRC | 17:02 | |
| jungleboyj | erlon: Nice catch on the creating volume from Snapshot! I will have to give that a try. | 17:02 |
| *** diablo_rojo_phon has joined #openstack-cinder | 17:03 | |
| jungleboyj | erlon: We will have work to do to get volume cloning working. | 17:03 |
| *** mriedem1 has joined #openstack-cinder | 17:04 | |
| *** lprice has joined #openstack-cinder | 17:04 | |
| *** lprice1 has quit IRC | 17:04 | |
| erlon | jungleboyj: No, they are running, and passing: http://logs.openstack.org/86/147186/32/experimental/gate-tempest-dsvm-full-devstack-plugin-nfs-nv/8998eff/console.html | 17:05 |
| *** gaurangt has left #openstack-cinder | 17:05 | |
| jungleboyj | Well, that is good. | 17:05 |
| erlon | jungleboyj: this last run just the clonning tests didn't passed, I think that the next run will sucess | 17:05 |
| *** mriedem has quit IRC | 17:06 | |
| *** krtaylor has joined #openstack-cinder | 17:06 | |
| erlon | jungleboyj: but, one think that might be happening is that the tempest tests is not covering the situation where nova needs libvirt>=1.2.7 | 17:06 |
| *** sdake has joined #openstack-cinder | 17:06 | |
| *** baumann has quit IRC | 17:07 | |
| jungleboyj | erlon: Interesting. I will take a look at the tests and trying pulling your latest patch down to my 14.04 VM. | 17:08 |
| *** jgregor has quit IRC | 17:09 | |
| erlon | jungleboyj: ok | 17:09 |
| *** sdake_ has quit IRC | 17:09 | |
| erlon | jungleboyj: I though also that should be interesting to create anoter gate job to test it with NAS secure enabled | 17:10 |
| erlon | jungleboyj: the actual uses nas_secure=False | 17:10 |
| erlon | jungleboyj: what do you think? | 17:10 |
| *** baumann has joined #openstack-cinder | 17:11 | |
| *** baumann has quit IRC | 17:12 | |
| jungleboyj | erlon: Oy, we probably should do that but I know I have been seeing issues with that enabled. | 17:12 |
| *** akshai has quit IRC | 17:14 | |
| erlon | jungleboyj: The issues are related to configurations in the compute nodes. As cinder-volume and qemu instances in the compute nodes run wth different users, some operations may fail. | 17:14 |
| erlon | jungleboyj: if you set rootsquash=true in the NFS server, enable nas_secure, and configure libvirt to run QEMU as ubuntu:ubuntu por example it works clean | 17:16 |
| erlon | jungleboyj: that is the ideal scenario of a nas_secure configuration, i.e., rootsquash + nas_secure_true, the only command run as root is 'mount' | 17:17 |
| jungleboyj | erlon: Ok, I will have to give that a try. | 17:17 |
| *** harlowja has joined #openstack-cinder | 17:19 | |
| openstackgerrit | Merged openstack/os-brick: standardize release note page ordering https://review.openstack.org/366988 | 17:19 |
| *** diogogmt has joined #openstack-cinder | 17:21 | |
| *** baumann has joined #openstack-cinder | 17:23 | |
| *** avishay has quit IRC | 17:25 | |
| *** alonma has joined #openstack-cinder | 17:26 | |
| *** baumann has quit IRC | 17:27 | |
| *** alonma has quit IRC | 17:31 | |
| *** e0ne has joined #openstack-cinder | 17:38 | |
| *** chris_morrell has joined #openstack-cinder | 17:41 | |
| EmilienM | hey, this patch would help to fix volume encryption with barbican/cinder, can we review it ? https://review.openstack.org/#/c/366750/ | 17:42 |
| *** harlowja has quit IRC | 17:43 | |
| *** harlowja has joined #openstack-cinder | 17:45 | |
| hemna | EmilienM, ping | 17:45 |
| *** chris_morrell has quit IRC | 17:46 | |
| EmilienM | hemna: yup | 17:47 |
| *** lkuchlan has joined #openstack-cinder | 17:47 | |
| *** mriedem1 is now known as mriedem | 17:47 | |
| hemna | did you take a look at dane-fichter's comment ? | 17:48 |
| hemna | I'm curious to find an example of that | 17:48 |
| EmilienM | hemna: I'm not the code author | 17:48 |
| hemna | np | 17:50 |
| hemna | any idea what removals library they are talking about ? | 17:50 |
| *** zul has joined #openstack-cinder | 17:51 | |
| *** akerr has quit IRC | 17:51 | |
| patrickeast | hemna: totally unrelated to that change, have you tried out xenial w/ the latest brick? | 17:52 |
| hemna | patrickeast, I haven't yet.... | 17:53 |
| patrickeast | hemna: i'm getting less than good results :*( | 17:53 |
| hemna | I'm still using trusty | 17:53 |
| hemna | shite | 17:53 |
| hemna | whats up | 17:53 |
| patrickeast | mm well, not 100% sure yet | 17:53 |
| patrickeast | there are two problems that i can see right now | 17:53 |
| hemna | should I fire up a xenial vagrant and try attaches ? | 17:53 |
| patrickeast | one is getting i/o errors, probably related to something i need to change | 17:53 |
| hemna | hrmm | 17:53 |
| patrickeast | the other is that i'm seeing terminate connections being called and the iscsi session not being logged out | 17:54 |
| patrickeast | my favorite of bugs rears its head again | 17:54 |
| hemna | it's always somethin | 17:54 |
| patrickeast | truth | 17:54 |
| hemna | iSCSI ? | 17:54 |
| patrickeast | yep, iscsi + multipathing | 17:54 |
| hemna | https://review.openstack.org/#/c/264678/ | 17:55 |
| patrickeast | fc seems to be happy | 17:55 |
| hemna | I was just looking at that one today | 17:55 |
| hemna | was going to try and test it | 17:55 |
| hemna | the iSCSI code is well...a mess IMHO | 17:55 |
| hemna | I'd like to refactor it at some point to make it cleaner like the FC stuff is | 17:55 |
| hemna | FC is so much simpler | 17:55 |
| patrickeast | yea, could definitely use some cleanup | 17:55 |
| patrickeast | its like bug fix on top of bug fix | 17:55 |
| hemna | yah | 17:55 |
| hemna | :( | 17:55 |
| *** eharney_ has joined #openstack-cinder | 17:56 | |
| patrickeast | sorta losing the tree in the forest... or however that goes | 17:56 |
| patrickeast | anywho, yea if you have a chance to try it out with like a 3par or something that would be sweet | 17:56 |
| hemna | https://pypi.python.org/pypi/debtcollector | 17:56 |
| hemna | I wonder if that's the lib mentioned in that review | 17:56 |
| hemna | I don't think we use it in Cinder | 17:56 |
| hemna | Glance does though | 17:57 |
| *** eharney has quit IRC | 17:57 | |
| *** eharney_ is now known as eharney | 17:57 | |
| *** kaisers has joined #openstack-cinder | 18:00 | |
| hemna | has openstack switched to xenial yet? | 18:01 |
| *** kaisers has quit IRC | 18:01 | |
| hemna | I was under the assumption we were still on trusty | 18:01 |
| *** kaisers_ has joined #openstack-cinder | 18:01 | |
| jgriffith | hemna: months ago | 18:01 |
| patrickeast | iiuc the switch is happening in gate | 18:01 |
| patrickeast | i saw an email a couple weeks ago announcing N packages on xenial | 18:02 |
| hemna | ah ok cool | 18:02 |
| patrickeast | thats why i figured it was prudent to see if it actually worked | 18:02 |
| patrickeast | XD | 18:02 |
| hemna | guess our dib images are still on trusty though for our ci | 18:02 |
| hemna | :) | 18:02 |
| jgriffith | hemna: the jobs are labelled "xenial-xxxx" in the jenkins responses | 18:02 |
| hemna | jgriffith, yah just noticed that. where have I been.... | 18:02 |
| hemna | smh | 18:02 |
| jgriffith | ha | 18:02 |
| patrickeast | i think tempest jobs are still trusty though, right? | 18:03 |
| * hemna updates my packer.json ...... | 18:03 | |
| patrickeast | or at least some of them | 18:03 |
| jgriffith | patrickeast: nope... "gate-tempest-dsvm-xxxx-xenial" | 18:03 |
| jgriffith | patrickeast: they're running in parallel | 18:03 |
| patrickeast | jgriffith: oooh nvm, i was looking at an old review | 18:04 |
| jgriffith | hehe | 18:04 |
| jgriffith | yeah, I think the trusty dsmv-full stopped a while back | 18:04 |
| *** akshai has joined #openstack-cinder | 18:04 | |
| jgriffith | that's just leaves Grenade | 18:04 |
| patrickeast | yea its grenade and looks like maybe the ceph one | 18:04 |
| hemna | looks like 3par ci is still on trusty | 18:05 |
| patrickeast | i think pretty much all the 3rd party ci's are | 18:05 |
| hemna | has the dib image builder not been updated to xenial ? | 18:05 |
| *** jgregor has joined #openstack-cinder | 18:05 | |
| patrickeast | you have to add in new configs for it for your system | 18:06 |
| patrickeast | new node labels, jobs to go with | 18:06 |
| patrickeast | or at least updated labels and jobs | 18:06 |
| patrickeast | annnd figure out new networking and multipath config changes :( | 18:06 |
| * patrickeast grumbles about new things | 18:06 | |
| hemna | ah well, I guess I'll send an IT request into the black hole to update it. | 18:06 |
| hemna | patrickeast, what multipath config changes ? | 18:07 |
| jgriffith | hemna: does anything ever come back out of that black-hole? | 18:07 |
| hemna | nova.conf [libvirt] ? | 18:07 |
| hemna | only gamma radiation | 18:07 |
| jgriffith | hemna: seems about right | 18:08 |
| patrickeast | hemna: nah there was a change to some defaults in /etc/multipath.conf and one of the key names was removed that used to be in ours | 18:08 |
| *** alonma has joined #openstack-cinder | 18:08 | |
| hemna | ah ok. please don't tell me they enabled friendly names by default | 18:08 |
| * hemna grumbles at another distro for doing that | 18:08 | |
| patrickeast | the two things that tripped my system up was teh naming change with nic's from ethX to ensX, some differences in the cloud-init starting nics automagically now, and that there is iscsid *and* open-iscsi services running now instead of just the one | 18:09 |
| patrickeast | guess that was three | 18:09 |
| *** GB21 has quit IRC | 18:10 | |
| hemna | bah | 18:10 |
| hemna | because eth0 is bad ens0 good | 18:11 |
| patrickeast | heh even more fun | 18:11 |
| patrickeast | its like eth0 turned into ens3 | 18:11 |
| patrickeast | cus... reasons | 18:11 |
| hemna | that's awesome sauce right there. | 18:11 |
| patrickeast | i do like the systemctl stuff its using now though | 18:12 |
| patrickeast | makes it easier to share scripts with centos | 18:12 |
| hemna | I think distros just change things to keep themselves busy | 18:12 |
| *** alonma has quit IRC | 18:13 | |
| patrickeast | haha yea, i'm sure there are lots of better things but its kinda incremental | 18:13 |
| patrickeast | not like big "wow" improvements that make me forget all the stupid little things that break | 18:14 |
| hemna | alrighty, IT request sent off into oblivion | 18:14 |
| *** lseki has joined #openstack-cinder | 18:14 | |
| lseki | hemna: ping | 18:15 |
| hemna | lseki, hey | 18:15 |
| lseki | hemna: hi, I'm testing attached volume migration in my environment, and I'm having troubles | 18:17 |
| lseki | hemna: could you help me please? | 18:17 |
| hemna | I can try | 18:17 |
| hemna | :) | 18:17 |
| lseki | well, I created a volume, attached it to an instance, and then tried to migrate it via cinder CLI | 18:19 |
| hemna | using the force flag I presume | 18:19 |
| lseki | I expected c-api to return an error like "You cannot migrate an attached volume!" | 18:20 |
| lseki | but instead, c-vol raises "Unauthorized: The request you have made requires authentication. (HTTP 401)" | 18:20 |
| scottda | lseki: YOu must be admin to migrate | 18:20 |
| hemna | c-vol raised that ? | 18:20 |
| lseki | yep | 18:20 |
| hemna | shouldn't c-api raise that ? | 18:20 |
| lseki | that's what I expected | 18:21 |
| lseki | scottda: I can migrate the volume if it's not attached to a VM, so I suppose I'm using correct credentials | 18:22 |
| hemna | have a log file? | 18:22 |
| patrickeast | sounds like the call to nova thats failing | 18:22 |
| scottda | lseki: But cinder must call nova for an attached volume.... | 18:23 |
| hemna | scottda, it does at the end of the process | 18:24 |
| lseki | http://paste.openstack.org/show/572412/ | 18:24 |
| scottda | lseki: Is this on master? or an older release? | 18:24 |
| lseki | it's on master | 18:25 |
| scottda | Well, that paste shows it's failing in the call to Nova | 18:25 |
| patrickeast | yea | 18:25 |
| hemna | yah it's the call to Nova that's puking out | 18:25 |
| patrickeast | well, failing the call to keystone for auth | 18:25 |
| patrickeast | so it can do the call to nova | 18:26 |
| patrickeast | which leads me to think its something misconfigured in the keystone stuff in cinder.conf | 18:26 |
| scottda | Maybe an auth version thing that is different for the call to Nova than for Cinder... | 18:26 |
| hemna | if keystone conf was borked, how could he auth to even create the volume ? | 18:26 |
| patrickeast | hemna: i think to scottda's point, the code cinder users to auth is probably different from what novaclient is using | 18:27 |
| hemna | unless cinder was pointing to a different keystone than nova ? | 18:27 |
| patrickeast | maybe different api calls | 18:27 |
| hemna | attach probably wouldn't work then either | 18:27 |
| scottda | cinder/compute/nova.py looks like it has a bunch of hard-coded keystone version stuff... | 18:28 |
| e0ne | scottda: hi. do you mean nova version? | 18:29 |
| scottda | e0ne: No, I'm looking at how cinder uses the novaclient... | 18:30 |
| *** wilson1 has quit IRC | 18:30 | |
| openstackgerrit | Merged openstack/cinder: Save volume_type/group_type uuid into db when creating group https://review.openstack.org/368610 | 18:30 |
| e0ne | scottda: it's used only for InstanceLocalityFilter now | 18:30 |
| *** wilson1 has joined #openstack-cinder | 18:31 | |
| scottda | e0ne: Well, in that paste ^^^ it's used for update_server_volume | 18:31 |
| hemna | e0ne, during migrate cinder calls nova to tell it to update | 18:31 |
| scottda | For when the volume ID has changed, to call nova's swap_volume | 18:31 |
| e0ne | scottda, hemna: thanks! I've missed that | 18:31 |
| scottda | hemna: But not always. Only when cinder driver cannot change the vol_id back to the original... | 18:31 |
| hemna | scottda, that's not what I see. maybe I'm missing something | 18:33 |
| hemna | but it seems to always call nova if there is an attachment | 18:33 |
| hemna | https://github.com/openstack/cinder/blob/master/cinder/volume/manager.py#L1777-L1785 | 18:33 |
| scottda | hemna: Well, I've been trying to test this call for a Tempest test, and for LVM it seems to skip this call... | 18:34 |
| scottda | hemna: just a sec, having browser issues... | 18:34 |
| *** bardia has joined #openstack-cinder | 18:36 | |
| scottda | hemna: That code path is only called if you get to here: https://github.com/openstack/cinder/blob/master/cinder/volume/manager.py#L1977 | 18:39 |
| scottda | so, 'moved' is returned from self.driver.migrate_volume(), it is driver-depenedent | 18:40 |
| hemna | right, if the migration stayed on the same backend | 18:40 |
| scottda | ok, that makes sense | 18:40 |
| hemna | basically, we don't have to involve nova at all at that point, because there is no re-attach process that needs to happen | 18:40 |
| scottda | For my tests, I've 2 lvm backends on the same host. I'm thinking there's some logic that is short-circuiting what I'm trying to do...But that's probably not relevant to this discussion. | 18:41 |
| hemna | for our drivers we always balk at attached volumes | 18:44 |
| scottda | hemna: Yeah, support for that is always driver-specific, right? | 18:44 |
| hemna | yah | 18:44 |
| *** baumann has joined #openstack-cinder | 18:44 | |
| *** baumann has quit IRC | 18:44 | |
| patrickeast | not for the generic migration though | 18:44 |
| *** baumann has joined #openstack-cinder | 18:45 | |
| patrickeast | at least it wasn't the last time i looked at it | 18:45 |
| hemna | patrickeast, right, which is the fallback I believe no ? | 18:45 |
| patrickeast | yea? | 18:45 |
| * patrickeast wants to believe it is | 18:45 | |
| scottda | patrickeast: Yes, I think that's what gets us to that code path ^^^, 'if not moved, we call generic_migrate_volume' | 18:45 |
| hemna | so, I' | 18:46 |
| hemna | bah. | 18:46 |
| hemna | so I'm not sure how lseki hit that issue with keystone | 18:46 |
| patrickeast | oh derp | 18:46 |
| patrickeast | yea | 18:46 |
| patrickeast | scottda: thats right | 18:46 |
| hemna | lseki, compare the nova.conf and cinder.conf for keystone ? | 18:46 |
| scottda | lseki: and what is your driver? | 18:47 |
| patrickeast | scottda: this reminds me... where are we at with retype/migration tests in tempest :D | 18:47 |
| hemna | :) | 18:47 |
| scottda | patrickeast: Well, we are exactly here | 18:47 |
| scottda | :) | 18:47 |
| patrickeast | haha | 18:48 |
| scottda | patrickeast: With me trying to figure out how to get the tests to call server_update_volume... | 18:48 |
| scottda | patrickeast: Which it is currently not doing, I *believe* because we have 2 lvm backends, but on a single devstack, so it's not actually moving the volume. | 18:48 |
| patrickeast | the tests are still in review right? | 18:48 |
| *** lpetrut has joined #openstack-cinder | 18:49 | |
| scottda | Might need to use a 2 node devstack for the tests. | 18:49 |
| patrickeast | i can probably repro this with my multi-backend jobs | 18:49 |
| scottda | patrickeast: Yes, still in review. But mriedem had pointed out that nova's swap_volume() is not being called. And it's because we are not hitting server_update_volume... | 18:49 |
| patrickeast | scottda: ah yea, that would do it | 18:50 |
| scottda | patrickeast: Yes, if you just ran the test on you multi-BE jobs, it should work. We'd have to look at the nova logs to see. | 18:50 |
| *** alonma has joined #openstack-cinder | 18:50 | |
| scottda | https://review.openstack.org/#/c/326681 | 18:51 |
| lseki | hemna, scottda, patrickeast: I'm checking nova.conf and cinder.conf | 18:51 |
| hemna | scottda, does that also work with cinder migrate ? | 18:52 |
| mriedem | scottda: this is why i thought it required a multinode job | 18:52 |
| mriedem | to actually migrate the volume from one host to another | 18:52 |
| openstackgerrit | Vipin Balachandran proposed openstack/cinder: Newton release notes cleanup https://review.openstack.org/367937 | 18:52 |
| mriedem | but i guess it could be different backends on a single host? | 18:52 |
| *** ganso has joined #openstack-cinder | 18:53 | |
| *** lpetrut has quit IRC | 18:53 | |
| patrickeast | mriedem: i think it depends on what type of backends they are | 18:53 |
| patrickeast | mriedem: lvm might need to be two different hosts | 18:53 |
| hemna | lvm seems to work on the same host | 18:53 |
| hemna | according to the code | 18:54 |
| hemna | so in that case, you'd need a multi-node setup | 18:54 |
| hemna | https://github.com/openstack/cinder/blob/master/cinder/volume/drivers/lvm.py#L750-L751 | 18:55 |
| *** alonma has quit IRC | 18:55 | |
| *** jgregor has quit IRC | 18:55 | |
| scottda | hemna yup, that would do it. | 18:56 |
| *** baumann has quit IRC | 18:56 | |
| patrickeast | scottda: in like 50 min look for logs of multi-backend with two pure arrays running w/ your tempest patch http://openstack-logs.purestorage.com/PureISCSIDriver-tempest-dsvm-trusty-aio-multipath-multi-backend-pure/1 | 18:56 |
| patrickeast | just manually kicked it off and hard coded TEMPEST_REPO and TEMPEST_BRANCH | 18:57 |
| *** lkuchlan has quit IRC | 18:57 | |
| hemna | ok off to lunch... | 18:57 |
| scottda | patrickeast: Let me know if you get any test results on your multi-be. But that still leaves a problem in the multi-node job, because the driver will still be LVMVolumeDriver | 18:57 |
| *** lpetrut has joined #openstack-cinder | 18:58 | |
| patrickeast | scottda: is it a problem for them to be on different nodes though? | 18:58 |
| patrickeast | scottda: guess i should just go read through the retype on the lvm driver | 18:58 |
| scottda | patrickeast: I'm not sure why? | 18:59 |
| *** baumann__ has joined #openstack-cinder | 18:59 | |
| patrickeast | scottda: err i guess my question is more of whats the problem you refer to in "But that still leaves a problem in the multi-node job, because the driver will still be LVMVolumeDriver" | 18:59 |
| scottda | patrickeast: This is retype with --migration-policy on-demand | 19:00 |
| lseki | hemna: the [keystone_authtoken] session at nova.conf and cinder.conf are similar | 19:00 |
| lseki | (except username and signing_dir that are nova and cinder, respectively) | 19:00 |
| scottda | See hemna 's gitub link ^^^ | 19:00 |
| *** zul has quit IRC | 19:00 | |
| scottda | https://www.irccloud.com/pastebin/frUzl3HU/ | 19:00 |
| patrickeast | scottda: yea it fails the hostname check, and goes into generic migration mode, isn't that what we want? | 19:00 |
| patrickeast | would fail* | 19:01 |
| lseki | scottda: I'm using hnas driver, but it doesn't implement volume migration, so it fallbacks to default behaviour | 19:01 |
| scottda | yes, I guess the question is "what will come back in host['capabilities']['location_info']"? | 19:01 |
| patrickeast | hmm yea, good point, i dunno what that is | 19:02 |
| scottda | If it is LVMVolumeDriver for both nodes, it will not pass back moved=false and will not call generic_volume_migration and server_volume_update | 19:02 |
| scottda | patrickeast: Crap, I was hoping you'd know off the top of your head :) | 19:03 |
| patrickeast | yea if the types match | 19:03 |
| patrickeast | my guess though is that the hostname matches up with the cinder.conf hostname | 19:03 |
| patrickeast | or better yet is the like host@backend string | 19:03 |
| scottda | patrickeast: OH yeah, that should be different, so it should work. | 19:04 |
| patrickeast | https://github.com/openstack/cinder/blob/master/cinder/volume/drivers/lvm.py#L93 | 19:04 |
| patrickeast | ^ should be good to go | 19:04 |
| *** pcaruana has joined #openstack-cinder | 19:04 | |
| patrickeast | assuming the nodes are not all named 'ubuntu' or something silly like that | 19:04 |
| * patrickeast may or may not have run in circles with a problem like that the other day | 19:04 | |
| *** lpetrut has quit IRC | 19:05 | |
| *** xyang1 has joined #openstack-cinder | 19:05 | |
| scottda | lseki: Well, you could file a bug. Or wait for patrickeast tests to run to see if he get's success. Meanwhile, I'll move those tempest tests to multi-node and get that to work to have this move toward testing in the gate. | 19:06 |
| *** tongli has joined #openstack-cinder | 19:06 | |
| *** jgregor has joined #openstack-cinder | 19:07 | |
| *** e0ne has quit IRC | 19:08 | |
| lseki | scottda: Ok, I'll do it | 19:08 |
| erlon | scottda: Is Cinder supporting migration with attached volumes? | 19:09 |
| lseki | hemna: could you please ping me when you're back? | 19:09 |
| erlon | scottda: last time I followed migration (1~2 releases) it wasn't supporting | 19:10 |
| scottda | erlon: Yes, if either the driver supports it or nova-libvirt does: http://docs.openstack.org/admin-guide/blockstorage-volume-migration.html | 19:10 |
| scottda | erlon: I'm not sure if we have a support matrix for it.. | 19:11 |
| *** nherciu has quit IRC | 19:11 | |
| erlon | scottda: don't think so either, at that time I manually tested with the backends I had avail, but nobody did anything automated or build a matrix | 19:13 |
| erlon | scottda: the tempest test was blocked as well while there was not job in the gate using it | 19:13 |
| scottda | erlon: which tempest test? I've this one in -flight: https://review.openstack.org/#/c/326681/6/tempest/scenario/test_volume_migrate_attached.py | 19:14 |
| patrickeast | fwiw i just did the generic migration between two pure arrays manually with an attached volume and things seem to have worked ok | 19:14 |
| scottda | erlon: But I want to change that to do the appropriate calls to nova. | 19:14 |
| patrickeast | manually as in not via the tempest tests | 19:14 |
| patrickeast | tempest job is still stacking | 19:15 |
| erlon | scottda: not this one, it was one submmited by vicent hou | 19:15 |
| scottda | erlon: Those Vincent hou tests were for retype-with-migration, but not with a volume attached. | 19:16 |
| scottda | erlon: https://review.openstack.org/#/c/195443 | 19:17 |
| erlon | scottda: yes, because IIRC at the first implementation migration with attached wasn't supported | 19:17 |
| scottda | ok | 19:18 |
| xinli | @smcginnis: after I reboot my devstack node, it seems that cinder api's server failed to accept the request? how to resolve the issue: http://paste.openstack.org/show/572425 | 19:19 |
| *** kfarr has quit IRC | 19:19 | |
| smcginnis | xinli: Did you restack? | 19:20 |
| patrickeast | +1 for restacking | 19:20 |
| xinli | @smcginnis: yes, I run ./stack.sh | 19:20 |
| scottda | xinli: After rebooting? | 19:21 |
| xinli | @smcginnis: correct | 19:21 |
| erlon | xinli: that's is the tempest failure, what the services logs says? | 19:21 |
| xinli | erlog: /var/log/ in which log file? | 19:23 |
| *** ganso has left #openstack-cinder | 19:23 | |
| *** mriedem has quit IRC | 19:24 | |
| erlon | scottda: It would be nice if was be possible to configure more than 2 backends, and run all those migration and multibackend stuff in the maximun of combinations as possible | 19:24 |
| erlon | xinli: in devstack logs, I think they are in /opt/stack/logs/ | 19:25 |
| smcginnis | xinli: You can connect to the screen session and look at the c-api output. Or the log file is typically in /opt/stack/screen/screen-c-vol.log | 19:25 |
| scottda | erlon: Yes, but these tests are the most time-consuming. At least if you have cinder migrate with attached boot-from-volume. | 19:25 |
| erlon | xinli: or you can screen -r and then browse in the service's tabs | 19:25 |
| erlon | scottda: even with small 1G volumes? | 19:27 |
| erlon | scottda: how long it usually takes?? | 19:27 |
| *** diablo_rojo has joined #openstack-cinder | 19:27 | |
| scottda | erlon: Depends, but the tempest/scenario/test_volume_boot_pattern.py is one of the longest, and this would add more to that... | 19:28 |
| jgriffith | scottda: I think the question is; what does "longest" mean? Quantitatively | 19:28 |
| jgriffith | scottda: because if it's a 90 second test then it's well worth it | 19:28 |
| erlon | scottda: do we always have to run all tempest tests? if we can run a job with only a few tests would be much more benefit | 19:29 |
| scottda | Yeah, I agree it's worth it to robustly test, but there was some push back at first, and might be if we do "the maximum of combinations...possible". | 19:29 |
| jgriffith | scottda: TVP rangest from 30 to 180 seconds most of the time depending on version etc | 19:29 |
| *** diablo_rojo_ has joined #openstack-cinder | 19:29 | |
| jgriffith | scottda: and frankly that's the worst, most unreliable test we have | 19:30 |
| erlon | scottda: it does not make sense to test all the tempest harness if I only want to test migration, for example | 19:30 |
| *** jgregor has quit IRC | 19:30 | |
| scottda | I've certainly got the migrate_attached_volume tests up...just not BFV yet. | 19:30 |
| *** baumann__ has quit IRC | 19:30 | |
| scottda | erlon: Sure. But in the gate, we want to test this code path. Currently, nothing tests this call from Cinder-> nova | 19:30 |
| *** alonma has joined #openstack-cinder | 19:31 | |
| erlon | scottda: not sure what you mean, you mean you want this test to run alongside all other tests? | 19:33 |
| *** baumann has joined #openstack-cinder | 19:33 | |
| *** mriedem has joined #openstack-cinder | 19:33 | |
| scottda | erlon: We want this in Tempest, running in the merge queueu, yes. | 19:33 |
| *** Lee1092 has quit IRC | 19:34 | |
| *** jgregor has joined #openstack-cinder | 19:35 | |
| *** liverpooler has quit IRC | 19:36 | |
| *** alonma has quit IRC | 19:36 | |
| erlon | scottda: mhm, yes, I agree we should have it, but the ideia would have additional tempest tests, or a configuration that allowed to run they in specific jobs, with more backends enabled | 19:36 |
| scottda | erlon: Sure, that could be done. | 19:37 |
| *** baumann has quit IRC | 19:38 | |
| xinli | erlon: the c-api log shows : http://paste.openstack.org/show/572427/ | 19:39 |
| *** kfarr has joined #openstack-cinder | 19:39 | |
| *** jgregor has quit IRC | 19:41 | |
| *** baumann has joined #openstack-cinder | 19:41 | |
| *** julim_ has joined #openstack-cinder | 19:42 | |
| erlon | xinli: the reboot you say you did, was a reboot or a crash? | 19:42 |
| xinli | erlon: it's a reboot, the system admin migrate my VM to another server and after that he told me that the node has been reboot. | 19:43 |
| xinli | erlog: do you think that I need to reboot and then rerun ./stack? | 19:43 |
| patrickeast | scottda: looks like the job failed on the new test http://openstack-logs.purestorage.com/PureISCSIDriver-tempest-dsvm-trusty-aio-multipath-multi-backend-pure/1/logs/ | 19:43 |
| patrickeast | dunno why yet | 19:43 |
| erlon | xinli: it looks that something very serious happened, like you filesystem was corrupted | 19:44 |
| erlon | xinli: the easier option is to start from 0 | 19:44 |
| xinli | destroy this VM and rebuild it now one | 19:45 |
| erlon | xinli: if you have an image of the vanila system, just just ./stack again | 19:45 |
| xinli | erlon: ok. | 19:45 |
| *** julim_ has quit IRC | 19:45 | |
| erlon | xinli: yes, I never spend more than 5 minutes to recover a lot devstack | 19:46 |
| *** jgregor has joined #openstack-cinder | 19:46 | |
| erlon | xinli: trying to... | 19:46 |
| smcginnis | erlon: Good call. ;) | 19:46 |
| erlon | smcginnis: revert snapshot + ./stack.sh while I have a coffee :) | 19:47 |
| *** zul has joined #openstack-cinder | 19:55 | |
| *** _ducttape_ has quit IRC | 19:57 | |
| *** lpetrut has joined #openstack-cinder | 19:59 | |
| jungleboyj | erlon: Were you still working on fixing up the gate jobs? Getting them fixed until we have the snapshot patch merged? | 20:00 |
| *** ducttape_ has joined #openstack-cinder | 20:00 | |
| *** tongli has quit IRC | 20:01 | |
| erlon | jungleboyj: yes, the gate job should work the way it is | 20:01 |
| jungleboyj | So you changed it? | 20:01 |
| erlon | jungleboyj: but I want to understand why is not failing | 20:01 |
| *** _coolsvap_ has quit IRC | 20:02 | |
| erlon | jungleboyj: no, apparentlly the tempest tests are not touching the cade where it should require libvirt 1.2.7, so, its just passing | 20:02 |
| erlon | s/cade/code/ | 20:02 |
| *** alonma has joined #openstack-cinder | 20:03 | |
| erlon | jungleboyj: I sent that change this morening but It hit an error in a unit test | 20:03 |
| jungleboyj | erlon: I just tried running everything by hand on libvirt 1.2.2 and it looks good to me. | 20:04 |
| erlon | jungleboyj: you run tempest? | 20:04 |
| jungleboyj | Eric explained to me the limitation in versions less than 1.2.7 and I am thinking it will only have problems if the volume you are trying to snapshot is attached. | 20:05 |
| jungleboyj | erlon: I just ran tempest and it looked like all the snapshot tests passed. Running again due to another configuration error. | 20:05 |
| erlon | jungleboyj: yes, if the volume is attached and you delete a snapshot | 20:05 |
| erlon | jungleboyj: I think we should also modify the tempest test to delete a volume while attached | 20:06 |
| *** alonma has quit IRC | 20:07 | |
| *** tongli has joined #openstack-cinder | 20:07 | |
| erlon | jungleboyj: https://github.com/openstack/tempest/blob/8dbbc297b460347451929abf611ef05d6ac56f4b/tempest/api/volume/test_volumes_snapshots.py#L75-L75 | 20:07 |
| jungleboyj | erlon: Not a bad idea. | 20:08 |
| erlon | jungleboyj: I think it could cover more situations, like, creating 3/4 snaps online, deleting online, creating offline then attaching, etc | 20:09 |
| *** porrua has quit IRC | 20:11 | |
| *** MVenesio has joined #openstack-cinder | 20:11 | |
| erlon | jungleboyj: can you do that? them I can wokr on the gate changes | 20:12 |
| *** tongli has quit IRC | 20:12 | |
| erlon | jungleboyj: a @ddt with a lot off possible cases would be nice | 20:12 |
| MVenesio | Hi guys, i'm using nova integrated with ceph, and i'm having a big issue. One ceph monitor is down and, when i boot a nova instance created from volume, it fails trying to connect with the dead monitor. But i have quorum in ceph with another two servers | 20:13 |
| MVenesio | any idea about this | 20:13 |
| MVenesio | ? | 20:13 |
| erlon | jungleboyj: but is tempests not UT lol, | 20:13 |
| *** tongli has joined #openstack-cinder | 20:13 | |
| erlon | jungleboyj: not even know if that is possible | 20:14 |
| jungleboyj | erlon: Tempest is not. I am not sure. That stuff is new to me. | 20:15 |
| *** diablo_rojo_phon has quit IRC | 20:15 | |
| erlon | jungleboyj: hmm, but its ok to use without ddt as well | 20:16 |
| jungleboyj | erlon: I just completed a tempest run, with NFS snapshots on Ubuntu 14.04 | 20:18 |
| erlon | jungleboyj: passed all? | 20:18 |
| jungleboyj | Oh, one more thing here. | 20:19 |
| *** nicolasbock has joined #openstack-cinder | 20:22 | |
| openstackgerrit | Erlon R. Cruz proposed openstack/cinder: NFS snapshots https://review.openstack.org/147186 | 20:24 |
| erlon | jungleboyj: what? | 20:24 |
| jungleboyj | erlon: Ooops, it looks like it was still using LVM for the test run. I switched over to only having the NFS driver running and now LOTS of things fail. So I will need to further investigate. | 20:24 |
| *** sdake_ has joined #openstack-cinder | 20:25 | |
| *** thumpba has quit IRC | 20:27 | |
| *** xinli has quit IRC | 20:27 | |
| *** xinli has joined #openstack-cinder | 20:27 | |
| *** sdake has quit IRC | 20:28 | |
| *** rmcall_ has joined #openstack-cinder | 20:28 | |
| scottda | patrickeast: It looks like the Tempest test failure was during cleanup...maybe a timeout, not sure. | 20:29 |
| *** thumpba has joined #openstack-cinder | 20:29 | |
| *** rmcall has quit IRC | 20:29 | |
| *** rmcall_ is now known as rmcall | 20:29 | |
| *** salv-orl_ has joined #openstack-cinder | 20:31 | |
| *** lpetrut has quit IRC | 20:31 | |
| scottda | patrickeast: But I don't see the tests making the call into Nova swap_volume. | 20:32 |
| *** salv-orlando has quit IRC | 20:34 | |
| *** alonma has joined #openstack-cinder | 20:34 | |
| *** eharney has quit IRC | 20:35 | |
| jungleboyj | erlon: I gotta run. I will keep looking into this. | 20:37 |
| *** mdenny has quit IRC | 20:37 | |
| erlon | jungleboyj: ok, Im leaving in a minute as well | 20:37 |
| *** cknight has quit IRC | 20:37 | |
| jungleboyj | erlon: Cool. Later. | 20:38 |
| erlon | jungleboyj: see you | 20:38 |
| *** alonma has quit IRC | 20:39 | |
| *** tongli has quit IRC | 20:39 | |
| *** akapil has joined #openstack-cinder | 20:41 | |
| *** jungleboyj has quit IRC | 20:42 | |
| *** MVenesio has quit IRC | 20:43 | |
| *** akapil has quit IRC | 20:45 | |
| *** pcaruana has quit IRC | 20:46 | |
| *** lpetrut has joined #openstack-cinder | 20:49 | |
| *** akshai has quit IRC | 20:56 | |
| *** sdague has quit IRC | 21:02 | |
| *** nicolasbock has quit IRC | 21:03 | |
| *** alonma has joined #openstack-cinder | 21:06 | |
| *** baumann has quit IRC | 21:06 | |
| *** mvk has joined #openstack-cinder | 21:07 | |
| *** pauloewerton has quit IRC | 21:09 | |
| *** enriquetaso has quit IRC | 21:10 | |
| *** alonma has quit IRC | 21:10 | |
| *** kaisers_ has quit IRC | 21:12 | |
| *** kaisers has joined #openstack-cinder | 21:14 | |
| *** rlrossit has quit IRC | 21:16 | |
| *** kaisers has quit IRC | 21:18 | |
| *** jgregor has quit IRC | 21:21 | |
| *** jgregor has joined #openstack-cinder | 21:22 | |
| *** jgregor has quit IRC | 21:28 | |
| *** enriquetaso has joined #openstack-cinder | 21:30 | |
| openstackgerrit | xing-yang proposed openstack/cinder: Add functional tests for groups https://review.openstack.org/362584 | 21:33 |
| *** alonma has joined #openstack-cinder | 21:37 | |
| *** lprice has quit IRC | 21:38 | |
| *** sdake_ has quit IRC | 21:38 | |
| *** sdake has joined #openstack-cinder | 21:39 | |
| *** alonma has quit IRC | 21:41 | |
| patrickeast | ruh roh | 21:46 |
| patrickeast | i think theres something wonkey going on with retype | 21:46 |
| hemna | oh? | 21:47 |
| * patrickeast pulls up github links | 21:47 | |
| *** sdake_ has joined #openstack-cinder | 21:47 | |
| patrickeast | this check is always failing | 21:47 |
| patrickeast | https://github.com/openstack/cinder/blob/master/cinder/volume/manager.py#L2290 | 21:47 |
| patrickeast | because diff looks liek {'encryption': {}, 'qos_specs': {}, 'extra_specs': {u'replication_enabled': (u'<is> True', None)}} | 21:48 |
| *** raunak has joined #openstack-cinder | 21:48 | |
| patrickeast | so checking if it is None is always false | 21:48 |
| patrickeast | for the encryption thingy | 21:48 |
| hemna | sweet, so lets just remove the rest of the code under it then. | 21:48 |
| hemna | done! | 21:48 |
| patrickeast | easy! | 21:48 |
| hemna | Cinder team cleans up code.... | 21:49 |
| patrickeast | but yea, so i don't think its ever going to call the driver retype as-is :( | 21:49 |
| patrickeast | unless i'm missing something | 21:49 |
| *** sdake has quit IRC | 21:49 | |
| *** thumpba has quit IRC | 21:49 | |
| *** sdake has joined #openstack-cinder | 21:50 | |
| hemna | man I just updated chrome and it's been crashing all day today | 21:50 |
| * hemna waits for a new tab to finish rendering... | 21:50 | |
| patrickeast | ouch | 21:51 |
| *** dustins has quit IRC | 21:51 | |
| *** sdake_ has quit IRC | 21:52 | |
| hemna | so did diff change from a dict to an ovo ? | 21:56 |
| patrickeast | nah, i see the problem | 21:56 |
| patrickeast | but confused as to how it was introduced and no one noticed | 21:57 |
| patrickeast | https://bugs.launchpad.net/cinder/+bug/1622762 | 21:57 |
| openstack | Launchpad bug 1622762 in Cinder "retype never calls driver method" [Undecided,New] | 21:57 |
| patrickeast | https://github.com/openstack/cinder/commit/2d1242ae69378dae8ce26a51af3d860a2875eba0#diff-724c53b8cd580ae533b4fb301df7c3e9R2269 | 21:57 |
| patrickeast | that check is wrong | 21:57 |
| hemna | probably because we fall back and do a migrate | 21:57 |
| hemna | so tests pass | 21:58 |
| patrickeast | yea, but if you don't specify to allow migration it never works | 21:58 |
| patrickeast | i guess we don't really test retype very well : / | 21:58 |
| hemna | evidently not :( | 21:58 |
| *** alyson_ has quit IRC | 21:59 | |
| *** lseki has quit IRC | 22:00 | |
| scottda | Well, there's an associated un-merged Tempest test that is a WIP. Perhaps we shouldn't merge the code until we get the tests working? | 22:01 |
| patrickeast | XD | 22:02 |
| patrickeast | scottda: maybe for like the P release | 22:02 |
| patrickeast | features > working code | 22:02 |
| patrickeast | obvs | 22:02 |
| scottda | and tests optional | 22:02 |
| patrickeast | thats what customers are for, amiright? | 22:03 |
| *** mriedem has quit IRC | 22:03 | |
| *** ducttape_ has quit IRC | 22:04 | |
| *** alonma has joined #openstack-cinder | 22:08 | |
| *** zigo has quit IRC | 22:08 | |
| *** martyt has joined #openstack-cinder | 22:10 | |
| *** zigo has joined #openstack-cinder | 22:11 | |
| *** zigo is now known as Guest41388 | 22:11 | |
| *** alonma has quit IRC | 22:13 | |
| *** cebruns__ has quit IRC | 22:13 | |
| *** cebruns has joined #openstack-cinder | 22:15 | |
| *** Guest41388 is now known as zigo_ | 22:17 | |
| *** cebruns has quit IRC | 22:19 | |
| *** thumpba has joined #openstack-cinder | 22:20 | |
| *** ociuhandu has quit IRC | 22:20 | |
| *** cebruns has joined #openstack-cinder | 22:21 | |
| *** cebruns has quit IRC | 22:22 | |
| *** thumpba has quit IRC | 22:24 | |
| openstackgerrit | Patrick East proposed openstack/cinder: Fix logic in retype check for encryption changes https://review.openstack.org/369064 | 22:25 |
| patrickeast | ayy with that my replication tempest tests pass | 22:26 |
| *** cebruns has joined #openstack-cinder | 22:29 | |
| *** alonma has joined #openstack-cinder | 22:40 | |
| *** cebruns has quit IRC | 22:41 | |
| *** cebruns has joined #openstack-cinder | 22:42 | |
| *** bardia has quit IRC | 22:44 | |
| *** martyt has quit IRC | 22:45 | |
| *** alonma has quit IRC | 22:45 | |
| *** krtaylor has quit IRC | 22:45 | |
| *** sdake has quit IRC | 22:48 | |
| *** openstackgerrit has quit IRC | 22:48 | |
| *** openstackgerrit has joined #openstack-cinder | 22:49 | |
| *** rmcall has quit IRC | 22:50 | |
| *** rmcall has joined #openstack-cinder | 22:50 | |
| *** salv-orlando has joined #openstack-cinder | 22:53 | |
| *** mriedem has joined #openstack-cinder | 22:53 | |
| *** xinli has quit IRC | 22:54 | |
| *** salv-orl_ has quit IRC | 22:54 | |
| *** sdague has joined #openstack-cinder | 22:56 | |
| *** kaisers has joined #openstack-cinder | 23:03 | |
| *** lpetrut has quit IRC | 23:03 | |
| *** sdague has quit IRC | 23:04 | |
| *** hemna is now known as hemnafk | 23:05 | |
| *** kfarr has quit IRC | 23:05 | |
| *** fernnest has quit IRC | 23:07 | |
| *** kaisers has quit IRC | 23:07 | |
| Swanson | anyway to up the number of times os-brick does a rescan? | 23:10 |
| *** alonma has joined #openstack-cinder | 23:11 | |
| *** martyt has joined #openstack-cinder | 23:11 | |
| *** sdague has joined #openstack-cinder | 23:15 | |
| *** martyt has quit IRC | 23:15 | |
| *** alonma has quit IRC | 23:16 | |
| *** cebruns has quit IRC | 23:16 | |
| *** cebruns has joined #openstack-cinder | 23:19 | |
| *** xyang1 has quit IRC | 23:20 | |
| *** raunak has quit IRC | 23:26 | |
| *** cebruns has quit IRC | 23:27 | |
| *** zul has quit IRC | 23:28 | |
| *** cebruns has joined #openstack-cinder | 23:30 | |
| *** mriedem has quit IRC | 23:34 | |
| *** alonma has joined #openstack-cinder | 23:43 | |
| *** alonma has quit IRC | 23:47 | |
| *** akshai has joined #openstack-cinder | 23:50 | |
| *** raunak has joined #openstack-cinder | 23:52 | |
| *** akshai_ has joined #openstack-cinder | 23:53 | |
| *** akshai has quit IRC | 23:56 | |
| *** leeantho has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!