Wednesday, 2016-08-17

*** Julien-zte has joined #openstack-cinder00:02
*** akapil has quit IRC00:02
*** merooney_ has joined #openstack-cinder00:06
*** yangyapeng has quit IRC00:08
*** akapil has joined #openstack-cinder00:08
*** merooney has quit IRC00:09
*** alonma has joined #openstack-cinder00:12
*** arch-nemesis has quit IRC00:14
*** alonma has quit IRC00:16
*** mtanino has quit IRC00:18
*** alonma has joined #openstack-cinder00:19
*** Julien-zte has quit IRC00:23
*** alonma has quit IRC00:24
*** Julien-zte has joined #openstack-cinder00:25
*** alonma has joined #openstack-cinder00:26
*** Julien-zte has quit IRC00:26
*** diogogmt has quit IRC00:28
*** alonma has quit IRC00:30
*** GB21 has joined #openstack-cinder00:31
*** mriedem has joined #openstack-cinder00:35
*** thumpba_ has quit IRC00:39
*** thumpba has joined #openstack-cinder00:39
*** akapil_ has joined #openstack-cinder00:43
*** thumpba has quit IRC00:44
*** akapil has quit IRC00:46
*** alonma has joined #openstack-cinder00:47
*** alonma has quit IRC00:51
*** Suyi has quit IRC00:55
*** alonma has joined #openstack-cinder00:55
*** chlong has joined #openstack-cinder00:58
*** alonma has quit IRC00:59
*** cheneydc has joined #openstack-cinder00:59
*** alonma has joined #openstack-cinder01:02
*** raunak has joined #openstack-cinder01:05
raunakHi Sean, I tried all the skipped tests and had a successful run. Would upload it as a comment on openstack review until our logs acess are been worked on01:05
*** alonma has quit IRC01:06
*** alonma has joined #openstack-cinder01:09
openstackgerritxing-yang proposed openstack/cinder: Add generic volume groups  https://review.openstack.org/32245901:11
*** leeantho has quit IRC01:12
*** Apoorva has quit IRC01:13
*** alonma has quit IRC01:14
openstackgerritxing-yang proposed openstack/cinder: Add group snapshots  https://review.openstack.org/32805201:14
*** alonma has joined #openstack-cinder01:16
openstackgerritxing-yang proposed openstack/cinder: WIP: Rename functions names involving groups.  https://review.openstack.org/35596101:19
*** cheneydc has quit IRC01:19
*** yangyapeng has joined #openstack-cinder01:20
*** alonma has quit IRC01:20
*** Apoorva has joined #openstack-cinder01:21
*** alonma has joined #openstack-cinder01:23
*** diogogmt has joined #openstack-cinder01:25
*** Asaithambi has joined #openstack-cinder01:27
*** alonma has quit IRC01:28
*** akapil_ has quit IRC01:31
*** jungleboyj has joined #openstack-cinder01:32
*** akapil has joined #openstack-cinder01:34
*** ducttape_ has quit IRC01:34
*** ducttape_ has joined #openstack-cinder01:35
*** GB21 has quit IRC01:35
*** Asaithambi has quit IRC01:38
openstackgerritDao Cong Tien proposed openstack/cinder: Stop using mox in unit/test_volume.py  https://review.openstack.org/35581501:42
*** EinstCrazy has joined #openstack-cinder01:42
*** lixiaoy1_ has quit IRC01:45
*** julim has joined #openstack-cinder01:46
*** cknight has joined #openstack-cinder01:46
openstackgerritxiexs proposed openstack/cinder: Refactor the using of dict.get() in the test assertion  https://review.openstack.org/35532301:50
*** raunak has quit IRC01:55
openstackgerritDao Cong Tien proposed openstack/cinder: Stop using mox in test_quota.py  https://review.openstack.org/35617401:55
*** bardia_ has quit IRC01:55
openstackgerritDao Cong Tien proposed openstack/cinder: Stop using mox in unit/test_quota.py  https://review.openstack.org/35617401:58
openstackgerritDao Cong Tien proposed openstack/cinder: Stop using mox in unit/test_qos_specs.py  https://review.openstack.org/35576601:59
*** EinstCrazy has quit IRC02:01
openstackgerritXu Ao proposed openstack/cinder: Change a typo error in the releasenote  https://review.openstack.org/35617602:02
*** EinstCrazy has joined #openstack-cinder02:03
openstackgerritSean McGinnis proposed openstack/cinder: Expand description of test types  https://review.openstack.org/35617702:07
*** sdake has quit IRC02:13
*** sdake has joined #openstack-cinder02:13
*** yangyapeng has quit IRC02:14
*** yangyapeng has joined #openstack-cinder02:14
*** EinstCra_ has joined #openstack-cinder02:16
*** GB21 has joined #openstack-cinder02:16
*** julim has quit IRC02:17
*** jungleboyj has quit IRC02:18
openstackgerritxiexs proposed openstack/cinder: Refactor the using of dict.get() in the test assertion  https://review.openstack.org/35532302:18
*** EinstCrazy has quit IRC02:18
*** Lee1092 has joined #openstack-cinder02:19
*** bkopilov has quit IRC02:21
*** lixiaoy1 has joined #openstack-cinder02:25
openstackgerritDao Cong Tien proposed openstack/cinder: Stop using mox in unit/api/v1/ tests  https://review.openstack.org/35618502:32
*** cknight1 has joined #openstack-cinder02:33
*** cknight has quit IRC02:33
openstackgerritKarthik Prabhu Vinod proposed openstack/cinder: Switch service capabilities to ovo  https://review.openstack.org/31904002:34
*** zhongjun_ has quit IRC02:34
*** Apoorva_ has joined #openstack-cinder02:37
openstackgerritMerged openstack/cinder: Add CI_WIKI_NAME to driver's devref  https://review.openstack.org/35586202:37
openstackgerritXu Ao proposed openstack/cinder: Change a typo error in the releasenote  https://review.openstack.org/35618602:38
*** Apoorva has quit IRC02:40
*** Apoorva_ has quit IRC02:41
*** haplo37_ has quit IRC02:43
*** akapil has quit IRC02:44
*** GB21 has quit IRC02:44
lixiaoy1Hi, I see several tempest tests in tempest.api.image.v1.test_images.CreateRegisterImagesTest.test_register_http_image, http://logs.openstack.org/60/317060/11/check/gate-tempest-dsvm-full-devstack-plugin-ceph/1f34c49/console.html#_2016-08-16_20_44_58_262300 . is it a known issue? Anyone is working on it?02:46
*** ducttape_ has quit IRC02:49
*** alonma has joined #openstack-cinder02:49
*** raunak has joined #openstack-cinder02:51
*** akapil has joined #openstack-cinder02:51
*** alonma has quit IRC02:54
*** zhongjun_ has joined #openstack-cinder02:55
*** EinstCrazy has joined #openstack-cinder02:56
*** alonma has joined #openstack-cinder02:57
openstackgerritxiexs proposed openstack/cinder: Refactor the case test_storwize_vdisk_copy_ops  https://review.openstack.org/35619402:57
*** EinstCra_ has quit IRC03:00
*** alonma has quit IRC03:01
*** alonma has joined #openstack-cinder03:09
*** mriedem has quit IRC03:11
*** ducttape_ has joined #openstack-cinder03:13
*** sticker has quit IRC03:13
*** alonma has quit IRC03:14
*** thumpba has joined #openstack-cinder03:19
openstackgerritxiexs proposed openstack/python-cinderclient: Add --is-public option to type-list  https://review.openstack.org/33588103:20
*** alonma has joined #openstack-cinder03:21
*** yuelongguang has quit IRC03:21
*** alonma has quit IRC03:25
*** bswartz has joined #openstack-cinder03:26
*** alonma has joined #openstack-cinder03:28
*** diogogmt has quit IRC03:28
*** alonma has quit IRC03:32
*** sdake has quit IRC03:34
*** alonma has joined #openstack-cinder03:35
*** akapil has quit IRC03:35
*** ducttape_ has quit IRC03:36
*** bkopilov has joined #openstack-cinder03:37
*** yangyape_ has joined #openstack-cinder03:38
*** yangyapeng has quit IRC03:38
*** alonma has quit IRC03:40
*** alonma has joined #openstack-cinder03:42
*** akapil has joined #openstack-cinder03:43
*** alonma has quit IRC03:47
*** raunak has quit IRC03:52
*** raunak has joined #openstack-cinder03:53
*** links has joined #openstack-cinder03:54
*** alonma has joined #openstack-cinder03:55
openstackgerritxiexs proposed openstack/cinder: Add validation for the metadata properties  https://review.openstack.org/33091303:56
openstackgerritxiexs proposed openstack/cinder: Convert InvalidVolumeMetadataSize to webob.exc.*  https://review.openstack.org/35621303:56
*** alonma has quit IRC03:59
*** Asaithambi has joined #openstack-cinder04:00
*** Asaithambi has quit IRC04:02
*** Asaithambi has joined #openstack-cinder04:03
*** coolsvap has joined #openstack-cinder04:04
*** GB21 has joined #openstack-cinder04:06
*** merooney_ has quit IRC04:06
*** Asaithambi has quit IRC04:09
*** Asaithambi has joined #openstack-cinder04:12
*** bardia has joined #openstack-cinder04:14
openstackgerritAnh Tran proposed openstack/cinder: api-ref: Fix parameters attributes  https://review.openstack.org/35621604:17
*** bardia has quit IRC04:18
*** GB21 has quit IRC04:23
*** raunak has quit IRC04:23
*** bswartz has quit IRC04:27
*** cknight1 has quit IRC04:28
*** sticker has joined #openstack-cinder04:29
openstackgerritGuy Kim proposed openstack/cinder: ITRI DISCO cinder volume driver  https://review.openstack.org/34710004:31
*** raunak has joined #openstack-cinder04:34
*** alonma has joined #openstack-cinder04:35
*** ducttape_ has joined #openstack-cinder04:36
*** alonma has quit IRC04:40
*** ducttape_ has quit IRC04:41
*** alonma has joined #openstack-cinder04:43
*** laughterwym has joined #openstack-cinder04:45
*** alonma has quit IRC04:47
openstackgerritSoffie Huang proposed openstack/cinder: Refactoring FalconStor FSS storage driver  https://review.openstack.org/35583604:48
*** lkuchlan has joined #openstack-cinder04:52
*** GB21 has joined #openstack-cinder04:54
*** ChubYann has quit IRC04:54
*** pbandark has joined #openstack-cinder04:55
*** narayrak has joined #openstack-cinder05:00
*** ChubYann has joined #openstack-cinder05:06
*** dramakri has quit IRC05:18
*** akapil has quit IRC05:21
*** akapil has joined #openstack-cinder05:23
*** alonma has joined #openstack-cinder05:34
*** gcb_ has joined #openstack-cinder05:35
*** sandanar has joined #openstack-cinder05:37
openstackgerritxiexs proposed openstack/cinder: Refactor the case test_storwize_vdisk_copy_ops  https://review.openstack.org/35619405:38
*** ducttape_ has joined #openstack-cinder05:38
*** alonma has quit IRC05:39
*** liverpooler has joined #openstack-cinder05:40
*** thumpba has quit IRC05:40
*** Poornima has joined #openstack-cinder05:41
*** alonma has joined #openstack-cinder05:41
*** ducttape_ has quit IRC05:43
*** alonma has quit IRC05:46
*** rcernin has joined #openstack-cinder05:46
*** alonma has joined #openstack-cinder05:49
*** thumpba has joined #openstack-cinder05:51
*** yangyape_ has quit IRC05:51
*** yangyapeng has joined #openstack-cinder05:52
*** alonma has quit IRC05:53
*** yangyape_ has joined #openstack-cinder05:54
*** yangyapeng has quit IRC05:54
*** zhangguoqing has joined #openstack-cinder05:54
zhangguoqingHi, I get a volumes list by cinderclient, but I can not get the user_id (volume.user_id) of anyone volume.05:55
zhangguoqinghow to get the volume's user_id?05:56
*** thumpba has quit IRC05:56
*** alonma has joined #openstack-cinder05:56
zhangguoqingIt's importmant for me, who can help me? thanks. :)05:57
zhangguoqing(Pdb) dir(volume)06:00
zhangguoqing['HUMAN_ID', 'NAME_ATTR', '__class__', '__delattr__', '__dict__', '__doc__', '__eq__', '__format__', '__getattr__', '__getattribute__', '__hash__', '__init__', '__module__', '__new__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__', '__weakref__', '_add_details', '_append_request_id', '_info', '_loaded', 'append_request_ids',06:00
zhangguoqing'attach', 'attachments', 'availability_zone', 'begin_detaching', 'bootable', 'created_at', 'delete', 'delete_image_metadata', 'detach', 'display_description', 'display_name', 'encrypted', 'extend', 'force_delete', 'get', 'get_pools', 'human_id', 'id', 'initialize_connection', 'is_loaded', 'manage', 'manager', 'metadata', 'migrate_volume', 'multiattach', 'os-vol-host-attr:host',06:00
zhangguoqing'os-vol-mig-status-attr:migstat', 'os-vol-mig-status-attr:name_id', 'os-vol-tenant-attr:tenant_id', 'promote', 'reenable', 'request_ids', 'reserve', 'reset_state', 'retype', 'roll_detaching', 'set_image_metadata', 'set_loaded', 'set_metadata', 'setup', 'show_image_metadata', 'size', 'snapshot_id', 'source_volid', 'status', 'terminate_connection', 'to_dict', 'unmanage', 'unreserve',06:00
zhangguoqing'update', 'update_all_metadata', 'update_readonly_flag', 'upload_to_image', 'volume_type', 'x_openstack_request_ids']06:00
*** alonma has quit IRC06:01
*** alonma has joined #openstack-cinder06:03
lixiaoy1zhangguoqing: seems summary result doesn't include user_id, you can check the detailed info of a volume which includes user_id06:06
*** akapil has quit IRC06:06
*** akapil has joined #openstack-cinder06:07
*** alonma has quit IRC06:08
zhangguoqingyes, I have added the detailed=true.06:08
zhangguoqingand I find a strange that it cant get user_id only in cinderclient in python shell env.06:09
lixiaoy1even you set detailed=true?06:09
zhangguoqingthen I use cinderclient in ceilometer, I can not obtain the volume.user_id06:09
zhangguoqingyes, i'm sure.06:10
zhangguoqingfrom cinderclient import client06:12
zhangguoqingcinder = client.Client('2', 'admin', 'password', 'admin', 'http://172.16.40.6:5000/v2.0')06:12
zhangguoqingsearch_opts = {'all_tenants': 1}06:12
zhangguoqingvolumes = cinder.volumes.list(detailed=True, search_opts=search_opts)06:12
zhangguoqingvolumes[0].user_id06:12
lixiaoy1I think you can check the url it accessed through this call06:14
zhangguoqingok, thank you. i will do more tests. :)06:15
openstackgerritOpenStack Proposal Bot proposed openstack/cinder: Imported Translations from Zanata  https://review.openstack.org/35624806:15
*** alonma has joined #openstack-cinder06:16
*** laughterwym has quit IRC06:18
*** laughterwym has joined #openstack-cinder06:20
*** alonma has quit IRC06:21
*** ebalduf has quit IRC06:21
*** alonma has joined #openstack-cinder06:24
*** laughterwym has quit IRC06:25
*** alonma has quit IRC06:28
*** e0ne has joined #openstack-cinder06:30
*** alonma has joined #openstack-cinder06:31
*** sofie has joined #openstack-cinder06:32
*** sandanar_ has joined #openstack-cinder06:32
*** sandanar has quit IRC06:35
*** alonma has quit IRC06:36
*** akapil_ has joined #openstack-cinder06:38
*** alonma has joined #openstack-cinder06:38
*** arecknag has joined #openstack-cinder06:39
*** ducttape_ has joined #openstack-cinder06:39
*** lkuchlan has quit IRC06:40
*** raunak has quit IRC06:41
*** akapil has quit IRC06:41
*** alonma has quit IRC06:43
*** sofie has quit IRC06:44
*** pgadiya has joined #openstack-cinder06:44
*** ducttape_ has quit IRC06:45
*** alonma has joined #openstack-cinder06:46
*** akapil_ has quit IRC06:48
*** raunak has joined #openstack-cinder06:49
*** alonma has quit IRC06:50
*** akapil has joined #openstack-cinder06:52
*** belmoreira has joined #openstack-cinder06:53
*** raunak has quit IRC06:55
*** alonma has joined #openstack-cinder06:56
*** sandanar__ has joined #openstack-cinder06:59
*** gouthamr has quit IRC07:00
*** markvoelker has quit IRC07:00
*** tesseract- has joined #openstack-cinder07:00
*** alonma has quit IRC07:00
*** sandanar_ has quit IRC07:02
*** raunak has joined #openstack-cinder07:03
*** alonma has joined #openstack-cinder07:03
*** lpetrut has joined #openstack-cinder07:04
*** akapil has quit IRC07:05
*** akapil has joined #openstack-cinder07:05
*** alonma has quit IRC07:07
*** raunak has quit IRC07:09
*** GB21 has quit IRC07:10
*** akapil has quit IRC07:10
*** alonma has joined #openstack-cinder07:10
openstackgerritMatan Sabag proposed openstack/cinder: Changes stats report for ScaleIO driver  https://review.openstack.org/35612007:14
*** alonma has quit IRC07:14
*** alonma has joined #openstack-cinder07:17
*** raunak has joined #openstack-cinder07:21
*** alonma has quit IRC07:22
*** alonma has joined #openstack-cinder07:25
*** andymaier has joined #openstack-cinder07:26
*** lpetrut has quit IRC07:29
*** alonma has quit IRC07:29
*** aohuanxuan has joined #openstack-cinder07:29
*** alonma has joined #openstack-cinder07:32
openstackgerrityuyafei proposed openstack/os-brick: Multipath device keeps old size when extending volume  https://review.openstack.org/35332507:36
*** alonma has quit IRC07:36
*** alonma has joined #openstack-cinder07:39
*** lkuchlan has joined #openstack-cinder07:43
*** alonma has quit IRC07:44
*** GB21 has joined #openstack-cinder07:44
*** jamielennox is now known as jamielennox|away07:45
*** raunak has quit IRC07:45
*** alonma has joined #openstack-cinder07:46
*** sumit has joined #openstack-cinder07:46
*** alonma has quit IRC07:51
*** e0ne has quit IRC07:51
*** Dinesh_Bhor has joined #openstack-cinder07:52
*** thumpba has joined #openstack-cinder07:52
*** raunak has joined #openstack-cinder07:53
*** thumpba has quit IRC07:57
*** Asaithambi has quit IRC07:58
*** zzzeek has quit IRC08:00
*** markvoelker has joined #openstack-cinder08:00
*** thumpba has joined #openstack-cinder08:01
*** zzzeek has joined #openstack-cinder08:01
*** alonma has joined #openstack-cinder08:01
*** markvoelker has quit IRC08:05
*** alonma has quit IRC08:06
*** thumpba has quit IRC08:06
*** akapil has joined #openstack-cinder08:06
*** ducttape_ has joined #openstack-cinder08:08
*** alonma has joined #openstack-cinder08:08
*** coolsvap is now known as coolsvap_08:09
*** raunak has quit IRC08:09
openstackgerritDavanum Srinivas (dims) proposed openstack/cinder: [WIP] Testing latest u-c  https://review.openstack.org/30376608:10
*** ducttape_ has quit IRC08:12
*** alonma has quit IRC08:13
*** raunak has joined #openstack-cinder08:14
*** alonma has joined #openstack-cinder08:16
*** belmoreira has quit IRC08:18
*** e0ne has joined #openstack-cinder08:19
*** alonma has quit IRC08:20
*** alonma has joined #openstack-cinder08:23
*** alonma has quit IRC08:27
*** alonma has joined #openstack-cinder08:30
*** alonma has quit IRC08:30
*** alonma has joined #openstack-cinder08:30
*** Asaithambi has joined #openstack-cinder08:32
*** dkehn_ has quit IRC08:32
*** smile-luobin has left #openstack-cinder08:33
*** Jaison has joined #openstack-cinder08:35
*** links has quit IRC08:37
openstackgerritGorka Eguileor proposed openstack/os-brick: Fix cmd execution stderr, stdout unicode errors  https://review.openstack.org/34108508:43
*** jamielennox|away is now known as jamielennox08:50
*** Asaithambi has quit IRC08:50
*** Asaithambi has joined #openstack-cinder08:50
*** dkehn_ has joined #openstack-cinder08:52
*** GB21 has quit IRC08:52
*** lixiaoy1 has quit IRC08:59
*** raunak has quit IRC08:59
*** markvoelker has joined #openstack-cinder09:01
*** thumpba has joined #openstack-cinder09:02
*** markvoelker has quit IRC09:06
*** GB21 has joined #openstack-cinder09:06
*** thumpba has quit IRC09:08
*** ducttape_ has joined #openstack-cinder09:08
openstackgerritYanLin Ren proposed openstack/cinder: Volume Manage/Unmanage Support for IBM FlashSystem  https://review.openstack.org/35385609:09
*** raunak has joined #openstack-cinder09:10
*** ducttape_ has quit IRC09:13
*** raunak has quit IRC09:15
openstackgerritMatan Sabag proposed openstack/cinder: Changes stats report for ScaleIO driver  https://review.openstack.org/35612009:16
xiexsIs gate-cinder-docs-ubuntu-xenial broken?09:19
*** gus has quit IRC09:22
*** andymaier has quit IRC09:23
*** alonma has quit IRC09:27
*** lpetrut has joined #openstack-cinder09:27
*** alonma has joined #openstack-cinder09:27
*** Digvijay has joined #openstack-cinder09:28
*** maestropandy has joined #openstack-cinder09:29
*** raunak has joined #openstack-cinder09:30
*** alonma has quit IRC09:32
*** digu has joined #openstack-cinder09:33
*** alonma has joined #openstack-cinder09:34
*** mvk has quit IRC09:35
*** Digvijay has quit IRC09:37
zhangguoqinglixiaoy1: I get the volume's user_id after I change the server type of cinder to volumev2.09:38
*** alonma has quit IRC09:39
zhangguoqingbut I can not obtain the user_id of volume's snapshot.09:39
zhangguoqingand cinder snapshot-show volume-snapshot also can not see the user_id09:40
zhangguoqinghowever, I can see the user_id in database table.09:41
zhangguoqingSo, how to get the user_id of a volume's snapshot?09:42
*** alonma has joined #openstack-cinder09:42
*** jamielennox is now known as jamielennox|away09:43
*** alonma has quit IRC09:46
*** yangyapeng has joined #openstack-cinder09:48
*** alonma has joined #openstack-cinder09:49
*** yangyape_ has quit IRC09:50
*** alonma has quit IRC09:53
*** andymaier has joined #openstack-cinder09:53
*** yangyape_ has joined #openstack-cinder09:55
*** yangyapeng has quit IRC09:55
*** lpetrut1 has joined #openstack-cinder09:57
*** openstack has joined #openstack-cinder10:18
*** sdake_ has quit IRC10:18
*** e0ne has quit IRC10:18
*** sdake_ has joined #openstack-cinder10:19
*** sdake has quit IRC10:19
*** Poornima has quit IRC10:19
openstackgerritTuan proposed openstack/cinder: Stop using mox in cinder/tests/unit  https://review.openstack.org/35635810:20
*** yangyape_ has quit IRC10:21
*** dkehn_ has joined #openstack-cinder10:22
*** sdague has joined #openstack-cinder10:27
openstackgerritAvishay Traeger proposed openstack/python-cinderclient: List manageable volumes and snapshots  https://review.openstack.org/35127510:33
*** EinstCrazy has quit IRC10:40
*** raunak has quit IRC10:42
ArsenCHi cinder team, please help review https://review.openstack.org/#/c/341345/ . Thanks~10:42
openstackgerritEllen Leahy proposed openstack/python-cinderclient: Changed backup-restore to accept backup name  https://review.openstack.org/35277810:44
*** ociuhandu has quit IRC10:50
openstackgerritHussain Chachuliya proposed openstack/cinder: Use %()d for integer substitution  https://review.openstack.org/35638010:50
*** mvk has quit IRC10:58
openstackgerritHelen Walsh proposed openstack/cinder: VMAX Driver - Live Migration for VMAX3  https://review.openstack.org/33077810:59
*** vladani has quit IRC11:00
*** andymaier has quit IRC11:00
openstackgerritHussain Chachuliya proposed openstack/cinder: Use min attribute from oslo_config  https://review.openstack.org/35638811:04
*** avishay has joined #openstack-cinder11:09
*** ducttape_ has joined #openstack-cinder11:10
*** ducttape_ has quit IRC11:15
*** andymaier has joined #openstack-cinder11:19
*** vbala has joined #openstack-cinder11:20
*** yangyapeng has joined #openstack-cinder11:22
*** sdake_ is now known as sdake11:24
*** ociuhandu has joined #openstack-cinder11:28
*** Digvijay has joined #openstack-cinder11:30
*** alyson_ has joined #openstack-cinder11:31
*** digu has quit IRC11:32
*** bkopilov has quit IRC11:39
*** Digvijay has quit IRC11:40
*** lkuchlan_ has joined #openstack-cinder11:43
*** lkuchlan has quit IRC11:44
swamireddyvbala: hi11:46
vbalaswamireddy: hi11:47
swamireddyvbala: Updated the volume summary patch as per the review comments...can you please review the same...https://review.openstack.org/#/c/335067/11:48
vbalaok11:49
swamireddyvbala: Thank you11:53
openstackgerritErlon R. Cruz proposed openstack/cinder: Fix HNAS stats reporting  https://review.openstack.org/34447711:53
*** aohuanxuan has quit IRC11:56
*** julim has joined #openstack-cinder12:01
*** sdake_ has joined #openstack-cinder12:01
*** andymaier has quit IRC12:02
*** coolsvap_ is now known as coolsvap12:03
*** gouthamr has joined #openstack-cinder12:04
*** tpsilva has joined #openstack-cinder12:04
*** sdake has quit IRC12:05
*** sdake_ is now known as sdake12:05
*** mtanino has joined #openstack-cinder12:07
*** avishay has quit IRC12:08
*** markvoelker has joined #openstack-cinder12:09
*** ducttape_ has joined #openstack-cinder12:10
*** jamielennox|away is now known as jamielennox12:11
*** andymaier has joined #openstack-cinder12:14
*** huanan_L has quit IRC12:15
*** cdelatte has joined #openstack-cinder12:16
*** huanan has joined #openstack-cinder12:16
*** sandanar__ has quit IRC12:17
*** alonma has quit IRC12:23
*** ducttape_ has quit IRC12:23
*** alonma has joined #openstack-cinder12:27
*** e0ne has joined #openstack-cinder12:28
*** GB21 has quit IRC12:28
*** gcb_ has quit IRC12:33
*** tommylikehu has joined #openstack-cinder12:33
*** nicolasbock has quit IRC12:35
*** vbala has quit IRC12:44
*** gcb_ has joined #openstack-cinder12:46
*** bswartz has joined #openstack-cinder12:48
*** mtanino has quit IRC12:49
openstackgerritxiexs proposed openstack/cinder: Add validation for the metadata properties  https://review.openstack.org/33091312:55
*** rlrossit has joined #openstack-cinder12:59
*** ducttape_ has joined #openstack-cinder13:00
*** tommylikehu has quit IRC13:00
*** Julien-zte has joined #openstack-cinder13:00
*** jungleboyj has joined #openstack-cinder13:02
*** EinstCrazy has joined #openstack-cinder13:03
*** alonma has quit IRC13:04
*** bkopilov has joined #openstack-cinder13:05
*** alonma has joined #openstack-cinder13:05
*** xyang1 has joined #openstack-cinder13:06
*** mvk has joined #openstack-cinder13:06
*** diablo_rojo has joined #openstack-cinder13:08
*** alonma has quit IRC13:09
*** alonma has joined #openstack-cinder13:09
*** xyang has joined #openstack-cinder13:09
*** coolsvap is now known as _coolsvap_13:11
nikeshmsmcginnis: hi13:13
*** Julien-zte has quit IRC13:14
scottdaWhat was the status of dynamic reconfiguration? was that diablo_rojo driving that?13:15
smcginnisnikeshm: Hi13:15
smcginnisscottda: Yes, I think so.13:16
xyange0ne: hi13:16
e0nexyang: hi13:16
smcginnisDoctor13:16
xyange0ne: so I used a constant in the migration script now.  is that ok13:16
e0nexyang: it should be13:16
scottdaI see Dynamic Config spec merged. I don't see any link to code...13:17
nikeshmsmcginnis: we want to add replication failback support in Kaminario driver, our code is ready, do it need bp13:17
xyange0ne: did  you see my other replies to your comments?  have they addressed your concerns?13:17
e0nexyang: I don't remember if we've got final agreement on that bug, but it sounds reasonable for me13:17
xyange0ne: ok, thanks13:17
smcginnisnikeshm: No, for a driver feature implementation like that it is not really necessary.13:18
xyange0ne: the link you gave has a patch that was abandoned13:18
e0nexyang: maybe :)13:18
e0nexyang: I've got only one consern about your patch now13:19
xyange0ne: what is it13:19
nikeshmsmcginnis: can i change version history because we are adding failback support and do we need releasenote?13:19
*** mriedem has joined #openstack-cinder13:20
e0nexyang: accodning to tests for https://review.openstack.org/#/c/322459/58/cinder/group/api.py13:20
e0nexyang: I would like to get more unit tests than cross-module testing13:20
*** Julien-zte has joined #openstack-cinder13:20
e0nexyang: but community could not agree with me on it13:21
xyange0ne: oh, so just test this class, not to start from api url?13:21
e0nexyang: yes13:21
*** merooney has joined #openstack-cinder13:21
e0nexyang: we can ask about this case to others cores and community13:21
xyange0ne: I could add some tests there if you want13:21
smcginnisnikeshm: A release note would be good to let users know of the capability. Not sure what you mean by change version history.13:21
xyange0ne: sure13:21
nikeshmVersion history:         1.0 - Initial driver         1.1 - Added manage/unmanage and extra-specs support for nodedup         1.2 - Added replication support         1.3 - Added retype support13:21
*** EinstCrazy has quit IRC13:22
xyange0ne: I probably just followed what are there for other api unit tests13:22
smcginnisnikeshm: Oh, totally up to you, but I think it's good practice to list stuff like that in the docstring.13:22
e0nexyang: IMO, if we test more code by one test, it's easier to support it13:22
e0nexyang: fair enouph :)13:22
e0nexyang: it's hard to -1 if we've already get such things in the code13:23
xyange0ne: :)13:23
xyange0ne: let me take a look and see what tests I can add.13:25
*** porrua has joined #openstack-cinder13:25
e0nexyang: thanks13:25
xyange0ne: thank  you for reviewing and testing!13:25
e0nexyang: anyway, I would not -1 on it. maybe, I'll just put +113:25
e0nexyang: np13:25
openstackgerritxiexs proposed openstack/cinder: Refactor the using of dict.get() in the test assertion  https://review.openstack.org/35532313:26
e0nexyang: I'm going review and test your snapshot-related patches later today13:26
xyange0ne: it can't go anywhere anyway now.  it can't pass ceph CI:(13:26
xyange0ne: thanks!13:26
*** ducttape_ has quit IRC13:26
*** akerr has joined #openstack-cinder13:29
cFoutsDuncanT: https://review.openstack.org/#/c/346135 would appreciate a review if you have time. fyi, the eseries CI failure doesn't apply to this patch. Thanks!13:29
*** Asaithambi has quit IRC13:29
DuncanTcFouts: Is somebody looking at the eseries failure? Broken CIs make everybody's life difficult13:31
cFoutsDuncanT: we are having trouble getting all 8 tests to pass successfully in one pass. We have a number of user stories in the backlog to improve internal and upstream CI once we get to feature freeze :)13:32
*** dustins has joined #openstack-cinder13:32
DuncanTcFouts: Thanks. Good to hear it's being looked at.13:33
*** Digvijay has joined #openstack-cinder13:36
*** cknight has joined #openstack-cinder13:36
*** edmondsw has joined #openstack-cinder13:37
*** mtanino has joined #openstack-cinder13:37
*** tongli has joined #openstack-cinder13:39
*** ametts has joined #openstack-cinder13:40
*** Jaison has quit IRC13:46
openstackgerritvenkata krishna proposed openstack/cinder: Add replication failback in Kaminario K2 drivers  https://review.openstack.org/35647513:48
openstackgerritMerged openstack/python-brick-cinderclient-ext: Add support of 'local-attach' feature for RBD volumes  https://review.openstack.org/35583113:50
*** sdake has quit IRC13:50
*** xyang1 has quit IRC13:51
*** xyang1 has joined #openstack-cinder13:51
scottdae0ne: I think that's the last python-brick-cinderclient-ext patch that needs to merge ^^^ . Can we remove the 'review reminder' from the meeting agenda?13:51
e0nescottda: sure. I'll do13:52
scottdae0ne: Cool.13:52
e0nescottda: and I'll add one more patch... but I don't want to take a meeting time on it13:52
e0nescottda: thanks13:52
*** thumpba has joined #openstack-cinder13:52
scottdae0ne: Not a big deal to have the patch in the meeting, I just wanted to remove that big, giant URL :)13:52
e0nescottda: I hope, we'll get more contribution to python-brick-cinderclient-ext in O release13:53
scottdae0ne: It sucks that the wiki won't let us use tinyURL.13:53
e0nescottda: I hope, we'll get more contribution to python-brick-cinderclient-ext in O release13:53
e0nescottda: I don't like to ask people for reviews, but python-brick-cinderclient-ext project is not popular now, so it's had to do it13:53
scottdae0ne: Yeah. Do you/we ever talk to Ironic people about their use cases and any wishlist from them?13:54
e0nescottda: we just merged attache feature last week13:54
*** Asaithambi has joined #openstack-cinder13:54
openstackgerritzhangguoqing proposed openstack/cinder: Show user_id when cinder snapshot-show  https://review.openstack.org/35647813:54
e0nescottda: I'll ask for tests and feedback on the next ironic meeting13:55
scottdae0ne: Sounds good.13:55
*** ducttape_ has joined #openstack-cinder13:55
e0nescottda: maybe, they'll want to wait for a release.. I don't know13:55
e0nescottda: and CI with Ironic is the next on my TODOs list13:55
*** xyang1 has quit IRC13:58
*** sdake_ has joined #openstack-cinder14:01
*** ociuhandu has quit IRC14:01
openstackgerritYanLin Ren proposed openstack/cinder: Volume Manage/Unmanage Support for IBM FlashSystem  https://review.openstack.org/35385614:01
*** zhangguoqing has quit IRC14:04
sdaguesmcginnis / hemna: did we get a plan forward on narrowing the races in os-brick ?14:08
openstackgerritMerged openstack/cinder: Change a typo error in the releasenote  https://review.openstack.org/35618614:10
smcginnissdague: In meeting. hemna put up a patch to remove the locks, but it didn't address retries.14:11
smcginnissdague: Was that retry piece we talked about on the nova side though?14:11
*** ociuhandu has joined #openstack-cinder14:12
*** hoobaman has joined #openstack-cinder14:14
hoobamanhi14:14
hoobamani got very bad performance while booting from image to a scaleio block volume14:15
hoobamanseems like the validation step qemu-img convert (validate image) is using non-bufffered direct IO14:15
hoobamanmax performance i can get is 8MB/s on a Gbit connection :/14:16
hoobamantake like 8-10 minutes to boot the instance14:16
hoobamanany pointers? :)14:16
*** Digvijay has quit IRC14:17
*** tommylikehu has joined #openstack-cinder14:19
diablo_rojoscottda: Yeah I haven't had much time to work on that, but when I start at the foundation next week I will get going on it.14:20
*** haplo37_ has joined #openstack-cinder14:22
hoobamanAny pointers on how I can increase this performance?14:23
*** Asaithambi has quit IRC14:24
smcginnisxyang: Have you seen this? ^14:24
xyang@smcginnis: which one?14:24
sdaguesmcginnis: removing the locks doesn't solve the issue, it makes the races worse14:24
smcginnisxyang: hoobaman was asking about image copy performance with scaleio.14:24
xyang@smcginnis: oh, okay14:24
sdaguesmcginnis: my understanding was the retry logic would be in os-brick itself14:25
smcginnissdague: Not sure it can be. The retry was to address not having to have locks.14:25
sdaguesmcginnis: the retry was to mitigate cinder logging out isciadm while nova was attaching14:26
scottdadiablo_rojo: Cool. It just came up at a standup meeting. I'm happy to help if I can, so keep me in the loop.14:26
xyanghoobaman: what's the size of the volume14:26
sdagueif you remove the locks all you end up with is now the possibility of nova racing itself14:26
smcginnissdague: Right. So if we don't lock, that is the risk. So with no locks we could theoretically just retry since a new attach won't have any IO going.14:26
sdaguebecause this is eventlet14:26
diablo_rojoscottda: Awesome. I definitely will.14:27
sdaguesmcginnis: no, there are multiple levels of locking here14:27
smcginnisThe locks keep that from happening (if they worked) by not allowing an attach and detach to overlap.14:27
sdaguethere is a single process don't screwing itself, still a good idea to happen14:27
sdagueand there is global cluster lock, which can not work14:27
sdaguebut you can mitigate against the lack of global cluster lock with a retry14:27
smcginnissdague: The root of this was that even with a single node we can't lock across nova-cinder, so effectively it is always a global lock, right?14:28
sdaguesmcginnis: no14:28
sdaguethat lock *also* prevents nova-compute from racing itself14:29
sdaguein a single process14:29
sdaguebecause it is a single process eventlet system14:29
sdagueso every io changes green threads14:29
smcginnissdague: Guess I don't know enough of the nova side to see how that happens.14:29
sdagueso nova-compute can be detaching and attaching at the same time14:29
sdaguethat's a thing you want to prevent14:30
smcginnissdague: Seems like that should be locked on the nova side, but with the locks as they are now that should be mitigated. So I don't see with the locks why we would need retries.14:30
sdaguesmcginnis: the lock is in os-brick14:30
sdaguesmcginnis: ....14:30
sdaguethe issue14:31
smcginnissdague: Right, but this came about because cinders lock dir is different than novas lock dir.14:31
smcginnisAt least that was my understanding.14:31
sdagueis that c-vol and n-cpu can be on the same host14:31
smcginnisYes14:31
sdagueand doing the same isciadm commands14:31
sdagueand there is no workable model to preventing that14:31
smcginnisThat's how it was described to me.14:31
sdague*in addition*14:31
sdaguethe locking constraints is os-brick, also prevent, correctly c-vol and n-cpu from doing pathelogically stupid things14:32
sdaguelike logging themself out14:32
sdagueyou have 2 levels of conflict here14:32
sdaguecan n-cpu stomp itself14:32
sdaguecan c-vol stomp itself14:32
sdaguecan n-cpu <=> c-vol stomp each other14:33
*** pbandark has quit IRC14:33
sdaguethe locks as is prevent the first 1 items14:33
sdagueand can't prevent the 3rd14:33
smcginnisThat third case was the only one where I was aware there was any kind of issue here.14:33
sdaguesmcginnis: because of the locks14:33
sdagueif you remove the locks14:33
sdagueyou get 1 & 214:33
sdagueas well14:33
hoobamanxyang: thx for your reply, the size of the volume is 6GB14:34
smcginnissdague: I agree. But someone was pushing last time that the locks were useless and we should just do retries if we actually hit an issue.14:34
smcginnisBecause of case 3.14:34
hoobamanxyang: however I did some tests with a qcow2 image of 900MB, the behaviour is the same14:34
sdaguesmcginnis: the lock is useless in case 314:34
sdagueso we need a retry there14:34
hoobamanxyang: i mean the size of the image is 6GB14:34
hoobamanxyang: volume is 16GB14:35
xyanghoobaman: I'll have to check with scaleio QA and see what they get.  do you have a email address I can get back to you14:35
sdagueremoving the locks otherwise is like saying, "I'm not sure I trust my airbags, so I'm going to cut out my seat belts"14:36
smcginnissdague: Right. I think we are arguing the same point here basically.14:36
smcginnisI wasn't the one pushing for lock removal.14:36
smcginnisBut I thought the plan was to add retries.14:36
smcginnisBut I also thought that had to happen above brick.14:37
sdaguesmcginnis: ok, that's what I was asking.14:37
sdagueno, I think that happens in brick14:37
sdagueinside the locked area14:37
sdaguebecause c-vol is going to have the same issues where nova can stomp it14:37
sdagueanyway, this is why I asked the question14:38
*** pgadiya has quit IRC14:38
smcginnisWell, since it can only happen in the third case, where the locks can become useless, I don't think it has to happen within the locked area.14:38
sdaguebecause I think the answer is "there is no plan"14:38
*** jdurgin1 has joined #openstack-cinder14:38
sdaguesmcginnis: THE LOCKS ARE NOT USELESS THERE14:38
sdaguebecause they are actively preventing #1/#2 from happening14:39
smcginnisIn the third case? Mostly useless.14:39
smcginnisI'm only talking about three.14:39
sdagueright, but it's not like while 3 is happening, 1 & 2 aren't also able to happen14:39
smcginnisIn 1 and 2 we don't have a problem because we ARE locking and therefore we can't have a logout happen while trying to login.14:39
sdagueyou could have 5 simultaneous conflicts happening, 3 x 1, 2 x 314:40
*** xinli has joined #openstack-cinder14:40
sdagueyou are defending against the case 114:40
sdagueand you need to do the retry while those defenses are up14:40
smcginnisOK, semantics, but in those cases I consider them all to be case 3.14:40
sdaguesmcginnis: but they aren't14:40
sdaguethey really really aren't14:40
smcginnisIt only becomes an issue for 1 & 2 if you also have n-cpu and c-vol on the same host as well as different hosts, so it ends up being case 3.14:41
sdaguesmcginnis: except, not entirely14:41
smcginnisWell again, semantics. Doesn't change the condition that needs to be protected against, no matter how we want to label them.14:42
sdaguebecause you can either decide that you are handling the concurency of case 3 inside the context where the case 1 concurrency is addressed, and those requests are going to process in some order14:42
sdagueor you can give up all handling, let everything race back in again, including the case 1 issues14:42
sdaguemaybe you can make them the same, but the handling seems a lot more straight forward if you do it in the context of case 1 being a solved problem14:43
*** kfarr has joined #openstack-cinder14:44
*** diogogmt has joined #openstack-cinder14:47
*** arecknag has quit IRC14:47
*** erhudy has joined #openstack-cinder14:51
openstackgerritIvan Kolodyazhny proposed openstack/cinder: Merge two fake drivers into the one FakeLoggingVolumeDriver  https://review.openstack.org/35652514:53
openstackgerritIvan Kolodyazhny proposed openstack/cinder: Merge two fake drivers into the one FakeLoggingVolumeDriver  https://review.openstack.org/34842814:55
*** laughterwym has joined #openstack-cinder14:59
scottdaeharney, xyang1, gouthamr, akerr, smcginnis, cFouts, e0ne, geguileo, dulek, flip214, ntpttr patrickeast, _alastor_, DuncanT, erlon14:59
scottdaTest-Lovers, this is your summons...14:59
cFoutso/14:59
gouthamrhey o/14:59
erlonhey14:59
* DuncanT waves14:59
scottda#startmeeting cinder_testing14:59
openstackMeeting started Wed Aug 17 14:59:47 2016 UTC and is due to finish in 60 minutes.  The chair is scottda. Information about MeetBot at http://wiki.debian.org/MeetBot.14:59
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.14:59
openstackThe meeting name has been set to 'cinder_testing'14:59
patrickeasthey15:00
scottda#link https://etherpad.openstack.org/p/Cinder-testing15:00
ntpttr___o/15:00
e0nehi15:01
*** _coolsvap_ is now known as coolsvap15:01
*** enriquetaso has joined #openstack-cinder15:01
*** rhagarty_ has quit IRC15:01
scottdaFrom the agenda:15:02
scottda    devstack-gate cinder multi-node patch has merged: https://review.openstack.org/#/c/327619/15:02
scottdaIs dulek around?15:02
openstackgerritSean McGinnis proposed openstack/os-brick: Add retries to iSCSI connect_volume  https://review.openstack.org/35653215:02
scottdaOK15:03
scottdaNext on agenda:     Add a multibackend list to tempest.conf has merged: https://review.openstack.org/#/c/32589515:03
scottdaI'm waiting on re-approval of next patch in the series....15:03
*** openstackgerrit has quit IRC15:03
scottdaMain thing is that I've updated the status of the patches under review...15:04
scottdaI'll go to the QA meeting as well as infra channel and look for help on the ones that need final +2.15:04
scottdaAny help with that would also be good.15:04
*** openstackgerrit has joined #openstack-cinder15:05
scottdaand some patches have zero reviews, so please have a look.15:05
smcginnisscottda: Is there a common topic we can pull up all of them?15:07
*** raunak has joined #openstack-cinder15:07
patrickeastsmcginnis: nope, not yet anyway15:07
scottdasmcginnis: maybe...not sure. they are in at least 2 different repos: devstack-gate and devstack.15:07
smcginnisscottda: OK, probably easier just pulling up patches where you're the owner then.15:07
*** alonma has quit IRC15:07
*** xyang1 has joined #openstack-cinder15:07
*** alonma has joined #openstack-cinder15:07
scottdaI'll try to keep the status up-to-date on the etherpad. That may be the best way to track 'em15:07
smcginnisscottda: That sounds good.15:07
scottdaHas anyone other than me tried geguileo 's manual tests for AA-HA?15:08
scottda#link http://gorka.eguileor.com/manual-validation-of-cinder-aa-patches/15:08
scottdaWorks great. Ship it.15:08
geguileoXD15:09
smcginnisgeguileo: I haven't been able to run through and test, but very nice write up!15:09
*** Digvijay has joined #openstack-cinder15:09
geguileoI hope it helps understand the flow a little bit better15:09
geguileoAlthough it doesn't replace a good devref for adding clenable resources/states15:10
xyang1geguileo: do you want to add that to cinder doc15:11
e0neI'm going to test them once I finish groups-related patches testing15:11
*** ducttape_ has quit IRC15:11
geguileoxyang1: Not that one, I'll create a doc with an explanation on how to add a new cleanable resource/state15:11
*** alonma has quit IRC15:11
xyang1geguileo: ok15:11
geguileoxyang1: I don't think adding that info to the cinder doc is that useful15:11
*** maestropandy has quit IRC15:11
geguileoIt was more to help people get started in the testing and understand the pieces at play15:12
smcginnisA "how to" sounds useful.15:12
e0ne:)15:12
scottdaI'm going to re-test today with the latest set. And I threatened geguileo I would record it an put the webcast on youtube, as I did with api microversions. Some found that helpful...15:12
*** lcurtis has joined #openstack-cinder15:12
geguileoBecause sometimes it's hard just reading the specs15:12
scottdaAnyone have anything test-wise to bring up?15:13
smcginnisscottda: I finally wrote up test description changes.15:13
smcginnisAnd switched around what we were saying for "integration" vs "functional", so interested in what folks think of that.15:14
e0neI'm working on fake drivers refactoring and devstack integration. I hope, all patches will be ready this week.15:14
scottda#link https://review.openstack.org/#/c/356177/15:14
xyang1e0ne: I saw you submitted a patch on the fake driver.  so we should use that one for functional tests?15:14
smcginnisscottda: Beat me to it!15:14
scottdasmcginnis: looks -1'd by you?15:14
*** alonma has joined #openstack-cinder15:14
e0neand yes, fake driver could be use for functional and integration tests15:14
e0nexyang1: I hope so15:14
smcginnisscottda: Just to make sure everyone has a chance to take a look and raise any objections before someone decides to just push it through.15:14
e0nexyang1: I didn't test it with devstack, so it's still in WIP status15:15
scottdasmcginnis: Makes sense. Everyone please review.15:15
xyang1e0ne: ok15:15
scottda#link https://review.openstack.org/35652515:16
e0newe need refatcor unit/test_volume.py to get split between unit, funcitonal and integration tests15:16
scottdae0ne: 's fake driver patch ^^15:16
smcginnise0ne: +115:16
e0necorrect #link https://review.openstack.org/34842815:16
scottdaoops15:17
e0nescottda: that one is abandoned. I missed correct change-id15:17
scottdak15:17
e0ne#link https://review.openstack.org/348449 - patch to devstack15:18
e0nexyang1: I'll add FakeGateDriver support to devstack patch ^^ too15:18
xyang1e0ne: thanks:)15:18
*** Julien-zte has quit IRC15:18
*** alonma has quit IRC15:19
scottdae0ne: Would you like to put those links in the etherpad?15:19
e0nescottda: I'll do once I test my patch with devstack15:20
scottdathanks15:20
e0neeverybody is welcome to review ot now.. maybe I missed something15:20
e0neTBH, I removed few asserts from test_volume15:20
*** raunak has quit IRC15:21
scottdaSo, in general, we're getting good reviews from cinder folks on these test patches. I still don't know how to get more QA/infra people involved, but there's been some progress.15:21
*** alonma has joined #openstack-cinder15:21
scottda┬┐Hay mas? Anything else?15:23
patrickeasti had one thing maybe worth discussion15:23
scottdapatrickeast: Bring it15:23
patrickeastin the in-tree cg tempest tests https://review.openstack.org/#/c/355271/ i added a new config opt group for tempest.conf15:23
openstackgerritMatan Sabag proposed openstack/cinder: ScaleIO over subscription support  https://review.openstack.org/35654715:23
patrickeastinstead of re-using/adding to existing groups like "volume" or "volume-features-enabled" (or whatever it is)15:24
patrickeastwanted to make sure no one was strongly opposed to that... it is kinda weird to have so many sections basically for configuring volume/cinder stuff15:24
*** kfarr has quit IRC15:25
scottdapatrickeast: Thanks. We'll see what people think in the review...15:26
*** akapil has quit IRC15:26
*** alonma has quit IRC15:26
patrickeastscottda: haha yea, i take silence as a resounding "ship it"15:26
*** akapil has joined #openstack-cinder15:26
e0ne:)15:27
scottdaalrighty then....we can finish unless there are other items?15:27
scottda#endmeeting15:28
openstackMeeting ended Wed Aug 17 15:28:13 2016 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)15:28
openstackMinutes:        http://eavesdrop.openstack.org/meetings/cinder_testing/2016/cinder_testing.2016-08-17-14.59.html15:28
openstackMinutes (text): http://eavesdrop.openstack.org/meetings/cinder_testing/2016/cinder_testing.2016-08-17-14.59.txt15:28
scottdathanks!15:28
openstackLog:            http://eavesdrop.openstack.org/meetings/cinder_testing/2016/cinder_testing.2016-08-17-14.59.log.html15:28
openstackgerritJustin A Wilson proposed openstack/cinder: Added config option to enable SSL  https://review.openstack.org/35295815:29
*** alonma has joined #openstack-cinder15:29
*** tommylikehu has quit IRC15:29
*** kfarr has joined #openstack-cinder15:32
*** alonma has quit IRC15:34
karthikp_dulek: Hi15:34
*** alonma has joined #openstack-cinder15:37
smcginnispatrickeast: If you happen have time, that large test where we were getting overlapping attach/dettach against this would be good: https://review.openstack.org/#/c/356532/15:37
patrickeastsmcginnis: can do15:39
*** ducttape_ has joined #openstack-cinder15:40
*** alonma has quit IRC15:41
patrickeastsmcginnis: unfortunately i have a hunch that its not going to be so easy :(, that should fix the case where the attach command got screwed up by a detach logging it out but the more ugly problem is when two detaches race and both decide not to log out, we get dangling multipath devices and iscsi sessions15:41
patrickeastthose ones require manual cleanup on the initiator :*(15:42
patrickeastsmcginnis: good news though, its super rare to have c-vol and n-cpu on the same node... so i'm not really concerned about this too much15:42
patrickeastwell other than devstack15:43
smcginnispatrickeast: True.15:43
smcginnispatrickeast: I'd rather have a few useless sessions hanging around than inaccessible volumes though. ;)15:43
smcginnispatrickeast: But we should probably think about a better way to handle this. Or have some kind of cleanup processs.15:44
*** alonma has joined #openstack-cinder15:44
patrickeastuh, no, the broken session break all future attachments with the same iqn+lun15:44
patrickeastits way bad15:44
smcginnispatrickeast: Oh, good point. Didn't think of that.15:44
*** thumpba has quit IRC15:44
patrickeastsmcginnis: it doesn't affect all backends though15:45
patrickeastsmcginnis: things like lvm that have unique iqns and always are lun 0 its no problemo15:45
smcginnisI think just yours and mine and the few others that have shared target ports.15:45
patrickeastyep15:45
patrickeastsmcginnis: anyway, i'll give that retry patch a try... seems like we should be doing that anyway, regardless of the locking situation15:47
smcginnispatrickeast: Cool, thanks!15:47
openstackgerritMerged openstack/os-brick: Wrong param makes exception message throws inaccurate  https://review.openstack.org/35389215:48
*** alonma has quit IRC15:49
openstackgerritJustin A Wilson proposed openstack/cinder: Added config option to enable SSL  https://review.openstack.org/35295815:49
*** lprice has quit IRC15:49
*** lprice has joined #openstack-cinder15:50
*** jdurgin1 has quit IRC15:50
*** alonma has joined #openstack-cinder15:51
*** lkuchlan_ has quit IRC15:52
*** leeantho has joined #openstack-cinder15:53
hemnascottda, ping15:55
*** alonma has quit IRC15:55
scottdahemna: Hi15:55
hemnascottda, do you happen to have links to the nova -> cinder workflows that I created15:56
hemnaI can't seem to find them15:56
scottdahemna: They are here https://etherpad.openstack.org/p/cinder-nova-api-changes15:56
hemnathank you!15:57
*** ebalduf has joined #openstack-cinder15:57
*** rcernin has quit IRC15:58
*** sdake_ is now known as sdake16:00
*** sdake is now known as sdake_16:00
*** alonma has joined #openstack-cinder16:02
*** andymaier has quit IRC16:03
*** wewe0901 has joined #openstack-cinder16:04
*** lprice1 has joined #openstack-cinder16:04
*** laughterwym has quit IRC16:05
*** dmellado is now known as dmellado|off16:05
*** lprice has quit IRC16:06
*** alonma has quit IRC16:07
*** Apoorva has joined #openstack-cinder16:08
*** Digvijay has quit IRC16:15
*** netapp has joined #openstack-cinder16:16
*** narayrak has quit IRC16:21
openstackgerritAnthony Lee proposed openstack/cinder: Disabling online volume copy in HPE 3PAR drivers  https://review.openstack.org/34806016:25
openstackgerritWalter A. Boring IV (hemna) proposed openstack/cinder: WIP Add new supported driver checks  https://review.openstack.org/35560816:29
*** cknight has quit IRC16:31
*** cknight has joined #openstack-cinder16:31
*** bwallis has quit IRC16:36
openstackgerritSteve Martinelli proposed openstack/python-cinderclient: deprecate command `cinder endpoints`  https://review.openstack.org/34960216:37
*** JoseMello has joined #openstack-cinder16:37
*** Suyi has joined #openstack-cinder16:40
*** raunak has joined #openstack-cinder16:46
*** kfarr has quit IRC16:46
*** xyang has quit IRC16:46
e0nejungleboyj: hi! I answered to your comment for https://review.openstack.org/#/c/352924/16:50
*** xyang has joined #openstack-cinder16:50
*** lpetrut has quit IRC16:52
*** venkhari has joined #openstack-cinder16:54
*** sandanar has joined #openstack-cinder16:56
*** bkopilov has quit IRC16:59
*** tesseract- has quit IRC17:00
*** sandanar_ has joined #openstack-cinder17:01
* hemna is off to HR re-education camp....17:02
* jungleboyj is ignoring the invite to HR re-education camp.17:02
* diablo_rojo sends a tasteful reply to the HR re-education camp invite17:03
*** chris_morrell has joined #openstack-cinder17:04
*** sandanar has quit IRC17:04
*** bwallis has joined #openstack-cinder17:05
*** e0ne has quit IRC17:05
*** jungleboyj has quit IRC17:06
*** liverpooler has quit IRC17:07
*** jdillaman has quit IRC17:12
*** jdillaman has joined #openstack-cinder17:13
*** lpetrut has joined #openstack-cinder17:15
*** bardia has joined #openstack-cinder17:16
openstackgerritYuriy Nesenenko proposed openstack/cinder: Separate create and update rules for volume metadata  https://review.openstack.org/35662417:21
*** rcernin has joined #openstack-cinder17:26
*** lpetrut has quit IRC17:27
*** mvk has quit IRC17:31
*** akapil has quit IRC17:36
openstackgerritMike Rooney proposed openstack/cinder: NetApp: Report multiattach as enabled  https://review.openstack.org/35663517:40
*** stevemar_ has joined #openstack-cinder17:43
*** stevemar_ has quit IRC17:43
*** bkopilov has joined #openstack-cinder17:49
scottdaWill the cinderclient version after 1.9.0 be 2.0.0 ? or would it be 1.10.0 ?17:50
smcginnisscottda: I think 1.10.017:53
smcginnisscottda: We probably don't want to move to 2.0 until there is a major change.17:53
smcginnisAlthough..17:53
smcginnisWhen did we add microversion support?17:53
*** alyson_ has quit IRC17:53
scottdasmcginnis: last release...I think that'd be 1.8.017:55
smcginnisscottda: OK. That maybe would have warranted bumping to 2.0, but too late for that. So 1.10.0.17:56
scottdacool, thx17:56
*** alyson_ has joined #openstack-cinder18:00
*** jordanP has joined #openstack-cinder18:04
*** pauloewerton has joined #openstack-cinder18:04
*** mvk has joined #openstack-cinder18:07
*** dkehn_ has quit IRC18:08
*** alonma has joined #openstack-cinder18:10
bswartzsmcginnis, hemna: who does release management for os-brick?18:12
*** mtanino has quit IRC18:13
*** sandanar_ has quit IRC18:13
*** alonma has quit IRC18:14
hemnabswartz, the cinder team.18:15
hemnait's owned and managed by cinder just like the cinderclient is18:15
bswartzhemna: who has the power to push tags though?18:15
bswartzI'm curious when the next tagged release of os-brick is planned18:15
hemnaI think anyone can post a review in gerrit to submit a new release18:16
bswartzhemna: oh18:16
hemnawell, our deadline is next week18:16
smcginnisbswartz: It's all managed through the normal release team process.18:16
hemnaso I presume we'll do another drop next week18:16
bswartzokay I figured next week18:16
smcginnisI'm thinking probably late Wednesday.18:16
bswartzI forgot about the fact that anyone can push a commit to the releases repo18:16
*** alonma has joined #openstack-cinder18:20
*** dkehn_ has joined #openstack-cinder18:20
*** merooney has quit IRC18:21
*** merooney has joined #openstack-cinder18:22
*** merooney has quit IRC18:22
*** rooneym has joined #openstack-cinder18:23
*** Lee1092 has quit IRC18:23
*** ociuhandu has quit IRC18:24
*** jungleboyj has joined #openstack-cinder18:24
*** alonma has quit IRC18:25
openstackgerritMatan Sabag proposed openstack/cinder: ScaleIO over subscription support  https://review.openstack.org/35654718:25
*** _ducttape_ has joined #openstack-cinder18:27
*** narayrak has joined #openstack-cinder18:29
*** ducttape_ has quit IRC18:30
*** rlrossit_ has joined #openstack-cinder18:38
*** liverpooler has joined #openstack-cinder18:38
*** rlrossit has quit IRC18:41
xyang@smcginnis: ping18:45
*** ametts has quit IRC18:45
xyang@smcginnis: do you know if anyone is looking at gate-tempest-dsvm-full-devstack-plugin-ceph failure?  I'm getting that in several patches and won't go away after rechecks18:46
hemnaxyang, DuncanT mentioned that he was proposing making it experimental because it's so unstable18:47
xyanghemna: ok, thanks18:47
hemnanot sure about the short term though, it's puking a bunch from what I see18:48
xyanghemna: can you take a look of this patch when  you get a chance? https://review.openstack.org/#/c/317060/  You reviewed it before18:48
hemnasure18:49
xyanghemna: thanks18:49
openstackgerritvenkata krishna proposed openstack/cinder: Add replication failback in Kaminario K2 drivers  https://review.openstack.org/35647518:52
*** hoonetorg has quit IRC18:53
nikeshmif some one get time please review https://review.openstack.org/#/c/35647518:54
*** ametts has joined #openstack-cinder18:56
*** venkhari has quit IRC18:56
*** kfarr has joined #openstack-cinder18:56
*** _ducttape_ has quit IRC18:56
openstackgerritxing-yang proposed openstack/cinder: Add generic volume groups  https://review.openstack.org/32245918:56
*** ducttape_ has joined #openstack-cinder18:57
*** lpetrut has joined #openstack-cinder18:57
openstackgerritxing-yang proposed openstack/cinder: Add generic volume groups  https://review.openstack.org/32245918:58
*** thumpba has joined #openstack-cinder19:05
*** jamielennox has quit IRC19:06
*** hoonetorg has joined #openstack-cinder19:06
*** fifieldt has quit IRC19:07
*** xinli has quit IRC19:09
*** jamielennox has joined #openstack-cinder19:09
*** wewe0901 has quit IRC19:13
*** tongli has quit IRC19:15
*** xyang has quit IRC19:15
*** akerr has quit IRC19:15
*** xyang has joined #openstack-cinder19:16
*** fifieldt has joined #openstack-cinder19:18
stevemarsmcginnis: why are you so hard to find in gerrit19:24
*** ociuhandu has joined #openstack-cinder19:25
stevemarahhh sean_... too many seans19:25
stevemarnot like there aren't too many steve's or anything19:25
*** edmondsw has quit IRC19:28
*** _ducttape_ has joined #openstack-cinder19:28
smcginnisstevemar: Hah!19:30
stevemar:)19:30
smcginnisxyang1: I'm not sure if anyone is working on that actually.19:31
*** xinli has joined #openstack-cinder19:32
*** ducttape_ has quit IRC19:32
xyang@smcginnis: oh, thanks19:33
mriedemdoes this really warrant an ERROR and stacktrace? http://logs.openstack.org/78/346278/5/check/gate-tempest-dsvm-neutron-multinode-full/71882bb/logs/screen-c-api.txt.gz?level=TRACE#_2016-08-16_15_45_02_98919:33
mriedemit's just a 40019:33
mriedemditto http://logs.openstack.org/78/346278/5/check/gate-tempest-dsvm-neutron-multinode-full/71882bb/logs/screen-c-api.txt.gz?level=TRACE#_2016-08-16_15_41_08_29119:34
mriedemand this is just a 409 http://logs.openstack.org/78/346278/5/check/gate-tempest-dsvm-neutron-multinode-full/71882bb/logs/screen-c-api.txt.gz?level=TRACE#_2016-08-16_15_23_06_31519:34
mriedemwhich isn't an ERROR19:34
smcginnismriedem: Well, it's kind of an error, right?19:34
smcginnismriedem: Probably better as just a warning though.19:35
mriedemis this something you'd page an operator for?19:35
smcginnisJust for fun.19:35
smcginnisBut fair point. I can put up a patch to change it to warning.19:35
mriedemas long as jungleboyj is getting that page19:35
mriedemi'll open a bug19:35
mriedemthere are a few of these19:35
smcginnismriedem: :)19:35
mriedemhttps://bugs.launchpad.net/cinder/+bug/161423619:37
openstackLaunchpad bug 1614236 in Cinder "c-api should not log ERROR tracebacks for 400/409 cases" [Undecided,New]19:37
smcginnismriedem: Hmm, it's that the check for the size raises an exception (which is valid) but the taskflow executor doesn't know the difference so it's at error level.19:37
mriedemrm -rf taskflow19:38
smcginnisDon't start that again. :]19:38
mriedemso,19:38
mriedemi'm guessing taskflow needs some kind of expected_exceptions list or something?19:38
mriedemoslo.messaging has a thing like that19:38
smcginnismriedem: Probably19:38
smcginnismriedem: That handling is fairly simple: https://github.com/openstack/taskflow/blob/master/taskflow/engines/action_engine/executor.py#L4719:39
mriedemwhere does it log it?19:40
smcginnisTaskflow let's it bubble up. Still looking...19:40
*** jdillaman has quit IRC19:41
mriedemthe task must be defined in cinder19:41
*** liverpooler has quit IRC19:42
mriedemlooks like it would be a subclass of taskflow.atom.Atom19:42
mriedemoh i see it's the ExtractVolumeRequestTask.execute19:44
mriedemin cinder.flows.api.create_volume19:44
-openstackstatus- NOTICE: The volume for logs.openstack.org filled up rather suddenly, causing a number of jobs to fail with a POST_FAILURE result and no logs; we're manually expiring some logs now to buy breathing room, but any changes which hit that in the past few minutes will need to be rechecked and/or approved again19:44
scottdathat explains that19:46
smcginnismriedem: I can't look into right now. Want to comment that in the bug? Wonder if we should add taskflow to it.19:46
mriedemharlowja: ^ where does taskflow log failed task executions?19:47
harlowjacinder afaik plugged in there own logger19:47
harlowjaso the cinder log?19:47
harlowjacinder afaik plugged in one of https://github.com/openstack/taskflow/blob/master/taskflow/listeners/logging.py#L7119:48
harlowjaand provided there own logger19:48
harlowja(i just repeated the same thing)19:48
harlowjaha19:48
mriedemheh19:49
mriedem_NO_TRACE_EXCEPTIONS = (exception.InvalidInput, exception.QuotaError)19:49
*** _ducttape_ has quit IRC19:49
karthikp_scottda: Hi Scott, got a minute to discuss about our efforts on partial grenade?19:49
*** ducttape_ has joined #openstack-cinder19:50
scottdakarthikp_: sure19:50
* jungleboyj doesn't want more pages.19:50
harlowjaright mriedem seems like the traceback on failures isn't happening for those exceptions19:50
harlowja(for better or worse)19:51
karthikp_scottda: we just had a few of our patches merged for multi-node grenade19:51
karthikp_scottda: but the grenade job keeps failing as it tries to upgrade the cinder c-vol service19:52
karthikp_the reason is that we have excluded c-vol and c-bkp from features19:53
karthikp_https://review.openstack.org/#/c/327619/5/features.yaml19:53
scottdalooking...19:54
*** jungleboyj has quit IRC19:55
scottdakarthikp_: I guess I cannot say for sure the proper way to do this. I wonder if patrickeast could help, since he has a CI running multi-node. Maybe we can figure it out using his system?19:57
karthikp_scottda: so we got this patch in to check if c-vol is enabled or not https://review.openstack.org/#/c/355678/219:58
karthikp_scottda: i see ..let me ping him now19:58
scottdaI think I just did :)19:58
*** xyang has quit IRC20:04
openstackgerritTom Swanson proposed openstack/cinder: Dell SC: Break_replication unhandled NoneType error  https://review.openstack.org/35668720:06
Swanson+2/+A ^^^20:06
scottdaSwanson: What's that, Jedi mind trick?20:07
*** xyang has joined #openstack-cinder20:07
hemna-2!20:08
hemnano unit tests....-220:08
hemnafail!20:08
Swansonscottda, I've stopped with the subliminal messaging and am going Superliminal.20:08
hemnagreen color in the diff on your patch....-2!20:08
*** jungleboyj has joined #openstack-cinder20:08
Swansonhemna and jungleboyj switch names?20:09
hemnassshhh don't tell jungleboy20:09
*** coolsvap has quit IRC20:11
jungleboyjwhat?20:12
hemnadamn, he found out.20:12
Swansongreat. gonna get nailed for that missing apostrophe.20:14
jungleboyjSwanson: WHAT!?!  hemna has come to the punctuation side?20:15
*** alonma has joined #openstack-cinder20:15
*** porrua has quit IRC20:16
mriedemjungleboyj: https://bugs.launchpad.net/cinder/+bug/161424920:18
openstackLaunchpad bug 1614249 in Cinder "Clone doesn't work for NFS driver" [Undecided,New]20:18
*** alonma has quit IRC20:20
*** e0ne has joined #openstack-cinder20:20
*** alonma has joined #openstack-cinder20:24
*** cdelatte has quit IRC20:25
*** alonma has quit IRC20:29
*** xyang has quit IRC20:32
openstackgerritMatan Sabag proposed openstack/cinder: Changes stats report for ScaleIO driver  https://review.openstack.org/35612020:34
jgriffithmriedem: good lord, really!20:36
* jgriffith pretty much gives up at this point20:36
mriedemjgriffith: true story20:36
jgriffithmriedem: nice thing about Wiki's... anybody can enter anything they want20:37
mriedemwe moved that in tree a long time ago20:37
mriedemso the matrix is tied to the actual code when it lands20:37
jgriffithmriedem: yeah, I tried and had a mutiny20:37
jgriffithmriedem: mostly from folks who aren't even actively engaged around here any longer... go figure20:37
mriedemf'em20:38
mriedempush again20:38
jgriffithmriedem: let's keep adding features in Cinder that we don't test and that don't work20:38
openstackgerritMatan Sabag proposed openstack/cinder: ScaleIO over subscription support  https://review.openstack.org/35654720:38
jgriffithmriedem: wiki's are pure evil with this many people invovled20:38
mriedemfwiw, https://github.com/openstack/nova/blob/master/doc/source/support-matrix.ini20:38
mriedemthe nova matrix is based on an ini file20:38
bswartzYes, wikis are evil, let us use etherpads20:39
mriedemhttps://github.com/openstack/nova/blob/master/doc/ext/support_matrix.py20:39
jgriffithmriedem: very nice, thanks20:39
mriedem^ parses the ini and throws it into the rst20:39
hemnathat's cool20:39
mriedemgives http://docs.openstack.org/developer/nova/support-matrix.html20:39
jgriffithI'll follow that lead and start some doc refs20:39
patrickeastscottda: karthikp_: oh interesting, that change seems pretty straight-forward... I don't run grenade tests on my ci20:40
* bswartz realizes laptop battery is almost zero20:40
mriedemi'm guessing you'll rage quit because we have like what, 5 drivers? cinder has 40020:40
patrickeastbut can try it out if there is some specific thing we need tested20:40
*** rcernin has quit IRC20:40
erlonmriedem: thats beautiful :)20:41
mriedemall credit goes to danpb20:41
patrickeastmriedem: jgriffith: somethign like that could be useful with our new plan of tagging drivers that aren't tested20:41
jgriffithbswartz: good... no more smart ass comments from you today!20:41
jgriffithpatrickeast: exactly!!20:41
mriedempatrickeast: yeah johnthetubaguy has been working on that for awhile now in nova20:42
scottdapatrickeast: That'd be great. I think karthikp_ Has a theory, and if you can test it we'd save some time. But I'll let him give the details....20:42
jgriffithbswartz: unless of course you have a power supply and an outlet... then I'm doomed20:42
erlonjgriffith: isnt possible to pull that information from the drivers interface checks?20:42
mriedemhttp://docs.openstack.org/developer/nova/feature_classification.html20:42
smcginniserlon: jgriffith mriedem: I was actually looking at that at one point.20:43
jgriffitherlon: don't get me started... you'll hear my ABC rant and nobody wants that20:43
jgriffitherlon: on a positive note, smcginnis has started a sane approach to doing that.... and yes20:43
erlonjgriffith: haha, I now you love ABC20:43
jgriffitherlon: I'd think that doc should be compiled by code, NOT humans20:43
smcginnisjgriffith: +120:44
erlonsmcginnis: that would be even better/accurate than maintaining an INI file20:44
smcginnisI actually looked at how Nova is doing it. Not bad, but I think with the interface checking we have now (once we untangle the ABC stuff) I think it would be very easy to generate and publish the matrix.20:44
smcginniserlon: +120:44
karthikp_scottda: thanks scott. patrickeast; in order to perform a multi-node grenade, we disabled a few services on primary by https://review.openstack.org/#/c/327619/5/features.yaml20:45
patrickeastanother cool thing with the ini file, if we put whether or not the driver is tested/validated/whatever... eventually we just put a link to their out of tree git repo and...20:45
jgriffithsmcginnis: so I'm thinking use the interface checking to generate the matrix doc20:45
jgriffithsmcginnis: erlon I think that's the same thing you both are saying no?20:45
smcginnisjgriffith: Yep20:46
karthikp_patrickeast:Now, its failing our grenade tests fail as there is no c-vol enabled.... do you think it would be right to patch it like this20:46
erlonjgriffith: mhm20:46
karthikp_https://review.openstack.org/#/c/355678/20:46
smcginnispatrickeast: Oh, that would be an advantage of the ini approach.20:46
jgriffithsmcginnis: so where are we with getting that interface checking online?  Just a matter of consolidating the classes back to something sane again?20:46
jgriffithpatrickeast: oh... I like that20:47
smcginnisjgriffith: Yeah. I just have the job to do the basic interface checking.20:47
patrickeastkarthikp_: hmm well it seems like it would be ok, i'm not 100% sure what the right way of configuring whether or not we expect c-vol to be there or not20:47
jgriffithsmcginnis: maybe next week you and I can sync up on that?20:47
*** cknight has quit IRC20:47
smcginnisjgriffith: But until we stop using the crazy inheritance we have now where pretty much every driver has every interface, we can't do much more.20:47
smcginnisjgriffith: Sounds good.20:47
jgriffithI'd like to somehow do that this release if possible20:47
erlonpatrickeast: jgriffith: that would be possible with the interface too, the same way we added the CI WIKIs20:47
patrickeastkarthikp_: because sometimes we do expect it to be there, right? like for normal grenade jobs20:47
smcginnisjgriffith: I think it would be a tight schedule, but not impossible at this point.20:47
jgriffithsmcginnis: I think I still have a patch out there that removes all of that?  It'd need an update/rebase, but it at least exists20:48
hemnathe xio driver needs to extend the VolumeDriver20:48
hemnahttps://review.openstack.org/#/c/355608/7/cinder/volume/drivers/xio.py20:48
hemnaI noticed that in my supported driver check patch20:48
patrickeastkind of funny though that apparently you don't actually *need* to20:49
hemnapatrickeast, as long as you implement the functions that the manager calls....20:49
smcginnisWell, really you shouldn't _have_ to inherit. As long as you implement all the right calls and do them right, should be OK.20:49
mriedemjgriffith: https://review.openstack.org/#/c/351918/ would help get the nfs job in a less crazy full 100% failure state20:49
hemnathe problem was, my patch was puking because the xio driver didn't have the inherited method to check if it was supported20:50
hemnait should inherit IMHO20:50
jgriffithmriedem: yeah, I'm ok with that I think... but what about the derivitives?20:50
hemnaor lets stop doing OOP20:50
karthikp_patrickeast: yes you are right ..in a normal grenade job.. we do expect it to present..... we tweaked in that way to  that we can test cinder in a multi-node setup like c-api c-sch (master) and subnode to run mitaka c-vol and cbak20:50
jgriffithmriedem: they work I believe... ie NetApp NFS, Gluster etc20:50
smcginnishemna: More like OOPs20:50
jgriffithmriedem: that protocol check would exclude them as well I think... have to look20:51
erlonjgriffith: I *think* the patch will only affect the NFS job20:51
mriedemjgriffith: hmm, good point, not sure about that20:51
mriedemi was basing that on this https://github.com/openstack/devstack-plugin-nfs/blob/master/devstack/override-defaults20:52
* erlon checking20:52
jgriffitherlon: I think a few are inheritting the protocol from the parent class which is the NFS driver20:52
patrickeastkarthikp_: looking at the code I think whats there should be ok, on the assumption that the is_service_enabled function looks at like ENABLED_SERVICES or something (like devstack does)20:53
jgriffithmriedem: erlon yeah... netapp for example would get nailed... appears that Gluster would because it pulls protocol from parent20:53
jgriffithlikely others as well20:53
erlonmriedem: jgriffith: hmm, yes it will affect, the other NFS drivers report 'nfs'20:53
patrickeastkarthikp_: to make sure that we only verify c-vol is alive when we've configured c-vol to be enabled at all20:53
mriedemok20:54
erlonjgriffith: mriedem: I though you where changing the CI regex when calling tempest20:54
mriedemi think we might be able to handle it in the nfs devstack plugin anyway20:54
erlonmriedem: that would not affect others20:54
stevemarscottda: o/20:54
stevemarscottda: your comment here: https://review.openstack.org/#/c/34960220:54
jgriffithI suppose I should make sure everybody isn't just adding a skip on that test in their nfs derivative driver20:55
jgriffithThat would pretty much make my head explode20:55
stevemarscottda: i say the "P" release, since that in 2 cycles, or next major release of cinderclient, i don't believe we're allowed to remove things in "feature" releases (bumping from 1.9.0 to 1.10.0), normally when something is removed you bump the "major" number20:55
*** xyang has joined #openstack-cinder20:56
jgriffithhmm... thought I'd poke at the gluster logs... that didn't work out so well20:56
jgriffithhttp://logs.openstack.org/45/355945/2/check/gate-cinder-tempest-dsvm-glusterfs-nv/b7903cd/20:56
openstackgerritxing-yang proposed openstack/cinder: Add group snapshots  https://review.openstack.org/32805220:56
erlonmriedem: where would you need to change to change the tempest regex? In my CI is the jenkins job config, but I know know about the gate job20:56
jgriffithYeah, Gluster is running the clone tests succesfully20:57
scottdastevemar: You're saying cinderclient would bump the major number to 2.0.0 for removing the 'cinder endpoints' command? I wouldn't think we would do that. But smcginnis might say differently.20:58
*** edmondsw has joined #openstack-cinder20:58
stevemarscottda: i believe that is the case for any python lib that follows semver. if you knowingly make a breaking change you bump the major version20:59
stevemarscottda: first line of http://semver.org/20:59
openstackgerritKaitlin Farr proposed openstack/cinder: Replace key manager with Castellan  https://review.openstack.org/28049220:59
scottdastevemar: Yeah, I get that.20:59
stevemarscottda: *shrugs* number are cheap :)21:00
stevemarnumbers21:00
*** xyang has quit IRC21:00
stevemari don't particularly care either way, it's up to you all21:00
stevemarjust wanted to explain my reasoning21:00
*** e0ne has quit IRC21:00
scottdastevemar: That all makes sense. I just don't think we've been that disciplined about our cinderclient versioning.21:01
*** bill_az has joined #openstack-cinder21:01
scottdastevemar: I'll leave it up to smcginnis and others to mull over for a bit.21:01
stevemarscottda: considering you're still at 2, i would think not :D21:01
karthikp_patrickeast: cool thanks! Do these changes make sense to you ? by this I mean do you know if its being tested in a multi-node  like i discussed above21:01
stevemarerrr 1.021:01
scottdahaha21:02
stevemarscottda: i thought all openstack libs were bumped by one when we dropped py26 support...21:02
patrickeastkarthikp_: the changes make sense, dunno if we've actually merged a job that runs with that configuration21:02
jgriffithstevemar: NOTHING is cheap21:02
stevemarjgriffith: free lunches are!21:02
jgriffithstevemar: especially things that *seem* like they are21:02
patrickeastscottda: did we turn on the multi-node grenade job? is it in experimental?21:02
stevemarjgriffith: :O21:02
jgriffith:)21:02
karthikp_patrickeast: we did.... this https://review.openstack.org/#/c/334537/16/jenkins/jobs/devstack-gate.yaml21:03
openstackgerritSteve Martinelli proposed openstack/python-cinderclient: deprecate command `cinder endpoints`  https://review.openstack.org/34960221:03
*** gouthamr has quit IRC21:04
*** xinli has quit IRC21:04
smcginnisstevemar, scottda: Makes sense on dropping support.21:04
smcginnisSo it would stay 1.x for now and marked deprecated.21:04
smcginnisThen when we actually remove it we can bump up to 2.0.21:05
karthikp_patrickeast: scottda: we got that merged and it is experimental: gate-grenade-dsvm-cinder-multinode21:05
*** xyang has joined #openstack-cinder21:05
mriedemerlon: the job def in project-config21:05
scottdasmcginnis: stevemar Yup, it makes sense. We'll just have to make sure we continue to do this properly in the future, as we've never bumped the major version in the past..21:05
mriedemerlon: for nfs with snapshot and clone, we can just configure tempest rather than blacklist tests with a regex21:05
mriedems/can/should/21:06
stevemarsmcginnis: coolio21:06
patrickeastkarthikp_: gotcha, so yea it seems like that should be ok... but I'm also not an expert at grenade testing so theres a chance Im missing something ;)21:06
*** sdague has quit IRC21:06
stevemarscottda right now "who the hell does this keystone schmuck think he is - coming in here and ruining everything"21:06
erlonmriedem: hmm, and that would be in the devstack plugin21:07
mriedemerlon: yes, i plan on working on it21:08
*** narayrak has quit IRC21:08
scottdastevemar: Your work is welcome. And it's good to force us to think about things like cinderclient versioning, since we obviously missed some stuff.21:10
stevemar:)21:11
scottdastevemar: Hey, what was the deal with https://bugs.launchpad.net/cinder/+bug/1614104 ? Did you previously fix something like that?21:11
openstackLaunchpad bug 1614104 in Cinder "Cinder endpoints throws error in Keystone V3" [Undecided,New] - Assigned to Jay Conroy (jayconroy)21:11
*** kfarr has quit IRC21:13
*** fifieldt has quit IRC21:13
stevemarscottda: yep, it was right after the initial migration to keystoneauth21:13
stevemarscottda: i noticed the functional tests were failing21:13
stevemarand fixed it up21:13
stevemari think 1614104 is using a funky version21:13
scottdastevemar: ok. Can you put something in that bug? i.e. about when that fix was, or a link to it?21:14
*** diablo_rojo has quit IRC21:17
*** arch-nemesis has joined #openstack-cinder21:18
openstackgerritSteve Martinelli proposed openstack/python-cinderclient: deprecate command `cinder endpoints`  https://review.openstack.org/34960221:18
stevemarscottda: i sure can21:18
*** enriquetaso has quit IRC21:21
*** rlrossit_ has quit IRC21:22
karthikp_patrickeast: cool :). Thanks. I guess I summon support to get this one merged from infra team21:24
*** fifieldt has joined #openstack-cinder21:25
*** jungleboyj has quit IRC21:27
*** alonma has joined #openstack-cinder21:37
*** pauloewerton has quit IRC21:39
*** dustins has quit IRC21:41
*** alonma has quit IRC21:41
*** lpetrut has quit IRC21:44
*** xyang has quit IRC21:48
*** mdenny has quit IRC21:50
patrickeastjgriffith: hey, looking at https://review.openstack.org/#/c/351275/ did you have any strong objections to merging it as-is?21:51
patrickeasti think all the questions were answered, but kind of with a "nah, i like it this way", so i figured i'd check21:51
*** xyang has joined #openstack-cinder21:51
jgriffithyes, but not strong enough to fight over21:51
patrickeastjgriffith: with the api in general, yea?21:52
patrickeasti'm wondering if we have any leeway to change it up, or if its too late and we might as well ship the client since the api's are locked in forever : /21:52
smcginnisBut microversions fix everything! :)21:57
scottdasmart Alec21:57
patrickeastsmcginnis: well, kinda, we can update the api with a new microversion to make it less bad21:57
*** mriedem has quit IRC21:58
openstackgerritMerged openstack/cinder: Clean imports in code  https://review.openstack.org/35575221:58
patrickeastsmcginnis: and could even just have the client not support the other version of that api, in theory21:58
smcginnisI know, just joking about the panacea it was made out to be.21:59
smcginnisYep.21:59
*** xyang has quit IRC21:59
*** xyang has joined #openstack-cinder22:00
patrickeasttotally unrelated, has anyone tried multi-node FC recently?22:02
hemnanot I22:02
hemnawhat's up22:02
openstackgerritMerged openstack/cinder: Refactor cinder.utils._copy_volume_with_path  https://review.openstack.org/35426922:02
patrickeasti can't for the life of me get tempest to pass on multi-node jobs for it.... going to spin up a manual env and see if it still has problems22:03
patrickeasthemna: just tests failing, issues with the volume attach/detach kinda thing22:03
patrickeastlemmie dig up a link22:03
hemnalogs ?22:03
*** alyson_ has quit IRC22:03
*** nikeshm has quit IRC22:04
patrickeasthemna: both of these failed http://openstack-logs.purestorage.com/PureFCDriver-tempest-dsvm-volume-multi-node-multipath/22:04
patrickeastalthough I'm not 100% convinced the env was setup right22:04
*** ametts has quit IRC22:04
patrickeastthere are a variety of errors in the logs, but not sure yet who-done-it22:04
hemnahttp://openstack-logs.purestorage.com/PureFCDriver-tempest-dsvm-volume-multi-node-multipath/2/logs/screen-n-cpu.txt.gz?#_2016-08-16_21_39_31_54722:06
hemnathat's odd22:06
patrickeastthats the new thing we added in for the fc wildcard business22:06
*** sdake_ has quit IRC22:06
patrickeastanother "expected" failure : /22:06
patrickeastgg privsep22:06
openstackgerritTom Swanson proposed openstack/cinder: Dell SC: Break_replication unhandled NoneType error  https://review.openstack.org/35668722:07
hemnafc wildcard business ?!22:07
* patrickeast just assumes logging issues with os-brick are privseps fault now22:07
patrickeasthemna: there was a bug fix just recently...uhh22:07
* patrickeast looks for the link22:08
*** jordanP has quit IRC22:08
hemnahttps://github.com/openstack/os-brick/commit/28a4d55a0a465ac36ed012a2d634cb64e8f5d59922:08
hemnathat thing ?22:08
patrickeasthttps://github.com/openstack/os-brick/commit/28a4d55a0a465ac36ed012a2d634cb64e8f5d59922:08
patrickeasthah22:08
patrickeastyea22:08
patrickeastso it runs grep, and if it fails no biggie, its not that type of hba22:09
*** xyang has quit IRC22:09
patrickeastand it does a wildcard scan22:09
patrickeastwell, wildcard + lun22:09
patrickeastbut the grep output is logged as that error22:10
hemnaah ok, so yet another example of why privsep is a PITA to use22:10
hemnaas it always logs stderr22:10
hemnaeven though we are try/excepting it22:10
hemna:(22:10
patrickeastactually, now that i look at the change again it shouldn't be logging at error level in brick when we catch the exception22:10
patrickeastwe're doing part of that one : /22:11
patrickeastwell, if it failed for a real reason we should...22:11
patrickeastbleh22:11
hemnalooks like that grep is borked22:12
hemnaI dunno22:12
*** bardia has quit IRC22:12
patrickeastoh well, no one uses fc anyway22:12
*** bardia has joined #openstack-cinder22:13
hemnahrmm22:14
hemnacome to think of it22:14
hemnaI have 3 entries on my server in /sys/class/fc_transport22:14
hemnabut no volumes attached22:14
hemnaso I'm not sure this is correct to do now22:14
patrickeasti thought those were just the hba pci devices?22:15
hemnaI only have 2 HBAs22:15
hemnabut there are 3 entries in there22:15
patrickeastweird22:16
patrickeastuhh22:17
patrickeasthow many ports?22:17
patrickeastlooking at http://linuxmantra.com/2013/10/adding-san-storage-path.html i think you get one per port22:18
patrickeast3 though is still weird, unless maybe ones in link down state?22:18
hemnayah that's what I don't get22:18
hemnait should be 2 or 422:18
hemna3 ?22:18
hemnawtf22:18
patrickeastor has been gobbled up by libvirt?22:18
hemnaI only have 1 vagrant running and it doesn't have any fc22:19
* hemna is confused22:19
patrickeastweird22:19
*** JoseMello has quit IRC22:20
openstackgerritJustin A Wilson proposed openstack/cinder: Added config option to enable SSL  https://review.openstack.org/35295822:20
patrickeastare any of the devices pci_stub'ed or vifo'd?22:20
*** sdake has joined #openstack-cinder22:21
hemnaI have a bunch of other vms shut down22:21
hemnamaybe one of those have it22:21
openstackgerritJustin A Wilson proposed openstack/cinder: Added config option to enable SSL  https://review.openstack.org/35295822:22
patrickeastcan see with like lspci -vv and look for vfio or pci_stub as the driver22:23
hemnanada22:24
*** amrith has joined #openstack-cinder22:31
amrithsmcginnis, hello22:32
*** sdake has quit IRC22:38
*** lcurtis has quit IRC22:40
smcginnisamrith: Hey!22:42
amrithhi smcginnis ... I see you are having good luck with technology today. just imagine what people had to suffer through 60 years ago22:42
smcginnisamrith: I think postal mail would be better at this point. ;)22:42
amrithI'm good anytime from 7am eastern to 5pm eastern22:43
amrithy'all pick22:43
smcginnisamrith: OK, thanks! Hopefully we can get feedback and finalize quickly now.22:43
* smcginnis crosses fingers22:43
amrithi submitted my answers ...22:45
amrithsorry to put you through this22:45
amrithnow, you blocked 1h slots but we'll need > 1h :)22:45
smcginnisamrith: Haha, no problem. I guess I'm just too distracted today. It shouldn't have been this difficult.22:45
smcginnisamrith: We can see who is open for multiple slots as well. If we can all only do one hour and then we have to break out into smaller groups for another hour, we can work with that. ;)22:46
amrithtrue22:46
amrithand surprisingly enough, the answer is NO!22:47
amrithwell, ifneedbe is good enough22:47
amrithsounds good22:48
amrithg'night smcginnis22:48
smcginnisamrith: Goodnight. Thanks amrith.22:48
*** xyang has joined #openstack-cinder22:48
*** xyang has quit IRC22:54
*** ducttape_ has quit IRC22:58
*** xyang has joined #openstack-cinder22:59
*** xyang has quit IRC23:02
*** erhudy has quit IRC23:02
*** sumit7990 has joined #openstack-cinder23:05
*** sumit has quit IRC23:08
*** xyang1 has quit IRC23:10
*** tpsilva has quit IRC23:18
*** gouthamr has joined #openstack-cinder23:28
*** arch-nemesis has quit IRC23:44
*** diogogmt has quit IRC23:56

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!