Wednesday, 2015-07-08

*** dannywilson has quit IRC00:00
*** gouthamr has joined #openstack-cinder00:00
vincent_hou_thingee: Hi Mike. If you have got a time, could you take a look at the spec https://review.openstack.org/#/c/186327/?00:00
*** salv-orlando has joined #openstack-cinder00:01
*** topshare has quit IRC00:02
*** Guest7393 has quit IRC00:04
*** vincent_hou_ has quit IRC00:05
*** IlyaG has joined #openstack-cinder00:10
*** jwcroppe has quit IRC00:11
*** mgarza has quit IRC00:11
*** topshare has joined #openstack-cinder00:11
*** chlong has joined #openstack-cinder00:12
*** garthb has quit IRC00:14
*** salv-orlando has quit IRC00:14
*** topshare has quit IRC00:15
*** gouthamr_ has joined #openstack-cinder00:18
*** vokt has quit IRC00:19
*** gouthamr has quit IRC00:20
*** tsekiyam_ has joined #openstack-cinder00:22
*** dims has joined #openstack-cinder00:23
*** jamielennox is now known as jamielennox|away00:24
*** tsekiyama has quit IRC00:26
patrickeastthingee: hey, when you get a chance could you target https://blueprints.launchpad.net/cinder/+spec/cinder-internal-tenant and https://blueprints.launchpad.net/cinder/+spec/image-volume-cache for liberty-2?00:26
openstackgerritRyan Lucio proposed openstack/cinder: VMEM v6000: Fix export verify routines to use proper state nodes  https://review.openstack.org/16565500:26
*** tsekiyam_ has quit IRC00:27
*** Apoorva has joined #openstack-cinder00:27
openstackgerritPatrick East proposed openstack/cinder: WIP Generic image cache.  https://review.openstack.org/19579500:28
openstackgerritPatrick East proposed openstack/cinder: Add Cinder internal tenant support.  https://review.openstack.org/19935200:28
openstackgerritAnthony Lee proposed openstack/cinder: Improve 3PAR driver VLUN creation and deletion  https://review.openstack.org/19616400:30
*** mtanino has quit IRC00:30
thingeepatrickeast: done00:31
patrickeastthingee: thanks!00:31
*** Apoorva has quit IRC00:32
*** jamielennox|away is now known as jamielennox00:35
*** leeantho has quit IRC00:42
*** jasondotstar has quit IRC00:43
*** diogogmt has quit IRC00:49
*** yamada-h has joined #openstack-cinder00:51
*** topshare has joined #openstack-cinder00:57
*** yamada-h has quit IRC00:57
*** kkeuning has joined #openstack-cinder01:01
*** kkeuning has quit IRC01:02
*** lixiaoy1 has joined #openstack-cinder01:11
*** salv-orlando has joined #openstack-cinder01:12
*** heyun has joined #openstack-cinder01:15
*** Longgeek has joined #openstack-cinder01:15
*** yamada-h has joined #openstack-cinder01:17
*** heyun has quit IRC01:18
*** jungleboyj has joined #openstack-cinder01:18
*** heyun has joined #openstack-cinder01:19
*** davechen1 has joined #openstack-cinder01:19
*** Longgeek has quit IRC01:20
*** yamada-h has quit IRC01:20
*** jdurgin1 has quit IRC01:22
*** annegentle has joined #openstack-cinder01:23
*** davechen has joined #openstack-cinder01:23
*** yamada-h has joined #openstack-cinder01:24
*** dims_ has joined #openstack-cinder01:24
*** dims_ has quit IRC01:24
*** kkeuning has joined #openstack-cinder01:25
*** yamada-h has quit IRC01:25
*** davechen1 has quit IRC01:25
*** salv-orlando has quit IRC01:26
*** kkeuning has quit IRC01:26
*** kkeuning has joined #openstack-cinder01:27
*** dims has quit IRC01:27
*** annegentle has quit IRC01:27
*** kkeuning has quit IRC01:28
*** Apoorva has joined #openstack-cinder01:28
*** Apoorva has quit IRC01:33
*** IlyaG has quit IRC01:38
*** diogogmt has joined #openstack-cinder01:38
*** Yogi11 has joined #openstack-cinder01:39
openstackgerritSean Chen proposed openstack/cinder: Tintri snapshot id  https://review.openstack.org/19882901:44
*** laughterwym_ has joined #openstack-cinder01:50
*** laughterwym has quit IRC01:53
*** asselin_ has joined #openstack-cinder01:55
*** kkeuning has joined #openstack-cinder01:57
*** akerr has joined #openstack-cinder01:57
*** Ctina has joined #openstack-cinder01:58
*** akerr_ has joined #openstack-cinder02:00
*** kkeuning has quit IRC02:00
*** jskinner has joined #openstack-cinder02:02
*** akerr has quit IRC02:03
*** jskinner has quit IRC02:04
*** jwcroppe has joined #openstack-cinder02:07
*** Yogi11 has quit IRC02:09
*** jwcroppe has quit IRC02:11
openstackgerritPatrick East proposed openstack/cinder: Add Cinder internal tenant support.  https://review.openstack.org/19935202:14
openstackgerritPatrick East proposed openstack/cinder: WIP Generic image cache.  https://review.openstack.org/19579502:14
*** Lee1092 has joined #openstack-cinder02:18
*** jasondotstar has joined #openstack-cinder02:22
*** vincent_hou has joined #openstack-cinder02:22
*** mutoulbj has joined #openstack-cinder02:22
openstackgerritLisaLi proposed openstack/cinder: Fix restore point if backup base is diff-format in ceph  https://review.openstack.org/19899302:23
*** Apoorva has joined #openstack-cinder02:29
*** patrickeast has quit IRC02:29
openstackgerritKuo-tung Kao proposed openstack/python-cinderclient: Add more details for replication  https://review.openstack.org/19656202:30
openstackgerritLisaLi proposed openstack/cinder: Fix restore point if backup base is diff-format in ceph  https://review.openstack.org/19899302:30
*** Apoorva has quit IRC02:34
*** annegentle has joined #openstack-cinder02:34
*** bkopilov has quit IRC02:36
*** gouthamr has joined #openstack-cinder02:37
*** annegentle has quit IRC02:37
*** gouthamr_ has quit IRC02:39
*** BharatK has joined #openstack-cinder02:45
*** changbl has joined #openstack-cinder02:48
*** hakimo_ has joined #openstack-cinder02:52
*** hakimo has quit IRC02:54
openstackgerritDave Chen proposed openstack/cinder: set/unset volume image metadata  https://review.openstack.org/14772602:54
*** kkeuning has joined #openstack-cinder02:55
vincent_housmcginnis: Hey.02:55
openstackgerritLisaLi proposed openstack/cinder: Fix restore point if backup base is diff-format in ceph  https://review.openstack.org/19899302:56
*** rajinir has joined #openstack-cinder02:59
*** jasondotstar has quit IRC03:02
*** jwcroppe has joined #openstack-cinder03:08
*** salv-orlando has joined #openstack-cinder03:13
*** jwcroppe has quit IRC03:13
*** vincent_hou has quit IRC03:15
*** Longgeek has joined #openstack-cinder03:16
openstackgerritwanghao proposed openstack/cinder: Validate value when user update quota  https://review.openstack.org/19793803:18
*** Longgeek has quit IRC03:22
*** kkeuning has quit IRC03:23
openstackgerritLin Yang proposed openstack/python-cinderclient: Improve error message when cinder quota exceeded  https://review.openstack.org/17894603:27
*** xyang1 has quit IRC03:28
*** salv-orlando has quit IRC03:28
*** Apoorva has joined #openstack-cinder03:30
openstackgerritKuo-tung Kao proposed openstack/cinder: cinder list fails with 'name' sort key  https://review.openstack.org/19861403:31
*** vincent_hou has joined #openstack-cinder03:31
vincent_housmcginnis: https://review.openstack.org/#/c/194223/ this is another important fix.03:32
*** Apoorva has quit IRC03:34
*** rajinir has quit IRC03:35
*** annegentle has joined #openstack-cinder03:37
*** garthb has joined #openstack-cinder03:38
*** BharatK has quit IRC03:40
*** annegentle has quit IRC03:41
openstackgerritLin Yang proposed openstack/python-cinderclient: Remove redundant statement and refactor  https://review.openstack.org/19654603:42
*** davechen is now known as davechen_away03:45
*** patrickeast has joined #openstack-cinder03:45
*** patrickeast has quit IRC03:49
*** patrickeast has joined #openstack-cinder03:49
*** patrickeast has quit IRC03:50
*** patrickeast has joined #openstack-cinder03:51
*** BharatK has joined #openstack-cinder03:51
*** boris-42 has quit IRC03:52
*** Apoorva has joined #openstack-cinder03:55
*** bkopilov has joined #openstack-cinder03:57
*** _cjones_ has quit IRC03:57
openstackgerritTeruaki Ishizaki proposed openstack/cinder: Sheepdog: Add class for dog command executor  https://review.openstack.org/19453204:00
*** annashen has joined #openstack-cinder04:01
*** gouthamr has quit IRC04:06
*** mtreinish has quit IRC04:07
openstackgerritTeruaki Ishizaki proposed openstack/cinder: Sheepdog: change create and delete operation  https://review.openstack.org/19739304:08
*** r-daneel has joined #openstack-cinder04:13
*** simondodsley has joined #openstack-cinder04:14
*** jwcroppe has joined #openstack-cinder04:15
*** mtreinish has joined #openstack-cinder04:16
*** simondodsley has quit IRC04:19
*** david-ly_ has joined #openstack-cinder04:19
*** diogogmt has quit IRC04:20
*** sgotliv has joined #openstack-cinder04:21
*** david-lyle has quit IRC04:22
*** Apoorva has quit IRC04:26
*** annashen has quit IRC04:28
*** salv-orlando has joined #openstack-cinder04:29
*** akerr_ is now known as akerr_away04:33
*** chlong has quit IRC04:35
*** lpetrut has joined #openstack-cinder04:40
*** salv-orlando has quit IRC04:42
*** laughterwym__ has joined #openstack-cinder04:43
*** harlowja_at_home has joined #openstack-cinder04:43
*** chlong has joined #openstack-cinder04:44
*** laughterwym_ has quit IRC04:46
*** vincent_hou_ has joined #openstack-cinder04:47
*** vincent_hou has quit IRC04:49
*** vincent_hou has joined #openstack-cinder04:49
*** chlong has quit IRC04:51
*** yamada-h has joined #openstack-cinder04:52
*** garthb has quit IRC04:52
*** shyama has joined #openstack-cinder04:52
*** changbl has quit IRC04:52
*** mutoulbj has quit IRC04:52
*** mutoulbj_ has joined #openstack-cinder04:53
*** vincent_hou_ has quit IRC04:53
*** davechen_away is now known as davechen04:57
*** nkrinner_afk is now known as nkrinner04:58
*** yamada-h_ has joined #openstack-cinder05:09
*** dims has joined #openstack-cinder05:11
*** nikeshm has joined #openstack-cinder05:12
*** yamada-h has quit IRC05:12
openstackgerritxing-yang proposed openstack/cinder: Non-disruptive backup  https://review.openstack.org/19393705:13
*** lpetrut has quit IRC05:13
*** Longgeek has joined #openstack-cinder05:16
*** deepakcs has joined #openstack-cinder05:17
*** mutoulbj has joined #openstack-cinder05:18
*** chlong has joined #openstack-cinder05:20
*** boris-42 has joined #openstack-cinder05:21
*** mutoulbj_ has quit IRC05:22
*** annegentle has joined #openstack-cinder05:26
*** rakesh_mishra__ has joined #openstack-cinder05:26
*** Apoorva has joined #openstack-cinder05:27
*** shausy has joined #openstack-cinder05:29
*** annegentle has quit IRC05:30
*** Apoorva has quit IRC05:32
*** vincent_hou has quit IRC05:33
*** davechen_afk is now known as jungler05:37
*** dims has quit IRC05:40
*** rakesh_mishra__ has quit IRC05:47
*** rakesh_mishra__ has joined #openstack-cinder05:48
*** ig0r__ has joined #openstack-cinder05:51
*** yamada-h_ has quit IRC05:52
*** sgotliv has quit IRC05:52
*** sgotliv has joined #openstack-cinder05:52
*** ig0r_ has quit IRC05:52
openstackgerritAndrey Pavlov proposed openstack/cinder: Fix doc string definitions  https://review.openstack.org/19312405:54
*** nlevinki has joined #openstack-cinder05:55
*** nihilifer has joined #openstack-cinder05:59
*** lpetrut has joined #openstack-cinder06:03
*** lpetrut has quit IRC06:04
*** nlevinki has quit IRC06:05
*** rakesh_mishra__ has quit IRC06:08
*** rakesh_mishra__ has joined #openstack-cinder06:10
*** bkopilov has quit IRC06:19
*** patrickeast has quit IRC06:19
*** tobe has joined #openstack-cinder06:19
*** nlevinki has joined #openstack-cinder06:20
openstackgerritTeruaki Ishizaki proposed openstack/cinder: Sheepdog: Add class for dog command executor  https://review.openstack.org/19453206:22
*** yamada-h has joined #openstack-cinder06:24
*** Apoorva has joined #openstack-cinder06:28
*** rakesh_mishra__ has quit IRC06:28
*** rakesh_mishra_ has joined #openstack-cinder06:28
*** salv-orlando has joined #openstack-cinder06:31
openstackgerritVincent Hou proposed openstack/python-cinderclient: Change cinderclient according to volume migration improvement  https://review.openstack.org/18954706:31
*** coolsvap|away is now known as coolsvap06:32
*** bkopilov has joined #openstack-cinder06:32
*** Apoorva has quit IRC06:32
openstackgerritVincent Hou proposed openstack/python-cinderclient: Change cinderclient according to volume migration improvement  https://review.openstack.org/18954706:35
openstackgerritTeruaki Ishizaki proposed openstack/cinder: Sheepdog: Add class for dog command executor  https://review.openstack.org/19453206:39
*** salv-orlando has quit IRC06:42
*** markus_z has joined #openstack-cinder06:46
openstackgerritVincent Hou proposed openstack/cinder: Add the notification and the migration progress indication  https://review.openstack.org/18961406:51
*** harlowja_at_home has quit IRC06:52
*** xiangfeiz has joined #openstack-cinder06:56
*** yrabl has quit IRC06:58
*** yrabl has joined #openstack-cinder06:58
*** markus_z has quit IRC06:58
openstackgerritTeruaki Ishizaki proposed openstack/cinder: Sheepdog: change create and delete operation  https://review.openstack.org/19739306:59
*** salv-orlando has joined #openstack-cinder07:03
*** yamada-h has quit IRC07:05
*** yamada-h has joined #openstack-cinder07:10
*** annegentle has joined #openstack-cinder07:14
*** annegentle has quit IRC07:19
*** yamada-h has quit IRC07:19
*** rakesh_mishra_ has quit IRC07:20
*** rakesh_mishra_ has joined #openstack-cinder07:22
*** xiangfeiz has quit IRC07:22
*** anshul has joined #openstack-cinder07:24
*** e0ne has joined #openstack-cinder07:25
*** yamada-h has joined #openstack-cinder07:26
*** Apoorva has joined #openstack-cinder07:28
*** Apoorva has quit IRC07:33
*** afazekas has joined #openstack-cinder07:34
*** jordanP has joined #openstack-cinder07:35
*** rakesh_mishra_ has quit IRC07:36
*** jwcroppe has quit IRC07:37
openstackgerritVincent Hou proposed openstack/cinder: Add the notification and the migration progress indication  https://review.openstack.org/18961407:38
*** dims has joined #openstack-cinder07:41
*** dims_ has joined #openstack-cinder07:42
*** akerr_away has quit IRC07:43
*** jistr has joined #openstack-cinder07:43
*** ndipanov has joined #openstack-cinder07:45
*** dims has quit IRC07:45
*** dims_ has quit IRC07:46
*** tobe has quit IRC07:46
*** bluex-pl has joined #openstack-cinder07:47
*** markvoelker has quit IRC07:47
*** e0ne is now known as e0ne_07:49
*** e0ne_ is now known as e0ne07:51
*** ronis has joined #openstack-cinder07:52
*** afazekas has quit IRC07:56
*** e0ne is now known as e0ne_07:57
*** e0ne_ is now known as e0ne07:59
*** e0ne has quit IRC08:00
*** anshul has quit IRC08:00
openstackgerritKuo-tung Kao proposed openstack/cinder: cinder list fails with 'name' sort key  https://review.openstack.org/19861408:00
*** ociuhandu has joined #openstack-cinder08:01
*** lpetrut has joined #openstack-cinder08:08
*** afazekas has joined #openstack-cinder08:09
*** ociuhandu has quit IRC08:13
*** tobe has joined #openstack-cinder08:16
openstackgerritDaniel Tadrzak proposed openstack/cinder: Fix saving tz aware datetimes in Versioned Objects  https://review.openstack.org/19607108:17
*** chlong has quit IRC08:21
*** tobe has quit IRC08:25
openstackgerritPradeep Sathasivam proposed openstack/cinder: Adds friendly zone name support  https://review.openstack.org/18051808:25
*** tobe has joined #openstack-cinder08:25
*** Apoorva has joined #openstack-cinder08:29
*** yamada-h has quit IRC08:29
*** boris-42 has quit IRC08:32
*** Apoorva has quit IRC08:33
*** belmoreira has joined #openstack-cinder08:38
*** tobe has quit IRC08:40
*** abehl has joined #openstack-cinder08:42
*** dims has joined #openstack-cinder08:43
*** lpetrut has quit IRC08:43
*** dims_ has joined #openstack-cinder08:44
*** laughterwym__ has quit IRC08:44
*** lpetrut has joined #openstack-cinder08:44
openstackgerritxiaoqin proposed openstack/cinder: IBM storwize driver configuration item default valume change  https://review.openstack.org/19869208:45
*** dims__ has joined #openstack-cinder08:45
openstackgerritVincent Hou proposed openstack/cinder: Add the notification and the migration progress indication  https://review.openstack.org/18961408:45
*** dims___ has joined #openstack-cinder08:46
*** lixiaoy1 has quit IRC08:47
*** dims has quit IRC08:47
*** markvoelker has joined #openstack-cinder08:48
*** dims_ has quit IRC08:49
*** dims__ has quit IRC08:49
*** dims___ has quit IRC08:50
*** markvoelker has quit IRC08:53
openstackgerritVincent Hou proposed openstack/cinder: Add the notification and the migration progress indication  https://review.openstack.org/18961408:58
*** lpetrut has quit IRC09:02
*** e0ne has joined #openstack-cinder09:03
*** annegentle has joined #openstack-cinder09:03
*** lpetrut has joined #openstack-cinder09:03
*** shausy has quit IRC09:07
*** shausy has joined #openstack-cinder09:07
*** annegentle has quit IRC09:08
*** r-daneel has quit IRC09:08
*** afazekas has quit IRC09:12
*** belmoreira has quit IRC09:12
*** markus_z has joined #openstack-cinder09:16
*** laughterwym has joined #openstack-cinder09:19
*** tobe has joined #openstack-cinder09:20
*** e0ne is now known as e0ne_09:21
*** afazekas has joined #openstack-cinder09:21
*** belmoreira has joined #openstack-cinder09:24
openstackgerritYuriy Nesenenko proposed openstack/cinder-specs: Checking the existence of volume  https://review.openstack.org/17482309:24
*** e0ne_ is now known as e0ne09:26
*** Apoorva has joined #openstack-cinder09:30
openstackgerritDaniel Tadrzak proposed openstack/cinder: CGSnapshot Object  https://review.openstack.org/19502709:30
openstackgerritDaniel Tadrzak proposed openstack/cinder: ConsistencyGroup Object  https://review.openstack.org/16364109:30
openstackgerritMichal Dulko proposed openstack/cinder: Service object  https://review.openstack.org/16041709:30
*** mutoulbj_ has joined #openstack-cinder09:33
*** haypo has joined #openstack-cinder09:34
*** Apoorva has quit IRC09:34
*** mutoulbj has quit IRC09:36
*** rakesh_mishra has joined #openstack-cinder09:36
*** e0ne is now known as e0ne_09:37
*** nikeshm has quit IRC09:38
*** IanGovett has joined #openstack-cinder09:39
*** rakesh_mishra has quit IRC09:44
*** ociuhandu has joined #openstack-cinder09:44
*** rakesh_mishra has joined #openstack-cinder09:45
*** jasondotstar has joined #openstack-cinder09:45
openstackgerritwanghao proposed openstack/cinder: Incremental backup improvements for L  https://review.openstack.org/18832809:45
*** nihilifer has quit IRC09:52
*** davechen has left #openstack-cinder09:53
*** aix has joined #openstack-cinder09:54
*** rakesh_mishra has quit IRC09:56
*** rakesh_mishra has joined #openstack-cinder09:56
openstackgerritVincent Hou proposed openstack/cinder: Add the notification and the migration progress indication  https://review.openstack.org/18961409:58
*** e0ne_ is now known as e0ne10:00
*** rakesh_mishra has quit IRC10:02
*** rakesh_mishra has joined #openstack-cinder10:03
*** rakesh_mishra has quit IRC10:03
*** rakesh_mishra has joined #openstack-cinder10:05
*** mutoulbj_ has quit IRC10:09
*** e0ne is now known as e0ne_10:11
*** e0ne_ is now known as e0ne10:13
*** zhenguo has quit IRC10:26
*** tobe has quit IRC10:29
*** afazekas has quit IRC10:29
*** Apoorva has joined #openstack-cinder10:31
*** coolsvap is now known as coolsvap|away10:33
*** Apoorva has quit IRC10:35
deepakcsjordanP: https://review.openstack.org/#/c/192736/8/nova/virt/libvirt/driver.py , quick Q...10:39
deepakcsjordanP: active_disk_object.source_path <-- does this work correct when disk of of type 'network' ? I haven't tried in recent time, thus asking10:39
*** afazekas has joined #openstack-cinder10:40
*** e0ne is now known as e0ne_10:40
*** e0ne_ is now known as e0ne10:42
*** ronis has quit IRC10:48
*** ronis has joined #openstack-cinder10:48
*** markvoelker has joined #openstack-cinder10:49
*** annegentle has joined #openstack-cinder10:52
*** rakesh_mishra has quit IRC10:53
*** markvoelker has quit IRC10:54
*** rakesh_mishra has joined #openstack-cinder10:54
openstackgerritTeruaki Ishizaki proposed openstack/cinder: Sheepdog: change snapshot and clone operation  https://review.openstack.org/19895210:54
*** rakesh_mishra has quit IRC10:54
*** rakesh_mishra has joined #openstack-cinder10:54
*** annegentle has quit IRC10:56
*** rakesh_mishra has quit IRC10:58
*** rakesh_mishra has joined #openstack-cinder10:58
*** heyun has quit IRC10:58
*** topshare has quit IRC11:02
*** ronis has quit IRC11:08
jordanPdeepakcs, you tell me :)11:11
deepakcsjordanP: i am not sure and right now i don't have a setup to validate that. I think it won't work11:11
deepakcsjordanP: you need to use _get_snap_dev if active_protocol is not None (which means its a network disk)11:11
jordanPdeepakcs, :                 rebase_base = _get_snap_dev(delete_info['file_to_merge'],11:12
jordanP                                            active_disk_object.backing_store)11:12
jordanPL192411:12
*** afazekas has quit IRC11:13
deepakcsjordanP: what are u trying to say ?11:13
jordanPyou mean _get_snap_dev(active_disk_object.source_path) ?11:13
deepakcsjordanP: yeah... i am not sure if active_disk_object.source_path is correct value when disk is of 'network' type11:13
jordanPdeepakcs, which driver uses "network" disk ? Mine is based on files...11:14
*** ronis has joined #openstack-cinder11:14
deepakcsjordanP: rbd, gluster, sheepdog also i guess. Its been a while for me to have tried network disk config11:14
*** mutoulbj has joined #openstack-cinder11:15
deepakcsjordanP: eharney (during US time) would be good to consult on this11:15
jordanPso I should setup a ceph ou glusterFS to test this corner case ?11:15
jordanPman, I just want to fix that for cinder drivers based on RemoteFS11:16
jordanPI cant"t fix it for every driver on eartch !11:16
deepakcsjordanP: network disk type is a supported disk type in libvirt! It so happens that ur driver doesn't utilise that11:17
deepakcsjordanP: i will sent a mail to eharney to check this out.11:17
jordanPdeepakcs, I understand that. But do I have to be the QA engineer for every drivers out there ?11:17
jordanPit's not me who disabled that tempest test for glusterfs driver, I am sorry11:18
deepakcsjordanP: no, but the fix should be generic enuf, it shouldn't break other disk types, right ? OTherwise there is no point of having remotefs abstraction11:18
jordanPit's not going to break anything. Because it's already broken11:18
deepakcsjordanP: i am not speaking as glusterfs driver person but speaking in general11:19
jordanPthe ceph CI is running fine with and without my patch. So i can assume I didnt break it.11:19
deepakcsjordanP: ceph doesn't use online snapshot, so it doesn't go thru that path ( i was wrong when i said rbd before)11:20
deepakcsjordanP: afaik gluster, sheepdog and anything else based on remotefs would use it11:20
jordanPdeepakcs, okay. So you want me to setup a glusterFS to test that ?11:21
deepakcsjordanP: the reason i m not sure is bcos libvirt doesn't provide me a simple way to figure what field (source_path or some otehr field) to use when disk is network type11:21
*** rakesh_mishra has quit IRC11:21
deepakcsjordanP: it would be easier to tlak with eharney who has dealt with this in the past. thats all i m saying!11:21
jordanPdeepakcs, okay :) no rush, I'll try to reach eharney.11:22
deepakcsjordanP: i will try to be there, but its late in the eve for me, in any case pls check with him11:22
*** ronis_ has joined #openstack-cinder11:22
deepakcsi will leave a comment in the patch, won't rate it11:23
jordanPgood. Sorry I a bit nervous. But each time I have to go through sending a patch to nova, I want to : ʕノ•ᴥ•ʔノ ︵ ┻━┻11:23
jordanP:)11:23
deepakcsjordanP: what was the last part of ur msg ? Its greek to me11:24
jordanPdeepakcs, just some kind of ascii art11:24
deepakcsjordanP: can't help it, since the code to manage file and network type disks are common part in driver.py.... I didn't design it that way11:24
openstackgerritKyrylo Romanenko proposed openstack/python-cinderclient: Add tests for python-cinderclient  https://review.openstack.org/19764311:24
deepakcsjordanP: definitely not ascii :) 'art' can't be sure11:25
jordanPutf8 art then :D11:25
deepakcsjordanP: let me do this, will send a mail to eharney, Ccing you, i am hoping that would be better11:26
jordanPdeepakcs, even better. thanks11:26
deepakcsjordanP: np11:26
*** topshare has joined #openstack-cinder11:26
*** ronis has quit IRC11:26
openstackgerritDaniel Tadrzak proposed openstack/cinder: Cleanup for SnapshotObject  https://review.openstack.org/19834111:28
openstackgerritKyrylo Romanenko proposed openstack/python-cinderclient: Add tests for python-cinderclient  https://review.openstack.org/19764311:29
*** zhenguo has joined #openstack-cinder11:29
*** topshare has quit IRC11:30
*** Apoorva has joined #openstack-cinder11:31
*** e0ne is now known as e0ne_11:34
*** e0ne_ is now known as e0ne11:35
deepakcsjordanP: mail sent, lets hope for a response11:36
*** Apoorva has quit IRC11:36
jordanPdeepakcs, it can wait one day or 2. It's not critical11:36
deepakcsjordanP: u never know how long it can take, so we do our best to close it :)11:37
jordanPyep11:37
*** e0ne has quit IRC11:38
*** Longgeek has quit IRC11:41
*** Longgeek has joined #openstack-cinder11:42
*** Longgeek has quit IRC11:42
*** Longgeek has joined #openstack-cinder11:42
*** Longgeek has quit IRC11:43
*** kaisers_ has joined #openstack-cinder11:46
*** dims has joined #openstack-cinder11:47
*** zul has quit IRC11:48
*** dims_ has joined #openstack-cinder11:48
*** kaisers_97 has joined #openstack-cinder11:49
*** dims__ has joined #openstack-cinder11:49
*** markvoelker has joined #openstack-cinder11:50
*** asselin_ has quit IRC11:51
*** asselin_ has joined #openstack-cinder11:52
*** dims has quit IRC11:52
*** dims_ has quit IRC11:53
*** markvoelker has quit IRC11:54
*** dims__ has quit IRC11:54
*** kaisers_ has quit IRC11:54
*** topshare has joined #openstack-cinder11:54
*** kaisers_97 has quit IRC11:56
*** salv-orlando has quit IRC11:58
*** abehl has quit IRC11:58
*** abehl has joined #openstack-cinder11:59
*** timcl has joined #openstack-cinder12:00
*** abehl has quit IRC12:09
*** afazekas has joined #openstack-cinder12:11
*** bswartz has quit IRC12:12
*** markvoelker has joined #openstack-cinder12:13
*** belmoreira has quit IRC12:17
*** abehl has joined #openstack-cinder12:21
*** salv-orlando has joined #openstack-cinder12:21
openstackgerritSergey Vilgelm proposed openstack/cinder: Switch to the oslo._utils.fileutils  https://review.openstack.org/19953712:22
openstackgerritSergey Vilgelm proposed openstack/cinder: Switch to the oslo_utils.fileutils  https://review.openstack.org/19953712:23
*** belmoreira has joined #openstack-cinder12:23
openstackgerritSergey Vilgelm proposed openstack/cinder: Switch to the oslo_utils.fileutils  https://review.openstack.org/19953712:23
*** Longgeek has joined #openstack-cinder12:25
*** jasondotstar has quit IRC12:25
*** annegentle has joined #openstack-cinder12:30
*** zul has joined #openstack-cinder12:30
*** afazekas has quit IRC12:31
*** chlong has joined #openstack-cinder12:31
*** ganso_ has joined #openstack-cinder12:32
*** tellesnobrega_ has joined #openstack-cinder12:32
*** kkeuning has joined #openstack-cinder12:32
*** kkeuning has quit IRC12:33
*** kjnelson has joined #openstack-cinder12:35
*** erlon has joined #openstack-cinder12:39
*** tellesnobrega__ has joined #openstack-cinder12:39
openstackgerritxing-yang proposed openstack/cinder: Attach snapshot  https://review.openstack.org/19855612:41
*** afazekas has joined #openstack-cinder12:42
*** tellesnobrega_ has quit IRC12:42
*** jungleboyj has quit IRC12:42
*** tellesnobrega_ has joined #openstack-cinder12:42
*** tellesnobrega__ has quit IRC12:46
*** akerr has joined #openstack-cinder12:48
*** dims has joined #openstack-cinder12:50
*** simondodsley has joined #openstack-cinder12:51
*** coolsvap|away is now known as coolsvap12:52
*** merooney has joined #openstack-cinder12:53
openstackgerritSergey Vilgelm proposed openstack/cinder: Switch to the oslo_utils.fileutils  https://review.openstack.org/19953712:54
*** dims has quit IRC12:54
*** afazekas has quit IRC12:55
*** eharney has joined #openstack-cinder12:59
*** j_king has quit IRC12:59
*** nkrinner has quit IRC12:59
*** Swanson has quit IRC13:00
*** abehl has quit IRC13:00
*** mfisch has quit IRC13:00
*** abehl has joined #openstack-cinder13:00
*** Swanson has joined #openstack-cinder13:00
*** j_king has joined #openstack-cinder13:00
*** tellesnobrega__ has joined #openstack-cinder13:01
*** julim has joined #openstack-cinder13:01
*** zhipeng has joined #openstack-cinder13:01
*** nkrinner has joined #openstack-cinder13:01
openstackgerritAdriano Freires Rosso proposed openstack/cinder: Adds manage/unmanage methods for HNAS drivers  https://review.openstack.org/19799913:02
*** cdelatte has joined #openstack-cinder13:03
*** tellesnobrega_ has quit IRC13:03
openstackgerritxing-yang proposed openstack/cinder: Attach snapshot  https://review.openstack.org/19855613:04
*** timcl has quit IRC13:04
*** tellesnobrega_ has joined #openstack-cinder13:04
*** jasondotstar has joined #openstack-cinder13:05
*** abehl has quit IRC13:06
*** tellesnobrega__ has quit IRC13:06
*** xyang1 has joined #openstack-cinder13:07
*** tellesnobrega_ has quit IRC13:09
*** tellesnobrega_ has joined #openstack-cinder13:09
*** xyang has joined #openstack-cinder13:10
*** dustins has joined #openstack-cinder13:12
*** deepakcs has quit IRC13:14
*** afazekas has joined #openstack-cinder13:15
*** asselin__ has joined #openstack-cinder13:16
*** tellesnobrega__ has joined #openstack-cinder13:16
*** tellesnobrega_ has quit IRC13:18
*** asselin_ has quit IRC13:19
*** changbl has joined #openstack-cinder13:19
*** akerr has quit IRC13:21
*** akerr_ has joined #openstack-cinder13:21
*** tellesnobrega_ has joined #openstack-cinder13:21
*** abehl has joined #openstack-cinder13:21
*** dims has joined #openstack-cinder13:22
*** tellesnobrega__ has quit IRC13:24
*** mriedem_away is now known as mriedem13:24
*** timcl has joined #openstack-cinder13:25
*** tellesnobrega__ has joined #openstack-cinder13:27
*** akshai has joined #openstack-cinder13:31
*** tellesnobrega_ has quit IRC13:31
*** Yogi1 has joined #openstack-cinder13:31
*** salv-orlando has quit IRC13:33
smcginnisDuncanT: You still alive?13:33
*** eharney has quit IRC13:34
*** bswartz has joined #openstack-cinder13:37
*** Apoorva has joined #openstack-cinder13:38
*** thangp has joined #openstack-cinder13:38
*** gouthamr has joined #openstack-cinder13:40
*** annegentle has quit IRC13:45
*** annegentle has joined #openstack-cinder13:45
*** rushil has joined #openstack-cinder13:46
*** jwcroppe has joined #openstack-cinder13:49
*** Apoorva has quit IRC13:53
*** Apoorva has joined #openstack-cinder13:54
*** rakesh_mishra has joined #openstack-cinder13:55
*** rakesh_mishra has quit IRC13:55
*** rakesh_mishra has joined #openstack-cinder13:57
*** Apoorva has quit IRC13:59
*** dustins has quit IRC13:59
*** mclaren has joined #openstack-cinder14:00
mclarenHowdy. Anyone available to  take a look at https://review.openstack.org/#/c/173977? Some kilo gating issues may be related to the cinder client14:01
*** Yogi1 has quit IRC14:01
openstackgerritSergey Vilgelm proposed openstack/cinder: Switch to the oslo_utils.fileutils  https://review.openstack.org/19953714:01
*** tellesnobrega_ has joined #openstack-cinder14:02
*** eharney has joined #openstack-cinder14:02
*** annegentle has quit IRC14:02
*** dustins has joined #openstack-cinder14:03
*** bkopilov has quit IRC14:03
*** rushil has quit IRC14:04
*** mtanino has joined #openstack-cinder14:04
*** edmondsw has joined #openstack-cinder14:05
*** shyama has quit IRC14:05
*** tellesnobrega__ has quit IRC14:05
*** tellesnobrega__ has joined #openstack-cinder14:06
*** tellesnobrega_ has quit IRC14:06
*** rushil has joined #openstack-cinder14:06
*** shyama has joined #openstack-cinder14:08
openstackgerritBharat Kumar Kobagana (BharatK) proposed openstack/cinder: GlusterFS backup driver  https://review.openstack.org/18225414:14
openstackgerritBharat Kumar Kobagana (BharatK) proposed openstack/cinder: Posix backup driver  https://review.openstack.org/16364714:14
DuncanTsmcginnis: Hi. Technically not dead yet, though I feel like I'm slowly cooking....14:17
*** esker has joined #openstack-cinder14:17
DuncanTthingee: jgriffith: Are you able to take a look at https://review.openstack.org/#/c/173977/ to unblock the stable gate, please?14:17
smcginnisDuncanT: ;)14:18
smcginnisDuncanT: Wonder about that script you had for abandoning old reviews.14:18
smcginnisDuncanT: Wondering if we should run that again.14:18
smcginnisDuncanT: Starting to get a lot out there that haven't been touched since early Kilo.14:18
*** BharatK has quit IRC14:18
DuncanTsmcginnis: I certainly can do, yeah. I wonder where I put it?14:18
*** timcl1 has joined #openstack-cinder14:19
DuncanTsmcginnis: I'll go have a dig14:19
smcginnisDuncanT: That's always my problem with those kinds of things. :)14:19
*** mfisch has joined #openstack-cinder14:19
DuncanTsmcginnis: I keep saying I should push everything to github14:20
*** mfisch has quit IRC14:20
*** mfisch has joined #openstack-cinder14:20
smcginnisDuncanT: Not a bad idea actually.14:20
DuncanTsmcginnis: Unfortuately, most of my scripts are bad enough I don't want to admit to having coded them ;-)14:20
smcginnisgit clone duncansrandomcrap14:21
smcginnis;)14:21
*** setmason has joined #openstack-cinder14:21
*** timcl has quit IRC14:21
setmasonI’m trying to get a total volume count across all tenants, I’m using cinder list —all-tenants and it seems to be stopping at 1000 volumes.  Is there a more efficient method of getting this data?14:22
*** zhipeng has quit IRC14:22
openstackgerritDuncan Thomas proposed openstack/cinder: Add support for backups to xiv_ds8k driver  https://review.openstack.org/11829814:25
scottdasetmason: What version of cinderclient are you using? I think you've hit bug https://bugs.launchpad.net/cinder/+bug/146826014:26
openstackLaunchpad bug 1468260 in Cinder "'cinder list' does not list more than 1000 volumes" [Undecided,Confirmed]14:26
scottda(for limit of 1000 volumes)14:26
setmasonok thanks14:26
*** shausy has quit IRC14:26
scottdaBug reports no problem with version 1.0 or 1.1, so that might be a workaround.14:27
*** annegentle has joined #openstack-cinder14:27
setmasonRunning Havana and v1 API14:28
openstackgerritDuncan Thomas proposed openstack/cinder: allow attaching status for multiattach volume  https://review.openstack.org/17213814:29
scottdaThat refers to cinderclient version ^^14:30
setmasonright.14:30
setmasonnot at system right now (on phone), will update bug with information when I get into the offfice and can run cinder --verison14:30
scottdacool14:31
*** tellesnobrega__ has quit IRC14:32
*** rushil has quit IRC14:34
*** salv-orlando has joined #openstack-cinder14:35
thingeeDuncanT: this requirements proposal doesn't match global requirements for stable/kilo14:35
thingeeDuncanT: keystoneclient for example14:36
DuncanTthingee: Ok, thanks, I'll pass that back to the people chasing it14:36
openstackgerritMichal Dulko proposed openstack/cinder: Remove unused arguments from c-vol's create_volume  https://review.openstack.org/19959014:36
xyangthingee: hi, can you please approve this blueprint and target L-2? https://blueprints.launchpad.net/cinder/+spec/non-disruptive-backup14:37
thingeedhellmann: is there a way to have proposal bot try again? looks like this update is out dated https://review.openstack.org/#/c/173977/314:38
erlonthingee: hi Mike14:39
erlonthingee: have you see my email on third-party list?14:39
erlonthingee: when you have a change please push that patch (https://review.openstack.org/#/c/197686/) fowardso we can enable the other CI account14:40
*** tellesnobrega_ has joined #openstack-cinder14:41
*** setmason has quit IRC14:42
xyangDuncanT: hi, I addressed your review comments on the non-disruptive backup patch: https://review.openstack.org/#/c/193937/  Also submitted a patch for attach snapshot: https://review.openstack.org/#/c/198556/ .  Can you provide some feedback on the attach snapshot interface?14:43
DuncanTxyang: Certainly, looking now14:43
xyangDuncanT: thanks!14:43
*** gouthamr has quit IRC14:44
*** nlevinki has quit IRC14:44
openstackgerritIan Cordasco proposed openstack/python-cinderclient: Remove unnecessary check for tenant information  https://review.openstack.org/19959514:45
*** severson_ has quit IRC14:46
*** ronis_ has quit IRC14:46
thingeeerlon: nope didn't see it. might've not been approved yet http://lists.openstack.org/pipermail/third-party-announce/2015-July/000232.html14:47
*** salv-orlando has quit IRC14:47
thingeeerlon: as you can see, there is no next message.14:47
erlonthingee: hmm, that's moderated :/14:47
thingeeerlon: yea14:48
thingeeerlon: can we get a ci run on that patch?14:48
*** gouthamr has joined #openstack-cinder14:49
erlonthingee: you mean Hitachi CI?14:49
erlonthingee: it run14:49
erlonthingee: but only the CI that was having problem (hitachi-horcm-fc)14:50
openstackgerritMichal Dulko proposed openstack/cinder: WIP: Refactoring of manager's create_volume flow  https://review.openstack.org/19316714:50
erlonthingee: about the fails, in general lines, we had a power failure this weekend and only manage to get it working yesterday in the end of the day. Is not 100% stable but we are working on14:51
*** belmoreira has quit IRC14:53
DuncanTxyang: Non-disruptive backup looks good now, I'll spin up a devstack and play a bit but I think I can +2 it soon14:53
thingeeerlon: ok, we'll look for a run from the hitachi ci with this patch. I'll check back on it later today.14:53
*** jdandrea has quit IRC14:54
*** jdandrea has joined #openstack-cinder14:55
mclarenthingee: quick query on https://review.openstack.org/#/c/173977 ... I'm not 100% sure why cinderclient can't just pin (<=) to the kilo keystone client. What else needs to happen? Thanks14:55
xyangDuncanT: great!  thanks!14:55
*** rhe00_ has quit IRC14:55
*** Apoorva has joined #openstack-cinder14:56
thingeemclaren: just like with any requirements proposal, we just match what's in global requirements https://github.com/openstack/requirements/blob/stable/kilo/global-requirements.txt#L12714:56
thingeemclaren: once the proposal bot updates the patch, we can merge it14:57
thingeeor someone else can update it. I don't mind.14:57
thingeeI'll just do it14:58
*** diogogmt has joined #openstack-cinder14:58
*** jdandrea has quit IRC14:59
mclarenthingee: ah, the current patch is stale ... I get it now -- thanks!15:00
*** jdurgin1 has joined #openstack-cinder15:00
*** kaisers_ has joined #openstack-cinder15:01
thingeemclaren: alright updated https://review.openstack.org/#/c/173977/15:02
*** bluex-pl has quit IRC15:02
*** dims has quit IRC15:04
*** coolsvap is now known as coolsvap|away15:04
*** dims has joined #openstack-cinder15:05
mclarenthanks!15:06
*** tellesnobrega_ has quit IRC15:06
*** MentalRay has joined #openstack-cinder15:09
*** jdandrea has joined #openstack-cinder15:09
*** dims has quit IRC15:10
*** bkopilov has joined #openstack-cinder15:10
*** coolsvap|away is now known as coolsvap15:12
*** nkrinner has quit IRC15:12
*** jdandrea has quit IRC15:12
*** kaisers_ has quit IRC15:13
*** mclaren has quit IRC15:14
*** bkopilov has quit IRC15:15
*** primechuck has joined #openstack-cinder15:15
*** rhe00_ has joined #openstack-cinder15:18
*** jgriffith has quit IRC15:20
*** liewegas has quit IRC15:20
*** boris-42 has joined #openstack-cinder15:21
*** liewegas has joined #openstack-cinder15:22
*** dims has joined #openstack-cinder15:23
*** topshare_ has joined #openstack-cinder15:25
*** kaisers_ has joined #openstack-cinder15:26
*** setmason has joined #openstack-cinder15:26
*** topshare has quit IRC15:26
*** tsekiyama has joined #openstack-cinder15:27
*** bkopilov has joined #openstack-cinder15:27
*** hemnafk is now known as hemna15:28
*** kaisers_ has quit IRC15:29
*** kaisers_ has joined #openstack-cinder15:31
*** asselin__ has quit IRC15:32
*** jgriffith has joined #openstack-cinder15:32
*** jdandrea has joined #openstack-cinder15:34
*** timcl1 has quit IRC15:34
*** kaisers_42 has joined #openstack-cinder15:35
*** kaisers_ has quit IRC15:36
*** pots has joined #openstack-cinder15:38
*** gouthamr has quit IRC15:39
*** sseverson has joined #openstack-cinder15:39
*** bswartz has quit IRC15:39
*** merooney has quit IRC15:39
*** gouthamr has joined #openstack-cinder15:40
*** e0ne has joined #openstack-cinder15:40
*** kaisers_42 has quit IRC15:40
openstackgerrithadi esiely proposed openstack/cinder: Store volume encryption metadata on each volume  https://review.openstack.org/15228415:40
*** zul has quit IRC15:41
*** david-ly_ is now known as david-lyle15:42
*** dustins has quit IRC15:43
*** annegentle has quit IRC15:45
*** laughterwym has quit IRC15:45
*** timcl has joined #openstack-cinder15:45
*** adurbin_ has joined #openstack-cinder15:45
eharneyis there a meeting today?15:46
*** annegentle has joined #openstack-cinder15:46
*** rhe00_ has quit IRC15:46
*** daneyon has joined #openstack-cinder15:46
e0neeharney: there is no agenda yet. but may be thingee would like to make some weekly anouncements15:46
*** coolsvap is now known as coolsvap|away15:47
*** salv-orlando has joined #openstack-cinder15:47
thingeee0ne: I don't have anything to announce this week.15:47
e0nethingee: thanks :)15:48
*** coolsvap|away is now known as coolsvap15:48
*** jasondotstar has quit IRC15:48
geguileoSo it's going to be a quick meeting :-)15:48
*** kaisers_ has joined #openstack-cinder15:48
dulekCan someone take a look on http://logs.openstack.org/90/199590/1/check/gate-cinder-python27/50c3959/console.html ?15:49
dulekI don't understand why it failed.15:49
*** patrickeast has joined #openstack-cinder15:49
*** abehl has quit IRC15:49
dulek"This test has failed because it attempted to execute commands"?15:49
eharneydulek: some test isn't mocking out commands, there's a script that runs at the end to detect commands that were run under sudo15:49
*** BharatK has joined #openstack-cinder15:50
dulekeharney: Okay, I'll take a look into that. Thanks!15:50
*** daneyon has quit IRC15:50
eharneyjobernar ran into this same thing yesterday on a similar code path...15:50
eharneyjbernard *15:50
geguileoeharney: I was thinking about jobernar problem yesterday as well  :-)15:51
jbernardyep15:51
jbernardsince many use vagrant, which requires passwordless-sudo, you don't catch it typically15:51
dulekAh, I think I understand that - when reworking the UT I've fixed one that wasn't doing what it was supposed to do.15:51
*** kaisers_ has quit IRC15:51
dulekNow it is and commands weren't mocked.15:51
*** yamada-h has joined #openstack-cinder15:51
*** yamada-h has quit IRC15:51
dulekAny possiblity to run such check locally?15:52
jbernarddulek: i put a 'raise' before the offending command from the gate logs15:52
jbernarddulek: and that turns up the culprit quickly15:52
dulekjbernard: Cool, thanks!15:53
*** patrickeast has quit IRC15:53
jbernarddulek: np15:53
eharneyif you run unit tests on an environment that requires passwords for sudo, it's usually pretty easy to narrow down15:53
eharneythe test will hang and time out and tell you which test it is15:54
openstackgerritTom Swanson proposed openstack/cinder: Dell SC: Add support for ManageableVD  https://review.openstack.org/19963115:54
*** jistr has quit IRC15:56
*** salv-orlando has quit IRC15:56
*** kaisers_ has joined #openstack-cinder15:56
*** jasondotstar has joined #openstack-cinder15:58
*** patrickeast has joined #openstack-cinder15:59
*** kaisers_ has quit IRC15:59
*** bswartz has joined #openstack-cinder16:00
*** salv-orl_ has joined #openstack-cinder16:00
*** tellesnobrega_ has joined #openstack-cinder16:00
*** rhe00_ has joined #openstack-cinder16:01
*** jungleboyj has joined #openstack-cinder16:04
*** kaisers_ has joined #openstack-cinder16:05
*** Yogi1 has joined #openstack-cinder16:05
jgriffithakerr_: did you see my comment on your trace logging patch?16:06
jungleboyjAnd we're back.16:06
smcginnisQuick one. :)16:06
akerr_jgriffith: don't think so... but that's ameade's patch if its the one i'm thinking of.  let me check16:07
jgriffithakerr_: ahh.. I think you're right :)16:07
jgriffithakerr_: in a nut-shell.... Seems like an ok idea, but I'm wondering isn't that just "debug"?16:07
*** rushil has joined #openstack-cinder16:08
jgriffithakerr_: and if we want "trace" level logging options, rather than write our own in Cinder, shouldn't we consider adding it to oslo.log as an option there instead?16:08
akerr_jgriffith: like, why have the option to turn it on and off?16:08
*** zul has joined #openstack-cinder16:08
jgriffithakerr_: well, no, not really16:08
jgriffithakerr_: I get that16:08
jgriffithakerr_: but why it's not just the existing "debug" level logging16:08
jgriffithakerr_: or like I said, add trace level option to oslo.log16:09
*** kaisers_ has quit IRC16:09
akerr_i thought it was using the debug level logging16:09
*** patrickeast has quit IRC16:09
*** patrickeast has joined #openstack-cinder16:10
*** rajinir_r has joined #openstack-cinder16:10
jgriffithOh.. well shit, DuncanT merged it16:10
*** dontalton has joined #openstack-cinder16:10
jgriffithOk, then, never mind16:10
thingeejgriffith: I was going to say, i think I saw that in the git logs this morning =/16:10
jgriffithakerr_: ameade https://review.openstack.org/#/c/157439/1516:10
tbarronjgriffith: ameade replied to you and he does want to go to oslo eventually16:11
ameadeo?16:11
ameadeo/16:11
ameadelol16:11
*** leeantho has joined #openstack-cinder16:11
*** coolsvap is now known as coolsvap|away16:11
rhe00_thingee: I am still catching up on things as I got back from vacation today. Will take a look at the CI shortly. I had some occasional trouble with nova network not giving out valid IP addresses before I left, but it should still have been reporting with a 66% success rate16:11
*** annasort has joined #openstack-cinder16:12
ameadejgriffith: there used to be the 'audit' log level that openstack added16:12
DuncanTHmmm, I did what?16:12
rhe00_thingee: Your emails ended up in a Openstack folder that wasn't synced to my phone so I didn't see them during my vacation. Sorry about that.16:12
thingeerhe00_: ok, will we already removed the driver, but we can of course readd things. Just wasn't sure if you disappeared since it has been almost 30 days.16:13
jgriffithDuncanT: oh.. nothing; I was just pointing out my questions in a patch to ameade16:13
DuncanTjgriffith: Ah, the tracing decorators?16:13
jgriffithDuncanT: but found it was moot as it merged this morning16:13
ameadeeyes one this? https://review.openstack.org/#/c/199170/16:13
thingeerhe00_: no thorough review or anything, just revert my removal after the CI is reporting again.16:13
*** kaisers_ has joined #openstack-cinder16:13
rhe00_thingee: I saw that. I will get it back up and running ASAP.16:13
*** markus_z has quit IRC16:15
*** dims has quit IRC16:15
DuncanTameade: audit was added for something very specific, it got misunderstood, misused then removed16:16
e0nejgriffith: fyi, updated spec about wsgi: https://review.openstack.org/192683. i think it won't be merged w/o your vote:)16:16
ameadeDuncanT: ah, figures i spose16:16
jgriffithsilly request... can someobdy ping me on another channel or PM?16:16
ameadejgriffith: done16:16
*** kaisers_ has quit IRC16:16
*** dims has joined #openstack-cinder16:16
*** merooney has joined #openstack-cinder16:16
jgriffithOk.. works, thanks16:16
jgriffithI completely rewrote my irssi config and thought I broke it16:17
ameadeoh i used to use irssi, wrote a plugin at some point16:17
smcginnisjgriffith: Using hilight?16:17
ameadein perl >.<16:17
ameadehttps://github.com/ameade/queryhistory_irssi16:18
jgriffithsmcginnis: Yeah16:18
ameadeshameless plug16:18
openstackgerritDaniel Tadrzak proposed openstack/cinder: Cleanup for SnapshotObject  https://review.openstack.org/19834116:18
jgriffithameade: cool!16:18
xyangthingee: XtremIO CI is reporting now.  Can you please remove -2 from this patch? https://review.openstack.org/#/c/189135/16:19
ameadeso that i could close all those darn query windows with people16:19
ameadeand still have context if i open them back up16:19
*** jdurgin1 has quit IRC16:19
ameadenow i use irccloud.com though16:19
*** zhenguo has quit IRC16:20
*** rakesh_mishra has quit IRC16:20
*** ronis_ has joined #openstack-cinder16:20
*** jordanP has quit IRC16:20
*** dims has quit IRC16:21
*** jbrc has joined #openstack-cinder16:21
DuncanTI'm a fan of irccloud16:24
*** jasondotstar has quit IRC16:25
winston-d5$ monthly, that's as much as a ting pod in digi-ocean.16:26
jungleboyjDuncanT: I just started using that on my iPad.  Haven't really messed with it though.16:26
DuncanTjungleboyj: For the most part, it just works[tm]16:26
DuncanTjungleboyj: The client side stuff is open source, so I might replace it with something self-hosted one day, but right now, laziness is winning out16:27
jungleboyjDuncanT: So, it is 'is a good thing' [tm]16:27
winston-ddoes it's app supports sending notifications to your Apple watch? (i'm android user)16:27
*** _cjones_ has joined #openstack-cinder16:28
openstackgerritWalter A. Boring IV (hemna) proposed openstack/os-brick: Added ABCMeta class to the InitiatorConnector  https://review.openstack.org/19964216:28
DuncanTwinston-d: I know nothing about apple watches. Took me a while to stop it sending notifications to my pebble though16:28
openstackgerritTom Barron proposed openstack/cinder: Fix NetApp cDOT driver use of Glance locations  https://review.openstack.org/19359716:28
jungleboyjDuncanT: Interesting, I hadn't even really looked at what all this does.  Need to look closer.16:29
jgriffithwinston-d: I'm buying you an Apple Watch for Christmas!16:29
*** changbl has quit IRC16:29
jgriffithwinston-d: and you MUST wear it in shame :)16:29
*** jaypipes has quit IRC16:29
openstackgerritWalter A. Boring IV (hemna) proposed openstack/os-brick: Added ABCMeta class to the InitiatorConnector  https://review.openstack.org/19964216:30
winston-djgriffith: buy me with all kinds of watches if you like, i'd be happy to wear them all.16:31
jgriffithwinston-d: LOL16:31
jgriffithwinston-d: Yeah I realized that wasn't such a cruel joke after I said it :)16:32
jgriffithhemna: there's only like two methods in there16:32
hemnayup16:33
jgriffithhehe16:33
hemnaI plan on adding a few more16:33
openstackgerritJon Bernard proposed openstack/cinder: Add support for file I/O volume migration  https://review.openstack.org/18727016:33
hemnaI'm working on something that will help do live migration verification16:33
hemnaand a new brick Connector method is going to be involved with that, so I wanted to get the ABC in first, and then add the new method next.16:33
jgriffithhemna: you mean like a tempest test that "runs live migration"16:33
hemnasorta16:34
tbarronwinston-d: i'm also using a $5 digital ocean vm to run my irc bouncer16:34
hemnabasically after a VM has been moved from the src to the destination, verify that all of the entries on the src are now gone.16:34
hemnano dangling volumes16:34
hemnawhich cause problems later.16:34
hemnathe brick connector classes will get a new public method to ask for discovered paths for a volume.16:35
*** jwcroppe has quit IRC16:35
hemnamake that call before the migration, and then the same call after migration16:35
hemnato verify that they are gone.16:35
jgriffithhemna: and if they're not?16:35
hemnaif they aren't, then we have a dirty state16:36
winston-dtbarron: i also used it to get over the invisible wall we have in China.16:36
jgriffithhemna: well... my question is how do you fix that?16:36
tbarronwinston-d: ++16:36
hemnait depends16:36
jgriffithhemna: isn't that a problem with the initiator?16:36
hemnafrom the testing that we've seen, there are bugs in drivers that cause this, and also there are problems with the way nova is calling cinder that can cause it as well.16:36
hemna:(16:36
hemnaand also there could be problems with the code that removes the paths16:37
jgriffithhemna: do you have an example of a driver with a bug?16:37
hemnathe disconnect_volume call16:37
hemnawe have an etherpad with our current testing results16:38
hemnajgriffith, https://etherpad.openstack.org/p/live-migration-results16:38
hemna3par iSCSI is borked, and lvm iscsi looks borked from what we've seen.16:38
hemna3par is our problem for sure and we are working on it.16:39
SwansonSooo... long() doesn't exist in python3.4?16:39
hemnabut I'm trying to get some tools in place that can help others verify live migration as well16:39
jgriffithhemna: odd, LVM worked in Kilo16:39
hemnaso it's not just something that we use in house.16:39
jgriffithbut who knows, I haven't tried it in a while16:39
smcginnisSwanson: I believe int is now long.16:39
jgriffithhemna: so I was confused I guess, you said "some drivers had bugs" I thought you identified something in code in other peoples drivers maybe16:40
jgriffithhemna: that's what I was curious about16:40
*** dims has joined #openstack-cinder16:40
hemnajgriffith, I have yes16:40
hemnait's just not on that etherpad16:40
hemnajust haven't had time to document it16:40
hemnabasically nova calls initialize_connection directly many times during live migration16:41
jgriffithhemna: oh... well maybe you should share that info so those drivers / maintainers can have a look?16:41
smcginnisSwanson: https://www.python.org/dev/peps/pep-0237/16:41
jgriffithhemna: and also to share the info16:41
patrickeastsomewhat related, we’ve seen issues with nova leaving behind dangling multipath connections https://bugs.launchpad.net/nova/+bug/138579816:41
openstackLaunchpad bug 1385798 in OpenStack Compute (nova) "Multipath ISCSI connections left open after disconnecting volume with libvirt" [Undecided,Confirmed] - Assigned to Danny Wilson (daniel-wilson)16:41
hemnaand some drivers do an export every time initialize_connection is called.16:41
patrickeastdoesn’t happen with brick though16:41
hemnawhen nova simply expects initialize_connection to return the connection_info dict for a volume.16:41
hemnaanyway16:41
*** dontalton is now known as bitblt16:41
*** bitblt has quit IRC16:41
*** bitblt has joined #openstack-cinder16:41
hemnaI was going to go through the drivers today or tomorrow and document which ones do that and which don't16:41
*** bitblt has quit IRC16:42
hemnaI just haven't had time yet.16:42
*** afazekas has quit IRC16:42
hemnaI want to get live migration to the point where it just works for everyone.16:42
winston-dhemna: ++16:43
*** xyang has quit IRC16:43
winston-dit should just works16:43
*** xyang has joined #openstack-cinder16:43
hemnaand also allow us a way to verify that it's not left in a dirty state, for everyone.16:43
openstackgerritxing-yang proposed openstack/cinder: Add multiple pools support to VMAX driver  https://review.openstack.org/18999016:43
winston-dwith conditions and terms applied16:43
jgriffithhemna: yeah, that would be good16:43
jgriffithhemna: just seems like there are only a few of use that have ever even tried it :(16:43
jgriffithwhich is too bad, and likely part of the problem16:43
hemnait's been the bane of my existence for a while now.16:43
hemnayah16:44
jgriffithI'm just lucky... after a couple patches to Nova it just "works" for me (and SF and LVM)16:44
hemnamy plan is to work with asselin to see how we can get CI for LM going and add the verification.16:44
*** primechuck has quit IRC16:44
hemnawe need several different scenarios tested, not just a single VM fired up with a single volume and it works migrated one way16:44
jgriffithsingle vm with single volume?16:45
hemnajgriffith, some drivers are lucky.  lefthand is also one of them where it seems to just work.16:45
*** xyang has quit IRC16:46
hemnayah that's the simplest of cases that's easy to try.16:46
hemnabut isn't really real world IMHO16:46
*** xyang has joined #openstack-cinder16:46
hemnawe laid out our test scenarios in that etherpad, 7 of them.16:46
jgriffithI guess I don't follow... you mean multi-compute nodes though?16:46
hemnaall of our tests are with multiple compute nodes16:47
jgriffithhemna: oh, I see... makes sense16:47
hemnawhere case 1. is you bring openstack up (multiple nodes) and you fire up 1 VM, attach a cinder volume to it, and migrate it one way.  done.16:47
*** garthb has joined #openstack-cinder16:47
jgriffithI think you're missing the most important one there....16:47
jgriffithboot from volume16:48
hemnasure16:48
jgriffithall the others I would think become sort of null if multi-attach ever lands and works16:48
hemnayah we should add boot from volume for sure16:48
hemnathanks for that one :)16:49
haypohemna: hi16:50
hemnahey16:51
haypohemna: we told me that you can release a new version of os-brick?16:51
hemnaI can't, as I'm not the PTL :)16:51
hemnaos-brick is owned by Cinder, which means only the PTL can do a release.16:51
haypojgriffith: hey, what about you? :) can you release a new version of os-brick?16:53
openstackgerritTom Swanson proposed openstack/cinder: Dell SC: Add support for ManageableVD  https://review.openstack.org/19963116:53
smcginnishaypo: You need to hit up thingee.16:53
jgriffithhaypo: as hemna mentioned I believe thingee owns that now16:54
jgriffithhaypo: I "could" but then he'd have to kill me after16:54
e0ne:)16:54
thingeehaypo, jgriffith, hemna: if we're ready, I can do one today. jgriffith  nor I can push these anymore since the changes to library releases.16:54
tbarrone0ne: thanks for the review of https://review.openstack.org/19917316:54
thingeeit has to be proposed16:55
hayposorry, jgriffith is not the current cinder PTL? sorry. so it's thingee?16:55
jgriffiththingee: oh?  Interesting16:55
thingeehaypo: correct'16:55
tbarrone0ne: w.r.t. the question about (and I understand that it's a moot question for this cherrypick):16:55
jgriffithhaypo: yes, thingee has been PTL for like a year16:55
haypothingee: i would prefer to see https://review.openstack.org/#/c/197624/ merged into os-brick, but hemna doesn't want it :)16:55
tbarrone0ne: (_, export_path) = self._get_export_ip_path(share=share)16:55
e0netbarron: you're welcome16:55
*** vokt has joined #openstack-cinder16:55
e0netbarron: it'ls ok for cherry pick16:56
jgriffiththingee: so what's the new rule on the libs?16:56
shyamahemna: https://review.openstack.org/#/c/198400/ should this be fixed in nova rather than in cinder?16:56
e0netbarron: i don't know if any guidelines exists16:56
tbarrone0ne: you make the good observation that the '_' can be confused with _ from i18n lib16:56
setmasonIs there a more efficient method of determining the number of cinder volumes you have, besides cinder list —all-tenants? I’ve got about 5000 volumes and using this command with —marker and then eventually piping into wc -l, seems a bit kludgy.16:56
thingeejgriffith: http://lists.openstack.org/pipermail/openstack-dev/2015-June/066346.html16:56
tbarrone0ne: for future reference, is there a good way in OpenStack to indicate an unused variable?16:57
jgriffiththingee: thanks16:57
haypohemna: so you're still unable to run tests using testr? https://review.openstack.org/#/c/197624/16:57
e0netbarron: but IMO, it would be better something like  (_your_unused_variable_short_name, export_path) = self._get_export_ip_path(share=share)16:57
haypohemna: it's probably not a big deal, just an annoying issue in your setup16:57
tbarrone0ne: yeah, I think that makes sense, will do going forwards16:57
e0netbarron: thank you, Tom16:57
jgriffithsetmason: sadly you best option might be a DB querie at that point if you ask me16:58
setmasoneww16:58
tbarrone0ne: no, thank you Ivan :-)  Your review on this one was very helpful.16:58
setmasonI dont’ want to increase osapi_max_limit…16:58
jgriffithsetmason: we should have an admin api call for that sort of thing16:59
haypothingee: i'm asking a new os-brick release to get hemna fixes for python 3. os-brick blocks four of my python 3 patches (because the current os-brick release doesn't work on python 3)16:59
haypothingee: (patches for cinder)16:59
jgriffithsetmason: if you file a bug, I'll mark it as wishlist and hack something up for it16:59
thingeehaypo: in the middle of a meeting, so I'll need to get back to you16:59
haypothingee: ok17:00
jgriffithsetmason: probably should be something under cinder-manage17:00
*** timcl1 has joined #openstack-cinder17:00
jgriffithsetmason: unless you have a global admin17:00
hemnameeting....17:00
*** hemna is now known as hemnafk17:00
jgriffithsetmason: somebody like DuncanT might have some suggestions/insights on solving that17:00
*** jwcroppe has joined #openstack-cinder17:01
*** dustins has joined #openstack-cinder17:02
*** topshare_ has quit IRC17:02
DuncanTsetmason: There's been a suggestion before that we should have a stats API, but nobody has actually coded it (volume count & total size per backend, snapshot counts and total size per backend, backups count and total size....)17:02
winston-dhemnafk:17:02
setmasonDuncanT: Thanks17:03
DuncanTsetmason: Shouldn't be particularly hard to write, the info is all in the DB17:03
*** timcl has quit IRC17:03
jgriffithDuncanT: setmason so that's a good point... but what about "I want *all* volumes across *all* tenants"17:03
jgriffithDuncanT: setmason seems like that crosses the line over to cinder-manage, which could just do the db querie for you?17:03
*** MentalRay has quit IRC17:03
jgriffithDuncanT: seem kosher?17:03
setmasonok. will look at cinder-manage17:04
jgriffithDuncanT: otherwise you have the whole "run it for 'n' tenants" problem.  Unless you have some uber tenant to set in the policy file on it17:04
*** MentalRay has joined #openstack-cinder17:04
setmasonwell it would seem that an extension to cinder list would be more appropriate17:04
jgriffithwhich I'm not quite sure how you'd do that17:04
jgriffithsetmason: I'd agree17:05
setmasonas we’re running this to determine cinder stats, as a provider.17:05
*** bswartz has quit IRC17:05
jgriffithsetmason: assuming you have a user that accesses *all* tenants17:05
jgriffithsetmason: then it's actually really easy17:05
DuncanTjgriffith: cinder-manage could do it, but it is the kind of thing that is really useful on a dashboard. admin-only by default I guess, but 'count * where' clauses shouldn't kill the db17:05
jgriffithsetmason: we already have a call to get all for all tenants17:05
DuncanTjgriffith: Just getting the count, sum etc is rather lighter weight than listing 10k volumes and doing the math on the client17:06
setmasonWe’re running this as an admin, really to populate dashboards and reports.17:06
jgriffithDuncanT: yeah... makes sense; I'm just unclear on how to deal with multiple projects and rules around that17:06
*** rushil has quit IRC17:07
DuncanTjgriffith: I'd say these were system stats, and just ignore projects etc. Filtering can be added later17:07
setmasonjgriffith: the cinder list —all-tenants does what I need, but it’s limited by osapi_max_limit.  The workaround is to use markers every $osapi_max_limit.17:07
jgriffithsetmason: DuncanT ok... that makes sense17:07
jgriffithsetmason: DuncanT if nobody else wants to tackle it I'm happy to add it to my list17:07
jgriffithbtw, isn't part of this supposed to be what ceilometer is all about :)17:08
*** bswartz has joined #openstack-cinder17:08
openstackgerritMichal Dulko proposed openstack/cinder: Remove unused arguments from c-vol's create_volume  https://review.openstack.org/19959017:09
DuncanTjgriffith: ceilometer would basically  have to keep a complete copy of the cinder db to know this info... I don't know where the line should be drawn, but it seems like a useful sort of API to have17:09
*** mutoulbj has quit IRC17:12
openstackgerritIan Cordasco proposed openstack/python-cinderclient: Remove unnecessary check for tenant information  https://review.openstack.org/19959517:12
*** daneyon has joined #openstack-cinder17:14
*** Apoorva has quit IRC17:15
openstackgerritAdriano Freires Rosso proposed openstack/cinder: Adds manage/unmanage methods for HNAS drivers  https://review.openstack.org/19799917:16
*** timcl1 has quit IRC17:17
*** timcl has joined #openstack-cinder17:18
*** rushil has joined #openstack-cinder17:18
*** lpetrut has quit IRC17:18
*** tellesnobrega_ has quit IRC17:19
*** daneyon has quit IRC17:19
*** primechuck has joined #openstack-cinder17:19
*** ronis_ has quit IRC17:20
*** rushil has quit IRC17:21
*** rushil has joined #openstack-cinder17:21
*** winston-d_ has joined #openstack-cinder17:21
*** Apoorva has joined #openstack-cinder17:23
*** Apoorva has quit IRC17:24
*** ociuhandu has quit IRC17:25
*** Apoorva has joined #openstack-cinder17:26
*** bkopilov has quit IRC17:28
*** e0ne has quit IRC17:28
*** kjnelson has quit IRC17:31
*** BharatK has quit IRC17:32
*** kjnelson1 has joined #openstack-cinder17:33
*** bkopilov has joined #openstack-cinder17:36
*** dannywilson has joined #openstack-cinder17:37
*** brmj has joined #openstack-cinder17:37
brmjjgriffith:ping17:38
*** dims has quit IRC17:39
*** dims has joined #openstack-cinder17:39
brmjjgriffith:NM17:40
*** Apoorva has quit IRC17:41
*** hemnafk is now known as hemna17:42
*** brmj has quit IRC17:43
*** raunak has joined #openstack-cinder17:45
*** MentalRay has quit IRC17:46
openstackgerritTom Swanson proposed openstack/cinder: Dell SC: Add support for ManageableVD  https://review.openstack.org/19963117:48
Swanson@#$!!$@!!!!17:48
*** Apoorva has joined #openstack-cinder17:49
*** lpetrut has joined #openstack-cinder17:51
*** delattec has joined #openstack-cinder17:52
*** annashen has joined #openstack-cinder17:53
*** crose has quit IRC17:53
*** sseverson has quit IRC17:53
*** crose has joined #openstack-cinder17:53
*** bkopilov has quit IRC17:54
*** cdelatte has quit IRC17:56
*** Longgeek has quit IRC17:57
smcginnisSwanson: Is your VD unmanageable?17:57
*** Longgeek has joined #openstack-cinder17:57
*** Longgeek has quit IRC17:57
*** aix has quit IRC17:57
Swansonsmcginnis: It is find so long as I don't implement TransferVD.17:58
Swansonfine17:58
openstackgerritWalter A. Boring IV (hemna) proposed openstack/os-brick: Added ABCMeta class to the InitiatorConnector  https://review.openstack.org/19964217:58
smcginnisSwanson: You don't want to transfer your VD.17:58
*** lpetrut has quit IRC17:58
*** Longgeek has joined #openstack-cinder17:59
*** kjnelson1 has quit IRC18:01
openstackgerritxing-yang proposed openstack/os-brick: Add connector driver for the ScaleIO cinder driver  https://review.openstack.org/18583518:01
*** xyang has quit IRC18:01
*** dannywilson has quit IRC18:02
*** dannywilson has joined #openstack-cinder18:03
*** kjnelson has joined #openstack-cinder18:03
*** xyang has joined #openstack-cinder18:06
*** xyang has quit IRC18:07
*** xyang has joined #openstack-cinder18:07
*** tellesnobrega_ has joined #openstack-cinder18:08
*** timcl has quit IRC18:09
*** bkopilov has joined #openstack-cinder18:09
*** skylerberg has joined #openstack-cinder18:10
*** resker has joined #openstack-cinder18:12
*** bpoulos has joined #openstack-cinder18:16
*** esker has quit IRC18:16
thingeexyang: looks like the recent emc xio runs are failing often, but I removed the -218:17
*** dannywilson has quit IRC18:18
*** xyang1 has quit IRC18:19
*** ronis_ has joined #openstack-cinder18:20
xyangthingee: there was another hardware issue a few days ago.  that's probably why.  it was just fixed.  hopefully it will run more stable now.  we'll keep monitoring it.  thanks!18:22
*** boris-42 has quit IRC18:22
openstackgerritErickson Filipe Guedes dos Santos proposed openstack/cinder: Nested Quota: Set default values to subproject  https://review.openstack.org/19440618:22
thingeexyang: this is recent runs from today http://paste.openstack.org/show/355957/18:24
*** resker has quit IRC18:25
xyangthingee: ok, we'll take a look.  I thought it is fixed18:25
openstackgerritKendall Nelson proposed openstack/cinder: Enable cinder-manage to delete old services from the database  https://review.openstack.org/19761318:29
thingeehaypo: back18:32
thingeehemna: can you recheck https://review.openstack.org/#/c/197624/218:33
hemnaok18:33
hemnaI had just tried it last night on a stock ubuntu 14.04.18:33
thingeehaypo: ok, let see about getting this change in and let me double check with hemna if we need anything else in the release18:33
hemnaI ran that on a clean system w/o an existing .testrepository18:34
thingeehemna: ok, I just tried it with the instruction haypo gave and it works for me18:34
*** bswartz has quit IRC18:34
hemnaand if python-gdbm is required, then that has to be handled by the patch.18:34
thingeehttps://www.irccloud.com/pastebin/0l6mVs2u/18:34
*** jbrc has quit IRC18:35
thingeehaypo: ^18:35
*** dustins has quit IRC18:36
*** rushil has quit IRC18:38
thingeehemna: ok I'll wait for things to be addressed there for release.18:39
*** esker has joined #openstack-cinder18:39
thingeehemna: anything else? https://review.openstack.org/#/q/project:openstack/os-brick+status:open,n,z18:39
*** timcl has joined #openstack-cinder18:39
*** dustins has joined #openstack-cinder18:39
thingeehemna: fyi, these connectors are low priority at this point. nova has already stopped accepting new bps18:39
hemnayah the new connectors are low IMHO18:40
thingeehemna: https://wiki.openstack.org/wiki/Nova/Liberty_Release_Schedule18:40
hemnathey need shims in nova for them to be used at all anyway18:40
thingeehemna: right18:40
thingeehemna: once we have the release, can we add rbd to this https://review.openstack.org/#/c/175569/23 ?18:40
hemnaso, I think the RBD one is special in libvirt18:41
hemnabecause libvirt itself has native RBD support18:41
thingeehemna: I was just talking to John Garbutt about this, and he would like things to be done in the same patch...oh ok so we don't need to update anything there18:41
hemnathat's why I didn't touch it.  also the RBD connector in os-brick isn't the same as what's in libvirt volume.py18:42
hemnaI'm not an RBD expert though18:42
hemnabut that was my take on it FWIW18:42
thingeehemna: ok, if you can reply to me in the review about that, just so others know.18:42
thingeejbernard: ^18:42
hemnathingee, ok added my response FWIW18:44
jbernardill take a look18:44
bpoulosi have a question about the python-cinderclient release from yesterday (1.3.0)18:44
bpoulosis it going to be re-released?18:44
bpoulosI saw there was a comment about permissions issues when pushing it, but I didn't see any follow up on the mailing list18:45
jbernardhemna: i didn't realize that was going to land so soon18:45
hemnajbernard, well that's the hope.  is to finally get os-brick used in nova for L18:45
jbernardhemna: would it help if post an etherpad on what common pieces of the rbd connector nova could potentially use?18:46
jbernardhemna: there appears to be a large portion of libvirt-specific code, but not all of it18:46
thingeebpoulos: the permission issues is just a change to the process for pushing something to pypi.18:46
hemnahonestly, I don't know it that well.  from what I remember the rdb libvirt volume.py code relied on the native libvirt RBD support to do volume attaches18:47
hemnawhich really wouldn't work in os-brick18:47
thingeegeguileo: is it safe for me to assume you're spearheading this https://etherpad.openstack.org/p/CinderNovaAPI ?18:47
*** jamielennox is now known as jamielennox|away18:47
bpoulosthingee: I downloaded 1.3.0 from pypi, and it looks to be out of date18:47
hemnasince os-brick is used by cinder outside of libvirt entirely.18:47
thingeebpoulos: how so?18:47
bpoulosfrom somewhere between 1.1.1 and 1.2.018:47
hemnajbernard, I'm not sure how you'd go about abstracting those 2 out.18:47
jbernardhemna: that was my understanding as well, but i have yet to thoroughly go through it18:47
bpoulosthingee: it doesn't have the latest commits18:47
hemnadoesn't mean it couldn't be done18:48
bpoulosthingee: everything in https://review.openstack.org/#/c/199242/1/ is missing18:48
geguileothingee: No, I think that'd be hemna18:48
hemnabut for L, I think we should punt on it.18:48
jbernardhemna: i will at least take a look at the review you posted18:48
hemnajbernard, ok great thanks18:48
bpoulosthingee: I downloaded it from https://pypi.python.org/pypi/python-cinderclient/1.3.018:48
jbernardhemna: i agree, getting it landed at all woudl be a huge win18:48
jbernardhemna: i can follow up with a secondar patch for rbd if needed18:48
hemnajbernard, ok great18:49
hemnathat works for me.18:49
jbernardcool18:49
bpoulosthingee: the release notes also look like they're out of date http://docs.openstack.org/developer/python-cinderclient/18:49
thingeebpoulos: whoa you're right. let me talk to the release team about this18:49
thingeehttps://github.com/openstack/python-cinderclient/commits/1.3.018:49
bpoulosthingee: thanks for looking into it!18:50
*** patrickeast_ has joined #openstack-cinder18:51
thingeehemna, geguileo are you spearheading this https://etherpad.openstack.org/p/CinderNovaAPI18:51
geguileothingee: I'm not working on that18:52
hemnathingee, scottda18:52
thingeehaha18:52
thingeescottda: Is it you? :D18:52
hemnathingee, scottda has done the lion's share of the work on that etherpad18:53
thingeehemna, scottda, I was talking to John Garbutt, and he mentioned it would be good if someone wanted to raise this in the next Nova meeting.18:54
hemnaheh yah18:54
*** jamielennox|away is now known as jamielennox18:54
thingeehemna, scottda: I went ahead and tagged these bugs as 'volumes' so that it would get the attention of the people who are focusing on these type of issues, per John's instructions.18:55
thingeehemna, scottda: but I think it would be good for someone to bring it up18:55
hemnaI'm also working with asselin to work on something that might help verify live migration issues, where volumes get left behind on the source.18:55
*** kjnelson has quit IRC18:55
hemnait's all sort of related18:55
hemnathingee, http://paste.openstack.org/show/356117/18:56
thingeehemna: and there is no .testrepository? http://haypo-notes.readthedocs.org/en/latest/openstack.html#tox-testr-db-type-could-not-be-determined-error18:58
jgriffiththingee: I updated the comments in that bug FWIW: https://bugs.launchpad.net/nova/+bug/142365418:59
openstackLaunchpad bug 1423654 in OpenStack Compute (nova) "Nova rescue causes LVM timeouts after moving attachments" [Medium,Confirmed] - Assigned to John Griffith (john-griffith)18:59
hemnarunning tox runs py27 first, which creates a .testrepository, and causes the py34 failure.18:59
*** MentalRay has joined #openstack-cinder19:00
*** jungleboyj has quit IRC19:00
jgriffiththingee: the root of the problem is that a temp lvm device is created for the rescue, and the dm mapper paths get all honked up19:00
scottdathingee: I can put something on the agenda for the next Nova meeting..19:01
scottdaI had been thinking we might talk about the Cinder side of things at the mid-cycle, and then present some type of well-formed plan to Nova after that.19:02
hemnathingee, ok posted another comment on that py34 patchset.  I show how 1 run works, and how another run fails.  both from a clean clone.19:07
hemnagotta get some chow....bbiab19:07
*** hemna is now known as hemnafk19:08
*** e0ne has joined #openstack-cinder19:09
*** rajinir_r has quit IRC19:09
scottdahemna: What do you think? Do we have enough consenus on Cinder changes to start involving Nova ? It seems there are still multiple ideas being considered around using the DB and '*-ing' states for mutual exclusion, or some type of DLM, not to mentino taskflow...19:09
scottdas/mentino/mention19:09
*** ndipanov has quit IRC19:10
*** simondodsley has quit IRC19:13
*** bitblt has joined #openstack-cinder19:13
thingeescottda: I'm fine with checking ing state19:13
thingeeI'm not convinced with some DLM19:14
thingeehemnafk: ^19:14
*** coolsvap|away is now known as coolsvap19:14
openstackgerritMatt Riedemann proposed openstack/cinder: Set encrypted key in connection_info during initialize  https://review.openstack.org/19367319:14
mriedemthere go all of my beautiful +2s19:15
*** Rockyg has joined #openstack-cinder19:15
*** simondodsley has joined #openstack-cinder19:15
thingeemriedem: pff I already gave it a +219:15
thingeeI can't believe this is still going..haha19:16
scottdaYes, I'm not a fan of DLM.19:16
*** bswartz has joined #openstack-cinder19:16
scottdaBut does the Cinder team in general have any consensus on fixing the Nova <-> Cinder API? It seems like it does not...19:17
scottdaI figured it might take the mid-cycle to sort things out, and then go to Nova with some type of plan.19:18
jgriffithscottda: if you want my 2 cents (which you probably don't) I think there is some concensus on the more "understood" thing19:19
jgriffithscottda: there's def low hanging fruit19:19
jgriffithscottda: clean up of the messy reservce/attach stuff19:19
thingeescottda: besides the ing state checking, what are the other cases?19:19
jgriffithscottda: consolidate the calls in Cinder (Cinder side is a bigger mess right now IMHO)19:19
thingeescottda: that we need consensus on19:20
jgriffithscottda: and making sure status's are checked sufficiently is of course a huge win, can't see an arg there19:20
scottdajgriffith: certainly want your opinion...19:20
thingeejgriffith: can you be more specific on the consolidating?19:20
scottdaI think the long bug list is a good starting point.19:20
jgriffiththingee: scottda so there's more problems on the cinder side IMO than anywhere else19:21
jgriffiththingee: scottda I tried to fix some of this up last year....19:21
jgriffiththingee: scottda some folks use intialize_con to create targets, others use export19:21
scottdajgriffith: agreed that cinder fixup is the first step...I just want to start getting Nova onboard for the changes that will have to take place in Nova as well.19:21
jgriffiththingee: scottda there's a hodge-podge of implementation that resulted from a lack of clear understanding/documentation of the flow and what was intended to do what19:22
jgriffithso folks just interpretted however they interpretted it :)19:22
thingeejgriffith: doesn't initialize_con result in create_export anyways?19:22
jgriffithscottda: for some drivers :)19:23
jgriffithscottda: that's kind of the problem IIRC19:23
thingeejgriffith: this is just coming from the volume manager.19:23
thingeejgriffith: if I'm reading things right19:23
*** r-daneel has joined #openstack-cinder19:23
jgriffithscottda: there was a divergence in there WRT what diff drivers were doing19:23
jgriffithscottda: thingee yeah, but one of the problems is the inheritance structure of the drivers caused some divergence19:24
jgriffithscottda: thingee and some drivers over-rode the methods19:24
jgriffithscottda: thingee anyway... if you guys want I can dig up the details later19:24
scottdaWell, things are pretty messy right now. I've listed a few ideas on what should be done in the etherpad under "Goal: Fixup the CinderNova API"19:24
jgriffithscottda: thingee but really the whole attach process on the Cinder side is not as consistent as it should be.  Unless somebody fixed it all up when I wasn't looking :)19:25
jgriffithscottda: yeah, I'm reading that now....19:25
scottdaPlease add/change/delete whatever...it's just my rough ideas.19:25
thingeejgriffith: I'm saying the manager's initialize connection method sets the order of calling create_export -> initialize_connection ... the driver can only override what those methods do, not the order.19:25
thingeejgriffith: maybe I'm confused here19:25
jgriffithscottda: you've got a lot of awesome stuff there19:25
*** shyama has quit IRC19:25
scottdajust a brain dump...19:26
bswartzhey is the midcycle meetup at the same venue as 12 months ago? HP hosting?19:26
scottdabswartz: yes19:26
scottdahttps://etherpad.openstack.org/p/cinder-liberty-midcycle-meetup19:26
bswartzscottda: ty that's exactly what I was looking for19:27
* bswartz notes that etherpad is all-knowing19:27
scottdaToo bad you cannot search for an etherpad by keywords.19:28
thingeebswartz: also useful https://wiki.openstack.org/wiki/Sprints#Liberty_sprints19:28
bswartzdoes google index the contents of etherpads?19:28
jgriffiththingee: yeah, but look at for example the 3par driver....19:28
jgriffiththingee: it's intialize connection does the "create_export" stuff, and just passes when create_export is called19:29
bswartzthingee: ty19:29
*** esker has quit IRC19:29
thingeejgriffith: haha ok fair19:29
thingeejgriffith: lets note those kind of drivers and fix that?19:29
*** esker has joined #openstack-cinder19:29
jgriffiththingee: I'm not saying that's an issue19:29
jgriffiththingee: just saying... some drivers do it that way, others do it a different way19:30
scottdaI'll add "audit drivers" to the etherpad...19:30
thingeejgriffith: help me fill in the gap...how does this relate with things being broken between nova and cinder? If nova is going to call initialize_connection...the export either happens in 3pars cases within it's initialize_connection call, or it happens later in the initiailize_connection from the manager19:31
jgriffiththingee: it may not matter any longer.  We used to however have create-export calls outside of intialize_connection19:33
jgriffiththingee: and I think I very clearly said "there's nothing wrong with it" just that it was inconsistent19:33
*** patrickeast_ has quit IRC19:33
jgriffiththingee: let me scroll back and find my comment....19:33
thingeejgriffith: ok... I might be more confused now trying to understand things here.19:33
thingee:)19:33
*** zul has quit IRC19:34
scottdathingee: I think that's a common issue. There's a bunch of different problems. I want to work on documenting some of the use cases and how things *should* work. That might help...19:34
jgriffithbahh... anyway, I said "not saying it's wrong"19:35
jgriffithI'll just keep my opinions to myself19:35
thingeejgriffith: I'm not upset in talking to you here. I'm genuinely trying to under here because I feel ignorant of this topic.19:36
thingeeand you might understand things here because you have opinions19:37
thingeescottda: thanks19:37
*** changbl has joined #openstack-cinder19:37
jgriffiththingee: not at all, I didn't take it that way19:38
thingeejgriffith: well I don't want you to feel you need to keep your opinions to yourself.19:39
jgriffiththingee: I don't think I can be accused of keeping opinions to myself :)  Even when I say I'm going to :)19:39
jgriffiththingee: so here's an example of what I'm looking at:https://github.com/openstack/cinder/blob/master/cinder/volume/manager.py#L108119:40
jgriffithin a pass that returns "None" right... so there's no model_update19:40
jgriffiththingee: it's stuffed in the driver.initialize_connection call19:42
jgriffiththingee: so I'm not sure but I *think* this may be part of why some drivers can't do live-migration19:42
jgriffiththingee: they never update the model info after the create19:42
jgriffiththingee: and IIRC I thought I heard some folks sayign that can't reuse connections/iqn's after a detach19:42
jgriffiththingee: so that would be a problem I think/maybe19:43
scottdajgriffith: Can you put that info somewhere in the etherpad? Maybe under "Random Notes"?19:45
jgriffithscottda: sure19:45
scottdathx19:45
*** e0ne has quit IRC19:47
*** timcl has quit IRC19:50
*** lpetrut has joined #openstack-cinder19:50
*** daneyon has joined #openstack-cinder19:55
thingeejgriffith: I see. I'm unsure what the provider_location has to do with live migration. since that's what you would usually get from the create_export return19:55
*** lpetrut has quit IRC19:55
thingeejgriffith: oh wait19:55
thingeejgriffith: got it reusing19:55
jgriffiththingee: yeah19:55
jgriffiththingee: but I'm not familiar with other devices enough to know if that's the problem or not19:56
jgriffiththingee: but it sounded like it could be19:56
thingeejgriffith: assume it is I suppose19:56
jgriffiththingee: and besides, having methods related to connectivity that just do a "pass" is kinda suspect19:56
jgriffiththingee: might be fine, but def seems like something to look at and make sure it's understood19:57
*** coolsvap is now known as coolsvap|away19:58
*** bpoulos has left #openstack-cinder20:00
*** vilobhmm has joined #openstack-cinder20:08
*** coolsvap|away is now known as coolsvap20:09
*** openstackgerrit has quit IRC20:10
*** openstackgerrit has joined #openstack-cinder20:10
*** e0ne has joined #openstack-cinder20:11
*** kjnelson has joined #openstack-cinder20:11
ericksonsantosvilobhmm, Hi :)20:11
*** e0ne has quit IRC20:12
vilobhmmericksonsantos : reviewing your code20:12
ericksonsantosvilobhmm, I've uploaded a new patch set adding more tests20:12
vilobhmmjust came back from vacation20:12
ericksonsantosvilobhmm, nice20:13
*** ronis_ has quit IRC20:13
vilobhmmthanks for moving the patch forward20:13
vilobhmmmuch appreciated20:13
ericksonsantosvilobhmm, I'll give you time to breathe haha20:13
ericksonsantosvilobhmm, no problem :)20:14
vilobhmmdone…looks ok to me20:14
ericksonsantosvilobhmm, awesome!20:15
*** jungleboyj has joined #openstack-cinder20:15
vilobhmmsmcginnis : If you can have a look at https://review.openstack.org/#/c/194406/ it will be helpful20:15
vilobhmmjungleboyj : ^^20:16
smcginnisvilobhmm: Sure. Tied up with something right now, but have it open in a tab now and will try to get to it soon.20:16
vilobhmmsmcginnis: whenever you are free :) no rush…thanks!20:16
openstackgerritPatrick East proposed openstack/cinder: Add Cinder internal tenant support.  https://review.openstack.org/19935220:16
openstackgerritPatrick East proposed openstack/cinder: WIP Generic image cache.  https://review.openstack.org/19579520:16
jungleboyjvilobhmm: Looking./20:17
vilobhmmjungleboyj : thanks !20:17
*** belmoreira has joined #openstack-cinder20:18
*** merooney has quit IRC20:20
ericksonsantossmcginnis, jungleboyj it is nice to have your eyes on this change \o20:21
openstackgerritAlex Meade proposed openstack/cinder: NetApp E-Series: Add debug tracing  https://review.openstack.org/17058020:21
*** crose has quit IRC20:22
*** dustins has quit IRC20:23
jungleboyjSeems relatively straight forward to me.20:25
*** sgotliv has quit IRC20:26
ericksonsantosjungleboyj, thanks! we will keep things moving!20:29
*** daneyon has quit IRC20:29
*** simondodsley has quit IRC20:30
*** bswartz has quit IRC20:33
*** annasort has quit IRC20:38
*** jasondotstar has joined #openstack-cinder20:40
*** Yogi1 has quit IRC20:41
*** lpetrut has joined #openstack-cinder20:44
*** jgriffith has quit IRC20:45
*** rushil has joined #openstack-cinder20:45
xyangeharney: hi, are you there20:53
eharneyxyang: hi20:53
eharneywhat's up? :)20:53
xyangeharney: so about your comments on LVM:)20:53
eharneyyeeeeeaah20:53
xyangeharney: right now LVM doesn't do attach snapshot in my patch20:53
xyangeharney: I can't do it because it is always local20:54
xyangeharney: so I tested it on my setup, but when I submit the patch, it is still a local snapshot, the path is already there, no need to attach20:54
eharneyhow do we end up needing the if "snapshot" path in create_export then?20:54
xyangeharney: because I tested it that way, so we could use it later20:55
*** Lee1092 has quit IRC20:55
openstackgerritSean Chen proposed openstack/cinder: Tintri snapshot id  https://review.openstack.org/19882920:55
xyangeharney: if we use a remote volume node in the future for backup20:55
*** annegentle has quit IRC20:55
xyangeharney: then it can be used.20:56
eharneyi see20:56
*** akerr_ has quit IRC20:56
xyangeharney: where if you think that will screw things up, maybe we need to remove those?20:57
tbarronxyang: eharney: when we decouple backup service from volume service we'll want to do remote attach even for lvm20:57
xyangeharney: somehow I need to provide a way to do attach snapshot as the default20:57
xyangtbarron: it that is the case, then this will be useful20:57
eharneytbarron: but the other half of my argument was that we could just do a volume clone instead and get the same performance on thin20:57
tbarronxyang: I agree20:57
eharneyisn't the idea to do a clone if attach_snapshot isn't supported?20:58
xyangeharney: yes, but snapshot is quicker20:58
eharneyi'm not sure it is if using thin lvm20:58
eharneythey should be doing the same amount of work20:58
xyangeharney: is it?  I don't know.  usually snapshot is quicker20:58
eharneyfor thick it will be faster, but i've been wanting to move us away from thick for a long time anyway20:59
tbarroneharney: I don't disagree with that.  whether snap or clone is faster likely depends on thin/thick IIUC20:59
eharneywe should just give up on thick at this point now that we have thin prov accounting20:59
eharneybut anyhow20:59
eharneyi'm just mostly not sure that the complexity of more LVM issues (which we have been poking around at for years) is worthwhile21:00
xyangeharney: I'm not sure about LVM thin, but for other drivers may still want to do attach snapshot21:00
eharneysure, and that's fine21:00
*** rushil has quit IRC21:00
xyangeharney: well, clone is always a choice if we don't want to introduce too much complexity into LVM21:01
eharneyi dunno, we can look at it for a bit and see what people think, but i'd lean toward just avoiding it21:02
eharneyor at least avoid it if using thin21:02
xyangeharney: sure, we could change the logic if there are concerns21:03
*** Apoorva has quit IRC21:04
xyangeharney: whether to use snapshot or not can be determined by backup_use_temp_snapshot().  we can do some check there to see if it is thin21:04
eharneythat makes sense other than maybe still causing issues for thick21:07
*** tellesnobrega_ has quit IRC21:07
*** esker has quit IRC21:07
*** jasondotstar has quit IRC21:08
xyangeharney: if you have strong concerns, I can remove it.  Let's see if there are other comments.  We don't need to attach snapshot right now any way as it is all local21:09
*** thangp has quit IRC21:09
eharneysure21:09
*** lpetrut has quit IRC21:12
*** annasort has joined #openstack-cinder21:13
*** hemnafk is now known as hemna21:15
*** IlyaG has joined #openstack-cinder21:18
*** coolsvap is now known as coolsvap|away21:22
*** tellesnobrega_ has joined #openstack-cinder21:23
*** jgriffith has joined #openstack-cinder21:24
*** annegentle has joined #openstack-cinder21:25
*** tellesnobrega_ has quit IRC21:26
haypohemna: "and if python-gdbm is required, then that has to be handled by the patch." hum, i disagree. it's a bug in testrepository21:26
hemnarunning "tox" shouldn't break by your patch.  and it does.21:27
hemnathat has to get fixed or it'll be -1 from me.21:27
*** jgriffith has quit IRC21:27
*** rajinir has joined #openstack-cinder21:27
hemnahaypo, I provided 2 example clean runs against your patch.  one works, the other doesn't.21:28
haypohemna: i discussed with a testrepository developer. they are moving to sqlite21:28
haypohemna: it would solve “db type could not be determined” and “gdbm is missing” errors21:28
*** e0ne has joined #openstack-cinder21:29
*** DericHorn-HP has joined #openstack-cinder21:29
hemnahaypo, whatever you have to do to make that patch work out of the box for folks.   it can't break running tox.21:29
*** fifieldt has quit IRC21:30
*** kjnelson has quit IRC21:31
haypoi checked other projects. heat runs python 3.4 tests using testr (https://github.com/openstack/heat/blob/master/tox.ini#L69), ironic too (https://github.com/openstack/ironic/blob/master/tox.ini#L16)21:33
haypohemna: testr is not widely used to run py34 tests yet because only a few tests pass on python 3.4 in most openstack projects21:33
haypo:-/21:34
*** belmoreira has quit IRC21:34
*** e0ne has quit IRC21:37
openstackgerritVictor Stinner proposed openstack/os-brick: Cleanup Python 3 changes  https://review.openstack.org/19762421:37
haypohemna: ok, i see you problem with "tox". it's fixed by ^^21:37
haypohemna: py34 must be before py27 in the envlist of tox.ini21:39
*** gouthamr has quit IRC21:40
hemnahaypo, ok I'll check it out in a bit.  totally slammed right now.21:41
*** Rockyg has quit IRC21:42
*** fifieldt has joined #openstack-cinder21:44
openstackgerritMitsuhiro Tanino proposed openstack/cinder: Remove unnecessary export after volume migration  https://review.openstack.org/19422321:46
*** jasondotstar has joined #openstack-cinder21:48
*** annegentle has quit IRC21:49
*** annegentle has joined #openstack-cinder21:49
haypothingee, hemna : again, https://review.openstack.org/197624 is not need for a new release of os-brick. it's a just a cleanup ;)21:49
lifelesshaypo: you can use testr easily for py34, just give it the filenames you want to test21:50
lifelesshaypo: tsetr run test_thing test_other test_quad21:51
lifelesshaypo: as long as the other files can be *parsed* it will work21:51
*** IlyaG has quit IRC21:51
haypolifeless: since all tests now pass on python 3.4, i don't want to pass any filename anymore21:51
haypolifeless: just run the full test suite21:51
lifelesshaypo: cool21:51
*** xyang1 has joined #openstack-cinder21:51
haypolifeless: it's the purpose of my change ;)21:51
hemnahaypo, still failing.21:52
hemnahttp://paste.openstack.org/show/356447/21:52
*** vokt has quit IRC21:53
*** jungleboyj has quit IRC21:53
haypohemna: that's because you have to install gdbm21:54
hemnahaypo, that patch needs to account for that.21:55
openstackgerritxing-yang proposed openstack/os-brick: Add connector driver for the ScaleIO cinder driver  https://review.openstack.org/18583521:56
*** xyang has quit IRC21:56
haypohemna: you cannot install gdbm using requirements.txt or test-requirements.txt. it's part of the system, it's a package21:56
haypohemna: well, gdbm is supposed to be part of the python stdlib, but ubuntu chose to split it in a different package21:56
haypolifeless: ^^21:56
*** mriedem has quit IRC21:56
hemna:(!!21:57
hemnathat completely blows.   this is a standalone library that can/will be used outside of openstack, so we can't rely on devstack installing that package just to make tox work :(21:57
hemnahaypo, I presume other projects are going to run into this same problem, once we nuke the tox.ini whitelisted mechanism for running py3421:58
lifelesshemna: yes21:59
hemnaouch21:59
haypogdbm is installed explicitly for testr in devstack: https://github.com/openstack-dev/devstack/blob/master/files/debs/general#L2021:59
lifelesshemna: its because python2 and python3 don't by default support *any* overlapping dbm format files.21:59
haypoi agree that it's annoying to have to install it manually21:59
lifelesshemna: We've been around this multiple times trying to make it Just Work.21:59
hemnadang21:59
haypoi'm using fedora, and it looks like gdbm is part of the base system, so i didn't have your issue21:59
lifelesshemna: what is needed is someone to work up a patch to use a different serialization method for the test timing database.21:59
hemnaI'm going to get pinged any time an os-brick dev runs tox on their box now. :(22:00
lifelesshemna: -or-22:00
*** jasondotstar has quit IRC22:00
lifelesshemna: for OpenStack to stop installing testr inside the tox venv.22:00
lifelesshemna: its not meant to be used like that.22:00
haypohemna: "I presume other projects are going to run into this same problem" yes. and as i wrote, heat & ironic already run python 3.4 tests using testr22:00
lifelesshemna: the problem *only* occurs because its being run with the same Python as the unittests are running under.22:00
lifelesshemna: but IMO testr, like tox, should be system installed.22:00
lifelesswe already tell people to install tox22:01
lifelessthere's no reason we can't tell them to install testr too.22:01
hemnayah we have to system install tox to even run it22:01
haypohemna: you only have two steps to do, but only once: (1) install gdbm (2) remove .testrepository (if you already ran before), and just after that run tox -e py3422:01
lifelessand it would entirely avoid this problem22:01
hemnathat would work for me.22:01
haypoin practice, (2) occurs again if you manually remove .testrepository and then run tox -e py2722:01
hemnahaypo, the steps to make it work are fine and all, but when someone else tries to run tox for the lib, they'll complain it doesn't work (just like I've been), and then hunt me down to find out why.22:02
haypohemna: i know i know... i lost many hours because of testr bugs :)22:02
haypohemna: i don't want to harass you forever with testr issues. i think i will just abandon my change22:02
lifelessso22:02
hemna:(22:02
lifelessI'm going to argue here22:02
lifelesstestr, with the subunit stream that gets indexed22:03
lifelessis part of the main openstack testing infrastructure22:03
hemnahaypo, or can we make a change as lifeless suggests, to require testr as a system install, like tox. ?22:03
lifeless*not* using it is a fairly significant deviation, and cuts you off from the infra statistics gathering22:03
lifelessit was ok as a bootstrap thing for py3422:03
hemnalifeless, +122:03
lifelessI'm obviously biased as the primary author of testr22:04
lifelessbut22:04
*** boris-42 has joined #openstack-cinder22:04
haypoi abandonned https://review.openstack.org/#/c/197624/22:04
lifelessI haven't driven any of the large adoption of it within openstack22:04
*** annasort has quit IRC22:04
haypoi don't want to spend all my nights on this silly pendantic change22:04
hemnaI'd be ok adopting whatever the rest of openstack chooses, as long as it works and doesn't require some hidden knowledge of needing to install python-gdbm.22:05
haypo"I'm obviously biased as the primary author of testr" ah, really? :-D22:05
hemna:)22:05
lifelesshaypo: you didn't know? Or is that sarcasm ?22:05
haypolifeless: i knew it :)22:05
hemnaso using testr outside of the venv seems like the right thing no?22:06
haypolifeless, hemna: feel free to takeover my abandonned my patch, but i just want to spend any more time on it22:06
lifelesshemna: so the rest of openstack just accepts the current limitation and installs python-gdbm everywhere22:06
haypoi have far more important stuff to do ;)22:06
hemnalifeless, heh ok.  I guess I'm just being a trouble make here.  I was just hoping not to break things.22:07
lifelesshemna: I think your lowest investment option is to just take haypo's patch, and add a readme entry ssaying 'to test this repo, install tox and python-gdbm'22:07
lifelesshemna: from there, you can obviously make things better22:07
*** IlyaG has joined #openstack-cinder22:07
lifelesshemna: but at least you won't be an outlier22:07
hemnayah that's good22:07
lifelesshaypo: please unabandon22:08
*** IlyaG has quit IRC22:08
haypolifeless: i'm not going to touch this patch again :-p22:08
hemnaok, so haypo just update the readme and document that as lifeless described.22:08
lifelesshaypo: noone else can unabandond22:08
lifelesshaypo: I'll add the readme bit for hemna but it needs to be unabandoned first22:08
hemnahaypo, please :)22:08
openstackgerritEric Harney proposed openstack/cinder: LVM Thin Provisioning auto-detect  https://review.openstack.org/10465322:08
hemnahaypo, then we can share a beer at the next conference and move on :)22:09
*** IlyaG_ has joined #openstack-cinder22:12
haypowhy did i need to unabandon my patch?22:12
hemnahaypo, so we can just update the readme with the new requirement of installing python-gdbm22:13
hemnathen +A the patch22:13
hemnasince I guess the lib is no worse off than other projects wrt to py34 and needing python-gdbm manually installed.22:14
lifelesshaypo: because gerrit won't let other people unabandon to fix it up for you22:15
lifelesshaypo: if you abandon, they have to start over22:15
lifelesshaypo: if you un-abandon, I can do what hemna is requesting22:15
hemnahaypo, thank you.22:17
haypohemna: i restored my change. you should now be able to take over the change id22:17
openstackgerritEric Harney proposed openstack/cinder: LVM Thin Provisioning auto-detect  https://review.openstack.org/10465322:18
*** MentalRay has quit IRC22:18
*** jgriffit1 has joined #openstack-cinder22:20
*** jgriffit1 has quit IRC22:20
openstackgerritlifeless proposed openstack/os-brick: Cleanup Python 3 changes  https://review.openstack.org/19762422:20
*** annegentle has quit IRC22:21
*** Guest69012 has joined #openstack-cinder22:23
*** IanGovett1 has joined #openstack-cinder22:26
hemnalifeless, thanks for the help.22:27
*** DericHorn-HP has quit IRC22:28
*** IanGovett has quit IRC22:28
*** DericHorn-HP has joined #openstack-cinder22:28
*** eharney has quit IRC22:28
haypohemna: i am a little bit exhausted by python 3. it takes a lot of energy to fix a million of projects and then harass people to get a release22:28
hemnahaypo, yah understandable.22:29
hemnahaypo, thanks a bunch for your help on this though.22:29
hemnasorry for being such a pest about the failures.22:29
hemnaI really appreciate your efforts.22:29
haypofor Routes, it took ~3 months to get my python 3 patch merged. and now i'm waiting a release since ~20 days22:30
haypofor pyeclib, i sent a pull request 3 months ago, and developers replied that they will check it and then nothing. i ping them every 15 days22:31
haypomy patch is dead simple, i don't understand22:31
hemna:(22:31
haypoi already did that a lot of times for many other projects22:31
*** Guest69012 has quit IRC22:32
*** akshai has quit IRC22:33
*** IlyaG_ has quit IRC22:37
*** eharney has joined #openstack-cinder22:42
*** mriedem has joined #openstack-cinder22:46
*** eharney has quit IRC22:47
*** primechuck has quit IRC22:47
*** edmondsw has quit IRC22:50
thingeehemna, haypo, lifeless happy you all came to this conclusion...I was reading through the backlog and about to suggest just updating the readme and be done with it :P22:54
hemna:)22:54
*** boris-42 has quit IRC22:56
*** Adriano__ has quit IRC22:56
*** boris-42 has joined #openstack-cinder22:57
*** DericHorn-HP has quit IRC22:58
*** Adriano__ has joined #openstack-cinder22:59
*** IlyaG has joined #openstack-cinder22:59
*** erlon has quit IRC23:00
openstackgerritWalter A. Boring IV (hemna) proposed openstack/os-brick: WIP: Add new Connector APIs for path validation  https://review.openstack.org/19976423:03
openstackgerritMike Perez proposed openstack/os-brick: Prep for 0.2.1 release  https://review.openstack.org/19976623:06
openstackgerritMike Perez proposed openstack/os-brick: Prep for 0.2.1 release  https://review.openstack.org/19976623:07
*** Adriano__ has quit IRC23:12
*** annashen has quit IRC23:12
*** zhenguo has joined #openstack-cinder23:14
*** jgriffith has joined #openstack-cinder23:18
*** ganso_ has quit IRC23:18
*** IlyaG has quit IRC23:20
*** jwcroppe has quit IRC23:26
*** jwcroppe has joined #openstack-cinder23:27
*** IanGovett1 has quit IRC23:29
*** zul has joined #openstack-cinder23:29
hemnaxyang1,23:29
hemnaxyang1, the ScaleIO brick CI is instantly failing again. :(23:31
*** jwcroppe has quit IRC23:31
*** diogogmt has quit IRC23:38
*** topshare has joined #openstack-cinder23:41
xyang1hemna: I think the way we run the CI is not reliable.  I don't know why it is different from cinder CI, but the way it is set up is different.23:46
hemnaxyang1, ok.  it failed in 5 seconds :)  so it's quick at least23:47
xyang1hemna: Has nothing to do with the code.23:47
xyang1hemna: It has to with cherrypicking, but honestly I don't know how to explain:).  Have to get another run later23:48
xyang1hemna: I'll need to ask asselin what is the trick23:48
hemnaxyang1, ok.  coolio.23:49
hemnagit.openstack.org seems borked at the moment anyway, so on that note...time to head home.23:50
*** hemna is now known as hemnafk23:50
*** jwcroppe has joined #openstack-cinder23:55
*** jungleboyj has joined #openstack-cinder23:59

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