Tuesday, 2018-01-09

*** claudiub has quit IRC00:00
*** r-daneel has quit IRC00:05
*** david-lyle has quit IRC00:07
*** dklyle has joined #openstack-cinder00:07
*** dklyle has quit IRC00:07
*** jmlowe has joined #openstack-cinder00:09
*** wanghao has quit IRC00:15
*** wanghao has joined #openstack-cinder00:17
*** wanghao has quit IRC00:18
*** itlinux has joined #openstack-cinder00:18
*** wanghao has joined #openstack-cinder00:23
*** wanghao has quit IRC00:28
*** sdague has quit IRC00:29
*** jmlowe has quit IRC00:31
*** edmondsw has joined #openstack-cinder00:36
*** wanghao has joined #openstack-cinder00:36
*** jmlowe has joined #openstack-cinder00:38
*** moshele has quit IRC00:42
*** felipemonteiro_ has joined #openstack-cinder00:44
*** Dinesh_Bhor has joined #openstack-cinder00:57
openstackgerritJohn Griffith proposed openstack/cinder master: Add multiattach policy  https://review.openstack.org/53156901:00
openstackgerritwanghao proposed openstack/python-cinderclient master: Support for reporting backend state in service list  https://review.openstack.org/53044601:04
*** zhaochao has joined #openstack-cinder01:09
*** felipemonteiro_ has quit IRC01:12
*** moshele has joined #openstack-cinder01:13
*** markvoelker has joined #openstack-cinder01:17
*** zhurong has joined #openstack-cinder01:21
*** hoangcx has joined #openstack-cinder01:23
*** armax has joined #openstack-cinder01:25
*** mriedem1 has joined #openstack-cinder01:40
*** mriedem has quit IRC01:42
*** liverpooler has joined #openstack-cinder01:43
*** Dinesh_Bhor has quit IRC01:45
openstackgerritzheng yin proposed openstack/python-cinderclient master: fix offset  https://review.openstack.org/53194601:45
*** liverpooler has quit IRC01:46
*** markstur_ has joined #openstack-cinder01:51
*** jmlowe has quit IRC01:52
*** zhongjun has joined #openstack-cinder01:52
*** markstur has quit IRC01:54
*** mriedem has joined #openstack-cinder01:56
*** mriedem1 has quit IRC01:58
*** felipemonteiro_ has joined #openstack-cinder02:02
*** Dinesh_Bhor has joined #openstack-cinder02:05
openstackgerritzheng yin proposed openstack/python-cinderclient master: add snapshot_id param note  https://review.openstack.org/48993102:05
*** Dinesh_Bhor has quit IRC02:07
*** jmlowe has joined #openstack-cinder02:07
*** dalgaaf has quit IRC02:12
*** dalgaaf has joined #openstack-cinder02:12
*** namnh has joined #openstack-cinder02:17
*** vivsoni has quit IRC02:19
*** vivsoni has joined #openstack-cinder02:19
*** jmlowe has quit IRC02:23
openstackgerritzheng yin proposed openstack/cinder master: backup/ceph: remove redundant pool  https://review.openstack.org/48994902:24
*** Dinesh_Bhor has joined #openstack-cinder02:26
openstackgerritzheng yin proposed openstack/cinder master: Ceph backup: eventlet.green.subprocess replace subprocess  https://review.openstack.org/46660702:29
*** edmondsw has quit IRC02:30
*** Apoorva has quit IRC02:32
*** harlowja has quit IRC02:32
*** armaan has quit IRC02:33
*** Dinesh_Bhor has quit IRC02:33
*** moshele has quit IRC02:34
*** jmlowe has joined #openstack-cinder02:37
*** jmlowe has quit IRC02:53
*** nicolasbock has quit IRC02:58
*** bkopilov has quit IRC03:04
*** markvoelker has quit IRC03:06
*** hoonetorg has quit IRC03:12
openstackgerritzheng yin proposed openstack/python-cinderclient master: Modify offset judgment conditions  https://review.openstack.org/53194603:12
*** Srinivas has joined #openstack-cinder03:13
SrinivasHi, I am facing issue while installing devstack in Openstack CI Jenkins. I am getting this error in devstack.log Policy doesn't allow volume_extension:types_manage to be performed. (HTTP 403) (Request-ID: req-0aaedf11-5696-4bcc-a5d5-90762d921dc4)03:14
SrinivasAnyone is facing same issue. Thanks03:14
*** jmlowe has joined #openstack-cinder03:16
openstackgerritzheng yin proposed openstack/python-cinderclient master: Add snapshot_id param note for backup-create  https://review.openstack.org/48993103:25
*** hoonetorg has joined #openstack-cinder03:25
openstackgerritYu Zhang proposed openstack/cinder master: Fix: validation error if the optional parameter's value is None  https://review.openstack.org/53128903:26
tommylikehu    hey Srinivas , when did you install your devstack environment?03:29
*** jmlowe has quit IRC03:39
Srinivastommylikehu: In jenkins devstack-gate is installing the devstack03:39
Srinivastommylikehu: i am not installing devstack separatly03:39
Srinivastommylikehu: From last 3 hours it is failing03:40
tommylikehuSrinivas: ok, AFAIK nothing related to this operation should have changed. it's always required as a admin context.https://github.com/openstack/cinder/blob/master/cinder/policies/volume_type.py#L2903:42
*** jmlowe has joined #openstack-cinder03:44
*** abhishekk has joined #openstack-cinder03:44
Srinivastommylikehu: i am using project:demo and user as admin in for creating nodepool vms03:47
*** zhurong has quit IRC03:55
*** Dinesh_Bhor has joined #openstack-cinder04:01
*** monjoshi has joined #openstack-cinder04:02
*** ykarel has joined #openstack-cinder04:14
*** Apoorva has joined #openstack-cinder04:15
*** bkopilov has joined #openstack-cinder04:23
*** yangyapeng has quit IRC04:31
*** yangyapeng has joined #openstack-cinder04:31
*** felipemonteiro_ has quit IRC04:34
*** chhavi has joined #openstack-cinder04:37
*** mriedem has quit IRC04:47
*** Dinesh_Bhor has quit IRC04:50
*** Dinesh_Bhor has joined #openstack-cinder04:51
*** Dinesh_Bhor has quit IRC04:52
*** Dinesh_Bhor has joined #openstack-cinder04:55
*** markvoelker has joined #openstack-cinder04:56
*** felipemonteiro_ has joined #openstack-cinder05:01
*** psachin has joined #openstack-cinder05:04
*** monjoshi has quit IRC05:07
*** links has joined #openstack-cinder05:07
*** monjoshi has joined #openstack-cinder05:07
*** Srinivas has quit IRC05:10
*** arnewiebalck_ has quit IRC05:24
*** markstur_ has quit IRC05:29
*** markstur has joined #openstack-cinder05:29
*** claudiub has joined #openstack-cinder05:29
*** markstur has quit IRC05:29
*** Apoorva has quit IRC05:49
*** markstur has joined #openstack-cinder05:50
*** hoonetorg has quit IRC05:55
*** markstur has quit IRC05:55
*** hoonetorg has joined #openstack-cinder05:59
*** sandanar has joined #openstack-cinder06:03
*** Dinesh_Bhor has quit IRC06:07
*** felipemonteiro_ has quit IRC06:08
*** Dinesh_Bhor has joined #openstack-cinder06:10
*** wanghao_ has joined #openstack-cinder06:12
*** wanghao has quit IRC06:15
*** gouthamr has quit IRC06:17
openstackgerritJesse Wu proposed openstack/cinder master: Synology: Rename iSCSI target config options  https://review.openstack.org/53202106:22
openstackgerritJesse Wu proposed openstack/cinder master: Synology: Rename iSCSI target config options  https://review.openstack.org/53202106:25
jiaopengjuFailed to list volumes in latest devstack environment. https://bugs.launchpad.net/cinder/+bug/174205206:29
openstackLaunchpad bug 1742052 in Cinder "Failed to list volumes in latest devstack environment" [Undecided,New]06:29
*** markstur has joined #openstack-cinder06:32
*** links has quit IRC06:35
*** moshele has joined #openstack-cinder06:37
*** markstur has quit IRC06:37
*** rcernin has quit IRC06:38
*** niraj_singh has joined #openstack-cinder06:39
*** gkadam has quit IRC06:43
*** zhurong has joined #openstack-cinder06:50
*** links has joined #openstack-cinder06:52
*** markstur has joined #openstack-cinder06:53
*** hoonetorg has quit IRC06:56
*** pcaruana has joined #openstack-cinder06:56
*** markstur has quit IRC06:57
*** gcb has joined #openstack-cinder06:58
*** links has quit IRC07:03
*** caisan_ has joined #openstack-cinder07:04
*** pgadiya has joined #openstack-cinder07:04
*** Dinesh_Bhor has quit IRC07:05
*** lkuchlan has joined #openstack-cinder07:06
*** armaan has joined #openstack-cinder07:10
*** caisan_ has quit IRC07:11
*** caisan has joined #openstack-cinder07:12
*** hoonetorg has joined #openstack-cinder07:12
*** markstur has joined #openstack-cinder07:14
*** links has joined #openstack-cinder07:14
*** Dinesh_Bhor has joined #openstack-cinder07:15
*** markstur has quit IRC07:18
*** Dinesh_Bhor has quit IRC07:27
*** felipemonteiro_ has joined #openstack-cinder07:27
*** links has quit IRC07:27
*** hoonetorg has quit IRC07:32
*** felipemonteiro_ has quit IRC07:39
openstackgerritChhavi Agarwal proposed openstack/cinder master: Run backup-restore operations on native thread  https://review.openstack.org/51831607:43
*** wanghao_ has quit IRC07:43
*** wanghao has joined #openstack-cinder07:44
*** e0ne has joined #openstack-cinder07:44
*** links has joined #openstack-cinder07:45
*** Dinesh_Bhor has joined #openstack-cinder07:48
*** rhagarty has quit IRC07:49
*** rhagarty has joined #openstack-cinder07:52
*** karmine has quit IRC07:55
*** AlexeyAbashkin has joined #openstack-cinder07:56
*** itlinux has quit IRC07:58
*** caisan_ has joined #openstack-cinder08:00
*** hoonetorg has joined #openstack-cinder08:01
*** caisan has quit IRC08:03
*** armax has quit IRC08:08
*** lkuchlan has quit IRC08:12
*** lkuchlan has joined #openstack-cinder08:12
*** gkadam has joined #openstack-cinder08:13
*** tesseract has joined #openstack-cinder08:20
*** alexchadin has joined #openstack-cinder08:22
*** markstur has joined #openstack-cinder08:26
*** chhavi__ has joined #openstack-cinder08:29
*** chhavi has quit IRC08:29
*** psachin has quit IRC08:30
*** caisan has joined #openstack-cinder08:30
*** markstur has quit IRC08:30
openstackgerritPeter Wang proposed openstack/cinder master: Unity: Add revert_to_snapshot support  https://review.openstack.org/53207108:31
*** caisan_ has quit IRC08:33
*** caisan has quit IRC08:33
*** caisan has joined #openstack-cinder08:34
*** caisan has quit IRC08:34
*** caisan has joined #openstack-cinder08:35
openstackgerritPeter Wang proposed openstack/cinder master: Unity: Add revert_to_snapshot support  https://review.openstack.org/53207108:35
*** daidv has quit IRC08:38
*** namnh has quit IRC08:38
*** hoangcx has quit IRC08:38
*** namnh has joined #openstack-cinder08:38
*** hoangcx has joined #openstack-cinder08:39
*** daidv has joined #openstack-cinder08:39
*** psachin has joined #openstack-cinder08:39
openstackgerritYu Zhang proposed openstack/cinder master: Fix: validation error if the optional parameter's value is None  https://review.openstack.org/53128908:48
*** Dinesh_Bhor has quit IRC08:49
*** markstur has joined #openstack-cinder08:52
*** markstur has quit IRC08:57
*** monjoshi has quit IRC09:03
*** monjoshi has joined #openstack-cinder09:03
*** caisan_ has joined #openstack-cinder09:06
*** Dinesh_Bhor has joined #openstack-cinder09:08
*** caisan has quit IRC09:09
openstackgerritlei zhang proposed openstack/python-cinderclient master: Fix the incorrect git.openstack.org source URL  https://review.openstack.org/53209909:10
*** markstur has joined #openstack-cinder09:13
*** monjoshi has quit IRC09:17
*** monjoshi has joined #openstack-cinder09:17
*** markstur has quit IRC09:18
*** wanghao has quit IRC09:25
*** wanghao has joined #openstack-cinder09:26
*** wanghao has quit IRC09:26
*** wanghao has joined #openstack-cinder09:26
*** wanghao has quit IRC09:26
*** wanghao has joined #openstack-cinder09:27
*** monjoshi has quit IRC09:27
*** monjoshi has joined #openstack-cinder09:27
*** wanghao has quit IRC09:27
*** wanghao has joined #openstack-cinder09:27
*** wanghao has quit IRC09:28
*** wanghao has joined #openstack-cinder09:28
*** wanghao has quit IRC09:29
*** wanghao has joined #openstack-cinder09:29
*** wanghao has quit IRC09:29
*** wanghao has joined #openstack-cinder09:30
*** wanghao has quit IRC09:30
*** wanghao has joined #openstack-cinder09:31
*** wanghao has quit IRC09:31
*** wanghao has joined #openstack-cinder09:31
*** caisan has joined #openstack-cinder09:32
*** wanghao has quit IRC09:32
*** hoonetorg has quit IRC09:32
*** wanghao has joined #openstack-cinder09:32
*** wanghao has quit IRC09:33
*** wanghao has joined #openstack-cinder09:33
*** jistr|off|mtg is now known as jistr09:36
*** caisan_ has quit IRC09:36
*** Dinesh_Bhor has quit IRC09:38
*** armaan has quit IRC09:46
*** armaan has joined #openstack-cinder09:47
*** chhavi__ has quit IRC09:48
*** chhavi__ has joined #openstack-cinder09:49
*** hoonetorg has joined #openstack-cinder09:49
*** erlon has joined #openstack-cinder09:54
*** stakeda has quit IRC09:59
*** lpetrut has joined #openstack-cinder10:01
*** caisan_ has joined #openstack-cinder10:06
*** armaan has quit IRC10:07
*** armaan has joined #openstack-cinder10:07
*** caisan has quit IRC10:09
*** mvk has joined #openstack-cinder10:20
*** caisan_ has quit IRC10:24
*** sdague has joined #openstack-cinder10:40
*** mayur_ind has joined #openstack-cinder10:40
*** mayur_ind has left #openstack-cinder10:41
*** hoonetorg has quit IRC10:46
*** hoonetorg has joined #openstack-cinder10:48
*** hemna_ has quit IRC10:48
openstackgerritCiara Stacke proposed openstack/cinder master: VMAX driver - Multiple storage group creation attempts  https://review.openstack.org/53215910:48
*** namnh has quit IRC10:48
*** abhishekk has quit IRC10:51
*** kholkina has joined #openstack-cinder10:52
*** alexchadin has quit IRC10:53
*** chhavi__ has quit IRC10:57
*** alexchadin has joined #openstack-cinder10:59
*** AlexeyAbashkin has quit IRC11:02
*** AlexeyAbashkin has joined #openstack-cinder11:02
*** monjoshi has quit IRC11:02
*** hoonetorg has quit IRC11:04
tommylikehu  hey smcginnis , jungleboyj  is it possible to keep track of the bootable attribute when upload image from volume and then create new one from image now?11:11
*** moshele has quit IRC11:19
*** moshele has joined #openstack-cinder11:21
*** bkopilov has quit IRC11:36
*** kwathore has joined #openstack-cinder11:43
*** mvk has quit IRC11:45
*** alexchadin has quit IRC12:14
*** alexchadin has joined #openstack-cinder12:14
*** nicolasbock has joined #openstack-cinder12:21
*** takedakn has joined #openstack-cinder12:23
*** moshele has quit IRC12:29
*** pgadiya has quit IRC12:32
*** hoonetorg has joined #openstack-cinder12:34
*** pgadiya has joined #openstack-cinder12:37
*** mvk has joined #openstack-cinder12:47
*** zhurong_ has joined #openstack-cinder12:56
openstackgerritKushal Wathore proposed openstack/cinder master: 3PAR: fix create group from source functionality  https://review.openstack.org/53217812:57
*** lkuchlan has quit IRC13:00
*** hoonetorg has quit IRC13:03
openstackgerritIlya Shakhat proposed openstack/cinder master: Initialize osprofiler in WSGI application  https://review.openstack.org/53218013:06
erlongeguileo: ping13:07
*** zhurong_ has quit IRC13:07
geguileoerlon: pong!13:07
*** zhurong has quit IRC13:07
erlongeguileo: hey that was fast13:07
geguileoerlon: yup, I'm back from PTO  :-)13:07
erlongeguileo: hmm, good, fresh head!13:08
erlongeguileo: Im working on that oveprovisioning improvements13:08
geguileoerlon: not really, I'm drowning in emails... XD13:08
geguileonice!13:08
erlongeguileo: there's  a bug happening where the scheduler allows the user to create as many thin volumes as he can within the 60seconds window13:09
*** abishop has joined #openstack-cinder13:10
erlongeguileo: is there a way to get around that? first thing that came to my mind was why the host updates are send via network and not only stored in the DB. Any particular reason or just historical?13:11
geguileoerlon: Is that something new?13:11
geguileoOr was it present in older releases?13:11
geguileoBecause it doesn't make sense to me, even if you have multiple schedulers13:11
erlongeguileo: that is happening in liberty, but I believe that this Cinder's behaviour does not have changed over the newer releases13:12
geguileothe most you would be able to do is N * max over provisioning13:12
geguileoWell, in my experience you cannot create as many as you want, the limit is known13:12
geguileoBut I could be wrong....13:12
geguileoAnd the 60 seconds would depend on the stats reporting rate configured in Cinder13:13
erlongeguileo: yes, but the thing is, the volume service reports the allocated_capacity_gb, or provisioned_capacity_gb only from minute to minute, the scheduler makes the calculation and filtering based on this stat, which is only updated at the regular basis (60sec default)13:14
*** _ix has joined #openstack-cinder13:14
geguileoerlon: no, it is also updated when you create a volume (but only on the node that does the scheduling)13:15
*** dave-mccowan has joined #openstack-cinder13:15
erlongeguileo: so, in that 60sec window the scheduler will not know that the allocated_capacity_gb/provisioned_capacity_gb has changed13:15
geguileoerlon: that's why I say it's not unlimited13:15
geguileoerlon: the node that does the scheduling knows this fact, the other schedulers are the ones that don't13:15
erlongeguileo: hmm, the scheduler does the update in its hosts fields right?13:16
*** bkopilov has joined #openstack-cinder13:17
geguileoerlon: in the stats table it keeps in memory13:17
erlongeguileo: that makes sense, I just need to know now if that is not implement in liberty and where that was added13:17
*** ykarel is now known as ykarel|away13:17
geguileoerlon: I don't know when it was added, I know it's there  ;-)13:18
erlongeguileo: lol that its not hard to find, thanks13:18
geguileonp13:19
*** dave-mccowan has quit IRC13:21
*** edmondsw has joined #openstack-cinder13:21
*** ykarel|away has quit IRC13:22
*** alexchadin has quit IRC13:23
*** alexchadin has joined #openstack-cinder13:24
*** dave-mccowan has joined #openstack-cinder13:26
*** yangyapeng has quit IRC13:27
*** _ix has quit IRC13:38
*** mchlumsky has joined #openstack-cinder13:41
*** markstur has joined #openstack-cinder13:42
*** markstur has quit IRC13:47
*** crushil has joined #openstack-cinder13:49
*** AlexeyAbashkin has quit IRC13:52
*** mriedem has joined #openstack-cinder13:54
*** yangyapeng has joined #openstack-cinder13:54
*** gman-tx has joined #openstack-cinder13:57
*** lhx_ has joined #openstack-cinder13:58
*** chhavi__ has joined #openstack-cinder13:58
*** takedakn has quit IRC13:58
*** AlexeyAbashkin has joined #openstack-cinder14:00
*** ykarel|away has joined #openstack-cinder14:00
*** ykarel|away is now known as ykarel14:01
*** markstur has joined #openstack-cinder14:03
*** alexchadin has quit IRC14:06
*** kholkina has quit IRC14:06
openstackgerritKushal Wathore proposed openstack/cinder master: 3PAR: fix create group from source functionality  https://review.openstack.org/53217814:07
*** markstur has quit IRC14:07
*** dustins has joined #openstack-cinder14:10
*** alexchadin has joined #openstack-cinder14:10
*** pgadiya has quit IRC14:12
*** moshele has joined #openstack-cinder14:14
kwathore@Team, This is very critical bug which we have observerd releated to multipath: https://bugs.launchpad.net/nova/+bug/172392814:16
openstackLaunchpad bug 1723928 in os-brick "In case of volume_use_multipath=True, Nova unable to fetch CONF.libvirt.volume_use_multipath value from nova.conf" [Undecided,New]14:16
kwathorewe have analyzed from our end and given input to it14:17
mriedemi sent kwathore here because i don't have an env to recreate or experience with multipath, so was hoping someone else working on cinder can confirm/recreate14:18
*** jmlowe has quit IRC14:19
*** abhishekk has joined #openstack-cinder14:20
kwathoreThis is critical as all drivers faces this issue of multipath14:20
kwathorethanks @eone and meriedem14:20
*** lkuchlan has joined #openstack-cinder14:20
jungleboyjkwathore:  Thanks for forwarding this along.  I am not sure that I have the right env.14:23
jungleboyjhemna:  ^^14:23
jungleboyjmriedem:  We will see what we can do.  Thanks.14:23
*** markstur has joined #openstack-cinder14:24
*** moshele has quit IRC14:24
kwathore@jungleboyj: Thanks,14:27
*** psachin has quit IRC14:28
*** markstur has quit IRC14:28
*** jmlowe has joined #openstack-cinder14:30
*** moshele has joined #openstack-cinder14:34
*** jmlowe has quit IRC14:34
*** jmlowe has joined #openstack-cinder14:35
*** jmlowe has quit IRC14:36
*** felipemonteiro_ has joined #openstack-cinder14:36
*** moshele has quit IRC14:38
*** markvoelker has quit IRC14:40
*** gman-tx has quit IRC14:41
*** eharney has joined #openstack-cinder14:41
*** markstur has joined #openstack-cinder14:45
*** kaisers has quit IRC14:45
*** markstur has quit IRC14:49
*** kaisers has joined #openstack-cinder14:51
*** alexchadin has quit IRC14:56
*** gouthamr has joined #openstack-cinder15:00
*** liverpooler has joined #openstack-cinder15:01
*** nicolasbock has quit IRC15:02
*** esker has joined #openstack-cinder15:09
*** gman-tx has joined #openstack-cinder15:10
*** markvoelker has joined #openstack-cinder15:13
*** wanghao_ has joined #openstack-cinder15:15
*** markvoelker has quit IRC15:16
*** hoonetorg has joined #openstack-cinder15:16
*** wanghao has quit IRC15:17
*** jmlowe has joined #openstack-cinder15:20
*** jmlowe has quit IRC15:21
*** abhishekk has quit IRC15:21
*** jmlowe has joined #openstack-cinder15:21
*** armax has joined #openstack-cinder15:21
*** jmlowe has quit IRC15:22
*** jmlowe has joined #openstack-cinder15:22
*** markvoelker has joined #openstack-cinder15:25
openstackgerritZhao Chao proposed openstack/python-cinderclient master: Migrate to keystoneauth identity cli opts.  https://review.openstack.org/53221715:28
*** crose has joined #openstack-cinder15:29
*** Apoorva has joined #openstack-cinder15:33
*** links has quit IRC15:36
*** esker has quit IRC15:39
*** hoonetorg has quit IRC15:41
*** sar has joined #openstack-cinder15:42
*** david-lyle has joined #openstack-cinder15:44
*** felipemonteiro__ has joined #openstack-cinder15:45
*** felipemonteiro_ has quit IRC15:49
*** hoonetorg has joined #openstack-cinder15:55
*** markstur has joined #openstack-cinder15:55
*** felipemonteiro__ has quit IRC16:03
*** felipemonteiro__ has joined #openstack-cinder16:03
*** sandanar has quit IRC16:03
*** markvoelker has quit IRC16:09
*** stvnoyes has joined #openstack-cinder16:11
openstackgerritSteve Noyes proposed openstack/cinder master: Improve multiattach error message  https://review.openstack.org/53187716:15
jgriffithkwathore: jungleboyj did ya'll check out the notes in there from William?16:17
jgriffithmight have the answer for you in there16:17
*** lpetrut has quit IRC16:18
*** itlinux has joined #openstack-cinder16:19
smcginnisjgriffith, mriedem: Based on those last couple comments, looks like an error in Nova.16:19
jgriffithsmcginnis: seems like it; not picking up the config setting somehow?16:19
jgriffithkwathore: https://github.com/openstack/nova/blob/master/nova/virt/libvirt/volume/iscsi.py#L3916:22
jbernardjgriffith: wasn't there a replication-status flag to one of the cinder-manage commands at some point?16:22
jbernardjgriffith: am i misremembering that?16:22
jgriffithjbernard: :)16:22
jgriffithjbernard: `cinder service-list --withreplication`16:22
jbernardyou're about to say you just removed it aren't you16:22
jgriffithhaha, no16:22
jgriffithbut I suspect you're fighting the microversion battle16:22
jbernardahh, not cinder-manage16:22
jbernardok, i was looking in the wrong place entirely16:23
jgriffithwhich I do on a daily basis16:23
jgriffithOH!  haha, that will do it16:23
jbernardim fighting many battles at the moment ;)16:23
jgriffithjbernard: sadly I know the feeling16:23
jgriffithjbernard: try service-list with the latest mv and see if that's what you were looking for16:24
jgriffithif not gimmie a shout16:24
jbernardok, will do; thanks16:24
*** esker has joined #openstack-cinder16:33
openstackgerritZhao Chao proposed openstack/python-cinderclient master: Migrate to keystoneauth identity cli opts.  https://review.openstack.org/53221716:33
*** lhx_ has quit IRC16:35
*** e0ne has quit IRC16:35
*** markvoelker has joined #openstack-cinder16:37
*** sar has quit IRC16:40
*** lkuchlan has quit IRC16:42
*** markstur_ has joined #openstack-cinder16:42
*** felipemonteiro_ has joined #openstack-cinder16:45
*** markstur has quit IRC16:46
*** crose has quit IRC16:46
*** felipemonteiro__ has quit IRC16:49
mriedemkwathore: the comments in the nova bug don't make sense,16:53
mriedemi.e. if the config option is set then nova has to be using it16:53
mriedemkwathore: you attached some n-cpu logs, but i need to see the start of the n-cpu service startup logs where it dumps the config options it has loaded into memory16:53
*** itlinux has quit IRC16:56
*** AlexeyAbashkin has quit IRC17:02
*** vivsoni__ has joined #openstack-cinder17:05
*** armax has quit IRC17:05
*** mvk has quit IRC17:06
*** sar has joined #openstack-cinder17:07
mriedemkwathore: which version of os-brick did this work with in ocata, and which version of os-brick does it not work with in pike?17:07
openstackgerritChhavi Agarwal proposed openstack/cinder master: Run backup-restore operations on native thread  https://review.openstack.org/51831617:09
*** felipemonteiro_ has quit IRC17:13
*** chhavi__ has quit IRC17:14
*** ykarel has quit IRC17:16
*** elmaciej has joined #openstack-cinder17:18
*** itlinux has joined #openstack-cinder17:18
*** moshele has joined #openstack-cinder17:18
vivsoni__mriedem: i have to re-create ocata devstack env to check os-brick :)17:20
mriedemvivsoni__: if you're using devstack, then it would be this version https://github.com/openstack/requirements/blob/stable/ocata/upper-constraints.txt#L22417:22
mriedem1.11.017:22
mriedemvs os-brick===1.15.4 in pike17:22
*** esker has quit IRC17:22
*** monjoshi has joined #openstack-cinder17:22
*** david-lyle has quit IRC17:23
vivsoni__yes mriedem. you are right17:23
*** esker has joined #openstack-cinder17:23
*** david-lyle has joined #openstack-cinder17:26
mriedemhere is a simple test to see the option is used https://review.openstack.org/53227017:27
*** esker has quit IRC17:28
*** esker has joined #openstack-cinder17:28
*** Apoorva has quit IRC17:31
*** monjoshi has quit IRC17:32
*** monjoshi has joined #openstack-cinder17:32
vivsoni__mriedem: you mean to say. its working properly or you have created unit test case to check whether its working or not ?17:33
mriedemi wrote a unit test to show that when changing the config option either way, it's used and the resulting host connector dict has the appropriate 'multipath' entry17:33
vivsoni__mriedem: it somehow in real scenario its not populating from nova.conf17:37
mriedemthat doesn't make sense,17:37
mriedemand as i said in the bug report, i need the config option dump from when nova-compute starts17:37
mriedemwhich is in the logs17:37
vivsoni__hmm17:38
*** elmaciej has quit IRC17:39
*** moshele has quit IRC17:40
mriedemvivsoni__: do you have something wrong with your env setup? like multiple nova configs getting loaded or something?17:42
mriedemi don't see how only your team can be hitting this but no one else has ever reported something that would be so blatantly obviously broken17:42
vivsoni__mriedem: i have checked my  env17:43
vivsoni__mriedem: i will provide you nova.config file as well as nova logs17:44
vivsoni__which has those nova.conf parsing details17:44
mriedemhere is another patch showing the iscsi/fc volume connectors working https://review.openstack.org/53227217:44
*** markvoelker has quit IRC17:47
*** gkadam has quit IRC17:49
vivsoni__Hi Sean & Team, please review https://review.openstack.org/#/c/525719/17:54
*** tesseract has quit IRC17:57
vivsoni__Thanks Sean :)17:57
*** edmondsw has quit IRC17:58
*** edmondsw has joined #openstack-cinder18:01
*** Apoorva has joined #openstack-cinder18:02
*** Apoorva has quit IRC18:02
*** rmcall has joined #openstack-cinder18:05
*** edmondsw has quit IRC18:06
*** david-lyle has quit IRC18:06
*** AlexeyAbashkin has joined #openstack-cinder18:26
openstackgerritJohn Griffith proposed openstack/cinder master: Enable multiattach capability  https://review.openstack.org/53228018:27
jgriffithsmcginnis: ildikov mriedem johnthetubaguy so thinking about updating/modifying this slightly:  https://specs.openstack.org/openstack/cinder-specs/specs/queens/enable-multiattach.html18:29
jgriffithin addition to some of the policy things we talked about already; I'm thinking it might be wise to make the default on secondary/tertiary attachments R/O by default18:30
jgriffithand only allow additional R/W attachments if the type specifies mod:rw explicitly... thoughts?18:30
*** AlexeyAbashkin has quit IRC18:30
jgriffiths/mod/mode/18:31
jgriffithotherwise, the series is complete enough to run this and watch your volumes go readonly due to fs errors if anyone is interested:  https://review.openstack.org/#/q/topic:bp/multi-attach-v3-attach+(status:open+OR+status:merged)18:31
*** Apoorva has joined #openstack-cinder18:32
ildikovjgriffith: fs errors?18:36
*** armax has joined #openstack-cinder18:36
jgriffithildikov: fs=='file system'18:36
*** crushil has left #openstack-cinder18:36
jgriffithildikov: it's nothing we didn't already know/talk about18:36
ildikovjgriffith: I know :)18:36
jgriffithjust wondering if there's a better setting for defaults18:37
ildikovjgriffith: I'm sensitive to the word 'error'18:37
jgriffithOh.. "you know", sorry then I'm not sure what the question is?18:37
jgriffithhaha18:37
jgriffithyeah, well that's what you get when trying to do multiattach with ext4 :)18:37
*** itlinux has quit IRC18:38
ildikovjgriffith: yeah, I know that too :)18:38
jgriffithildikov: haha, ok so then what's the question?18:39
* jgriffith is all out of guesses :)18:39
*** monjoshi has quit IRC18:42
jgriffithstvnoyes: let me know what you think here:  https://review.openstack.org/#/c/531877/318:42
*** itlinux has joined #openstack-cinder18:42
jgriffithildikov: hmmm... does that mean you're not going to tell me :)18:42
*** monjoshi has joined #openstack-cinder18:42
stvnoyesjgriffith, I'm not crazy about the new wording. "is not of a type' is somewhat clunky18:44
jgriffithstvnoyes: fair enough... sold!18:44
jgriffithHave a +A18:44
stvnoyesgreat thanks!18:44
jgriffiththank you!18:44
jgriffithDOH18:44
jgriffithpep8 just failed :(18:45
*** gouthamr has quit IRC18:46
*** itlinux has quit IRC18:46
*** gouthamr has joined #openstack-cinder18:48
smcginnisjgriffith: I like the idea of defaulting to ro. I missed part of the discussion in Denver, but I thought that was one of the things proposed at the time.18:49
ildikovjgriffith: no question, I just couldn't check the log earlier and as *everything* is failing all the time I'm in problem solving mode by default, so if you say 'something error' I will ask back :)18:49
*** itlinux has joined #openstack-cinder18:49
jgriffithildikov: LOL18:49
ildikovjgriffith: just to see if it's something new or something that we know/expected :)18:49
jgriffithildikov: so FWIW I just load the cinder patches, don't even need/use the Nova patches in this case18:50
*** moshele has joined #openstack-cinder18:50
jgriffithwhich of course means the caching stuff is not turned off; but that's sort of irrelevant for what I'm looking at18:50
jgriffithsmcginnis: yeah, honestly we've talked about sooooo much about different things I can't keep it straight :)18:50
jgriffithsmcginnis: quite possible that I was supposed to do this in the first place but my feable old mind got confused :)18:51
ildikovjgriffith: well, the idea is to have Cinder function nicely on its own so whomever is using it without Nova for multi-attach, have fun :)18:51
*** felipemonteiro has joined #openstack-cinder18:51
smcginnisjgriffith: Nah, it was probably just the last thing I heard mentioned before I left the room.18:51
ildikovjgriffith: I'm kinda fine with the 'RO' idea18:51
jgriffithildikov: "kinda" ?18:51
ildikovjgriffith: I'm not sure at this point what people expect as 'default' scenario18:51
smcginnisjgriffith: But totally agree there's been so much talked about that it's hard to keep track some times what was actually decided.18:51
ildikovjgriffith: is the 'kinda' part18:52
ildikovjgriffith: but having the additional attachments as 'RO' sounds safer18:52
smcginnisThat's what the specs are for, and we all agreed to the spec that did not state this, so I think what we have right now is what we've officially agreed to.18:52
smcginnisThat said, I wouldn't mind changing it to be "safe"ish.18:52
jgriffithwell, I'll be offline for a few hours so people can mull it over and send me nasty DM's or objections :)18:52
*** itlinux has quit IRC18:52
jgriffithsmcginnis: I'm with you18:52
smcginnisMore saferer.18:53
jgriffithIt doesn't change the ability to do things, just picks what I think is "more saner" (or more safer)18:53
jgriffith:)18:53
ildikovsmcginnis: we were always unsure of how 'RO' is set and how it can or can't be changed so we said to go for the easiest route and have all 'RW' and the ability to turn this off completely with policy18:53
jgriffithsanerer18:53
jgriffithAhh... wait; now I remember why I did it that way18:53
smcginnisildikov: Ah, that could be why that didn't happen.18:53
jgriffithit requires the nova changes18:53
smcginnisFair enough.18:53
*** itlinux has joined #openstack-cinder18:54
ildikovsmcginnis: the only thing I'm actually good at is remembering some of the crazy shit we talk about...18:54
jgriffithie has to be set/enforced on the hypervisor for some backends18:54
jgriffithildikov: bahh, that's not true18:54
smcginnis:)18:55
* jgriffith resists temptation to make a bad joke18:55
*** itlinux has quit IRC18:55
*** moshele has quit IRC18:55
ildikovjgriffith: now you made me curious :)18:55
jgriffithildikov: LOL18:56
smcginnisjgriffith: Weren't you heading out the door?18:56
smcginnis*quick*18:56
jgriffithI was going to make a smart ass zinger but it wasn't so far from the truth that I decided not to say it18:56
jgriffithsmcginnis: that's right18:56
jgriffithcya18:56
jgriffith:)18:56
*** jmlowe has quit IRC18:56
*** gkadam has joined #openstack-cinder18:57
ildikovjgriffith: :)18:57
*** crose has joined #openstack-cinder19:01
*** rwellum has joined #openstack-cinder19:07
*** david-lyle has joined #openstack-cinder19:10
*** moshele has joined #openstack-cinder19:12
vivsoni__mriedem: i have provided the n-cpu logs for issue 'multipath' value not populated from nova.conf19:12
vivsoni__https://bugs.launchpad.net/nova/+bug/172392819:13
openstackLaunchpad bug 1723928 in OpenStack Compute (nova) "In case of volume_use_multipath=True, Nova unable to fetch CONF.libvirt.volume_use_multipath value from nova.conf" [Undecided,Incomplete]19:13
vivsoni__if time permits, please have a look at it19:13
*** markvoelker has joined #openstack-cinder19:13
*** markvoelker has quit IRC19:13
*** monjoshi has quit IRC19:16
*** itlinux has joined #openstack-cinder19:18
*** vivsoni__ has quit IRC19:18
*** dustins has quit IRC19:19
*** moshele has quit IRC19:21
*** jmlowe has joined #openstack-cinder19:22
*** markvoelker has joined #openstack-cinder19:23
*** itlinux has quit IRC19:24
*** esker has quit IRC19:24
*** esker has joined #openstack-cinder19:25
*** hoonetorg has quit IRC19:25
*** esker has quit IRC19:28
*** esker has joined #openstack-cinder19:29
*** gkadam has quit IRC19:32
*** harlowja has joined #openstack-cinder19:33
*** esker has quit IRC19:34
*** elmaciej has joined #openstack-cinder19:34
*** itlinux has joined #openstack-cinder19:36
*** jmlowe has quit IRC19:40
*** harlowja_ has joined #openstack-cinder19:41
*** itlinux has quit IRC19:41
*** markvoelker has quit IRC19:42
*** markvoelker has joined #openstack-cinder19:43
*** jmlowe has joined #openstack-cinder19:43
*** harlowja has quit IRC19:43
*** moshele has joined #openstack-cinder19:44
*** markvoelker has quit IRC19:48
*** gouthamr has quit IRC19:53
*** harlowja_ has quit IRC19:55
*** harlowja has joined #openstack-cinder19:55
*** gouthamr has joined #openstack-cinder20:02
*** elmaciej has quit IRC20:04
mriedemaha20:05
mriedemJan 09 10:56:10 cld6b16 nova-compute[54607]: DEBUG oslo_service.service [None req-2bb753f4-bde2-4181-9cce-f3d68bc1d3e2 None None] config files: ['/etc/nova/nova-cpu.conf'] {{(pid=54607) log_opt_values /usr/local/lib/python2.7/dist-packages/oslo_config/cfg.py:2870}}20:05
mriedemwrong config file bro!20:05
mriedemjgriffith: on your ro thing, idk20:05
mriedemjgriffith: so thinking through scenarios,20:08
mriedemwhat if i create a multiattach volume and attach it to server1, it's rw right?20:08
mriedemthen attach the volume to server2, it's then ro (if you make that change),20:09
mriedemthen i detach it from server1 (or delete server1),20:09
mriedemthe volume would still be ro on server2 even though that's the primary / sole attachment, yeah?20:09
ildikovin my understanding yes20:10
ildikovI got asked earlier whether it could be changed, I mean 'RW' and 'RO' for cases like this20:10
mriedemso i guess if you made all secondary attachments ro, you'd likely need some logic to do attachment counting on detach to see if there is 1 left and change it to be rw?20:10
ildikovbut I didn't bring that up as we are far far away from even playing with that thought...20:10
mriedemalternatively, rely on the user to provide which mode they want, i.e. pass the mode to the compute API when attaching a volume and that overrides the default in cinder20:11
*** mvk has joined #openstack-cinder20:11
mriedemat this point in queens, ^ isn't going to happen in nova, but could be tee'd up for rocky ptg for discussion?20:11
stvnoyesI like that ^ approach (mriedem)20:12
*** moshele has quit IRC20:12
ildikovI support the discussion on the PTG20:12
mriedemML might not be a bad start for kicking of discussion too20:12
mriedemsince we need to know how people are going to actually use this20:12
ildikovmriedem: I would ask it on the ops ML (too)20:12
jungleboyjI think going the safe route and leaving it RO is best for the initial stab at things and we can discuss/improve it for the future.20:13
*** markstur_ has quit IRC20:13
ildikovjungleboyj: not having any 'RW' attachments is for sure safe, but hmmm20:14
mriedemfor a volume-backed instance, the attachment has to be RW doesn't it?20:14
*** markstur has joined #openstack-cinder20:14
jungleboyjildikov: Right, but the first one needs to be RW.  I mean the subsequent ones.20:14
ildikovjungleboyj: and then you delete the first...20:14
jungleboyjildikov: I didn't mean all R/O.  Right, you delete the first and leave the subsequent ones.20:14
jungleboyjI personally think changing that could be dangerous.20:15
mriedembut when you are back down to 1 attachment, do you change that from RO to RW automatically?20:15
*** moshele has joined #openstack-cinder20:15
*** dustins has joined #openstack-cinder20:15
jungleboyjmriedem:  I don't think so.20:15
ildikovmriedem: also if there are cases when the guest on the hypervisor needs some config too it's even worse20:16
mriedemi can't really speak to how people are going to use this, so i'm just throwing out ideas20:17
mriedemreally seems like ML/PTG discussion is needed to try and get some user input20:17
mriedembut actual users will probably not be at the PTG,20:17
mriedemunless their vendor is there to speak for htem20:17
mriedem*them20:17
mriedemstvnoyes and the army of oracle people behind him are as close as i get :)20:18
ildikov:)20:18
ildikovmriedem: we will have the scientific SIG people there and they want multi-attach too20:18
mriedemwhat's blair's irc?20:19
*** moshele has quit IRC20:19
ildikovit's more oneswig who mentioned multi-attach20:19
ildikovwhich means even johnthetubaguy has info about the case20:20
mriedemok i don't know oneswig20:20
ildikovI can reach out to E/// folks whether they still want it20:20
ildikovor ask gibi to chat with the vMRS guys :)20:21
*** hoonetorg has joined #openstack-cinder20:21
mriedemE/// that's nice20:21
mriedemit is funny that gibi doesn't seem to have to care about this as much as you did when you were still there20:22
*** armaan has quit IRC20:22
ildikovI think starting with the mail thread is a great idea and we have a handful of people who knows about the usage side on the PTG and get to ideas20:22
mriedemcan i say 'not it' to starting that email thread?20:22
ildikovmriedem: well, I watched two people failing over it and I kind of got to the stage that I had to prove it's doable, I guess... :)20:23
*** armaan has joined #openstack-cinder20:23
mriedemspeaking of doable,20:23
mriedemdo you have those tests working yet?20:23
*** eharney has quit IRC20:24
ildikovmriedem: I ran all the functional tests locally except the last bunch, which then failed... Murphy...20:24
ildikovand then I switched 'multiattach' in the unit test fakes to False and now I'm rerunning the unit tests to see whether that screws up anything20:25
*** eharney has joined #openstack-cinder20:26
*** elmaciej has joined #openstack-cinder20:28
openstackgerritMerged openstack/cinder-tempest-plugin master: Updated from global requirements  https://review.openstack.org/53135320:28
*** imacdonn has quit IRC20:32
*** imacdonn has joined #openstack-cinder20:32
stvnoyeswhat army? ;-)20:32
*** markvoelker has joined #openstack-cinder20:33
ildikovstvnoyes: ;)20:36
smcginnisI plan on going to the ops midcycle. I can ask around there about multiattach and planned usage too.20:38
*** sar has quit IRC20:39
*** rwellum has left #openstack-cinder20:41
ildikovsmcginnis: that would be great!20:41
ildikovsmcginnis: it's after the PTG I think though, but still20:41
smcginnisildikov: Yeah, the following week.20:42
ildikovsmcginnis: that's good enough20:42
ildikovI mean you could share plans where we got and report back if there are too many objections as it'll still be in time to change20:43
smcginnisildikov: Hopefully the discussion is still fresh enough that I can feed back and input I get there.20:43
*** jmlowe has quit IRC20:44
*** armaan has quit IRC20:44
jgriffithmriedem: hmmm... yeah, I was thinking about that while I was out; it's sort of ugly.  It's possible, but it's ugly.20:49
jgriffithmriedem: and yes, I already intend to have the mode setting in there, just trying to determine the best way to do that.20:50
jgriffithmriedem: maybe that's enough20:50
jgriffithmriedem: but I think it needs to be an explicit option to the attach command20:50
jgriffithI can't come up with a better way to do it than that20:50
jgriffitheverything else is super wonky20:51
jgriffithso maybe add an attach option for mode with another mv bump20:51
jgriffithildikov: I'm fine with the email thread; with the caveat that it doesn't mean we can or will do what people say/want :)20:52
ildikovjgriffith: I think that's true just in general :)20:52
mriedemjgriffith: yeah i think passing the mode to the compute attach api is good for rocky, and nova will pass it through to cinder20:52
ildikovmriedem: jgriffith: +120:53
jgriffithmriedem: that seems like the best solution to me; ok... we'll plan on that route.  Can still throw it out on ML to get feedback too20:53
*** armax has quit IRC20:53
mriedemthe workaholic in me thinks it could be added to 2.59 in nova yet since that's not done, but with me being out next week and still working through test issues, it would be super tight to introduce that now i think20:54
ildikovmriedem: the workaholic? what? ;)20:54
ildikovmriedem: after you said no to the new API microversion to fix the 'reserved' state for shelved_offloaded I say no to this one20:55
jgriffithmriedem: sorry to tell you this man, but I think the "workaholic" in you and has completely consumed your inner self20:55
ildikovI sense FFE even without adding this20:55
jgriffithanywho; I'll get the changes up on this side later today or first thing tomorrow when I find some time to write code20:56
ildikovjgriffith: just skip meetings :)20:57
*** belmoreira has joined #openstack-cinder20:57
jgriffithildikov: Ok, I'll send the decline responses out with you as my authorizing agent20:58
*** esker has joined #openstack-cinder20:58
ildikovjgriffith: approved, go ahead :)21:02
*** crose has quit IRC21:02
*** esker has quit IRC21:02
*** jmlowe has joined #openstack-cinder21:06
*** esker has joined #openstack-cinder21:07
*** liverpooler has quit IRC21:11
*** markvoelker has quit IRC21:13
*** markvoelker has joined #openstack-cinder21:13
*** markvoelker has quit IRC21:17
*** armax has joined #openstack-cinder21:34
*** esker has quit IRC21:35
*** pcaruana has quit IRC21:38
*** david-lyle has quit IRC21:40
*** dave-mccowan has quit IRC21:40
*** eharney has quit IRC21:41
*** threestrands has joined #openstack-cinder21:44
*** threestrands has quit IRC21:44
*** threestrands has joined #openstack-cinder21:44
*** rmcall has quit IRC21:52
*** esker has joined #openstack-cinder21:53
*** dillaman has quit IRC21:53
*** moshele has joined #openstack-cinder21:59
*** AlexeyAbashkin has joined #openstack-cinder22:00
*** moshele has quit IRC22:04
*** mchlumsky has quit IRC22:04
*** esker has quit IRC22:05
*** AlexeyAbashkin has quit IRC22:05
*** rcernin has joined #openstack-cinder22:06
*** belmoreira has quit IRC22:10
*** gman-tx has quit IRC22:16
*** dustins has quit IRC22:24
*** abishop has quit IRC22:24
*** elmaciej has quit IRC22:27
*** elmaciej has joined #openstack-cinder22:28
*** markvoelker has joined #openstack-cinder22:29
*** markvoelker has quit IRC22:30
*** gouthamr has quit IRC22:35
*** itlinux has joined #openstack-cinder22:47
*** gouthamr has joined #openstack-cinder22:53
*** _pewp_ has quit IRC22:56
*** _pewp_ has joined #openstack-cinder22:57
*** moshele has joined #openstack-cinder22:57
*** e0ne has joined #openstack-cinder22:58
*** moshele has quit IRC23:00
*** abishop has joined #openstack-cinder23:02
*** sdague has quit IRC23:04
*** chhavi__ has joined #openstack-cinder23:11
*** markvoelker has joined #openstack-cinder23:11
*** markvoelker has quit IRC23:11
*** chhavi__ has quit IRC23:15
*** stvnoyes has quit IRC23:19
*** AlexeyAbashkin has joined #openstack-cinder23:22
*** AlexeyAbashkin has quit IRC23:27
*** sdague has joined #openstack-cinder23:34
*** gmann has joined #openstack-cinder23:35
*** moshele has joined #openstack-cinder23:43
*** markvoelker has joined #openstack-cinder23:43
*** markvoelker has quit IRC23:43
*** moshele has quit IRC23:49
*** netapp has quit IRC23:54
*** netapp has joined #openstack-cinder23:58

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!