Wednesday, 2015-12-02

« Tuesday, 2015-12-01 Index Thursday, 2015-12-03 »
*** dilip has joined #openstack-cinder00:00
*** e0ne_ has quit IRC00:00
*** diogogmt has joined #openstack-cinder00:00
hemnamtanino, ping00:02
mtaninohemna: pong00:02
hemnamtanino, hey, so I'm just poking around00:02
*** zhangjn has joined #openstack-cinder00:02
hemnamtanino, noticed you authored the nova/virt/libvirt/storage/lvm.py00:02
mtaninohemna: ah, I fixed it if I remember correctly00:03
hemnaI presume that's nova's code for creating local lvm volumes for VMs00:03
hemnaat least it seems to be00:03
mtaninohemna: Yes, I think so.00:04
*** salv-orlando has quit IRC00:04
hemnaso we also have cinder/brick/local_dev/lvm.py00:04
mtaninoThat's code is for ephemeral disk of Nova.00:04
hemnawhich seems to do a lot of the same stuff00:05
hemnaand looks a bit more feature rich from what I can tell00:05
mtaninoyes, so it is better to merge these stuffs.00:05
hemnado you think that makes sense to migrate into brick and have nova and cinder share that ?00:05
hemnathat was one of the original goals of 'brick' as we discussed it way back in Havana00:05
hemnaI haven't looked at this stuff in detail much until today00:06
mtaninohemna: I think it's make sense. Maintaining duplicate code is costful,00:06
hemnacinder's lvm driver uses the the cinder/brick/local_dev/lvm.py code to do work00:06
mtaninohmm00:06
*** ZZelle has quit IRC00:06
hemnayah, I'd rather see bug fixes for lvm related volumes get seen/used by nova00:07
mtaninoso do we try to move lvm code into Brick?00:07
hemnathat's what I'm thinking00:07
hemnait would need a few more methods00:07
mtaninoI see.00:07
hemnathe nova code has some helper functions to list volumes, get size etc.00:07
hemnathat the cinder local_dev doesnt'00:08
hemnathere is also a concept of clearing the volume in nova before deleting it. (writing zero's to it)00:08
hemnaquite a bit of differences between how cinder does delete and nova00:09
mtaninoso, like a LVM library is fine for both Cinder and nova00:09
hemnayah00:09
hemnaI like how the cinder local_dev/lvm.py is written00:09
hemnaI think jgriffith did almost all of it.00:09
hemnaeharney as well looks like.00:10
mtaninohmm00:10
*** yhayashi has quit IRC00:11
hemnaI don't think it would be too hard to do actually,.00:11
*** dilip has quit IRC00:12
mtaninoYes, I think it can be done00:13
hemnaok, maybe I'll throw together a poc00:13
mtaninoI see. It's nice.00:14
openstackgerritMerged openstack/cinder: Cleanup orphaned code in sqlalchemy API  https://review.openstack.org/25094400:14
*** terryyao has quit IRC00:14
*** jerrygb has joined #openstack-cinder00:16
*** EinstCrazy has joined #openstack-cinder00:16
*** zhangjn has quit IRC00:17
*** EinstCrazy has quit IRC00:17
*** EinstCrazy has joined #openstack-cinder00:18
openstackgerritAlex O'Rourke proposed openstack/cinder: 3PAR: Implement v2 replication (managed)  https://review.openstack.org/23120100:20
openstackgerritAlex O'Rourke proposed openstack/cinder: 3PAR: Implement v2 replication (unmanaged)  https://review.openstack.org/23124500:21
*** mc_nair has quit IRC00:22
*** EinstCrazy has quit IRC00:22
*** yhayashi has joined #openstack-cinder00:23
DuncanTIs the cinder dashboard link (00:26
DuncanThttps://review.openstack.org/#/dashboard/?foreach=project%3A%5Eopenstack%2F.%2Acinder.%2A+status%3Aopen+NOT+owner%3Aself+NOT+label%3AWorkflow%3C%3D%252D1+label%3AVerified%3E%3D1+NOT+label%3ACode%252DReview%3C%3D%252D1%252cself+NOT+label%3ACode%252DReview%3E%3D1%252cself&title=Cinder+Review+Inbox&Cinder+Specs=project%3Aopenstack%2Fcinder%252Dspecs&Bug+Fixes=t00:26
DuncanTopic%3A%5Ebug%2F.%2A&Needs+Feedback+%28Changes+older+than+5+days+that+have+not+been+reviewed+by+anyone%29=NOT+label%3ACode%252DReview%3C%3D2+age%3A5d&You+are+a+reviewer%252c+but+haven%27t+voted+in+the+current+revision=reviewer%3Aself&Needs+final+%2B2=label%3ACode%252DReview%3E%3D2+limit%3A50&New+Contributors=reviewer%3A10068&Passed+Jenkins%252c+No+Negative+F00:26
DuncanTeedback=NOT+label%3ACode%252DReview%3E%3D2+NOT+label%3ACode%252DReview%3C%3D%252D1+limit%3A50&Wayward+Changes+%28Changes+with+no+code+review+in+the+last+2days%29=NOT+label%3ACode%252DReview%3C%3D2+age%3A2d ) broken for everybody, or just me?00:26
DuncanT(link at the bottom of https://wiki.openstack.org/wiki/Cinder#Project_Meetings )00:26
*** shhjwang has joined #openstack-cinder00:27
mtaninoDuncanT: ping,00:28
DuncanTmtanino: Hi00:28
mtaninoDuncanT: Hi, thank you for the comennt on https://review.openstack.org/#/c/252065/100:29
*** ildikov has quit IRC00:29
mtaninoDuncanT: I'm thinking what kind of approach is best way for Nova boot volume case.00:29
*** ildikov has joined #openstack-cinder00:29
DuncanTmtanino: Ah yes. An interesting idea, I'm just not sure it solves enough of the whole problem to be the right solution00:29
openstackgerritAlex O'Rourke proposed openstack/cinder: 3PAR: Implement v2 replication (managed)  https://review.openstack.org/23120100:30
openstackgerritAlex O'Rourke proposed openstack/cinder: 3PAR: Implement v2 replication (unmanaged)  https://review.openstack.org/23124500:30
DuncanTmtanino: Am I right in thinking that none boot-volumes will have exactly the same issue?00:30
*** gouthamr has joined #openstack-cinder00:31
hemnaDuncanT,https://goo.gl/00:31
*** IlyaG has quit IRC00:31
mtaninoDuncanT: If user don't know the relation between volume_type and hypervisor, I think it happnes.00:31
hemnaDuncanT, the review inbox url ?00:31
*** mdbooth has quit IRC00:32
DuncanThttps://goo.gl/M1YhMH and yes, the review inbox00:32
hemnaDuncanT, works for me00:32
hemnait's my 'home page' in chrome00:32
mtaninoDuncanT: even if it is not boot from volume case.00:32
DuncanThemna: Ok, thanks, must be something weird I've done00:32
DuncanThemna: I upgraded a few crap block extensions, might eb that00:33
DuncanTmtanino: So the solution is to set up suitable volume types - none boot volumes are just as problematic as boot volumes. Or just fix all the backends to work everywhere, which would be even better00:34
*** gouthamr_ has joined #openstack-cinder00:34
*** bswartz has joined #openstack-cinder00:34
mtaninoDuncanT: so volume_type kvm.SSD, VMware.SSD or etc is one of solution which you mentioned00:35
mtaninoDuncanT: I think VMware backend or Windows backend can't support everywhere00:35
DuncanTmtanino: So an operator with a hetrogenious cloud should seriously consider not using either of those backends... and if they do then they'll need appropriate types00:36
mtaninoDuncanT: We have a issue boot from volume issue using KVM and VMware multihypervisor environment.00:37
*** gouthamr has quit IRC00:37
*** angela-s has quit IRC00:37
mtaninoDuncanT: and Cinder multi backend00:37
*** dilip has joined #openstack-cinder00:38
DuncanTmtanino: So create a VMware-only volume type - you'll need it for none boot-from-volumes anyway00:38
mtaninoDuncanT: yes, and If nova can specify volume type during boot from volume, problem is not happen.00:39
mtaninoDuncanT: Should user know which volume type can be applied to specific hypervisor?00:39
DuncanTmtanino: The user is the only one who can know in the general case, and it changes over time, e.g. some but not all hypervisors have gained native ceph support00:40
DuncanTmtanino: Where 'user' means 'operator' just to be clear00:41
mtaninoDuncanT: not an operator. Real end user of cloud.00:41
mtaninoDuncanT: in my case.00:41
*** smoriya_ has joined #openstack-cinder00:42
DuncanTmtanino: They'll have to know, since there's no way of hinting it for a none boot-from-volume volume even if we add the code in your spec. The cloud operator should add it to the description of the type if it is significant. In general I'd suggest that you want to pick a universal backend for most clouds, unless you have fairly skilled users or a weirrd00:42
DuncanTrequirements set00:42
mtaninoDuncanT: Yes, I thought at first, then proposed this. https://review.openstack.org/#/c/245986/00:43
*** dilip has quit IRC00:44
mtaninoDuncanT: patch set 4 is approach using volume type.00:44
*** markvoelker has quit IRC00:45
*** zhenguo has joined #openstack-cinder00:45
DuncanTmtanino: Yes, that (patch set 4) looks like the right solution to the problem, and is useful for other general volume type uses, like picking an SSD type as boot volume00:47
mtaninoDuncanT: thank you for the comment.00:47
*** hemna is now known as hemnafk00:48
mtaninoDuncanT: I should think that approach again.00:48
DuncanTmtanino: You're welcome. Please ping me again if you have any questions or if nova hate taht idea - I'm happy to talk to nova devs about it if needed00:48
mtaninoDuncanT: It' very helpful. Thank you00:49
*** alonma has joined #openstack-cinder00:52
*** diogogmt has quit IRC00:55
*** ctina_ has joined #openstack-cinder00:55
*** alonma has quit IRC00:56
*** alonma has joined #openstack-cinder00:59
*** ctina_ has quit IRC01:00
*** zhangjn has joined #openstack-cinder01:01
*** EinstCrazy has joined #openstack-cinder01:01
*** salv-orlando has joined #openstack-cinder01:02
*** alonma has quit IRC01:04
*** tull has quit IRC01:05
*** alonma has joined #openstack-cinder01:08
tbarronDuncanT: thanks for the comment on https://review.openstack.org/24649801:08
tbarroneharney: ^^^^^01:08
*** IlyaG has joined #openstack-cinder01:09
tbarronDuncanT, eharney, npotter: my role on this one is mostly instigator :-)01:10
tbarronwe clearly haven't closed all the race windows here and it prob. isn't fair to put closing them all on npotter's fix01:10
tbarronI'm just trying to figure the best way forward beyond that fix, as this race is one of many.01:11
DuncanTtbarron: My argument here would be that the service might be down momentarily and so the correct behaviour would be to leave the delete request in the queue to be dealt with later, like we do for volume requests, with the backup in the deleting state01:11
DuncanTtbarron: Unless I'm miss-understanding the issue01:11
tbarroncompare and swap on DB state within the api itself seems to be the way we want to go.01:12
tbarronbut eharney is pointing to potential races when we cast out, and I dunno if that has been thought through.01:12
*** mtanino has quit IRC01:12
*** alonma has quit IRC01:13
tbarronDuncanT: your argument may well be sound, I'm just digesting, learning, and very interested.01:13
tbarronThese small window races will be more problematic with scale-out backup service, and with active-active volume service.01:13
DuncanTtbarron: I don't think returning an error to the user if the service is down is the right thing to do - put the backup in deleting in the api, fire the work into the queue and forget it. Sure the backup will sit in deleting until the service comes back up, but who cares?01:14
DuncanTtbarron: I'm not sure I see the race01:14
tbarronDuncanT: so you are -1 on the npotter fix in the first place?  If one endorses the fix, then there are races.01:15
DuncanTtbarron: I might be missing something, please don't discount that, it's late here01:15
tbarronDuncanT: well, I thought it was late here on the east coast of the US :-), but then I go to bed and get up early.01:16
openstackgerritAtsushi SAKAI proposed openstack/python-cinderclient: Fix comma location in comment  https://review.openstack.org/25210801:16
tbarronDuncanT: we should probably pick this up sometime tomorrow when you've had lunch and eharney and npotter are online.01:16
DuncanTtbarron: Fine by me. It is late enough here to count as early, but I was making progress on stuff so decided to carry on01:17
tbarronDuncanT: :-)01:17
DuncanTNo doubt I'll regret this in the morning01:17
tbarronDuncanT: in my expeience the best work is done in the wee hours, either late or early.01:18
tbarronDuncanT: but get some sleep.  I'm interested in what you'll say when we return to the subject.01:18
DuncanTtbarron: I find it is a different state of mind - if I can't fix it during the day, sometimes early or late provides a new insight01:19
tbarronDuncanT: completely agree. The business day is all just dealing with meetings, people, etc.01:19
DuncanTtbarron: Sometimes I look at the code I write late at night again in the cold light of day and burn the hard drive to hdie the evidence....01:19
tbarronhehe01:19
tbarronand sometimes you say, well, there was something here!01:20
tbarronttyl, I'm going afk for a while.  Get some sleep!01:21
smcginnishemnafk: Not still around, are you?01:22
*** ildikov has quit IRC01:23
smcginnisDuncanT: bswartz had an issue with the dashboard last week IIRC.01:26
smcginnisDuncanT: Works fine for me.01:26
smcginnisDuncanT: I can't remember how he got around it, but maybe try FF vs Chrome or something.01:27
smcginnisDuncanT: And make sure you're logged in too.01:27
DuncanTsmcginnis: It works for me under chrome but not firefox01:27
smcginnisDuncanT: Cool. At least it works somewhere. :)01:27
smcginnisDuncanT: Maybe when Ben's around you can ping him on how he fixed it.01:28
smcginnisDuncanT: Isn't it pretty late there?01:28
*** Lee1092 has joined #openstack-cinder01:29
DuncanTsmcginnis: Yeah, I will call it a night shortly. Not pulled a late night for a while, and it seems to have been productive01:29
smcginnisDuncanT: Awesome! It's a good feeling. :)01:30
*** lixiaoy1 has joined #openstack-cinder01:40
*** RA has joined #openstack-cinder01:42
openstackgerritMerged openstack/cinder: xio: fix regression in authentication  https://review.openstack.org/25175501:42
*** RA is now known as Guest7921201:42
*** mriedem has joined #openstack-cinder01:44
*** rajinir_ has quit IRC01:45
*** jwcroppe has joined #openstack-cinder01:45
*** zhangjn has quit IRC01:46
*** jerrygb has quit IRC01:50
*** diogogmt has joined #openstack-cinder01:54
*** jerrygb has joined #openstack-cinder01:55
*** leeantho has quit IRC01:57
*** ctracey has quit IRC02:05
*** zhangjn has joined #openstack-cinder02:06
*** ctracey has joined #openstack-cinder02:06
openstackgerritwanghao proposed openstack/python-cinderclient: support for snapshot management  https://review.openstack.org/24909002:07
*** nug has joined #openstack-cinder02:07
*** alonma has joined #openstack-cinder02:09
*** ctracey has left #openstack-cinder02:11
*** alonma has quit IRC02:13
*** alonma has joined #openstack-cinder02:16
*** gcb has joined #openstack-cinder02:20
*** alonma has quit IRC02:20
*** salv-orlando has quit IRC02:23
*** alonma has joined #openstack-cinder02:24
*** alonma has quit IRC02:29
*** smoriya_ has quit IRC02:30
*** david-lyle has joined #openstack-cinder02:31
*** alonma has joined #openstack-cinder02:34
*** terryyao has joined #openstack-cinder02:35
openstackgerritMerged openstack/cinder: Add Mitaka-1 release notes  https://review.openstack.org/25151702:37
openstackgerritMerged openstack/cinder: Add retype in lvm driver  https://review.openstack.org/24496402:37
openstackgerritMerged openstack/cinder: Sheepdog: Optimization of error handling  https://review.openstack.org/24555202:38
openstackgerritMerged openstack/cinder: Add some missing fields to Volume object  https://review.openstack.org/24084702:38
openstackgerritMerged openstack/cinder: EMC VMAX - Change naming convention for MV and SG for FAST  https://review.openstack.org/24700802:38
*** alonma has quit IRC02:38
*** alonma has joined #openstack-cinder02:42
*** lprice has quit IRC02:42
*** lprice has joined #openstack-cinder02:42
*** haomaiwa_ has joined #openstack-cinder02:42
openstackgerritCedric Zhuang proposed openstack/cinder: VNX: Get volume type from location throws error  https://review.openstack.org/24623002:44
openstackgerritLisaLi proposed openstack/cinder: Delete unused codes in rbd.retype  https://review.openstack.org/24446002:44
*** alonma has quit IRC02:46
*** houming has joined #openstack-cinder02:47
lixiaoy1DuncanT: ping. Could you help to review the retype spec https://review.openstack.org/#/c/248593?02:50
lixiaoy1DuncanT: and plus scaling backup spec https://review.openstack.org/#/c/24097802:51
*** dilip has joined #openstack-cinder02:51
*** dilip has quit IRC02:51
*** dilip has joined #openstack-cinder02:51
*** haomaiwa_ has quit IRC02:52
*** haomaiwang has joined #openstack-cinder02:53
*** bardia has quit IRC02:53
*** alonma has joined #openstack-cinder02:56
*** haomaiwang has quit IRC02:57
*** gouthamr has joined #openstack-cinder02:59
*** gouthamr_ has quit IRC02:59
*** david-lyle has quit IRC03:00
*** alonma has quit IRC03:00
*** apoorvad has quit IRC03:01
*** bardia has joined #openstack-cinder03:04
*** Guest79212 has quit IRC03:06
*** changbl has joined #openstack-cinder03:13
*** [2]Thelo has joined #openstack-cinder03:13
*** dilip has quit IRC03:13
*** gouthamr has quit IRC03:14
*** [1]Thelo has quit IRC03:15
*** dongc has joined #openstack-cinder03:18
*** salv-orlando has joined #openstack-cinder03:20
*** mriedem has quit IRC03:22
*** dilip has joined #openstack-cinder03:23
*** dims has quit IRC03:26
*** bardia has quit IRC03:33
*** jerrygb has quit IRC03:34
*** dilip has quit IRC03:34
*** bardia has joined #openstack-cinder03:34
*** jerrygb has joined #openstack-cinder03:34
*** jungleboyj has joined #openstack-cinder03:37
*** uberjay has quit IRC03:38
openstackgerritLisaLi proposed openstack/cinder: Update retype limitation in volume/api  https://review.openstack.org/24546003:38
*** jerrygb has quit IRC03:39
*** uberjay has joined #openstack-cinder03:40
openstackgerritSean McGinnis proposed openstack/cinder: Remove deprecated LVM ISCSI and ISER Drivers  https://review.openstack.org/25094803:41
*** links has joined #openstack-cinder03:43
wanghaosmcginnis: ping03:45
openstackgerritLisaLi proposed openstack/cinder: Cleanup in backup reset status  https://review.openstack.org/24674803:47
wanghaosmcginnis: hi Sean,  Could you help me to review query volume glance metadata spec: https://review.openstack.org/#/c/174778/03:48
wanghaosmcginnis: I have updated this for your comment.03:48
*** jiangliucheng has joined #openstack-cinder03:56
*** alonma has joined #openstack-cinder03:56
openstackgerritShu Muto proposed openstack/python-cinderclient: Delete python bytecode before every test run  https://review.openstack.org/25217103:57
*** alonma has quit IRC04:01
*** zhangjn has quit IRC04:03
*** gouthamr has joined #openstack-cinder04:04
openstackgerritVincent Hou proposed openstack/cinder: Storwize: Implement v2 replication (mirror)  https://review.openstack.org/24904204:06
openstackgerritVincent Hou proposed openstack/cinder: Storwize: Implement v2 replication (split IO)  https://review.openstack.org/23738704:06
openstackgerritVincent Hou proposed openstack/cinder: Storwize: Implement v2 replication (mirror)  https://review.openstack.org/24904204:09
openstackgerritVincent Hou proposed openstack/cinder: Storwize: Implement v2 replication (split IO)  https://review.openstack.org/23738704:09
openstackgerritxing-yang proposed openstack/cinder: Add a GateDriver  https://review.openstack.org/24495604:10
*** elonden has joined #openstack-cinder04:17
*** bardia has quit IRC04:18
*** houming has quit IRC04:19
*** IlyaG has quit IRC04:26
*** zhangjn has joined #openstack-cinder04:31
*** divab has joined #openstack-cinder04:37
openstackgerritMerged openstack/cinder: Fix the bug of can't get the desired image info  https://review.openstack.org/24524104:39
*** dilip has joined #openstack-cinder04:40
*** p0rtal_ has quit IRC04:40
*** david-lyle has joined #openstack-cinder04:44
*** salv-orlando has quit IRC04:51
*** salv-orlando has joined #openstack-cinder04:51
*** dilip has quit IRC04:56
*** alonma has joined #openstack-cinder04:56
*** gouthamr has quit IRC04:58
*** alonma has quit IRC05:01
*** alonma has joined #openstack-cinder05:03
*** alonma has quit IRC05:07
*** alonma has joined #openstack-cinder05:11
*** zhangjn has quit IRC05:12
*** alonma has quit IRC05:16
*** deepakcs has joined #openstack-cinder05:18
*** divab is now known as abhi05:23
*** markvoelker has joined #openstack-cinder05:24
*** zhangjn has joined #openstack-cinder05:32
*** markvoelker has quit IRC05:41
*** zhonghua-lee has quit IRC05:42
*** zhonghua-lee has joined #openstack-cinder05:43
*** links has quit IRC05:43
*** mudassirlatif has joined #openstack-cinder05:47
*** rcernin has joined #openstack-cinder05:52
*** harlowja_at_home has joined #openstack-cinder05:53
*** links has joined #openstack-cinder05:54
*** mudassirlatif has quit IRC05:57
*** mudassirlatif has joined #openstack-cinder05:57
openstackgerritManjeet Singh Bhatia proposed openstack/python-cinderclient: Pass proxy environment variable to tox  https://review.openstack.org/25219806:06
*** shhjwang has quit IRC06:07
*** bardia has joined #openstack-cinder06:10
*** houming has joined #openstack-cinder06:10
*** bardia has quit IRC06:14
*** alonma has joined #openstack-cinder06:15
*** harlowja has quit IRC06:16
*** alonma has quit IRC06:21
*** alonma has joined #openstack-cinder06:25
*** rcernin has quit IRC06:25
openstackgerritOpenStack Proposal Bot proposed openstack/cinder: Imported Translations from Zanata  https://review.openstack.org/25220906:26
*** alonma has quit IRC06:29
*** mudassirlatif has quit IRC06:32
*** alonma has joined #openstack-cinder06:36
*** p0rtal has joined #openstack-cinder06:37
*** ChubYann has quit IRC06:38
*** alonma has quit IRC06:40
*** alonma has joined #openstack-cinder06:42
*** bardia has joined #openstack-cinder06:44
*** alonma has quit IRC06:47
*** p0rtal has quit IRC06:48
*** alonma has joined #openstack-cinder06:49
*** bardia has quit IRC06:50
*** alonma has quit IRC06:53
openstackgerritwanghao proposed openstack/cinder-specs: Get the replicated volume ref for DR  https://review.openstack.org/18215006:54
*** mudassirlatif has joined #openstack-cinder06:54
*** alonma has joined #openstack-cinder06:55
*** alonma has quit IRC06:59
*** harlowja_at_home has quit IRC07:01
*** alonma has joined #openstack-cinder07:01
*** terryyao has quit IRC07:03
*** dilip has joined #openstack-cinder07:04
*** terryyao has joined #openstack-cinder07:05
*** alonma has quit IRC07:05
*** alonma has joined #openstack-cinder07:07
*** zhenguo has quit IRC07:07
*** jwcroppe has quit IRC07:09
*** rcernin has joined #openstack-cinder07:09
*** dilip has quit IRC07:10
*** jwcroppe has joined #openstack-cinder07:11
openstackgerritSilvan Kaiser proposed openstack/os-brick: Add quobyte protocol for remotefs connections  https://review.openstack.org/25037107:12
*** alonma has quit IRC07:12
*** mudassirlatif has quit IRC07:15
*** alonma has joined #openstack-cinder07:18
*** terryyao has quit IRC07:21
*** jerrygb has joined #openstack-cinder07:21
*** alonma has quit IRC07:22
*** alonma has joined #openstack-cinder07:25
*** jerrygb has quit IRC07:25
openstackgerritPeter Penchev proposed openstack/cinder: Reintroduce the StorPool driver  https://review.openstack.org/22015507:26
*** abehl has joined #openstack-cinder07:27
openstackgerritRico Lin proposed openstack/python-cinderclient: deprecate python26 support  https://review.openstack.org/25222807:29
*** alonma has quit IRC07:29
*** salv-orlando has quit IRC07:29
*** alexschm has joined #openstack-cinder07:36
*** alonma has joined #openstack-cinder07:37
*** sgotliv__ has quit IRC07:37
*** salv-orlando has joined #openstack-cinder07:40
*** alonma has quit IRC07:41
*** alonma has joined #openstack-cinder07:45
openstackgerritShu Muto proposed openstack/python-cinderclient: Remove py26 support  https://review.openstack.org/25223207:46
*** ildikov has joined #openstack-cinder07:49
*** alonma has quit IRC07:50
*** ociuhandu has joined #openstack-cinder07:50
*** [2]Thelo has quit IRC07:50
*** [1]Thelo has joined #openstack-cinder07:51
*** elonden has quit IRC07:53
*** alonma has joined #openstack-cinder07:57
*** alonma has quit IRC08:02
*** alonma has joined #openstack-cinder08:03
*** haomaiwang has joined #openstack-cinder08:04
*** zhangjn has quit IRC08:05
*** nkrinner has joined #openstack-cinder08:05
*** smoriya_ has joined #openstack-cinder08:06
*** alonma has quit IRC08:08
*** alonma has joined #openstack-cinder08:09
*** liverpooler has joined #openstack-cinder08:11
*** zhangjn has joined #openstack-cinder08:13
*** alonma has quit IRC08:14
*** smoriya_ has quit IRC08:15
*** alonma has joined #openstack-cinder08:16
*** chenying has quit IRC08:20
*** alonma has quit IRC08:20
*** alonma has joined #openstack-cinder08:22
*** vgridnev has joined #openstack-cinder08:22
*** markus_z has joined #openstack-cinder08:22
*** alonma has quit IRC08:27
*** shakamunyi has quit IRC08:27
*** alonma has joined #openstack-cinder08:28
openstackgerritwanghao proposed openstack/cinder: Get the replicated volume ref for DR  https://review.openstack.org/25225008:29
*** smoriya_ has joined #openstack-cinder08:29
*** egonzalez has joined #openstack-cinder08:29
*** chenying_ has quit IRC08:31
*** chenying has joined #openstack-cinder08:31
*** alonma has quit IRC08:33
*** smoriya_ has quit IRC08:33
*** alonma has joined #openstack-cinder08:36
*** gcb has quit IRC08:39
openstackgerritxiaoqin proposed openstack/cinder: Only source-id is accepted for IBM storwize/SVC volume manage command right now. volume manage command using source-name or source-id should be both working.  https://review.openstack.org/25225908:41
*** alonma has quit IRC08:41
*** smoriya__ has joined #openstack-cinder08:42
*** alonma has joined #openstack-cinder08:42
*** smoriya__ has quit IRC08:43
*** smoriya_ has joined #openstack-cinder08:43
*** gcb has joined #openstack-cinder08:44
*** smoriya_ has quit IRC08:44
*** smoriya_ has joined #openstack-cinder08:45
*** lpetrut has joined #openstack-cinder08:47
*** alonma has quit IRC08:47
*** salv-orlando has quit IRC08:48
*** salv-orlando has joined #openstack-cinder08:48
*** alonma has joined #openstack-cinder08:49
openstackgerritxiaoqin proposed openstack/cinder: volume manage using source-name  https://review.openstack.org/25225908:49
openstackgerritShay Halsband proposed openstack/cinder: XtremIO: fix generic glance cache with XtremIO  https://review.openstack.org/24981408:50
*** salv-orlando has quit IRC08:53
*** alonma has quit IRC08:53
*** smoriya_ has quit IRC08:54
*** haomaiwang has quit IRC08:54
*** alonma has joined #openstack-cinder08:55
*** smoriya_ has joined #openstack-cinder08:55
*** haomaiwa_ has joined #openstack-cinder08:57
*** houming has quit IRC08:58
*** alonma has quit IRC08:59
*** houming has joined #openstack-cinder09:00
*** alonma has joined #openstack-cinder09:01
*** alonma has quit IRC09:06
*** divab has joined #openstack-cinder09:06
*** abhi has quit IRC09:07
*** alonma has joined #openstack-cinder09:07
*** alonma has quit IRC09:12
*** alonma has joined #openstack-cinder09:13
*** egonzalez has quit IRC09:14
openstackgerritwanghao proposed openstack/cinder: Get the replicated volume ref for DR  https://review.openstack.org/25225009:15
*** mdbooth has joined #openstack-cinder09:17
*** alonma has quit IRC09:18
*** alonma has joined #openstack-cinder09:20
*** alonma has quit IRC09:20
*** vgridnev has quit IRC09:20
*** alonma has joined #openstack-cinder09:20
*** aix has joined #openstack-cinder09:21
*** vinayp has quit IRC09:26
*** deepakcs has quit IRC09:27
markus_zThere is a bug reported for nova which sounds a lot like it should be for cinder. Could one of you please have a look at https://bugs.launchpad.net/nova/+bug/152189409:29
openstackLaunchpad bug 1521894 in OpenStack Compute (nova) "Datastore selection in cinder vmdk driver is too slow" [Undecided,New]09:29
*** deepakcs has joined #openstack-cinder09:32
wanghaomarkus_z: Looks like should be a issue in vmware driver.09:33
wanghaomarkus_z: not for nova.09:33
*** [2]Thelo has joined #openstack-cinder09:37
*** hogepodge has quit IRC09:38
*** [1]Thelo has quit IRC09:40
*** hogepodge has joined #openstack-cinder09:41
*** jordanP has joined #openstack-cinder09:43
*** zhangjn has quit IRC09:44
*** alonma has quit IRC09:50
*** jistr has joined #openstack-cinder09:50
*** alonma has joined #openstack-cinder09:50
*** haomaiwa_ has quit IRC09:51
*** alonma has quit IRC09:52
*** alonma has joined #openstack-cinder09:53
*** houming has quit IRC09:53
*** dongc has quit IRC10:00
*** e0ne has joined #openstack-cinder10:03
*** zhangjn has joined #openstack-cinder10:06
*** e0ne has quit IRC10:06
*** e0ne_ has joined #openstack-cinder10:07
*** e0ne_ has quit IRC10:09
*** e0ne has joined #openstack-cinder10:09
*** e0ne_ has joined #openstack-cinder10:14
*** smoriya_ has quit IRC10:15
*** e0ne has quit IRC10:16
markus_zwanghao: thanks, I moved it to the cinder project and added a person from vmware to the subscriber list.10:28
*** dzamboni has joined #openstack-cinder10:29
*** e0ne_ has quit IRC10:30
*** alonma has quit IRC10:30
*** sgotliv__ has joined #openstack-cinder10:31
*** alonma has joined #openstack-cinder10:31
*** alonma has quit IRC10:32
*** alonma has joined #openstack-cinder10:32
*** vgridnev has joined #openstack-cinder10:34
*** zhangjn has quit IRC10:35
openstackgerritMerged openstack/cinder: Delete unused codes in rbd.retype  https://review.openstack.org/24446010:36
*** alonma has quit IRC10:37
*** weihan has joined #openstack-cinder10:39
*** zhangjn has joined #openstack-cinder10:40
DuncanTThe nova blueprint to move dm-setup crypt code to brick just got approved10:49
*** gcb_ has joined #openstack-cinder11:00
*** flip214 has quit IRC11:01
*** gcb has quit IRC11:03
*** alonma has joined #openstack-cinder11:04
*** gcb_ has quit IRC11:07
*** alonma has quit IRC11:08
*** alonma has joined #openstack-cinder11:09
*** alonma has quit IRC11:14
*** alonma has joined #openstack-cinder11:16
*** flip214 has joined #openstack-cinder11:16
*** flip214 has quit IRC11:16
*** flip214 has joined #openstack-cinder11:16
*** alonma has quit IRC11:20
*** zhangjn has quit IRC11:21
*** alonma has joined #openstack-cinder11:22
openstackgerritSzymon Borkowski proposed openstack/cinder: Use proper config option to connect to keystone  https://review.openstack.org/25193711:23
*** zhangjn has joined #openstack-cinder11:23
*** erlon has joined #openstack-cinder11:26
*** alonma has quit IRC11:26
*** alonma has joined #openstack-cinder11:28
*** alonma has quit IRC11:33
*** aix has quit IRC11:34
*** gcb has joined #openstack-cinder11:36
*** alonma has joined #openstack-cinder11:41
*** gcb has quit IRC11:45
openstackgerritMerged openstack/cinder: Fix typo in LIO terminate_connection error msg  https://review.openstack.org/23538311:47
*** gcb has joined #openstack-cinder11:50
*** zhangjn has quit IRC11:57
*** EinstCrazy has quit IRC11:57
*** deepakcs has quit IRC12:01
*** aix has joined #openstack-cinder12:04
openstackgerritMichal Dulko proposed openstack/cinder: Add metadata aliases to Volume object  https://review.openstack.org/24776312:04
openstackgerritMichal Dulko proposed openstack/cinder: Recognize extra fields in CinderObjectDictCompat  https://review.openstack.org/24776212:04
*** leo__ has joined #openstack-cinder12:05
*** gouthamr has joined #openstack-cinder12:11
*** raildo-afk is now known as raildo12:12
*** egonzalez has joined #openstack-cinder12:13
*** houming has joined #openstack-cinder12:14
*** salv-orlando has joined #openstack-cinder12:15
*** LiuNanke has joined #openstack-cinder12:16
*** nikeshm has quit IRC12:19
*** alonma has quit IRC12:20
*** alonma has joined #openstack-cinder12:21
*** sgotliv__ has quit IRC12:22
*** EinstCrazy has joined #openstack-cinder12:22
*** LiuNanke has quit IRC12:22
*** alonma has quit IRC12:25
*** alonma has joined #openstack-cinder12:27
*** LiuNanke has joined #openstack-cinder12:28
*** LiuNanke has quit IRC12:29
*** alonma has quit IRC12:31
*** gouthamr has quit IRC12:32
*** alonma has joined #openstack-cinder12:33
*** alonma has quit IRC12:38
*** sgotliv__ has joined #openstack-cinder12:38
*** alonma has joined #openstack-cinder12:39
*** gouthamr has joined #openstack-cinder12:40
*** zhangjn has joined #openstack-cinder12:41
openstackgerritVipin Balachandran proposed openstack/cinder: VMware: Unit test refactoring (image to vol - 1/2)  https://review.openstack.org/24917412:42
*** zhangjn has quit IRC12:42
*** alonma has quit IRC12:44
*** zhangjn has joined #openstack-cinder12:44
*** zhangjn has quit IRC12:45
*** alonma has joined #openstack-cinder12:46
*** zhangjn has joined #openstack-cinder12:49
*** alonma has quit IRC12:51
*** salv-orlando has quit IRC12:51
*** zhangjn has quit IRC12:52
*** zhangjn_ has joined #openstack-cinder12:52
*** alonma has joined #openstack-cinder12:52
*** salv-orlando has joined #openstack-cinder12:53
openstackgerritVipin Balachandran proposed openstack/cinder: VMware: Unit test refactoring (image to vol - 2/2)  https://review.openstack.org/24917512:53
openstackgerritVipin Balachandran proposed openstack/cinder: VMware: Replace mox with mock  https://review.openstack.org/24979712:56
*** alonma has quit IRC12:56
*** haomaiwa_ has joined #openstack-cinder12:57
*** alonma has joined #openstack-cinder12:58
dulekShould I expect that migration between two LVM hosts works?13:00
dulekIn my fresh env it fails with "iscsiadm ['-m', 'session']: stdout= stderr=iscsiadm: No active sessions.".13:00
dulekI'm not sure if it's broken in current master or what…?13:01
*** salv-orlando has quit IRC13:02
*** alonma has quit IRC13:03
*** alonma has joined #openstack-cinder13:04
*** salv-orlando has joined #openstack-cinder13:06
*** zhipeng has joined #openstack-cinder13:06
*** zhangjn_ has quit IRC13:08
*** alonma has quit IRC13:08
*** ociuhandu has quit IRC13:09
*** alonma has joined #openstack-cinder13:11
*** zhangjn has joined #openstack-cinder13:14
*** zhangjn has quit IRC13:15
*** alonma has quit IRC13:15
*** salv-orlando has quit IRC13:15
*** [2]Thelo has quit IRC13:15
*** zhangjn has joined #openstack-cinder13:15
*** zhangjn has quit IRC13:16
*** zhangjn has joined #openstack-cinder13:17
*** alonma has joined #openstack-cinder13:17
*** zhangjn has quit IRC13:17
*** [1]Thelo has joined #openstack-cinder13:18
*** zhangjn has joined #openstack-cinder13:18
*** zhangjn has quit IRC13:18
*** zhangjn has joined #openstack-cinder13:19
*** genunix_ has joined #openstack-cinder13:19
genunix_Hello, does anyone know what's the status of implementation of this blueprint? https://blueprints.launchpad.net/cinder/+spec/bare-metal-volumes13:20
*** alonma has quit IRC13:21
genunix_It would satisfy our use-cases so we would like to possibly participate on testing and development of using cinder for bare-metal.13:22
*** alonma has joined #openstack-cinder13:23
*** alonma has quit IRC13:27
dulekgenunix_: Have you seen BlockDeviceDriver?13:27
dulekgenunix_: I believe it aims to solve this.13:28
dulekgenunix_: It has some major issues, but e0ne and Yuriy Nesenenko (not sure on his IRC nick) are working to make it usable.13:29
*** alonma has joined #openstack-cinder13:29
dulekBTW - my issues with migration are solved, needed to tweak my_ip in cinder.conf.13:30
*** jungleboyj has quit IRC13:32
*** dilip has joined #openstack-cinder13:33
*** itzdilip has joined #openstack-cinder13:33
genunix_dulek: thank you for pointing me, it seems promising. Our use-case is basically about using Cinder separately from Openstack infrastructure only to maintain volumes for physical nodes.13:33
*** alonma has quit IRC13:33
dulekgenunix_: So right now this driver is race-condition prone. If I recall correctly can happen that two clients will reserve same physical device.13:34
*** diablo_rojo has joined #openstack-cinder13:35
dulekgenunix_: I've explained the details somewhere in this review: https://review.openstack.org/22229213:35
*** markvoelker has joined #openstack-cinder13:36
genunix_dulek: ad. our use-case, it seems that it will also require a little bit more than the driver. Eg. how to use cinder for attaching volumes to physical nodes (which are not nova or ironic managed or anyway part of openstack infrastructure)?13:38
*** alonma has joined #openstack-cinder13:39
*** markvoelker_ has joined #openstack-cinder13:41
*** houming has quit IRC13:42
*** markvoelker has quit IRC13:42
genunix_basically it would be similar setup as eg. Swift which is also possible to setup standalone.13:42
*** alonma has quit IRC13:43
*** zhipeng has quit IRC13:44
*** edmondsw has joined #openstack-cinder13:46
*** alonma has joined #openstack-cinder13:46
*** alonma has quit IRC13:51
*** alonma has joined #openstack-cinder13:52
*** links has quit IRC13:53
*** arnewiebalck has joined #openstack-cinder13:55
dulekgenunix_: Sorry, manager called.13:56
dulekgenunix_: There are some efforts to make it possible to attach volumes without Nova.13:57
*** alonma has quit IRC13:57
dulekgenunix_: I think e0ne and jgriffith are contact points on that.13:57
genunix_dulek: I tried to contact jgriffith with no success so far :-/13:58
dulekgenunix_: I believe he's in PST timezone, so you need to13:59
*** alonma has joined #openstack-cinder13:59
*** arnewiebalck_ has quit IRC13:59
dulekgenunix_: Sorry, you need to take that into account.13:59
*** akerr has joined #openstack-cinder14:00
*** timcl has joined #openstack-cinder14:01
*** dims has joined #openstack-cinder14:01
*** eharney has joined #openstack-cinder14:02
*** porrua has joined #openstack-cinder14:02
*** jungleboyj has joined #openstack-cinder14:03
*** alonma has quit IRC14:03
*** yhayashi has quit IRC14:03
*** bswartz has quit IRC14:04
*** breitz has quit IRC14:05
*** alonma has joined #openstack-cinder14:05
*** breitz has joined #openstack-cinder14:05
*** pradipta has joined #openstack-cinder14:07
*** alonma has quit IRC14:10
*** jerrygb has joined #openstack-cinder14:10
*** zhangjn has quit IRC14:10
*** alonma has joined #openstack-cinder14:11
*** zhangjn has joined #openstack-cinder14:11
itzdilipMy location: Alappuzha, India14:12
*** zhangjn has quit IRC14:12
*** haomaiwa_ has quit IRC14:13
DuncanTgenunix_: e0ne has a draft brick client on github14:13
*** itzdilip has quit IRC14:13
DuncanTgenunix_: There isn't actually much/anything on the cinder side, unless you want to work without keystone14:13
*** dilip has quit IRC14:13
*** arnewiebalck has quit IRC14:14
*** dilip has joined #openstack-cinder14:14
*** itzdilip has joined #openstack-cinder14:14
*** e0ne has joined #openstack-cinder14:14
*** arnewiebalck has joined #openstack-cinder14:14
*** zhangjn has joined #openstack-cinder14:15
*** zhangjn has quit IRC14:15
*** zhangjn has joined #openstack-cinder14:16
*** alonma has quit IRC14:16
*** alonma has joined #openstack-cinder14:17
*** ociuhandu has joined #openstack-cinder14:21
*** itzdilip has quit IRC14:22
*** alonma has quit IRC14:22
*** dilip has quit IRC14:22
*** e0ne_ has joined #openstack-cinder14:23
*** jwcroppe has quit IRC14:23
*** alonma has joined #openstack-cinder14:23
*** e0ne has quit IRC14:24
*** mriedem has joined #openstack-cinder14:25
*** haomaiwang has joined #openstack-cinder14:25
*** alonma has quit IRC14:28
*** abehl has quit IRC14:28
*** houming has joined #openstack-cinder14:29
*** jgregor has joined #openstack-cinder14:29
*** alonma has joined #openstack-cinder14:29
*** zhangjn has quit IRC14:31
*** houming has quit IRC14:32
*** arnewiebalck_ has joined #openstack-cinder14:33
*** xyang has joined #openstack-cinder14:33
*** alonma has quit IRC14:34
*** arnewiebalck__ has joined #openstack-cinder14:34
*** xyang has quit IRC14:35
*** arnewiebalck has quit IRC14:36
*** arnewiebalck__ is now known as arnewiebalck14:36
*** arnewiebalck_ has quit IRC14:37
*** zhangjn has joined #openstack-cinder14:38
*** zhangjn has quit IRC14:38
*** xyang has joined #openstack-cinder14:38
*** zhangjn has joined #openstack-cinder14:39
*** alonma has joined #openstack-cinder14:39
*** zhangjn has quit IRC14:41
*** alonma has quit IRC14:41
*** alonma has joined #openstack-cinder14:41
genunix_DuncanT: thanks, that brick thing seems to be what I am looking for :-)14:44
*** baumann has joined #openstack-cinder14:45
*** houming has joined #openstack-cinder14:47
DuncanTgenunix_: Great. Bug reports (or success reports) welcome14:48
*** adrianofr has joined #openstack-cinder14:50
*** bswartz has joined #openstack-cinder14:51
*** arnewiebalck_ has joined #openstack-cinder14:55
*** ntpttr has joined #openstack-cinder14:55
*** arnewiebalck has quit IRC14:57
*** arnewiebalck has joined #openstack-cinder14:58
*** julim has quit IRC15:00
*** fergal has joined #openstack-cinder15:00
*** divab has quit IRC15:00
*** arnewiebalck_ has quit IRC15:01
*** EinstCra_ has joined #openstack-cinder15:01
*** jordanP has quit IRC15:03
*** abhi has joined #openstack-cinder15:03
*** dustins has joined #openstack-cinder15:04
*** EinstCrazy has quit IRC15:04
*** salv-orlando has joined #openstack-cinder15:06
*** fmccrthy has joined #openstack-cinder15:07
*** leo__ has quit IRC15:08
*** fergal has quit IRC15:08
*** zhipeng has joined #openstack-cinder15:08
*** lprice has quit IRC15:09
*** timcl has quit IRC15:09
*** fmccrthy has quit IRC15:14
*** salv-orl_ has joined #openstack-cinder15:14
*** fmccrthy has joined #openstack-cinder15:14
*** fmccrthy has quit IRC15:15
*** haomaiwang has quit IRC15:15
*** delattec has joined #openstack-cinder15:16
*** fmccrthy has joined #openstack-cinder15:17
*** salv-orlando has quit IRC15:18
openstackgerritTom Hancock proposed openstack/cinder: Fix dictionary key error  https://review.openstack.org/25245115:19
*** delatte has joined #openstack-cinder15:19
*** delattec has quit IRC15:19
*** cdelatte has quit IRC15:19
*** jordanP has joined #openstack-cinder15:21
*** timcl has joined #openstack-cinder15:21
*** rhedlind has quit IRC15:22
*** zhipeng has quit IRC15:23
*** ildikov has quit IRC15:23
*** rhedlind has joined #openstack-cinder15:24
*** liverpooler has quit IRC15:24
openstackgerritMikhail Khodos proposed openstack/cinder: Re-add Nexenta driver  https://review.openstack.org/18887715:24
*** mc_nair has joined #openstack-cinder15:25
*** arnewiebalck has quit IRC15:27
*** arnewiebalck has joined #openstack-cinder15:28
*** markvoelker_ has quit IRC15:28
*** klkumar has joined #openstack-cinder15:29
*** weihan has quit IRC15:30
openstackgerritMichal Dulko proposed openstack/cinder: Switch request_spec in create volume calls to ovo  https://review.openstack.org/24660015:33
openstackgerritMichal Dulko proposed openstack/cinder: OVO request_spec in migrate/manage/retype calls  https://review.openstack.org/25245815:33
scottdakmartin: hemnafk https://review.openstack.org/#/c/252451/15:34
*** ChanServ has quit IRC15:34
*** Thambirajan has quit IRC15:34
*** mdbooth has quit IRC15:34
*** grumpycatt has quit IRC15:34
*** hogepodge has quit IRC15:34
*** thingee has quit IRC15:34
*** jungleboyj has quit IRC15:34
*** diablo_rojo has quit IRC15:34
*** gouthamr has quit IRC15:34
*** egonzalez has quit IRC15:34
*** erlon has quit IRC15:34
*** vgridnev has quit IRC15:34
*** dzamboni has quit IRC15:34
*** markus_z has quit IRC15:34
*** nkrinner has quit IRC15:34
*** jiangliucheng has quit IRC15:34
*** changbl has quit IRC15:34
*** diogogmt has quit IRC15:34
*** NightKhaos has quit IRC15:34
*** zz_john5223 has quit IRC15:34
*** krtaylor has quit IRC15:34
*** tsufiev has quit IRC15:34
*** stevemar has quit IRC15:34
*** kaisers has quit IRC15:34
*** sileht has quit IRC15:34
*** chlong has quit IRC15:34
*** xek has quit IRC15:34
*** amit213 has quit IRC15:34
*** Fdaisuke has quit IRC15:34
*** mkoderer has quit IRC15:34
*** swebb has quit IRC15:34
*** mgagne has quit IRC15:34
*** timcl has quit IRC15:34
*** baumann has quit IRC15:34
*** alonma has quit IRC15:34
*** jgregor has quit IRC15:34
*** e0ne_ has quit IRC15:34
*** jerrygb has quit IRC15:34
*** breitz has quit IRC15:34
*** klkumar has quit IRC15:34
*** aix has quit IRC15:34
*** alexschm has quit IRC15:34
*** david-lyle has quit IRC15:34
*** lixiaoy1 has quit IRC15:34
*** dulek has quit IRC15:34
*** rhagarty has quit IRC15:34
*** mdavidson has quit IRC15:34
*** ankit_ag has quit IRC15:34
*** cbader has quit IRC15:34
*** jwang has quit IRC15:34
*** bluex has quit IRC15:35
*** aorourke has quit IRC15:35
*** anteaya has quit IRC15:35
*** thrawn01 has quit IRC15:35
*** _fortis has quit IRC15:35
*** dosaboy has quit IRC15:35
*** SamYaple has quit IRC15:35
*** rmstar has quit IRC15:35
*** mordred has quit IRC15:35
*** afazekas has quit IRC15:35
*** winston-d has quit IRC15:35
*** genunix_ has quit IRC15:35
*** kfox1111 has quit IRC15:35
*** lifeless has quit IRC15:35
*** pck has quit IRC15:35
*** raildo has quit IRC15:35
*** leseb_ has quit IRC15:35
*** ccesario has quit IRC15:35
*** kbyrne has quit IRC15:35
*** eikke has quit IRC15:35
*** jgriffith has quit IRC15:35
*** cburgess has quit IRC15:35
*** pocketprotector has quit IRC15:35
*** _amrith_ has quit IRC15:35
*** rmk has quit IRC15:35
*** fnordahl has quit IRC15:35
*** smcginnis has quit IRC15:35
*** geguileo has quit IRC15:35
*** mtreinish has quit IRC15:35
*** zzzeek has quit IRC15:35
*** bapalm has quit IRC15:35
*** jistr has quit IRC15:35
*** anish has quit IRC15:35
*** guitarzan has quit IRC15:35
*** dansmith has quit IRC15:35
*** zhiyan has quit IRC15:35
*** ekarlso has quit IRC15:35
*** jbernard has quit IRC15:35
*** noqa_v_g1ovnie has quit IRC15:35
*** DuncanT has quit IRC15:35
*** cppforlife_ has quit IRC15:35
*** zul has quit IRC15:35
*** cvstealt1 has quit IRC15:35
*** scottda has quit IRC15:35
*** wanghao has quit IRC15:35
*** onder has quit IRC15:35
*** rushiagr_away has quit IRC15:35
*** patrickeast has quit IRC15:35
*** sweston has quit IRC15:35
*** ameade has quit IRC15:35
*** tpeoples has quit IRC15:35
*** rhefner has quit IRC15:35
*** briancurtin has quit IRC15:35
*** arnewiebalck has quit IRC15:35
*** rhedlind has quit IRC15:35
*** jordanP has quit IRC15:35
*** delatte has quit IRC15:35
*** salv-orl_ has quit IRC15:35
*** EinstCra_ has quit IRC15:35
*** ntpttr has quit IRC15:35
*** houming has quit IRC15:35
*** xyang has quit IRC15:35
*** mriedem has quit IRC15:35
*** ociuhandu has quit IRC15:35
*** porrua has quit IRC15:35
*** dims has quit IRC15:35
*** akerr has quit IRC15:35
*** [1]Thelo has quit IRC15:35
*** gcb has quit IRC15:35
*** zhonghua-lee has quit IRC15:35
*** lyarwood has quit IRC15:35
*** jaypipes has quit IRC15:35
*** skraynev has quit IRC15:35
*** asselin has quit IRC15:35
*** frakt has quit IRC15:35
*** coolsvap has quit IRC15:35
*** markstur has quit IRC15:35
*** krotscheck has quit IRC15:35
*** Tridde has quit IRC15:35
*** Lickitysplitted_ has quit IRC15:35
*** SergeyLukjanov has quit IRC15:35
*** j_king_ has quit IRC15:35
*** ir2ivps7__ has quit IRC15:35
*** ericksonfgds has quit IRC15:35
*** kragniz has quit IRC15:35
*** sb- has quit IRC15:35
*** rwall has quit IRC15:35
*** cebruns has quit IRC15:35
*** nug- has quit IRC15:35
*** boichev2 has quit IRC15:35
*** aarefiev has quit IRC15:35
*** hemnafk has quit IRC15:35
*** HenryG has quit IRC15:35
*** sbadia has quit IRC15:35
*** zigo has quit IRC15:35
*** Nakato has quit IRC15:35
*** logan- has quit IRC15:35
*** ByPasS has quit IRC15:35
*** ircuser-1 has quit IRC15:35
*** wN has quit IRC15:35
*** Swanson has quit IRC15:35
*** EmilienM has quit IRC15:35
*** notmorgan has quit IRC15:35
*** netapp has quit IRC15:35
*** med_ has quit IRC15:35
*** wolsen has quit IRC15:35
*** kambiz has quit IRC15:35
*** Anticimex has quit IRC15:35
*** toabctl has quit IRC15:35
*** nug has quit IRC15:35
*** doffm has quit IRC15:35
*** tobasco has quit IRC15:35
*** bswartz has quit IRC15:35
*** adrianofr has quit IRC15:35
*** ndipanov has quit IRC15:35
*** theanalyst has quit IRC15:35
*** serverascode has quit IRC15:35
*** johnthetubaguy has quit IRC15:35
*** jamielennox has quit IRC15:35
*** dobson has quit IRC15:35
*** sparr has quit IRC15:35
*** notmorgan has joined #openstack-cinder15:36
*** alonma has joined #openstack-cinder15:37
*** jgriffith has joined #openstack-cinder15:37
*** cburgess has joined #openstack-cinder15:37
*** pocketprotector has joined #openstack-cinder15:37
*** _amrith_ has joined #openstack-cinder15:37
*** rmk has joined #openstack-cinder15:37
*** fnordahl has joined #openstack-cinder15:37
openstackgerritEric Harney proposed openstack/cinder: LIO: Handle initiator IQNs as case insensitive  https://review.openstack.org/25246015:37
*** sparr has joined #openstack-cinder15:38
*** notmorgan has quit IRC15:38
*** adelia has joined #openstack-cinder15:38
*** jungleboyj has joined #openstack-cinder15:39
*** diablo_rojo has joined #openstack-cinder15:39
*** gouthamr has joined #openstack-cinder15:39
*** egonzalez has joined #openstack-cinder15:39
*** erlon has joined #openstack-cinder15:39
*** vgridnev has joined #openstack-cinder15:39
*** dzamboni has joined #openstack-cinder15:39
*** markus_z has joined #openstack-cinder15:39
*** nkrinner has joined #openstack-cinder15:39
*** jiangliucheng has joined #openstack-cinder15:39
*** changbl has joined #openstack-cinder15:39
*** diogogmt has joined #openstack-cinder15:39
*** NightKhaos has joined #openstack-cinder15:39
*** zz_john5223 has joined #openstack-cinder15:39
*** krtaylor has joined #openstack-cinder15:39
*** tsufiev has joined #openstack-cinder15:39
*** stevemar has joined #openstack-cinder15:39
*** kaisers has joined #openstack-cinder15:39
*** sileht has joined #openstack-cinder15:39
*** chlong has joined #openstack-cinder15:39
*** xek has joined #openstack-cinder15:39
*** amit213 has joined #openstack-cinder15:39
*** Fdaisuke has joined #openstack-cinder15:39
*** mkoderer has joined #openstack-cinder15:39
*** swebb has joined #openstack-cinder15:39
*** mgagne has joined #openstack-cinder15:39
*** jistr has joined #openstack-cinder15:39
*** anish has joined #openstack-cinder15:39
*** guitarzan has joined #openstack-cinder15:39
*** zhiyan has joined #openstack-cinder15:39
*** ekarlso has joined #openstack-cinder15:39
*** jbernard has joined #openstack-cinder15:39
*** noqa_v_g1ovnie has joined #openstack-cinder15:39
*** DuncanT has joined #openstack-cinder15:39
*** cppforlife_ has joined #openstack-cinder15:39
*** zul has joined #openstack-cinder15:39
*** cvstealt1 has joined #openstack-cinder15:39
*** scottda has joined #openstack-cinder15:39
*** wanghao has joined #openstack-cinder15:39
*** onder has joined #openstack-cinder15:39
*** rushiagr_away has joined #openstack-cinder15:39
*** mdbooth has joined #openstack-cinder15:39
*** grumpycatt has joined #openstack-cinder15:39
*** dansmith has joined #openstack-cinder15:39
*** nug has joined #openstack-cinder15:40
*** zigo has joined #openstack-cinder15:40
*** Nakato has joined #openstack-cinder15:40
*** logan- has joined #openstack-cinder15:40
*** ByPasS has joined #openstack-cinder15:40
*** wN has joined #openstack-cinder15:40
*** Swanson has joined #openstack-cinder15:40
*** EmilienM has joined #openstack-cinder15:40
*** netapp has joined #openstack-cinder15:40
*** med_ has joined #openstack-cinder15:40
*** wolsen has joined #openstack-cinder15:40
*** kambiz has joined #openstack-cinder15:40
*** Anticimex has joined #openstack-cinder15:40
*** toabctl has joined #openstack-cinder15:40
*** doffm has joined #openstack-cinder15:40
*** tobasco has joined #openstack-cinder15:40
*** nug is now known as 7YUAAB7RZ15:40
*** alonma has quit IRC15:41
*** smcginnis has joined #openstack-cinder15:42
*** geguileo has joined #openstack-cinder15:42
*** mtreinish has joined #openstack-cinder15:42
*** zzzeek has joined #openstack-cinder15:42
*** bapalm has joined #openstack-cinder15:42
*** Thambirajan has joined #openstack-cinder15:42
*** patrickeast has joined #openstack-cinder15:42
*** sweston has joined #openstack-cinder15:42
*** ameade has joined #openstack-cinder15:42
*** tpeoples has joined #openstack-cinder15:42
*** rhefner has joined #openstack-cinder15:42
*** briancurtin has joined #openstack-cinder15:42
*** hogepodge has joined #openstack-cinder15:42
*** thingee has joined #openstack-cinder15:42
*** alonma has joined #openstack-cinder15:43
*** johnthetubaguy has joined #openstack-cinder15:44
*** bswartz has joined #openstack-cinder15:44
*** adrianofr has joined #openstack-cinder15:44
*** ndipanov has joined #openstack-cinder15:44
*** theanalyst has joined #openstack-cinder15:44
*** serverascode has joined #openstack-cinder15:44
*** jamielennox has joined #openstack-cinder15:44
*** dobson has joined #openstack-cinder15:44
*** gnarld_ has joined #openstack-cinder15:44
*** alonma has quit IRC15:47
*** alonma has joined #openstack-cinder15:49
wanghaoeharney: hi Eric, Could you do me a favor? Please help me to remove L freeze -2 in this patch:https://review.openstack.org/#/c/182276/.15:49
*** _fortis has joined #openstack-cinder15:49
*** ctina has joined #openstack-cinder15:51
eharneywanghao: that patch needs to hold at -2 for now anyway because the format detection is a security hole15:51
*** aix has joined #openstack-cinder15:51
wanghaoeharney: OK, got it. I will try to fix this.15:52
*** alonma has quit IRC15:53
*** markvoelker has joined #openstack-cinder15:55
*** xyang has joined #openstack-cinder15:56
openstackgerritNate Potter proposed openstack/cinder: Use the catalog to retrieve glance_api_servers  https://review.openstack.org/25247615:56
openstackgerritMichael Price proposed openstack/cinder: Implement CHAP Authentication for E-Series Driver  https://review.openstack.org/24699015:57
*** ircuser-1 has joined #openstack-cinder15:59
*** crose has joined #openstack-cinder15:59
*** klkumar has joined #openstack-cinder15:59
*** timcl has joined #openstack-cinder15:59
*** baumann has joined #openstack-cinder15:59
*** jgregor has joined #openstack-cinder15:59
*** e0ne_ has joined #openstack-cinder15:59
*** jerrygb has joined #openstack-cinder15:59
*** breitz has joined #openstack-cinder15:59
*** genunix_ has joined #openstack-cinder15:59
*** alexschm has joined #openstack-cinder15:59
*** david-lyle has joined #openstack-cinder15:59
*** lixiaoy1 has joined #openstack-cinder15:59
*** dulek has joined #openstack-cinder15:59
*** rhagarty has joined #openstack-cinder15:59
*** mdavidson has joined #openstack-cinder15:59
*** ankit_ag has joined #openstack-cinder15:59
*** cbader has joined #openstack-cinder15:59
*** eikke has joined #openstack-cinder15:59
*** jwang has joined #openstack-cinder15:59
*** bluex has joined #openstack-cinder15:59
*** aorourke has joined #openstack-cinder15:59
*** anteaya has joined #openstack-cinder15:59
*** thrawn01 has joined #openstack-cinder15:59
*** dosaboy has joined #openstack-cinder15:59
*** SamYaple has joined #openstack-cinder15:59
*** rmstar has joined #openstack-cinder15:59
*** mordred has joined #openstack-cinder15:59
*** winston-d has joined #openstack-cinder15:59
*** kfox1111 has joined #openstack-cinder15:59
*** lifeless has joined #openstack-cinder15:59
*** pck has joined #openstack-cinder15:59
*** raildo has joined #openstack-cinder15:59
*** ccesario has joined #openstack-cinder15:59
*** leseb_ has joined #openstack-cinder15:59
*** kbyrne has joined #openstack-cinder15:59
openstackgerritNate Potter proposed openstack/cinder: Use the catalog to retrieve glance_api_servers  https://review.openstack.org/25247615:59
*** rcernin has quit IRC15:59
*** zul has quit IRC16:00
*** pots has joined #openstack-cinder16:00
*** SergeyLukjanov has joined #openstack-cinder16:01
*** dustins has quit IRC16:01
DuncanTAre we meeting today?16:03
*** kragniz has joined #openstack-cinder16:03
*** dzamboni has quit IRC16:03
erlonanyone here?16:04
*** jseiler has joined #openstack-cinder16:04
*** ctina has quit IRC16:07
*** ctina has joined #openstack-cinder16:07
openstackgerritThang Pham proposed openstack/cinder: Update migrate_volume API to use versionedobjects  https://review.openstack.org/22891616:09
openstackgerritThang Pham proposed openstack/cinder: Update initialize_connection to use versionedobjects  https://review.openstack.org/23889416:09
*** ircuser-1 has quit IRC16:09
*** crose has quit IRC16:09
*** timcl has quit IRC16:09
*** baumann has quit IRC16:09
*** jgregor has quit IRC16:09
*** e0ne_ has quit IRC16:09
*** jerrygb has quit IRC16:09
*** breitz has quit IRC16:09
*** klkumar has quit IRC16:09
*** alexschm has quit IRC16:09
*** david-lyle has quit IRC16:09
*** lixiaoy1 has quit IRC16:09
*** dulek has quit IRC16:09
*** rhagarty has quit IRC16:09
*** mdavidson has quit IRC16:09
*** ankit_ag has quit IRC16:09
*** cbader has quit IRC16:09
*** jwang has quit IRC16:09
*** bluex has quit IRC16:09
*** aorourke has quit IRC16:09
*** anteaya has quit IRC16:09
*** thrawn01 has quit IRC16:09
*** dosaboy has quit IRC16:09
*** SamYaple has quit IRC16:09
*** rmstar has quit IRC16:09
*** mordred has quit IRC16:09
*** winston-d has quit IRC16:09
*** genunix_ has quit IRC16:09
*** kfox1111 has quit IRC16:09
*** lifeless has quit IRC16:09
*** pck has quit IRC16:09
*** raildo has quit IRC16:09
*** leseb_ has quit IRC16:09
*** ccesario has quit IRC16:09
*** kbyrne has quit IRC16:09
*** eikke has quit IRC16:09
openstackgerritThang Pham proposed openstack/cinder: Update replication to use versionedobjects  https://review.openstack.org/23890616:09
*** baumann1 has joined #openstack-cinder16:09
*** rajinir_ has joined #openstack-cinder16:10
*** Erlon_ has joined #openstack-cinder16:12
*** alonma has joined #openstack-cinder16:14
*** jaypipes has joined #openstack-cinder16:14
openstackgerritMerged openstack/cinder: Imported Translations from Zanata  https://review.openstack.org/25220916:15
*** EinstCrazy has joined #openstack-cinder16:16
*** diogogmt has quit IRC16:17
*** EinstCrazy has quit IRC16:18
*** alonma has quit IRC16:19
openstackgerritShay Halsband proposed openstack/cinder: XtremIO: fix generic glance cache with XtremIO  https://review.openstack.org/24981416:19
flip214hogepodge: what needs to be done to get included on the marketplace driver list?16:24
flip214hogepodge: we're passing since ~5 months or so ... the DRBD driver is my concern.16:25
flip214hogepodge: thanks for the quick answer, BTW!16:25
hogepodgeflip214: np, these links might help http://www.openstack.org/brand/interop/16:26
hogepodgeflip214: http://www.openstack.org/brand/openstack-compatible/16:27
*** p0rtal has joined #openstack-cinder16:27
flip214hogepodge: thanks, reading.16:28
flip214does that talk about how to get on https://www.openstack.org/marketplace/drivers/, too?16:28
hogepodgeThe marketplace home page has the contact information for requesting a product page in the marketplace http://www.openstack.org/marketplace16:30
hogepodgehttp://www.openstack.org/assets/marketplace/join-the-marketplace.pdf16:30
flip214thanks for all the pointers.... will read and come back if there are any more questions!16:31
flip214thanks a lot!16:31
*** alonma has joined #openstack-cinder16:33
*** sbadia has joined #openstack-cinder16:34
*** HenryG has joined #openstack-cinder16:34
*** hemna has joined #openstack-cinder16:34
*** aarefiev has joined #openstack-cinder16:34
*** boichev2 has joined #openstack-cinder16:34
*** cebruns has joined #openstack-cinder16:34
*** rwall has joined #openstack-cinder16:34
*** sb- has joined #openstack-cinder16:34
*** ericksonsantos has joined #openstack-cinder16:34
*** ir2ivps7__ has joined #openstack-cinder16:34
*** j_king_ has joined #openstack-cinder16:34
*** Lickitysplitted_ has joined #openstack-cinder16:34
*** Tridde has joined #openstack-cinder16:34
*** krotscheck has joined #openstack-cinder16:34
*** markstur has joined #openstack-cinder16:34
*** coolsvap has joined #openstack-cinder16:34
*** frakt has joined #openstack-cinder16:34
*** skraynev has joined #openstack-cinder16:34
*** lyarwood has joined #openstack-cinder16:34
*** zhonghua-lee has joined #openstack-cinder16:34
*** gcb has joined #openstack-cinder16:34
*** [1]Thelo has joined #openstack-cinder16:34
*** akerr has joined #openstack-cinder16:34
*** ociuhandu has joined #openstack-cinder16:34
*** mriedem has joined #openstack-cinder16:34
*** ntpttr has joined #openstack-cinder16:34
*** salv-orl_ has joined #openstack-cinder16:34
*** delatte has joined #openstack-cinder16:34
*** jordanP has joined #openstack-cinder16:34
*** arnewiebalck has joined #openstack-cinder16:34
*** notmorgan has joined #openstack-cinder16:34
*** afazekas has joined #openstack-cinder16:34
*** rs007 has joined #openstack-cinder16:34
*** timcl has joined #openstack-cinder16:34
*** lprice has joined #openstack-cinder16:34
*** cbader has joined #openstack-cinder16:34
*** zul has joined #openstack-cinder16:34
*** rhedlind has joined #openstack-cinder16:34
*** mtanino has joined #openstack-cinder16:34
*** Yogi1 has joined #openstack-cinder16:34
*** ircuser-1 has joined #openstack-cinder16:34
*** crose has joined #openstack-cinder16:34
*** klkumar has joined #openstack-cinder16:34
*** jgregor has joined #openstack-cinder16:34
*** jerrygb has joined #openstack-cinder16:34
*** breitz has joined #openstack-cinder16:34
*** genunix_ has joined #openstack-cinder16:34
*** alexschm has joined #openstack-cinder16:34
*** david-lyle has joined #openstack-cinder16:34
*** lixiaoy1 has joined #openstack-cinder16:34
*** dulek has joined #openstack-cinder16:34
*** rhagarty has joined #openstack-cinder16:34
*** mdavidson has joined #openstack-cinder16:34
*** ankit_ag has joined #openstack-cinder16:34
*** eikke has joined #openstack-cinder16:34
*** jwang has joined #openstack-cinder16:34
*** aorourke has joined #openstack-cinder16:34
*** anteaya has joined #openstack-cinder16:34
*** thrawn01 has joined #openstack-cinder16:34
*** dosaboy has joined #openstack-cinder16:34
*** SamYaple has joined #openstack-cinder16:34
*** rmstar has joined #openstack-cinder16:34
*** mordred has joined #openstack-cinder16:34
*** winston-d has joined #openstack-cinder16:34
*** kfox1111 has joined #openstack-cinder16:34
*** lifeless has joined #openstack-cinder16:34
*** pck has joined #openstack-cinder16:34
*** raildo has joined #openstack-cinder16:34
*** ccesario has joined #openstack-cinder16:34
*** leseb_ has joined #openstack-cinder16:34
*** kbyrne has joined #openstack-cinder16:34
*** alonma has quit IRC16:37
*** cbader_ has joined #openstack-cinder16:41
*** lpetrut has quit IRC16:41
*** cbader has quit IRC16:45
*** zul has quit IRC16:45
*** rhedlind has quit IRC16:45
*** mtanino has quit IRC16:45
*** Yogi1 has quit IRC16:45
*** ircuser-1 has quit IRC16:45
*** crose has quit IRC16:45
*** jgregor has quit IRC16:45
*** jerrygb has quit IRC16:45
*** breitz has quit IRC16:45
*** klkumar has quit IRC16:45
*** alexschm has quit IRC16:45
*** david-lyle has quit IRC16:45
*** lixiaoy1 has quit IRC16:45
*** dulek has quit IRC16:45
*** rhagarty has quit IRC16:45
*** mdavidson has quit IRC16:45
*** ankit_ag has quit IRC16:45
*** jwang has quit IRC16:45
*** aorourke has quit IRC16:45
*** anteaya has quit IRC16:45
*** thrawn01 has quit IRC16:45
*** dosaboy has quit IRC16:45
*** SamYaple has quit IRC16:45
*** rmstar has quit IRC16:45
*** mordred has quit IRC16:45
*** winston-d has quit IRC16:45
*** genunix_ has quit IRC16:45
*** kfox1111 has quit IRC16:45
*** lifeless has quit IRC16:45
*** pck has quit IRC16:45
*** raildo has quit IRC16:45
*** leseb_ has quit IRC16:45
*** ccesario has quit IRC16:45
*** kbyrne has quit IRC16:45
*** eikke has quit IRC16:45
*** lprice has quit IRC16:45
*** timcl has quit IRC16:45
*** rs007 has quit IRC16:45
*** afazekas has quit IRC16:45
*** notmorgan has quit IRC16:45
*** arnewiebalck has quit IRC16:45
*** jordanP has quit IRC16:45
*** delatte has quit IRC16:45
*** salv-orl_ has quit IRC16:45
*** ntpttr has quit IRC16:45
*** mriedem has quit IRC16:45
*** ociuhandu has quit IRC16:45
*** akerr has quit IRC16:45
*** [1]Thelo has quit IRC16:45
*** gcb has quit IRC16:45
*** zhonghua-lee has quit IRC16:45
*** lyarwood has quit IRC16:45
*** skraynev has quit IRC16:45
*** frakt has quit IRC16:45
*** coolsvap has quit IRC16:45
*** markstur has quit IRC16:45
*** krotscheck has quit IRC16:45
*** Tridde has quit IRC16:45
*** Lickitysplitted_ has quit IRC16:45
*** j_king_ has quit IRC16:45
*** ir2ivps7__ has quit IRC16:45
*** ericksonsantos has quit IRC16:45
*** sb- has quit IRC16:45
*** rwall has quit IRC16:45
*** cebruns has quit IRC16:45
*** boichev2 has quit IRC16:45
*** aarefiev has quit IRC16:45
*** hemna has quit IRC16:45
*** HenryG has quit IRC16:45
*** sbadia has quit IRC16:45
*** pradipta has quit IRC16:45
thingeeflip214: that's a good question.16:46
*** DuncanT1 has joined #openstack-cinder16:46
flip214I'm trying to sneak it in via os-brick now16:46
*** hemna has joined #openstack-cinder16:46
*** kbyrne has joined #openstack-cinder16:46
*** leseb_ has joined #openstack-cinder16:46
*** ccesario has joined #openstack-cinder16:46
*** raildo has joined #openstack-cinder16:46
*** pck has joined #openstack-cinder16:46
*** lifeless has joined #openstack-cinder16:46
*** kfox1111 has joined #openstack-cinder16:46
*** winston-d has joined #openstack-cinder16:46
*** mordred has joined #openstack-cinder16:46
*** rmstar has joined #openstack-cinder16:46
*** SamYaple has joined #openstack-cinder16:46
*** dosaboy has joined #openstack-cinder16:46
*** thrawn01 has joined #openstack-cinder16:46
*** anteaya has joined #openstack-cinder16:46
*** aorourke has joined #openstack-cinder16:46
*** jwang has joined #openstack-cinder16:46
*** eikke has joined #openstack-cinder16:46
*** ankit_ag has joined #openstack-cinder16:46
*** mdavidson has joined #openstack-cinder16:46
*** rhagarty has joined #openstack-cinder16:46
*** dulek has joined #openstack-cinder16:46
*** lixiaoy1 has joined #openstack-cinder16:46
*** david-lyle has joined #openstack-cinder16:46
*** alexschm has joined #openstack-cinder16:46
*** genunix_ has joined #openstack-cinder16:46
*** breitz has joined #openstack-cinder16:46
*** jerrygb has joined #openstack-cinder16:46
*** jgregor has joined #openstack-cinder16:46
*** klkumar has joined #openstack-cinder16:46
*** crose has joined #openstack-cinder16:46
*** ircuser-1 has joined #openstack-cinder16:46
*** Yogi1 has joined #openstack-cinder16:46
*** mtanino has joined #openstack-cinder16:46
*** rhedlind has joined #openstack-cinder16:46
*** zul has joined #openstack-cinder16:46
*** lprice has joined #openstack-cinder16:46
*** timcl has joined #openstack-cinder16:46
*** rs007 has joined #openstack-cinder16:46
*** afazekas has joined #openstack-cinder16:46
*** notmorgan has joined #openstack-cinder16:46
*** arnewiebalck has joined #openstack-cinder16:46
*** jordanP has joined #openstack-cinder16:46
*** delatte has joined #openstack-cinder16:46
*** salv-orl_ has joined #openstack-cinder16:46
*** ntpttr has joined #openstack-cinder16:46
*** mriedem has joined #openstack-cinder16:46
*** ociuhandu has joined #openstack-cinder16:46
*** akerr has joined #openstack-cinder16:46
*** [1]Thelo has joined #openstack-cinder16:46
*** gcb has joined #openstack-cinder16:46
*** zhonghua-lee has joined #openstack-cinder16:46
*** lyarwood has joined #openstack-cinder16:46
*** skraynev has joined #openstack-cinder16:46
*** frakt has joined #openstack-cinder16:46
*** coolsvap has joined #openstack-cinder16:46
*** markstur has joined #openstack-cinder16:46
*** krotscheck has joined #openstack-cinder16:46
*** Tridde has joined #openstack-cinder16:46
*** Lickitysplitted_ has joined #openstack-cinder16:46
*** j_king_ has joined #openstack-cinder16:46
*** ir2ivps7__ has joined #openstack-cinder16:46
*** ericksonsantos has joined #openstack-cinder16:46
*** sb- has joined #openstack-cinder16:46
*** rwall has joined #openstack-cinder16:46
*** cebruns has joined #openstack-cinder16:46
*** boichev2 has joined #openstack-cinder16:46
*** aarefiev has joined #openstack-cinder16:46
*** HenryG has joined #openstack-cinder16:46
*** sbadia has joined #openstack-cinder16:46
flip214because that's where it should be implemented16:46
thingeeflip214: I haven't really thought of it for vendors. so far cross-project specs have been general ideas like quotas, or logging practices...16:47
flip214never mind, I'll just try to go on there.16:47
flip214thanks for listening, that eases the mind.16:47
flip214sorry, kids need taking care of16:47
*** rhedlind_ has joined #openstack-cinder16:48
*** philipw has joined #openstack-cinder16:48
*** aix has quit IRC16:48
*** aix has joined #openstack-cinder16:48
*** xyang has quit IRC16:48
*** xyang has joined #openstack-cinder16:48
*** baumann1 has quit IRC16:48
*** baumann1 has joined #openstack-cinder16:48
*** rajinir_ has quit IRC16:48
*** rajinir_ has joined #openstack-cinder16:48
*** Erlon_ has quit IRC16:48
*** Erlon_ has joined #openstack-cinder16:48
*** rhedlind_ has quit IRC16:48
*** rhedlind_ has joined #openstack-cinder16:48
*** ircuser-1 has quit IRC16:48
*** ildikov has joined #openstack-cinder16:48
hemnaflip214, ?16:48
*** salv-orl_ has quit IRC16:49
thingeeflip214: don't you still need some hook in nova with the os-brick piece?16:49
smcginnisscottda: Any response for the comments on here: https://review.openstack.org/#/c/23259916:49
hemnaguess I missed the first part of this conversation16:49
hemnaopenstack-meeting is empty for me as well16:49
hemnableh16:49
smcginnishemna: Freenode issues.16:49
DuncanT1hemna: freenode lost a hub and has been going nuts since16:49
smcginnisHopefully the bot captured it all.16:50
thingeehemna: he was wondering if drbd transport needs a cross-project spec.16:50
thingeehemna: since it touches nova and cinder16:50
DuncanT1I've got local logs from when I joined, if we need to reconstruct16:50
thingeehemna: haven't really considered the specs being used for vendor initiatives16:50
wanghaoit's terrible irc.....16:50
hemnathingee, ah ok.  drbd support has to land in brick first, but if there is no cinder driver for it....16:50
hemnasmcginnis, ah ok.16:50
hemna:(16:50
smcginnishemna: Looks like it caught it: http://eavesdrop.openstack.org/meetings/cinder/2015/cinder.2015-12-02-15.59.html16:51
markus_zsmcginnis: I've seen you do provide bug stats in the cinder meetings. Can I maybe interest you in this? https://review.openstack.org/#/c/250903/16:51
smcginnismarkus_z: Definitely interested. :)16:51
*** jordanP has quit IRC16:52
smcginnismarkus_z: I have some scripts I use that I was going to combine with openstack-infra reviewstats to publish here:16:52
smcginnishttp://cinderstats-dellstorage.rhcloud.com/16:52
smcginnismarkus_z: But right now I manually run them.16:52
hemnasmcginnis, we should probably do a build of os-brick as well no ?16:52
hemnakeeping with the client/lib releases on milestones mantra16:53
markus_zsmcginnis: Ah, look like the thing johnthetubaguy provides for nova :)16:53
markus_z*looks16:53
*** alexschm has quit IRC16:53
DuncanT1markus_z: Very similar, yes16:53
smcginnismarkus_z: Yep, both use the reviewstats stuff.16:53
smcginnisscottda pointed me to what John was doing for nova.16:53
smcginnisI like it.16:53
smcginnisNever hurts to have data. ;)16:53
* DuncanT1 is reminded to push a pull request to russel to get review-stats to count 0 reviews too16:53
markus_zyeah, true.16:54
smcginnishemna: Yeah, I've meant to check with you on that. Do you think it's a good time for it?16:54
smcginnisDuncanT1: I believe Russel is just using reviewstats too.16:54
smcginnisDuncanT1: I have those reports generated here: http://cinderstats-dellstorage.rhcloud.com/16:54
*** jordanP has joined #openstack-cinder16:54
markus_zsmcginnis: ok, cool, just wanted to make you aware of the thing I build. I'll keep you in the loop when it lands.16:55
hemnasmcginnis, there are a few reviews needing a final +216:55
smcginnisThe advantage of mine is that until my patch went through yesterday, it didn't have os-brick and it was looking for volume-api.16:55
smcginnishemna: I'll take a look.16:55
hemnawould be nice to get some other eyes on those16:55
smcginnismarkus_z: Awesome, thanks!16:55
DuncanT1smcginnis: Russel wrote/hosts reviewstats I believe16:55
smcginnisDuncanT1: Ah, cool.16:56
smcginnisDuncanT1: I think he should pick up my change then, if he's running that from the official repo.16:56
smcginnisDuncanT1: Waiting for a +A on this. Look OK to you? https://review.openstack.org/#/c/248383/16:58
smcginnisDuncanT1: Done it elsewhere already.16:58
*** rhedlind_ has quit IRC16:59
DuncanT1smcginnis: Done17:00
smcginnisDuncanT1: THanks!17:00
thingeewanghao: wrt cinder specs needing reviews, I just approved a bunch yesterday.17:01
*** rs007 has quit IRC17:01
thingeewanghao: might17:01
thingee've not been yours though =/17:01
*** jistr has quit IRC17:01
*** adurbin_ has joined #openstack-cinder17:03
smcginnisDuncanT1: Another easy one if you want: https://review.openstack.org/#/c/246023/17:03
smcginniswanghao: I do have a tab open.17:03
smcginnisWe do need to get some more of the specs moving along.17:04
DuncanT1smcginnis: Looking17:04
winston-dDuncanT1: oops, i've +A it17:04
*** ChanServ has joined #openstack-cinder17:04
*** wolfe.freenode.net sets mode: +o ChanServ17:04
smcginniswinston-d: :) Thanks!17:05
winston-dit's been on my gerrty for a while17:05
*** bluex has joined #openstack-cinder17:05
*** e0ne has joined #openstack-cinder17:05
*** apoorvad has joined #openstack-cinder17:05
smcginnishemna: That was quick. os-brick should be ready to go this afternoon hopefully.17:05
*** notmorgan is now known as Guest6639117:05
xyangjgriffith: ping17:05
hemnasmcginnis, ok rockin thanks17:05
jgriffithxyang: :)17:05
openstackgerritTom Hancock proposed openstack/cinder: Fix dictionary key error  https://review.openstack.org/25245117:05
jgriffithxyang: I  knew I'd be hearing from you soon17:06
xyangjgriffith: :)17:06
xyangjgriffith: so we do have a default implementation for drivers that doesn't attach snapshot17:06
xyangDuncanT1: we are talking about backup snapshots spec17:07
*** Guest66391 is now known as morganfainberg17:07
*** gnarld_ is now known as nug17:07
jgriffithyes, DuncanT1 pointed that out and I updated the comments17:07
*** morganfainberg has quit IRC17:07
*** morganfainberg has joined #openstack-cinder17:07
*** nug is now known as Guest6990317:07
*** morganfainberg is now known as notmorgan17:07
xyangjgriffith: I think we should still provide a way to backup a volume in-use, that is a use case too17:07
*** Yogi1 has left #openstack-cinder17:07
scottdasmcginnis: Comments on the A-A/HA spec? I'm not sure which comments you mean17:08
xyangjgriffith: maybe seems redundant, but from user's point of view, it is convenient to have both ways17:08
xyangjgriffith: they can do it in one step instead of two17:08
*** belmoreira has joined #openstack-cinder17:08
openstackgerritTom Hancock proposed openstack/cinder: Fix dictionary key error  https://review.openstack.org/25245117:09
DuncanT1scottda: there's the confusing bit of language around cleanup for starters17:09
*** andrew____ has joined #openstack-cinder17:10
*** Erlon_ has quit IRC17:10
xyangjgriffith: also if you take a look of the code change, it is small17:10
andrew____hi, im trying to add another custom attribute just to learn more about cinder. from what i understand, i would need to change the db, is that correct?17:10
jgriffithxyang: yes, I know it's a simple change, and probably a good one17:10
jgriffithxyang: I just wish we would only do snapshots and not volumes17:11
xyangjgriffith: I think backup in-use volume has its value too17:11
scottdaDuncanT1: smcginnis OK, I see the cleanup comments and they are vague. I'll put my 2cents in a comment.17:11
*** rcernin has joined #openstack-cinder17:11
jgriffithmaybe deprecate the existing volume-backup method and go forward with doing snaps only?17:11
jgriffithxyang: my issue is that we're creating a confusing API and user experience17:12
jgriffithxyang: and weird rules and code-paths17:12
xyangjgriffith: we do have volume backup to start with17:12
xyangjgriffith: deprecating may not be good user experience17:12
xyangjgriffith: they expect to backup an in-use volume too17:12
xyangjgriffith: I think the complexity is because we are finding the parent backup automatically for incremental, other than that, it doesn't seem complicated17:14
jgriffithxyang: yeah, I know you're not going to agree with me (and that's ok) :)17:14
*** julim has joined #openstack-cinder17:14
jgriffithxyang: I completely expect to be completely over-turned on this17:14
jgriffithxyang: I also made it a -1 and not a -2 only to voice my opinion but to NOT block anything17:15
*** markstur has quit IRC17:15
xyangjgriffith: I do hope to get your support on this though, a +1 rather than -1 or -2:)17:15
jgriffithxyang: I'd say it is somewhat complicated because we get into a maintenance case of "what is the code path/flow"17:15
openstackgerritGerald McBrearty proposed openstack/cinder: Add config option to control flash copy rate  https://review.openstack.org/24825017:15
jgriffithxyang: sometimes it's one way, sometimes another... and when it's attached it's a little of both17:15
xyangjgriffith: It does add more to the code.  I'll add more comments to clarify it17:16
jgriffithxyang: don't get me wrong, I think it's the right way to do it.  I just wish we did it ONE way for everything17:16
jgriffithand just focused on that17:16
xyangjgriffith: sure17:17
jgriffithxyang: when it's time to fix a bug or add an enhancement or whatever it sucks to have to try and remember to clean it up in multiple places in different ways17:17
Swansonjgriffith, Should there be an update_replications function?  In case someone changes their replication destinations?  Driver would tear down repls that are not listed on the backend any more and create those that are there now?  Thots?17:17
jgriffithxyang: multiple code-paths == badness IMO17:17
jgriffithSwanson: retype17:17
xyangjgriffith:  I'll add more comments and hopefully it is clear what is expected17:17
openstackgerritThang Pham proposed openstack/cinder: Implement refresh() for cinder objects  https://review.openstack.org/24899817:18
jgriffithxyang: thanks... I'll remove my vote and butt out of it :)17:18
Swansonjgriffith, Oh me on my.  I'll look into that.  Thanks!17:18
xyangjgriffith: thanks!17:18
openstackgerritEric Harney proposed openstack/cinder: LIO: Handle initiator IQNs as case insensitive  https://review.openstack.org/25246017:19
*** EinstCrazy has joined #openstack-cinder17:19
*** p0rtal has quit IRC17:20
jgriffithSwanson: :)17:20
jgriffithSwanson: that was the idea anyway... if that doesn't work for some reason lemme know17:20
jgriffithSwanson: but there's retype, and theres the repstatus update info17:20
jgriffithso we should be covered17:21
jgriffithkeeps things rather simple IMHO17:21
xyangjgriffith:  someone was asking a question on failover yesterday.  After failover, the volume status become failed over.   Is there any way to make it enabled again, other than manually change the db?17:21
jgriffithxyang: we've had some discussions on that....17:21
Swansonjgriffith, so if the vol  type is the same in retype I should take the opportunity to resetup the repls.  Easy enough.17:21
jgriffithxyang: I've been leaning towards that should require admin intervention17:21
Swansonjgriffith, repstatus update info?17:22
Swansonjgriffith, is this the get_replication_updates bit?17:22
xyangjgriffith: sure.  it's just that we currently cannot reset status17:22
jgriffithxyang: I like the concept of "hey... you can fail over in bad scenario, BUT an admin needs to get involved to figure out what went wrong and how to recitify things"17:22
xyangjgriffith: with an API17:22
jgriffithxyang: ahhh17:22
xyangjgriffith: I think if admin change the volume status in the db, then we are good again.  just lack of api for that17:23
jgriffithxyang: we'll certainly need to address that17:23
xyangjgriffith: sure, thanks17:23
jgriffithxyang: absolutely17:23
xyangjgriffith: by the way, I submitted a WIP patch to add tempest tests for CG: https://review.openstack.org/#/c/252213/17:24
*** thangp has joined #openstack-cinder17:24
jgriffithxyang: oh cool!17:25
jgriffithxyang: so just change the backup config to CG-config and it should be good no?17:25
xyangjgriffith: I think so17:25
jgriffithxyang: awesome!  Thanks for going that route17:25
xyangjgriffith: I just need to test more17:26
*** EinstCrazy has quit IRC17:26
*** vgridnev has quit IRC17:26
*** belmoreira has quit IRC17:26
*** gouthamr has quit IRC17:26
*** vgridnev has joined #openstack-cinder17:26
xyangjgriffith: I don't think we can run them in the gate even if we have a gatedriver there though because the policies are disabled by default17:26
*** markstur has joined #openstack-cinder17:27
scottdaDuncanT1: smcginnis I commented on the HA/AA spec. Perhaps geguileo will have something to add. I also don't think the spec needs to be re-spun to reflect this clarification.17:27
*** gouthamr has joined #openstack-cinder17:28
*** egonzalez has quit IRC17:29
*** dims has joined #openstack-cinder17:29
*** abehl has joined #openstack-cinder17:30
*** klkumar has quit IRC17:30
jgriffithxyang: I'm fairly opposed to the gate-driver strategy anyway17:32
jgriffithxyang: it's one more thing to support with little or no value17:32
hemnasmcginnis, https://review.openstack.org/#/c/250371/17:32
jgriffithxyang: if we key off of the support config settings in tempest I think we have a win / win17:32
xyangjgriffith: ok, did you come up with that name?:)17:32
xyangjgriffith: I do intend to have our CI's run those tests17:33
jgriffithxyang: the gate-driver name?  No, not me17:33
xyangjgriffith: I thought when I brought this up at the summit, someone picked this name17:33
jgriffithxyang: yeah, i think that's the right answer... maybe even take it a step further at some point and key off of the drivers capaibilities report17:33
jgriffithxyang: maybe... I dont' remember :)17:33
xyangjgriffith: I was initially thinking about adding it to tests folder17:33
jgriffithxyang: but it's all recorded for historical purposes and to protect the innocent :)17:34
jgriffithxyang: yeah, that's what I'd really like at some point17:34
openstackgerritGerald McBrearty proposed openstack/cinder: Storwise: add config option to control flash copy rate  https://review.openstack.org/24825017:34
xyangjgriffith: I'll go check on youtube:)17:34
jgriffithxyang: but crawl, walk run (CWR)17:34
jgriffithLOL... please don't!17:34
xyangjgriffith: it could be you, don't be surprised:)17:34
jgriffithI contradict myself enough, don't need evidence from youtube :)17:34
xyangjgriffith: I'm fine not to submit that.17:34
xyangjgriffith: tempest tests are needed any way17:35
*** alonma has joined #openstack-cinder17:35
xyangjgriffith: it will be easier for people who want to make changes to CG but doesn't have a backend that really supports it17:35
xyangjgriffith: that was my original intention17:36
jgriffithxyang: yeah, I think it's the right answer17:36
xyangjgriffith: may be I can add it to the fakedriver in the tests?17:36
jgriffithxyang: oh... no, wait :)17:36
jgriffithxyang: what I meant was relying on third party CI seems like the right answer to me17:37
jgriffithI guess adding it to the existing fake-driver might be a good compromise though17:37
xyangjgriffith: sure, relying on 3rd party CI is good.  It is just easier if someone wants to run it themselves, debug, etc17:38
jgriffithfair point17:38
xyangjgriffith: otherwise they have to rely on logs only17:38
DuncanT1Adding it to the fake driver is a good thing anyway, you can do some nice scale testing with it easily17:38
xyangDuncanT1: scale testing?17:38
*** alonma has quit IRC17:39
DuncanT1xyang: The fake driver is great for things like testing the scheduler with 200 backends, that sort of thing. I'm hoping to use it for more complex tests in future17:39
*** kfarr has joined #openstack-cinder17:39
xyangDuncanT1: sure17:39
*** shausy has joined #openstack-cinder17:40
*** alonma has joined #openstack-cinder17:42
*** mudassirlatif has joined #openstack-cinder17:43
*** alonma has quit IRC17:46
*** alonma has joined #openstack-cinder17:48
*** alonma has quit IRC17:52
*** changbl has quit IRC17:54
*** gouthamr has quit IRC17:54
openstackgerritMerged openstack/cinder-specs: Support HA Active/Active configurations  https://review.openstack.org/23259917:54
*** shausy has quit IRC17:54
*** _amrith_ is now known as amrith17:55
*** amrith has left #openstack-cinder17:55
*** shausy has joined #openstack-cinder17:56
openstackgerritMikhail Khodos proposed openstack/cinder: Re-add Nexenta driver  https://review.openstack.org/18887717:59
*** shausy has quit IRC18:00
*** shausy has joined #openstack-cinder18:00
*** julim_ has joined #openstack-cinder18:01
*** alonma has joined #openstack-cinder18:03
*** julim has quit IRC18:04
*** jordanP has quit IRC18:06
*** changbl has joined #openstack-cinder18:07
*** alonma has quit IRC18:08
hemnaxyang, from what I remember we named it the gate driver in the meetup in Tokyo18:11
*** harlowja has joined #openstack-cinder18:12
*** e0ne has quit IRC18:13
xyanghemna: that's what I remembered:)18:15
openstackgerritMerged openstack/os-brick: Correct a log message  https://review.openstack.org/25102418:15
xyanghemna: we have youtube to prove it too:)18:16
*** vgridnev has quit IRC18:16
hemnaI'm sure it's in there somewhere.18:17
*** alonma has joined #openstack-cinder18:19
*** dustins has joined #openstack-cinder18:19
*** nkrinner has quit IRC18:20
*** rebase_ has joined #openstack-cinder18:20
*** p0rtal has joined #openstack-cinder18:23
*** alonma has quit IRC18:25
*** alonma has joined #openstack-cinder18:27
smcginnisall: py26 support officially dropped from python-cinderclient18:29
smcginnishttps://review.openstack.org/25228618:29
*** salv-orlando has joined #openstack-cinder18:30
*** alonma has quit IRC18:31
*** diablo_rojo has quit IRC18:33
jgriffithsmcginnis: py26 is dead, long live py26!18:34
*** alonma has joined #openstack-cinder18:35
smcginnis:)18:35
*** alonma has quit IRC18:39
openstackgerritMikhail Khodos proposed openstack/cinder: Re-add Nexenta driver  https://review.openstack.org/18887718:39
*** sweston has quit IRC18:41
*** patrickeast has quit IRC18:41
*** adrianofr has quit IRC18:41
*** alonma has joined #openstack-cinder18:44
*** p0rtal_ has joined #openstack-cinder18:47
*** alonma has quit IRC18:48
smcginnisAnyone else having issues running tox -ecover18:49
smcginnisGetting an error, reports success, no output.18:49
*** alonma has joined #openstack-cinder18:50
openstackgerritSean McGinnis proposed openstack/os-brick: Add requests to project requirements  https://review.openstack.org/24838318:50
*** bardia has joined #openstack-cinder18:51
*** p0rtal has quit IRC18:51
*** vgridnev has joined #openstack-cinder18:52
eharneysmcginnis: it's running for me18:54
tbarronsmcginnis: seems like it worked recently, trying again now18:54
*** alonma has quit IRC18:54
smcginniseharney, tbarron: Thanks. Could very well just be me. I'll try on another machine.18:55
*** markus_z has quit IRC18:55
*** alonma has joined #openstack-cinder18:57
*** andrew____ has quit IRC18:58
*** alonma has quit IRC19:02
*** alonma has joined #openstack-cinder19:06
*** sweston has joined #openstack-cinder19:08
*** alonma has quit IRC19:10
*** ociuhandu has quit IRC19:20
*** aix has quit IRC19:20
*** diablo_rojo has joined #openstack-cinder19:20
*** adrianofr has joined #openstack-cinder19:21
*** leeantho has joined #openstack-cinder19:23
*** sgotliv__ has quit IRC19:31
*** shausy has quit IRC19:32
*** briancurtin has quit IRC19:33
*** serverascode has quit IRC19:33
*** rhefner has quit IRC19:33
*** scottda has quit IRC19:33
*** wanghao has quit IRC19:33
*** ameade has quit IRC19:33
*** zhiyan has quit IRC19:33
*** leseb_ has quit IRC19:34
*** angela-s has joined #openstack-cinder19:34
*** leseb has joined #openstack-cinder19:36
ildikovhemna: hi. are you around?19:36
ildikovhemna: I would like to talk to you about the multi-attach blueprint19:36
hemnaildikov, hey19:37
hemnaildikov, I'm guessing you saw my review.  heh19:37
ildikovhemna: hehe, yeap, you're pretty right :)19:38
*** abehl has quit IRC19:38
ildikovhemna: so the thing is that I'm getting confused as if you check the previous reviews we got to the point that volume attachments should be handled on host level19:38
*** 7YUAAB7RZ is now known as nug19:39
*** nug has left #openstack-cinder19:39
ildikovhemna: we had a bug on Cinder side, when we ried to handle the attachments per instance: https://bugs.launchpad.net/cinder/+bug/147127819:39
openstackLaunchpad bug 1471278 in Cinder "target removal on detaching volume with multi-attach flag" [Undecided,Invalid]19:39
ildikovhemna: and it was also marked as invalid with the same explanation as the spec is written now in Nova19:40
*** apoorvad has quit IRC19:41
ildikovhemna: could you help me clarify this?19:43
ildikovhemna: the other side of the issue is that Nova spec freeze is tomorrow19:43
hemnahold on a sec...on a phone call...19:43
*** patrickeast has joined #openstack-cinder19:45
*** DuncanT has quit IRC19:54
*** erlon has quit IRC19:54
*** julim_ has quit IRC19:54
*** markvoelker has quit IRC19:55
*** ndipanov has quit IRC19:56
*** gsilvis has joined #openstack-cinder19:57
*** Lee1092 has quit IRC19:59
*** cppforlife_ has quit IRC19:59
*** markvoelker has joined #openstack-cinder19:59
*** alkhod has joined #openstack-cinder20:05
*** alonma has joined #openstack-cinder20:06
openstackgerritNate Potter proposed openstack/cinder: Check backup service before backup delete  https://review.openstack.org/24649820:06
hemnaildikov, ok back20:06
*** erlon has joined #openstack-cinder20:06
openstackgerritMerged openstack/os-brick: Fix iopsLimit parameter in ScaleIO connector  https://review.openstack.org/25158220:07
ildikovhemna: ok, I'm still here20:07
openstackgerritMerged openstack/os-brick: Multipath Device Action Being Parsed as Name  https://review.openstack.org/24588220:07
*** julim has joined #openstack-cinder20:07
hemnaildikov, reading through the 1471278 bug..20:07
hemnaildikov, ok so I'm not sure that bug is overly descriptive20:08
hemnaI'm unclear if the nova side was passing the attachment_uuid along with the detach call to cinder20:08
ildikovhemna: according to my understanding Cinder removed the target regardless of the fact that there were still existing attachment on the given host20:09
hemnaI presume this was an lvm cinder driver20:09
ildikovhemna: also could you please check smcginnis's comments on the spec patch set #12?20:10
openstackgerritMerged openstack/os-brick: Brick add param documentation to connectors  https://review.openstack.org/24602320:10
openstackgerritMerged openstack/os-brick: Add quobyte protocol for remotefs connections  https://review.openstack.org/25037120:10
hemnaok I'll go back through the review comments and see if I can wrap my head around it all20:10
*** alonma has quit IRC20:10
hemnabut if my initial assumptions end up being correct, it could be a bug in the lvm driver's terminate connection call.20:11
hemnasome backends will export a single volume to a host, no matter how many attachments there are to vms on that host.20:11
SwansonAre replication-promote and replication-reenable replication v2 functions?20:11
*** Lee1092 has joined #openstack-cinder20:11
hemnaother backends will export a target for every attachment.20:11
ildikovI'm not sure about the driver itself, it most probably was the lvm one we tested, but unfortunately I wasn't the one who did it20:12
hemnaildikov, fair enough :)20:12
hemnalvm is a bit of a special case in some respects20:12
ildikovbut the bug is not that important at this point, I'm more worried about the Nova spec freeze tomorrow20:12
hemnaeven though it's the reference driver20:13
hemnacrap20:13
hemnanova spec freeze is already?!?!20:13
hemna*smh*20:13
ildikovwhen we had the fix for the bug up for review the main concern was that the different drivers have different behavior20:13
hemnaso nova gets 1 milestone to get all specs in place for the entire release?20:13
hemnathat doesn't give folks a lot of time to work out issues.  :(!20:14
*** serverascode has joined #openstack-cinder20:14
ildikovyes, because the last milestone is for high prio items20:14
ildikovand this one is not a high prio one, although still smth that many people would be happy about20:14
ildikov(including me and I assume you too :) )20:14
hemnawe've been trying to get this in for a few years now20:14
ildikovso we have code on the Nova side already20:15
hemnahttps://review.openstack.org/#/c/85852/20:15
*** ameade has joined #openstack-cinder20:15
hemnathat's my original patch way back when20:15
ildikovI rebased it, I need to double check it though, but most of the tests look ok20:15
ildikovthe API part we agreed with John to skip as they want to remove volume creation from the Nova API20:16
hemnaildikov, cool!20:16
hemnaphew20:16
hemna:)20:16
ildikovthe rest of the code is more or less in shape assuming attaching a volume per instance and not per host20:17
hemnaildikov, ok so I don't know what to do.20:17
hemnacinder has the volume_attachments table that is supposed to track attachments20:17
ildikovhttps://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/volume-multi-attach,n,z20:18
*** zhiyan has joined #openstack-cinder20:19
ildikovif you check the comments around patch set #12 on the spec, then you will see why I modified the blueprint to handle the attachments per host20:19
ildikovI checked the Cinder blueprints and wiki pages20:19
ildikovthere is one, which talks about host level attaching a volume, all the rest refers to instances20:20
hemnaildikov, ok I'll read through it.20:21
ildikovhemna: is the intention on Cinder side to track all the attachments to actual VM instances?20:21
hemnaildikov, yes20:21
hemnaildikov, that was the intention20:21
hemnaand the attachment_uuid uniquely identifies each attachment to each instance20:22
ildikovand how does it work in reality when we're attaching a second volume?20:22
ildikovor what's the most common way for the drivers?20:23
ildikovhaving let's say one target and then the initiator on the host opens a new iSCSI connection?20:23
hemnathe attach call comes in for the new instance, we save the attachment against that new instance and we get a 2nd record in the volume_attachments table20:23
hemnaand that gets a unique attachment_uuid20:23
hemnaso when nova calls detach, it's supposed to sent the attachment_uuid of the instance it's detaching from.  Then cinder updates that record20:24
hemnaand is able to track correctly the state of the volume. (in-use, available)20:24
hemnaildikov, so what the cinder driver actually does is entirely dependent on it's backend and how it does attachments.20:25
smcginnisildikov, hemna: Sorry, just got back from a block of meetings.20:25
*** IlyaG has joined #openstack-cinder20:25
ildikovthe bug is about this part as Cinder completes the whole detach for the first call, maybe only for lvm case, I would need to check that20:25
smcginnisildikov: And sorry, looks like I may have given bad information. :{20:25
hemnawe can't make any assumptions on how each cinder backend does it's exports (targets) and detaches.20:25
ildikovsmcginnis: no probs, we're in the middle of the story still :)20:25
hemnaildikov, I think that bug was really against the lvm driver20:25
smcginnisildikov: hemna  will straighten us out. ;)20:25
*** cppforlife_ has joined #openstack-cinder20:26
hemnasmcginnis, or confuse us even more :)20:26
ildikovhemna: but we need to ensure somehow still that the volume does not get detached from a host when there's still active attachment20:26
smcginnisI honestly wasn't aware we tracked per vm instance.20:26
hemnayup20:26
smcginnishemna: What is the reason for that?20:26
hemnathat's the job of the cinder driver to know if it still has active attachments on that host20:27
smcginnishemna: As a backend, I don't really care what nova does with it once it's exposed on the compute host.20:27
hemnasmcginnis, because some backends export a new LUN per attachment, even on the same host.20:27
ildikovhemna: ok, then we can re-open the bug with this additional comment20:27
hemnaand we need to know which one of those to nuke at detach time.20:27
smcginnishemna: Is there a technical need for that?20:28
*** briancurtin has joined #openstack-cinder20:28
dulekdiablo_rojo: Hi, can you take a look and advice on https://review.openstack.org/#/c/183537/ ?20:28
smcginnisUltimately nova just needs a scsi device to use for the vm. Why make multiple scsi devices for the same volume?20:28
hemnasmcginnis, other than that's the way the backend works20:28
dulekdiablo_rojo: I believe it's failing because of config options generator.20:28
hemnathis is how the 3PAR works fwiw.20:29
hemnaour other driver (lefthand) does a single target per host, regardless of how many instances on that host have the volume.20:29
dulekdiablo_rojo: Ah, it's more complicated. Patch is adding a new options group. How to proceed with that and genopts?20:29
hemnabut without tracking in cinder you could get the volume in available state by just accidentally calling the cinder detach api20:30
smcginnishemna: Still not sure why 3par would need to know there are more than one instances using the volume.20:30
smcginnisGuess I need to learn more about other backends.20:31
hemnaalso, the attachment tracking comes into play when you attach to multiple hosts20:32
hemnare: attaches to ironic, and attaches to cinder nodes.20:32
diablo_rojodulek: You need to run tox -e genopts to get a new opt.py file20:32
hemnathose also get entries in the volume_attachments table20:32
smcginnisI see multiple hosts, because ultimately from the storage side it's that target<>initiator connectivity.20:33
smcginnisI still don't see how instances have any visibility to a backend.20:33
diablo_rojodulek: That should get all the recently added options and then you should be able to check in that opts.py file and pep8 should pass.20:33
*** rhefner has joined #openstack-cinder20:33
dulekdiablo_rojo: But generate_cinder_opts.py has the sections hardcoded.20:34
ildikovI assume the question here whether we would need different LUNs for each instance in case of 3PAR or it's enough to have one per host and have multiple instances using it20:34
diablo_rojoOH20:34
dulekdiablo_rojo: And we're adding a new [coordination] section.20:34
Swansonaorourke, jgriffith are replication-promote and replication-reenable repl v1 functions?20:34
diablo_rojodulek: Oh okay.20:34
hemnaso if we only use 1 attachment per host, then nova has state that cinder doesn't20:34
jgriffithSwanson: yup20:34
ildikovalthough I'm not really dying for refactor the Nova code if it's not what was the intention here oon the first place either20:35
smcginnishemna: But why does cinder care about that state.20:35
hemnameaning only nova has the source of truth for attachments.20:35
smcginnisIf they want 1 vm or 10 vms, why do we care?20:35
hemnasmcginnis, force detach issues20:35
Swansonjgriffith, Excellent.  Mentally flushed.20:35
hemnaand reset state20:35
hemna:(20:35
dulekdiablo_rojo: Is there a reason sections are hardcoded? Should we just add it to generate_cinder_opts? This seems a little odd.20:35
smcginnishemna: But a force is a "force". Still, why does storage care?20:35
ildikovI'm fine if Cinder handles the state, from the perspective hemna highlighted that sounds more clean20:35
hemnausers could be dumb and try and force detach for 1 dead instance, which nukes all of them for the rest.20:35
smcginnisNot trying to be difficult, just don't understand.20:35
diablo_rojodulek: Yeah currently those sections aren't being pulled out dynamically. One of the things on my plate is to do that. At this point you're probably going to need to hard code it.20:36
hemnasmcginnis, well but you may have 1 dead vm on the same n-cpu node and need to force detach it.20:36
smcginnisThis is different than any other storage management behavior I've ever dealt with.20:36
*** scottda has joined #openstack-cinder20:36
smcginnishemna: Well, yeah. :)20:36
hemnawhich will bypass all the nova checks to call terminate connection and detach20:36
smcginnishemna: That would be on the nova side, no?20:36
hemnaand now you've lost the target for the rest of the vms.20:36
smcginnisWhy would you be calling detach to your storage if you are detaching a vm?20:36
smcginnisCinder detach shouldn't ever be called unless that host no longer needs the SCSI device.20:37
hemnawe call it today w/ terminate_connection and detach.20:37
*** crose has quit IRC20:37
hemnabut there is nothing preventing a user from calling it via the cinderclient to try and force detach20:37
hemnaand cinder would have no idea that it's attached to another 5 vms on that host.20:37
hemnaoops, there goes your target.20:37
hemnapoof20:37
dulekdiablo_rojo: Okay. bluex - you may want to take a look on my conversation with diablo_rojo.20:37
*** wanghao has joined #openstack-cinder20:38
jgriffithhuh?20:38
smcginnisBut again, your telling cinder to force a detach. So wheterh a vm or container or whatever is using the volume, as a block storage service we just remove the device.20:38
hemnaanyway, I have to get some chow.... /me is starving20:38
smcginnisildikov: Sorry, we'll obviously need more discussion on this. :/20:38
jgriffithmake clean; make all20:39
jgriffithgrrr20:39
ildikovsmcginnis: well, I don't know how much you checked our discussion with hemna, but the Nova spec deadline is tomorrow :S :( :.( etc. :)20:39
hemna:(20:39
smcginnisildikov: Yeah, saw that. :{20:39
*** ociuhandu has joined #openstack-cinder20:39
ildikovsmcginnis: but I'm on the side of clarifying this for sure20:39
smcginnisildikov: Maybe we can at least get to the point of the spec being OK, then fix up in the implementation.20:40
ildikovsmcginnis: hemna: and obviously I would like to understand the behaviour also20:40
ildikovyeah, that would be great, but that part is what needs more discussion as much as I ccan see here :)20:40
hemnaok so20:41
ildikovsmcginnis: hemna: can we continue this a bit later today to get to a consensus?20:41
hemnaon one hand, this sort of violates the volume_attachments table20:41
hemnabut I wonder if we could just get away with it in the short term20:41
hemnaand see how it goes ?20:41
hemnawe could eventually migrate to what I've been discussing later ?20:42
smcginnishemna: To clarify - you mean get away with how the spec is written right now?20:42
hemnayah20:42
hemnameaning nova doesn't call cinder after the 1st attach20:42
hemnait bugs me and I think we might run into problems20:43
ildikovhemna: well, if we go for the per host direction that means refactoring on Nova side20:43
hemnabut that's almost always the case w/ anything20:43
ildikovhemna: I'm not sure about the Cinder part, there might not20:43
hemnaI'd love to see the end to end ability land finally20:43
ildikovhemna: so it's easier to go for the per instance version and then refactor if that seems better longer term20:43
hemnaildikov, the downside is that we'd have to add checks in the drivers to make sure it doesn't nuke the target if there are other attachments to the same host.20:44
hemnaand I'm not sure the lvm driver has any concept of that at all20:44
ildikovhemna: I mean the pervious version of the spec looks easier not the one you've just commented on20:44
hemnaonly the attachments table has that, and it doesn't track the host for the instance20:44
ildikovhemna: sure, this is what we had that bug report for originally20:45
smcginnisildikov: I think as it is is where it needs to be, but we need to hash some of this out on the cinder side.20:45
ildikov:(, yeah, why life would be easy just out of the box20:45
hemnaildikov, would be nice!20:45
smcginnishemna: Maybe change your vote for now, let's try to get the spec through while we can, then we can mess with it as we go? :)20:46
jgriffithhemna: curious, didn't we go with just making attachments a list in Cinder?20:46
jgriffithhemna: or did that suggestion flop20:46
ildikovsmcginnis: I have the code for the earlier version of the spec and I have a gut feeling that the refactorings on Nova side will not be that straight forward20:46
hemnajgriffith, that has other problems20:46
* jgriffith supposes he could look20:46
jgriffithbut there's the answer20:47
hemnathose problems related to force detach and detaching on baremetal.20:47
hemnaanyway....I have to go get food before the cafeteria here closes.20:47
jgriffithhemna: that seems to be the "phrase of the month"20:47
jgriffithFOOOD!!!!20:47
jgriffith:)20:47
hemna:)20:47
ildikov:)20:48
*** hemna is now known as hemnafood20:48
*** DuncanT has joined #openstack-cinder20:48
ildikovsmcginnis: sorry for favoring the per instance direction as a first step20:49
smcginnisildikov: No, I get it from your perspective.20:49
smcginnisI just don't think it makes sense for a block storage service to know or care about vm instances.20:49
smcginnisEspecially as we start looking at containers and bare metal as consumers of the block storage.20:50
ildikovwe tested the initial version of that code and it's feature is delayed for so long now20:50
smcginnisIt really doesn't matter to me who or what is consuming my storage.20:50
ildikovyeah, that sounds reasonable20:50
smcginnisildikov: :(20:50
smcginnisildikov: Hopefully we can get this through.20:50
smcginnisildikov: I would love to see this implemented.20:50
ildikovso if we can start with the current as a first step, than we would have a better likelyhood to get this into Mitaka20:51
ildikovand then we can refactor it on both sides as for the user it should be transparent how we handle the attachments in the background20:51
smcginnisildikov: Yeah, let's just get things going. We can correct down the road. Totally agree.20:51
ildikovby current I meant the code for the per instance base20:51
smcginnisildikov: If that's what we have to do, then I'm fine with that.20:52
ildikovsmcginnis: believe me, I'm the kind of let's make things the right way, but realistically the winter cycle is too short for that20:52
ildikovespecially with the Nova deadlines20:52
smcginnisildikov: I do think it needs to change, but this is obviously a bigger issue than just this spec.20:52
ildikovyeah, I agree20:53
scottdaildikov: Sorry to be late to the conversation. IRC client acting up on me. I just added comments to your spec20:53
scottdaAnd I think I missed an early part of this conversation.  Prior to 20:36 UTC20:54
smcginnisildikov: If you get agreement from hemnafood and scottda, I'll get in there later and give a +1 for my support.20:54
ildikovsmcginnis: cool, thanks :)20:55
smcginnisscottda: http://eavesdrop.openstack.org/irclogs/%23openstack-cinder/%23openstack-cinder.2015-12-02.log.html20:55
smcginnisscottda: Search for "ildikov, ok back" :)20:55
scottdawell, that is handy. Thanks20:55
ildikovscottda: we've just been discussing the main directions here which are either handling attachments per host or per instance20:55
ildikovsmcginnis: lol :)20:56
openstackgerritNate Potter proposed openstack/cinder: Use the catalog to retrieve glance_api_servers  https://review.openstack.org/25247620:56
*** EinstCrazy has joined #openstack-cinder20:56
scottdayeah. And is the direction going towards per instance?20:56
*** raildo is now known as raildo-afk20:58
smcginnisscottda: I believe so. Is that you're take ildikov?20:58
ildikovthat's my preference at least as a first step for now20:58
ildikovscottda: what would be your 2 cents for the topic?20:59
*** IlyaG has quit IRC21:00
scottdaI think that Nova should not have to manage the attachments, and Cinder should have to deal with it, even if that means individual drivers will need some way to keep track in the case of 1 export per host.21:00
*** EinstCrazy has quit IRC21:00
scottdaI should have said exports instead of attachments21:00
smcginnisscottda: Why does block storage care about VMs?21:01
smcginnisscottda: When we do bare metal, do we need to track which apps are using it? :)21:01
smcginnisscottda: We should just care about exposing storage to an initiator.21:01
scottdaIt should not care about VMs or bare metal. But the problem seems to be at the level of terminate_connection, not detach()21:01
smcginnisscottda: If that initiator runs VMs and attaches one or more to the volume, then that's the domain of that host.21:02
ildikovscottda: we were also discussing these options as short term and long term, where per instance is the short term one21:02
smcginnisscottda: It's up to the service controlling the inititator to say when it no longer needs the attachment.21:02
smcginnisscottda: So nova should only call cinder detach if it no longer needs access to that block storage.21:02
ildikovdumb question do we have/can we have multiple connections for the same volume per initiator?21:03
scottdasmcginnis: Fair enough. I think that should be called out in the spec, since there is both the detach() management of the Cinder DB and the terminate_connection to manage the export.21:03
smcginnisildikov: Multipathing?21:03
smcginnisscottda: That's where i think we should be. For the purposes of getting this spec through I think we just have to go with the per-instance bit.21:04
smcginnisBut I don't like it. :)21:04
smcginnisSorry, gotta run. bbl21:04
*** xyang has quit IRC21:05
ildikovsmcginnis: yeah, i think my question falls there, but let's continue later21:05
scottdaildikov: What time zone are you in?21:06
ildikovscottda: CET, it's 10pm now21:06
scottdayikes, getting late for you.21:06
*** EinstCrazy has joined #openstack-cinder21:06
ildikovscottda: but Nova spec deadline is tomorrow...21:06
*** alonma has joined #openstack-cinder21:06
scottdayes!21:07
ildikovyeah, I gave my soul to open source, I'm not always sure it was a good deal... :)21:07
ildikovso I hope we can agree one the Nova side of the mult-attach spec here as i still need to hunt down Nova folks to get this through21:08
*** xyang has joined #openstack-cinder21:08
ildikov*hoped21:08
scottdaYes, when hemnafood turns into hemna we should be able to resolve this.21:09
ildikovyeap, hopefully he does not take the afternoon off :)21:09
ildikovI hope we can agree on a step by step option and start with the per instance direction as it seems that was the intention on Cinder side and then refactor to per host and then let Nova ddeal with the attachments21:10
*** alonma has quit IRC21:11
*** EinstCrazy has quit IRC21:12
scottdaI'm OK with that and I am not very adamant about either direction. I'm just interested in making the design as smart as possible.21:13
scottdaAnd I also don't worry about the 'force detach' case where a user calls the detach API directly without Nova. That's dangerous and unsupported AFAIC21:14
ildikovlong term that sounds good, I mean the as smart as possible21:14
ildikovbut with Nova deadlines we need something that works first IMHO21:14
ildikovhehe, it's quite hard to fully prepare a system for users, they will eventually break it anyway21:15
scottdaThe main issue seems to be the case where there is >1 attachment to a given host. If backend storage can handle this in different ways, by exporting a single LUN or  >1 LUN, then someone (NOva or Cinder) needs to keep track of when the last detach is occuring.21:15
scottdaAnd I fully sympathize with the deadline. I'm for whatever it takes to get this through.21:16
*** ntpttr has quit IRC21:16
ildikovif for the 2nd+ instance the attach happens only on Nova side, then there should not be additional LUNs per host21:16
Swansonjgriffith, replication_status seems to have undergone some changes.  Are the expected states captured in a document or should I just work my way through driver.py again?21:17
*** ntpttr has joined #openstack-cinder21:17
ildikovscottda: but Nova needs refactoring to handle it this way21:17
ildikovit seems that attaching a volume is quite centralized, I'm more concerned about detach, I mean when we get there to have the last instance detaching a volume so we would need to call out to Cinder to terminate_connection also21:19
jgriffithSwanson: this one https://github.com/openstack/cinder/commit/c30ce044b713113317bbeb1c3df2593bd6a6ec2b#diff-28844a4bdd73ce7acb2b2cb32a4ae27621:19
jgriffithSwanson: or status?21:19
Swansonjgriffith, replication_status.21:19
scottdaildikov: Yes, either Nova needs the logic or Cinder. I see smcginnis  arguement that Cinder should not care how the volume is used on a host once the export is presented.21:20
scottdaildikov: Have you looked in detail at the scope of the Nova refactor?21:20
jgriffithSwanson: so what's the question exactly?21:21
ildikovscottda: to be honest as mid/long term goal I share that view21:21
*** alonma has joined #openstack-cinder21:21
Swansonjgriffith, When is what returned?21:21
ildikovscottda: not deeply21:21
jgriffithSwanson: so the status fields haven't been changed since V121:21
scottdaSorry, could you repeat the short term solution? Doing it per instance? I'm not sure how we'd deal with terminate_connection in that case.21:21
ildikovscottda: I got some pointers for the attach part, and I was told that deatch seems harder to handle, but I'm not full time on Nova :(21:22
jgriffithSwanson: valid states are still: enabled, enabling, disabled, disabling, error21:22
scottdaWould Nova call terminate_connection along with each detach whenever 'nova volume-detach' is called, as it is done today?21:22
Swansonenabling_secondary looks to be in there.  active, copying, etc.21:22
ildikovscottda: from what hemnafood said earlier that was the original idea that Cinder is supposed to handle this21:22
Swansonjgriffith, Just going to go through manager.py and do what is appropriate.21:23
jgriffithSwanson: https://github.com/openstack/cinder/blob/master/cinder/volume/manager.py#L313421:23
scottdaildikov: OK. We'll have to wait for him to see if there is still a gap on the Cinder side. He wrote that code :)21:23
ildikovscottda: if that's how it works today when we call detach, then I assume this should be changed21:23
jgriffithSwanson: notice the nightmare V1 stuff is still in there... just focus on V2 if I were you21:24
openstackgerritMikhail Khodos proposed openstack/cinder: Re-add Nexenta driver  https://review.openstack.org/18887721:24
scottdayes, but Multi-attach is already in Cinder. So I'm not sure what would be missing.21:24
ildikovscottda: then it's better to not make decisions without him :)21:24
*** alonma has quit IRC21:25
ildikovscottda: well, right, so the terminate_connection part should be handled somehow too, but we had issues with detach so it should be double checked for sure21:27
Swansonjgriffith, That looks like a plan.  I'm checking statuses I do not need to be.  I think I'll just remove them.  If the layers above are checking that I'm in the right state I'll call it good.21:27
jgriffithSwanson: yeah... that V1 stuff is actually a bit of a nightmare, and confusing IMO21:28
jgriffithand having both versions in the code base at the same time is not helping anyone :(21:28
*** IlyaG has joined #openstack-cinder21:30
*** adrianofr has quit IRC21:30
*** hemnafood is now known as hemna21:30
openstackgerritManjeet Singh Bhatia proposed openstack/python-cinderclient: Pass proxy environment variable to tox  https://review.openstack.org/25219821:31
*** alonma has joined #openstack-cinder21:33
*** alonma has quit IRC21:37
ildikovhemna: please ping us with scottda, if you're ready to discuss and agree on the next steps21:39
*** xyang1 has joined #openstack-cinder21:44
*** ctina_ has joined #openstack-cinder21:45
*** earlephilhower has joined #openstack-cinder21:47
hemnahey I'm kinda here21:48
hemnadid you guys solve everything?21:49
hemna:)21:49
*** ctina has quit IRC21:49
scottdaSo, it seems there are 2 basic issues:21:49
scottdaOne is that the spec doesn't refer to the attachment_id for calling detach. That doesn't seem too controversial.21:49
*** ctina_ has quit IRC21:50
scottdaThe other is whether Nova should call terminate_connection each time the end user calls 'nova volume-detach' and let Cinder deal with the export...21:50
scottdaor whether Nova should manage the connections on it's side and determine when the last volume on a particular host is detached, and call terminate_connection at that time.21:51
scottdaWith the problem being complicated by the fact that a backend storage driver might export 1 lun for multiple volumes on the same host, or might export >1 lun21:51
scottdaDid I get that right?21:51
ildikovthe attachment_id is referred in the spec21:51
ildikovso when we go to Cinder we use it, the question is what complexity we should have in Nova and what in Cinder as much as I can see21:52
scottdaildikov: Yes, but see the comments hemna and I have made regarding attachment_id. It's really a small thing and not the important one, I don't think.21:53
*** apoorvad has joined #openstack-cinder21:53
*** thangp has quit IRC21:54
scottdaThe controversy seems to be around how/when to call terminate_connection. Unless I've gotten this wrong.21:54
hemnacinder's unreserve_volume checks attachments in the volume_attachment table as well21:55
ildikovscottda: yeah, so I would go with patch set #12 for now21:55
hemnanm, I think that's safe21:56
ildikovscottda: that basicly leaves tracking and responsibility on Cinder's side, which was the intention if I understood hemna correctly21:56
hemnaI'm trying to see if there are any gotchas with the volume state if we don't have nova call cinder for secondary attaches21:56
hemnaildikov, yah21:56
openstackgerritCyril Roelandt proposed openstack/cinder: Python 3: fix cinder.tests.unit.api.contrib.test_snapshot_{,un}manage  https://review.openstack.org/25264121:57
openstackgerritCyril Roelandt proposed openstack/cinder: Python 3: fix cinder.tests.unit.volume.drivers.netapp.eseries.test_utils  https://review.openstack.org/25264221:57
openstackgerritCyril Roelandt proposed openstack/cinder: Python 3: fix cinder.tests.unit.api.contrib.test_scheduler_hints  https://review.openstack.org/25264321:57
openstackgerritCyril Roelandt proposed openstack/cinder: Python 3: Use oslo_serialization.jsonutils.loads instead of json.loads  https://review.openstack.org/25264421:57
openstackgerritCyril Roelandt proposed openstack/cinder: Python 3: fix cinder.tests.unit.api.contrib.test_volume_{,un}manage  https://review.openstack.org/25264521:57
openstackgerritCyril Roelandt proposed openstack/cinder: Python 3: fix cinder.tests.unit.api.contrib.test_cgsnapshots  https://review.openstack.org/25264621:57
openstackgerritCyril Roelandt proposed openstack/cinder: Python3: fix cinder.tests.unit.api.contrib.test_volume_encryption_metadata  https://review.openstack.org/25264721:57
openstackgerritCyril Roelandt proposed openstack/cinder: Python 3: fix cinder.tests.unit.api.v2.test_volumes  https://review.openstack.org/25264821:57
openstackgerritCyril Roelandt proposed openstack/cinder: Python 3: fix cinder.tests.unit.api.contrib.test_snapshot_actions  https://review.openstack.org/25264921:57
hemnawhoa21:57
scottdayes, but I believe smcginnis is of the opposite opinion.21:58
smcginnisRight now I'm of the opinion of what is the path of least resistance to get this nova spec through.21:59
smcginnisThen I'll be of the opinion of how do we fix it. :)21:59
hemnacrud21:59
ildikovsmcginnis: I like your supportive attitude :)21:59
hemnaso, we'd need to fix the lvm driver to support multiattach21:59
hemnaor disable it22:00
ildikovI can write up a backlog spec for Nova to cover the direction to handle the attachments on host level22:00
hemnait would need to know that it has attachents left on that host and terminate_connection would be a noop.22:00
*** bswartz has quit IRC22:01
openstackgerritMitsuhiro Tanino proposed openstack/cinder: Robustify writing iscsi target persistence file  https://review.openstack.org/24788622:01
ildikovsmcginnis: the spec would go through somehow with the current form too, I'm more worried about the corrsponding code22:01
*** bswartz has joined #openstack-cinder22:01
scottdaBut we can worry about the code after tomorrow :)22:01
*** zohar has joined #openstack-cinder22:01
ildikovsmcginnis: refactoring takes time and then we need to fight for review attention, which is tough according to my experience :S22:02
scottdahehe , yes mine too.22:02
ildikovscottda: well, non-priority code should be up for review nowadays too :)22:02
scottdahemna: Won't we need to fix LVM to support multi-attach anyway?22:03
smcginnisildikov: Yeah, I know hemna has had a lot of fun getting stuff in nova. hehe22:03
zoharhi all! can i get some pointers about writing a new nbd driver please?22:03
zoharbasically what our appliance can do is create nbd devices (like /dev/nbd1, etc...) from cluster objects22:03
ildikovsmcginnis: hehe, then I will ask him for help, as he already knows how to do that :)22:04
scottdazohar: Here's an overview: https://wiki.openstack.org/wiki/Cinder/how-to-contribute-a-driver22:04
zoharin this case, it is block device (like iscsi)22:04
hemna*sigh*   meeting....22:04
*** hemna is now known as hemnafk22:04
ildikovoh, we lost hemnafk :(22:05
scottdaSo, does ildikov 's spec need an update? I'm just not clear where it says anything about dealing with terminate_connection....22:05
scottdaIt should probably be called out whether it happens each time a 'nova volume-detach' is called, and Cinder does the management, or whether Nova will track the number of uses of each connection.22:06
zoharso my question is simple, how can i serve a local device (on the node that should consume it, ie hyper-convergance) as a cinder volume?22:06
ildikovscottda: so I would like to fall back to patch set #12 and there we kind of need to decide about the terms we use22:07
zoharand the driver logic of creating/deleting/etc it will take care of of course22:07
zoharit would be somewhat similar to lvm22:07
zoharexcept no partitions, just raw devices22:07
zoharnbd devices22:07
zoharany pointers? :)22:08
zoharsorry for spamming btw ^_^22:08
*** adurbin_ has quit IRC22:08
smcginniszohar: I would recommend taking a look at the lvm driver as a basis.22:09
smcginniszohar: And modify from there where it is different.22:09
scottdaildikov: I am actually fine with either choice for managing the connections myself. So let's make sure smcginnis and hemnafk are on board with the decision.22:09
zohari did, had troubles figuring it out22:09
zoharthank you though22:09
zohari will look more22:09
smcginnisscottda, ildikov: hemnafk understands the way things are better than me. I just understand the way I think they should be. :)22:10
smcginnisSo hopefully he's back soon and can give his input.22:10
smcginnisFrom my perspective, I'm fine rolling the spec back to how it was before I "helped22:10
ildikovscottda: as much as I saw Nova calls terminate_connection now along with detach22:10
smcginnis"22:10
*** julim has quit IRC22:11
scottdaildikov: That's correct. but they are 2 separate calls. And I'm fine with Nova calling terminate_connection for each instance of a volume being detached...22:12
scottdait seems the only way to deal with drivers that export a new lun for each attach of the same volume22:12
ildikovscottda: https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L475022:12
scottda(or add a bunch or really complex code on the nova side to check this)22:12
ildikovscottda: Nova folks would be really sad by adding more complexity to the code22:13
scottdaI certainly don't like the idea of sad Nova folk22:13
smcginnis:)22:14
scottdaI think that it will be easier to have Cinder deal with the complexity. And mostly because of the aforementioned sadness.22:14
scottdaIn which case I'd guess your spec might not need to be changed. As long as hemna doesn't see a problem.22:15
*** dustins has quit IRC22:15
ildikovscottda: single responsibility is a good thing in theory and I think in steps we can move the parts where they belong22:15
scottda(Not that he needs to deal with the Cinder changes)22:15
smcginnisildikov: +122:15
ildikovyou know in amounts of sadness that can be handled at a time :)22:15
smcginnisildikov: Evenly spread the sadness. :)22:16
openstackgerritKendall Nelson proposed openstack/cinder: Adding devref about genconfig  https://review.openstack.org/25265522:16
scottdaYeah, push some sadness out to summer time.22:16
ildikovsmcginnis: exactly :)22:16
diablo_rojosmcginnis: ^^22:17
smcginnisdiablo_rojo: Already open in a tab.22:17
diablo_rojosmcginnis: Ninja fast today.22:17
ildikovscottda: sunshine makes it easier to handle22:17
smcginnisHah22:17
smcginnisSorry, I'll be online later. I'll watch for any nova-spec update and make sure I weigh in as soon as I can.22:18
scottdaYeah, I'm ready to +1 as well.22:18
*** ntpttr has quit IRC22:19
*** timcl has quit IRC22:20
ildikovscottda: smcginnis: I can go and revert to patch set #12 and add a link to the log of this discussion and hemnafk can check it also, when he's back from his meeting22:20
alkhodHi everyone. I have an issue with attaching nfs volume to instance. The only traceback I could find in logs was in n-cpu. Here is the link http://paste.openstack.org/show/480673/. Any thought on this would be great (I am totally stuck here), thanks in advance. Hope this is a correct channel to ask...22:20
scottdaildikov: Sounds good. I'll remind hemnafk to look, if you need to be done for the day.22:21
ildikovI can also add a comment on considering to move to the direction of what is outlined in patch set #13, but in a step-by-step approach22:21
*** Lee1092 has quit IRC22:21
ildikovscottda: I upload the spec now, then take a shower, if hemnafk is not back by then, I would really appreciate if you do this favor for me :)22:22
bardiaalkhod: What driver backend are you using?22:23
ildikovscottda: smcginnis: hemnafk: thanks all for the chat and clarifications!22:23
alkhodbardia: I have this error using generic nfs22:23
*** vgridnev has quit IRC22:24
alkhodbardia: which tells me this is probably a backend issue, but I can't find out what could this be. The error output is kinda not really informant22:25
*** erlon has quit IRC22:26
*** lprice has quit IRC22:27
*** alonma has joined #openstack-cinder22:28
*** lprice has joined #openstack-cinder22:29
scottdaildikov: Thanks for the work and perseverance.22:30
*** baumann1 has left #openstack-cinder22:31
ildikovscottda: np, we're trying to get this done for too long to just simply let it go22:32
*** diablo_rojo has quit IRC22:32
*** alonma has quit IRC22:32
*** lprice has quit IRC22:33
*** alonma has joined #openstack-cinder22:34
bardiaalkhod: sorry not really familiar with the nova side. If your volume is created and you can see it then I'm assuming its a nova issue22:34
openstackgerritChris Morrell proposed openstack/cinder: Address potential races in SolidFire VAG.  https://review.openstack.org/25266022:34
bardiaalkhod: maybe someone else here has seen this. I would try #openstack-nova as well22:35
*** ianbrown has joined #openstack-cinder22:36
alkhodbardia: ok, thank you, will try asking on nova channel too22:36
openstackgerritHelen Walsh proposed openstack/cinder: EMC VMAX - get_short_host_name not called in find_device_number  https://review.openstack.org/25209322:36
*** hemnafk is now known as hemna22:37
hemnasorry guys, had a meeting22:37
*** alonma has quit IRC22:38
*** angela-s has quit IRC22:40
*** angela-s has joined #openstack-cinder22:41
*** edmondsw has quit IRC22:41
*** adelia has quit IRC22:41
*** angela-s has quit IRC22:42
ildikovhemna: np, I've just uploaded a new version of the spec22:42
hemnaildikov, ok22:43
openstackgerritxing-yang proposed openstack/cinder: Add a FakeGateDriver  https://review.openstack.org/24495622:43
ildikovhemna: I basically reverted to patch set #12 and addressed your cinderclient comment and scottda's comment about the docs with data corruption22:43
*** alonma has joined #openstack-cinder22:43
hemnaok sounds good.22:43
hemnasorry about all of this22:43
hemnathere are probably going to be issues with either approach really.22:44
ildikovhemna: no problem, I think it's important to get to the same page and it usually takes some time and effort22:44
xyangjgriffith: I moved it to the fake_driver: https://review.openstack.org/#/c/244956/22:44
ildikovhemna: I think so too, this is why I'm more on the side of starting with the original intention and then fix things22:44
jgriffithxyang: cool, that' looks good22:45
hemnaildikov, ok sounds good.22:45
xyangjgriffith: I can't log warning messages any more as there's a hacking check that prevents any logs in unit tests22:45
jgriffithxyang: :(22:45
ildikovhemna: I will link the log for this IRC discussion to the review too22:45
jgriffithalthough I think that's partially my fault22:45
hemnaildikov, ok that's probably a good idea :)22:45
xyangjgriffith: this one have "Fake" in its name, I hope that is clear enough:)22:45
*** sgotliv__ has joined #openstack-cinder22:46
ildikovhemna: we can add a follow up spec later if we decide to handle the attach/detach/terminate_connection on the host level way22:46
hemnaildikov, ok thats good.  we may eventually end up there22:46
jgriffithxyang: :)22:46
hemnaas long as we aren't painted into a corner22:46
jgriffithxyang: at some point I guess we need to fix up the mess of fake drivers too22:46
ildikovhemna: that refactoring should be transparent to the user anyway22:46
jgriffithxyang: https://github.com/openstack/cinder/blob/master/cinder/volume/driver.py#L255622:46
hemnaildikov, +!22:47
hemna122:47
jgriffithxyang: https://github.com/openstack/cinder/blob/master/cinder/volume/driver.py#L272722:47
xyangjgriffith: I wonder why we have so many of them22:47
jgriffithxyang: at least FakeLVM is gone and LoggingDriver is gone22:47
hemnaildikov, thank you so much for working on this.  much appreciated!22:48
hemnaildikov,  I tried a few times22:48
xyangjgriffith: are they?22:48
*** alonma has quit IRC22:48
jgriffithxyang: haha.. I think so :)22:48
ildikovhemna: I cannot guarantee this time either, but at least we have a better view on where and how we want to go22:48
xyangjgriffith: oh, they are in driver.py22:49
hemnaildikov, ok if there is anything else I can help with, please let me know.22:49
xyangjgriffith: just noticed that22:49
hemnatesting, etc.22:49
jgriffithxyang: who would know anymore... there's like 20 classes in there now :(22:49
xyangjgriffith: ya, it is hard to navigate it:(22:50
jgriffithxyang: worst thing EVER!22:50
xyang:)22:50
xyangjgriffith: I agree it is very confusing22:50
ildikovhemna: scottda: smcginnis: please check the spec review and add your votes there so that I can ping Nova folks tomorrow22:51
ildikovhemna: sure, will do, thanks!22:51
jgriffithxyang: so one thing I wonder....22:51
*** kfarr has quit IRC22:51
jgriffithxyang: not for your patch...22:51
jgriffithxyang: but at some point in the future, should we just collapse all those fakes in fake_driver into a single "class FakeDriver"22:51
*** mriedem has quit IRC22:52
xyangjgriffith: sure, that sounds good22:52
jgriffithxyang: some of those honestly I don't think have any business being in there anyway22:52
jgriffithxyang: ok.. maybe I'll actually work on it :)22:52
xyangjgriffith: :)22:52
*** alonma has joined #openstack-cinder22:53
*** ianbrown has quit IRC22:54
*** jseiler has quit IRC22:55
SwansonSoooo.....  I do a replication-failover.  Volume is on the new backend.  host seems right(ish).  volume_type is the original volume_type.22:56
*** lprice has joined #openstack-cinder22:57
*** alonma has quit IRC22:58
*** jgregor has quit IRC22:58
*** jungleboyj has quit IRC23:01
*** alonma has joined #openstack-cinder23:02
*** alonma has quit IRC23:06
*** EinstCrazy has joined #openstack-cinder23:09
*** akerr has quit IRC23:09
*** xyang has quit IRC23:09
*** alonma has joined #openstack-cinder23:10
openstackgerritMerged openstack/cinder: Port Windows drivers to Python 3  https://review.openstack.org/23494623:11
*** xyang has joined #openstack-cinder23:12
*** EinstCrazy has quit IRC23:13
tbarronjgriffith, xyang: FakeVD ABC23:13
* tbarron ducks23:13
*** [2]Thelo has joined #openstack-cinder23:14
xyangtbarron: we can easily add 20 of those:)23:14
tbarronxyang: -)23:14
hemnaFake Fakes23:14
*** alonma has quit IRC23:15
tbarronclass PseudoFakeVD(FakeVD):23:15
tbarronhemna's idea I thingk ^^23:16
*** [1]Thelo has quit IRC23:16
tbarronok, I'll stop now23:16
hemna:)23:16
xyang:)23:16
openstackgerritMerged openstack/cinder: Fix dictionary key error  https://review.openstack.org/25245123:18
*** p0rtal has joined #openstack-cinder23:19
*** vgridnev has joined #openstack-cinder23:22
*** p0rtal_ has quit IRC23:23
jgriffithtbarron: I'll get you for that!23:25
jgriffith:)23:25
tbarronjgriffith: thank you, thought you just didn't notice.23:26
tbarron:-)23:26
jgriffithJust delayed, ina meeting23:26
tbarronsorry to hear that23:27
jgriffithhaha23:27
xyangjgriffith, tbarron: maybe we should have a class navigator, just like the project navigator:)23:30
jgriffithxyang: LOL23:30
jgriffithxyang: I kinda think we need one23:30
xyang:)23:30
tbarronxyang: :)23:32
*** ntpttr has joined #openstack-cinder23:34
*** ntpttr has left #openstack-cinder23:35
*** p0rtal_ has joined #openstack-cinder23:35
*** vgridnev has quit IRC23:36
*** dulek has quit IRC23:36
openstackgerritBardia Keyoumarsi proposed openstack/cinder: Volume driver for Coho Data storage solutions  https://review.openstack.org/24669023:36
*** jaypipes has quit IRC23:38
*** p0rtal has quit IRC23:38
*** dulek has joined #openstack-cinder23:39
*** hemna is now known as hemnafk23:42
openstackgerritAlex O'Rourke proposed openstack/cinder: 3PAR: Implement v2 replication (managed)  https://review.openstack.org/23120123:42
*** smoriya_ has joined #openstack-cinder23:44
openstackgerritAlex O'Rourke proposed openstack/cinder: 3PAR: Implement v2 replication (unmanaged)  https://review.openstack.org/23124523:47
*** zohar has quit IRC23:52
*** yhayashi has joined #openstack-cinder23:52
*** bardia has quit IRC23:54
*** ildikov has quit IRC23:56
« Tuesday, 2015-12-01 Index Thursday, 2015-12-03 »

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