Wednesday, 2015-09-30

*** openstackgerrit has quit IRC00:01
*** openstackgerrit has joined #openstack-cinder00:02
*** openstackstatus has quit IRC00:02
*** openstack has joined #openstack-cinder00:03
*** julim has joined #openstack-cinder00:08
mgagneI have no idea on how to backport that patch to stable/kilo, so much changes since :-/ https://review.openstack.org/#/c/222756/300:08
*** devlaps has quit IRC00:09
jgriffithmgagne: I should have it for you tomorrow00:09
mgagnejgriffith: are you the merge conflicts solver as a service? :D00:10
jgriffithmgagne: :)00:11
jgriffithmgagne: so I've been working on backports for other versions00:11
jgriffithmgagne: that's the last one on my list00:11
mgagneoh, so there is work in progress already00:11
jgriffithmgagne: yeah, that's what i meant00:12
mgagnecool, will wait until tomorrow to work on that one then and test it00:14
jgriffithmgagne: sounds good00:14
* jgriffith goes to make some coffee00:15
*** tsekiyam_ has joined #openstack-cinder00:19
*** mtanino has quit IRC00:19
*** tsekiyama has quit IRC00:23
*** tsekiyam_ has quit IRC00:23
*** gouthamr has quit IRC00:28
*** dustins has quit IRC00:36
*** leeantho_ has quit IRC00:37
*** stevemar has joined #openstack-cinder00:40
openstackgerritAnish Bhatt proposed openstack/cinder: Obtain target authentication for CXT from database same as LIO target  https://review.openstack.org/22829800:41
*** Yogi11 has joined #openstack-cinder00:46
*** _cjones_ has quit IRC00:56
*** duobei has joined #openstack-cinder00:57
*** EinstCrazy has joined #openstack-cinder00:58
*** duobei has quit IRC01:02
*** brad[] has joined #openstack-cinder01:05
*** chenke has joined #openstack-cinder01:07
*** sghanekar_ has quit IRC01:08
*** davechen1 has joined #openstack-cinder01:23
*** ociuhandu has joined #openstack-cinder01:24
*** apoorvad_ has joined #openstack-cinder01:30
*** morgan is now known as notreallymorgan01:30
*** notreallymorgan is now known as morgan01:30
*** cdelatte has quit IRC01:31
*** bill_az has quit IRC01:33
*** apoorvad has quit IRC01:33
*** Lee1092 has joined #openstack-cinder01:33
*** apoorvad_ has quit IRC01:34
*** mudassirlatif has quit IRC01:38
openstackgerritWilson Liu proposed openstack/cinder: Add hypermetro support for Huawei driver  https://review.openstack.org/20202301:40
openstackgerritWilson Liu proposed openstack/cinder: Encode username and password in Huawei driver  https://review.openstack.org/21608501:41
openstackgerritWilson Liu proposed openstack/cinder: Fix capacity report error in Huawei driver  https://review.openstack.org/22630801:43
*** ociuhandu has quit IRC01:44
*** kevincarr1991 has joined #openstack-cinder01:46
*** ociuhandu has joined #openstack-cinder01:46
*** IanGovett has quit IRC01:52
*** kevincarr1991 has quit IRC01:55
*** kevincarr1991 has joined #openstack-cinder01:57
*** pv_ has joined #openstack-cinder02:01
pv_hey does run_tests.sh log its results somewhere?02:01
*** stevemar has quit IRC02:02
*** akerr_away is now known as akerr02:03
*** zhenguo has joined #openstack-cinder02:06
*** stevemar has joined #openstack-cinder02:12
*** haomaiwang has joined #openstack-cinder02:15
*** stevemar has quit IRC02:17
*** martyturner has joined #openstack-cinder02:17
*** changbl has joined #openstack-cinder02:18
*** haomaiwang has quit IRC02:22
*** haomaiwa_ has joined #openstack-cinder02:22
*** changbl has quit IRC02:23
*** changbl has joined #openstack-cinder02:25
*** chenying11 has joined #openstack-cinder02:35
*** stevemar has joined #openstack-cinder02:36
*** haomaiwa_ has quit IRC02:41
*** haomaiwang has joined #openstack-cinder02:42
*** changbl has quit IRC02:42
*** stevemar has quit IRC02:49
*** stevemar has joined #openstack-cinder02:50
*** stevemar has quit IRC02:54
*** dims_ has quit IRC02:57
*** salv-orlando has joined #openstack-cinder03:00
*** haomaiwang has quit IRC03:01
*** haomaiwang has joined #openstack-cinder03:01
*** dave-mccowan has quit IRC03:06
*** davechen has joined #openstack-cinder03:16
*** davechen1 has quit IRC03:19
*** stevemar has joined #openstack-cinder03:22
*** stevemar_ has joined #openstack-cinder03:24
*** salv-orlando has quit IRC03:25
*** stevemar has quit IRC03:27
*** stevemar_ has quit IRC03:28
*** changbl has joined #openstack-cinder03:31
*** stevemar has joined #openstack-cinder03:31
*** changbl has quit IRC03:39
*** kevincarr1991 has quit IRC03:44
*** kevincarr1991 has joined #openstack-cinder03:45
*** kevincarr1991 has quit IRC03:47
*** martyturner has quit IRC03:51
*** links has joined #openstack-cinder03:53
*** dims has joined #openstack-cinder03:57
*** Yogi11 has quit IRC03:58
*** haomaiwang has quit IRC04:01
*** haomaiwang has joined #openstack-cinder04:01
*** chenke has quit IRC04:02
*** dims has quit IRC04:03
*** pradipta has joined #openstack-cinder04:05
*** akerr is now known as akerr_away04:06
*** davechen1 has joined #openstack-cinder04:08
*** davechen has quit IRC04:10
*** davechen has joined #openstack-cinder04:15
*** davechen1 has quit IRC04:18
*** chenying11 has quit IRC04:19
*** ociuhandu has quit IRC04:22
*** links has quit IRC04:36
*** kambiz has quit IRC04:37
*** kambiz has joined #openstack-cinder04:41
*** shausy has joined #openstack-cinder04:44
*** mudassirlatif has joined #openstack-cinder04:45
jgriffithpv_: nope, not that I know of04:51
*** salv-orlando has joined #openstack-cinder04:55
*** haomaiwang has quit IRC05:01
*** haomaiwang has joined #openstack-cinder05:01
*** earlephilhower has quit IRC05:02
*** edtubill has quit IRC05:03
*** BharatK has joined #openstack-cinder05:05
*** kfox1111 has quit IRC05:09
*** salv-orlando has quit IRC05:09
*** links has joined #openstack-cinder05:12
*** BharatK has quit IRC05:28
*** alexpilotti has quit IRC05:33
*** nikeshm has quit IRC05:37
*** mudassirlatif has quit IRC05:38
*** BharatK has joined #openstack-cinder05:40
*** mudassirlatif has joined #openstack-cinder05:45
*** deepakcs has joined #openstack-cinder05:51
*** thingee has quit IRC05:56
*** vgridnev has joined #openstack-cinder06:00
openstackgerritDave Chen proposed openstack/python-cinderclient: Add commands to show image metadata  https://review.openstack.org/20936006:00
*** haomaiwang has quit IRC06:01
*** haomaiwang has joined #openstack-cinder06:01
*** nkrinner has joined #openstack-cinder06:01
*** stevemar has quit IRC06:04
*** stevemar has joined #openstack-cinder06:04
*** stevemar has quit IRC06:08
*** mudassirlatif has quit IRC06:12
*** sileht has quit IRC06:17
*** sileht has joined #openstack-cinder06:28
*** ankit_ag has joined #openstack-cinder06:38
*** salv-orlando has joined #openstack-cinder06:41
openstackgerritJuPing proposed openstack/cinder: Fix the bug of "spelling error of a word"  https://review.openstack.org/22926306:47
*** salv-orlando has quit IRC06:48
*** smcginnis has quit IRC06:56
*** haomaiwang has quit IRC07:01
*** haomaiwang has joined #openstack-cinder07:01
*** afazekas_ has joined #openstack-cinder07:01
*** markus_z has joined #openstack-cinder07:02
*** nikeshm has joined #openstack-cinder07:04
nikeshmhi07:04
*** alexpilotti has joined #openstack-cinder07:24
*** chenying11 has joined #openstack-cinder07:25
*** davechen has left #openstack-cinder07:32
*** ronis has joined #openstack-cinder07:32
*** afazekas_ has quit IRC07:40
*** vgridnev has quit IRC07:42
*** vgridnev has joined #openstack-cinder07:43
*** alexpilotti has quit IRC07:58
*** haomaiwang has quit IRC08:00
*** chenying has joined #openstack-cinder08:01
*** chenying11 has quit IRC08:01
*** haomaiwang has joined #openstack-cinder08:01
*** jistr has joined #openstack-cinder08:02
*** e0ne has joined #openstack-cinder08:03
*** salv-orlando has joined #openstack-cinder08:05
*** salv-orlando has quit IRC08:06
*** salv-orlando has joined #openstack-cinder08:07
*** brad[] has quit IRC08:07
*** alexpilotti has joined #openstack-cinder08:13
*** yangyapeng has joined #openstack-cinder08:14
*** stevemar has joined #openstack-cinder08:16
*** breitz has quit IRC08:16
*** breitz has joined #openstack-cinder08:17
*** markvoelker has quit IRC08:20
*** stevemar has quit IRC08:21
*** vgridnev has quit IRC08:22
*** rlucio_ has joined #openstack-cinder08:30
*** rlucio has quit IRC08:34
*** lpetrut has joined #openstack-cinder08:37
*** bluex-pl has joined #openstack-cinder08:40
*** bluex-pl has quit IRC08:41
*** bluex-pl has joined #openstack-cinder08:41
*** jordanP has joined #openstack-cinder08:51
*** brad[] has joined #openstack-cinder08:52
*** e0ne has quit IRC09:00
*** e0ne has joined #openstack-cinder09:00
*** haomaiwang has quit IRC09:01
*** EinstCrazy has quit IRC09:01
*** haomaiwang has joined #openstack-cinder09:01
*** BharatK has quit IRC09:01
*** yangyapeng has quit IRC09:02
*** shausy has quit IRC09:04
*** shausy has joined #openstack-cinder09:04
*** alexpilotti has quit IRC09:04
*** aix has joined #openstack-cinder09:06
*** BharatK has joined #openstack-cinder09:15
*** vgridnev has joined #openstack-cinder09:18
*** markvoelker has joined #openstack-cinder09:21
openstackgerritGorka Eguileor proposed openstack/cinder-specs: Remove Cinder API races  https://review.openstack.org/20710109:21
nikeshmwe tried to install liberty devstack in ubuntu 14.04 with  http://paste.openstack.org/show/474783/ localrc, we are not getting cinder part in dashboard09:22
nikeshmany idea09:23
nikeshmcinder is working in command line only09:24
*** markvoelker has quit IRC09:26
*** chenying has quit IRC09:28
*** EinstCrazy has joined #openstack-cinder09:28
*** dims has joined #openstack-cinder09:32
*** dims has quit IRC09:39
*** EinstCrazy has quit IRC09:40
*** dims has joined #openstack-cinder09:42
*** lprice1 has joined #openstack-cinder09:44
*** lprice has quit IRC09:44
*** martyturner has joined #openstack-cinder09:52
*** martyturner has quit IRC09:56
*** rajesht_ has joined #openstack-cinder09:58
*** aagrawal has joined #openstack-cinder09:58
*** Yogi1 has joined #openstack-cinder09:58
*** haomaiwang has quit IRC10:01
*** ankit_ag has quit IRC10:01
*** haomaiwang has joined #openstack-cinder10:01
*** rajesht has quit IRC10:01
*** Yogi1 has quit IRC10:02
*** EinstCrazy has joined #openstack-cinder10:07
*** IanGovett has joined #openstack-cinder10:13
*** haomaiwang has quit IRC10:15
*** haomaiwang has joined #openstack-cinder10:16
*** jasondotstar_afk is now known as jasondotstar10:19
*** haomaiwang has quit IRC10:21
*** BharatK has quit IRC10:27
*** e0ne has quit IRC10:36
*** BharatK has joined #openstack-cinder10:40
*** shausy has quit IRC10:40
*** shausy has joined #openstack-cinder10:41
*** wanghao has quit IRC10:42
*** shausy has quit IRC10:50
*** jordanP has quit IRC10:59
dulekDuncanT: Hi, I've compiled an Etherpad summarizing the issue with Cinder's AZs and outcomes of discussion on ML.11:09
dulekDuncanT: https://etherpad.openstack.org/p/cinder-availability-zones11:09
dulekDuncanT: Any review if that's a helpful approach would be appreciated. :)11:09
DuncanTDulek: Thanks, I am working on the proposed solution part of the spec, and I can leverage that etherpad for alternatives and references11:10
dulekDuncanT: Okay, feel free to expand it, so it will be helpful to get everybody into discussion at the fishbowl session.11:12
dulekDuncanT: And once you have spec up for reviews it would be cool if you could ping me.11:13
dulekDuncanT: I'm not sure if you're aware of that, but I've compiled a spec for one of the ideas a few weeks earlier: https://review.openstack.org/#/c/22097111:15
dulekDuncanT: May be useful for you. :)11:15
DuncanTI didn't see that spec, no, reading now, thanks11:15
*** e0ne has joined #openstack-cinder11:18
*** EinstCra_ has joined #openstack-cinder11:19
*** EinstCrazy has quit IRC11:21
*** markvoelker has joined #openstack-cinder11:22
*** BharatK has quit IRC11:25
*** markvoelker has quit IRC11:27
*** pradipta has quit IRC11:29
flip214what services do I need to install on a node to be able to export iSCSI LUNs? c-vol? The storage below is done via drbd --11:29
flip214so that is for the case that the c-vol service node doesn't have the exported data locally; then another node needs to do the iSCSI export, so which services need to be active there?11:30
*** julim has quit IRC11:34
*** aarefiev has quit IRC11:35
flip214the controller node would need to tell another node to provide the iSCSI export of a (there-local) block storage device11:35
*** aarefiev has joined #openstack-cinder11:38
*** cdelatte has joined #openstack-cinder11:39
*** julim has joined #openstack-cinder11:39
*** anshul has quit IRC11:41
*** julim has quit IRC11:44
*** akerr_away is now known as akerr11:45
*** akerr has quit IRC11:47
*** takedakn has joined #openstack-cinder11:48
nikeshmhi11:50
nikeshmdid any face the issue?  > installed liberty devstack in ubuntu 14.04 with  http://paste.openstack.org/show/474783/ localrc, not getting Volume Tab dashboard11:52
e0nenikeshm: hi. you probably should ask it in #openstack-horizon11:56
nikeshmok thanks11:56
*** markvoelker has joined #openstack-cinder12:03
*** julim has joined #openstack-cinder12:03
*** Yogi1 has joined #openstack-cinder12:03
*** Yogi11 has joined #openstack-cinder12:06
*** salv-orlando has quit IRC12:06
*** salv-orlando has joined #openstack-cinder12:07
*** Yogi1 has quit IRC12:09
*** deepakcs has quit IRC12:10
*** raildo-afk is now known as raildo12:12
e0necinder api v1 is enabled again:(12:19
*** Yogi11 has quit IRC12:20
*** akerr has joined #openstack-cinder12:20
*** smcginnis has joined #openstack-cinder12:22
*** aix has quit IRC12:22
*** smcginnis has quit IRC12:25
*** smcginnis has joined #openstack-cinder12:27
*** anshul has joined #openstack-cinder12:27
openstackgerritPhilipp Marek proposed openstack/cinder: DRBD: new option "drbdmanage_devs_on_controller".  https://review.openstack.org/22939612:27
*** marcusvrn_ has joined #openstack-cinder12:30
*** julim has quit IRC12:30
*** merooney has joined #openstack-cinder12:34
*** julim has joined #openstack-cinder12:34
*** porrua has joined #openstack-cinder12:35
*** bswartz has quit IRC12:36
*** juzuluag has joined #openstack-cinder12:42
*** haomaiwang has joined #openstack-cinder12:43
*** martyturner has joined #openstack-cinder12:47
*** dave-mccowan has joined #openstack-cinder12:51
*** edmondsw has joined #openstack-cinder12:54
*** aix has joined #openstack-cinder12:55
*** nug has joined #openstack-cinder12:57
*** nug is now known as cfouts12:59
*** jaypipes has quit IRC13:00
*** martyturner has quit IRC13:01
*** haomaiwang has quit IRC13:01
*** haomaiwang has joined #openstack-cinder13:01
openstackgerritYuriy Nesenenko proposed openstack/cinder: Implement extend_volume method to Block Device driver  https://review.openstack.org/22130313:01
*** haomaiwang has quit IRC13:02
*** haomaiwang has joined #openstack-cinder13:02
e0ne"Currently almost no vendors support cinder v2, this means that v2 as a default is a thing that is more likely to cause a user pain. HOWEVER ..." :'(13:02
*** martyturner has joined #openstack-cinder13:03
openstackgerritYuriy Nesenenko proposed openstack/cinder: Implement snapshots-related features for Block Device Driver  https://review.openstack.org/22229213:05
*** zhipeng has joined #openstack-cinder13:07
*** raildo is now known as raildo-afk13:12
*** jungleboyj has quit IRC13:12
*** julim has quit IRC13:13
*** julim has joined #openstack-cinder13:15
*** julim has quit IRC13:16
*** xyang1 has joined #openstack-cinder13:16
*** Yogi1 has joined #openstack-cinder13:18
*** dustins has joined #openstack-cinder13:18
*** mriedem has joined #openstack-cinder13:19
*** eharney has joined #openstack-cinder13:21
*** mriedem has quit IRC13:23
*** mriedem has joined #openstack-cinder13:26
*** aagrawal has quit IRC13:26
*** julim has joined #openstack-cinder13:27
*** jgregor has joined #openstack-cinder13:30
*** alejandrito has joined #openstack-cinder13:30
*** juzuluag has quit IRC13:31
*** Yogi11 has joined #openstack-cinder13:32
*** takedakn has quit IRC13:32
*** raildo-afk is now known as raildo13:33
*** Yogi1 has quit IRC13:34
*** ccesario has joined #openstack-cinder13:37
*** takedakn has joined #openstack-cinder13:40
*** william has joined #openstack-cinder13:44
*** william is now known as Guest4963913:45
*** julim has quit IRC13:45
*** takedakn has quit IRC13:47
*** baumann has joined #openstack-cinder13:47
*** dims has quit IRC13:49
*** dims has joined #openstack-cinder13:49
*** ociuhandu has joined #openstack-cinder13:50
*** julim has joined #openstack-cinder13:51
*** mc_nair has joined #openstack-cinder13:52
*** gouthamr has joined #openstack-cinder13:54
*** thangp has joined #openstack-cinder13:54
*** stevemar has joined #openstack-cinder13:56
*** jaypipes has joined #openstack-cinder13:56
*** mriedem is now known as mriedem_away13:57
*** merooney has quit IRC13:58
*** mc_nair has quit IRC13:58
*** lprice has joined #openstack-cinder13:59
*** julim has quit IRC14:00
*** haomaiwang has quit IRC14:01
*** ntpttr has joined #openstack-cinder14:01
*** haomaiwang has joined #openstack-cinder14:01
openstackgerritJoel Coffman proposed openstack/cinder-specs: Add volume encryption metadata spec  https://review.openstack.org/17605014:01
*** lprice1 has quit IRC14:01
*** merooney has joined #openstack-cinder14:02
*** mc_nair has joined #openstack-cinder14:02
*** timcl1 has quit IRC14:04
*** timcl has joined #openstack-cinder14:04
*** bswartz has joined #openstack-cinder14:06
*** julim has joined #openstack-cinder14:07
*** haomaiwang has quit IRC14:14
*** haomaiwa_ has joined #openstack-cinder14:15
*** j_king has left #openstack-cinder14:17
jgriffithsmcginnis: api v1/v2 might be a good topic for todays meeting ya think?14:18
smcginnisjgriffith: Ya, maybe. :)14:18
jgriffith:)14:18
*** porrua has quit IRC14:18
smcginnisjgriffith: I did see you have v1 removal on there. Good!14:19
*** porrua has joined #openstack-cinder14:19
jgriffithIt'd be good if we good set up a tiger team or something to get all these incompatabilities figured out/tested14:19
jgriffithOh... yeah... sorry14:19
e0ne:)14:19
jgriffithI wanted to check with you, I put two things on there14:19
smcginnisYeah, I definitely think we need a wider picture of everything.14:20
jgriffithI pinged you the other night but it was late, so I just posted them :)14:20
smcginnisjgriffith: Those look good.14:20
jgriffithk14:20
smcginnisYeah, cool. Deep asleep by then.14:20
smcginnisSomehow I've been turning in to a morning person.14:20
smcginnisNot sure if I like that or not. ;)14:20
jgriffith:)  welcome!!!14:20
jgriffithMornings are awesome, quiet, birds.. that fresh morning smell14:21
smcginnis+114:21
jgriffithIn a couple weeks though I'm going back to sleeping in.  Nothing but cold, gray and probably snow :(14:21
smcginnisMuch better when you're next to a mountain, I'm sure.14:21
jgriffithThat does help :)14:21
*** lcurtis has joined #openstack-cinder14:22
e0newe can't disable v1 w/o https://review.openstack.org/#/c/229039/ :(14:22
*** lprice has left #openstack-cinder14:22
*** Swanson has quit IRC14:22
*** asselin__ has joined #openstack-cinder14:22
smcginnise0ne: I'm thinking at this point we can't disable v1 for a little longer yet.14:22
jgriffithsmcginnis: I'm honestly wondering if V2 is a failed effort :(14:23
*** lprice has joined #openstack-cinder14:23
jgriffithbut yeah, I agree with smcginnis I don't think we can disable it for quite a while14:23
e0nesmcginnis: we need to option for devstack to disable it. and try this for every project:(14:23
smcginnisjgriffith: I don't think failed...14:23
*** Swanson has joined #openstack-cinder14:24
smcginnisBut not as easy as a transition as hoped for sure.14:24
e0nesmcginnis: I don't know how it could be tested on infra14:24
jgriffithnahh... not failed, but dropped14:24
jgriffithnot done14:24
jgriffithneeds help14:24
jgriffithand it's sadly been a very very long evolution14:25
smcginnisVery true.14:25
jgriffiththe bummer is that most of the changes are response codes, not even missing/modified methods14:25
jgriffithwhich is the most annoying thing of all14:25
*** asselin_ has quit IRC14:26
smcginnisjgriffith: They are "more correct"14:26
smcginnisBut still different14:26
jgriffithindeed14:26
jgriffithsometimes correct doesn't matter14:26
jgriffith:(14:26
smcginnisHonestly in most cases I think we are to explicit in checking results. Whether it's 200 or 201, both mean "SUCCESS".14:26
jgriffithI agree with you, but others don't seem to see it that way14:28
openstackgerritTom Swanson proposed openstack/cinder: Dell SC: cgsnapshot-delete doesn't actually delete  https://review.openstack.org/22900714:30
*** anshul has quit IRC14:30
jgriffithsmcginnis: the only thing I find is interesting is how many times keystone and glance break the world, and I mean break as in take away interfaces, change them etc14:31
mtreinishjgriffith: the issues with the v1 drop here are the same as those projects have been going through for a long time14:31
mtreinishthere's a reason major api version drops have not really happened before14:32
jgriffithmtreinish: well, kinda14:32
smcginnismtreinish: Sounds like as a whole we need to decide how to migrate to new versions?14:32
smcginnisjgriffith: Double standards? No way!14:32
smcginnis:)14:32
jgriffithsmcginnis: oh.. wait, no... I wasn't saying that14:33
mtreinishsmcginnis: my opinion is we need to support old and new, dropping the old just doesn't happen14:33
jgriffithIt may be true or not. but I didn't say it :)14:33
mtreinishjgriffith: heh, I see what you did there :)14:33
smcginnisjgriffith: ;)14:33
smcginnismtreinish: Yeah, I think we need to find a good way to help folks move to the new API.14:33
mtreinishsmcginnis: which is why things like microversions were introduced14:33
smcginnismtreinish: So it actually is possible some day to remove the old one.14:33
jgriffithHonestly what I was getting at is that I think we need a better design for all of this as a community (all of OpenStack)14:33
mtreinishto make moving foward easier but keep support for the old14:34
smcginnisjgriffith: Definitely agree with that.14:34
*** mriedem_away is now known as mriedem14:34
jgriffithOr change our philosophy on API's altogether14:34
*** mudassirlatif has joined #openstack-cinder14:34
jgriffithLike, nothing ever goes away; anything post V1 you provide a version in the call itself14:35
jgriffithso for example, you add a new V3 method that foos a volume;14:35
mtreinishjgriffith: have you looked at how microversions work?14:36
jgriffiththe call looks like:  volumes.foo(arg, version=3)14:36
jgriffithmtreinish: briefly14:36
jgriffithI need to spend more time on that14:36
mtreinishit's the same basic idea, you're very explicit with versioning up front and they're tied to specific changes to the api14:37
mtreinishand then the server supports a range of versions14:37
jgriffithinteresting14:37
jgriffithYeah, that's pretty much exactly the sort of thing I was thinking.  I"ll have to look at the Nova code and see what they have there14:38
smcginnisjgriffith: Manila has been going through the conversion.14:38
mtreinishjgriffith: sdague had a blog post that explained it at a high level pretty well14:38
smcginnisI think bswartz agreed to be a resource for us on that.14:38
jgriffithI work on another API currently that has a model like that, and it works pretty well, it means you never take anything away or break anything but I kinda think that's good14:38
jgriffithmtreinish: looking for it now...14:38
johnthetubaguyjgriffith: in case this helps: http://specs.openstack.org/openstack/nova-specs/specs/kilo/implemented/api-microversions.html14:39
jgriffithjohnthetubaguy: thanks14:39
*** edtubill has joined #openstack-cinder14:39
johnthetubaguyjgriffith: I think ironic had quide a detailed spec too, I think there were plans for an API-WG doc on microversions, but I don't remember how far that got14:39
bswartzdid someone say my name and microversions?14:40
jgriffithjohnthetubaguy: interesting14:40
* jgriffith has apparantly been under a rock for a while14:40
bswartzin manila we have experience with microversions and it's mostly good, but there are some surprises14:41
smcginnisbswartz: Just mentioning you and Manila have done a lot of work there.14:41
*** dave-mccowan has quit IRC14:41
bswartzI'm happy to talk about it in tokyo or wherever14:41
jgriffithbswartz: yeah, I do recall a very brief conversation with you on that... but I never went and did my homework :(14:41
smcginnisbswartz: We would love any warning signs you can give us on that path.14:41
bswartzjgriffith and I were considering a joint talk on microversions+experimental APIs14:41
bswartzit could still happen14:42
scottdaI have a spec up for Cinder microversions: https://review.openstack.org/#/c/223803/314:42
smcginnisbswartz: +114:42
scottdaAnd POC code: https://review.openstack.org/#/c/224910/314:42
bswartzscottda: awesome14:42
*** haomaiwa_ has quit IRC14:42
jgriffithjohnthetubaguy: mtreinish smcginnis cool stuff, yeah, this is exactly the sort of thing I had in mind (only because I've worked on a project that did something similar)14:42
scottdaIt's mostly ported from Manila14:42
*** haomaiwang has joined #openstack-cinder14:43
bswartzscottda: did you port anything related to experimental yet?14:43
johnthetubaguyjgriffith: sdague has a good post on the nova side of this: https://dague.net/2015/06/05/the-nova-api-in-kilo-and-beyond-2/14:43
mtreinishjgriffith: https://dague.net/2015/06/05/the-nova-api-in-kilo-and-beyond-2/14:43
scottdaNo, I skipped the experimental for now...I figured on just getting the basic microversion in first.14:43
jgriffithhaha14:43
mtreinishah, johnthetubaguy beat me to it14:43
johnthetubaguy:)14:43
*** jungleboyj has joined #openstack-cinder14:44
*** anshul has joined #openstack-cinder14:44
scottdabswartz: I probably need to talk to you Manila folk about the surprises. I'm seeing what I think is a race in how one of the decorators for microversions is working.14:45
scottdas/is/are/14:45
*** tsekiyama has joined #openstack-cinder14:45
bswartzscottda: pull in me and cknight14:45
jgriffithmtreinish: johnthetubaguy very nice blog from sdague14:45
bswartzalso cfouts for microversion client-related changes14:46
bswartzthe client was where most of the surprises came from14:46
jgriffithAnd it does touch on some other things I've thought about in the past and points out why they suck :)14:46
scottdaWhat time zone are they in? I'll query you in the Manila channel14:46
bswartzregarding experimental -- it's not as simple as I'd hoped but we're working through the complexity14:46
bswartzscottda: we're all east coast14:46
scottdacool.14:47
bswartzhowever wednesday is meeting hell day around here14:47
scottdaBTW, patrickeast has already put a fishBowl talk on the Summit agenda for experimental APIs, and there's a working group for microversions.14:47
bswartzso I suggest you aim for thurs or fri14:47
scottdawill do.14:47
*** mtanino has joined #openstack-cinder14:49
jgriffithexcellent14:52
jgriffith:q14:52
jgriffithoops14:52
e0ne:wq!14:53
e0ne:)14:53
*** bswartz has quit IRC14:53
*** kevincarr1991 has joined #openstack-cinder14:55
jgriffithe0ne: :)14:55
*** dave-mccowan has joined #openstack-cinder14:57
*** haomaiwang has quit IRC15:01
*** r-daneel has joined #openstack-cinder15:01
*** haomaiwang has joined #openstack-cinder15:01
*** timcl1 has joined #openstack-cinder15:02
*** zhipeng has quit IRC15:02
*** ronis has quit IRC15:03
*** ntpttr has quit IRC15:03
*** timcl has quit IRC15:03
*** merooney has quit IRC15:05
*** merooney has joined #openstack-cinder15:05
*** bluex-pl has quit IRC15:06
*** jistr has quit IRC15:07
*** ntpttr has joined #openstack-cinder15:09
*** haomaiwang has quit IRC15:10
*** haomaiwa_ has joined #openstack-cinder15:10
*** vgridnev has quit IRC15:12
*** hemnafk is now known as hemna15:14
hemnamornin15:15
*** pv_ has quit IRC15:15
*** dims_ has joined #openstack-cinder15:16
*** haomaiwa_ has quit IRC15:19
*** dims has quit IRC15:20
*** haomaiwang has joined #openstack-cinder15:20
*** links has quit IRC15:21
scottdamorning15:24
*** doffm has joined #openstack-cinder15:26
doffmHi all, does anyone know the status of snapshot rollback work in cinder?15:28
*** julim has quit IRC15:28
doffmI found a couple of old patches / blueprints. https://review.openstack.org/#/c/104127/15:29
doffmHas anything more recent happened?15:29
*** julim has joined #openstack-cinder15:31
jgriffithdoffm: nope, that's pretty much dead15:35
jgriffithdoffm: the answer for now is just "create volume from snapshot"15:35
*** julim has quit IRC15:36
*** nkrinner has quit IRC15:37
doffmjgriffith: Thanks.15:37
*** julim has joined #openstack-cinder15:37
*** haomaiwang has quit IRC15:37
*** haomaiwang has joined #openstack-cinder15:38
*** thangp has quit IRC15:41
*** srikanth_p has joined #openstack-cinder15:42
*** salv-orl_ has joined #openstack-cinder15:43
*** diablo_rojo1 has joined #openstack-cinder15:44
*** diablo_rojo has quit IRC15:44
*** jgregor has quit IRC15:44
*** mriedem1 has joined #openstack-cinder15:44
*** mriedem has quit IRC15:44
*** baumann1 has joined #openstack-cinder15:44
*** baumann has quit IRC15:44
*** jgregor has joined #openstack-cinder15:44
*** diablo_rojo has joined #openstack-cinder15:44
*** salv-orlando has quit IRC15:46
*** adurbin_ has joined #openstack-cinder15:46
*** jungleboyj has quit IRC15:47
*** diablo_rojo1 has quit IRC15:49
*** rhagarty has joined #openstack-cinder15:50
*** rhagarty_ has joined #openstack-cinder15:51
smcginnisjgriffith: Do you know what I need to do to get launchpad permissions to target blueprints and things?15:52
jgriffithsmcginnis: hmm.. lemme look15:52
*** haomaiwang has quit IRC15:53
smcginnisjgriffith: Thanks15:53
*** haomaiwang has joined #openstack-cinder15:53
*** dosaboy_ is now known as dosaboy15:54
jgriffithsmcginnis: try it now15:54
jgriffithOh wait.. one other thing I need to change15:55
smcginnisjgriffith: Hey, look at all the pretty yellow circles.15:55
jgriffithsmcginnis: ok, and you are now an admin15:55
jgriffithgo forth and administrate15:56
smcginnisAlready feeling drunk with power.15:56
smcginnisjgriffith: Thank you!15:56
*** aix has quit IRC15:57
*** thingee has joined #openstack-cinder15:57
*** merooney has quit IRC15:59
*** rajinir has joined #openstack-cinder16:00
*** martyturner has quit IRC16:00
scottdasmcginnis: Will you use your power to start the meeting, or are you too drunk?16:00
jgriffithLOL16:00
*** jungleboyj has joined #openstack-cinder16:00
*** haomaiwang has quit IRC16:01
*** haomaiwa_ has joined #openstack-cinder16:01
*** baumann1 has left #openstack-cinder16:02
*** baumann1 has joined #openstack-cinder16:02
*** haomaiwa_ has quit IRC16:04
*** gouthamr has quit IRC16:04
*** erlon has joined #openstack-cinder16:05
*** jwcroppe has quit IRC16:05
*** bswartz has joined #openstack-cinder16:06
*** dims_ has quit IRC16:06
*** haomaiwang has joined #openstack-cinder16:07
*** Yogi11 has quit IRC16:07
*** Yogi1 has joined #openstack-cinder16:09
*** haomaiwang has quit IRC16:11
*** haomaiwa_ has joined #openstack-cinder16:11
*** breitz has quit IRC16:13
*** zhenguo has quit IRC16:14
*** raildo is now known as raildo-afk16:14
*** breitz has joined #openstack-cinder16:14
*** thangp has joined #openstack-cinder16:15
*** leeantho has joined #openstack-cinder16:16
*** adurbin__ has joined #openstack-cinder16:17
*** adurbin_ has quit IRC16:17
*** IlyaG has joined #openstack-cinder16:18
*** haomaiwa_ has quit IRC16:18
*** haomaiwang has joined #openstack-cinder16:19
*** haomaiwang has quit IRC16:20
*** haomaiwa_ has joined #openstack-cinder16:20
*** lpetrut has quit IRC16:25
*** shausy has joined #openstack-cinder16:27
*** raildo-afk is now known as raildo16:30
*** haomaiwa_ has quit IRC16:33
*** vgridnev has joined #openstack-cinder16:33
*** dims_ has joined #openstack-cinder16:33
*** stevemar has quit IRC16:35
*** ronis has joined #openstack-cinder16:35
*** markus_z has quit IRC16:36
*** garthb has joined #openstack-cinder16:36
*** _cjones_ has joined #openstack-cinder16:42
*** _cjones_ has joined #openstack-cinder16:42
*** rhedlind-m has quit IRC16:44
*** jwcroppe has joined #openstack-cinder16:44
*** shausy has quit IRC16:47
*** martyturner has joined #openstack-cinder16:47
*** apoorvad has joined #openstack-cinder16:48
*** devlaps has joined #openstack-cinder16:50
*** ntpttr has quit IRC16:51
*** harlowja has quit IRC16:52
*** merooney has joined #openstack-cinder16:52
*** Yogi1 has quit IRC16:54
kmartingood job smcginnis, your hired!17:00
*** gouthamr has joined #openstack-cinder17:00
eharneywe still need to land https://review.openstack.org/#/c/228160/17:00
eharneyand https://review.openstack.org/#/c/228107/17:01
jungleboyjkmartin: Hmmm, he missed that first topic change though.   ;-)17:01
hemnaeharney, thierry still has that one blocked17:01
eharneyhemna: sure, but, we need to get these in rc2, so... when are we doing that?17:01
smcginniskmartin: Thank you sir. Not too drunk I guess. LOL17:01
thingeexyang1, smcginnis: yes, we'll have a backport once we know what to base that off of.17:01
kmartinjungleboyj, one mistake is okay for his first time17:01
*** mudassirlatif_ has joined #openstack-cinder17:01
smcginnisjgriffith: Wha?!?17:01
xyang1thingee: thanks17:01
smcginnisthingee: Thanks!17:02
xyang1kmartin: we will find out pretty soon if he gets drunk:)17:02
*** jgregor has quit IRC17:02
jungleboyjkmartin: ++17:02
smcginnisxyang1: What happens at the summit, stays at the summit. :)17:02
smcginnisUnless it gets put in an etherpad.17:03
xyang1smcginnis: :)17:03
smcginnisBut then you can never find it again anyway, so that's safe.17:03
*** hemna is now known as hemnafk17:03
jungleboyjxyang1: I might have insider information.  ;-)17:03
smcginnisjungleboyj: Shut your mouth.17:03
xyang1jungleboyj: I'll chat with you at the summit:)17:04
jungleboyjsmcginnis: I didn't say anything.17:04
*** mudassirlatif has quit IRC17:04
*** mudassirlatif_ is now known as mudassirlatif17:04
*** Guest49639 has quit IRC17:04
thingeeThe Cinder def core work happening https://review.openstack.org/#/c/22163117:04
*** william has joined #openstack-cinder17:04
*** william is now known as Guest1560817:04
smcginnisthingee: Oh, nice.17:05
smcginnisOff to food before my next meeting...17:05
jgriffiththingee: what does "advisory" mean?17:06
jgriffithnm... reading the doc17:07
*** stevemar has joined #openstack-cinder17:07
*** earlephilhower has joined #openstack-cinder17:08
thingeejgriffith: a recommendation17:10
thingeewhoops smcginnis ^17:10
*** srikanth_p has quit IRC17:11
*** Yogi1 has joined #openstack-cinder17:12
*** IlyaG has quit IRC17:13
*** IlyaG has joined #openstack-cinder17:14
*** e0ne has quit IRC17:14
*** aix has joined #openstack-cinder17:19
openstackgerritPhilipp Marek proposed openstack/cinder: DRBD: new option "drbdmanage_devs_on_controller".  https://review.openstack.org/22939617:21
*** mriedem1 has quit IRC17:21
*** ociuhandu has quit IRC17:22
*** mriedem has joined #openstack-cinder17:22
*** thingee_ has joined #openstack-cinder17:23
*** stevemar has quit IRC17:24
*** thingee has quit IRC17:25
*** stevemar has joined #openstack-cinder17:26
*** mudassirlatif has quit IRC17:32
*** akerr has quit IRC17:32
*** lpetrut has joined #openstack-cinder17:34
*** IlyaG has quit IRC17:34
*** shakamunyi has quit IRC17:35
*** IlyaG has joined #openstack-cinder17:35
*** ntpttr has joined #openstack-cinder17:37
*** thingee_ has quit IRC17:38
*** baumann1 has left #openstack-cinder17:43
*** garthb has quit IRC17:46
*** ntpttr has quit IRC17:48
*** ntpttr has joined #openstack-cinder17:48
*** stevemar has quit IRC17:57
*** stevemar has joined #openstack-cinder17:57
*** ntpttr has quit IRC18:00
*** shakamunyi has joined #openstack-cinder18:01
*** stevemar has quit IRC18:01
openstackgerritDiem Tran proposed openstack/cinder: Fix ZFSSA drivers' local cache bugs  https://review.openstack.org/22955418:03
*** hemnafk is now known as hemna18:04
*** ntpttr has joined #openstack-cinder18:04
*** apoorvad_ has joined #openstack-cinder18:05
*** dzamboni has joined #openstack-cinder18:05
* nikeshm slaps smcginnis around a bit with a large fishbot18:08
openstackgerritSonia Ghanekar proposed openstack/cinder: Add multi-initiator extra-spec for Nimble driver  https://review.openstack.org/22918518:08
*** apoorvad has quit IRC18:08
smcginnisnikeshm: Fine, be that way. :)18:10
patrickeasttsekiyama: hemna: you guys know whats up with https://review.openstack.org/#/c/196482/  in regards to it making it in to L?18:10
hemnapatrickeast, so I was working with tsekiyama on that one for a while18:11
hemnato fix the searching paths for the multipath device18:11
patrickeastyea it looks like its in good shape now18:11
hemnaI think it's good to go18:11
patrickeastjust sitting there with a bunch of +1's18:12
hemnaI think it's backport potential for L18:12
*** harlowja has joined #openstack-cinder18:13
*** e0ne has joined #openstack-cinder18:13
*** openstackgerrit has quit IRC18:16
nikeshmsmcginnis: that message generated by mistake :)18:16
*** openstackgerrit has joined #openstack-cinder18:17
*** anshul has quit IRC18:19
openstackgerritNate Potter proposed openstack/cinder: Add backup_swift_auth_url to swift backup driver  https://review.openstack.org/22864618:20
*** lpetrut has quit IRC18:20
*** garthb has joined #openstack-cinder18:22
*** e0ne has quit IRC18:23
*** mudassirlatif has joined #openstack-cinder18:23
*** Guest15608 has quit IRC18:24
*** e0ne has joined #openstack-cinder18:29
*** aix has quit IRC18:30
*** tsekiyama has quit IRC18:34
*** jwang_ has quit IRC18:34
*** julim has quit IRC18:41
*** Guest15608 has joined #openstack-cinder18:42
*** baumann has joined #openstack-cinder18:42
*** julim has joined #openstack-cinder18:43
*** Guest15608 has quit IRC18:43
*** apoorvad_ has quit IRC18:43
*** apoorvad has joined #openstack-cinder18:44
*** william has joined #openstack-cinder18:44
*** william is now known as Guest8237018:45
SwansonAny cores around to take a peek at https://review.openstack.org/#/c/229007/ ?  Got one +2 now.  I would like to port it to liberty.  (I'm presuming I can't make the release.)18:46
*** dims_ has quit IRC18:46
openstackgerritNate Potter proposed openstack/cinder: Add backup_swift_auth_url to swift backup driver  https://review.openstack.org/22864618:46
*** openstackgerrit has quit IRC18:46
*** aix has joined #openstack-cinder18:46
*** dims_ has joined #openstack-cinder18:46
*** openstackgerrit has joined #openstack-cinder18:47
*** merooney has quit IRC18:50
*** martyturner has quit IRC18:51
*** martyturner has joined #openstack-cinder18:53
*** aix has quit IRC18:55
*** david-lyle has quit IRC18:55
*** david-lyle has joined #openstack-cinder18:55
*** angela-s has joined #openstack-cinder18:59
e0nejgriffith, hemna: fyi, I tried to address all your comments about block device driver in the openstack-dev ml19:03
openstackgerritxing-yang proposed openstack/cinder: CG driver function should not access db  https://review.openstack.org/22958019:05
xyang1hemna: patrickeast ^19:06
hemnae0ne, url ?19:07
*** mudassirlatif has quit IRC19:07
hemnaxyang1, ooh19:07
e0nehemna: btw, I almost finished first version of brickclient, I need to do some code cleanup before push it to github19:07
xyang1hemna: let me know if anything else needs to be passed in19:07
patrickeastxyang1: \o/ yay19:08
e0nehemna: http://lists.openstack.org/pipermail/openstack-dev/2015-September/075883.html19:08
*** mudassirlatif has joined #openstack-cinder19:09
*** mudassirlatif has quit IRC19:11
*** dave-mccowan has quit IRC19:11
hemnaxyang1, https://github.com/openstack/cinder/blob/master/cinder/volume/drivers/san/hp/hp_3par_common.py#L451-L45619:11
*** merooney has joined #openstack-cinder19:12
hemnaxyang1, so delete_consistencygroup returns a model_update and a list of volumes.19:12
hemnaxyang1, is that still needed?19:12
hemnaxyang1, I have a feeling that updating the volume.status is probably not a great thing to do there.19:12
xyang1hemna: i did not change driver funtions yet19:12
hemnaxyang1, yah, I'm asking19:12
hemnaxyang1, I could submit a follow up for that file for us19:13
xyang1hemna: so manager is expecting a status change19:13
hemnaxyang1, it feels wrong for a driver to be updating the status field of a volume object19:13
xyang1hemna: based on that, manager will update19:14
xyang1hemna: that is the same model as other methods19:14
hemnawhere?19:14
xyang1hemna: if there something happens, manager does not know19:14
xyang1hemna: create volume19:15
hemnaxyang1, ?!  afaik drivers send back a model update which the volume manager persists.19:16
hemnaI've never seen a driver manually change the status field of a volume itself.19:16
hemnamaybe I'm high19:16
jgriffithhemna: not directly usually, but typically something embedded in the response or a known exception19:17
jgriffithhemna: that then lets the manager do the db update19:17
xyang1hemna: the idea is driver needs to tell manager the status of volumes, this is a little different too as there is a group and ind vol status19:17
*** _cjones_ has quit IRC19:17
hemnahttps://github.com/openstack/cinder/blob/master/cinder/volume/drivers/san/hp/hp_3par_common.py#L45419:17
hemnathat just seems 'wrong' ?!19:17
smcginnisxyang1: Were you saying that? That the driver should be returning a model update?19:17
hemnaI dunno, maybe I'm paranoid19:18
smcginnisSo the 3par imlementation should change?19:18
xyang1smcginnis: right19:18
smcginnisxyang1: Thanks, that's what I thought.19:18
xyang1smcginnis: then manager updates based on that19:18
smcginnisI think you're both saying the same thing ultimately.19:18
jgriffithxyang1: +119:18
jgriffithsmcginnis: :)19:18
jgriffithhemna: you still there?19:20
hemnasorry19:20
hemnahad a local interrupt19:21
hemnaso, what's the right thing to do here?19:21
hemnaso the change that xyang1 has to pass in the list of volumes to delete_consistencygroup19:21
hemnaso we don't have to fetch them.19:21
hemnashouldn't the manager set the status to 'deleted' itself after a successful call to the driver's delete_cg ?19:22
*** david-lyle has quit IRC19:22
jgriffithhemna: it does19:22
smcginnisYeah, so I think that 3par implementation needs to just pass back a model update instead of performing that internally.19:22
*** david-lyle has joined #openstack-cinder19:22
jgriffithhemna: https://github.com/openstack/cinder/blob/master/cinder/volume/manager.py#L261519:23
hemnajgriffith, was just looking at that19:23
jgriffithhemna: is that what you're looking for?19:23
*** IlyaG has quit IRC19:23
hemnathe volume manager sets the status what exists in the volume object itself19:23
hemnaso the 3PAR code as it stands is required in order to mark that volume as deleted.19:23
hemnameaning we can't remove this line: https://github.com/openstack/cinder/blob/master/cinder/volume/drivers/san/hp/hp_3par_common.py#L45419:24
xyang1hemna: so manager depends on driver to return status, because here we have a group and volumes19:24
hemnayah I get that19:25
*** jgregor has joined #openstack-cinder19:25
xyang1hemna: not as straight forward as one vol19:25
hemnaI guess I'm asking is it 'ok' for drivers to be touching the status field of a volume itself.19:25
hemnajust in general19:25
*** stevemar has joined #openstack-cinder19:26
hemnaseems like a grey area to me19:26
smcginnisOh, looking at the code a little closer. I _think_ that's right as is.19:26
jgriffithhemna: xyang1 the other alternative you guys could look at is changing the method to return a 'update struct' with the info that gets used instead of the volume object19:26
xyang1hemna: in this case, if you don't tell me, you will have to throw exception19:26
xyang1jgriffith: i am looking at that too19:27
hemnajgriffith, yah19:27
xyang1jgriffith: but the snapshot one needs volume id19:27
xyang1jgriffith: so I figured it is easier to keep as is19:27
hemnais it wrong to assume a successful call to a driver's delete_cg results in each volume being deleted ?19:27
xyang1hemna: if everything is successful, that is fine19:28
xyang1hemna: problem is if you get it half way19:28
Swansonhemna: xyang1: thanks for the +2's19:29
hemnaxyang1, so if the driver pukes on deleting the 3rd volume out of 5, and raises, then we bail down into the exception area in the volume manager anyway19:29
xyang1hemna: so we could set everything as succesful if no exception19:29
xyang1hemna: otherwise everything fail19:29
*** IlyaG has joined #openstack-cinder19:29
hemnaso looking at this closer it seems that the volume manager assumes that a driver may not raise an exception during the delete_cg call19:30
hemnaand will update the 3rd of 5 volumes status to !deleted19:30
*** IlyaG has quit IRC19:31
hemnathen https://github.com/openstack/cinder/blob/master/cinder/volume/manager.py#L2622-L262519:31
*** timcl has joined #openstack-cinder19:31
xyang1hemna: there is exception catch there19:31
xyang1hemna: I was trying to handle both19:31
hemnayah19:31
hemnaour driver might raise during a delete_volume call19:32
hemnaso19:32
xyang1hemna: the current code handles both19:32
hemnashould we wrap our loop over delete_volume and catch those ourselves?19:32
xyang1hemna: so if you raise, it shoukd be fine too19:32
hemnaso that prevents the entire action from getting stuck in an error_deleting19:32
hemnaat least on the cg19:33
*** timcl1 has quit IRC19:33
hemnaok I guess that happens anyway19:34
xyang1hemna: the group will be error deleting if you set status to error19:34
xyang1hemna: if you catch, what are you going to return19:34
hemnaok, so if we don't update the volume itself, then the vmanager has no way of making sure the entire thing worked19:35
*** david-lyle has quit IRC19:35
xyang1hemna: right, as it does not know about ind vol19:35
hemnakambiz, well it just seems odd to be touching the volume status directly from the driver's perspective19:35
*** baumann has quit IRC19:35
*** timcl has quit IRC19:36
hemna*sigh* xchat19:36
kambizhemna: ?19:36
kambizoh .. sorry19:36
* kambiz frowns at <tab>19:36
hemnakambiz, my bad.  it's xchat's tab completion. sorry.19:36
*** diablo_rojo1 has joined #openstack-cinder19:36
kambizno worries ... weechat does the same to me at times.19:37
*** akerr has joined #openstack-cinder19:38
*** diablo_rojo has quit IRC19:38
hemnaxyang1, so i'm just trying to figure out if we can somehow get away from touching the volume status in the driver itself19:38
*** baumann has joined #openstack-cinder19:38
hemnaand still report failure/success correctly19:38
jgriffithhemna: I'll try this one more time....19:40
jgriffithhemna: You could modify the manager call to take an update structure on response19:40
hemnaso in general in other driver calls, if there is a failure, isn't it assumed there is an exception raised to the vmanager ?19:40
jgriffithand let it do the updates on the volume objects itself19:40
hemnayah that's kind what I was thinking19:40
hemnaI mean if we are doing this sort of thing in other places, then I suppose it's not a big deal19:41
hemnajust trying to understand the vmanager -> driver contract as best as I can.19:41
xyang1jgriffith: so right now it returns model_updat for group and volumes19:41
jgriffithxyang1: yeah, I know19:41
jgriffithxyang1: hemna just doesn't like that it touches the volume object directly19:41
xyang1jgriffith: we can change volumes to be an [] of dictionaries19:41
hemnajgriffith, am I just being silly?19:42
jgriffithxyang1: I have no opinion on that really... but was just pointing out that there's a fairly trivial middle ground19:42
xyang1jgriffith: is that what you mean19:42
jgriffithhemna: I don't know.. I kinda see your point.  It's more about consistency IMHO19:42
xyang1hemna: will you be ok with dictionay19:42
*** dave-mccowan has joined #openstack-cinder19:42
jgriffithhemna: consistency in terms of "drivers don't muck with the volume objects"19:42
hemnayah, that's kinda what was getting at in a terribly worded and completely convoluted/confusing way19:42
*** akerr_ has joined #openstack-cinder19:42
jgriffithI don't know if that's good/bad or matters19:42
xyang1hemna: jgriffith in other words, model update for a volume19:43
hemnayah19:43
jgriffithxyang1: yeah... so you could just say something like delete_cgs ... return (model_update, [{volume_status_updates: { blah blah}}]19:44
xyang1jgriffith: ya, that is better19:44
hemnaaye19:44
jgriffithxyang1: it's really NO different than what you have, it just separates the two interests19:44
xyang1jgriffith: hemna that was the intention19:44
xyang1But everyone takes the short cut19:44
jgriffithxyang1: :)19:44
jgriffithxyang1: Yeah, it's really just a short cut19:45
*** thingee has joined #openstack-cinder19:45
xyang1I just need to figure out create_cgsnapshot19:45
xyang1That one needs volume id19:45
jgriffithxyang1: but not even necessarily, because if the manager was ever changed to not do the db updates off of those they'd be lost anyway19:45
*** akerr has quit IRC19:46
jgriffithxyang1: you have code for that?19:46
hemnaok sorry for causing a fuss19:46
xyang1jgriffith: which one?19:46
jgriffithhemna: nahh, I think it was a really good discussion19:46
*** markstur has joined #openstack-cinder19:46
xyang1Cgsnapshot?19:46
jgriffithxyang1: for the 'create cg snaps'19:46
jgriffithyeah19:46
*** markstur has left #openstack-cinder19:47
xyang11 sec19:47
jgriffithcan't find your patch19:47
hemnagotta grab some chow....bbiab.19:47
*** hemna is now known as hemnafk19:47
*** merooney has quit IRC19:47
*** diablo_rojo1 is now known as diablo_rojo19:47
*** xyang has joined #openstack-cinder19:49
xyangjgriffith: https://github.com/openstack/cinder/blob/master/cinder/volume/manager.py#L289419:49
xyangjgriffith: my patch https://review.openstack.org/#/c/229580/19:49
jgriffithOh.. DERP19:49
jgriffithI thought you were working on something new19:49
jgriffithsorry19:49
xyang1jgriffith: no, just bug fixing19:50
*** timcl has joined #openstack-cinder19:52
jgriffithxyang: yeah.. that ones kinda tricky huh?  You have to do some thing like take the group, get the cg ID off of that, then querie the backrefs of the volumes19:52
jgriffithxyang: like a get_volumes and filter on the cg ID19:52
*** david-lyle has joined #openstack-cinder19:52
xyangjgriffith: ya, I can get that from the original snapshots, I think19:52
xyangjgriffith: I have that before calling driver19:53
jgriffithxyang: or add a volume backref like we do with normal snapshots19:53
*** esker has joined #openstack-cinder19:53
xyangjgriffith: I'll take a look of that19:54
jgriffithxyang: so you don't like the way your patch is doing it?19:56
jgriffithxyang: I mean... that works, seems pretty straight forward19:57
xyangjgriffith: it works, it is just that if I change snapshots to model_update, it won't have volume_id any more19:57
xyangjgriffith: I can't expect driver to provide that in model_update19:58
jgriffithxyang: looking....19:58
*** mudassirlatif has joined #openstack-cinder19:58
xyangjgriffith: so I think I can use the original snapshots which are objects19:58
jgriffithxyang: why don't you just use the same trick :)19:59
xyangjgriffith: what same trick?19:59
jgriffithxyang: modify the signature and pass in the volume?19:59
jgriffithxyang: or did we just come full circle :)19:59
xyangjgriffith: the snapshots here were returned by the driver20:00
xyangjgriffith: it is a circle, I think:)20:00
jgriffith:)20:00
xyangjgriffith: so if I rename the returned parameter from driver, that will work20:00
xyangjgriffith: so I passed in snapshots, that has volume_id.  when driver returns back, it is snapshot_model_updates20:01
xyangjgriffith: snapshot_model_updates has uuid of snapshot and status only20:02
openstackgerritDiem Tran proposed openstack/cinder: Fix ZFSSA drivers' local cache bugs  https://review.openstack.org/22955420:02
*** IlyaG has joined #openstack-cinder20:04
*** akerr_ has quit IRC20:04
*** akerr has joined #openstack-cinder20:05
*** david-ly_ has joined #openstack-cinder20:07
*** timcl has quit IRC20:07
*** david-lyle has quit IRC20:07
*** david-ly_ is now known as david-lyle20:08
*** martyturner has quit IRC20:10
openstackgerritxing-yang proposed openstack/cinder: CG driver function should not access db  https://review.openstack.org/22958020:11
openstackgerritEric Harney proposed openstack/cinder: LVM: Make sparse_copy_volume private, use for capabilities  https://review.openstack.org/22698920:12
*** martyturner has joined #openstack-cinder20:12
*** baumann has quit IRC20:13
*** baumann has joined #openstack-cinder20:16
*** Lee1092 has quit IRC20:23
*** bswartz has quit IRC20:29
*** salv-orl_ has quit IRC20:31
*** martyturner has quit IRC20:31
*** IlyaG has quit IRC20:36
*** esker has quit IRC20:37
openstackgerritNate Potter proposed openstack/cinder: Add backup_swift_auth_url to swift backup driver  https://review.openstack.org/22864620:37
*** _cjones_ has joined #openstack-cinder20:37
*** garthb has quit IRC20:38
*** baumann has quit IRC20:38
*** stevemar has quit IRC20:39
*** stevemar has joined #openstack-cinder20:40
*** mc_nair has quit IRC20:40
*** adurbin__ has quit IRC20:41
*** hemnafk is now known as hemna20:42
*** stevemar has quit IRC20:45
*** ronis has quit IRC20:47
*** _cjones_ has quit IRC20:48
*** _cjones_ has joined #openstack-cinder20:49
*** IlyaG has joined #openstack-cinder20:49
*** ndipanov has quit IRC20:53
*** akerr has quit IRC20:53
*** stevemar has joined #openstack-cinder20:53
*** akerr has joined #openstack-cinder20:54
*** baumann has joined #openstack-cinder20:55
*** raildo is now known as raildo-afk20:55
*** Yogi1 has quit IRC20:57
*** ntpttr has quit IRC20:57
*** dustins has quit IRC21:01
*** david-lyle has quit IRC21:03
*** david-ly_ has joined #openstack-cinder21:03
openstackgerritKendall Nelson proposed openstack/cinder: Hacking check for opt name registration  https://review.openstack.org/22337521:03
*** diablo_rojo has quit IRC21:04
*** Guest82370 has quit IRC21:04
*** david-ly_ is now known as david-lyle21:05
*** IlyaG has quit IRC21:07
*** apoorvad_ has joined #openstack-cinder21:12
*** apoorvad_ has quit IRC21:12
*** apoorvad has quit IRC21:13
*** apoorvad_ has joined #openstack-cinder21:13
*** ociuhandu has joined #openstack-cinder21:14
*** porrua has quit IRC21:16
*** IlyaG has joined #openstack-cinder21:19
*** david-lyle has quit IRC21:26
*** david-lyle has joined #openstack-cinder21:29
*** albertom has quit IRC21:33
jgriffithcongrats thingee21:33
*** sgotliv has joined #openstack-cinder21:34
thingeejgriffith: thanks!21:34
jgriffithpretty cool21:35
*** baumann has quit IRC21:35
thingeeyeah feels good to finally let the news out :)21:35
jgriffith:)21:35
jgriffithSo what's your role going to be?21:36
*** jungleboyj has quit IRC21:36
thingeecross-project developer coordinator21:36
openstackgerritxing-yang proposed openstack/cinder: CG driver function should not access db  https://review.openstack.org/22958021:36
jgriffithnice21:36
thingeefocusing on cross project initiatives like openstack specs, def core, product working group, etc21:36
xyanghemna: ^21:37
xyanghemna: if you return "None, None" in delete_consistencygroup, manager will set status to 'error' for group and all volumes21:38
xyanghemna: sorry21:38
xyanghemna: let me try again21:38
xyanghemna: if you return "None, None" in delete_consistencygroup, manager will set status to 'deleted' for group and all volumes21:38
xyanghemna: if you throw exception, status will be set to "error" for group and volumes21:39
xyanghemna: how is that sound21:39
*** thangp has quit IRC21:41
*** garthb has joined #openstack-cinder21:41
*** gouthamr has quit IRC21:41
*** thingee has quit IRC21:46
*** IlyaG has quit IRC21:47
*** akerr has quit IRC21:47
*** david-lyle has quit IRC21:48
openstackgerritAngela Smith proposed openstack/cinder: Adds friendly zone name support  https://review.openstack.org/18051821:49
*** dzamboni has quit IRC21:52
*** david-lyle has joined #openstack-cinder21:54
*** kevincarr1991 has quit IRC21:54
*** jgregor has quit IRC21:59
openstackgerritxing-yang proposed openstack/cinder: CG driver function should not access db  https://review.openstack.org/22958022:00
*** xyang has quit IRC22:03
*** garthb has quit IRC22:03
*** eharney has quit IRC22:03
*** svasheka has quit IRC22:05
*** thingee has joined #openstack-cinder22:06
*** edtubill has quit IRC22:07
*** kevincarr1991 has joined #openstack-cinder22:07
tbarronthingee: "cross-project developer coordinator": you'll do great and I expect we'll be seeing a lot of you in that role :-)22:09
*** IlyaG has joined #openstack-cinder22:09
*** aarefiev has quit IRC22:09
*** tsekiyama has joined #openstack-cinder22:11
*** kevincar_ has joined #openstack-cinder22:15
*** david-lyle has quit IRC22:15
*** diogogmt has joined #openstack-cinder22:16
*** kevincarr1991 has quit IRC22:18
*** garthb has joined #openstack-cinder22:19
*** kevincar_ has quit IRC22:19
*** ociuhandu has quit IRC22:22
*** mriedem has quit IRC22:22
hemnaxyang1, that sounds good22:23
*** ociuhandu has joined #openstack-cinder22:24
*** marcusvrn_ has quit IRC22:25
xyang1hemna: great. I made similar change for create/delete cgsnapshot22:26
*** devlaps has quit IRC22:30
*** dims__ has joined #openstack-cinder22:30
*** earlephilhower has quit IRC22:32
*** e0ne has quit IRC22:33
*** dims_ has quit IRC22:33
*** IanGovett has quit IRC22:35
*** david-lyle has joined #openstack-cinder22:36
ociuhanduthingee, asselin: Hello, did you have by any chance a moment to go over the Microsoft iSCSI CI logs?22:36
asselinociuhandu, yes i think its fine....22:37
asselinwould've been nice if the layout & flow  matched better upstream, and if you used os-loganalyze, but those aren't requirements22:38
ociuhanduasselin: thank you. I will try to implement also these asap, so we are matching the layout and flow better22:38
ociuhanduasselin: what can we do as next steps to re-enble the gerrit account?22:39
asselinask thingee or smcginnis22:40
ociuhanduasselin: ok, thank you22:40
*** xyang1 has quit IRC22:45
*** thingee has quit IRC22:45
*** salv-orlando has joined #openstack-cinder22:46
openstackgerritMitsuhiro Tanino proposed openstack/cinder: Obtain target authentication from database same as LIO target  https://review.openstack.org/22808422:49
*** angela-s has quit IRC22:52
*** edtubill has joined #openstack-cinder22:56
*** mriedem has joined #openstack-cinder23:00
*** ntpttr has joined #openstack-cinder23:02
*** alejandrito has quit IRC23:04
*** bill_az has joined #openstack-cinder23:04
*** ociuhandu has quit IRC23:05
*** jaypipes has quit IRC23:07
*** erlon has quit IRC23:08
*** tsekiyama has quit IRC23:10
*** tsekiyama has joined #openstack-cinder23:10
*** EinstCra_ has quit IRC23:22
*** stevemar has quit IRC23:23
*** stevemar has joined #openstack-cinder23:24
*** brad[] has quit IRC23:25
*** IlyaG has quit IRC23:27
*** stevemar has quit IRC23:28
*** mudassirlatif_ has joined #openstack-cinder23:32
*** sgotliv has quit IRC23:33
*** ntpttr has quit IRC23:34
*** mudassirlatif has quit IRC23:34
*** mudassirlatif_ is now known as mudassirlatif23:34
*** mudassirlatif has quit IRC23:35
*** diogogmt has quit IRC23:36
*** jwcroppe has quit IRC23:40
*** jwcroppe has joined #openstack-cinder23:41
*** mudassirlatif has joined #openstack-cinder23:42
*** mudassirlatif has quit IRC23:43
*** jwcroppe has quit IRC23:45
*** diogogmt has joined #openstack-cinder23:48
*** rajinir has quit IRC23:48

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