Wednesday, 2015-05-13

*** dannywilson has quit IRC00:05
*** markvoelker has joined #openstack-cinder00:06
openstackgerritEdwin Wang proposed openstack/cinder: Add iSCSI protocol support for IBM FlashSystem
*** MentalRay has joined #openstack-cinder00:13
*** primechuck has quit IRC00:14
*** esker has joined #openstack-cinder00:16
*** annegentle has joined #openstack-cinder00:17
*** chlong_ has joined #openstack-cinder00:20
*** annegentle has quit IRC00:23
kmartinasselin, AMQP server on is unreachable: [Errno 104] Connection reset by peer. Trying again in 1 seconds.
*** tsekiyam_ has joined #openstack-cinder00:25
*** mtanino has quit IRC00:27
*** david-lyle has quit IRC00:28
*** tsekiyama has quit IRC00:28
*** tsekiyam_ has quit IRC00:29
*** rmesta has quit IRC00:32
*** annegentle has joined #openstack-cinder00:33
*** esker has quit IRC00:34
openstackgerritTom Barron proposed openstack/cinder: Add standard QoS spec support to cDOT drivers
*** IanGovett has quit IRC00:40
*** heyun has joined #openstack-cinder00:43
*** Yogi1 has joined #openstack-cinder00:43
*** Yogi1 has quit IRC00:46
*** Yogi1 has joined #openstack-cinder00:48
*** jdurgin has quit IRC00:50
*** rwsu has quit IRC00:54
*** _cjones_ has quit IRC00:55
*** esker has joined #openstack-cinder01:01
tbarronsmcginnis: winston-1: winston-d: lost your votes on b/c I had to rebase due to brick moving out of cinder01:03
tbarron^^^^ code unchanged except for some imports01:05
*** leeantho has quit IRC01:14
*** esker has quit IRC01:18
*** Lee1092 has joined #openstack-cinder01:25
openstackgerritJohn Griffith proposed openstack/cinder: Check volume_backend in retype
*** tobe43 has joined #openstack-cinder01:38
*** esker has joined #openstack-cinder01:44
*** vilobhmm1 has quit IRC01:49
*** vilobhmm1 has joined #openstack-cinder01:49
*** Yogi1 has quit IRC01:53
*** vilobhmm1 has quit IRC01:54
openstackgerritPatrick East proposed openstack/cinder-specs: Generic image volume cache functionality
*** Apoorva has quit IRC02:00
patrickeastjgriffith: ^ a quick stab at something for the image cache stuff, lemmie know if that gels with what you were thinking02:00
patrickeasti’m also only like 90% sure what i’m proposing would work :D02:01
*** patrickeast has quit IRC02:03
*** esker has quit IRC02:03
*** annegentle has quit IRC02:07
*** garthb has joined #openstack-cinder02:23
*** IlyaG has joined #openstack-cinder02:25
*** IlyaG has quit IRC02:28
*** IlyaG has joined #openstack-cinder02:28
*** esker has joined #openstack-cinder02:29
*** annegentle has joined #openstack-cinder02:40
*** annegentle has quit IRC02:41
*** esker has quit IRC02:47
*** tobe43 has quit IRC02:58
*** alexpilotti has quit IRC03:00
*** chlong has quit IRC03:05
*** annegentle has joined #openstack-cinder03:10
*** akerr_ has quit IRC03:13
*** esker has joined #openstack-cinder03:13
*** MentalRay has quit IRC03:13
*** dims_ has quit IRC03:18
*** rongze has joined #openstack-cinder03:19
*** annegentle has quit IRC03:20
*** esker has quit IRC03:24
*** esker has joined #openstack-cinder03:24
*** Apoorva has joined #openstack-cinder03:26
*** links has joined #openstack-cinder03:30
*** esker has quit IRC03:32
*** vilobhmm1 has joined #openstack-cinder03:33
jgriffithwinston-1: thingee DuncanT If we can get this merged:
jgriffithI'd like to turn on client testing in zuul03:41
*** patrickeast has joined #openstack-cinder03:42
*** david-lyle has joined #openstack-cinder03:48
*** anuragpalsule has joined #openstack-cinder03:48
*** chlong_ has quit IRC03:52
*** anuragpalsule has quit IRC03:53
*** coolsvap|afk is now known as coolsvap03:56
*** BharatK has joined #openstack-cinder03:56
openstackgerritJessy Lee proposed openstack/cinder: Add volume drivers for Infortrend Storage
openstackgerritJessy Lee proposed openstack/cinder: Add volume drivers for Infortrend Storage
*** anshprat has quit IRC04:06
*** dims has joined #openstack-cinder04:18
openstackgerritPatrick East proposed openstack/cinder-specs: Generic image volume cache functionality
*** rongze has quit IRC04:23
*** dims has quit IRC04:25
*** vilobhmm1 has quit IRC04:25
jgriffithpatrickeast: if you add something about the ability to use snapshots I'll give it my vote04:26
jgriffithpatrickeast: thanks for hitting the other issues for me04:26
patrickeastjgriffith: np, was just typing up some more comments04:26
*** vilobhmm1 has joined #openstack-cinder04:26
jgriffithpatrickeast: ahh... cool04:26
jgriffithpatrickeast: like "hell no I won't do snapshots" :)04:26
patrickeastjgriffith: the one thing i’m not sure on yet is whether or not we should or shouldn’t use the volume objects04:26
jgriffithpatrickeast: yeah... there's arguments both ways04:27
patrickeasttheres already so much re-usable goodness that knows about volume objects… and if we use a special tenant we don’t have to touch the db table at all04:27
jgriffithpatrickeast: understood04:27
patrickeastmaybe if its just limited to volumes and snapshots we could use both, and the cache code knows to detect which is which and make the right calls04:27
patrickeastie clone volume or create from snapshot04:28
jgriffithpatrickeast: I'm willing to keep an open mind on it for sure04:28
jgriffithpatrickeast: well... there's a way around that04:28
jgriffithpatrickeast: template-volume--->snap--->volume04:28
jgriffithpatrickeast: which the more I think about it is going to be required04:28
jgriffithpatrickeast: since we can't "attach" snaps04:28
patrickeastoh right04:29
patrickeastit has to be a volume initially either way04:29
jgriffithpatrickeast: yeah04:29
jgriffithpatrickeast: so maybe just a note in ther pointing out how people that are snapshot-biased can still benefit (including the LVM driver)04:29
jgriffithpatrickeast: I was actually thinking the snapshot would be the byproduct of the first create04:30
jgriffithfor those that need it04:30
jgriffithpatrickeast: anyway, I think we can adjust as needed04:30
patrickeastyea i figure we can always add more support in, i’ll just be careful to not box us into anything too much04:31
jgriffithI do like the new table with a backref to a volume though04:31
jgriffithpatrickeast: yeah, exact-a-mundo04:31
jgriffithand on that note... I'm going to go have a beeer :o04:31
jgriffithcatch ya in the morning04:31
*** anshprat has joined #openstack-cinder04:36
*** deepakcs has joined #openstack-cinder04:41
*** anshprat has quit IRC04:44
*** anshprat has joined #openstack-cinder04:44
*** tshefi has joined #openstack-cinder04:45
*** dannywilson has joined #openstack-cinder04:47
*** dannywilson has quit IRC04:49
*** dannywilson has joined #openstack-cinder04:49
*** heyun has quit IRC04:53
*** nkrinner has joined #openstack-cinder04:58
*** markvoelker has quit IRC04:58
*** patrickeast has quit IRC05:03
*** markvoelker has joined #openstack-cinder05:06
*** vilobhmm11 has joined #openstack-cinder05:11
deepakcseharney: ping05:12
openstackgerritBharat Kumar Kobagana proposed openstack/cinder: GlusterFS backup driver
openstackgerritBharat Kumar Kobagana proposed openstack/cinder: Add Posix backup driver. Supports simple nas too.
*** hemna_ has joined #openstack-cinder05:13
*** bill_az has joined #openstack-cinder05:13
*** xiaohui_ has joined #openstack-cinder05:14
*** vilobhmm1 has quit IRC05:15
*** sgotliv has joined #openstack-cinder05:15
*** dosaboy_ has joined #openstack-cinder05:18
*** guitarza1 has joined #openstack-cinder05:19
*** skraynev_ has joined #openstack-cinder05:22
*** garthb has quit IRC05:23
*** coolsvap- has joined #openstack-cinder05:23
*** chlong has joined #openstack-cinder05:23
*** bkopilov is now known as bkopilov_wfh05:24
*** rongze has joined #openstack-cinder05:26
*** xiaohui has quit IRC05:26
*** coolsvap has quit IRC05:26
*** hemna has quit IRC05:26
*** anteaya has quit IRC05:26
*** ir2ivps7 has quit IRC05:26
*** guitarzan has quit IRC05:26
*** dosaboy has quit IRC05:27
*** skraynev has quit IRC05:27
*** heyun has joined #openstack-cinder05:27
*** anteaya has joined #openstack-cinder05:28
*** dalgaaf has joined #openstack-cinder05:30
*** ir2ivps7 has joined #openstack-cinder05:31
*** chlong has quit IRC05:34
*** sks has joined #openstack-cinder05:37
openstackgerritBharat Kumar Kobagana proposed openstack/cinder: RemoteFS: Reporting configured reserved_percentage in _update_volume_stats
*** nlevinki has joined #openstack-cinder05:43
*** chlong has joined #openstack-cinder05:48
*** dulek has joined #openstack-cinder05:48
*** rongze has quit IRC05:55
*** lpetrut has joined #openstack-cinder05:56
*** rongze has joined #openstack-cinder05:58
*** dims has joined #openstack-cinder06:01
*** BharatK has quit IRC06:01
*** rongze has quit IRC06:02
*** dims has quit IRC06:06
*** rongze has joined #openstack-cinder06:06
*** afazekas_ has joined #openstack-cinder06:11
*** Apoorva has quit IRC06:11
*** anuragpalsule has joined #openstack-cinder06:13
*** BharatK has joined #openstack-cinder06:16
*** IlyaG has quit IRC06:19
*** IlyaG has joined #openstack-cinder06:20
*** chlong has quit IRC06:24
*** IlyaG has quit IRC06:24
*** chlong has joined #openstack-cinder06:27
*** dannywilson has quit IRC06:33
*** bill_az has quit IRC06:34
*** BharatK has quit IRC06:44
*** smallbig has left #openstack-cinder06:48
*** lpetrut has quit IRC06:50
*** dulek has quit IRC06:51
*** dulek has joined #openstack-cinder06:51
*** yrabl has quit IRC06:54
*** skraynev_ is now known as skraynev06:54
*** yrabl has joined #openstack-cinder06:55
*** BharatK has joined #openstack-cinder06:55
*** fifieldt has joined #openstack-cinder06:56
*** fifieldt has quit IRC06:56
openstackgerritRick Chen proposed openstack/cinder: Port ProphetStor driver to use new driver model
*** tshefi has quit IRC07:05
*** markus_z has joined #openstack-cinder07:06
*** jordanP has joined #openstack-cinder07:08
*** dulek has quit IRC07:09
*** x3k is now known as xek07:19
*** tshefi has joined #openstack-cinder07:20
openstackgerritBharat Kumar Kobagana proposed openstack/cinder: Using 'fallocate' instead of 'dd' to create raw volumes
*** vilobhmm11 has quit IRC07:27
*** pschaef has joined #openstack-cinder07:29
*** xek has quit IRC07:40
*** xek has joined #openstack-cinder07:41
*** ronis has joined #openstack-cinder07:41
*** agarciam has joined #openstack-cinder07:48
*** jistr has joined #openstack-cinder07:49
*** dulek has joined #openstack-cinder07:50
*** dulek has quit IRC07:55
*** heyun has quit IRC07:57
*** Longgeek has joined #openstack-cinder08:00
*** rongze has quit IRC08:06
*** e0ne has joined #openstack-cinder08:25
openstackgerritJeegn Chen proposed openstack/cinder: Address 0x712d8e0e error in VNX Cinder Driver
*** e0ne has quit IRC08:28
*** tshefi has quit IRC08:28
*** e0ne has joined #openstack-cinder08:30
*** e0ne has quit IRC08:31
*** tshefi has joined #openstack-cinder08:41
openstackgerritYuriy Nesenenko proposed openstack/cinder: Checking the existence of volume
*** yuriy_n17 has joined #openstack-cinder09:02
*** dulek has joined #openstack-cinder09:09
openstackgerritGorka Eguileor proposed openstack/cinder: Create iSCSI lio portals with right IPs and port
*** alecv has joined #openstack-cinder09:14
*** ankit_ag has joined #openstack-cinder09:19
*** rongze has joined #openstack-cinder09:20
*** e0ne has joined #openstack-cinder09:22
*** dulek has quit IRC09:28
*** dulek has joined #openstack-cinder09:29
*** dims has joined #openstack-cinder09:35
*** e0ne is now known as e0ne_09:36
*** e0ne_ is now known as e0ne09:38
*** 5EXAA60VI has quit IRC09:40
*** dims has quit IRC09:40
ekarlsohow can I delete a volume that says it's attached to a instance but the instance is gone ?09:45
*** dims has joined #openstack-cinder10:01
*** lpetrut has joined #openstack-cinder10:02
*** IanGovett has joined #openstack-cinder10:06
*** jamielennox is now known as jamielennox|away10:08
e0neekarlso: you can try cinder reset-state <vol_id> first and then delete a volume10:13
*** dulek has quit IRC10:19
*** e0ne is now known as e0ne_10:25
*** wanghao has quit IRC10:25
*** wanghao has joined #openstack-cinder10:25
*** jbernard has quit IRC10:26
*** e0ne_ is now known as e0ne10:26
openstackgerritTom Barron proposed openstack/cinder: Use nfs_oversub_ratio when reporting pool capacity
*** angela-s has quit IRC10:28
*** dulek has joined #openstack-cinder10:28
*** aarefiev has quit IRC10:29
*** aarefiev has joined #openstack-cinder10:30
*** EmilienM has quit IRC10:33
*** danieru has quit IRC10:33
*** ekarlso has quit IRC10:33
*** danieru has joined #openstack-cinder10:34
*** aarefiev has quit IRC10:35
*** EmilienM has joined #openstack-cinder10:35
*** aarefiev has joined #openstack-cinder10:35
*** jbernard has joined #openstack-cinder10:36
*** e0ne is now known as e0ne_10:39
*** e0ne_ is now known as e0ne10:40
*** ekarlso has joined #openstack-cinder10:41
*** cbader has quit IRC10:41
*** deepakcs has quit IRC10:41
*** mdenny has quit IRC10:41
*** ekarlso has quit IRC10:45
*** ekarlso has joined #openstack-cinder10:45
*** dulek has quit IRC10:50
*** dulek has joined #openstack-cinder10:50
*** dulek_ has joined #openstack-cinder10:53
*** dulek_ has quit IRC10:53
*** dulek_ has joined #openstack-cinder10:54
openstackgerritAlfredo Moralejo proposed openstack/cinder: rbd driver in cinder does not manage glance images multi-location
*** dulek has quit IRC10:54
*** kmartin has quit IRC10:56
*** Trozz has quit IRC10:58
openstackgerritAnton Arefiev proposed openstack/cinder: Add entry create and cast tasks to manage workflow
openstackgerritAnton Arefiev proposed openstack/cinder: Fix lvm manage existing volume
*** asselin has quit IRC11:01
DuncanT   <--- VM breakout vunerability. Ooops11:04
openstackgerritAnton Arefiev proposed openstack/cinder: Add entry create and cast tasks to manage workflow
openstackgerritAnton Arefiev proposed openstack/cinder: Fix lvm manage existing volume
*** Trozz has joined #openstack-cinder11:05
*** dalgaaf has quit IRC11:07
*** pschaef has quit IRC11:11
*** ociuhandu has joined #openstack-cinder11:13
*** julim has quit IRC11:18
*** e0ne is now known as e0ne_11:26
openstackgerritSwapnil Kulkarni proposed openstack/python-cinderclient: Fixed typos and repeated docstrings
*** tobe4333 has joined #openstack-cinder11:35
openstackgerritPranali Deore proposed openstack/cinder: Cinder os-force_detach api returns 500
*** coolsvap- is now known as coolsvap|afk11:41
*** IanGovett has quit IRC11:43
*** IanGovett has joined #openstack-cinder11:49
*** delattec has quit IRC11:50
*** cdelatte has quit IRC11:50
*** BharatK has quit IRC11:53
*** fthiagogv has joined #openstack-cinder11:54
*** tobe4333 has quit IRC11:57
*** IanGovett1 has joined #openstack-cinder11:58
*** lpabon_ has joined #openstack-cinder12:01
*** lpabon_ has quit IRC12:01
*** IanGovett has quit IRC12:01
*** BharatK has joined #openstack-cinder12:05
*** cdelatte has joined #openstack-cinder12:07
*** e0ne_ is now known as e0ne12:08
*** delattec has joined #openstack-cinder12:09
*** Tross1 has quit IRC12:09
*** cdelatte has quit IRC12:12
*** aix has joined #openstack-cinder12:13
*** bswartz has quit IRC12:14
*** lpabon has quit IRC12:14
*** lpabon has joined #openstack-cinder12:19
*** cbader has joined #openstack-cinder12:26
*** timcl has joined #openstack-cinder12:28
*** akerr has joined #openstack-cinder12:29
*** dulek_ has quit IRC12:31
*** alexpilotti has joined #openstack-cinder12:32
*** annegentle has joined #openstack-cinder12:34
*** Yogi1 has joined #openstack-cinder12:43
*** BharatK has quit IRC12:44
*** annegentle has quit IRC12:45
*** annegentle has joined #openstack-cinder12:46
openstackgerrityogeshprasad proposed openstack/cinder: Add chap support to CloudByte cinder driver
*** julim has joined #openstack-cinder12:53
*** fthiagogv has quit IRC12:58
*** sks has quit IRC12:59
*** bill_az has joined #openstack-cinder13:00
*** merooney has joined #openstack-cinder13:03
*** bkopilov_wfh has quit IRC13:04
*** timcl1 has joined #openstack-cinder13:05
*** Yogi1 has quit IRC13:05
*** nkrinner has quit IRC13:06
*** annegentle has quit IRC13:06
*** bkopilov_wfh has joined #openstack-cinder13:07
*** annegentle has joined #openstack-cinder13:07
*** timcl has quit IRC13:08
*** primechuck has joined #openstack-cinder13:08
*** mriedem has left #openstack-cinder13:10
*** mriedem has joined #openstack-cinder13:11
*** sks has joined #openstack-cinder13:13
openstackgerritTina Tang proposed openstack/cinder: Create consistgroup from cgsnapshot support in VNX driver
*** BharatK has joined #openstack-cinder13:23
*** guitarza1 is now known as guitarzan13:26
*** xyang has joined #openstack-cinder13:26
openstackgerritEdwin Wang proposed openstack/cinder: Add iSCSI protocol support for IBM FlashSystem
*** fthiagogv has joined #openstack-cinder13:31
*** Yogi1 has joined #openstack-cinder13:32
*** merooney has quit IRC13:39
*** ankit_ag has quit IRC13:39
*** rongze has quit IRC13:43
*** primechuck has quit IRC13:46
openstackgerritSean McGinnis proposed openstack/cinder: Fix incorrect reraising of exceptions
*** merooney has joined #openstack-cinder13:48
*** links has quit IRC13:48
*** IlyaG has joined #openstack-cinder13:51
*** merooney has quit IRC13:52
*** sks has quit IRC13:54
*** rongze has joined #openstack-cinder13:56
DuncanTHave we documented what the units of cinder create (etc) are (1000 or 1024 mb)?13:57
DuncanTPretty sure there's a bug in one of the HP drivers but I can't see a doc for it. I guess the default answer is to do what the reference driver does... I might even look at a tempest test for this13:58
smcginnisDuncanT: Good question. Our storage uses 1024, so if we are requested to create a 1G volume it is 1024 mb.13:58
*** markvoelker has quit IRC13:59
*** anuragpalsule has quit IRC13:59
*** Yogi11 has joined #openstack-cinder14:00
e0neDuncanT: afaik, we it is not documened. tha same question was here about month ago14:01
DuncanTThanks, I'll look at adding it to the docs14:01
*** primechuck has joined #openstack-cinder14:02
*** merooney has joined #openstack-cinder14:03
*** Yogi1 has quit IRC14:03
*** rushil has joined #openstack-cinder14:04
*** sks has joined #openstack-cinder14:07
*** mtanino has joined #openstack-cinder14:07
*** jungleboyj has joined #openstack-cinder14:11
jgriffithDuncanT: yes, it's documented in the code as 1024MiB14:12
jgriffitherr... you know... the 1024 one14:12
jgriffithlemme see if I can find it14:12
akerroslo units are MiB (1024)14:13
akerror rather GiB14:13
*** eharney has quit IRC14:14
jgriffithakerr: yeah, that's what I was getting at14:15
*** haomaiwang has joined #openstack-cinder14:15
jgriffithakerr: thanks, I can't talk yet this morning ;)14:15
*** jaypipes has joined #openstack-cinder14:15
*** bswartz has joined #openstack-cinder14:17
jgriffiththingee: DuncanT Not sure if you guys wanted to setup something on Monday night or not:
jgriffithThere's a link to do so, I was going to set something but I won't be able to make it, so that didn't make sense14:20
thingeeDuncanT: can you let us know the mid cycle meetup results in the meeting today?14:22
thingeejgriffith: what is this?14:22
DuncanTthingee: Sure14:23
*** breitz has quit IRC14:23
jgriffiththingee: response to this
*** breitz has joined #openstack-cinder14:24
jgriffithThat's the answer to "There must be an app" for that14:24
thingeejgriffith: ah ok!14:24
*** haomaiwang has quit IRC14:25
smcginnisI haven't heard too much about Summit parties so far. Is there a list somewhere?14:25
thingeejgriffith: yeah let me catch up on things and I'll see about doing something.14:25
DuncanTjgriffith: good call!14:25
*** akerr has quit IRC14:25
jgriffiththingee: might be worth gauging availability in todays meeting14:25
thingeeGood idea14:25
*** bkopilov has joined #openstack-cinder14:26
jgriffiththingee: DuncanT FYI
*** akerr has joined #openstack-cinder14:26
*** dulek has joined #openstack-cinder14:26
jgriffiththose get us LVM-Thin gate testing, as well as turn on the cinderclient functional tests14:27
jgriffithI thnk grenade is fixed up now, so the devstack one should finally get through check14:27
*** bkopilov_wfh has quit IRC14:27
*** afazekas has joined #openstack-cinder14:29
openstackgerritAnton Arefiev proposed openstack/cinder: Add entry create and cast tasks to manage workflow
openstackgerritAnton Arefiev proposed openstack/cinder: Fix lvm manage existing volume
*** chlong has quit IRC14:30
*** sks has quit IRC14:31
openstackgerritSean McGinnis proposed openstack/cinder: Remove use of deprecated LOG.warn
smcginnis^^ Last one I swear. After this I'm going to run as far away from logging as I can.14:33
e0nesmcginnis: :)14:34
*** nlevinki has quit IRC14:35
*** deepakcs has joined #openstack-cinder14:37
*** fanyaohong has joined #openstack-cinder14:37
*** tsekiyama has joined #openstack-cinder14:39
*** MentalRay has joined #openstack-cinder14:40
*** fthiagogv has quit IRC14:48
jgriffithDuncanT: would you be so kind as to just make this f'ing go away:
jgriffithDuncanT: ie by way of +2/A so I never see the stupid waste of time again14:49
DuncanTI've got your other 3 links open, I will look before the meeting but I'm in a sprint planning session not14:50
*** e0ne is now known as e0ne_14:50
*** e0ne_ is now known as e0ne14:52
jgriffithDuncanT: don't worry about it, was just FYI14:52
DuncanTOk, thanks14:52
DuncanTOne was waiting for jenkins anyway14:52
*** patrickeast has joined #openstack-cinder14:52
jgriffithThey're all waiting for *something* :)14:53
*** hemna_ is now known as hemna14:53
DuncanTAren't we all....14:53
DuncanTMorning, Walt14:53
*** Yogi11 has quit IRC14:54
*** Yogi1 has joined #openstack-cinder14:56
*** bkopilov has quit IRC14:57
*** kmartin has joined #openstack-cinder14:59
*** reed_ has joined #openstack-cinder15:00
*** fthiagogv has joined #openstack-cinder15:01
scottdathingee: I put Discuss Liberty mid-cycle on the meeting agenda. Did I remember correctly that you had wanted to finalize things this week?15:01
thingeescottda: yes I was wondering who did that. Thank you!15:01
*** asselin has joined #openstack-cinder15:02
thingeejgriffith: great work on switch lvm to thin. I was noticing the devstack work yesterday.15:02
*** marcusvrn_ has joined #openstack-cinder15:02
thingeejgriffith: Definitely interested in getting that in. I will post comments for support15:02
jgriffiththingee: yeah, we'll see how it fares in the gate15:02
*** e0ne is now known as e0ne_15:02
jgriffiththingee: also need to think about any possible migration plan for upgrades15:03
*** ebalduf has joined #openstack-cinder15:03
*** BharatK has quit IRC15:03
thingeejgriffith: you're our migration expert :D. nova volume -> cinder ;)15:03
jgriffiththingee: haha... that was easy compared to this.  Nobody would "lose data" :)15:04
jgriffiththingee: I've got a crazy script that uses a ramdisk and fstrim, but I don't think I'd trust it for "real" people to use15:05
*** akerr has quit IRC15:07
*** akerr has joined #openstack-cinder15:07
*** akerr is now known as akerr_away15:08
DuncanTjgriffith: Assuming you have enough disk space, create a new vg, start a second c-vol, use cinder-migrate and shrink / grow the VGs as space becomes an issue?15:08
jgriffithDuncanT: well, the problem is that means dd15:08
*** akerr_away has quit IRC15:08
jgriffithDuncanT: which means no point in using thin because you end up allocating all the blocks15:08
jgriffithDuncanT: so you end up right back where you started for the existing volumes15:09
DuncanTAh. Write a smarter dd that spots zero runs?15:09
jgriffithDuncanT: I'll get right on that :)15:09
jgriffithDuncanT: LOL15:09
jgriffithDuncanT: there's "stuff" out there to do that sort of thing15:09
jgriffithDuncanT: just not sure how well they work etc15:09
jgriffithDuncanT: honestly it might be perfectly fine to say "it's up to you ops folks or customers"15:10
jgriffithDuncanT: just run multi-backend for now, and new volumes are just "better" than your oldones15:10
DuncanTjgriffith: That's probably a good initial story, sure15:10
*** mdenny has joined #openstack-cinder15:11
DuncanTjgriffith: A smarter dd would hopefully just plug into migrate transparently, so it can be aadded by whoever is desparate for it15:11
*** bkopilov has joined #openstack-cinder15:11
*** akerr has joined #openstack-cinder15:11
jgriffithDuncanT: yeah for sure15:12
DuncanTjgriffith: I've been wondering about writing a dumb (but fast) gzip encoder to speed up backups... just compress zero runs and literal everything else15:13
*** bswartz has quit IRC15:14
jgriffithThere are compression algo's that do that15:14
jgriffithDuncanT: would be cool to look into it15:14
openstackgerritPatrick East proposed openstack/cinder: Refactor PureISCSIDriver into base and iSCSI classes.
DuncanTjgriffith: I fancy writing some C for a change, and the gzip FSM doesn't look so very bad15:15
*** dims has quit IRC15:15
*** dims has joined #openstack-cinder15:16
*** rmesta has joined #openstack-cinder15:18
*** jistr is now known as jistr|mtg15:19
*** bswartz has joined #openstack-cinder15:19
aarefievjgriffith: hi, got a min?15:23
*** afazekas_ has quit IRC15:24
*** afazekas has quit IRC15:24
jgriffithaarefiev: sure, what's up?15:26
aarefievjgriffith: have a question about your comment on
aarefiev I don't quite understand what you mean by 'The lvrename exception was "intended" to catch this case ...'15:27
jgriffithaarefiev: just commentary, don't worry about it15:27
jgriffithaarefiev: the point was when I wrote that I had your bug in mind15:27
jgriffithaarefiev: the idea was that "lvrename foo foo"15:28
jgriffithraises an exception15:28
jgriffithbut what I WASNT thinking was how we create a new db ref/ID15:28
jgriffithso the name would be different15:28
jgriffithsee what I'm saying?15:28
*** jistr|mtg is now known as jistr15:29
aarefievjgriffith: I don't understand your point about return bool15:29
aarefievin which cases it will be useful15:29
*** markstur has quit IRC15:29
jgriffithaarefiev: so my point there was rather than just let the exception escape all the way out to taskflow or manager.py15:30
jgriffithaarefiev: make that utils method an explicit True/False check15:30
*** merooney has quit IRC15:30
jgriffithaarefiev: so actually return True of False15:30
*** markstur has joined #openstack-cinder15:32
aarefievjgriffith: hmm, and if it is already managed it will return True and ...?15:32
*** merooney has joined #openstack-cinder15:33
*** daneyon_ has joined #openstack-cinder15:33
jgriffithaarefiev: so either way.... so if you change it to something like:15:34
jgriffithYou return True if it is, then raise the exception in the calling method15:34
jgriffithIf it's False, then you carry on and finish the manage routine15:34
jgriffithaarefiev: if you hate the idea that's fine, I'll change my vote15:35
jgriffithaarefiev: I'm likely the only one that cares about this :)15:35
*** thangp has joined #openstack-cinder15:35
aarefievjgriffith: you mean not utils method raise exception, but caller should do it15:36
jgriffithaarefiev: correct15:36
*** mriedem has left #openstack-cinder15:36
jgriffithaarefiev: Just thinkingit may be more "re-usable"15:37
*** daneyon has quit IRC15:37
openstackgerritThang Pham proposed openstack/cinder: Switch to oslo_versionedobjects
*** mriedem has joined #openstack-cinder15:37
openstackgerritSean McGinnis proposed openstack/cinder: Fix incorrect reraising of exceptions
aarefievjgriffith: ok got it , I don't really see use cases, I will update my patch , thanks for review15:38
jgriffithaarefiev: thanks15:39
*** _cjones_ has joined #openstack-cinder15:39
openstackgerritRick Chen proposed openstack/cinder: Port ProphetStor driver to use new driver model
*** e0ne_ is now known as e0ne15:43
*** tshefi has quit IRC15:43
*** bkopilov has quit IRC15:44
openstackgerritxing-yang proposed openstack/cinder: EMC VMAX Manage/Unmanage Volume
*** xyang has quit IRC15:44
*** _cjones_ has quit IRC15:44
openstackgerritxing-yang proposed openstack/cinder: Fix a problem with FAST support in VMAX driver
*** eharney has joined #openstack-cinder15:45
*** _cjones_ has joined #openstack-cinder15:45
*** bkopilov has joined #openstack-cinder15:45
e0nejgriffith, aarefiev: fyi, - "Checking the existence of volume" spec15:46
*** ebalduf has quit IRC15:48
e0nesmcginnis: i've posed one very minor comment to your patch if you'll make one more change-set15:49
smcginnise0ne: Thanks, I'll take a look.15:49
*** garthb has joined #openstack-cinder15:50
*** rushil has quit IRC15:51
openstackgerritSean McGinnis proposed openstack/cinder: Fix incorrect reraising of exceptions
*** shakamunyi has quit IRC15:51
*** barra204 has quit IRC15:51
*** ronis has quit IRC15:51
*** rushil has joined #openstack-cinder15:51
openstackgerritSean McGinnis proposed openstack/cinder: Remove use of deprecated LOG.warn
smcginnise0ne: Good point, I had meant to do that. Updated.15:52
*** esker has joined #openstack-cinder15:53
openstackgerritTom Swanson proposed openstack/cinder: Dell: Added verify cert option for REST calls
SwansonUm, that should be interesting.15:55
openstackgerritxing-yang proposed openstack/cinder: Fix a problem with FAST support in VMAX driver
smcginnisSwanson: Why's that?15:56
e0nesmcginnis: thanks!15:56
smcginnise0ne: NP, thanks for pointing it out.15:56
*** ebalduf has joined #openstack-cinder15:56
openstackgerritPatrick East proposed openstack/cinder-specs: Generic image volume cache functionality
openstackgerritWalter A. Boring IV (hemna) proposed openstack/os-brick: Updated from global requirements
jgriffithe0ne: oh, yeah... there ya go15:58
*** adurbin_ has joined #openstack-cinder15:58
jgriffithaarefiev: so build your check off of that15:58
openstackgerritWalter A. Boring IV (hemna) proposed openstack/os-brick: ISCSI be careful parsing iscsiadm output
openstackgerritxing-yang proposed openstack/cinder: EMC VMAX Manage/Unmanage Volume
*** markus_z has quit IRC16:00
*** vilobhmm1 has joined #openstack-cinder16:00
jgriffithe0ne: thanks... I actually think I saw that the other day but forgot :(16:01
*** ebalduf has quit IRC16:01
openstackgerritxing-yang proposed openstack/cinder: Fix issues with extra specs in VMAX driver
e0nejgriffith: you're welcome! actually, i sent you a link to code, not to spec:(16:01
*** Zhongjun has joined #openstack-cinder16:02
jgriffithe0ne: that's better for me actually16:02
e0nejgriffith: :)16:03
aarefievjgriffith: ok, thanks, I got your point, I just don't see any benefit of it, so I've asked you what I missed16:04
jgriffithaarefiev: sure, but the point should be obvious now.16:05
*** xyang has joined #openstack-cinder16:05
jgriffithaarefiev: there's another patch in flight that does the exact same sort of thing for a different area of the code16:05
jgriffithaarefiev: that was mypoint16:05
*** dulek has quit IRC16:08
*** Apoorva has joined #openstack-cinder16:08
*** erlon has joined #openstack-cinder16:12
*** rwsu has joined #openstack-cinder16:13
*** merooney has quit IRC16:14
*** nlevinki has joined #openstack-cinder16:14
*** timcl1 has quit IRC16:15
*** ebalduf has joined #openstack-cinder16:15
*** IlyaG has quit IRC16:15
*** IlyaG has joined #openstack-cinder16:16
*** ebalduf has quit IRC16:16
*** ebalduf has joined #openstack-cinder16:16
aarefievjgriffith: it doesn't do the same, in my patch I just check db entry for volume existing16:17
openstackgerritPatrick East proposed openstack/cinder-specs: Generic image volume cache functionality
*** setmason has joined #openstack-cinder16:19
*** rushil has quit IRC16:20
jgriffithaarefiev: your implementation is different/simpler, but the goal and end result is the same16:21
jgriffithaarefiev: not sure what you want me to say here16:21
*** lpetrut has quit IRC16:21
jgriffithaarefiev: feel free to get input from others, maybe I'm completely off base16:22
*** vilobhmm1 has quit IRC16:25
aarefievjgriffith: in patch you pointed checks that volume exists on the storage backend, in my patch I just check db entry16:25
*** jistr has quit IRC16:25
aarefievjgriffith: it is diff level16:25
jgriffithaarefiev: yes, implementation is different, agreed16:26
jgriffithaarefiev: frankly some things about yours I like better16:26
jgriffithaarefiev: but my point is stil the same as it was16:26
*** primechuck has quit IRC16:26
jgriffithaarefiev: what would you like me to do?16:26
*** leeantho has joined #openstack-cinder16:26
*** rongze has quit IRC16:27
openstackgerritWalter A. Boring IV (hemna) proposed openstack/cinder: 3PAR enable multiattach capability reporting
aarefievjjgriffith: I just say it is diff functionality, it doesn't close the bug16:27
jgriffithaarefiev: and I'm saying respin your patch with the referenced work as a dependency and just use the method that he's implemented for the check and get on with life16:29
jgriffithaarefiev: that will close the bug, you'll have resolved, I'll GREATLY appreciate your time and effort16:29
jgriffithand we can all just be happy and sing Kumba-ya16:29
*** jordanP has quit IRC16:29
jgriffithaarefiev: or, you can just think I'm a JackA$$ and ignore me16:30
jgriffithaarefiev: maybe some other reviewers or core members will agree with you16:30
aarefievjgriffith: ok, I think we don't understand each other:), I will update my patch16:31
openstackgerritOpenStack Proposal Bot proposed openstack/cinder: Updated from global requirements
*** dannywilson has joined #openstack-cinder16:31
jgriffithaarefiev: ok, thanks, and sorry if I'm being dense16:33
*** primechuck has joined #openstack-cinder16:35
*** crose has joined #openstack-cinder16:35
aarefievjgriffith: np, I'll update patch and add more explanation,  maybe I'll be able convince you:)16:38
*** MentalRay has quit IRC16:38
*** merooney has joined #openstack-cinder16:40
jgriffithaarefiev: Very possible that you will :)16:40
*** MentalRay has joined #openstack-cinder16:41
aarefievjgriffith: :)16:41
*** timcl has joined #openstack-cinder16:43
anteayapatrickeast: are you about? there was a question the other day about what zuul version you are using16:48
*** alecv has quit IRC16:49
anteayapatrickeast: apparently you are using a zuul 2.0.1, which is confusing for me since when I look at the git refs for zuul it shows the latest tag is 2.0.016:49
*** fanyaohong has quit IRC16:49
anteayapatrickeast: where are you getting your zuul 2.0.1 from?16:49
patrickeastanteaya: uhh, good question, lemmie go check on my master real quick16:50
anteayathank you16:50
patrickeastanteaya: one sec16:50
* anteaya stands by16:50
*** patrickeast_ has joined #openstack-cinder16:51
patrickeast_forgot my irc client doesnt like switching over to the vpn...16:52
anteayapicky client16:52
patrickeast_anteaya: i get
*** lpetrut has joined #openstack-cinder16:52
anteayaso you do16:52
anteayaany idea how you get zuul versions?16:52
patrickeast_its installed from asselin’s scripts, my guess is its tip of tree?16:52
*** patrickeast has quit IRC16:52
*** patrickeast_ is now known as patrickeast16:52
anteayaasselin: any thoughts here?16:53
anteayaI'll swing over to infra eventually just thought I would see what I could learn here first16:53
asselinanteaya, it uses tip of master.16:55
*** daneyon_ has quit IRC16:56
asselinanteaya, signing off...will be back later...16:56
anteayaasselin: okay thanks16:56
*** daneyon has joined #openstack-cinder16:56
anteayaso pbr cooks up a version number based on semver16:57
anteayathat is where that version number is coming from16:57
jgriffithhemna: the tric is typcially we avoid capping16:57
*** IlyaG_ has joined #openstack-cinder16:57
jgriffithhemna: at the same time requiring 3'rd party CI run against brick seems like a waste16:58
patrickeastjgriffith: hemna: i’m actually excited to be able to test on brick, primarliy for once nova uses brick16:58
patrickeasti don’t want to run on all nova patches16:58
hemnapatrickeast, same here :)16:58
patrickeastbut i do want to make sure it can attach and what not16:59
patrickeastso its a perfect way to test that16:59
hemnapatrickeast, we need to get some of the os-brick 'bugs' fixed.16:59
*** vilobhmm1 has joined #openstack-cinder16:59
*** rushil has joined #openstack-cinder16:59
hemnapatrickeast, which are basically feature requests that bring the os-brick code up to par with some of the features that were added in nova's libvirt but didn't make it into cinder/brick before the os-brick cutoff16:59
e0nehemna: maybe os-brick<=1.0 be better16:59
*** merooney has quit IRC16:59
hemnae0ne, but we want to force CI testing against the next drop, which will be 0.2.0 no?17:00
*** IlyaG has quit IRC17:00
e0neaccording to semver we should not breack api compatibilitiies in minor versions17:00
hemnaif we do <=1.0 then it'll just update everyone?17:00
patrickeasthemna: heh yea, there are some bugs we are seeing with the nova iscsi/multipath code that don’t occur in brick… i’m hoping they go away when brick comes into the picture for nova17:01
*** asselin has quit IRC17:01
patrickeasthopefully they aren’t part of those features17:01
e0nehemna: ok. <0.2 will be ok for me17:01
hemnaI'm open to whatever17:01
*** deepakcs has quit IRC17:03
hemnaso the current patch
hemnameans that no one will get 0.2.0 when we release it.17:03
hemnauntil we update global-requirements.txt again17:03
jgriffithhemna: e0ne so typically capping is frowned upon unless there's a known issue in a version17:03
hemnaand then cinder gets the bot patch, which we can test?17:03
*** patrickeast has quit IRC17:03
hemnaso I'm just trying to figure out how we want to force CI testing against a bump of os-brick17:03
jgriffithhemna: e0ne I don't think it' a big deal to run like we do with every other external lib17:03
*** winston-d is now known as winston-d_zzZ17:04
*** sks has joined #openstack-cinder17:05
*** sks has quit IRC17:06
openstackgerritTom Swanson proposed openstack/cinder: Dell SC: Added support for alternate iscsi portals
*** rushil has quit IRC17:10
thangpjgriffith, thingee: could we try to merge this again?  simple rebase17:11
*** ronis has joined #openstack-cinder17:12
*** rushil has joined #openstack-cinder17:13
openstackgerritYuriy Nesenenko proposed openstack/cinder: Checking the existence of volume
e0nejgriffith, hemna: i prefer to not have strict versions in requirements.17:17
e0nejgriffith, hemna: any solution has pros and cons17:17
*** vilobhmm1 has quit IRC17:18
*** vilobhmm1 has joined #openstack-cinder17:19
*** nlevinki has quit IRC17:19
*** harlowja has quit IRC17:20
*** harlowja has joined #openstack-cinder17:20
jgriffithe0ne: yeah, I agree17:20
jgriffithe0ne: I say just leave it uncapped and go17:20
*** MRay has joined #openstack-cinder17:23
*** MentalRay has quit IRC17:23
e0neanyway, if we break something after new release of os-brick, we'll be notified via CIs and openstack-dev ml17:25
*** dannywilson has quit IRC17:26
*** bswartz has quit IRC17:28
*** mdenny has quit IRC17:29
*** leeantho has quit IRC17:29
*** Yogi1 has quit IRC17:30
*** leeantho has joined #openstack-cinder17:33
*** 17SACKCIT has joined #openstack-cinder17:34
*** 7YUAAALBR has joined #openstack-cinder17:34
*** thangp has quit IRC17:34
*** 7YUAAALBR has quit IRC17:34
*** 17SACKCIT has quit IRC17:34
*** rushil has quit IRC17:34
*** rushil has joined #openstack-cinder17:35
*** asselin has joined #openstack-cinder17:35
*** hemna_ has joined #openstack-cinder17:36
DuncanTjgriffith: Just reading your replication spec... couple of questions, starting with the easy one: Why not a replication table joined by primary key to volume? Means that 'normal' volumes aren't affected17:36
openstackgerritTom Swanson proposed openstack/cinder: Dell SC: Added support for alternate iscsi portals
jgriffithDuncanT: certainly could do that17:36
DuncanTjgriffith: 2) Does this design work ok with backends that can only replicate by pool (i.e. netapp)? I guess you'd end up with a bunch of replicated volumes you didn't ask for?17:36
*** marcusvrn1 has joined #openstack-cinder17:36
*** thangp has joined #openstack-cinder17:37
*** hemna_ has quit IRC17:37
*** rushil_ has joined #openstack-cinder17:37
jgriffithDuncanT: which is kinda why I wanted to push as much as possible in to type and driver17:37
jgriffithand let them deal with it :)17:37
jgriffithso the idea is that it will work regardless of the impl in the driver17:37
jgriffithas ugly as it might be17:37
jgriffithDuncanT: so in the case of folks that require a pool level replication....17:38
DuncanTI'd like to see their thoughts on this approach... I like the new API but worth getting the thoughts of NetApp since they're a special case. If they don't have useful feedback I'm fine ignoring them17:38
jgriffithDuncanT: 1. Sucks to be them17:38
jgriffithDuncanT: 2. It should work by create --type replicated; tells the driver "hey put this in an appropriate pool"17:39
*** crose has quit IRC17:39
DuncanTBut then your enable/disable replication calls make less sense17:39
thingeejgriffith, xyang1 driver email announcement
jgriffithDuncanT: yeah, so I believe this will actually make this "easier" for them17:39
jgriffithyeah, the enable/disable was a requirement from "others"17:39
thingeeand anyone else who wants to give feedback on the new driver email
jgriffithDuncanT: personally my thought was that wasn't needed17:40
*** rushil has quit IRC17:40
jgriffithDuncanT: but I can see where it's useful I guess17:40
jgriffithDuncanT: one thought I had was just "retype" to non-replicated17:40
*** reed_ has quit IRC17:41
jgriffithDuncanT: problem there becomes "what do I do with the target volumes"17:41
*** marcusvrn has quit IRC17:41
*** marcusvrn_ is now known as marcusvrn17:41
*** belmoreira has joined #openstack-cinder17:41
*** alexpilotti has quit IRC17:41
jgriffiththingee: LGTM17:42
thingeejgriffith: thanks17:42
jgriffiththingee: "merged by" should eliminate confusion I think17:42
*** crose has joined #openstack-cinder17:43
jgriffiththingee: only other thing pending here is stuff like adding what you call a "fabric"17:43
DuncanTjgriffith: Yeah, I'm not sure how to square that circle entirely. It is certainly no worse than what is already there for them17:43
*** belmoreira has quit IRC17:43
thingeejgriffith: oh yeah, that's a good point.17:43
DuncanTthingee: I'd use the term 'volume drivers' in the first line, personally17:44
jgriffiththingee: or what I call a "data-transport" or what someobdy else calls a "what's-it-ma-jig"17:44
thingeejgriffith: we should also include that in the deadline discussion. I would like us all to be on the same page.17:44
thingeejgriffith: definitely17:44
thingeeDuncanT: +117:44
jgriffiththingee: yeah, I still have to convince people why the "seperate" driver inheritting from multiple base's is a bad idea17:44
eharneye0ne: some interesting comments came up about the RBD tpool patch in
DuncanTjgriffith: When designing replication initially, the ability to test that replication was actually working, without stopping it, was high on the list of desires... I guess that would be 'trigger a snap on the slave'? Thoughts on that requirement?17:45
*** porrua has joined #openstack-cinder17:46
thingeeok, all I need is xyang1's approval and I'll send it out17:46
jgriffithDuncanT: yeah... perhaps17:46
jgriffithDuncanT: so this is where I strayed from others17:46
jgriffithDuncanT: IMO the admin piece is kinda "seperate"17:46
jgriffithDuncanT: I don't think end user should have acces to things lke "start/stop" replicate, or fail over17:47
jgriffithDuncanT: they should just "pay for replciated volume" and move on17:47
DuncanTIs that test an admin feature though?17:47
*** fthiagogv has quit IRC17:47
jgriffithDuncanT: it shold "just work"17:47
jgriffithDuncanT: and yes, admin needs to have ability to manage it17:47
jgriffithDuncanT: and test it is probably good too17:48
*** rushil_ has quit IRC17:48
jgriffithDuncanT: although I do feel that Vendors should provide easy mechanisms for that17:48
jgriffithDuncanT: so maybe an admin API call that says "verify_replcation_status"17:48
jgriffithDuncanT: and it's up to the driver to implement "some" way of reporting that17:48
jgriffithDuncanT: but if you want to actually 'fail it' and test it that way, that's another ball of yarn I'll have to think about17:49
e0neeharney: thanks for notification!17:49
jgriffithDuncanT: FYI I'm also trying to not "boil the ocean"17:49
DuncanTjgriffith: One piece of feedback we got is that prospective users of replication have relatively little faith in admins to have set things up correctly17:49
*** patrickeast has joined #openstack-cinder17:50
jgriffithDuncanT: yeah, I get that... but "welcome to the cloud"17:50
DuncanTjgriffith: +1 on not boiling the ocean, more a 'could this be added later without breaking the design'17:50
jgriffithDuncanT: if you're not getting what you're paying for, find another cloud :)17:50
jgriffithDuncanT: yes, for sure (added later)17:50
jgriffithDuncanT: and quite possibly up front17:50
DuncanTjgriffith: Like backups, replication doesn't exist if you can't test it regularly17:50
jgriffithDuncanT: just wasn't super high on my priority list17:50
jgriffithDuncanT: yeah, that's definitely fair17:51
*** merooney has joined #openstack-cinder17:51
jgriffithDuncanT: so simple first pass is an API call that just says "fail-over"17:51
jgriffithor whatever17:51
DuncanTSo we could just add a call that makes the slave do a snap, that would fullfil the requirement, I think?17:51
jgriffithI don't even mind if that's an end-user available thing I suppose17:51
DuncanT(I don't mean add it now, I mean make sure it logically should work)17:51
DuncanTpolicy.js can decide if it is end-user accessable17:52
jgriffithDuncanT: or yeah, your snap idea.  I have to think about that some more but it seems like that might be perfectly legit17:52
*** timcl1 has joined #openstack-cinder17:52
jgriffithDuncanT: one twist though is if we keep the ability to migrate to backends that aren't "configured" in the OS deployment17:52
*** timcl has quit IRC17:52
jgriffithDuncanT: maybe we modify that aspect so it's "always" configured, but not available for scheduling17:53
jgriffithDuncanT: probably cleaner that way17:53
DuncanTProbably cleaner, yeah17:53
jgriffithDuncanT: so you can access, including snaps etc, you just can't create on it17:53
jgriffithDuncanT: yeah, I like that a lot better I think17:53
DuncanTMakes fully remote replication (cross cloud) harder though... I guess once you add that, you get volumes that magically appear on the other side, and you do your snap or whatever there, that can work17:54
jgriffithDuncanT: so that's the thing, I have a number of customers that we're doing that with right now17:55
jgriffithDuncanT: it was actually fairly easy17:55
jgriffithDuncanT: I stuffed all of the replication code in extra-specs and in the driver17:55
jgriffithDuncanT: nothing else17:55
DuncanTSo what happens on the other side of the replication?17:55
jgriffithDuncanT: what do you mean exactly?17:56
jgriffithDuncanT: so SF-Region-1 and SF-Region-217:56
jgriffithDuncanT: replciate SF-R1 to SF-R217:56
jgriffithDuncanT: users in cloud in region 2 can access the volume from SF-Region-217:57
jgriffithDuncanT: but they can't "write" to it17:57
jgriffithdidn't mess with 2 way rep17:57
*** haomaiwa_ has joined #openstack-cinder17:57
*** rongze has joined #openstack-cinder17:58
jgriffithbut I can give them a clone/snap of the volume and they can do whatever they like17:58
jgriffithand if something blows up on the first cluster, they have access to the second (although it's not optimal)17:58
jgriffithuntil the original device is back online and syncs back up17:59
DuncanTGot you17:59
jgriffithDuncanT: have to run for a bit... bbl17:59
DuncanTHow do things get into the cinder db in region2? Or shared db?17:59
jgriffithDuncanT: good chat17:59
jgriffithDuncanT: :)17:59
DuncanTOk, chat later17:59
jgriffithDuncanT: so that was the "fun" part17:59
*** marcusvrn has quit IRC17:59
jgriffithDuncanT: I did some hacks to the DB code17:59
jgriffithDuncanT: it's not pretty, but it's for a very specific use case18:00
jgriffithDuncanT: it would never fly in a proper cinder review18:00
DuncanTJust calling 'cinder import' on the far side pretty much does it AFAICT18:00
*** e0ne is now known as e0ne_18:00
DuncanTYou could do it with a cron job18:00
jgriffithDuncanT: yes, in combination with a special table18:00
jgriffithDuncanT: I used a periodic18:00
jgriffithdid it all in/from the driver18:00
jgriffithDuncanT: and created a special R/O type18:01
xyang1thingee: Hi, in a meeting, I'll read it now18:01
DuncanTLooks like that can be added in phase 2 then18:01
*** merooney has quit IRC18:01
jgriffithDuncanT: yeah, maybe18:01
jgriffithalright... really going now :)18:01
*** porrua has left #openstack-cinder18:02
*** Yogi1 has joined #openstack-cinder18:02
*** rongze has quit IRC18:03
*** annegentle has quit IRC18:04
*** e0ne_ is now known as e0ne18:04
*** e0ne has quit IRC18:06
*** Zhongjun has quit IRC18:09
xyang1thingee: Hi, I added a suggestion/comments there, especially about os-brick18:11
*** luv_ has joined #openstack-cinder18:12
*** rushil has joined #openstack-cinder18:12
*** luv has quit IRC18:13
*** luv_ is now known as luv18:13
*** angela-s has joined #openstack-cinder18:18
*** leeantho has quit IRC18:22
*** leeantho has joined #openstack-cinder18:23
*** sgotliv has quit IRC18:24
thingeexyang1: correct, os-brick drivers are a separate discussion18:30
openstackgerritMitsuhiro Tanino proposed openstack/cinder: LVM: Support efficient data copy using "dd" for create_cloned_volume
xyang1thingee: Ok, thanks18:30
*** MRay has quit IRC18:30
*** merooney has joined #openstack-cinder18:32
*** MentalRay has joined #openstack-cinder18:34
thingeexyang1: I revised it18:37
*** merooney has quit IRC18:37
xyang1thingee: Looks good now. Thanks!18:37
thingeexyang1: thanks for the feedback18:38
thingeealright going to send it out18:38
xyang1thingee: Sounds good18:38
*** merooney has joined #openstack-cinder18:40
*** ebalduf has quit IRC18:41
angela-sthingee: Thanks for changing the Assignee of these BPs to my name.
angela-sthingee: we'll have CI running in early June18:43
*** e0ne has joined #openstack-cinder18:44
*** aix has quit IRC18:47
*** ebalduf has joined #openstack-cinder18:55
*** rongze has joined #openstack-cinder18:59
*** annegentle has joined #openstack-cinder19:00
*** MentalRay has quit IRC19:01
*** Longgeek has quit IRC19:01
*** Longgeek has joined #openstack-cinder19:01
*** Svedrin has quit IRC19:01
*** Longgeek has quit IRC19:02
*** Longgeek has joined #openstack-cinder19:02
*** Longgeek has quit IRC19:04
*** rongze has quit IRC19:04
*** Longgeek has joined #openstack-cinder19:04
*** Rockyg_ has joined #openstack-cinder19:12
*** merooney has quit IRC19:13
*** Svedrin has joined #openstack-cinder19:17
*** merooney has joined #openstack-cinder19:21
jgriffithhemna: xyang1 smcginnis jungleboyj
jgriffithI'll leave it up to the rest of you how you want to run the prez, I'm a moron and thought it was a panel :(19:31
*** rmesta has quit IRC19:31
*** belmoreira has joined #openstack-cinder19:32
*** e0ne has quit IRC19:32
*** rmesta has joined #openstack-cinder19:32
xyang1jgriffith: Hi, I thought we have a panel after the presentation?  So part prez, part panel19:33
smcginnisxyang1, jgriffith: That sounds good to me.19:33
*** belmoreira has quit IRC19:33
jgriffithxyang1: cool19:33
smcginnisI think it will be nice to run through the preso to give some background.19:33
*** e0ne has joined #openstack-cinder19:33
smcginnisThen plenty of time for Q&A/discussion.19:33
smcginnisMy impression is some people aren't even aware FC is an option for cinder.19:34
smcginnisSo it would be good to give the overview of the preso so they understand the current state.19:34
*** lpabon has quit IRC19:34
xyang1jgriffith: have you got a moderator for the cinder dev session?19:35
smcginnisjgriffith: Presentation looks awesome!19:35
jungleboyjxyang1: That was my understanding as well.19:35
jgriffithxyang1: yes19:35
jgriffithsmcginnis: that's all hemna 's work, not mine19:36
jgriffithxyang1: Ed Balduf is going to moderate for us19:36
smcginnishemna: Presentation looks awesome! ;)19:36
jgriffithxyang1: We're finishing up questions and will get them out to everybody later today or tomorrow19:36
*** nkrinner has joined #openstack-cinder19:36
xyang1jgriffith: ok, that will be great19:36
*** eharney has quit IRC19:38
jungleboyjhemna: Nice work.  Good looking slides!19:39
* jungleboyj owes hemna or kmartin a beer.19:39
xyang1hemna: jgriffith jungleboyj smcginnis , for the FC Session Q&A, should we also prepare some questions if no one asks any?19:40
jungleboyjxyang1: That isn't a bad idea.19:40
smcginnisxyang1: Wouldn't hurt. Hopefully their is enough interest in this that there will be enough people asking questions, but it would be good to fill any gaps.19:42
xyang1jungleboyj: smcginnis, We can add some questions to the etherpad19:42
jungleboyj1) can you mix fabric types ... No.19:42
*** e0ne has quit IRC19:43
*** turul_ has quit IRC19:45
*** turul_ has joined #openstack-cinder19:46
*** jwcroppe has joined #openstack-cinder19:48
*** e0ne_ has joined #openstack-cinder19:48
*** IanGovett has joined #openstack-cinder19:48
*** IanGovett1 has quit IRC19:50
*** jwcroppe_ has quit IRC19:50
*** erlon has quit IRC19:51
*** xyang has quit IRC19:52
*** e0ne_ has quit IRC19:55
openstackgerritTom Swanson proposed openstack/cinder: Dell SC: Added support for alternate iscsi portals
*** e0ne_ has joined #openstack-cinder19:57
*** Rockyg_ has quit IRC19:59
*** e0ne_ has quit IRC19:59
*** e0ne has joined #openstack-cinder19:59
*** merooney has quit IRC19:59
*** rongze has joined #openstack-cinder20:00
*** merooney has joined #openstack-cinder20:00
*** Rockyg has joined #openstack-cinder20:01
e0nethingee: hi. what help do you need with removing api v1?20:01
*** rongze has quit IRC20:04
thingeee0ne: nova needs to use discovery
*** merooney has quit IRC20:06
thingeee0ne: heat as well
thingeee0ne: openstack client assumes just v1
*** delattec has quit IRC20:08
e0nethingee: oh, openstack client is very and very outdated:)20:08
*** dannywilson has joined #openstack-cinder20:09
thingeee0ne: looks like the openstack sdk requires v2
*** bswartz has joined #openstack-cinder20:10
*** eharney has joined #openstack-cinder20:10
e0nethingee: i could take a part with openstack client. i started looking on it to work with keystone api v320:11
thingeee0ne: ok thanks :)20:11
thingeee0ne: I'll make a launchpad bug and start adding other projects.20:12
e0nethingee: i'll try to ping heat team20:12
thingeeand assign you to the openstack client one?20:12
*** jungleboyj has quit IRC20:12
e0nethingee: it would be great20:12
thingeee0ne: ok, making myself a note. because I need to get some lunch before I have a headache20:12
e0nethingee: sure. i'll tale a break to sleep soon20:13
thingeeDuncanT: I got a question about allowing cinder backup-create to have a destination option where the user can specify a destination that's not within the same cloud.20:15
thingeeDuncanT: obviously some latency problems20:15
e0nethingee: do we have agenda for summit sprints?20:15
thingeee0ne: no, do you want to spearhead that? :)20:16
*** jwcroppe_ has joined #openstack-cinder20:16
*** esker has quit IRC20:16
*** merooney has joined #openstack-cinder20:17
e0nethingee: i'll create an etherpad20:17
thingeee0ne: that would be great. We do have all day, but I would like there to be time to actually work on things. Some things need a small discussion window.20:18
*** jwcroppe has quit IRC20:18
e0nei don't know how do you call it in English, in Russian we'ge got a proverb20:18
e0ne[from google translate]: initiative is punishable20:18
smcginnise0ne: +1 :)20:19
*** dustins has joined #openstack-cinder20:20
*** kmartin has quit IRC20:21
*** MentalRay has joined #openstack-cinder20:24
*** Lee1092 has quit IRC20:31
e0neactually, it was copy-pasted from
*** julim has quit IRC20:34
*** marcusvrn1 has quit IRC20:35
*** marcusvrn has joined #openstack-cinder20:35
openstackgerritMitsuhiro Tanino proposed openstack/cinder: LVM: Pass volume size in MiB to copy_volume() during volume migration
*** kmartin has joined #openstack-cinder20:38
*** annegentle has quit IRC20:39
*** merooney has quit IRC20:39
*** annegentle has joined #openstack-cinder20:39
*** timcl1 has quit IRC20:40
akerre0ne: i think that's like "the tallest blade of grass gets cut" or something20:42
*** jwcroppe has joined #openstack-cinder20:43
e0neakerr: thanks!20:43
*** jungleboyj has joined #openstack-cinder20:43
openstackgerritPatrick East proposed openstack/cinder-specs: Generic image volume cache functionality
e0nenoted it:)20:44
akerre0ne: that google translate one is much nicer though20:44
openstackgerritPatrick East proposed openstack/cinder-specs: Generic image volume cache functionality
*** jwcroppe_ has quit IRC20:45
e0neakerr: usally i use it only for separate words. it works good in such cases20:45
akerre0ne: I use it to talk to my russian in-laws; it makes for some interesting converasations20:46
e0neakerr: :)20:47
*** belmoreira has joined #openstack-cinder20:49
*** jwcroppe_ has joined #openstack-cinder20:51
*** nkrinner has quit IRC20:51
openstackgerritOpenStack Proposal Bot proposed openstack/cinder: Updated from global requirements
openstackgerritTom Barron proposed openstack/cinder: Use nfs_oversub_ratio when reporting pool capacity
*** jwcroppe has quit IRC20:53
*** MentalRay has quit IRC20:53
tbarronok, who is going to merge something that makes me rebase that one ^^^ a third time in two days :-)20:54
smcginnistbarron: Is that a challenge. ;)20:56
*** rushil has quit IRC20:56
*** rushil has joined #openstack-cinder20:57
tbarronsmcginnis: I don't know the emoticon for pointing my first two fingers at my eyes, then pointing them at you20:57
smcginnistbarron: Hah, that would be awesome. Someone needs to invent that.20:57
tbarronsmcginnis: actually I'm grateful for all the log cleanup in our drivers, will make it much easier to see good correct examples to copy going forwards20:58
*** akerr has quit IRC20:58
*** ronis has quit IRC20:59
*** rongze has joined #openstack-cinder21:00
*** thangp has quit IRC21:00
smcginnistbarron: I hope so. Especially after all of that.21:01
*** rongze has quit IRC21:05
*** mriedem has quit IRC21:06
*** dustins has quit IRC21:09
*** ekarlso has quit IRC21:09
*** larsks has quit IRC21:09
*** wolsen has quit IRC21:09
*** crose has quit IRC21:09
*** MentalRay has joined #openstack-cinder21:10
*** wolsen has joined #openstack-cinder21:10
*** ekarlso has joined #openstack-cinder21:10
*** primechu_ has joined #openstack-cinder21:10
*** larsks has joined #openstack-cinder21:11
*** crose has joined #openstack-cinder21:11
*** primechu_ has quit IRC21:11
*** primechuck has quit IRC21:12
*** primechu_ has joined #openstack-cinder21:12
hemnajgriffith, thanks for the slides.   I pulled your slides into the deck.21:12
hemnakmartin, jgriffith, xyang1, jungleboyj, smcginnis,
smcginnishemna: Thanks!21:13
hemnacan everyone check it out and make sure it's good?21:13
smcginnisPreview Failed21:13
* smcginnis shakes fist at dropbox21:14
smcginnisAw, Snap! Something went wrong21:15
* smcginnis shakes fist at chromium21:15
smcginnisMaybe this is a sign I should leave early today.21:15
smcginnisNitpick - is it "defacto" or "de facto"? I believe it's two words.21:16
tbarronping hemna21:18
hemnawell, it's not defucto21:18
*** Yogi1 has quit IRC21:18
hemnatbarron, yah21:18
Swansonde facto21:18
tbarronhemna: u see my pm?21:19
xyang1hemna: thanks! Can you remove the slides after Q&A?21:19
hemnano sorry21:19
tbarronno prob21:19
hemnaok updated to 'de facto'21:19
tbarronjust so there are no secrets, I'm asking hemna if he'll list NetApp on page 14, kilo line21:19
*** vilobhmm11 has joined #openstack-cinder21:20
tbarronwe did get our CI going, FC is supported in kilo21:20
hemnatbarron, sure...21:20
tbarronand there's space on the line afer X-IO :-)21:21
hemnajust 'NetApp' ?21:21
*** vilobhmm12 has joined #openstack-cinder21:21
*** haomaiwa_ has quit IRC21:21
smcginnisHah, yeah, I think there's room for you. :)21:21
*** vilobhmm11 has quit IRC21:21
tbarronyes please21:21
hemnaok done21:21
tbarronOK if I email you a logo to go with the others?21:21
tbarronthanks!  I'm going home now, outa your hair.21:22
hemnano worries21:22
*** jwcroppe has joined #openstack-cinder21:22
*** vilobhmm11 has joined #openstack-cinder21:22
*** vilobhmm1 has quit IRC21:23
*** vilobhmm1 has joined #openstack-cinder21:24
*** mriedem has joined #openstack-cinder21:25
*** jwcroppe_ has quit IRC21:25
*** vilobhmm12 has quit IRC21:26
*** vilobhmm11 has quit IRC21:27
*** annegentle has quit IRC21:32
patrickeasthemna: if you are taking requests, wanna put Pure Storage on that list of FC drivers for Liberty? :D me and dannywilson have one just about ready to go up for review21:34
*** kmartin has quit IRC21:34
hemnapatrickeast, updated21:35
patrickeasthemna: yay thanks!21:36
patrickeastim sure i would have gotten a talking to from the marketing folks if we didn’t make it on that slide21:36
patrickeastlooks like that will be a great session though, judging from the slides21:37
patrickeasti’m excited to attend21:37
hemnayah that preso is a Cinder party21:37
mtaninojungleboyj: hi, thank you for the review.21:40
mtaninojungleboyj: Could you judge the importance of that bug?
openstackLaunchpad bug 1454835 in Cinder "[LVM] Volume migration destroys volume data" [Undecided,In progress] - Assigned to Mitsuhiro Tanino (mitsuhiro-tanino)21:41
*** crose has quit IRC21:44
jungleboyjmtanino: Welcome.  Seems like a 'high' to me.  Targeted to L1.21:47
mtaninojungleboyj: Thank you :) Also I will try to backport it to stable branch.21:48
jungleboyjOk, that seems like a reasonable plan.21:50
*** belmoreira has quit IRC21:54
*** IanGovett has quit IRC21:57
*** IanGovett has joined #openstack-cinder21:57
*** mriedem is now known as mriedem_away21:58
*** vmtrooper has joined #openstack-cinder21:59
*** rongze has joined #openstack-cinder22:01
*** vmtroope_ has joined #openstack-cinder22:02
*** vmtrooper has quit IRC22:03
*** rongze has quit IRC22:06
openstackgerritPavel Boldin proposed openstack/cinder: Add `iscsi_bsoflags' configuration option
*** lpetrut has quit IRC22:13
*** eharney has quit IRC22:14
*** vilobhmm1 has quit IRC22:16
*** IlyaG_ has quit IRC22:17
*** vilobhmm1 has joined #openstack-cinder22:17
*** jamielennox|away is now known as jamielennox22:19
*** julim has joined #openstack-cinder22:20
*** rushil has quit IRC22:24
*** rushil has joined #openstack-cinder22:25
*** e0ne has quit IRC22:28
*** dannywilson has quit IRC22:30
*** dannywilson has joined #openstack-cinder22:32
*** patrickeast has quit IRC22:33
*** patrickeast_ has joined #openstack-cinder22:33
*** Rockyg has quit IRC22:33
*** ekarlso has quit IRC22:34
*** ekarlso has joined #openstack-cinder22:34
nikeshmpatrickeast: hi22:36
*** IlyaG has joined #openstack-cinder22:36
patrickeast_nikeshm: hey22:37
-openstackstatus- NOTICE: Gerrit and Zuul are going offline for reboots to fix a security vulnerability.22:38
*** ChanServ changes topic to "Gerrit and Zuul are going offline for reboots to fix a security vulnerability."22:38
nikeshmpatrickeast : quick question,i want to ping floating ip of devstack vms created by nodepool,ip is in my network range,but on pinging its giving "ping: sendmsg: Operation not permitted"22:39
nikeshmis any link to enable that22:39
*** dims_ has joined #openstack-cinder22:40
nikeshmnikeshm:i am using nova-network in devstack22:40
patrickeast_nikeshm: do you have the security groups set up to allow ping?
*** dims has quit IRC22:41
*** madskier has joined #openstack-cinder22:42
*** dims has joined #openstack-cinder22:43
*** rushil has quit IRC22:44
*** ebalduf has quit IRC22:46
nikeshmpatrickeast_ thanks22:47
nikeshmnow able to ping22:47
*** dims_ has quit IRC22:47
*** madskier has quit IRC22:47
*** openstackgerrit has quit IRC22:47
nikeshmbut not in old vm22:47
nikeshmin newly created vm, i am able to ping22:47
patrickeast_nikeshm: hmm yea it might not change the access rules for vm’s that were already created22:48
jgriffithmtanino: if gerrit ever comes back I'll have a look at that right away22:48
*** angela-s has quit IRC22:48
*** openstackgerrit has joined #openstack-cinder22:49
*** vmtroope_ has quit IRC22:49
jgriffithnikeshm: the firewall won't update for an instance tht's already up22:49
jgriffithnikeshm: you can try "nova network-reset"22:49
mtaninojgriffith: thanks.22:49
nikeshmjgriffith: thanks,BTW see you all in summit :)22:51
jgriffithnikeshm: cool!22:51
nikeshmnikeshm:did you attend cloud foundry summit in santa clara?22:52
nikeshmjgriffith:did you attend cloud foundry summit in santa clara??22:52
jgriffithcloud foundry... what's that :P22:52
jgriffithno, I did not22:52
nikeshmits a PaaS22:52
nikeshmPlatform as a service22:52
jgriffithnikeshm: I was joking22:53
jgriffithnikeshm: but no, I did not attend22:53
nikeshmi attended since it was near to my stay22:53
*** vmtrooper has joined #openstack-cinder22:54
jgriffithnikeshm: was it a good event?22:54
*** rushil has joined #openstack-cinder22:54
nikeshmyes it was22:54
nikeshmlot of companies participated22:55
*** ChanServ changes topic to "The OpenStack Block Storage Project Cinder | The New Kids On the Block |"22:56
-openstackstatus- NOTICE: Gerrit and Zuul are back online.22:56
*** asselin_ has joined #openstack-cinder22:57
*** dannywilson has quit IRC22:58
*** patrickeast_ has quit IRC22:58
*** patrickeast has joined #openstack-cinder23:01
*** asselin_ has quit IRC23:01
*** dannywilson has joined #openstack-cinder23:01
*** rongze has joined #openstack-cinder23:02
*** rongze has quit IRC23:06
*** Rockyg has joined #openstack-cinder23:07
uberjayquick (?) question about our openstack CI setup -- we should only be triggering on changes to the openstack/cinder repository, correct? (i'm also a little bit confused about *when* a change should be run through CI. I assume zuul is doing the right thing, but it's not triggering test runs as often as I assumed it would. :)23:09
patrickeastuberjay: what do you have your zuul triggers set up for?23:12
patrickeastuberjay: it should be running one for a few different scenarios, for example mine looks like which (afaik) triggers off of everything it should be (and maybe a tad more)23:14
patrickeastuberjay: i think some folks set theirs up to only trigger a build after jenkins has given a +123:14
ameadeuberjay: patrickeast: +1 that's what I do23:14
uberjaypatrickeast: ah, ok. thanks for the example, it's very helpful! i can see wanting to not waste cycles until jenkins gives it a +1, too. i'll have to give that some thought.23:16
patrickeastthe only reason I don’t do that is sometimes changes move fast and by the time jenkins has given a +1 if you then have to wait maybe 45 mins the patch might have already gotten a +2/A23:16
patrickeastdefinitely worth looking into if your system can’t keep up with the queue23:16
uberjayriight...ok. totally makes sense.23:16
*** primechu_ has quit IRC23:19
*** sc has quit IRC23:20
jgriffithmtanino: jungleboyj see my notes in the patch as well as the bug:
jgriffithmtanino: I'm actually now not convinced that bug is exactly critical as it's been like that for almost a year :)23:22
*** sc has joined #openstack-cinder23:22
*** rmesta has quit IRC23:30
*** vmtrooper has quit IRC23:32
*** setmason has quit IRC23:32
jgriffithjungleboyj: ping23:34
mtaninojgriffith: looking23:35
mtaninojgriffith: so, you recommend to change copy routine taking a GiB and fix related places to take GiB, right?23:37
mtaninohandling MiB is very confusiong.23:38
jgriffithmtanino: so probably but that's something to be careful23:39
*** MentalRay has quit IRC23:43
jgriffithmtanino: we can go with your patch, then figure out something better later23:44
mtaninojgriffith: it's an ideal but there is risks to fix many places, so we need to consider carefully.23:44
mtaninojgriffith: sure23:44
jgriffithmtanino: agreed 100%23:45
*** ociuhandu has quit IRC23:45
mtaninomight cause data corruptions for the volume :(23:46
*** tsekiyam_ has joined #openstack-cinder23:53
*** _cjones_ has quit IRC23:56
*** ebalduf has joined #openstack-cinder23:56
*** tsekiyama has quit IRC23:57
*** ebalduf has quit IRC23:57
*** mtanino has quit IRC23:57
*** tsekiyam_ has quit IRC23:58

Generated by 2.14.0 by Marius Gedminas - find it at!