Friday, 2016-03-04

openstackgerritTom Barron proposed openstack/cinder: Fix invalid uuid warnings in backup unit tests
*** andymaier_ has quit IRC00:06
*** yangyapeng has quit IRC00:08
*** ntpttr has quit IRC00:08
openstackgerritKurt Martin proposed openstack/cinder: Don't fail on clearing 3PAR object volume key
openstackgerritTom Barron proposed openstack/cinder: Fix retype failure when original has no volume type
openstackgerritPatrick East proposed openstack/cinder: Add backend id to Pure Volume Driver trace logs
*** laughterwym has joined #openstack-cinder00:27
*** diogogmt has quit IRC00:28
*** hemna is now known as hemnafk00:30
*** daneyon_ has quit IRC00:31
*** andymaier_ has joined #openstack-cinder00:32
*** Rodrigo_BR has joined #openstack-cinder00:34
Rodrigo_BRLive Migration failure: Unsafe migration: Migration may lead to data corruption if disks use cache != none00:34
Rodrigo_BRCan I perform live migration using NFS backend ?00:34
tbarronRodrigo_BR: hmm, can you tell us more about this? looks like there might be a launchpad bug, but I don't see the link.00:40
patrickeastRodrigo_BR: i think you should be able to, assuming you understand the risks you can get around that error with VIR_MIGRATE_UNSAFE in your migrations flags in nova.conf00:40
tbarronk, patrickeast knows more about this, I'll listen.00:42
patrickeasttbarron: haha, not sure about that... just quick to googling :p00:43
*** laughterwym has quit IRC00:43
patrickeasttbarron: Rodrigo_BR: found this and did a quick look at
openstackLaunchpad bug 1222682 in openstack-manuals "Live Migration does not work unless cache=none" [High,Won't fix]00:43
*** laughterwym has joined #openstack-cinder00:44
tbarronpatrickeast: Rodrigo_BR : so is there something special about NFS that would imply that "it's configured00:45
tbarronin a way that does not ensure data integrity once it is migrated"?00:45
* tbarron has a improbable thought that Rodrigo_BR might be on vacation from manila, where he knows a whole lot about such stuff, slumming over here after FF00:46
*** laughterwym has quit IRC00:48
patrickeasttbarron: from the docs it seems like having the cache enabled might mean that the files are not in sync for both systems, which could cause the unsafe migration00:49
Rodrigo_BRI am using the netapp driver in cinder00:50
Rodrigo_BRa post the configuration in
*** gouthamr has quit IRC00:51
tbarronpatrickeast: k, that makes sense.00:51
tbarronRodrigo_BR: looking ...00:51
tbarronRodrigo_BR: I think the safest thing to do is ask on #openstack-netapp00:52
tbarronRodrigo_BR: I used to work there and I don't know about the cacheiing.00:53
tbarronRodrigo_BR: offhand I would think it's a client-side thing, not NetApp or any server-side specific.00:53
tbarronRodrigo_BR: but I haven't checked that answer in detail.00:53
tbarronRodrigo_BR: you should be aware that it's Feature Freeze crunch right now, so people's attentions are on that.00:54
tbarronRodrigo_BR: unless it's absolutely critical, you'll probably get better response on that channel in a couple days.00:54
tbarronpatrickeast: Rodrigo_BR: thanks for pointing me to this NFS caching issue.  Learned a bit.00:56
Rodrigo_BR<tbarron> Thank You00:56
*** ntpttr has joined #openstack-cinder00:56
Rodrigo_BR<tbarron> I will try in ocenstack-netapp00:56
*** laughterwym has joined #openstack-cinder00:57
Rodrigo_BR* Sorry00:57
*** akshai has quit IRC00:58
*** akshai has joined #openstack-cinder01:02
*** laughter_ has joined #openstack-cinder01:04
*** laughter_ has quit IRC01:04
*** jaypipes has quit IRC01:04
*** laughter_ has joined #openstack-cinder01:04
*** laughter_ has quit IRC01:05
*** laughter_ has joined #openstack-cinder01:05
*** laughter_ has quit IRC01:06
*** laughterwym has quit IRC01:06
*** laughterwym has joined #openstack-cinder01:06
*** akshai has quit IRC01:08
*** laughter_ has joined #openstack-cinder01:10
*** laughterwym has quit IRC01:10
*** laughter_ has quit IRC01:11
*** yangyapeng has joined #openstack-cinder01:11
*** laughterwym has joined #openstack-cinder01:11
*** sasukeh has joined #openstack-cinder01:15
*** ntpttr has quit IRC01:16
*** yangyapeng has quit IRC01:17
*** Suyash has quit IRC01:17
*** garthb has quit IRC01:19
*** gouthamr has joined #openstack-cinder01:25
*** EinstCrazy has joined #openstack-cinder01:26
*** gouthamr_ has joined #openstack-cinder01:28
*** akshai has joined #openstack-cinder01:31
*** gouthamr has quit IRC01:31
*** bardia has quit IRC01:34
*** diogogmt has joined #openstack-cinder01:35
openstackgerritMerged openstack/cinder: Tintri image direct clone
*** laughterwym has quit IRC01:36
*** laughterwym has joined #openstack-cinder01:37
*** cheneydc has joined #openstack-cinder01:37
*** mtanino has quit IRC01:37
*** akshai_ has joined #openstack-cinder01:39
*** laughte__ has joined #openstack-cinder01:39
*** laught___ has joined #openstack-cinder01:40
*** laughterwym has quit IRC01:41
*** akshai has quit IRC01:42
*** laughte__ has quit IRC01:44
*** laught___ has quit IRC01:45
*** laughterwym has joined #openstack-cinder01:48
openstackgerritAnthony Lee proposed openstack/python-brick-cinderclient-ext: Query volume paths implementation
openstackgerritMerged openstack/cinder: Update unittest for Storwize pool-aware-cinder-scheduler
*** apoorvad has quit IRC01:53
*** bjornar has quit IRC01:55
*** esker has joined #openstack-cinder02:01
*** rhagarty_ has quit IRC02:04
*** esker has quit IRC02:07
*** vilobhmm11 has quit IRC02:09
*** rhagarty has joined #openstack-cinder02:12
openstackgerritPeter Wang proposed openstack/cinder: VNX: Update replication for v2.1
*** bardia has joined #openstack-cinder02:16
*** baojg has joined #openstack-cinder02:17
*** ntpttr has joined #openstack-cinder02:22
*** andymaier_ has quit IRC02:26
*** ntpttr has quit IRC02:28
*** Lee1092 has joined #openstack-cinder02:31
*** Rodrigo_BR has quit IRC02:32
*** gcb has quit IRC02:34
*** apoorvad has joined #openstack-cinder02:36
*** rhagarty has quit IRC02:36
openstackgerritWilson Liu proposed openstack/cinder: Huawei: Record and check LUN wwn
*** haomaiwang has joined #openstack-cinder02:39
*** esker has joined #openstack-cinder02:43
*** houming has joined #openstack-cinder02:44
*** haomaiwang has quit IRC02:45
*** esker has quit IRC02:48
*** akshai_ has quit IRC02:51
*** gcb has joined #openstack-cinder02:53
*** akshai has joined #openstack-cinder02:56
*** haomaiwang has joined #openstack-cinder02:56
*** cdelatte has joined #openstack-cinder02:59
*** akshai has quit IRC03:00
*** haomaiwang has quit IRC03:01
*** links has joined #openstack-cinder03:01
*** mmandel has joined #openstack-cinder03:08
*** haomaiwang has joined #openstack-cinder03:12
*** vilobhmm11 has joined #openstack-cinder03:15
*** haomaiwang has quit IRC03:17
*** esker has joined #openstack-cinder03:18
*** vilobhmm111 has joined #openstack-cinder03:22
*** vilobhmm11 has quit IRC03:25
openstackgerritMerged openstack/cinder: 3PAR use same LUN id for each export path
openstackgerritMerged openstack/cinder: EMC VMAX - Limit SG and MV to 64 characters
*** vilobhmm111 has quit IRC03:39
*** dims has quit IRC03:40
*** dims has joined #openstack-cinder03:43
*** dims has quit IRC03:44
*** links has quit IRC03:44
*** apoorvad has quit IRC03:45
*** gouthamr_ has quit IRC03:48
*** esker has quit IRC03:49
*** yangyapeng has joined #openstack-cinder03:50
*** Poornima has joined #openstack-cinder03:51
*** hemna has joined #openstack-cinder03:55
*** asselin__ has joined #openstack-cinder04:02
*** mmandel has quit IRC04:02
*** jdurgin1 has joined #openstack-cinder04:04
*** laughterwym has quit IRC04:13
*** laughterwym has joined #openstack-cinder04:15
*** hemna has quit IRC04:16
*** garthb has joined #openstack-cinder04:17
*** laughterwym has quit IRC04:27
*** esker has joined #openstack-cinder04:27
*** sborkows has quit IRC04:27
*** links has joined #openstack-cinder04:27
*** yangyape_ has joined #openstack-cinder04:27
*** openstack has joined #openstack-cinder14:08
openstackgerritPhilipp Marek proposed openstack/cinder: DRBD: Policy-based waiting for completion
*** haomaiwang has joined #openstack-cinder14:13
*** mriedem has joined #openstack-cinder14:15
*** knikolla has joined #openstack-cinder14:21
*** edtubill has joined #openstack-cinder14:22
*** ildikov has joined #openstack-cinder14:24
*** dustins has joined #openstack-cinder14:25
*** Julien-zte has joined #openstack-cinder14:26
*** dansmith is now known as superdan14:29
smcginnisflip214: You commented "new version coming up Monday", but is that it? ^^14:31
*** merooney has quit IRC14:32
e0nesmcginnis: hi. are we going to get this in Mitaka or it's too late?14:33
smcginnise0ne: It's too late. I had to cut the release yesterday for that.14:34
smcginnise0ne: So we have a very limited 0.1.0 release now.14:34
e0nesmcginnis: ok, thanks14:34
smcginnise0ne: Wish we could have gotten more in there, but it's a good start.14:34
e0nesmcginnis: do you know when will we got stable branches for clients?14:35
openstackgerritIvan Kolodyazhny proposed openstack/python-brick-cinderclient-ext: Require root permissions for CLI operations
smcginnise0ne: If I remember right they get branched at RC time.14:35
*** merooney has joined #openstack-cinder14:35
*** omolchanov1 has quit IRC14:35
e0nesmcginnis: ok, I was not sure about clients14:36
*** omolchanov has joined #openstack-cinder14:36
*** openstackgerrit_ has quit IRC14:36
smcginnise0ne: Yeah, they are a little different.14:36
*** dave-mccowan has joined #openstack-cinder14:37
*** mmandel has joined #openstack-cinder14:37
*** openstackgerrit_ has joined #openstack-cinder14:38
*** eharney has joined #openstack-cinder14:39
mriedemcan one of the cinder drivers accept the liberty nomination for this?
openstackLaunchpad bug 1523646 in OpenStack Compute (nova) liberty "Nova/Cinder Key Manager for Barbican Uses Stale Cache" [High,In progress] - Assigned to Matt Riedemann (mriedem)14:41
mriedemthe fix is released on liberty, i'm just doing LP paperwokr14:41
*** links has quit IRC14:41
smcginnismriedem: Don14:42
*** diogogmt has quit IRC14:42
*** rhagarty has joined #openstack-cinder14:45
*** omolchanov has quit IRC14:46
openstackgerritMerged openstack/cinder: Imported Translations from Zanata
*** rhagarty_ has joined #openstack-cinder14:46
*** omolchanov has joined #openstack-cinder14:47
*** jgregor has joined #openstack-cinder14:48
*** rhagarty has quit IRC14:48
openstackgerritOpenStack Proposal Bot proposed openstack/cinder: Updated from global requirements
*** cknight has joined #openstack-cinder14:55
*** jwcroppe has joined #openstack-cinder14:55
EmilienMHi Cinder folks, I've been trying (and failed) to deploy Cinder API with SSL enabled (without apache) - is it something that is supposed to work?14:56
eharneythingee: the dependency for this patch has landed in openstack/requirements now:
*** martyturner has quit IRC14:57
*** diogogmt has joined #openstack-cinder14:57
*** merooney has quit IRC14:58
smcginnisEmilienM: I haven't tried.14:58
*** baumann has joined #openstack-cinder14:58
smcginnisEmilienM: I know we recently had a patch to switch some wsgi stuff over to oslo. Hopefully we didn't miss something there.14:58
EmilienMsmcginnis: and when I deploy with apache, it does not work either, I've got 500 errors14:58
EmilienMin cinder.conf; I give all ssl options but cinder keeps starting in http14:59
smcginnisEmilienM: Does it work for you with non-SSL and Apache?14:59
EmilienMno !14:59
EmilienMI have some logs, etc I can show you, a sec14:59
smcginnisEmilienM: Maybe you're apache setup isn't right?14:59
EmilienMI copied devstack ;è)14:59
smcginnisEmilienM: If you could file a bug that would be great.15:00
EmilienMand I've got
*** haomaiwang has quit IRC15:01
smcginnisEmilienM: This is off of master?15:01
*** tonyb has quit IRC15:01
EmilienMsmcginnis: running mitaka b215:01
*** tonyb has joined #openstack-cinder15:01
*** haomaiwang has joined #openstack-cinder15:01
e0nesmcginnis: to be sure: we're accepting only bugfixes now, aren't we? E.g.
smcginnise0ne: There were some of those that were going to make upgrades easier that we might still want to get in.15:02
eharneye0ne: xyang1: could you look at my comment here before this patch merges?
smcginnise0ne: Bug fix only, but at the discretion of the core team if we want to accept anything else based on risk of introducing changes that might break things.15:02
*** timcl has quit IRC15:02
smcginnise0ne: And replication support.15:02
*** yuriy_n17 has joined #openstack-cinder15:03
e0nesmcginnis: yes. I remember about replication support15:03
*** timcl has joined #openstack-cinder15:03
smcginnisEmilienM: This came in after b2:
smcginnisEmilienM: I wonder if that might have an impact.15:03
*** martyturner has joined #openstack-cinder15:04
e0neeharney: I'm sorry, I thought it was addressed15:04
*** merooney has joined #openstack-cinder15:04
e0neeharney, xyang1: can we -2 or remove +A on it?15:04
smcginnise0ne: We don't need to block the lib patches either. Those are released, so anything that goes in now will just be in the next release we do post-Mitaka.15:05
*** kenhui has quit IRC15:05
openstackgerritEric Harney proposed openstack/cinder: Correcting thin provisioning behavior
eharneyhave to hit rebase to kick it out of the queue ^15:05
*** arecknag has quit IRC15:06
e0nesmcginnis: it's bacame too complicated. is it documented anywhere?15:07
smcginnise0ne: Is what documented?15:07
e0nesmcginnis: what do we accept now before RC1 and what should be -2'd15:07
xyang1e0ne: eharney removed +A15:09
e0nesmcginnis: to get released library for Mitaka without a stable branch is confusing, IMO15:09
smcginnise0ne: Really cinder is the only one that we need to worry about. The rest of the projects are libraries that are released on demand.15:09
smcginnise0ne: We don't _really_ do anything with stable branches on libs.15:09
xyang1eharney: are you suggesting to give warning now and throw exception in Newton?15:09
e0nesmcginnis: but we've got stable branch for cinderclient too15:10
eharneyxyang1: yes, that's how we normally do this15:10
xyang1eharney: ok15:10
*** markvoelker has quit IRC15:10
*** markvoelker has joined #openstack-cinder15:10
*** omolchanov has quit IRC15:10
*** omolchanov has joined #openstack-cinder15:11
eharneycould someone drop by this patch to get some missing requirements added?
*** Julien-zte has quit IRC15:13
e0neeharney: if it's only a driver requirements, do we really need it in  requirements.txt?15:13
eharneye0ne: we do if the driver doesn't do any handling around it not being present15:14
*** esker has joined #openstack-cinder15:14
openstackgerritMerged openstack/cinder: Fix invalid uuid warnings in backup unit tests
*** esker has quit IRC15:15
*** esker has joined #openstack-cinder15:15
*** timcl has quit IRC15:17
*** mvk has joined #openstack-cinder15:22
*** david_lyle__ has joined #openstack-cinder15:34
*** rhagarty_ has quit IRC15:36
*** david-lyle has quit IRC15:37
*** yangyapeng has joined #openstack-cinder15:38
*** yangyapeng has quit IRC15:39
*** lpetrut has quit IRC15:39
*** lpetrut has joined #openstack-cinder15:39
*** crose has joined #openstack-cinder15:41
*** laughterwym has quit IRC15:44
*** laughterwym has joined #openstack-cinder15:45
*** lpetrut has quit IRC15:47
*** kenhui has joined #openstack-cinder15:48
*** mtanino has joined #openstack-cinder15:49
*** jungleboyj has joined #openstack-cinder15:50
*** rhagarty has joined #openstack-cinder15:51
*** ntpttr has joined #openstack-cinder15:52
*** rhagarty_ has joined #openstack-cinder15:52
*** dims is now known as dimsum_15:53
*** merooney has quit IRC15:54
*** martyturner has quit IRC15:54
*** mriedem is now known as mriedem_meeting15:55
*** rhagarty has quit IRC15:56
*** timcl has joined #openstack-cinder15:56
*** martyturner has joined #openstack-cinder15:58
*** omolchanov has quit IRC15:58
*** omolchanov1 has joined #openstack-cinder15:58
*** merooney has joined #openstack-cinder15:58
*** haomaiwang has quit IRC16:01
*** hemnafk is now known as hemna16:04
openstackgerritScott DAngelo proposed openstack/cinder: support new HTTP microversion header
*** daneyon has joined #openstack-cinder16:07
*** laughterwym has quit IRC16:07
*** belmoreira has quit IRC16:07
*** laughterwym has joined #openstack-cinder16:08
*** e0ne has quit IRC16:12
*** merooney has quit IRC16:13
*** daneyon_ has joined #openstack-cinder16:13
openstackgerritGorka Eguileor proposed openstack/cinder: Refactor sqlalchemy service methods
openstackgerritGorka Eguileor proposed openstack/cinder: Add service_nodes table and related methods
openstackgerritGorka Eguileor proposed openstack/cinder: Add ServiceNode Versioned Object
openstackgerritGorka Eguileor proposed openstack/cinder: Change Job Distribution for HA A-A
*** daneyon has quit IRC16:16
*** ivase has joined #openstack-cinder16:17
*** david_lyle__ is now known as david_lyle16:17
*** kmartin__ has quit IRC16:17
*** laughter_ has joined #openstack-cinder16:18
*** timcl has quit IRC16:22
*** laughterwym has quit IRC16:22
mriedem_meetinge0ne must have left?16:23
mriedem_meetingjungleboyj: since you're stable CPL16:23
jungleboyjmriedem_meeting: Looking.16:25
*** laughter_ has quit IRC16:25
*** laughterwym has joined #openstack-cinder16:25
*** tongli has joined #openstack-cinder16:26
jungleboyjmriedem_meeting: Since it is a test impact I am fine with it.  What is up with the py34 failure though?16:26
*** andymaier_ has quit IRC16:27
jungleboyjmriedem_meeting: I looked at this yesterday and debated.  Thought I ahd voted.16:27
mriedem_meetingoh the py34 thing is the hyperv fialure that we fixed the other day16:27
mriedem_meetingi'll recheck16:27
jungleboyjmriedem_meeting: Ok.  Cool.16:27
mriedem_meetingi also replied to ivan in there16:28
jungleboyjYeah, I saw that.16:28
mriedem_meetingthe point is the 65K debug logging messages in a console log per unit test run16:28
mriedem_meetingwhich kills our ELK stack in infra16:28
*** rhagarty has joined #openstack-cinder16:28
*** rhagarty_ has quit IRC16:30
jungleboyjmriedem_meeting: +2 from me.  smcginnis can you take a look when you have a chance.  It is rechecking right now.16:30
smcginnisjungleboyj, mriedem_meeting: Done16:33
*** yuriy_n17 has quit IRC16:35
*** mriedem_meeting is now known as mriedem16:40
*** jaypipes has joined #openstack-cinder16:41
*** jaypipes is now known as leakypipes16:42
*** leeantho has joined #openstack-cinder16:42
*** rhagarty_ has joined #openstack-cinder16:45
*** rhagarty has quit IRC16:47
*** fthiagogv__ is now known as fthiagogv16:47
*** apoorvad has joined #openstack-cinder16:49
*** rcernin has quit IRC16:51
*** mmandel has quit IRC16:54
*** leeantho_ has joined #openstack-cinder16:59
*** omolchanov1 has quit IRC17:02
*** omolchanov has joined #openstack-cinder17:02
*** leeantho has quit IRC17:02
*** daneyon_ has quit IRC17:02
*** kmartin has joined #openstack-cinder17:02
*** daneyon has joined #openstack-cinder17:03
*** laughterwym has quit IRC17:09
*** laughterwym has joined #openstack-cinder17:10
*** mmandel has joined #openstack-cinder17:10
diablo_rojoDoes anyone know if there are volume attach_statuses besides attached, attaching, invalid and detached?  Or is that it?17:11
*** bjornar has joined #openstack-cinder17:13
*** daneyon has quit IRC17:13
*** laughterwym has quit IRC17:14
mc_nairdiablo_rojo: I'm seeing error_attaching and error_detaching also17:16
mc_nairdiablo_rojo: in the manager. Trying to see if anything else. Seems like maybe a detaching should be in there too? Haven't seen it yet17:16
diablo_rojomc_nair: Yeah I thought there would be a detaching too, but I haven't seen one17:17
openstackgerritMitsuhiro Tanino proposed openstack/cinder: Does not reset volume status when resetting migration status
openstackgerritMerged openstack/cinder: Convert huawei ISCSIDriver unit tests to RFC5737 addrs
*** martyturner has quit IRC17:22
scottdaattach_status does not completely map to volume status17:24
scottdaIt's really clunky that we need both, IMO17:24
*** martyturner has joined #openstack-cinder17:24
diablo_rojoscottda: Yeah I noticed not everything carried over, I just wanted to be sure I wasn't missing any valid statuses.17:24
*** haplo37 has joined #openstack-cinder17:28
*** Suyash has joined #openstack-cinder17:29
openstackgerritMerged openstack/cinder: EMC VMAX - Recreating SG when it has been deleted
openstackgerritOpenStack Proposal Bot proposed openstack/cinder: Updated from global requirements
*** ndipanov has quit IRC17:43
*** merooney has joined #openstack-cinder17:46
*** jistr has quit IRC17:46
*** vilobhmm11 has joined #openstack-cinder17:46
*** jwcroppe has quit IRC17:47
*** jwcroppe has joined #openstack-cinder17:48
*** vilobhmm111 has joined #openstack-cinder17:49
*** Thelo has quit IRC17:49
*** Thelo has joined #openstack-cinder17:50
*** vilobhmm11 has quit IRC17:51
*** salv-orlando has joined #openstack-cinder17:52
diablo_rojoscottda: Is availble a valid attach status? I just saw an instance of it in test_dothill.py17:53
*** akerr_ has quit IRC17:54
*** timcl has joined #openstack-cinder17:55
*** salv-orl_ has quit IRC17:55
*** leakypipes has quit IRC17:56
*** kenhui has quit IRC17:56
*** timcl has quit IRC18:02
scottdadiablo_rojo: I'm not positive. I can look in a little while...18:02
*** timcl1 has joined #openstack-cinder18:02
*** esker has quit IRC18:02
*** chris_morrell has joined #openstack-cinder18:04
*** ntpttr_ has joined #openstack-cinder18:07
*** ntpttr has left #openstack-cinder18:10
*** permalac has quit IRC18:12
*** merooney has quit IRC18:12
*** kenhui has joined #openstack-cinder18:14
*** ntpttr_ has quit IRC18:15
*** vilobhmm111 has quit IRC18:16
*** ntpttr_ has joined #openstack-cinder18:16
*** ntpttr_ has quit IRC18:16
*** jdurgin1 has quit IRC18:17
*** ntpttr_ has joined #openstack-cinder18:19
*** tongli has quit IRC18:21
*** merooney has joined #openstack-cinder18:21
*** ntpttr_ has quit IRC18:25
*** jordanP has quit IRC18:25
*** shakamunyi has joined #openstack-cinder18:29
*** ntpttr_ has joined #openstack-cinder18:30
*** ntpttr__ has joined #openstack-cinder18:31
*** ntpttr__ has quit IRC18:32
*** mrrowden has joined #openstack-cinder18:32
*** JoseMello has joined #openstack-cinder18:34
*** akerr has joined #openstack-cinder18:35
*** eharney has quit IRC18:35
*** ntpttr has joined #openstack-cinder18:37
*** ntpttr_ is now known as ntpttr18:37
*** yangyapeng has joined #openstack-cinder18:40
*** anshul has quit IRC18:44
*** yangyapeng has quit IRC18:45
*** merooney has quit IRC18:46
*** merooney has joined #openstack-cinder18:47
*** mriedem has quit IRC18:47
*** lprice1 has quit IRC18:48
*** mriedem has joined #openstack-cinder18:48
*** vilobhmm11 has joined #openstack-cinder18:48
*** lprice has joined #openstack-cinder18:48
*** lprice1 has joined #openstack-cinder18:49
*** lprice has quit IRC18:53
*** kenhui has quit IRC18:55
*** omolchanov has quit IRC18:56
*** omolchanov has joined #openstack-cinder18:56
*** akerr has quit IRC18:57
*** akerr has joined #openstack-cinder18:58
xyang1eharney: hi18:58
*** rlrossit has quit IRC19:00
*** omolchanov has quit IRC19:05
*** omolchanov1 has joined #openstack-cinder19:05
*** crose has quit IRC19:07
*** rlrossit has joined #openstack-cinder19:08
*** rlrossit_ has joined #openstack-cinder19:09
jgriffithdiablo_rojo: BTW, the detaching is in the "begin_detaching" method of volume.api as part of the conditional update19:09
*** martyturner has quit IRC19:09
jgriffithmc_nair: ^^19:09
*** rlrossit has quit IRC19:10
jgriffithsuppose it would be good to continue dulek 's enum work on the objects early in Newton19:10
mc_nairjgriffith: gotcha.  Interesting, so it looks like "detaching" is only part of the status, whereas "attaching" also gets pushed to attach_status and status19:12
jgriffithmc_nair: soemtimes we get sloppy :)19:13
*** ociuhandu has quit IRC19:13
*** eharney has joined #openstack-cinder19:13
jgriffithmc_nair: the attach/detach cod has gotten a bit wonky over the years I think.  But easy enough to fix up19:14
mc_nairjgriffith: we'll call it "code personality" :)19:14
jgriffithmc_nair: ohhh... I like that19:14
mc_nairjgriffith: sure, I don't think it's a big deal.  Was just curious cause grepping was getting attaching but no detaching19:14
jgriffithmc_nair: yeah, it does add confusion though :(19:14
jgriffithmc_nair: I hope to finish my attach/detach refactor when N opens up.  I"ll make a note to clean up the status stuff then as well19:15
smcginnisSeems kind of odd that we don't keep that in an enum.19:15
mc_nairsmcginnis: yea, would be nice.  Guessing that's why diablo_rojo was asking perhaps?19:16
smcginnisIt does make it pretty darn difficult to figure out what the valid states are as it is at the moment.19:17
*** timcl1 has quit IRC19:17
kmartinI guess this missing a few as well then
diablo_rojojgriffith: smcginnis Yeah I am working on the volume attach_status enum19:21
jgriffithdiablo_rojo: Ohh.. NICE!!19:21
diablo_rojojgriffith: I already have the snapshot status patch out there19:22
*** fthiagogv has quit IRC19:22
*** apoorvad has quit IRC19:22
jgriffithdiablo_rojo: Yeah, I think I already reviewed that19:22
jgriffithdiablo_rojo: prgress... progress :)19:22
diablo_rojojgriffith: I see that you reran solidfire on it and its happy19:23
jgriffithdiablo_rojo: oh.. stupid *&)( job failed again19:23
xyang1eharney: in this patch , check_for_setup_error is abstract method19:23
xyang1eharney: are you suggesting moving the check to
jgriffithdiablo_rojo: ahh.. yes, I remember now19:24
jgriffithdiablo_rojo: I'm +2, just waiting to get ceph to run now19:24
diablo_rojojgriffith: Not sure why gate-temptest-dsvm-full-ceph keeps failing though. I tried a recheck a few days ago and it never happened. So I rechecked again today.19:24
jgriffithdiablo_rojo: we'll see if your recheck does the trick19:24
diablo_rojojgriffith: yep :)19:24
* diablo_rojo crosses fingers19:24
eharneyxyang1: no...19:24
jgriffithdiablo_rojo: it's test_list_server_filters that fails, which is odd19:25
jgriffithdiablo_rojo: what i mean is it's in the server setup routines19:25
diablo_rojojgriffith: Yeah I thought so too. I know there was an issue a few days back with Ceph, but I dont recall the details19:25
eharneyxyang1: that verification could also be done in the manager19:26
*** apoorvad has joined #openstack-cinder19:26
openstackgerritWalter A. Boring IV (hemna) proposed openstack/cinder: Clean up image tmp file if c-vol gets restarted
smcginnisdiablo_rojo: The ceph job was switched to non-voting now.19:26
xyang1eharney: where in manager?19:27
diablo_rojosmcginnis: Cool. Hopefully it rechecks soon then.19:27
smcginnisdiablo_rojo: There is a bug in there and the version available on Ubuntu 14.04 is so old that it won't be patched.19:27
eharneyxyang1: around where the driver is loaded19:28
diablo_rojosmcginnis: So it will never be voting now?19:28
xyang1eharney: _init_?19:29
smcginnisdiablo_rojo: I'm actually not sure what the plan is there.19:29
smcginnisHopefully it's resolved somehow.19:29
smcginnisCeph is so widely used.19:29
smcginnisI don't really like that it's non-voting.19:29
smcginnisOther than that it now keeps us from having to constantly recheck.19:29
diablo_rojosmcginnis: Yeah I thought so. Makes me a tad nervous too.19:29
xyang1eharney: in manager init_host, it calls driver.check_for_setup_error19:30
*** ociuhandu has joined #openstack-cinder19:31
*** sdague has quit IRC19:32
*** Lee1092 has quit IRC19:32
*** merooney has quit IRC19:32
eharneyxyang1: not sure i understand the question... the point was it shouldn't be done in the driver's __init__19:33
openstackgerritMerged openstack/cinder: Fix issue with Pure drivers delete_snapshot exception handling
xyang1eharney: I agree19:33
*** timcl has joined #openstack-cinder19:33
xyang1eharney: right now check for setup error is an abstract method in BaseVD19:33
openstackgerritTom Barron proposed openstack/cinder: Run py34 tests with plain 'tox' command
xyang1eharney: so the question is which check for setup error shoukd this check be moved to19:34
*** merooney has joined #openstack-cinder19:34
eharneyxyang1: i don't think it's possible to move it to check_for_setup_error because it looks like most driver's don't call a super's version of that method anyway19:35
openstackgerritOpenStack Proposal Bot proposed openstack/cinder: Updated from global requirements
xyang1eharney: then where should it go?19:35
*** omolchanov1 has quit IRC19:35
*** omolchanov has joined #openstack-cinder19:35
eharneyxyang1: i don't know.19:36
eharneyxyang1: we're trying to add error checking of a sort we didn't have before, it seems19:36
xyang1eharney: get rid of the check?  is that better?19:37
eharneyxyang1: IMO this check shouldn't be necessary anyway, because <1.0 should work anyway and just reserve space19:39
eharneyxyang1: but i dunno19:39
eharneyi just didn't want it in driver __init__19:40
xyang1eharney: I am fine removing the check.  it is handled by the scheduler already19:40
mrrowdeneharney, I'm working on the patch that xyang1is talking about - we're already doing a warning in, is there a need for another warning in
jgriffithjungleboyj: please don't +A this patch
jgriffithjungleboyj: well, I guess you wouldn't actually... never mind19:41
*** martyturner has joined #openstack-cinder19:41
jungleboyjjgriffith: Wouldn't ...19:41
jgriffithjungleboyj: yeah, sorry... wasn't thinking when I asked that :)19:42
*** vilobhmm111 has joined #openstack-cinder19:42
openstackgerritTom Swanson proposed openstack/cinder: Dell SC: Incorrect values in REST API Login call
jungleboyjjgriffith: I removed my +2 to be safe.  :-)19:42
jgriffithjungleboyj: hehe :)19:42
jgriffithjungleboyj: wasn't sure if you'd agree with my comments there or not19:43
xyang1mrrowden: so remove the change in _init_ in  keep the  help string change.  keep the scheduler change19:43
mrrowdenxyang1, and remove second paragraph from commit message?19:43
xyang1mrrowden: yes19:44
eharneymrrowden: what does the check in the filter do?19:44
eharneymrrowden: just ignores it right?19:45
xyang1mrrowden: add a line in the commit msg saying you fixed the help string19:45
*** vilobhmm11 has quit IRC19:45
jungleboyjjgriffith: They look like good questions.19:45
xyang1eharney: if thin is true but ratio is invalid, the host will be skipped19:46
mc_nairI guess a question is - should a ratio < 1 be valid?19:47
mc_nairthink eharney and patrickeast have brought up19:47
xyang1mc_nair: no19:47
mrrowdeneharney, it gives a warning message about >= 1 then returns false19:47
eharneyultimately i think >0 should be valid, but we don't have to fix that right now19:47
patrickeasteharney: +119:48
eharneythere is no real reason to disallow setting 0.8 etc19:48
xyang1eharney: there is no real reason to support it either19:48
xyang1eharney: I have not heard a real use case for that19:48
eharneyxyang1: sure there is, you get more functionality via the same calculations19:48
eharneybut anyway19:49
xyang1eharney: I have not heard a real customer wanting that, unless if it is sfor some testing19:49
eharneyxyang1: then why do we have stuff like "reserved_percentage"?19:50
mc_naireharney: am I missing something or do we just need to remove the "host_state.max_over_subscription_ratio >= 1" part of the check to fix this for < 1 also?19:50
xyang1eharney: anyway, we don't support that earlier, so we are not losing anything19:50
xyang1eharney: reserved percentage has a specific parameter to achieve it19:51
eharneymc_nair: that should be the case ideally, but i'd need to look.  but then you need a check for >0 anyway, so that's probably for later19:52
mc_naireharney: Good point. Doesn't help with the question of *should* we, was just curious on that19:53
*** martyturner has quit IRC19:54
xyang1mrrowden: I added my comments in the patch19:54
patrickeastjungleboyj: hey, if you get a chance can you peek at again? lost your +2 with a rebase/merge conflict19:54
jungleboyjpatrickeast: Sure.19:55
patrickeastjungleboyj: sweet, thanks!19:55
eharneyi think just removing those pieces from is sensible19:55
mrrowdenxyang1, thanks, new patch coming up soon19:55
xyang1mrrowden: thanks19:55
openstackgerritMichael Rowden proposed openstack/cinder: Correcting thin provisioning behavior
*** msbrogli has joined #openstack-cinder19:59
mrrowdenthanks everyone for your input20:00
*** mrrowden has quit IRC20:00
msbrogliHey! I have one node running the VMs and one storage. But my storage has 2 IP addresses. The volumes are mounted using “the wrong IP”. I tried settings the config iscsi_ip_address but it has not worked. Can anyone help me?20:00
msbrogliI’ve set the iscsi_ip_address, restarted cinder-api cinder-scheduler cinder-volume. Then, I restarted the VM using the nova client. Is it enough?20:02
*** sheel has quit IRC20:07
*** timcl has quit IRC20:11
*** vilobhmm111 has quit IRC20:13
jungleboyjpatrickeast: Done.20:13
*** vilobhmm11 has joined #openstack-cinder20:13
*** merooney has quit IRC20:19
*** merooney has joined #openstack-cinder20:23
*** Suyash has quit IRC20:23
*** Suyash has joined #openstack-cinder20:24
*** diablo_rojo has quit IRC20:27
*** lpetrut has joined #openstack-cinder20:27
*** salv-orlando has quit IRC20:33
*** salv-orlando has joined #openstack-cinder20:33
*** timcl has joined #openstack-cinder20:35
*** mmandel has quit IRC20:38
*** ericksonsantos has quit IRC20:45
*** bardia has joined #openstack-cinder20:49
*** ivase has quit IRC20:50
openstackgerritMitsuhiro Tanino proposed openstack/python-cinderclient: Don't reset volume status when resetting migration status
openstackgerritJay Bryant proposed openstack/cinder: Add CGSnapshotStatus enum field
*** merooney has quit IRC20:55
openstackgerritMitsuhiro Tanino proposed openstack/python-cinderclient: Don't reset volume status when resetting migration status
*** delattec has quit IRC20:58
Swansonjgriffith, Any chance this can get a shove through the gate?20:58
mriedemare we gauranteed to have target_lun in connection_info['data'] for scsi attachments?20:59
smcginnismriedem: I'm not aware of any SCSI backends that do not set that.21:01
smcginnisPretty sure it's required.21:01
*** kenhui has joined #openstack-cinder21:02
*** kenhui1 has joined #openstack-cinder21:02
*** kenhui has quit IRC21:02
*** timcl has quit IRC21:02
hemnamriedem, I'm pretty sure that's required afaik21:02
hemnamriedem, some backends may always have a LUN of 021:03
hemnamriedem, but their IQN typically changes in that case21:03
mriedemwas just looking at
hemnayah target_lun should be set afaik21:04
*** rhagarty_ has quit IRC21:07
*** rhagarty_ has joined #openstack-cinder21:07
*** jgregor has quit IRC21:07
*** ivase_ has joined #openstack-cinder21:07
*** JoseMello has quit IRC21:08
*** laughterwym has joined #openstack-cinder21:09
*** apoorvad has quit IRC21:10
mriedemjwcroppe: do you have a fix for this else i'll push one up quick
openstackLaunchpad bug 1553319 in OpenStack Compute (nova) "When CPU metric collection fails, stack trace not in nova logs" [Undecided,New] - Assigned to Joe Cropper (jwcroppe)21:14
*** laughterwym has quit IRC21:15
*** ivase_ is now known as ivase21:16
*** raildo is now known as raildo-afk21:17
*** edtubill has quit IRC21:20
*** baumann has left #openstack-cinder21:33
*** akerr has quit IRC21:34
*** apoorvad has joined #openstack-cinder21:36
*** porrua has quit IRC21:39
*** jgregor has joined #openstack-cinder21:42
*** gouthamr has quit IRC21:43
msbrogliI changed the provider_location field in the volumes table. How to do I propagate the changes to the VMs?21:45
*** salv-orlando has quit IRC21:47
*** knikolla has quit IRC21:48
*** salv-orlando has joined #openstack-cinder21:48
*** mrrowden has joined #openstack-cinder21:50
*** salv-orlando has quit IRC21:51
*** kfarr has joined #openstack-cinder21:51
*** salv-orlando has joined #openstack-cinder21:52
mrrowdenxyang1 eharney can please I get your input on patrickeast comments for to see if another patch is in order?21:56
xyang1mrrowden: I am about to add my comments21:59
jgriffithhemna: ping22:05
jgriffithhemna: hey... your requests patch doesn't seem to be working for some reason22:06
jgriffithhemna: but it's weird... I can't get the config setting to work which I don't get22:06
hemnaare you getting errors reading the conf entry ?22:08
jgriffithlemme look22:08
hemnaI think the conf entry needs to be in the driver section ?22:08
jgriffithhemna: oh, because it's in the manager :)22:09
jgriffithhemna: that makes sense :(22:09
hemnathat way you can disable it for 1 driver, and not for others.22:09
hemnaI thought that was safer, than globally disabling it for all22:09
jgriffithhemna: yeah, that's it :)22:09
ntpttrsorry if I'm late to the party, but is something up with the gate? Looks like only 5 patches are in the 'check' stage and my patches aren't getting run
openstackgerritMerged openstack/cinder: Add backend id to Pure Volume Driver trace logs
*** dustins has quit IRC22:10
xyang1jgriffith: hemna when you two have time, can you take a look of this replication patch again?
hemnaxyang1, sure22:11
*** e0ne has joined #openstack-cinder22:11
*** laughterwym has joined #openstack-cinder22:13
hemnaxyang1, isn't backend_id a single entry ?
xyang1hemna: yes, it is fixed, right?22:15
*** jwcroppe has quit IRC22:16
hemnamsg = (_('Invalid secondary_backend_id specified. '22:16
hemna                             'Choose one entry from: %s.') % backend_id)22:16
hemnaI'm confused22:16
xyang1hemna: why?22:16
hemnathe wording says choose one entry from: backend_id22:16
jgriffithhemna: what's that coming from?22:16
xyang1hemna: oh, the text22:16
*** jwcroppe has joined #openstack-cinder22:16
hemnaif backend_id is a single entry.....22:16
xyang1hemna: got it22:16
jgriffithhemna: it's a list22:16
patrickeasthemna: each replication target has one, you specify one to failover too, so at times you are dealing with more than one22:17
hemnathat's a list ?22:17
hemnathat smells like a single entry to me...but maybe I'm confused.22:17
patrickeasti see22:18
hemnaline 3968 -> 397222:18
*** jgregor has quit IRC22:18
patrickeastyea that code is just looking at the first replication target22:18
jgriffithhemna: let me catch up here :)22:18
xyang1hemna: we only support 122:18
*** diablo_rojo has joined #openstack-cinder22:18
hemnamy question is, what is backend_id in that msg = (_( ... line ?22:18
*** jungleboyj has quit IRC22:18
hemnaI think it's a single entry, not a list22:18
hemnawhich makes the wording in the message confusing.22:19
jgriffithhemna: yeah22:19
hemnaok, phew.22:19
jgriffithhemna: i see what you're saying now22:19
xyang1hemna: the message is confusing22:19
xyang1hemna: he removed [ ], but did not change text22:19
xyang1hemna: are we on the same page?22:19
hemnaxyang1, ok thanks.  I thought I was nuts there for a second hehe22:20
hemnaxyang1, yup, I think so now. :)22:20
xyang1hemna: sorry, I did not pay close attention to the wording22:20
hemnanah it's cool :)22:20
hemnawell, -1 cool, but cool22:20
jgriffithhemna: xyang1 just don't forget that self.configuration.replication_device *can* be a list though22:21
xyang1jgriffith: sure, vnx only supports 1 device, is it ok22:21
hemnajgriffith, so at worst, they are hard coding looking for the 1st entry always then22:21
hemnain this case22:21
jgriffithhemna: yeah, and based on xyang1 's statement that they only support '1' that's cool22:22
jgriffithand I shouldn't say *can* I should say "it is" a list22:22
jgriffithbut never mind22:22
jgriffithI'm just talking out loud here ;)22:22
xyang1jgriffith: add your comments there if you have concerns22:22
jgriffithxyang1: no no, I don't have concerns22:23
*** gouthamr has joined #openstack-cinder22:23
jgriffithxyang1: I was just pointing out that even though your backend only supports one, that config option IS a list so using [0] is appropriate22:23
jgriffithxyang1: but that's aside from the comment about the wording :)22:23
xyang1jgriffith: ok:)22:23
*** akshai has quit IRC22:25
*** tpsilva has quit IRC22:27
*** lascii is now known as alaski22:28
openstackgerritOpenStack Proposal Bot proposed openstack/cinder: Updated from global requirements
*** [1]Thelo has joined #openstack-cinder22:29
*** kfarr has quit IRC22:31
*** Thelo has quit IRC22:31
*** [1]Thelo is now known as Thelo22:31
*** rlrossit_ has quit IRC22:34
*** arch-nemesis has quit IRC22:34
*** chris_morrell has quit IRC22:37
*** ircuser-1 has quit IRC22:39
openstackgerritTom Swanson proposed openstack/cinder: Dell SC: Incorrect values in REST API Login call
*** ircuser-1 has joined #openstack-cinder22:43
*** laughterwym has quit IRC22:43
*** laughterwym has joined #openstack-cinder22:44
*** sheel has joined #openstack-cinder22:45
*** mriedem has quit IRC22:46
*** erlon has quit IRC22:46
*** laughterwym has quit IRC22:48
*** arch-nemesis has joined #openstack-cinder22:51
*** e0ne has quit IRC22:55
*** msbrogli has quit IRC22:55
patrickeasthey so for the replication v2.1 implementations in drivers and release notes... how do we want to handle it?22:55
patrickeasttheres a mix of strategies right now, which might make for confusing release notes22:56
patrickeastsome drivers have ones that say implement v2 and implement v2.122:56
patrickeastsome (like mine) only have implement v222:56
patrickeastand i think the dell one removed the v2 and only have a note for v2.122:56
jgriffithpatrickeast: update your to say "cheesecake" :)22:56
patrickeastjgriffith: of course!22:56
jgriffithpatrickeast: my opinion is that it should be 2.122:56
jgriffithnot 222:56
*** haplo37 has quit IRC22:57
patrickeastjgriffith: yea im kinda thinking any of the leftover v2 release notes should be removed22:57
jgriffithpatrickeast: I would agree for sure22:57
patrickeastSwanson: smcginnis: aorourke: ^^22:57
patrickeasti was going to put up a patch to fix the pure one, but can fixup the others all in one go22:58
Swansonpatrickeast, v2.122:58
patrickeastany the in-flight ones can be fixed in review22:58
patrickeastSwanson: cool, yea so far thats the favorite22:58
Swansoncause I don't support v2 anymore and I think it is important to make sure we make that distinction.22:58
*** ivase has quit IRC23:00
aorourkeoops. I added a release note for supporting v2.1 but did not remove the old ones23:00
patrickeastaorourke: yea thats what i did23:00
patrickeastaorourke: i can remove it with mine if you want23:00
aorourkepatrickeast, that works for me23:01
Swansonv2 didn't comeout until early liberty, right?  So no release ever supported v2, right?23:01
aorourkepatrickeast, the LeftHand and 3PAR ones need to be removed23:01
patrickeastSwanson: yep, it wasn't ever really supported officially by anyone afaik23:01
patrickeastput a note on the reviews that are up now (I think I got them all..)23:03
*** lpetrut has quit IRC23:03
*** edmondsw has quit IRC23:05
openstackgerritPatrick East proposed openstack/cinder: Fixup release notes for v2 -> v2.1 replication impls
patrickeastaorourke: ^23:06
openstackgerritPatrick East proposed openstack/cinder: Fixup release notes for v2 -> v2.1 replication impls
patrickeastwhew, almost forgot to change it to say cheesecake23:12
*** kenhui1 has quit IRC23:13
*** e0ne has joined #openstack-cinder23:15
xyang1patrickeast: why remove release note?23:17
patrickeastxyang1: the driver wont support v2 anymore23:17
xyang1patrickeast: driver has a new feature, still need a release note23:17
patrickeastxyang1: so a release note saying support for v2 is kinda confusing, right?23:17
patrickeastxyang1: it would be supporting v2.123:18
xyang1patrickeast: that is what it says23:18
xyang1patrickeast: ?23:18
patrickeastxyang1: right so says the driver added support for v2.1 in mitaka, and is going to say it also suports v223:19
patrickeastxyang1: for the mitaka release notes23:19
patrickeastxyang1: so keep the new one, or modify the original23:19
patrickeastxyang1: does that make sense?23:19
xyang1patrickeast: oh, I see23:20
xyang1patrickeast: we should remove the old one23:20
xyang1patrickeast: the name is wrong too23:20
patrickeastxyang1: cool, sounds good23:21
xyang1patrickeast: thanks23:21
*** mrrowden has quit IRC23:23
*** msbrogli has joined #openstack-cinder23:26
*** xyang1 has quit IRC23:29
*** Trident has quit IRC23:33
*** msbrogli has quit IRC23:35
openstackgerritVasanthi Thirumalai proposed openstack/cinder: Violin Memory ISCSI storage for 7000 series AFA
*** msbrogli has joined #openstack-cinder23:38
*** daneyon has joined #openstack-cinder23:49
*** diablo_rojo has quit IRC23:51
*** daneyon_ has joined #openstack-cinder23:52
*** salv-orl_ has joined #openstack-cinder23:52
*** daneyon has quit IRC23:55
*** salv-orlando has quit IRC23:55

Generated by 2.14.0 by Marius Gedminas - find it at!