Wednesday, 2015-08-26

*** smoriya has joined #openstack-cinder00:05
*** dims_ has joined #openstack-cinder00:06
*** dims has quit IRC00:07
*** 64MADOR5X has quit IRC00:09
openstackgerritCory Stone proposed openstack/cinder: Dont eager load volume type specs on volume list
*** haomaiwang has joined #openstack-cinder00:10
*** simondodsley has joined #openstack-cinder00:11
guitarzanjgriffith: hey00:12
*** dims has joined #openstack-cinder00:12
guitarzanjgriffith: I just did some weird extra specs patch...00:12
*** IanGovett has quit IRC00:13
guitarzanoh yours is different :)00:13
*** asd112z has joined #openstack-cinder00:14
*** tsekiyam_ has joined #openstack-cinder00:15
*** dims_ has quit IRC00:15
*** garthb has quit IRC00:17
*** mtanino has quit IRC00:18
*** dflorea has quit IRC00:18
*** tsekiyama has quit IRC00:19
*** tsekiyam_ has quit IRC00:20
*** diogogmt has joined #openstack-cinder00:23
*** takedakn has left #openstack-cinder00:23
*** salv-orlando has quit IRC00:30
*** openstack has joined #openstack-cinder00:34 [freenode-info] channel trolls and no channel staff around to help? please check with freenode support:
*** _cjones_ has quit IRC00:34
*** dflorea has joined #openstack-cinder00:36
guitarzanextra specs...what a hassle00:37
openstackgerritwanghao proposed openstack/cinder: Add pagination to backups
*** diogogmt has quit IRC00:39
*** diogogmt has joined #openstack-cinder00:40
*** dflorea has quit IRC00:41
*** changbl has quit IRC00:42
*** Zhang has joined #openstack-cinder00:42
*** jwcroppe has quit IRC00:47
*** jwcroppe has joined #openstack-cinder00:47
*** pckizer has quit IRC00:49
*** pckizer has joined #openstack-cinder00:49
*** jwcroppe has quit IRC00:52
*** dflorea has joined #openstack-cinder00:56
*** asselin__ has joined #openstack-cinder00:56
*** asselin has quit IRC00:58
*** lixiaoy1 has joined #openstack-cinder01:04
*** lixiaoy1 has quit IRC01:05
*** haomaiwang has quit IRC01:09
*** haomaiwang has joined #openstack-cinder01:10
*** IlyaG has joined #openstack-cinder01:12
openstackgerritwanghao proposed openstack/cinder: Incremental backup improvements for L
*** skylerberg has quit IRC01:16
*** zhenguo has joined #openstack-cinder01:17
*** jwcroppe has joined #openstack-cinder01:19
*** theanalyst has quit IRC01:19
openstackgerritxing-yang proposed openstack/cinder: ScaleIO driver should use os-brick connector
*** lixiaoy1 has joined #openstack-cinder01:21
*** theanalyst has joined #openstack-cinder01:22
*** humble__ has joined #openstack-cinder01:23
*** asselin has joined #openstack-cinder01:24
*** dflorea has quit IRC01:24
*** bnemec has quit IRC01:26
*** asselin__ has quit IRC01:27
*** lixiaoy11 has joined #openstack-cinder01:28
*** lixiaoy1 has quit IRC01:30
openstackgerritxing-yang proposed openstack/cinder: ScaleIO driver should use os-brick connector
*** zhongjun has joined #openstack-cinder01:33
*** Apoorva has quit IRC01:33
*** Lee1092 has joined #openstack-cinder01:33
*** edmondsw has joined #openstack-cinder01:39
*** edmondsw has quit IRC01:39
*** jwcroppe_ has joined #openstack-cinder01:40
*** jwcroppe has quit IRC01:42
openstackgerritxing-yang proposed openstack/cinder: ScaleIO driver should use os-brick connector
*** davechen has joined #openstack-cinder01:51
*** gouthamr has joined #openstack-cinder01:54
*** julim has quit IRC01:57
Zhangzhenguo Hi02:04
*** haomaiwang has quit IRC02:09
*** haomaiwang has joined #openstack-cinder02:10
*** skylerberg has joined #openstack-cinder02:14
openstackgerritwanghao proposed openstack/cinder: Add pagination to backups
*** rushil has joined #openstack-cinder02:19
*** boris-42 has quit IRC02:20
*** IlyaG has quit IRC02:22
*** rushil has quit IRC02:23
*** rushil has joined #openstack-cinder02:24
ZhangI have some problem when Integrated 18000 storage.02:24
*** zhongjun has quit IRC02:26
*** chenying has joined #openstack-cinder02:32
*** annegent_ has quit IRC02:32
zhenguoZhang: you mean huawei driver?02:38
*** gouthamr_ has joined #openstack-cinder02:41
*** mtanino has joined #openstack-cinder02:41
*** annegentle has joined #openstack-cinder02:42
openstackgerritVincent Hou proposed openstack/cinder: WIP: Update the devref for volume migration
*** gouthamr has quit IRC02:44
zhenguoliuxg: around?02:46
*** dims has quit IRC02:46
*** gouthamr_ has quit IRC02:51
*** hakimo has joined #openstack-cinder02:52
*** hakimo_ has quit IRC02:54
*** haomaiwang has quit IRC03:09
*** haomaiwang has joined #openstack-cinder03:10
openstackgerritxing-yang proposed openstack/cinder: ScaleIO driver should use os-brick connector
*** skylerberg has quit IRC03:15
*** BharatK has quit IRC03:19
openstackgerritJuPing proposed openstack/cinder: Fix description for "Barbarism of editting a file"
*** garthb has joined #openstack-cinder03:31
*** jwcroppe_ is now known as jwcroppe03:35
*** baojg has joined #openstack-cinder03:39
*** haomaiwang has quit IRC03:40
*** hemna has joined #openstack-cinder03:41
*** hemna has quit IRC03:42
*** links has joined #openstack-cinder03:45
*** boris-42 has joined #openstack-cinder03:45
*** changbl has joined #openstack-cinder03:46
*** baojg has quit IRC03:52
*** rushil has quit IRC03:53
*** baojg has joined #openstack-cinder03:53
*** p0rtal has quit IRC03:54
*** dflorea has joined #openstack-cinder04:03
openstackgerritwanghao proposed openstack/python-cinderclient: Add to_dict method to Resource class
*** haomaiwang has joined #openstack-cinder04:05
*** annegentle has quit IRC04:07
*** haomaiwang has quit IRC04:09
*** haomaiwang has joined #openstack-cinder04:10
*** hemna has joined #openstack-cinder04:13
openstackgerritwanghao proposed openstack/cinder: Implement function to manage/unmanage snapshots
*** skylerberg has joined #openstack-cinder04:26
*** chenying has quit IRC04:27
*** humble__ has quit IRC04:28
*** sgotliv has joined #openstack-cinder04:35
*** deepakcs has joined #openstack-cinder04:35
openstackgerritTakeaki Matsumoto proposed openstack/cinder: Fix url in API response to get original
*** timcl has quit IRC05:04
*** lpetrut has joined #openstack-cinder05:06
*** rbradbury has quit IRC05:07
*** haomaiwang has quit IRC05:09
*** haomaiwang has joined #openstack-cinder05:10
lixiaoy11winston-d: here?05:12
*** hemna has quit IRC05:13
openstackgerritOpenStack Proposal Bot proposed openstack/cinder: Updated from global requirements
*** salv-orlando has joined #openstack-cinder05:16
*** zhongjun has joined #openstack-cinder05:16
openstackgerritTakeaki Matsumoto proposed openstack/cinder: Fix url in API response to get original
*** salv-orlando has quit IRC05:21
*** sgotliv has quit IRC05:21
*** RaySun has joined #openstack-cinder05:21
*** asd112z has quit IRC05:35
*** p0rtal has joined #openstack-cinder05:36
openstackgerritTakeaki Matsumoto proposed openstack/cinder: Fix url in API response to get original
*** dims has joined #openstack-cinder05:36
*** yrabl has quit IRC05:39
*** shausy has joined #openstack-cinder05:41
*** dims has quit IRC05:42
*** lpetrut has quit IRC05:43
*** shyama has joined #openstack-cinder05:46
openstackgerritTakeaki Matsumoto proposed openstack/cinder: Fix url in API response to get original
*** zhongjun has quit IRC05:50
*** garthb has quit IRC05:52
*** shausy has quit IRC05:53
*** shausy has joined #openstack-cinder05:54
openstackgerritTakeaki Matsumoto proposed openstack/cinder: Fix url in API response to get original
*** afazekas has joined #openstack-cinder06:08
*** haomaiwang has quit IRC06:09
*** lpetrut has joined #openstack-cinder06:09
*** haomaiwa_ has joined #openstack-cinder06:10
*** nkrinner has joined #openstack-cinder06:15
openstackgerritwanghao proposed openstack/cinder: Implement function to manage/unmanage snapshots
openstackgerritTeruaki Ishizaki proposed openstack/cinder: Sheepdog: Add class for dog command executor
*** RaySun has quit IRC06:20
*** chenying has joined #openstack-cinder06:24
openstackgerritWilson Liu proposed openstack/cinder: Add hypermetro support for Huawei driver
*** lpetrut has quit IRC06:27
*** p0rtal has quit IRC06:27
*** p0rtal has joined #openstack-cinder06:28
*** skylerberg has quit IRC06:28
wilson2Zhang: what's the problem with 18000 storage?06:29
openstackgerritwanghao proposed openstack/cinder: Incremental backup improvements for L
Zhangcreate volume size 1Gb, but lookup the size 5Gb when attach volume.06:31
*** p0rtal has quit IRC06:32
openstackgerritwanghao proposed openstack/cinder: Incremental backup improvements for L
openstackgerritOpenStack Proposal Bot proposed openstack/cinder: Imported Translations from Transifex
*** agarciam has joined #openstack-cinder06:37
*** salv-orlando has joined #openstack-cinder06:38
*** Zhang has quit IRC06:40
*** bluex-pl has joined #openstack-cinder06:43
*** bluex-pl has quit IRC06:43
*** vincent_hou has joined #openstack-cinder06:46
geguileojgriffith: I am now06:46
vincent_hougeguileo: Morning.06:46
geguileovincent_hou: Morning06:46
vincent_hougeguileo: Thank you for your review.06:47
*** ociuhandu has quit IRC06:47
geguileovincent_hou: No problem06:47
vincent_hougeguileo: I still have some minor questions for you. Is it OK?06:48
geguileovincent_hou: Sure06:48
geguileovincent_hou: Did you write them in the comments or are you going to ask them here?06:48
vincent_hougeguileo: I am putting some of them there, but there are something else here.06:49
geguileovincent_hou: Ok06:49
vincent_hougeguileo: why is if a in ('a', 'b') better than if a in ['a', 'b']?06:49
geguileovincent_hou: [] is a list, which is a "costly" operation06:50
geguileovincent_hou: () is a immutable tuple, which is cheaper06:50
geguileovincent_hou: And a dictionary is more expensive as well06:50
geguileovincent_hou: Because it has to build the hash table06:50
vincent_hougeguileo: [] takes more space and time.06:50
geguileovincent_hou: If you are not going to be modifying the items you should use a tuple06:51
geguileovincent_hou: Exactly06:51
geguileovincent_hou: And you can modify it by "mistake"06:51
*** ronis has joined #openstack-cinder06:51
geguileovincent_hou: So [] should only be used when you are going to be adding/removing items06:51
*** e0ne has joined #openstack-cinder06:52
vincent_hougeguileo: Wow, you did awesome in python. Thank you for letting me know.06:53
geguileovincent_hou: No problem06:53
vincent_hougeguileo: Then you said webob.exc.HTTPServiceUnavailable is not correct, what could be the right one for the exception raised for 'maintenance' volumes?06:53
openstackgerritVilobh Meshram proposed openstack/cinder: Cinder Nested Quota Driver
*** ronis_ has joined #openstack-cinder06:53
geguileovincent_hou: I would send 409 Conflict06:56
*** ronis has quit IRC06:57
vincent_hougeguileo: HTTPConflict?06:58
geguileovincent_hou: That's what I would return06:58
vincent_hougeguileo: I will put them in another patch .06:58
geguileovincent_hou: According to the RFC: The 409 (Conflict) status code indicates that the request could not06:59
geguileo   be completed due to a conflict with the current state of the target06:59
geguileo   resource.  This code is used in situations where the user might be06:59
geguileo   able to resolve the conflict and resubmit the request.06:59
*** ociuhandu has joined #openstack-cinder06:59
vincent_houHmm, it sounds more reasonable.06:59
geguileovincent_hou: But that error only for the maintenance status error06:59
geguileovincent_hou: Not for all the errors where the input is not valid07:00
vincent_houI see.07:00
vincent_hougeguileo: Thanks.07:00
geguileovincent_hou: No problem07:01
*** ankit_ag has joined #openstack-cinder07:01
*** yrabl has joined #openstack-cinder07:03
*** alexschm has joined #openstack-cinder07:04
openstackgerritXinXiaohui proposed openstack/cinder: Calculate virtual free capacity and notify
*** dflorea has quit IRC07:07
*** baojg has quit IRC07:08
*** haomaiwa_ has quit IRC07:09
*** Zhang has joined #openstack-cinder07:09
*** haomaiwang has joined #openstack-cinder07:10
*** baojg has joined #openstack-cinder07:10
*** yrabl has quit IRC07:15
*** yrabl has joined #openstack-cinder07:15
*** Zhang has quit IRC07:16
geguileoe0ne: ping07:16
e0negeguileo: hi07:19
geguileoe0ne: Hi, thanks for your review on the snapshot pagination07:19
geguileoe0ne: Any reason why you didn't +1 the workflow?07:20
e0negeguileo: you're welcome! thanks for addressing my comments07:20
e0negeguileo: only because it was too late (about midnight)07:20
e0negeguileo: i didn't want to miss anything:)07:20
geguileoe0ne: Thank you for the comment, I actually missed that method  :-)07:20
geguileoe0ne: That's usually a good idea at those hours  XD07:21
e0negeguileo: yep, i did some mistakes in the past, so i try to put only +2 late07:21
e0negeguileo: i'll be at the office in ~2 hours and +A your patch if it is still needed07:22
geguileoe0ne: Smart decission, I will take note  XD07:22
geguileoe0ne: Awesome!07:22
geguileoe0ne: Is it in your review list the patch that fixes the backup imports?07:22
e0negeguileo: it's on a top of my list07:23
geguileoe0ne: Awesome!!07:23
geguileoe0ne: Thanks07:23
e0negeguileo: yesterday was my first day after vacation, i've got very long review and todos lists07:24
*** Zhang has joined #openstack-cinder07:24
geguileoe0ne: I hope you enjoyed your vacation, and good luck with your backlog07:24
*** dims has joined #openstack-cinder07:24
*** dims has quit IRC07:29
openstackgerritVipin Balachandran proposed openstack/cinder: Add unit test for backup get_all
*** rushiagr_away is now known as rushiagr07:31
*** e0ne has quit IRC07:45
openstackgerritVincent Hou proposed openstack/cinder: Volume status management during migration
*** ronis has joined #openstack-cinder07:52
*** ronis_ has quit IRC07:54
openstackgerritVincent Hou proposed openstack/cinder: Volume status management during migration
openstackgerritLisaLi proposed openstack/cinder: Raise HTTPBadRequest once invalid filters are given
openstackgerritLisaLi proposed openstack/cinder: Raise HTTPBadRequest when invalid filters are given
openstackgerritLisaLi proposed openstack/cinder: Raise HTTPBadRequest when invalid filters are given
*** IanGovett has joined #openstack-cinder07:56
openstackgerritVincent Hou proposed openstack/cinder: Volume status management during migration
openstackgerritTeruaki Ishizaki proposed openstack/cinder: Sheepdog: improve create and delete operation
*** ndipanov has quit IRC08:05
*** ndipanov has joined #openstack-cinder08:06
*** ndipanov has quit IRC08:06
*** chenying has quit IRC08:09
*** haomaiwang has quit IRC08:09
*** haomaiwang has joined #openstack-cinder08:10
*** jwcroppe has quit IRC08:16
*** geguileo has quit IRC08:16
*** geguileo has joined #openstack-cinder08:17
*** chenying has joined #openstack-cinder08:20
*** bluex-pl has joined #openstack-cinder08:24
*** bluex-pl has quit IRC08:24
*** bluex-pl has joined #openstack-cinder08:24
*** anshul has joined #openstack-cinder08:25
*** stefan_amann has joined #openstack-cinder08:26
*** yuriy_n17 has joined #openstack-cinder08:28
yuriy_n17geguileo: Please review
stefan_amannthingee: Mike, may I ask you for your help with bug #1488233? System z / s390 will not be able to attach volumes when the FC LUN ID is >255. I proposed a fix for os-brick for Liberty. We need to back port to Kilo (Cinder, Nova, and os-brick since os-brick is not integrated in Kilo). But there is no option to nominate it for Kilo. At least not for Cinder, and os-brick. Any help would be appreciated!08:31
openstackbug 1488233 in os-brick "FC with LUN ID >255 not recognized" [Undecided,In progress] - Assigned to Stefan Amann (stefan-amann)08:31
openstackgerritHuang Zhiteng proposed openstack/os-brick: iSCSI Connector: force disconnect iscsi session
*** jistr has joined #openstack-cinder08:39
geguileoyuriy_n17: Done08:41
*** Zhang has quit IRC08:43
*** jordanP has joined #openstack-cinder08:49
*** lixiaoy11 has quit IRC08:56
yuriy_n17geguileo: Thanks a lot.08:57
yuriy_n17geguileo: What about to please review Thanks for advance.09:00
*** enterprisedc has joined #openstack-cinder09:02
yuriy_n17jgriffith: Hi! Please review
openstackgerritHuang Zhiteng proposed openstack/cinder: Update volume status AFTER terminate_connection is done
*** e0ne has joined #openstack-cinder09:08
*** haomaiwang has quit IRC09:09
openstackgerritlifeless proposed openstack/cinder: Remove unused dependency discover.
*** enterprisedc has quit IRC09:09
*** haomaiwang has joined #openstack-cinder09:10
*** Zhang has joined #openstack-cinder09:12
*** h00327910__ has joined #openstack-cinder09:14
*** Zhang has quit IRC09:14
openstackgerritJordan Pittier proposed openstack/cinder: Rework Scality SOFS driver to use RemoteFS class
*** agarciam has quit IRC09:16
openstackgerritWilson Liu proposed openstack/cinder: Add hypermetro support for Huawei driver
*** agarciam has joined #openstack-cinder09:16
*** vincent_hou has quit IRC09:17
*** e0ne has quit IRC09:18
*** agarciam has quit IRC09:19
*** Zhang has joined #openstack-cinder09:19
*** Zhang has quit IRC09:22
*** e0ne has joined #openstack-cinder09:23
*** wilson1 has joined #openstack-cinder09:24
*** agarciam has joined #openstack-cinder09:24
*** daemontool_ has joined #openstack-cinder09:25
*** ndipanov has joined #openstack-cinder09:26
openstackgerritVincent Hou proposed openstack/cinder: Volume status management during migration
*** gnarld_ has joined #openstack-cinder09:27
*** jgriffit1 has joined #openstack-cinder09:28
*** nug has quit IRC09:28
*** wilson2 has quit IRC09:28
*** geguileo has quit IRC09:28
*** rushiagr has quit IRC09:28
*** jgriffith has quit IRC09:28
*** gnarld_ is now known as nug09:28
*** daemontool__ has quit IRC09:28
*** geguileo has joined #openstack-cinder09:29
*** zhongjun has joined #openstack-cinder09:33
*** goodygum has joined #openstack-cinder09:38
openstackgerritzhaohua proposed openstack/cinder: Huawei: fix multi REST-URLs bug
*** rushiagr has joined #openstack-cinder09:47
*** dims has joined #openstack-cinder09:48
*** davechen has left #openstack-cinder09:50
e0negeguileo: are you around?09:52
geguileoe0ne: Yep09:52
e0negeguileo: few questions about import backup patches09:53
geguileoe0ne: I hope I still remember what it does  ;-)09:53
e0neit fixe a bug, i hope09:53
geguileoe0ne: That was the idea, yes   XD09:54
e0negeguileo: it wil never be merged until depended on patch get -1 from jenkins09:55
geguileoe0ne: We have a cyclic dependency09:55
geguileoe0ne: Tests in tempest are "wrong", since they assume current behavior is desired one09:56
e0negeguileo: oh...09:56
geguileoe0ne: And I have a tempest patch to fix that09:56
geguileoe0ne: But with current code it doesn't pass either09:56
e0negeguileo: why we've got a cyclic deps there>09:56
geguileoe0ne: Because Cinder patch needs tempest patch and viceversa09:56
geguileoe0ne: Cyclic dependency and Zuul doesn't like that09:57
e0negeguileo: can we try to add dependency inly for cinder patch?09:57
e0negeguileo: afair, tests for import backup were disabled09:57
geguileoe0ne: Not anymore09:58
geguileoe0ne: Cinder patch has dependency in the commit message09:58
e0negeguileo: we need to ask infra guys how to fix it09:59
e0negeguileo: i can't +2 w/o jenkins vote:(10:00
geguileoe0ne: I'll get Jenkins to give it -1 (with the error I mention)10:01
*** agarciam has quit IRC10:01
e0negeguileo: but we can't merge with -1 from jenkins10:01
*** agarciam has joined #openstack-cinder10:01
geguileoe0ne: Yeah, we need to disable that test first in tempest10:02
geguileoe0ne: Then merge this code in Cinder10:02
geguileoe0ne: And then update the test and re-enable it10:02
e0negeguileo: looks like a good plan to get it fixed and merged10:02
geguileoe0ne: Ok, I'll look into disabling that test10:02
e0negeguileo: thanks! i'll be happy to +2 once jenkins passed10:03
geguileoe0ne: Ok, I'll ping you again when I have it sorted out and passing Jenkins10:03
e0negeguileo: thanks!10:03
*** baojg has quit IRC10:06
*** ociuhandu_ has joined #openstack-cinder10:06
*** ociuhandu has quit IRC10:08
*** ociuhandu_ is now known as ociuhandu10:08
*** haomaiwang has quit IRC10:09
*** shausy has quit IRC10:10
*** haomaiwang has joined #openstack-cinder10:10
*** Zhang has joined #openstack-cinder10:16
*** zhongjun has quit IRC10:24
*** vgridnev has joined #openstack-cinder10:26
vgridnevis there someone who can review that  change: ?10:28
yuriy_n17smcginnis: Hi! Please review
*** chenying has quit IRC10:34
openstackgerritAnton Arefiev proposed openstack/cinder: WIP:Add rootwrap daemon mode support
openstackgerritAnton Arefiev proposed openstack/cinder: WIP: Switch to use execute from cinder utils
openstackgerritAnton Arefiev proposed openstack/cinder: Add rootwrap daemon mode support
openstackgerritAnton Arefiev proposed openstack/cinder: Switch to use execute from cinder utils
*** lpetrut has joined #openstack-cinder10:41
*** wilson2 has joined #openstack-cinder10:48
openstackgerritAnton Arefiev proposed openstack/cinder: Add rootwrap daemon mode support
*** wilson1 has quit IRC10:52
openstackgerritGorka Eguileor proposed openstack/cinder: Fix backup metadata import missing fields
aarefievDuncanT: hi, are u around ?10:58
openstackgerritVipin Balachandran proposed openstack/cinder: Replace urllib.unquote with urllib.parse.unquote
dulekGuys, I'm working on a bugfix and made changes in db.sqlalchemy.api. Where should I put unit tests?11:00
dulekI'm changing snapshot_get_all methods, so doesn't seem like a proper place - it's testing volume manager mostly.11:01
dulekAnd tests.unit.db also doesn't hold such tests...11:02
aarefievdulek: tests.unit.test_db_api ?11:02
dulekaarefiev: Ah, that should be it. Thanks!11:03
aarefievdulek: np11:05
*** haomaiwang has quit IRC11:09
*** agarciam1 has joined #openstack-cinder11:10
*** agarciam has quit IRC11:10
*** EinstCrazy has joined #openstack-cinder11:21
openstackgerritMitsuhiro Tanino proposed openstack/cinder: Efficient volume copy for generic volume migration
mtaninoDo we still have pthon27 error for VMwareVcVmdkDriverTestCase on the CI?11:42
*** Zhang has quit IRC11:42
*** EinstCrazy has quit IRC11:43
mtaninogeguileo: Hi, I added Get Volume Driver Capabilities patch (mtanino) on meeting topics.11:43
geguileomtanino: I saw it :-)11:44
mtaninoIf my understanding is different from your proposal, please let me know.11:44
*** dims_ has joined #openstack-cinder11:48
openstackgerritWilson Liu proposed openstack/cinder: Fix way of get LUN id in Huawei driver
*** bswartz has quit IRC11:51
*** mewald has joined #openstack-cinder11:52
openstackgerritWilson Liu proposed openstack/cinder: Encode username and password in Huawei driver
*** dims has quit IRC11:52
*** h00327910__ has quit IRC11:53
DuncanTaarefiev: Hi. I'm around right now11:56
*** agarciam1 has quit IRC11:56
aarefievDuncanT: hi, just want to know your thoughts about rootwrap daemon mode testing11:58
*** agarciam has joined #openstack-cinder11:59
aarefievDuncanT: and specifically about the sheepdog backup11:59
*** marcusvrn_ has joined #openstack-cinder12:01
aarefievDuncanT: it should be separate job or what?12:02
openstackgerritWilson Liu proposed openstack/cinder: Fix way of get LUN id in Huawei driver
*** mewald has left #openstack-cinder12:07
DuncanTaarefiev: A separate job, at least initially, will help narrow down the problems faster12:08
DuncanTaarefiev: At some point, it can probably be rolled into one, since non-daemon mode is tested by lots of jobs12:08
*** Zhang has joined #openstack-cinder12:08
*** mewald_ has joined #openstack-cinder12:09
*** EinstCrazy has joined #openstack-cinder12:10
*** simondodsley has quit IRC12:10
aarefievDuncanT : yep, we are worrying about the number of jobs)12:10
aarefievthey are a lot12:10
mewald_I need help understanding how cinder-backup works: Will cinder-backup connect to the volume backends and backup backends and take care of copying? In other words: Will data flow through the cinder-backend service?12:11
*** raildo-afk is now known as raildo12:14
*** bswartz has joined #openstack-cinder12:14
dulekmewald_: Here is extensive documentation of current state of how backups work and the state we want to achieve in Liberty.12:18
dulekmewald_: Ah, sorry, wrong link.12:18
dulekThat's the right one.12:18
*** mewald_ has quit IRC12:19
*** nikeshm has quit IRC12:19
*** nikeshm has joined #openstack-cinder12:19
*** erlon has joined #openstack-cinder12:23
openstackgerritMichal Dulko proposed openstack/cinder: Validate filters to snapshot_get_all* in db.api
DuncanTaarefiev: Infra will have opinions on that I guess. Combining them is ok, just makes debugging slightly harder12:27
*** annegentle has joined #openstack-cinder12:27
*** adrianofr has joined #openstack-cinder12:28
aarefievDuncanT: what do you mean  'Combining them...' ?12:28
*** smoriya has quit IRC12:28
*** aix has quit IRC12:29
openstackgerritMichal Dulko proposed openstack/cinder: Validate filters in snapshot_get_all* in db.api
*** timcl has joined #openstack-cinder12:32
DuncanTaarefiev: Just test everything in one job12:35
*** akerr has joined #openstack-cinder12:35
DuncanTaarefiev: as long as daemon mode is tested in one job somewhere, and non-daemon mode somewhere else, and every driver gets tested, we still get reasonable coverage12:36
DuncanTaarefiev: A vendor with interest in a specific scenario is welcome to add more CI on their own hardware12:37
aarefievDuncanT: I'm not quite understand about 'test  everything in one job', how do you see it ?12:38
aarefievwhat job will run12:39
*** edmondsw has joined #openstack-cinder12:40
yuriy_n17smcginnis: Thanks a lot.12:43
smcginnisyuriy_n17: no12:43
yuriy_n17smcginnis: I mean thanks for your reviewing.12:44
smcginnisyuriy_n17: Hah! I meant to type np. No problem reviewing. Looked good to me. :)12:44
*** EinstCrazy has quit IRC12:46
*** EinstCrazy has joined #openstack-cinder12:49
*** annasort has joined #openstack-cinder12:50
DuncanTaarefiev: Combine daemon mode into e.g. the sheepdog job12:50
*** simondodsley has joined #openstack-cinder12:50
aarefievDuncanT: ahh , I see :) , sorry, you mean  daemon mode and sheepdog12:51
DuncanTaarefiev: In general, we can probably test more config options by mixing up the jobs a bit12:52
aarefievDuncanT: I think it's good idea, the number of jobs have been increasing very fast12:54
aarefievDuncanT: thanks for your time12:55
*** deepakcs has quit IRC12:56
*** aix has joined #openstack-cinder12:58
*** mtanino has quit IRC12:59
*** diablo_rojo has joined #openstack-cinder13:00
*** eharney has joined #openstack-cinder13:00
*** diablo_rojo has left #openstack-cinder13:00
*** jwcroppe has joined #openstack-cinder13:00
*** mriedem has joined #openstack-cinder13:03
*** takedakn has joined #openstack-cinder13:04
jordanPsmcginnis, hi :)13:04
*** Zhang has quit IRC13:04
smcginnisjordanP: Hey. Sorry I didn't catch that earlier.13:05
*** merooney has joined #openstack-cinder13:05
jordanPsmcginnis, no problem.13:05
*** ericksonsantos has quit IRC13:05
jordanPhere's the thing. You can see the logs of the job if you click on the "console" link in the left panel of the jenkins screen. All the logs are available if you click on the "build artifacats" link. I am using gerrit trigger for now. I am not sure how customizable that is13:07
*** porrua has joined #openstack-cinder13:07
*** xyang1 has joined #openstack-cinder13:07
jordanPsmcginnis, I agree it's less convenient, you have to click on 2 links but it was like this since the start13:07
smcginnisjordanP: I don't use Jenkins myself, but most third party CIs do. You need to collect all of those artifacts and publish them on a log server. Not just make your jenkins server accessible.13:08
smcginnisjordanP: Which really doesn't seem like a good idea to me anyway. :)13:08
jordanP( is public)13:08
jordanPsmcginnis, I agree but then I have another log server to maintain13:09
smcginnisWelcome to third party CI> :)13:09
eikkesmcginnis: that jenkins instance is meant to be public, consciously13:09
smcginniseikke: That's fine. That's not really an issue.13:09
smcginnisJust commenting that I don't have enough confidence in my own admin skills that I would feel comfortable doing that.13:10
smcginnisBut the issue is the logs are not collected as required.13:10
openstackgerritAnton Arefiev proposed openstack/cinder: Add rootwrap daemon mode support
openstackgerritAnton Arefiev proposed openstack/cinder: Switch to use execute from cinder utils
*** diablo_rojo has joined #openstack-cinder13:10
eikkesmcginnis: which logs that aren't on e.g. ?13:11
*** martyturner has joined #openstack-cinder13:11
smcginniseikke: I'll look. But that's part of the point. I shouldn't have to go digging all through your stuff to find them.13:11
*** agarciam has quit IRC13:12
*** martyturner has quit IRC13:13
*** agarciam has joined #openstack-cinder13:13
smcginniseikke: How are you addressing the requirement to not have to download the file to view it. These are all gzipped, but via Jenkins it doesn't allow me to view in browser.13:13
*** martyturner has joined #openstack-cinder13:13
smcginniseikke: It does look like at least all of the files are there. You just need to get them to an accessible place that allows viewing.13:13
*** dustins has joined #openstack-cinder13:14
eikkesmcginnis: ok, that's something we'll work on (with some nginx filter or so)13:14
eikkewe'll also let subunit dump an xunit xml so the test results can be listed/visualized/... in the jenkins UI13:15
*** dustins has quit IRC13:15
*** links has quit IRC13:15
*** martyturner has quit IRC13:18
*** dustins has joined #openstack-cinder13:18
*** merooney has quit IRC13:20
*** diablo_rojo has quit IRC13:20
eikkesmcginnis: fwiw, the whole scality openstack team literally dropped whatever we were doing to handle this asap ;-)13:21
smcginniseikke: Awesome. Hopefully it won't be that big of a deal for you. Glad to see your giving it the attention though. Thanks!13:22
*** gouthamr has joined #openstack-cinder13:22
*** julim has joined #openstack-cinder13:22
stefan_amannthingee: may I ask you for your help with bug #1488233? System z / s390 will not be able to attach volumes when the FC LUN ID is >255. I proposed a fix for os-brick for Liberty. We need to back port to Kilo (Cinder, Nova, and os-brick since os-brick is not integrated in Kilo). But there is no option to nominate it for Kilo. At least not for Cinder, and os-brick. Any help would be appreciated!13:22
openstackbug 1488233 in os-brick "FC with LUN ID >255 not recognized" [Undecided,In progress] - Assigned to Stefan Amann (stefan-amann)13:22
smcginnisstefan_amann: thingee is on vacation right now.13:23
stefan_amannok.. thanks for letting me know! Anybody else who may be able to help?13:23
smcginnisstefan_amann: I think jgriffit1 has the ability to do that.13:23
* smcginnis throws jgriffit1 under the bus.13:23
*** Zhang has joined #openstack-cinder13:24
*** kvidvans has quit IRC13:25
*** ericksonsantos has joined #openstack-cinder13:25
*** annegentle has quit IRC13:27
*** eduardo has joined #openstack-cinder13:29
*** jgriffit1 is now known as jgriffith13:32
eharneystefan_amann: i nominated the bug for kilo, but the main thing is just to get the patch submitted for kilo after it lands in master13:33
*** diablo_rojo has joined #openstack-cinder13:34
*** Yogi1 has joined #openstack-cinder13:34
stefan_amannthanks... did you do it for os-brick and cinder? I released it to Liberty/os-brick already. There is no need to release it to Liberty/Cinder.13:35
jgriffithstefan_amann: I'm mixed on it, but I think everyone will agree on and I've seen more questionable patches for IBM sail through Nova and Cinder13:35
jgriffithstefan_amann: You'll need to propose it and of course modify it as there's no brick in kilo13:35
stefan_amannIt only affects System z because OpenStack has to use a special command to add each LUN. And the command is sensitive to the format. We discovered that the current implementation does not support LUN IDs >0xff. Hence OpenStack users can attach a max of 255 LUNs.13:37
stefan_amannSystem z partitions can be pretty large and powerful. And we have products integrating the Kilo code. That's why we would like to backport13:37
*** nug is now known as cFouts13:38
stefan_amanneharney: thanks for nominating! Can you also nominate it for os-brick / Kiloß13:38
eharneystefan_amann: there's no kilo release of os-brick, you have to fix it in Cinder13:40
stefan_amannok :-)13:40
stefan_amannThanks much, guys!!13:40
*** merooney has joined #openstack-cinder13:41
*** annasort has quit IRC13:43
*** annasort_ has joined #openstack-cinder13:43
*** annasort_ is now known as annasort13:43
*** kvidvans has joined #openstack-cinder13:48
aarefievjgriffith, hi hcould you share your thoughts about
jordanPsmcginnis, you can now "view" a gzipped file without downloading it : (you have to click on the "view" link)13:52
*** breitz has quit IRC13:52
*** breitz has joined #openstack-cinder13:52
*** annasort has quit IRC13:52
eikkethanks to @jordanP's mad nginx skills ^^13:53
openstackgerritOpenStack Proposal Bot proposed openstack/cinder: Updated from global requirements
jgriffithaarefiev: seems ok13:56
jgriffithDuncanT: I would be curious to see why you're so interested in this particular patch?13:56
jgriffithaarefiev: DuncanT I believe that sometimes we give people too many options, and create things that we can't possibly support or maintain well :(13:56
jgriffithaarefiev: it would probably be good if we started being selective in our options if you ask me13:57
smcginnisjordanP, eikke: Cool! A little unconventional, but I think that probably meets most requirements.13:57
smcginnisjordanP, eikke: Can you update your comment posting to include the /artifact/jenkins-logs/ at the end of the URL.13:57
smcginnisjordanP, eikke: That way it's obvious and no need to click around to find your logs.13:58
jordanPsmcginnis, yep I'll look at it13:58
smcginnisjordanP, eikke: I think that would be good enough (at least for now) to meet the requirements.13:58
smcginnisjordanP: Thanks!13:58
*** mtanino has joined #openstack-cinder13:58
aarefievjgriffith: yep, but it's really nice option, especially for big envs with a lot concurrent access13:59
smcginnisaarefiev: I see value in it, even though I don't personally have need for it.13:59
smcginnisaarefiev: Seems like a good deployment option.13:59
smcginnisjgriffith: Good point though about more options to maintain.14:00
jgriffithaarefiev: :)14:00
jgriffithaarefiev: anybody looking at doing the same across the other services?14:00
smcginnisjgriffith: Keystone appears to have it.14:00
jgriffithsmcginnis: I was mainly thinking Nova14:01
aarefievjgriffith: unfortunately nova and ironic already did it(14:01
*** diogogmt has quit IRC14:01
jgriffithWell in that case....14:01
jgriffithI'm just behind the times :)14:01
aarefievI thought cinder will be the first one:)14:01
*** eharney has quit IRC14:02
aarefievafter keystone14:02
*** thangp has joined #openstack-cinder14:03
*** rushiagr is now known as rushiagr_away14:03
*** annegentle has joined #openstack-cinder14:03
*** diogogmt has joined #openstack-cinder14:04
*** links has joined #openstack-cinder14:05
jgriffithstefan_amann: Given this hasn't even landed in os-brick yet you might want to focus on that first :)14:09
jgriffithhemnafk: ^^  when you get in this morning lemme know14:10
stefan_amannjgriffith: I'm confused now :-) isn't that the patch for os-brick? Or what do you mean with 'landed'14:12
stefan_amanndoes 'landed' mean merged?14:14
stefan_amannsorry for the dumb question :-)14:14
*** annasort has joined #openstack-cinder14:15
*** yrabl has quit IRC14:16
*** annasort_ has joined #openstack-cinder14:18
jgriffithstefan_amann: no worries :)14:18
*** diogogmt has quit IRC14:18
jgriffithstefan_amann: yes, by landed I mean merged14:18
wilson2angela-s: are you Angela Smith?14:19
stefan_amannok :-)14:19
wilson2angela-s: :)14:19
jgriffithstefan_amann: and there's also now a tricky point because even after it merges in os-brick, it's still not picked up by Cinder until a new lib is released14:19
jgriffithstefan_amann: so it makes things rather sloowwwwww going :)14:19
*** annasort has quit IRC14:20
*** annasort_ is now known as annasort14:20
*** takedakn has quit IRC14:22
*** rushil has joined #openstack-cinder14:22
openstackgerritDiem Tran proposed openstack/cinder: Local cache feature of Oracle ZFSSA drivers
*** rushiagr_away is now known as rushiagr14:22
*** haomaiwang has joined #openstack-cinder14:22
*** lifeless1 has joined #openstack-cinder14:23
*** lifeless has quit IRC14:24
yuriy_n17jgriffith: Hi!14:25
*** edtubill has joined #openstack-cinder14:27
*** ankit_ag has quit IRC14:27
*** EinstCrazy has quit IRC14:28
jordanPkaisers, hi. are you silvain kaiser ?14:30
jordanPsilvan sorry14:30
jgriffithyuriy_n17: What's up?14:31
yuriy_n17jgriffith: sorry for persistence. what about to review Thanks a lot for advance.;)14:32
jgriffithyuriy_n17: sorry, didn't know you updated it14:36
yuriy_n17jgriffith: Thanks a lot ;)14:37
*** Zhang has quit IRC14:37
*** mtanino has quit IRC14:37
*** eharney has joined #openstack-cinder14:37
jgriffithyuriy_n17: no problem14:43
*** mtanino has joined #openstack-cinder14:43
DuncanTjgriffith: My main interest with that patch is in API performance - it's currently pretty bad (in terms of requests per second), and running it under apache seems to give a noticable boost for little effort14:44
*** tpsilva has joined #openstack-cinder14:45
jgriffithDuncanT: but that patch doesn't run it under Apache :)14:45
jgriffithit's nginx14:45
jgriffithregardless I'm cool with the patch for sure14:45
DuncanTjgriffith: Bah, so I'm going senile, no surprise14:45
jgriffithI just wanted to see if somebody had some insight that I wasn't aware of as folks are pretty adament about merging it14:46
jgriffithand I personally haven't tried it14:46
*** Zhang has joined #openstack-cinder14:46
stefan_amannjgriffith: to your previous comment: is this related to Liberty only? We're less concerned about Liberty. Kilo is more pressing...14:46
jgriffithWas just curious is all14:46
DuncanTjgriffith: I tried it, seems to work fine and give up to 20% on the db-light API queries rate14:47
e0nejgriffith: actually, it allows run c-api both under apache & nginx. may be adding sample config for nginx if a good idea14:47
DuncanTjgriffith: For team wide values of 'I'14:47
jgriffithstefan_amann: so now that we have os-brick library, it means that changes merge there.... then eventually a never version of the lib is pushed out for release, at which time it shows up in Cinder and no in Nova14:47
jgriffithDuncanT: cool14:47
*** Zhang has quit IRC14:47
e0nejgriffith: i didn't put my +2 on it becasue of i was author of spec and helped a bit with patch.14:48
jgriffithe0ne: YES, we'll def need docs14:48
e0nejgriffith: agree14:48
jgriffithe0ne: that should be proopsed to openstack-docs for config options etc14:48
e0nejgriffith: we'll do it14:48
jgriffithe0ne: yeah, figured that migt be the case14:48
*** Zhang has joined #openstack-cinder14:48
jgriffithe0ne: I knew you were involved with it14:49
jgriffithe0ne: thanks!14:49
*** tsekiyama has joined #openstack-cinder14:49
*** Zhang has quit IRC14:49
e0nejgriffith: also i'm going to post some banchmarks results once i'll have a time for it14:49
jgriffithe0ne: cool... always nice to have data14:49
eikkesmcginnis: fwiw, our CI now posts a link straight to the artifacts/log page. I sent a recheck message to the review you commented on, but later runs (e.g. for 214595) already have the updated link14:49
*** Zhang has joined #openstack-cinder14:49
e0nejgriffith: yea, i did it for keystone, it's time to make cinder perf tests :)14:50
jgriffithe0ne: DuncanT one thing I've been reading (true or not... it's on the internet so who knows) is that there's in some cases an equal trade off for the performance14:52
jgriffithe0ne: DuncanT in error rate14:52
jgriffithe0ne: DuncanT and mixed data on memory usage14:53
e0nejgriffith: which cases do you mean?14:53
jgriffithe0ne: DuncanT in other words, "yeah" it's really fast using nginx, but it has a higher error-rate and larger memory footprint14:53
jgriffithe0ne: DuncanT take that with a grain of salt of course :)14:54
DuncanTjgriffith: Memory usage definitely goes up a bit, haven't tested nearly nearly enough to comment on errors - that worries me rather more14:54
DuncanTjgriffith: The memory usage compared to a modern service is still small, so that doesn't worry me at all14:54
jgriffithDuncanT: e0ne I certainly think it's a fine option to have, but as I commented on the patch the commit message is a bit biased :)14:54
jgriffithWould love to hear feedback from folks that are maybe running Nova with it in production14:55
jgriffithI'm not knocking the patch, I'm trying to figure out if I want to invest in using it or not :)14:55
* thrawn01 is happy we are looking at preformance, I was planning on looking at it soonish, our cinder api in production currently preforms pretty badly14:55
*** anshul has quit IRC14:55
jgriffiththrawn01: what version of Cinder?14:56
jgriffiththrawn01: frankly we've gone metadata crazy the last few releases and I think we're suffering from it14:56
jgriffiththrawn01: that and really "bad" programming in some of the API calls14:56
smcginniseikke: Awesome! Thanks Scality team for the attention and quick turnaround. Nice work!14:57
thrawn01lol, that is the best question! it's currently icehouse, but we are planning a quick upgrade cycle so we can get on the latest and start squashing bugs/performance issues that effect us.14:57
jgriffiththrawn01: like iterating over a list of all volumes and doing multiple fetches on each one :(14:57
jgriffiththrawn01: icehouse had a MAJOR problem in list-volumes14:57
thrawn01yeah, icehouse is better than havana for sure. but it's still not great.14:57
*** Zhang has quit IRC14:57
thrawn01oh yeah, I know it well14:58
jgriffiththrawn01: did a get on all volumes, then iterated as I described and did multiple fetches to get various metadata etc14:58
*** topshare has joined #openstack-cinder14:58
jgriffiththrawn01: :)14:58
*** topshare has quit IRC14:58
*** topshare has joined #openstack-cinder14:58
thrawn01jgriffith: =( we had to disable some extensions as they made the situation worse. we counted 4 N+1 selects with all the extensions running14:58
jgriffithanywho... some day it would be awesome if we cleaned up our code :)14:59
*** ronis has quit IRC14:59
*** links has quit IRC14:59
jgriffiththrawn01: ouch!!14:59
jgriffiththrawn01: you should file a bug and list those out so folks can take a look at them14:59
thrawn01once we get closer to liberty, I'll do just that!14:59
jgriffiththrawn01: it's hard to believe how badly we abuse the DB :(14:59
thrawn01I'm currently working through some smaller easy to fix patches to get my feet wet.15:00
jgriffiththrawn01: just willie-nillie throw in some db calls anywhere you like... what could go wrong :)15:00
thrawn01and trying to figure out thest testing stuff, I just now got tox working.... I think15:00
jgriffiththrawn01: :)  Well let me know if I can help with tox15:01
*** topshare has quit IRC15:01
jgriffiththrawn01: I recently updated the devref docs in tree and hopefully that will help people at least a little15:01
*** topshare has joined #openstack-cinder15:01
* thrawn01 looks15:01
*** kvidvans has quit IRC15:01
*** daneyon has joined #openstack-cinder15:01
openstackgerritSzymon Wróblewski proposed openstack/cinder: Support initialization state in Backup Manager
*** asd112z has joined #openstack-cinder15:02
thrawn01jgriffith: in cinder or cinderclient source?15:02
*** topshare has quit IRC15:02
jgriffiththrawn01: cinder15:03
*** topshare has joined #openstack-cinder15:03
*** vgridnev has quit IRC15:03
jgriffiththrawn01: same work could/should be done in client15:03
jgriffithI think15:03
*** topshare has quit IRC15:03
*** topshare has joined #openstack-cinder15:04
smcginnisjgriffith, DuncanT, geguileo, e0ne, hemnafk, xyang1: Not sure what to think of this one:
openstackgerritMichal Dulko proposed openstack/cinder: Validate filters in snapshot*, backup* in db.api
e0nesmcginnis: just -1 on it! :)15:04
smcginnise0ne: :)15:04
e0nenow, the long version....15:05
openstackgerritSzymon Wróblewski proposed openstack/cinder: utils.synchronized lock name formatting
DuncanTsmcginnis: While I'd love to see the block device driver die a quick death15:05
e0nethat patch will be updated soon to support loopback devices15:05
smcginnisDuncanT: That's my dilemma.15:05
*** crose has joined #openstack-cinder15:05
xyang1smcginnis: No snapshot support still?15:06
e0neDuncanT: why?15:06
*** topshare has quit IRC15:06
DuncanTe0ne: There was talk at the midcycle 6 months ago about getting it fixed up to minimal feature support, not happened though15:07
e0nexyang1: it's by design. we can't create snapshot from a whole devide15:07
*** Zhang has joined #openstack-cinder15:07
jgriffithsmcginnis: see my comments on the patch:
*** r-daneel has joined #openstack-cinder15:07
DuncanTe0ne: Do what rax do for snaps and copy the whole lot somewhere else?15:07
smcginnisDuncanT: I think that's another topic that has been discussed multiple times with no resolution.15:07
e0neDuncanT: that's an option15:07
guitarzanthat's gonna be difficult without dm snapshots helping out...15:07
jgriffithe0ne: see my comment in the patch15:07
DuncanTguitarzan: Offline only snaps? It will at least pass tempest15:08
jgriffithe0ne: DuncanT smcginnis my proposal was and still is, remove it from Cinder and have it be part of Sahara15:08
guitarzanDuncanT: I do suppose that's right15:08
openstackgerritSzymon Wróblewski proposed openstack/cinder: Python 3 incompatible expression fix
*** hemnafk is now known as hemna15:09
smcginnisjgriffith: I think I agree with that approach.15:09
*** haomaiwang has quit IRC15:09
smcginnisjgriffith: Either meet minimum requirements and have CI coverage, or move it out and maintain it in Sahara - the one and only place it's needed.15:09
jgriffithsmcginnis: or if you want to do big-data *stuff* use Cinder storage that's capable of doing it15:09
DuncanTguitarzan: Having a driver in that only supports a fraction of our 'mandatory' features seems rather broken, and the performance .v. lvmthick seems to be lost in the noise according to my tests15:09
smcginnisjgriffith: Yeah, I don't think it was every proven the the block device driver is absolutely needed to meet the performance requirements needed.15:10
*** boris-42 has quit IRC15:10
e0newhat if we'll implement required features and ci?15:10
hemnajgriffith, +1 on your comment on that patch15:10
*** edtubill has left #openstack-cinder15:10
smcginnisWhat DuncanT said.15:10
*** asd112z has quit IRC15:10
*** haomaiwang has joined #openstack-cinder15:10
guitarzanDuncanT: yeah, I don't understand the "lvm is crippling" idea myself15:10
DuncanTe0ne: Then it becomes just another driver and most of us stop caring I think15:10
smcginnisjgriffith: Yes, good comment on patch.15:11
e0nei'll propose patch for infra job once devstack integration will be finished15:11
smcginnise0ne: Yeah, I'd be fine with it then.15:11
*** asd112z has joined #openstack-cinder15:11
*** timcl1 has joined #openstack-cinder15:11
jgriffithe0ne: FYI it won't run them currently15:11
DuncanTguitarzan: Even with ramdisks as the backing store straight R/W performance was withing 5% or so for me15:11
hemnabut, if there is no CI on it, aren't we supposed to remove it ?15:11
jgriffithe0ne: it will fail miserable :(15:11
hemnathat's the standard now anyway15:11
e0neDuncanT, smcginnis: fair enough15:11
smcginnisjgriffith: Then we can remove it for failing CI. :)15:11
guitarzanDuncanT: I think you're preaching to the choir :)15:11
DuncanThemna: It has been a special snowflake for a while now15:11
jgriffithhemna: I don't know if you noticed but there are a quite a few drivers that don't have CI right now :(15:11
DuncanTguitarzan: :-)15:12
smcginnisWe probably need to do an audit of that again.15:12
hemnajgriffith, which is ok for a time, as long as the CI is being worked on and communication with the driver maintainer15:12
hemnanone of which is true for bdd ?15:12
*** timcl has quit IRC15:12
smcginnisWhat ever happened to the CI dashboard work?15:12
*** jistr is now known as jistr|mtg15:12
jgriffithWell here's a quick sample of just the last 48 hours:
e0neok. we need working CI and minimum features set in bdd15:13
*** jaypipes has joined #openstack-cinder15:13
jgriffithI have no idea how patrickeast 's tool there actually works, but it seems pretty :)15:13
e0nehemna: tbh, i didn't get any asks about adding new features to the bdd driver:(. it's my fault tooo15:14
*** Zhang has quit IRC15:14
*** diogogmt has joined #openstack-cinder15:15
*** zhangjn has joined #openstack-cinder15:15
*** zhangjn has quit IRC15:16
*** zhangjn has joined #openstack-cinder15:16
*** bnemec has joined #openstack-cinder15:17
jgriffithhemna:, 2015-08-26 15:11:52.44915:17
jgriffithhemna: looks suspect15:17
*** zhangjn has quit IRC15:17
*** stefan_amann has quit IRC15:18
hemnajgriffith, hrmm...wth is that15:18
jgriffithhemna: I have no idea :)15:18
*** vgridnev has joined #openstack-cinder15:18
jgriffithhemna: but didn't that rootwrap patch for brick land recently?15:19
asselinthere's a new ci dashboard in town:
*** zhangjinnan has joined #openstack-cinder15:19
jgriffithhemna: so I'm testing against os-brick master15:19
*** zhangjinnan has quit IRC15:19
hemnajgriffith, hasn't landed yet15:19
jgriffithhemna: yeah, just saw that... wth....15:20
*** zhangjn has joined #openstack-cinder15:20
hemnano idea what that one is15:20
hemnanever seen it before :(15:20
jgriffithI've hit it a few times now15:20
jgriffithlooking into it15:20
hemnalooks like the self.instream isn't exactly what they expected then15:21
hemnaprobably some other object that doesn't have read ?15:21
*** changbl has quit IRC15:21
hemnavery strange15:21
jgriffithhemna: the thing is it's in the os-brick connector that it's puking it seems15:21
*** BharatK has joined #openstack-cinder15:22
*** zhangjn has quit IRC15:22
hemnayah, it looks like it's trying to read the initiator file to get the host iqn15:22
*** zhangjn has joined #openstack-cinder15:22
*** rushil has quit IRC15:22
*** zhangjn has quit IRC15:22
jgriffithhemna: yeah.. the call to 'cat'15:22
*** jistr|mtg is now known as jistr15:22
*** zhangjn has joined #openstack-cinder15:23
jgriffithbut it's actually the shlex lib?15:23
*** kvidvans has joined #openstack-cinder15:24
hemnaI that's the oslo processutils doing that15:25
hemnano idea what that is.15:25
jgriffithhemna: I think I might know... just a sec15:25
*** zhangjn has quit IRC15:25
*** zhangjn has joined #openstack-cinder15:26
*** zhangjn has quit IRC15:26
*** zhangjn has joined #openstack-cinder15:26
*** zhangjn has quit IRC15:26
*** martyturner has joined #openstack-cinder15:27
*** zhangjn has joined #openstack-cinder15:27
*** zhangjn has quit IRC15:27
*** zhangjn has joined #openstack-cinder15:28
jgriffithhmm...or not :(15:28
*** zhangjn has quit IRC15:28
*** zhangjn has joined #openstack-cinder15:29
*** jdurgin1 has joined #openstack-cinder15:30
hemnawish I knew how this was reproduced15:30
jgriffithhemna: looks like it's 100% for me right now15:31
jgriffithwoner if shlex updated?15:31
jgriffithand something broke....15:31
jgriffithhmm... nope, that's part of python, not an external lib... weird15:32
hemnais shlex itself part of python?15:33
hemnaoh heh15:33
hemnawas just looking for it15:33
*** zhangjn has quit IRC15:33
hemnaoslo_concurrency 2.5.0 was released on 8/2415:34
*** haomaiwang has quit IRC15:34
*** edtubill has joined #openstack-cinder15:35
aarefievjgriffith: do you looking on ci on my patch?15:35
jgriffithaarefiev: I have no idea how to parse what you just said :)15:36
*** bluex-pl has quit IRC15:36
aarefievjgriffith: if you have something like 'RootwrapProcessHelper' object has no attribute 'read'15:37
hemnajgriffith, but I don't see anything in around that call that's changed recently15:37
*** vgridnev has quit IRC15:37
jgriffithaarefiev: that's the error I'm getting yes, when shlex tries to parse the tokenized root_helper that's passed in15:37
jgriffithI need to run down the hall before our meeting... bb in a few15:38
aarefievjgriffith: don't worry, this is one causes it15:38
hemnais gerrit just horribly slow today?15:39
smcginnishemna: Today?15:39
hemnasmcginnis, :)15:39
hemnaaarefiev, that hasn't landed though15:40
*** afazekas has quit IRC15:40
Swansonhemna: It certainly was yesterday.  And review pages are taking forever to load today.  I just assumed Dell added a few extra firewalls between here and the rest of humanity.15:41
*** vgridnev has joined #openstack-cinder15:41
aarefievhemna: jgriffith told he get the same error, my patch introduce 'RootwrapProcessHelper' :)15:41
hemnaSwanson, it's abnormally slow for me here today.  I thought it just might be the super awesome HP network15:41
thrawn01anyone here run tests on osx?15:42
thrawn01ran into some trouble getting pycrypto installed for pypy15:42
smcginnishemna: Is it down now? Just got a failure posting CI results. Network is unreachable.15:43
-openstackstatus- NOTICE: restarting gerrit due to a slow memory leak15:44
hemnasmcginnis, lol, I guess that answers your question.15:44
*** alexschm has quit IRC15:44
smcginnisThird times a charm. Worked now.15:44
smcginnishemna: Darn internets.15:44
hemnasmcginnis, openstackstatus/#openstack-cinder- NOTICE: restarting gerrit due to a slow memory leak15:44
hemnasmcginnis, in case you missed that15:44
*** garthb has joined #openstack-cinder15:45
smcginnishemna: Yep, didn't see that yet.15:45
*** mtanino has quit IRC15:45
*** mtanino has joined #openstack-cinder15:48
*** dannywilson has joined #openstack-cinder15:48
*** scohen1 has joined #openstack-cinder15:49
openstackgerritGorka Eguileor proposed openstack/cinder: Fix backup metadata import missing fields
*** agarciam has quit IRC15:50
*** dflorea has joined #openstack-cinder15:51
*** vincent_hou has joined #openstack-cinder15:52
*** p0rtal has joined #openstack-cinder15:53
*** vincent_hou_ has joined #openstack-cinder15:53
smcginnisAnyone else going to be at a large vendor conference in San Francisco next week?15:54
jgriffithaarefiev: ahhh :)15:54
hemnasmcginnis, ?15:54
jgriffithaarefiev: curious why it passes gate though ?15:54
jgriffithbut yes, it was that one and it's dep15:54
smcginnishemna: Going to be in San Fran next week through Thursday. Just wondering if anyone else here happens to be going.15:54
aarefievjgriffith: there are still places where code uses processutils directly15:55
*** adurbin_ has joined #openstack-cinder15:55
jgriffithaarefiev: yeah, and you'll need to figure out some way to coordinate the with os-brick it appears15:55
* hemna is confused how the RootwrapProcessHelper is already being used when it hasn't landed yet in Cinder15:56
SwansonGot a simple patch that really should only impact our CI (as the function in question should really only happen in the tempest tests)
*** vincent_hou has quit IRC15:56
*** vincent_hou_ is now known as vincent_hou15:56
*** vgridnev has quit IRC15:56
aarefievjgriffith: yeah, a big patch, touches all drivers in diff way (15:57
*** rajinir has joined #openstack-cinder15:57
rhe00smcginnis: Are you going to promote OpenStack at that conference? ;)15:57
hemnajgriffith, oh, your CI log is from the solidfire CI test of that patch.15:57
*** mtanino_ has joined #openstack-cinder15:57
smcginnisrhe00: Where I can. ;)15:57
jgriffithhemna: :)15:57
smcginnisrhe00: Are you going?15:58
hemnaI catch up....eventually.15:58
jgriffithhemna: I was suspicious because I had two in a row... didn't notice the second was dependent on the first15:58
rhe00smcginnis: No, didn't make it on the list.15:58
jgriffithhemna: haha... that's alright, luckily aarefiev was paying attention for both of us :)15:58
jgriffithhemna: aarefiev I knew I saw that rootwrap change, but confused it with the one in os-brick that hasn't merged yet15:59
*** skylerberg has joined #openstack-cinder15:59
jgriffithshall we have a meeting?15:59
smcginnisrhe00: Oh well. Not too excited about going. Been to enough of them now.15:59
aarefievjgriffith: we should refactor the way how we execute cmds, all drivers do it in diff way, we need common approach15:59
jgriffithaarefiev: agreed!!!!15:59
smcginnisMeeting time.15:59
hemnajgriffith, looks like the Netapp cDOT-iSCSI CI is failing for the exact same reason16:00
*** markstur has quit IRC16:00
*** mtanino has quit IRC16:00
*** p0rtal has quit IRC16:00
*** dims_ has quit IRC16:00
*** p0rtal has joined #openstack-cinder16:00
*** changbl has joined #openstack-cinder16:00
*** dims has joined #openstack-cinder16:01
anteayaeharney: how did your patch make out after you removed the circular dependency?16:02
*** timcl has joined #openstack-cinder16:04
*** _afazekas is now known as afazekas16:05
*** p0rtal has quit IRC16:05
*** DTadrzak has joined #openstack-cinder16:05
*** timcl1 has quit IRC16:05
*** markstur has joined #openstack-cinder16:07
*** nkrinner has quit IRC16:08
thrawn01anyone ever encounter an issue where you just can't install ANYTHING within a virtualenv?16:10
thrawn01I create a virtualenv with 2.6, and pip is unable to install anything.16:10
*** jistr has quit IRC16:12
*** jordanP has quit IRC16:13
*** Apoorva has joined #openstack-cinder16:13
*** kvidvans has quit IRC16:14
*** yuriy_n17 has quit IRC16:15
openstackgerritPatrick East proposed openstack/cinder: Generic image-volume cache
*** leeantho has joined #openstack-cinder16:21
*** agarciam has joined #openstack-cinder16:24
*** dubutante has joined #openstack-cinder16:24
*** _cjones_ has joined #openstack-cinder16:26
*** ronis has joined #openstack-cinder16:27
openstackgerritHuang Zhiteng proposed openstack/os-brick: iSCSI Connector: force disconnect iscsi session
*** martyturner1 has joined #openstack-cinder16:30
*** martyturner has quit IRC16:32
*** skylerberg has quit IRC16:34
*** jdurgin1 has quit IRC16:34
*** dflorea has quit IRC16:35
bswartzjgriffith: I'm trying to find the patch you referred to about capabilities during the meeting16:40
*** p0rtal has joined #openstack-cinder16:40
*** p0rtal has quit IRC16:40
*** vincent_hou_ has joined #openstack-cinder16:45
*** vincent_hou has quit IRC16:46
*** vincent_hou_ is now known as vincent_hou16:46
*** rushil has joined #openstack-cinder16:48
*** nikeshm_ has joined #openstack-cinder16:49
*** nikeshm_ has quit IRC16:50
*** lpetrut has quit IRC16:51
*** navneet_ has joined #openstack-cinder16:52
navneet_experimental thing makes sense for feature with no wide acceptability16:52
*** navneet_ has quit IRC16:53
*** simondodsley has quit IRC16:53
*** simondodsley has joined #openstack-cinder16:53
*** mudassirlatif has joined #openstack-cinder16:58
jbernardvincent_hou: ping regarding in-use volume migration17:00
*** martyturner1 has quit IRC17:01
*** jseiler_ has quit IRC17:02
bswartzjgriffith: ty17:03
*** dubutante has quit IRC17:03
jgriffithvincent_hou: sorry we ran out of time17:03
jgriffithvincent_hou: but I think everybody is here17:03
*** winston-d_ has joined #openstack-cinder17:03
vincent_houThe good news with is that we can report the progress for the volume migration.17:03
vincent_houNo problem.17:03
vincent_houfolks have concerns that I make the driver too heavy.17:03
*** tpsilva has quit IRC17:04
*** jseiler has joined #openstack-cinder17:04
vincent_houdoing things like getting the process id, reading the files, calculating, etc.17:04
eharneythe architecture here just isn't right... it's pushing too much stuff into the driver that shouldn't be there17:04
jgriffitheharney: I agree17:04
hemnaI like the feature and what it brings to Cinder.   I just think it's in the wrong place.  eharney +117:04
eharneysame here17:04
winston-d_process id? that's crazy, why does anybody cares about that.17:05
vincent_houShall I move them to manager?17:05
jgriffithwinston-d_: hehe17:05
eharneyit's a great idea, it just needs to be done more in the volume manager17:05
jgriffithI'd propose we fine tune it and fix thing up to use manager and bumpt it out to M at this point17:05
vincent_houI need to get the process id for "dd" to check how much transferred.17:05
hemnajgriffith, +117:06
eharneyvincent_hou: it's reasonable to add some driver hooks if we figure out that that's necessary... i'm not sure what the final interface would look like17:06
*** mistermarkwest has joined #openstack-cinder17:06
jgriffithvincent_hou: make a genrator that returns chunk completion :)17:06
*** trree has joined #openstack-cinder17:06
guitarzanit's a question of how much of the "default implementation" is a driver detail vs a manager detail17:06
eharneybut because of questions like that, i'm also thinking what jgriffith said17:06
*** Yogi1 has quit IRC17:06
*** boris-42 has joined #openstack-cinder17:07
*** jungleboyj has joined #openstack-cinder17:07
*** rushiagr is now known as rushiagr_away17:07
vincent_houFor the migration with dd command, I think I can figure out a sort of interface.17:08
*** rushiagr_away is now known as rushiagr17:08
jgriffithvincent_hou: you want to take some time on it and postpone to M?17:08
jgriffithvincent_hou: I don't see it as critical for L at this point17:08
jgriffithvincent_hou: fact is there are still problems with migrate outside of progress reporting17:09
jgriffithvincent_hou: although you've fixed up a good deal of things17:09
*** rushiagr is now known as rushiagr_away17:09
*** mistermarkwest has quit IRC17:09
vincent_houWell, it is part of the BP. frastrate me a bit.17:09
jgriffithvincent_hou: understand completely17:10
vincent_houif it is delayed.17:10
jgriffithvincent_hou: but it's a pretty heavy change17:10
*** rushiagr_away is now known as rushiagr17:10
jgriffithvincent_hou: if you think you can redesign things in the next week I guess that's up to you17:10
*** diablo_rojo has quit IRC17:10
*** trree has quit IRC17:10
jgriffithvincent_hou: the problem as I see it though is you have 7 other patches outstanding associated with that BP17:11
jgriffithvincent_hou: it doesn't seem like it's possible to get all of them finalized and stable in the next few days17:11
*** skylerberg has joined #openstack-cinder17:11
openstackgerritSonia Ghanekar proposed openstack/cinder: Add manage/unmanage volume support for Nimble
jgriffithvincent_hou: it's of course up to you17:11
vincent_houThings getting much for me here17:11
jgriffithvincent_hou: you have the approved bp's and the reviews up17:11
jgriffithvincent_hou: but I don't think anybody is going to rush them through as they are now17:12
vincent_houI see.17:12
jgriffithvincent_hou: and I don't know that it's worth you killing yourself writing code 24/7 for the next week17:12
jgriffithvincent_hou: but that's completely up to you17:12
hemnaok off to a meeting/beating.17:13
vincent_houThank you,, John. I know what you mean.17:13
*** hemna is now known as hemnafk17:13
*** winston-d is now known as winston-d_zZZ17:13
*** trree has joined #openstack-cinder17:13
jgriffithvincent_hou: it's great that you've stepped up and are working so hard on fixing this up17:13
jgriffithvincent_hou: I'm just saying you need to decide how to spend your time :)17:14
vincent_houI will do my best, but will leave the option open for the patches.17:14
jgriffithvincent_hou: thanks!!17:14
*** e0ne has quit IRC17:15
vincent_houThank you, John. Thanks everyone for the reviews.17:15
jgriffithpatrickeast: kmartin hemnafk winston-d_ geguileo e0ne So I'm proposing that for the *next* release (M) we have an earlier deadline for core features17:15
patrickeastjgriffith: i like that idea17:15
jgriffithpatrickeast: kmartin hemnafk winston-d_ geguileo e0ne at least two weeks before feature freeze17:16
jgriffithxyang1: ^^17:16
patrickeastjgriffith: i had been wondering if it makes sense to push new drivers only into the third milestone and have 1 and 2 for core features17:16
patrickeastthen do a similar milestone 2 freeze like nova17:16
*** lpetrut has joined #openstack-cinder17:16
jgriffithpatrickeast: TBH we used to do it that way17:16
jgriffithpatrickeast: well... not exactly, and not like Nova17:16
jgriffithpatrickeast: but driver mods were open all the way to the third milestone17:17
*** tpsilva has joined #openstack-cinder17:17
patrickeastjgriffith: gotcha17:17
vincent_houjbernard: hi17:17
jgriffithpatrickeast: that was changed to stop the "new" driver submissions, then kinda extended itself to other things17:17
jgriffithpatrickeast: but the key IMO is there should be a buffer there17:17
jgriffithpatrickeast: and using Nova's model is not really the same in Cinder because of our plugin model17:18
xyang1jgriffith: I thought we tried last time but difficult to get things merged that early17:18
jgriffithxyang1: I don't recall17:18
jgriffithxyang1: and by early I'm just saying a couple weeks17:18
xyang1jgriffith: I'll have to check17:18
jgriffithxyang1: so even saying driver updates all the way to 3'rd milestone17:18
jgriffithxyang1: and stil the two week buffer for core features17:19
jgriffithxyang1: but I could be completely wrong... maybe we did try it17:19
xyang1jgriffith: I think the 2 nd milestone seems difficult to meet, but not against trying it again17:19
jgriffithxyang1: I also think that maybe we should tie it to the feature itself17:19
jgriffithxyang1: oh, I'm certainly not proposing that!!!17:20
jgriffithxyang1: I'm saying a couple of weeks is all17:20
jgriffithxyang1: and I'm thinking what if you tied dates to the feature's release in core17:20
xyang1jgriffith: sure17:20
jgriffithxyang1: so for example, if it landed in core early... well then you're doomed17:20
jgriffithxyang1: but if it lands at the last hour of the freeze date there's an amnisty period of sorts for drivers to implement it17:21
jgriffithxyang1: I think it should be that or nothing until the next release... not sure which I think is better17:21
xyang1jgriffith: Why you are doomed if it land early?17:21
jgriffithxyang1: but that will be a great summit topic IMO17:22
jgriffithxyang1: I mean if it lands in core early, you have no excuses and no grace period17:22
xyang1jgriffith: Ya, good to discuss at the summit17:22
xyang1jgriffith: Oh, I see:)17:22
xyang1jgriffith: So what does the voting suggest?  Any action now?17:24
dannywilsonthat sounds good to discuss for next cycle but why are you trying to change the rules for L too? Changing rules mid cycle seems like a slippery slope17:25
*** stefan_amann has joined #openstack-cinder17:27
jgriffithxyang1: dannywilson bswartz I did the voting all wrong :(17:27
dannywilsonoh noes17:27
jgriffithI should have proposed it a choice between the two options in the vote itself17:27
*** daneyon has quit IRC17:27
xyang1jgriffith: oh:(17:27
jgriffithbut it seems like the majority were in favor of extending deadline for driver to implement late landing features17:28
xyang1jgriffith: You can still count manually:)17:28
jgriffithxyang1: yeah, that's what I did :)17:28
jgriffithbut it would've solved the issue of people voting the same on both options (like I did)17:29
xyang1jgriffith: :)17:29
jgriffithxyang1: I'm out of practice using meetbot after the past year :)17:29
xyang1jgriffith: :)17:30
jgriffithvote-bot at any rate17:30
*** agarciam has quit IRC17:31
*** kvidvans has joined #openstack-cinder17:33
bswartzjgriffith: at least you don't have to deal with the parliamentary differences between "yea", "aye", "no", and "nay".17:33
*** stefan_amann has quit IRC17:34
*** martyturner has joined #openstack-cinder17:34
xyang1bswartz: We should count those as well:)17:35
jgriffithbswartz: LOL17:35
*** dflorea has joined #openstack-cinder17:38
*** martyturner has quit IRC17:39
*** rushiagr is now known as rushiagr_away17:41
*** dims_ has joined #openstack-cinder17:41
*** p0rtal has joined #openstack-cinder17:41
*** dims has quit IRC17:42
swamireddyAny one working cinder active-active functionality support ?17:42
*** winston-d_ has quit IRC17:43
*** aix has quit IRC17:44
*** hemnafk is now known as hemna17:44
hemnaswamireddy, there are notes on that from the mid cycle meetup17:44
*** dflorea has quit IRC17:46
*** trree has quit IRC17:47
*** stefan_amann has joined #openstack-cinder17:47
*** trree has joined #openstack-cinder17:47
*** diablo_rojo has joined #openstack-cinder17:48
*** jdurgin has quit IRC17:49
*** Yogi1 has joined #openstack-cinder17:49
*** rushiagr_away is now known as rushiagr17:53
openstackgerritRyan Lucio proposed openstack/cinder: VMEM v7000: support storage pools / volume types
*** andreykurilin__ has joined #openstack-cinder18:07
*** trree has quit IRC18:08
*** jdurgin has joined #openstack-cinder18:09
*** annasort has quit IRC18:10
*** yrabl has joined #openstack-cinder18:10
*** annegentle has quit IRC18:14
*** openstackgerrit has quit IRC18:17
*** openstackgerrit has joined #openstack-cinder18:17
*** zhenguo has quit IRC18:23
*** e0ne has joined #openstack-cinder18:25
*** tpsilva has quit IRC18:27
*** aix has joined #openstack-cinder18:30
*** jaypipes has quit IRC18:31
*** vincent_hou has quit IRC18:32
*** ociuhandu has quit IRC18:33
*** vincent_hou has joined #openstack-cinder18:34
*** shyama has quit IRC18:36
swamireddyhemna: scottda: Thanks for the link..18:37
*** lifeless1 is now known as lifeless18:39
scottdaswamireddy: sure. Sorry that there's no quick, easy answer there...18:39
*** lpetrut has quit IRC18:39
*** tpsilva has joined #openstack-cinder18:39
scottdapeople are working on this, it's just a bunch of different stuff in different places.18:39
swamireddyOne more Q is - I am upgrading the cinder from J ->K (or I->J). Here, after upgrading the binaries/pkgs, I will run the db_sync cmd to sync dB tables as per the upgrade.  Is the db_sync is safe to run or do we need to take any sfaty steps before db_sync cmd?18:40
*** jaypipes has joined #openstack-cinder18:40
*** esker has joined #openstack-cinder18:40
asselin_patrickeast, did find an issue. pushed a small change to fix18:41
swamireddyscottda:  yep, thats correct...its discussion stage ATM18:41
*** esker has quit IRC18:41
scottdaswamireddy: Sorry, I don't think I can help with the upgrade question.18:42
*** esker has joined #openstack-cinder18:42
patrickeastasselin_: sweet, thanks! i still haven't had any time to upgrade any of my systems :(18:42
jgriffithswamireddy: make a bckup of your db just to be safe18:43
jgriffithswamireddy: You should be fine, but always recommend at least a mysql-dump prior to the migration18:44
jgriffithswamireddy: it's easy and a small precaution to take :)18:44
swamireddyscottda: Thanks. nop.18:44
*** martyturner has joined #openstack-cinder18:44
*** tpsilva has quit IRC18:45
*** diegows has joined #openstack-cinder18:47
*** vincent_hou has quit IRC18:47
*** jdurgin has quit IRC18:48
*** e0ne has quit IRC18:50
*** jwcroppe has quit IRC18:50
*** asd112z_ has joined #openstack-cinder18:52
*** mtanino_ has quit IRC18:53
*** tsekiyam_ has joined #openstack-cinder18:53
*** dims_ has quit IRC18:53
*** dims has joined #openstack-cinder18:54
*** mtanino has joined #openstack-cinder18:54
*** asd112z has quit IRC18:56
*** tsekiyama has quit IRC18:57
*** tpsilva has joined #openstack-cinder18:58
openstackgerritVilobh Meshram proposed openstack/cinder: Cinder Nested Quota Driver
*** eduardo has quit IRC19:03
*** rushiagr is now known as rushiagr_away19:03
*** sidbhatt13 has joined #openstack-cinder19:05
*** jdurgin has joined #openstack-cinder19:07
*** lpetrut has joined #openstack-cinder19:07
*** jwcroppe has joined #openstack-cinder19:08
*** sidbhatt13 has quit IRC19:09
*** tpsilva has quit IRC19:12
*** e0ne has joined #openstack-cinder19:13
*** tsekiyam_ has quit IRC19:13
*** krtaylor has quit IRC19:14
*** esker has quit IRC19:15
*** lpetrut has quit IRC19:18
*** sidbhatt13 has joined #openstack-cinder19:19
*** yrabl has quit IRC19:20
*** BharatK has quit IRC19:21
*** Lee1092 has quit IRC19:23
*** stefan_amann has quit IRC19:24
openstackgerritMichael Price proposed openstack/cinder: Reduce runtime of E-Series iSCSI tests
*** ronis has quit IRC19:28
*** daneyon has joined #openstack-cinder19:29
openstackgerritDerrick Wippler proposed openstack/python-cinderclient: Fixed test_password_prompted
*** martyturner1 has joined #openstack-cinder19:31
openstackgerritPatrick East proposed openstack/cinder: Implement manage/unmanage snapshot in Pure drivers
*** martyturner has quit IRC19:33
*** daneyon has quit IRC19:33
*** adurbin_ has quit IRC19:34
*** tsekiyama has joined #openstack-cinder19:37
*** BharatK has joined #openstack-cinder19:37
*** mtanino_ has joined #openstack-cinder19:37
openstackgerritEric Harney proposed openstack/cinder: Remove driver.set_execute()
*** mtanino has quit IRC19:39
*** rushil has quit IRC19:41
*** lpetrut has joined #openstack-cinder19:41
*** ociuhandu has joined #openstack-cinder19:44
*** daneyon has joined #openstack-cinder19:48
openstackgerritEric Harney proposed openstack/cinder: Remove driver.set_execute()
*** daneyon has quit IRC19:54
*** daneyon has joined #openstack-cinder19:55
*** scohen1 has quit IRC19:58
*** sidbhatt13 has quit IRC20:00
*** scohen1 has joined #openstack-cinder20:01
*** scohen1 has quit IRC20:01
*** jwcroppe has quit IRC20:03
*** lpetrut1 has joined #openstack-cinder20:04
*** daneyon has quit IRC20:09
*** jwcroppe has joined #openstack-cinder20:10
*** krtaylor has joined #openstack-cinder20:11
*** Yogi1 has quit IRC20:12
*** jwcroppe has quit IRC20:14
*** merooney has quit IRC20:14
smcginnisLooks like Summit presentation acceptance letters are starting to go out.20:15
hemnasmcginnis, yup.   Looks like I lucked out from doing preso's this summit :P20:18
smcginnishemna: Dang it. Want to switch? :)20:18
*** akerr has quit IRC20:20
*** porrua has quit IRC20:20
hemnasmcginnis, how many do you get to do?20:20
*** martyturner1 has quit IRC20:21
cebrunssmcginnis: Everyone wants to hear the new core's song and dance.  :)20:22
e0nehemna: :). the next challenge - to get visa. you won it too20:24
smcginniscebruns: :P20:24
hemnae0ne, heh20:24
smcginnishemna: Just one. We've got the conversation with Cinder developers session again. One more as an alternate.20:24
hemnaah cool that one20:24
*** daneyon has joined #openstack-cinder20:24
hemnaI'm not sure I was invited to that one20:25
hemnaI'll sit in the audience and ask standardization and HA questions.20:25
smcginnishemna: I can see if I can add you as a speaker yet. ;)20:25
e0neneed to take a break to sleep. bb in ~12 hours:)20:27
hemnathat sleep thing....20:27
*** e0ne has quit IRC20:28
*** rushil has joined #openstack-cinder20:29
*** rushil_ has joined #openstack-cinder20:29
*** sidbhatt13 has joined #openstack-cinder20:30
*** rushil has quit IRC20:34
*** simondodsley has quit IRC20:34
smcginnishemna: You ever see leftover scsi devices on the host after a volume is detached?20:41
jgriffithsmcginnis: LOL20:41
smcginnisjgriffith: Is this a known thing? :)20:42
*** bswartz has quit IRC20:42
jgriffithsmcginnis: I'll let hemna answer that one :)20:42
hemnaoh man20:43
smcginnishemna: Oh, that's good.20:43
smcginnishemna: This was reported to me by one of our test guys.20:43
hemnagood :)20:44
smcginnishemna: Wouldn't be at all surprised if he's running older code than that.20:44
hemnaseems your test guys are good then :)20:44
smcginnishemna: Has that been released in a new os-brick library? Or still pending?20:44
hemnanot to pypi yet20:44
smcginnishemna: Cool, awesome. I can at least let them know the status. Thanks!20:45
smcginnishemna: I need to check, but it's possible they are still hitting Kilo. Is/was it an issue there, and if so has that been backported?20:47
hemnaso Kilo is a different story, because Nova hasn't used os-brick until a few weeks ago20:48
hemnaI think we switched Cinder to use os-brick in Juno20:48
hemnaand prior to that it was the internal version of brick20:48
smcginnisSounds right.20:49
*** asd112z_ has quit IRC20:51
*** asd112z has joined #openstack-cinder20:52
*** asd112z has quit IRC20:52
*** asd112z has joined #openstack-cinder20:53
Swansonsmcginnis: test dude reported it on the 6th.  So, well before the merge of that defect.20:57
*** lpetrut1 has quit IRC20:57
*** dustins has quit IRC20:57
*** julim has quit IRC20:58
*** lpetrut has quit IRC20:58
*** dflorea has joined #openstack-cinder20:58
*** raildo is now known as raildo-afk20:59
*** jwcroppe has joined #openstack-cinder20:59
*** kambiz has quit IRC21:00
*** kvidvans has quit IRC21:01
*** agarciam has joined #openstack-cinder21:01
*** openstackgerrit has quit IRC21:01
*** openstackgerrit has joined #openstack-cinder21:02
*** winston-d_zZZ has quit IRC21:03
*** winston-d_zZZ has joined #openstack-cinder21:03
*** kambiz has joined #openstack-cinder21:03
*** garthb_ has joined #openstack-cinder21:03
*** jdandrea has joined #openstack-cinder21:05
*** r-daneel has quit IRC21:05
*** garthb_ has quit IRC21:06
openstackgerritDaniel Tadrzak proposed openstack/cinder: CGSnapshot Object
jdandreaI seem to recall a presentation that illustrates how to use the QoS features in cinder. Is anyone aware of this and, if so, where to locate it? (Searching online. Found a NetApp presentation so far. Looking for others.)21:06
*** garthb_ has joined #openstack-cinder21:06
openstackgerritDaniel Tadrzak proposed openstack/cinder: Cleanup for cinder tests with CGSnapshot
*** garthb has quit IRC21:06
*** sgotliv has joined #openstack-cinder21:07
*** dflorea has quit IRC21:10
*** crose has quit IRC21:11
*** daneyon_ has joined #openstack-cinder21:16
*** daneyon has quit IRC21:19
*** edtubill has left #openstack-cinder21:20
*** rushil_ has quit IRC21:22
*** timcl has quit IRC21:23
jungleboyjhemna: You there?21:24
*** agarciam has quit IRC21:25
jungleboyjkmartin: You around?21:28
*** IanGovett has quit IRC21:29
kmartinjungleboyj, yeah21:30
jungleboyjkmartin: Hey man, I have a silly FC related question.  Guess I could have pinged angela-s as well.21:31
jungleboyjkmartin: Where is it that the lookup service gets its credentials from for logging into the fabric?  I can't find that in the code for some reason.21:32
kmartinjungleboyj, sure, is it how to configure the IBM FC driver?21:32
jungleboyjkmartin: No ... How do you guys set up the creds to talk to the FC switch ?21:33
kmartinin cinder.conf21:33
kmartinjungleboyj, just for you
kmartinjungleboyj, page 11 fc_fabric_user = admin21:35
kmartin fc_fabric_password = password21:35
jungleboyjkmartin: Thank you.  Do they not use SSH keys at all?21:36
*** jamielennox has quit IRC21:36
jungleboyjkmartin: Only username/password?21:37
*** jaypipes has quit IRC21:37
kmartinjungleboyj, correct21:37
jungleboyjkmartin: That was the piece I was missing.  Thank you!21:37
*** esker has joined #openstack-cinder21:37
jungleboyjI had assumed that they would also support SSH keys and couldn't figure out why I couldn't find a place to define that.21:38
*** jamielennox has joined #openstack-cinder21:38
kmartinyeah, angela-s was talking about using the REST interface on the switch and getting rid of SSH altogether, but not sure where that ended up21:39
*** alejandrito has joined #openstack-cinder21:39
*** mriedem has quit IRC21:40
openstackgerritVictor Stinner proposed openstack/cinder: Port test_volume to Python 3
*** esker has quit IRC21:46
*** esker has joined #openstack-cinder21:46
*** krtaylor has quit IRC21:48
*** diablo_rojo has quit IRC21:49
*** alejandrito has quit IRC21:51
*** jaypipes has joined #openstack-cinder21:52
*** eharney has quit IRC21:53
*** jungleboyj has quit IRC21:55
*** esker has quit IRC21:56
*** dims_ has joined #openstack-cinder21:56
openstackgerritVictor Stinner proposed openstack/cinder: Port volume transfer to Python 3
*** edmondsw has quit IRC21:59
*** dims has quit IRC21:59
*** diogogmt has quit IRC22:08
*** sidbhatt13 has quit IRC22:09
*** merooney has joined #openstack-cinder22:12
*** merooney has joined #openstack-cinder22:12
*** merooney has quit IRC22:13
openstackgerritVictor Stinner proposed openstack/cinder: py3: Port pure driver test to Python 3
*** sgotliv has quit IRC22:19
*** dims_ has quit IRC22:24
*** dims has joined #openstack-cinder22:24
openstackgerritxing-yang proposed openstack/cinder: ScaleIO driver should use os-brick connector
*** jwcroppe has quit IRC22:30
openstackgerritMitsuhiro Tanino proposed openstack/cinder: Adds framework for get_capabilities() feature
*** gouthamr has quit IRC22:32
*** jaypipes has quit IRC22:33
*** dims has quit IRC22:34
openstackgerritVictor Stinner proposed openstack/cinder: Fix Python 3 issues in wsgi
rhagarty__xyang1: hello - regarding consistency groups... is there any way to determine the assigned volume type that is set when the CG is created?22:35
*** hemna is now known as hemnafk22:36
openstackgerritMitsuhiro Tanino proposed openstack/cinder: Allow specified backend capabilities to be retrieved
openstackgerritMitsuhiro Tanino proposed openstack/cinder: Efficient volume copy for generic volume migration
*** krtaylor has joined #openstack-cinder22:39
openstackgerritPatrick East proposed openstack/cinder: Generic image-volume cache
*** markvoelker has quit IRC22:46
*** thangp has quit IRC22:46
*** dflorea has joined #openstack-cinder22:46
*** gouthamr has joined #openstack-cinder22:53
*** markvoelker has joined #openstack-cinder22:54
*** dflorea has quit IRC23:06
*** dflorea has joined #openstack-cinder23:07
*** dflorea has quit IRC23:07
*** diegows has quit IRC23:10
angela-sjungleboyj: hey, last year our team added support for SSH key, but looks like we never added it to the upstream driver.  let me find out what happened and get back to you23:14
angela-skmartin: we have a HTTPS connector in the VF code review:
*** esker has joined #openstack-cinder23:16
*** BharatK has quit IRC23:18
kmartinangela-s, thanks, I knew I had seen it before.23:18
*** bkopilov has quit IRC23:19
*** arif-ali has quit IRC23:19
*** bkopilov has joined #openstack-cinder23:19
*** arif-ali has joined #openstack-cinder23:19
angela-sjungleboyj: the SSH key param addition is also in the VF code review ^^23:19
*** esker has quit IRC23:22
*** diegows has joined #openstack-cinder23:22
*** esker has joined #openstack-cinder23:23
*** diegows has quit IRC23:23
kmartinangela-s, on launchpad defect it seems if someone has attached volumes w/o FCZM then turns on FCZM any detach will fail? So would the process be detach all volumes from host that were pre-zoned, delete the pre-defined zone from the fabric, then enable FCZM and reattach all volumes?23:25
openstackLaunchpad bug 1486613 in Cinder "Existing volume detach fails after modifying the zoning_mode=fabric from none" [Undecided,Invalid]23:25
*** BharatK has joined #openstack-cinder23:30
angela-skmartin: it's drastic, but seems like the only option. Unless by some miracle the pre-defined zone names were formatted in the same way that we do for Openstack...23:30
angela-skmartin: to support this scenario, FCZM would have to be reworked to not look up by zone name, but by zone members, which would increase the time required to process the call, somewhat drastically i think, as we'd have to process the entire defined zone database23:32
kmartinangela-s, yeah it's pretty ugly but these seems like a valid use case23:32
kmartinI'll bring it up in the FC meeting tomorrow23:33
angela-ssure, good venue to discuss this item23:34
*** mriedem has joined #openstack-cinder23:36
kmartinangela-s, I'll add a comment to the defect with the above.23:38
angela-skmartin: ok, thanks!23:40
tbarronso I'm seeing cinder.tests.unit.test_vmware_vmdk.VMwareVcVmdkDriverTestCase.test_create_backing_from_stream_optimized_file fail on a bunch of unrelated changes.23:45
tbarronDoes that ring a bell with anyone?23:45
*** martyturner has joined #openstack-cinder23:53
*** zhangjn has joined #openstack-cinder23:55
patrickeasttbarron: i was seeing that on the image cache patchsets too23:56
patrickeasttbarron: couldn't reproduce it locally on my machine though23:56
patrickeasttbarron: i spent some time looking into what the test is doing... but couldn't really figure out what the problem is23:57
*** asd112z has quit IRC23:57
patrickeastthe weird part is why a unit test would be reading from a file23:58

Generated by 2.14.0 by Marius Gedminas - find it at!