Monday, 2015-10-05

*** gouthamr has quit IRC00:01
*** julim has joined #openstack-cinder00:09
*** dave-mccowan has quit IRC00:12
*** dave-mccowan has joined #openstack-cinder00:15
*** dims_ has quit IRC00:17
*** wilson1 has joined #openstack-cinder00:26
*** wilson-1 has quit IRC00:29
*** bswartz has quit IRC01:01
*** abehl has quit IRC01:02
*** bswartz has joined #openstack-cinder01:03
*** david-lyle has quit IRC01:13
*** lcurtis has joined #openstack-cinder01:21
*** julim has quit IRC01:22
*** Lee1092 has joined #openstack-cinder01:31
*** ronis has quit IRC01:31
*** david-lyle has joined #openstack-cinder01:44
openstackgerritxing-yang proposed openstack/cinder: CG driver function should not access db  https://review.openstack.org/22958001:58
*** gouthamr has joined #openstack-cinder02:07
*** gouthamr_ has quit IRC02:10
*** gouthamr has quit IRC02:12
openstackgerritxing-yang proposed openstack/cinder: VMAX Target iSCSI IP Address  https://review.openstack.org/22971202:13
*** gouthamr has joined #openstack-cinder02:20
*** bswartz has quit IRC02:20
*** dimsum__ has joined #openstack-cinder02:21
*** gouthamr has quit IRC02:30
openstackgerritZhang Jinnan proposed openstack/cinder: Volume extend error not catch exception  https://review.openstack.org/22797102:32
*** bswartz has joined #openstack-cinder02:39
*** dimsum__ has quit IRC02:40
*** lcurtis has quit IRC02:42
*** asselin__ has quit IRC02:51
*** julim has joined #openstack-cinder03:03
*** dave-mccowan has quit IRC03:07
*** julim has quit IRC03:09
*** david-lyle has quit IRC03:32
*** smoriya has joined #openstack-cinder03:35
*** asselin_ has joined #openstack-cinder03:43
*** david-lyle has joined #openstack-cinder03:43
*** david-lyle has quit IRC03:48
*** haomaiwang has joined #openstack-cinder03:52
*** david-lyle has joined #openstack-cinder03:53
*** haomaiwang has quit IRC04:01
*** haomaiwang has joined #openstack-cinder04:01
*** david-lyle has quit IRC04:02
*** david-lyle has joined #openstack-cinder04:06
*** david-lyle has quit IRC04:12
*** dimsum__ has joined #openstack-cinder04:21
*** david-lyle has joined #openstack-cinder04:22
*** amit213 has quit IRC04:28
*** amit213 has joined #openstack-cinder04:28
*** david-lyle has quit IRC04:29
*** edtubill has joined #openstack-cinder04:32
*** fifieldt has joined #openstack-cinder04:46
*** merooney has joined #openstack-cinder04:49
*** merooeny_ has joined #openstack-cinder04:51
*** merooney has quit IRC04:54
*** haomaiwang has quit IRC05:01
*** haomaiwang has joined #openstack-cinder05:01
*** merooeny_ has quit IRC05:03
*** pradipta has joined #openstack-cinder05:09
*** stevemar_ has quit IRC05:20
*** lpetrut has joined #openstack-cinder05:23
*** shausy has joined #openstack-cinder05:34
*** BharatK has joined #openstack-cinder05:34
*** edtubill has quit IRC05:34
*** nkrinner has joined #openstack-cinder05:35
*** et10man has joined #openstack-cinder05:46
*** lpetrut has quit IRC05:47
openstackgerritwatanabe isao proposed openstack/cinder: Don't merge  https://review.openstack.org/23088005:51
*** BharatK has quit IRC05:53
*** lpetrut has joined #openstack-cinder05:58
*** haomaiwang has quit IRC06:01
*** haomaiwang has joined #openstack-cinder06:01
*** ankit_ag has joined #openstack-cinder06:09
*** BharatK has joined #openstack-cinder06:09
*** deepakcs has joined #openstack-cinder06:11
*** lpetrut has quit IRC06:14
*** ishant has joined #openstack-cinder06:19
*** ankit_ag has quit IRC06:24
*** deepakcs has quit IRC06:26
*** aarefiev_ has quit IRC06:28
*** aarefiev has joined #openstack-cinder06:30
*** vlaza has joined #openstack-cinder06:31
*** links has joined #openstack-cinder06:35
*** vgridnev has joined #openstack-cinder06:37
*** deepakcs has joined #openstack-cinder06:41
*** alexschm has joined #openstack-cinder06:45
*** haomaiwang has quit IRC07:01
*** haomaiwang has joined #openstack-cinder07:01
openstackgerritxiaoqin proposed openstack/cinder: Add option for user to disable fast format at thick-provisioned volume creation in IBM  Storwize cinder  https://review.openstack.org/23082507:14
openstackgerritxiaoqin proposed openstack/cinder: Add option to disable fast format for thick volume creation  https://review.openstack.org/23082507:26
openstackgerritxiaoqin proposed openstack/cinder: Add option to disable fast format for thick volume creation  https://review.openstack.org/23082507:26
*** fifieldt has quit IRC07:28
*** geguileo has joined #openstack-cinder07:39
openstackgerritSergey Gotliv proposed openstack/python-cinderclient: Adding pagination to snapshots and backups lists  https://review.openstack.org/22558207:40
*** aix has quit IRC07:45
*** aix has joined #openstack-cinder07:45
*** jordanP has joined #openstack-cinder07:48
openstackgerritoliver-leahy-l proposed openstack/cinder: encryption_api_url requires a version  https://review.openstack.org/23003107:49
*** et10man has quit IRC07:52
*** haomaiwang has quit IRC08:01
*** haomaiwang has joined #openstack-cinder08:01
*** arif-ali has quit IRC08:05
*** arif-ali has joined #openstack-cinder08:09
*** e0ne has joined #openstack-cinder08:10
*** arif-ali has quit IRC08:16
*** arif-ali has joined #openstack-cinder08:19
*** nkrinner_ has joined #openstack-cinder08:22
*** lpetrut has joined #openstack-cinder08:22
*** nkrinner_ has quit IRC08:23
*** david-lyle has joined #openstack-cinder08:28
*** arif-ali has quit IRC08:34
*** jistr has joined #openstack-cinder08:38
*** arif-ali has joined #openstack-cinder08:39
*** arif-ali has quit IRC08:47
*** david-lyle has quit IRC08:51
*** arif-ali has joined #openstack-cinder08:51
*** ankit_ag has joined #openstack-cinder08:58
openstackgerritxiaoqin proposed openstack/cinder: Add fast format option for thick volume creation  https://review.openstack.org/23082508:58
*** jistr has quit IRC09:00
*** david-lyle has joined #openstack-cinder09:00
*** haomaiwang has quit IRC09:01
*** haomaiwang has joined #openstack-cinder09:01
*** jistr has joined #openstack-cinder09:01
*** BharatK has quit IRC09:01
*** markvoelker has joined #openstack-cinder09:05
*** markvoelker_ has joined #openstack-cinder09:06
*** david-lyle has quit IRC09:07
*** abehl has joined #openstack-cinder09:08
*** markvoelker has quit IRC09:09
*** BharatK has joined #openstack-cinder09:13
*** ronis has joined #openstack-cinder09:16
*** vgridnev has quit IRC09:19
*** vgridnev_ has joined #openstack-cinder09:19
openstackgerritVictor Stinner proposed openstack/cinder: emc vmax driver: use integer division for Python 3  https://review.openstack.org/22999809:20
*** markvoelker_ has quit IRC09:20
*** david-lyle has joined #openstack-cinder09:23
*** david-lyle has quit IRC09:27
*** david-lyle has joined #openstack-cinder09:30
*** jordan_ has joined #openstack-cinder09:35
*** ankit_ag has quit IRC09:36
*** jordanP has quit IRC09:39
*** ociuhandu has joined #openstack-cinder09:44
*** asselin__ has joined #openstack-cinder09:44
openstackgerritWilson Liu proposed openstack/cinder: Fix capacity report error in Huawei driver  https://review.openstack.org/22630809:44
*** asselin_ has quit IRC09:48
openstackgerritWilson Liu proposed openstack/cinder: Encode username and password in Huawei driver  https://review.openstack.org/21608509:52
*** david-lyle has quit IRC09:54
*** haomaiwang has quit IRC10:01
*** david-lyle has joined #openstack-cinder10:01
*** haomaiwang has joined #openstack-cinder10:01
*** e0ne has quit IRC10:10
*** e0ne has joined #openstack-cinder10:14
*** IanGovett has joined #openstack-cinder10:14
*** david-lyle has quit IRC10:16
*** david-lyle has joined #openstack-cinder10:19
*** IanGovett has quit IRC10:22
*** shausy has quit IRC10:27
*** shausy has joined #openstack-cinder10:27
*** david-lyle has quit IRC10:29
openstackgerritSzymon Wr√≥blewski proposed openstack/cinder: CGSnapshot Object  https://review.openstack.org/19502710:29
openstackgerritYuriy Nesenenko proposed openstack/cinder: Implement snapshots-related features for Block Device Driver  https://review.openstack.org/22229210:30
*** david-lyle has joined #openstack-cinder10:32
*** IanGovett has joined #openstack-cinder10:41
*** david-lyle has quit IRC10:42
*** david-lyle has joined #openstack-cinder10:49
*** BharatK has quit IRC10:53
*** david-lyle has quit IRC10:53
*** david-lyle has joined #openstack-cinder10:56
*** smoriya has quit IRC10:57
*** haomaiwang has quit IRC11:01
*** haomaiwang has joined #openstack-cinder11:01
*** aix has quit IRC11:03
*** haomaiwang has quit IRC11:05
*** haomaiwa_ has joined #openstack-cinder11:05
*** BharatK has joined #openstack-cinder11:06
openstackgerritSzymon Wr√≥blewski proposed openstack/cinder: Cleanup for cinder tests with CGSnapshot  https://review.openstack.org/20786611:07
*** david-lyle has quit IRC11:07
*** david-lyle has joined #openstack-cinder11:08
*** haomaiwa_ has quit IRC11:09
*** jordan_ has quit IRC11:14
*** Yogi1 has joined #openstack-cinder11:18
*** david-lyle has quit IRC11:18
haypohello. any progress on the failures of the python 3 gate?11:24
*** david-lyle has joined #openstack-cinder11:25
*** timcl has joined #openstack-cinder11:25
*** jordan_ has joined #openstack-cinder11:27
*** aix has joined #openstack-cinder11:35
*** kevincarr1991 has joined #openstack-cinder12:00
*** e0ne has quit IRC12:01
*** bswartz has quit IRC12:05
*** sb has quit IRC12:08
*** strictlyb has joined #openstack-cinder12:08
*** e0ne has joined #openstack-cinder12:13
*** dzamboni has joined #openstack-cinder12:14
e0nehaypo: hi. afaik, no:(12:16
*** raildo-afk is now known as raildo12:16
e0nehaypo: I'm going to take a look on it again a bit later12:16
*** BharatK has quit IRC12:18
*** Yogi1 has quit IRC12:19
*** bill_az has joined #openstack-cinder12:22
openstackgerritIvan Kolodyazhny proposed openstack/cinder: Convert  Retry-After header parameter value to string  https://review.openstack.org/23097712:25
*** martyturner has joined #openstack-cinder12:27
*** bill_az has quit IRC12:28
*** ishant has quit IRC12:29
*** diablo_rojo has joined #openstack-cinder12:32
haypoe0ne: any clue? last time i checked, i was still unable to reproduce, even with "tox -e py34" on a fresh clone12:33
haypoe0ne: should i have to run an external service like a MySQL server, RabbitMQ server or something else?12:33
e0nehaypo: I'm checking it with a frech tox env12:33
*** lpetrut has quit IRC12:35
kevincarr1991Good Morning! I have cinder installed on my controller node and I have cinder using the NFS drivers. I am able to create volumes from Horizon, but when I do they have improper permissions which restricts me from being able to attach the volumes to instances.12:37
kevincarr1991Has anyone has this issue or something like it before?12:37
*** edmondsw has joined #openstack-cinder12:37
*** dave-mccowan has joined #openstack-cinder12:38
*** bswartz has joined #openstack-cinder12:40
*** gouthamr has joined #openstack-cinder12:43
*** Lee1092 has quit IRC12:43
*** takedakn has joined #openstack-cinder12:45
*** porrua has joined #openstack-cinder12:48
e0nehaypo: what is your's concurrency param value in test run?12:50
e0nehaypo: "By default ostestr will set concurrency to ncpu, to specify something else use the concurrency=<n> option call ie: 'tox -epy27 -- --concurrency=4'"12:50
*** dustins has joined #openstack-cinder12:52
*** kevincarr1991 has quit IRC12:53
*** marcusvrn_ has joined #openstack-cinder12:53
*** akerr has joined #openstack-cinder12:53
*** julim has joined #openstack-cinder12:56
*** Yogi1 has joined #openstack-cinder12:57
*** akerr has quit IRC12:57
*** akerr has joined #openstack-cinder12:58
*** takedakn has quit IRC13:03
e0nehaypo: I've got something strange on my env. cinder.tests.unit.test_cmd failed even with python2713:03
*** dimsum__ is now known as dims13:08
*** julim_ has joined #openstack-cinder13:09
*** jaypipes has joined #openstack-cinder13:09
*** julim has quit IRC13:12
diablo_rojosmcginnis: DuncanT hemna_ When you all get a sec, would you re review https://review.openstack.org/#/c/223375/ Eric has voiced some concerns with the approach and I am curious to know if you agree or if you think this is an acceptable approach. Thanks!13:14
*** willsama has joined #openstack-cinder13:19
*** changbl has quit IRC13:22
*** links has quit IRC13:22
*** bill_az has joined #openstack-cinder13:23
*** chlong has joined #openstack-cinder13:26
*** lpetrut has joined #openstack-cinder13:27
*** erhudy has joined #openstack-cinder13:32
*** jasondotstar|afk is now known as jasondotstar13:32
*** superdan is now known as dansmith13:34
tbarronkevincarr1991: permissions on NFS exported filesystems are a frequent issue, quite apart form cinder and nova.13:36
tbarronkevincarr1991: you need to check how you are doing exports, mounts, whether root is squashed, etc.13:37
*** bill_az has quit IRC13:37
tbarronkevincarr1991: as I've suggested when you asked about other issues with NFS, determine the specific commands being run13:38
*** diogogmt has quit IRC13:39
tbarronkevincarr1991: by looking at logs (in this case nova cpu log).  Run them by hand while logged in as the nova user, after unmounting,, re-exporting from the NFS server, remounting, etd.13:39
tbarronetc.13:39
tbarronkevincarr1991: cinder and nova do not control how your NFS filesystem is exported, and that (plus any nfs mount options that you supply)13:40
tbarronkevincarr1991: is what determines permissions for the cinder user or nova user when they interact with that filesystem.13:41
*** deepakcs has quit IRC13:41
*** jungleboyj has joined #openstack-cinder13:41
*** xyang1 has joined #openstack-cinder13:45
*** mc_nair has joined #openstack-cinder13:45
*** vgridnev_ has quit IRC13:52
openstackgerritGorka Eguileor proposed openstack/cinder: Detach volume on device validation failure  https://review.openstack.org/23042913:52
*** haomaiwang has joined #openstack-cinder13:54
*** bswartz has quit IRC13:55
*** openstackstatus has joined #openstack-cinder13:56
*** ChanServ sets mode: +v openstackstatus13:56
*** mriedem has joined #openstack-cinder13:56
*** diablo_rojo has quit IRC13:57
*** diablo_rojo has joined #openstack-cinder13:58
*** jungleboyj has quit IRC14:00
*** haomaiwang has quit IRC14:01
*** dave-mccowan has quit IRC14:01
*** stevemar_ has joined #openstack-cinder14:01
*** links has joined #openstack-cinder14:05
xyang1smcginnis: ping14:05
*** haomaiwa_ has joined #openstack-cinder14:06
smcginnisxyang1: Hey!14:07
xyang1smcginnis: I was trying to backport, but got some issues14:08
xyang1smcginnis: did you get my email14:08
smcginnisxyang1: Just looking at it now.14:08
*** xyang has joined #openstack-cinder14:08
smcginnisxyang1: Not sure why that would be.14:08
*** xyang has quit IRC14:08
*** shausy has quit IRC14:09
xyang1smcginnis: very odd, the patch looks clean14:09
smcginnisxyang1: Yeah, I thought it looked ok. Hmm...14:09
xyang1smcginnis: but the commit id does not bring you to the original14:09
xyang1smcginnis: i wonder if I can just remove wip14:10
xyang1smcginnis: should I manually change the commit msg14:10
*** changbl has joined #openstack-cinder14:10
smcginnisxyang1: I suppose. Unless someone else here has more experience eith cherry pick issues, that might have to be what to do.14:11
smcginnisxyang1: I know jungleboyj has had to do quite a few. Maybe wait until he's online and check with him.14:11
xyang1smcginnis: ok, I'll update the commit msg then14:11
*** thangp has joined #openstack-cinder14:11
smcginnisUnless someone else chimes in.14:11
xyang1smcginnis: sure14:11
smcginnisxyang1: We've got a little time yet.14:12
xyang1smcginnis: when do you have to make the cut14:12
*** jgregor has joined #openstack-cinder14:13
smcginnisxyang1: I think Wednesday at the latest.14:14
*** dave-mccowan has joined #openstack-cinder14:14
*** mdenny has joined #openstack-cinder14:14
xyang1smcginnis: ok, thanks14:14
smcginnisxyang1: Do you know - are there any string changes in your patch? Translations could be an issue.14:14
*** jungleboyj has joined #openstack-cinder14:15
xyang1smcginnis: good point, let me check14:15
xyang1smcginnis: there is new warn msg14:16
*** raildo is now known as raildo-afk14:16
smcginnisxyang1: I think that's fine.14:16
smcginnisxyang1: But that probably means we should get it in sooner rather than later to let the translators have some time to translate it.14:16
*** nkrinner has quit IRC14:17
*** Lee1092 has joined #openstack-cinder14:17
xyang1smcginnis: i can change to debug if necessary14:17
smcginnisxyang1: OK. Let's wait a bit and see.14:19
*** lcurtis has joined #openstack-cinder14:19
xyang1smcginnis: sure.  let me add some comments there so people know what issue I ran into14:19
smcginnisxyang1: Good plan.14:19
*** merooney has joined #openstack-cinder14:20
*** salv-orl_ has joined #openstack-cinder14:22
*** eharney has joined #openstack-cinder14:23
*** salv-orlando has quit IRC14:24
*** raildo-afk is now known as raildo14:25
*** lcurtis has quit IRC14:25
*** lcurtis has joined #openstack-cinder14:26
*** mtanino has joined #openstack-cinder14:26
*** jgregor has quit IRC14:29
xyang1smcginnis: I added comments there to explain the problem:  https://review.openstack.org/#/c/230861/14:30
smcginnisxyang1: +114:31
xyang1smcginnis: thanks14:31
*** vivekd has joined #openstack-cinder14:32
*** vlaza has left #openstack-cinder14:35
*** changbl has quit IRC14:36
*** chlong has quit IRC14:38
*** garthb has joined #openstack-cinder14:38
*** CBADER has joined #openstack-cinder14:39
*** diogogmt has joined #openstack-cinder14:42
smcginnisxyang1: Looks like eharney knows what's going on (as usual). :)14:46
*** lpetrut has quit IRC14:46
xyang1smcginnis: let me see what he says:)14:46
xyang1eharney: thanks!  Let me try that14:47
*** baumann has joined #openstack-cinder14:49
*** edtubill has joined #openstack-cinder14:50
*** tsekiyama has joined #openstack-cinder14:50
*** jgregor has joined #openstack-cinder14:51
*** EmilienM has quit IRC14:55
*** EmilienM has joined #openstack-cinder14:55
*** ntpttr has joined #openstack-cinder14:55
*** david-ly_ has joined #openstack-cinder14:56
*** david-lyle has quit IRC14:57
*** stevemar_ has quit IRC14:57
*** david-ly_ is now known as david-lyle14:57
*** stevemar_ has joined #openstack-cinder14:57
*** haomaiwa_ has quit IRC15:01
*** haomaiwang has joined #openstack-cinder15:01
*** stevemar_ has quit IRC15:02
*** aix has quit IRC15:03
*** david-ly_ has joined #openstack-cinder15:03
*** david-lyle has quit IRC15:03
*** harlowja_at_home has joined #openstack-cinder15:04
*** david-ly_ is now known as david-lyle15:04
*** vivekd has quit IRC15:05
*** asselin_ has joined #openstack-cinder15:09
*** jungleboyj has quit IRC15:09
*** setmason has joined #openstack-cinder15:10
*** asselin__ has quit IRC15:12
*** diogogmt_ has joined #openstack-cinder15:13
*** shausy has joined #openstack-cinder15:13
*** diogogmt has quit IRC15:13
*** diogogmt_ is now known as diogogmt15:13
*** hemna_ is now known as hemna15:17
*** stevemar_ has joined #openstack-cinder15:18
*** thingee has joined #openstack-cinder15:19
*** alejandrito has joined #openstack-cinder15:23
hemnadiablo_rojo, so I'm fine with the current approach.  I think adding a cinder.utils.register wrapper is probably not something we want to do.  It could lead to other issues15:25
hemnadiablo_rojo, I do think Gorka has some good points in his review that need addressing fwiw.15:25
*** mc_nair has quit IRC15:28
*** lcurtis has quit IRC15:29
*** jungleboyj has joined #openstack-cinder15:29
*** merooney has quit IRC15:29
diablo_rojohemna: Yeah I just saw Gorka's comments and should have a new patch up as soon as pep8 passes :)15:30
diablo_rojohemna: Thanks for looking at it. I will give you a heads up when I get the new patch up.15:30
hemnaok coolio15:30
*** jdurgin1 has joined #openstack-cinder15:30
*** sparr has quit IRC15:32
*** tbarron has quit IRC15:32
*** dsariel_ has joined #openstack-cinder15:32
*** asselin__ has joined #openstack-cinder15:32
*** tbarron has joined #openstack-cinder15:33
*** sparr has joined #openstack-cinder15:34
*** changbl has joined #openstack-cinder15:35
*** asselin_ has quit IRC15:36
*** Guest43458 is now known as med_15:36
*** med_ has quit IRC15:36
*** med_ has joined #openstack-cinder15:36
*** dustins is now known as dustins|afk15:36
*** med_ is now known as med15:37
*** med is now known as med_15:38
*** dims has quit IRC15:40
*** Yogi1 has quit IRC15:40
*** mc_nair has joined #openstack-cinder15:42
*** dims has joined #openstack-cinder15:47
*** dims has quit IRC15:47
*** Guest96374 has joined #openstack-cinder15:47
*** Guest96374 has quit IRC15:48
*** david-ly_ has joined #openstack-cinder15:48
*** david-lyle has quit IRC15:48
*** dims_ has joined #openstack-cinder15:48
*** david-ly_ is now known as david-lyle15:49
*** mriedem has quit IRC15:50
*** haomaiwang has quit IRC15:50
*** haomaiwang has joined #openstack-cinder15:51
*** alexschm has quit IRC15:52
*** alejandrito has quit IRC15:55
*** shausy has quit IRC15:59
*** haomaiwang has quit IRC16:01
*** haomaiwang has joined #openstack-cinder16:01
openstackgerritKendall Nelson proposed openstack/cinder: Hacking check for opt name registration  https://review.openstack.org/22337516:03
*** hemna_ has joined #openstack-cinder16:04
*** haypo has left #openstack-cinder16:04
diablo_rojohemna geguileo ^^16:04
geguileodiablo_rojo: Ok, reviewing16:05
diablo_rojogeguileo: Thank you!16:06
*** CBADER has quit IRC16:07
*** cbader has joined #openstack-cinder16:07
*** akerr is now known as akerr_away16:07
*** raildo is now known as raildo-afk16:10
*** baumann1 has joined #openstack-cinder16:11
*** bnemec has joined #openstack-cinder16:11
*** jdurgin1 has quit IRC16:12
*** baumann has quit IRC16:12
*** raildo-afk is now known as raildo16:12
*** chris_morrell has joined #openstack-cinder16:13
*** mriedem has joined #openstack-cinder16:13
*** _cjones_ has joined #openstack-cinder16:14
SwansonAnyone know when (if) there is going to be a kilo bug fix release?16:14
*** alejandrito has joined #openstack-cinder16:16
*** hemna_ has quit IRC16:16
*** stevemar_ has quit IRC16:17
*** leeantho has joined #openstack-cinder16:17
*** angela-s has joined #openstack-cinder16:17
*** stevemar_ has joined #openstack-cinder16:18
*** pradipta has quit IRC16:22
*** jistr has quit IRC16:26
*** e0ne has quit IRC16:28
*** dustins|afk has quit IRC16:28
*** timcl has quit IRC16:31
*** harlowja_at_home has quit IRC16:33
*** alejandrito has quit IRC16:37
*** raildo is now known as raildo-afk16:38
*** raildo-afk is now known as raildo16:41
*** salv-orl_ has quit IRC16:42
*** apoorvad has joined #openstack-cinder16:44
*** haomaiw__ has joined #openstack-cinder16:45
*** haomaiwang has quit IRC16:46
*** crose has joined #openstack-cinder16:53
*** baumann1 has quit IRC17:00
openstackgerritEric Harney proposed openstack/cinder: Fix broken format string in vol mgr log  https://review.openstack.org/23109317:00
*** haomaiw__ has quit IRC17:01
*** 32NAAH197 has joined #openstack-cinder17:01
hemnalots of CI failures17:02
hemnableh17:02
smcginnishemna: I've still been get some residual stacking failures.17:05
smcginnisMostly shaken out, but some occasional ones yet.17:05
hemnasmcginnis, I was just looking at this guy https://review.openstack.org/#/c/22337517:05
hemnaand seeing the initial list of failures right away17:05
smcginnisDang17:05
hemnahaven't looked into them to see if the patch itself is causing it or not17:05
hemnajust seems odd17:06
*** Yogi1 has joined #openstack-cinder17:07
*** merooney has joined #openstack-cinder17:07
*** subscope has joined #openstack-cinder17:09
*** bswartz has joined #openstack-cinder17:10
*** timcl has joined #openstack-cinder17:10
*** aix has joined #openstack-cinder17:11
*** lcurtis has joined #openstack-cinder17:13
*** tsekiyama has quit IRC17:13
*** akerr_away is now known as akerr17:15
*** e0ne has joined #openstack-cinder17:17
*** markstur has joined #openstack-cinder17:17
*** raildo is now known as raildo-afk17:18
*** stevemar_ has quit IRC17:21
*** kjelly has quit IRC17:24
*** kjelly has joined #openstack-cinder17:25
*** earlephilhower has joined #openstack-cinder17:28
*** raildo-afk is now known as raildo17:30
*** ociuhandu has quit IRC17:30
*** openstackgerrit has quit IRC17:31
*** openstackgerrit has joined #openstack-cinder17:32
*** alejandrito has joined #openstack-cinder17:34
*** stevemar_ has joined #openstack-cinder17:36
*** alejandrito has quit IRC17:38
*** salv-orlando has joined #openstack-cinder17:42
*** 32NAAH197 has quit IRC17:44
*** salv-orl_ has joined #openstack-cinder17:45
*** ociuhandu has joined #openstack-cinder17:46
xyang1smcginnis: eharney jungleboyj : I re-submitted the backport patch: https://review.openstack.org/#/c/231111/17:46
*** salv-orlando has quit IRC17:46
*** raildo is now known as raildo-afk17:48
*** timcl has quit IRC17:54
*** crose has quit IRC17:57
*** ccesario_ has quit IRC17:58
smcginnisxyang1: Great, thanks. Just waiting for CI.17:58
xyang1smcginnis: thanks! We are only running CI against the master though17:59
smcginnisxyang1: Oh yeah, I just meant Jenkins check.18:00
xyang1smcginnis: Ok, yes18:00
smcginnis:)18:00
*** openstackgerrit has quit IRC18:01
*** openstackgerrit has joined #openstack-cinder18:02
*** dsariel_ has quit IRC18:02
*** apoorvad_ has joined #openstack-cinder18:02
*** timcl has joined #openstack-cinder18:03
openstackgerritEric Harney proposed openstack/cinder: Do not merge: Debugging bug 1501745  https://review.openstack.org/23112218:03
openstackbug 1501745 in Cinder "gate-cinder-python34 failed" [High,Confirmed] https://launchpad.net/bugs/1501745 - Assigned to Eric Harney (eharney)18:03
*** jaypipes has quit IRC18:03
*** apoorvad has quit IRC18:05
ntpttrHey everyone, I'm pretty fresh at contributing to Cinder, and I'm wondering if there are any resources people would recommend for learning the architecture of the project and how to get to know my way around the code. Is the wiki the best place? Is it good to just look around the github repo for a while feeling your way around? Thanks18:06
*** merooney has quit IRC18:12
*** aix has quit IRC18:14
*** raildo-afk is now known as raildo18:14
*** tsekiyama has joined #openstack-cinder18:14
*** dustins has joined #openstack-cinder18:19
*** salv-orl_ has quit IRC18:25
scottdantpttr: One way to learn is to follow what's going on here, in cinder-irc...18:26
scottdaclick on some of the links ^^^^ and read the code.18:26
scottdaIf someone is discussing something you don't understand, grep  the code, look in google, or ask here in IRC18:26
ntpttrscottda: Sounds good, thanks you18:27
*** dims_ has quit IRC18:27
scottdasure. Welcome to Cinder18:27
ntpttrscottda: Thanks :)18:28
scottdaWe've plenty of bugs you could look at: https://bugs.launchpad.net/cinder/+bugs?orderby=status&start=018:29
*** abehl has quit IRC18:31
*** salv-orlando has joined #openstack-cinder18:32
*** IanGovett has quit IRC18:32
openstackgerritGorka Eguileor proposed openstack/os-brick: Fix iSCSI multipath cleanup  https://review.openstack.org/23113418:35
openstackgerritGorka Eguileor proposed openstack/os-brick: Fix FC multipath cleanup  https://review.openstack.org/23113518:36
*** IanGovett has joined #openstack-cinder18:38
*** links has quit IRC18:44
*** raildo is now known as raildo-afk18:45
*** raildo-afk is now known as raildo18:45
*** diegows has joined #openstack-cinder18:48
diablo_rojogeguileo: I see you just +2ed my patch which I greatly appreciate. I had just gotten done addressing your replies and was going to push a new patch if you'd like. Otherwise we can just leave it.18:49
diablo_rojogeguileo: Thoughts?18:49
geguileodiablo_rojo: If you've already done the work I'd hate it go to waste18:50
*** stevemar_ has quit IRC18:50
geguileodiablo_rojo: Just push it and I'll review it before closing for shop for today18:50
*** stevemar_ has joined #openstack-cinder18:50
diablo_rojogeguileo: Ha okay I will push it now. One sec.18:51
geguileodiablo_rojo: No rush18:51
*** merooney has joined #openstack-cinder18:51
openstackgerritKendall Nelson proposed openstack/cinder: Hacking check for opt name registration  https://review.openstack.org/22337518:52
openstackgerritIvan Kolodyazhny proposed openstack/cinder-specs: Remove hardcoded releases list from unit tests  https://review.openstack.org/21822618:54
*** stevemar_ has quit IRC18:55
*** kevincarr1991 has joined #openstack-cinder18:56
*** dims_ has joined #openstack-cinder18:58
*** dims_ has quit IRC18:58
*** dims_ has joined #openstack-cinder18:59
*** apoorvad_ has quit IRC19:01
*** apoorvad has joined #openstack-cinder19:01
*** dzamboni has quit IRC19:03
*** dims_ has quit IRC19:03
*** akerr has quit IRC19:05
*** e0ne has quit IRC19:06
*** pv_ has joined #openstack-cinder19:10
pv_hey quick question other than run_tests.sh script is there another way to run unit test? i wanna test the output of my unit tests but when i do ./run_tests.sh my_test_file it runs the unit tests for emc_vmax? and errors out19:11
lifelesspv_: tox19:12
*** wolsen_ is now known as wolsen19:13
pv_ooooo for sure thanks19:13
pv_i just looked it up http://docs.openstack.org/developer/cinder/devref/unit_tests.html lol shoulda checked there first instead of asking here19:13
pv_thanks19:13
*** Lee1092 has quit IRC19:13
*** kevincar_ has joined #openstack-cinder19:15
smcginnispv_: We are considering removing run_tests.sh partly to avoid this confusion of multiple ways to do this.19:16
*** stevemar_ has joined #openstack-cinder19:16
*** kevincarr1991 has quit IRC19:18
*** dsariel_ has joined #openstack-cinder19:20
*** thingee has quit IRC19:27
*** thingee has joined #openstack-cinder19:28
pv_oh yeah i see that now on the page for unit tests thanks19:28
*** akerr has joined #openstack-cinder19:33
*** kevincar_ has quit IRC19:35
*** akerr_ has joined #openstack-cinder19:38
*** kevincarr1991 has joined #openstack-cinder19:40
*** rhe00 has quit IRC19:40
*** akerr has quit IRC19:41
kevincarr1991I have setup cinder on my controller node and use a nfs share for the volumes. I am able to create volumes using horizon but I am unable to attach the volumes to instances using horizon19:44
*** ronis has quit IRC19:50
*** thingee has quit IRC19:50
*** dims_ has joined #openstack-cinder19:53
*** thingee has joined #openstack-cinder19:56
*** thingee has quit IRC19:56
*** thingee has joined #openstack-cinder19:56
*** timcl has quit IRC19:57
*** dims_ has quit IRC19:57
*** apoorvad_ has joined #openstack-cinder20:00
*** DericHorn-HP has joined #openstack-cinder20:03
*** apoorvad has quit IRC20:03
*** merooney has quit IRC20:04
*** markvoelker has joined #openstack-cinder20:12
*** raildo is now known as raildo-afk20:13
*** e0ne has joined #openstack-cinder20:15
*** dims_ has joined #openstack-cinder20:16
*** dims_ has quit IRC20:16
*** dims_ has joined #openstack-cinder20:16
*** raildo-afk is now known as raildo20:18
*** lcurtis_ has joined #openstack-cinder20:19
*** raildo is now known as raildo-afk20:23
*** thangp has quit IRC20:24
*** jordan_ has quit IRC20:25
openstackgerritIvan Kolodyazhny proposed openstack/cinder-specs: Remove hardcoded releases list from unit tests  https://review.openstack.org/21822620:31
*** diegows has quit IRC20:35
*** anshul has joined #openstack-cinder20:37
*** erhudy has quit IRC20:39
*** dustins has quit IRC20:42
*** bswartz has quit IRC20:48
*** Yogi1 has quit IRC20:48
*** aix has joined #openstack-cinder20:49
openstackgerritRajini Ram proposed openstack/cinder: This patch adds "over subscription in thin provisioning" support to Dell Equallogic driver. The following changes are proposed: * Dell eqlx driver reports the following new capacities and ratios:   * provisioned_capacity   * max_over_subscription_ratio     https://review.openstack.org/23118621:02
openstackgerritRajini Ram proposed openstack/cinder: This patch adds "over subscription in thin provisioning" support to Dell Equallogic driver. The following changes are proposed: * Dell eqlx driver reports the following new capacities and ratios:   * provisioned_capacity   * max_over_subscription_ratio     https://review.openstack.org/23118621:03
*** lcurtis_ has quit IRC21:04
*** baumann has joined #openstack-cinder21:05
*** baumann has left #openstack-cinder21:06
*** zigo has quit IRC21:09
*** zigo has joined #openstack-cinder21:10
openstackgerritRajini Ram proposed openstack/cinder: Dell Eqlx: Support over subscription in thin provisioning  https://review.openstack.org/23118621:12
diablo_rojogeg21:14
openstackgerritKendall Nelson proposed openstack/cinder: Hacking check for opt name registration  https://review.openstack.org/22337521:15
*** apoorvad_ has quit IRC21:16
*** apoorvad has joined #openstack-cinder21:17
openstackgerritRajini Ram proposed openstack/cinder: Dell Eqlx: Support over subscription in thin provisioning  https://review.openstack.org/23118621:17
*** martyturner has quit IRC21:21
*** diablo_rojo has quit IRC21:22
*** thingee has quit IRC21:22
*** ntpttr has quit IRC21:25
*** mriedem is now known as mriedem_away21:26
*** kevincarr1991 has quit IRC21:26
*** kevincarr1991 has joined #openstack-cinder21:27
*** asselin__ has quit IRC21:30
*** dave-mccowan has quit IRC21:31
*** eharney has quit IRC21:31
*** subscope has quit IRC21:33
*** kevincarr1991 has quit IRC21:33
*** kevincarr1991 has joined #openstack-cinder21:34
Swansonjgriffith, patrickeast: what exactly is expected to happen with replication_failover?21:37
*** kevincarr1991 has quit IRC21:38
patrickeastSwanson: it should switch over to the replicated volume for the one being failed over21:38
patrickeastSwanson: so like for Pure arrays we start re-routing management commands for that volume to the secondary array (if possible), in theory subsequent attaches or whatever then are using the other array with the replicated volume21:40
*** jgregor has quit IRC21:40
Swansonpatrickeast: At the server level?  I should connect the former replication target to any servers that were attached to the source volume?21:40
patrickeastSwanson: heh yea, so i'm not sure exactly how that is supposed to work for the initiators21:41
Swansonpatrickeast: Or just make the target the primary and let nova attach?21:41
patrickeastSwanson: that was my interpretation of it21:42
Swansonpatrickeast: That simplifies things.21:42
patrickeastjgriffith: ^ does that line up with how you envisioned the failover to go?21:44
*** edmondsw has quit IRC21:44
Swansonpatrickeast, jgriffith: a concern would be that the repl destination (at least for Dell) is actually another volume with the same data but not the same id.  That's fine so long as initialize_connection and export are called on the new volume.21:46
*** julim_ has quit IRC21:46
Swansonpatrickeast, jgriffith: And that would still be an issue even if the id were the same as the repl destination would have different iscsi ports and whatnot.21:46
patrickeastSwanson: yea, i would imagine in your drivers failover_replication (or replication_failover... whatever we called it) you swap the provider_id or something so you can reference the new volume21:47
openstackgerritAlex O'Rourke proposed openstack/cinder: WIP - 3PAR: Implement v2 replication (managed)  https://review.openstack.org/23120121:47
patrickeastSwanson: yea same here, the connection info is different for it after the failover21:47
patrickeastSwanson: so at least for Pure (and sounds like Dell) it requires being reattached through the normal flow21:47
Swansonpatrickeast: Okay.  This is making some sense.  A little bit magical but it does make sense.21:49
*** jwang_ has joined #openstack-cinder21:50
Swansonpatrickeast: Is replication_failover the only way to kick off a failover?  Or is there some expectation the driver should do it?  Cause if the driver just up and does it I'm assuming we need to poke things above us somehow.21:51
*** e0ne has quit IRC21:51
Swansonjgriffith: ^21:52
patrickeastSwanson: I think the API is the only way it gets kicked off, at one point there was a cool idea of doing one if the backend went down and stopped responding but that got dropped from the initial v2 implementation21:52
hemnapatrickeast, it depends on if you are implementing 'managed' or 'unmanaged'21:54
*** e0ne has joined #openstack-cinder21:54
patrickeasthemna: which part?21:54
hemnawhat to do inside of failover_replication21:54
patrickeastahh yea21:55
patrickeastwell21:55
patrickeastkinda21:55
patrickeastdepends on the way the backend does replication i suppose21:55
hemnafor managed, you basically do whatever you need to do on the array, and then return a model update.21:55
SwansonI went straight to unmanaged.21:55
hemnaswitching the host to the new driver host instance21:55
hemnafor unmanaged, it's more complicated21:55
patrickeastfor us it works pretty much the same way whether its managed or unmanaged21:56
hemnawe are doing 2 separate patches that basically first implement managed, and then a follow up (dep) for unmanaged21:56
openstackgerritAlex O'Rourke proposed openstack/cinder: WIP - 3PAR: Implement v2 replication (managed)  https://review.openstack.org/23120121:56
patrickeast^ good timing :D21:56
hemnapatrickeast, :)21:56
*** edtubill has quit IRC21:57
apoorvadkevincarr1991: did you try with command line? Whats the error?21:58
Swansonhemna, so if I'm doing managed I just do whatever I need to do to break the replication and return a model update with a new volume_type/whatever...21:58
*** apoorvad_ has joined #openstack-cinder21:59
*** harlowja has quit IRC21:59
*** abhi has joined #openstack-cinder21:59
hemnaSwanson, yup, basically22:00
hemnaSwanson, if you look in the solidfire driver, that's what they are doing.22:00
patrickeastSwanson: so for managed you need to update the host field in the model update along with the other stuff so that after the failover it is the other c-vol backend handling the request22:00
patrickeastSwanson: unmanaged is more like the same driver managing both targets22:00
hemnapatrickeast, I still had a few issues that I wanted to chat with you about22:00
hemnabut it's not a huge hurry22:00
*** setmason has quit IRC22:01
hemnabut I think the list replication targets needs to be standardized in it's response from the drivers22:01
*** e0ne has quit IRC22:01
patrickeasthemna: yea it seems like for consumption by someone calling the API we should have a standard format22:01
*** stevemar_ has quit IRC22:02
hemnaI was thinking at least having a standard key for the target_name22:02
*** setmason has joined #openstack-cinder22:02
hemnaso when the cinderclient lists the available targets, the thing that folks copy/paste for failover is the same key22:02
*** stevemar_ has joined #openstack-cinder22:02
*** apoorvad has quit IRC22:02
hemnaor we'll end up with different things for a target between solidfire, 3par, pure, etc22:02
patrickeastah yea that would be great22:02
hemnathat would suxors for users.22:03
patrickeastbut maybe i wanna be a beautiful snowflake!!22:03
hemna:)22:03
jgriffithhemna: patrickeast that sort of defeats the whole purpose if I follow what you two are saying22:03
hemnawe are even allowing mappings from primary pool name, to secondary pool name, for where to replication the volumes.22:03
patrickeastjgriffith: for the standarized keys in the response?22:04
hemnawell,just 1 standard key really22:04
hemnaand the value can be anything22:04
jgriffithYou're talking about the admin show targets cmd?22:04
patrickeastjgriffith: it would make it easier to figure out what field you need to use for calling the failover and specifying a target, right?22:04
hemnapatrickeast, exactly22:04
patrickeastyea so hemna is right, just the one key22:04
patrickeastthe rest can be free-form22:04
jgriffithpatrickeast: sure, so long as everybody can use the "same" thing :)22:04
patrickeasthaha yea22:05
hemnajgriffith, yah, that's the idea.....hopefully.22:05
jgriffithpatrickeast: I thought that's how I set it already... maybe not22:05
hemnatarget_name22:05
*** jungleboyj has quit IRC22:05
patrickeastjgriffith: easy, we add a functional test and validate the return field ;)22:05
*** stevemar_ has quit IRC22:05
*** stevemar_ has joined #openstack-cinder22:05
jgriffith:)22:05
patrickeastjgriffith: its very possible that is already there and i missed it22:06
patrickeastiirc when implementing it we just made up the format loosly following the comments in the code and may have done it wrong22:06
jgriffithpatrickeast: hemna https://github.com/openstack/cinder/blob/master/doc/source/devref/replication.rst22:06
*** haigang has joined #openstack-cinder22:07
jgriffithpatrickeast: hemna managed is "EZ", but I don't know how/if you'd want to do something different on unmanaged?22:07
*** setmason has quit IRC22:07
jgriffithpatrickeast: hemna the idea was to just spit out what they set in the config file22:08
hemnayah, unmanaged is the more complex case22:08
patrickeastjgriffith: yea they end up pretty much the same for us, but depends on how the arrays work22:08
jgriffithhemna: and you can't "force" some structure there IMO22:08
hemnaand the case I was referring to in returning the standardized 'target_name' for the replication targets list22:09
jgriffithhemna: you mean "list_replication_targets" ?22:09
hemnayes22:09
hemnasorry22:09
hemnaI'm terrible at communicating what I'm thinking22:09
jgriffithhemna: which is what I'm talking about too :)  And is the last item in that doc22:09
jgriffithhaha22:09
jgriffithI'll work on my telepathy :)22:09
hemnahehe22:10
hemnaso, what I was thinking was adding a target_name in there22:10
*** chlong has joined #openstack-cinder22:10
patrickeastjgriffith: yea so in the doc we have different keys for the manage vs unmanaged, but imo it makes sense to have a standard one that is the 'secondary' string you copy pasta into the api command to failover22:10
hemnathat everyone will add to each of the dict entries22:10
patrickeastat least one common one22:10
hemnapatrickeast, +122:11
jgriffithpatrickeast: hemna sure if it works, I'd need to see what is proposed though22:12
jgriffithpatrickeast: hemna when I looked through the code that was out there everybody needed some pretty "different" things22:12
patrickeastyea def22:12
hemnamaybe I'll put together a quick json dict to show22:13
jgriffithcool22:13
hemnahttp://pastebin.com/WcQDpewB22:15
*** openstackgerrit has quit IRC22:16
*** openstackgerrit has joined #openstack-cinder22:17
*** akerr_ has quit IRC22:17
*** jungleboyj has joined #openstack-cinder22:19
*** porrua has quit IRC22:19
openstackgerritPatrick East proposed openstack/cinder: Add a standard key for replication target dicts  https://review.openstack.org/23121022:20
patrickeastjgriffith: hemna: ^ i put it into the doc hopefully for more context22:21
* patrickeast should mark that was a do not merge22:21
hemnapatrickeast, nice.  much better than what I did.22:21
hemnathat's basically what I had in mind as well22:21
hemnawe could bikeshed for a few hours on 'name' vs. 'target_name'22:21
patrickeasthaha yea22:22
hemna:)22:22
hemnaor roll some dice and be done with it.22:22
hemnaeither way, I'd be happy with it.22:22
patrickeastmy reasoning is, and i can see the for-loop now, you would end up doing for target in targets: target['target_name']22:22
patrickeastid be like psh i just want target['name']22:22
*** tsekiyama has quit IRC22:23
hemnacool22:23
hemnasold22:23
hemnaheh22:23
hemnaI'm not married to 'target_name'22:23
hemnajust as long as the name idea is there so we can all output something that is there for the copy/paste for failover22:24
patrickeastsettle it with http://justflipacoin.com/ :D22:24
*** anshul has quit IRC22:24
patrickeastyep totally agree22:24
jgriffithOh... sure, you just want/need a label22:25
jgriffiththat makes sense22:25
*** tsekiyama has joined #openstack-cinder22:25
patrickeastyea should be simple for everyone to do it22:25
hemnayah, it's just a label22:25
jgriffithreplication_target_name22:26
patrickeasti dunno man, we've only got 80 chars per line to work with...22:26
* patrickeast is kidding... mostly22:26
hemna:)22:26
hemnathe_common_name_that_everyone_hated22:27
patrickeasthaha22:27
hemna:)22:27
hemnaliterally, that key22:27
hemnahave to do something to make it through Monday22:28
*** jamielennox|away is now known as jamielennox22:31
*** akerr has joined #openstack-cinder22:32
*** setmason has joined #openstack-cinder22:33
patrickeastgah this cisco fczm is annoying, can't run the tempest tests in parallel with it puking all over22:33
hemna?22:35
patrickeastbasically this bug https://bugs.launchpad.net/cinder/+bug/148239822:35
openstackLaunchpad bug 1482398 in Cinder "cisco zone manager fails to delete zone" [Undecided,New] - Assigned to Al Lau (alau2)22:35
*** kevincarr1991 has joined #openstack-cinder22:35
patrickeastthe zone manager (or i guess maybe the switch?) can't handle doing parallel management requests to modify the zone22:36
hemnadon't they have a local file lock in there?22:36
*** akerr has quit IRC22:36
patrickeastdunno22:36
* patrickeast checks22:36
hemnapatrickeast, https://github.com/openstack/cinder/blob/master/cinder/zonemanager/drivers/cisco/cisco_fc_zone_driver.py#L11122:37
hemnathey also have one on delete_connection22:38
patrickeasthmm that is interesting22:38
patrickeastjust now on a test i'm seeing http://paste.openstack.org/show/475379/22:39
patrickeasterrors deleting from a zone and adding to one22:39
patrickeastmaybe there is something asynchronous on the switch? so even if its locked in the cinder code it isn't ready yet..22:40
*** mc_nair has quit IRC22:40
hemnaFC just sucks22:40
openstackgerritAlex O'Rourke proposed openstack/cinder: WIP - 3PAR: Implement v2 replication (managed)  https://review.openstack.org/23120122:40
*** kevincarr1991 has quit IRC22:40
*** akerr has joined #openstack-cinder22:40
patrickeasti might just go add a retry with like 10 tries on it22:40
hemnaso22:40
hemnaI wonder22:40
hemnaif a previous delete command wasn't done on the switch22:40
hemnaif they fire off the command and expect it to finish async at some later time22:41
hemnaand they get a follow up delete before the async finishes, and so it pukes22:41
patrickeastyea : / thats probably exactly whats happening22:41
openstackgerritAlex O'Rourke proposed openstack/cinder: WIP - 3PAR: Implement v2 replication (managed)  https://review.openstack.org/23120122:41
hemnapatrickeast, https://supportforums.cisco.com/discussion/11089316/nexus-7000-copy-run-start-config-change-already-progress22:45
*** raildo-afk is now known as raildo22:45
*** wilson-1 has joined #openstack-cinder22:46
*** harlowja has joined #openstack-cinder22:48
*** wilson1 has quit IRC22:49
*** dims__ has joined #openstack-cinder22:49
*** dims__ has quit IRC22:49
*** merooney has joined #openstack-cinder22:50
*** lcurtis has quit IRC22:53
patrickeasthemna: mmm so it looks like there is already some retry logic built in, i'm going to test out just setting this from 1 to 5 or something https://github.com/openstack/cinder/blob/master/cinder/zonemanager/drivers/cisco/cisco_fc_zone_client_cli.py#L31822:53
*** merooeny_ has joined #openstack-cinder22:53
*** dims_ has quit IRC22:53
*** willsama has quit IRC22:54
hemnaok22:55
*** merooney has quit IRC22:56
hemnaoh the attempts=122:56
hemnainteresting22:56
patrickeastfrom what i'm seeing, it recovers fine so hopefully it isn't the bug mentioned in that link you sent22:56
patrickeastyea22:56
patrickeastit will keep trying with a sleep between tries22:56
hemnathat's quite odd they have a retry built in to executing ssh commands22:58
patrickeasthaha yea22:58
patrickeastmore so in that i don't see anywhere with it being >122:58
hemnayah, we know ssh sucks, so just try a bunch22:58
patrickeastits really going to suck with A-A if we don't have a distributed lock22:59
patrickeastno way this thing is going to work correctly22:59
hemnayah22:59
hemnabut we are pushing the boundaries of what some of these devices were intended22:59
patrickeasthaha true22:59
hemnaa device that was designed to be updated by a single dude22:59
patrickeastso we just have to properly emulate that dude23:00
hemnaput a REST interface on everything23:00
patrickeasthaha depends how they implement it though... easy enough to slap a REST API on top of whatever handled those CLI commands and still have the same problem23:01
hemnano comment!23:01
hemna:P23:01
patrickeastlol23:01
*** merooeny_ has quit IRC23:01
openstackgerritAlex O'Rourke proposed openstack/cinder: 3PAR: Implement v2 replication (managed)  https://review.openstack.org/23120123:02
*** dave-mccowan has joined #openstack-cinder23:02
*** david-ly_ has joined #openstack-cinder23:04
*** david-lyle has quit IRC23:04
patrickeasthemna: oh, so this fc stuff was another side track along poking at that permission denied/unable to write bug... i added some code to check what /sys/block/xxx/ro is... but what i'm wondering is if that is actually definitive, like for sure if its 1 the read-only, do you know if a 0 means for sure it is writable?23:06
*** kevincarr1991 has joined #openstack-cinder23:06
*** wilson1 has joined #openstack-cinder23:07
patrickeasti'm concerned it might be more of a best guess23:07
patrickeastby the kernel23:07
hemnayou mean the return/result code 1 ?23:08
*** raildo is now known as raildo-afk23:08
patrickeastjust whatever the value is, so like 'cat /sys/block/sdb/ro' is either 0 or 123:08
patrickeastor it shows up in the output of lsblk23:09
hemnahrmm23:09
*** gouthamr has quit IRC23:10
hemnahave you tried dumping out the perms on the device prior to the qemu-img convert call ?23:10
*** wilson-1 has quit IRC23:10
patrickeastyea the file system permissions are always the same between ones that work and ones that fail23:10
hemnawth23:10
hemnamaybe disable directio ?23:11
*** kevincarr1991 has quit IRC23:11
patrickeastso im going the next level down to the block device read-only info23:11
hemnathe test that pukes is right after it thinks it can do directio on it23:11
hemnaleaving out the device buffers23:11
*** apoorvad_ has quit IRC23:11
hemnacrap...off to soccer practice23:12
*** hemna is now known as hemnafk23:12
patrickeasthmm thats worth a shot23:12
patrickeasti'll try it out23:12
patrickeasthemnafk: thanks23:12
*** apoorvad has joined #openstack-cinder23:12
openstackgerritOpenStack Proposal Bot proposed openstack/cinder: Updated from global requirements  https://review.openstack.org/23122523:13
*** earlephilhower has quit IRC23:14
*** chris_morrell has quit IRC23:15
*** dims_ has joined #openstack-cinder23:17
openstackgerritOpenStack Proposal Bot proposed openstack/os-brick: Updated from global requirements  https://review.openstack.org/23123923:17
openstackgerritAlex O'Rourke proposed openstack/cinder: WIP - 3PAR: Implement v2 replication (unmanaged)  https://review.openstack.org/23124523:19
*** tsekiyama has quit IRC23:23
patrickeast:o23:23
patrickeasthemnafk: finally got one to repro... they are showing up as read-only at that level http://paste.openstack.org/show/475384/23:23
openstackgerritAlex O'Rourke proposed openstack/cinder: WIP - 3PAR: Implement v2 replication (unmanaged)  https://review.openstack.org/23124523:24
*** zhiyan has quit IRC23:24
*** xyang1 has quit IRC23:25
*** zhiyan has joined #openstack-cinder23:26
patrickeastjungleboyj: hey, just saw your comment on https://review.openstack.org/#/c/230781/23:27
patrickeastjungleboyj: so uh... the short answer is 'its complicated'23:28
patrickeastjungleboyj: i have a list of them over in the Nova side of things https://review.openstack.org/#/c/229152/23:28
*** angela-s has quit IRC23:29
openstackgerritJohn Griffith proposed openstack/cinder: Add ability to set prefix in SolidFire Volume name  https://review.openstack.org/23125023:30
*** IanGovett has quit IRC23:32
*** gouthamr has joined #openstack-cinder23:32
*** gouthamr_ has joined #openstack-cinder23:34
*** tsekiyama has joined #openstack-cinder23:34
*** gouthamr has quit IRC23:37
openstackgerritJohn Griffith proposed openstack/cinder: Add ability to set prefix in SolidFire Volume name  https://review.openstack.org/23125023:41
*** stevemar_ has quit IRC23:42
*** setmason has quit IRC23:42
*** stevemar_ has joined #openstack-cinder23:45
patrickeastjgriffith: hey, just curious for that change with the prefix ^ any reason in particular why you aren't using the volume_name_template and volume['name'] instead of volume['id']23:47
patrickeastjgriffith: i ask because i wasn't sure what exactly the intended use was for it... but my driver was implemented with it, just curious if there are any downsides i should know about23:48
*** dims_ has quit IRC23:50
*** DericHorn-HP has quit IRC23:51
jgriffithpatrickeast: Name is optional... but I could just use the existing conf option I guess23:51
jgriffithpatrickeast: ie volume[name] is optional23:52
jgriffithwanted guaranteed unique  also23:52
jgriffithpatrickeast: template_name might be a good idea though....23:53
patrickeastjgriffith: eh other name... this confused me for a while, https://github.com/openstack/cinder/blob/master/cinder/db/sqlalchemy/models.py#L12723:53
*** haigang has quit IRC23:53
patrickeastjgriffith: volume['name'] will be the UUID passed into the template23:53
jgriffithpatrickeast: Ohhh.. that name :)23:53
patrickeastjgriffith: and does the "right" thing after migrations23:53
jgriffithpatrickeast: well lookie there :)23:54
*** gouthamr has joined #openstack-cinder23:55
jgriffithpatrickeast: cool... I marked it as wip and noted your suggestion23:55
jgriffithI'll take a look at that tomorrow23:55
jgriffithgreat idea, thanks!23:55
patrickeastjgriffith: ah ok, cool, so i guess the answer was 'no' there wasn't any downside to it i should be concerned about with my driver ;)23:55
jgriffithpatrickeast: No way23:56
*** gouthamr_ has quit IRC23:56
*** I has joined #openstack-cinder23:56
* patrickeast was accidentally helpful23:56
jgriffithpatrickeast: haha23:56
*** I is now known as Guest7130723:56
jgriffithpatrickeast: I have to look at it, it doesn't help the problem of people not having it set or if I introduce it with something other than UUID-23:56
jgriffithpatrickeast: because existing folks will upgrade and while it works on the Cinder side, the lookups on the SolidFire side will all fail :(23:57
patrickeastjgriffith: yea : ( that does suck23:57
jgriffithwhich is why I went the route I did.... but I'm now thinking there may be a way around it23:57

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