Tuesday, 2015-09-29

*** gouthamr_ has joined #openstack-cinder00:00
*** EinstCrazy has joined #openstack-cinder00:01
*** gouthamr has quit IRC00:03
*** angela-s has quit IRC00:07
*** asselin_ has joined #openstack-cinder00:08
*** EinstCrazy has quit IRC00:09
*** NightKhaos has quit IRC00:10
*** IlyaG has quit IRC00:11
*** NightKhaos has joined #openstack-cinder00:16
*** IlyaG has joined #openstack-cinder00:18
*** dims has joined #openstack-cinder00:24
*** ociuhandu has joined #openstack-cinder00:27
*** diegows has quit IRC00:30
openstackgerritThang Pham proposed openstack/cinder: Conversion to volume object  https://review.openstack.org/17705400:31
openstackgerritThang Pham proposed openstack/cinder: Update retype to use versionedobjects  https://review.openstack.org/22821800:31
*** tsekiyama has quit IRC00:33
*** DericHorn-HP has joined #openstack-cinder00:34
*** gouthamr has joined #openstack-cinder00:38
*** aix has quit IRC00:39
*** IlyaG has quit IRC00:39
*** aix has joined #openstack-cinder00:40
*** leeantho has quit IRC00:40
*** gouthamr_ has quit IRC00:41
*** amit213 has quit IRC00:45
*** gouthamr_ has joined #openstack-cinder00:46
*** edmondsw has quit IRC00:47
*** gouthamr has quit IRC00:49
*** DericHorn-HP has quit IRC00:49
*** aix has quit IRC00:52
*** DericHorn-HP has joined #openstack-cinder00:52
*** aix has joined #openstack-cinder00:55
*** shakamunyi has joined #openstack-cinder00:55
*** _cjones_ has quit IRC01:00
*** zhangjn has joined #openstack-cinder01:01
*** EinstCrazy has joined #openstack-cinder01:01
*** r-daneel has quit IRC01:04
*** chenke has joined #openstack-cinder01:06
*** dave-mcc_ has joined #openstack-cinder01:07
*** mtanino has quit IRC01:09
*** dave-mccowan has quit IRC01:10
*** zhangjn_ has joined #openstack-cinder01:17
*** dtroyer has left #openstack-cinder01:18
*** zhangjn has quit IRC01:19
*** DericHorn-HP has quit IRC01:20
openstackgerritThang Pham proposed openstack/cinder: Conversion to volume object  https://review.openstack.org/17705401:22
openstackgerritThang Pham proposed openstack/cinder: Update retype to use versionedobjects  https://review.openstack.org/22821801:22
*** apoorvad_ has joined #openstack-cinder01:27
*** apoorvad has quit IRC01:31
*** apoorvad_ has quit IRC01:31
*** diablo_rojo has quit IRC01:41
*** dustins has joined #openstack-cinder01:47
*** DericHorn-HP has joined #openstack-cinder01:48
*** changbl has joined #openstack-cinder01:55
*** DericHorn-HP has quit IRC01:56
*** julim has joined #openstack-cinder01:57
*** duobei has joined #openstack-cinder01:59
nikeshmtrying to use devstack for external ceph-cluster01:59
nikeshmusing ubuntu 14.04 juno devstack02:00
*** edtubill has joined #openstack-cinder02:00
nikeshmgetting error http://paste.openstack.org/show/474556/02:01
*** DericHorn-HP has joined #openstack-cinder02:02
*** changbl has quit IRC02:04
openstackgerritThang Pham proposed openstack/cinder: Update extend_volume to use versionedobjects  https://review.openstack.org/22869002:09
*** dims has quit IRC02:09
*** DericHorn-HP has quit IRC02:11
*** stevemar has joined #openstack-cinder02:19
*** rushil has joined #openstack-cinder02:25
*** sghanekar has quit IRC02:25
*** salv-orlando has joined #openstack-cinder02:28
*** zhenguo has joined #openstack-cinder02:34
*** salv-orlando has quit IRC02:37
*** chenying11 has joined #openstack-cinder02:39
*** garthb has joined #openstack-cinder02:44
*** Lee1092 has joined #openstack-cinder02:45
openstackgerritNathaniel proposed openstack/cinder: Add backup_keystone_url to swift backup driver  https://review.openstack.org/22864602:47
*** haomaiwang has joined #openstack-cinder02:51
*** Yogi1 has joined #openstack-cinder02:54
*** ks1 has joined #openstack-cinder02:59
*** haomaiwang has quit IRC03:01
*** haomaiwang has joined #openstack-cinder03:01
*** Yogi1 has quit IRC03:02
*** Yogi1 has joined #openstack-cinder03:08
*** dims has joined #openstack-cinder03:09
*** Yogi1 has quit IRC03:12
*** chenying11 has quit IRC03:13
*** DericHorn-HP has joined #openstack-cinder03:14
*** dims has quit IRC03:16
*** bnemec has quit IRC03:17
*** jdurgin1 has joined #openstack-cinder03:18
*** boris-42 has quit IRC03:19
*** hemna has joined #openstack-cinder03:20
*** harlowja_at_home has joined #openstack-cinder03:20
*** dave-mcc_ has quit IRC03:24
openstackgerritWilson Liu proposed openstack/cinder: Encode username and password in Huawei driver  https://review.openstack.org/21608503:27
*** boris-42 has joined #openstack-cinder03:29
openstackgerritWilson Liu proposed openstack/cinder: Fix capacity report error in Huawei driver  https://review.openstack.org/22630803:29
openstackgerritWilson Liu proposed openstack/cinder: Fix update Huawei driver issue  https://review.openstack.org/22720003:31
*** DericHorn-HP has quit IRC03:32
openstackgerritWilson Liu proposed openstack/cinder: Add hypermetro support for Huawei driver  https://review.openstack.org/20202303:33
*** dustins has quit IRC03:33
*** BharatK has joined #openstack-cinder03:34
*** links has joined #openstack-cinder03:36
*** chenke has quit IRC03:37
*** chenke has joined #openstack-cinder03:38
*** ks1 has quit IRC03:41
*** bill_az has quit IRC03:44
*** DericHorn-HP has joined #openstack-cinder03:46
*** DericHorn-HP has quit IRC03:49
*** zhangjn_ has quit IRC03:54
*** markvoelker has quit IRC03:56
*** ociuhandu has quit IRC04:00
*** haomaiwang has quit IRC04:01
*** haomaiwang has joined #openstack-cinder04:01
*** rushil has quit IRC04:02
*** hemna has quit IRC04:06
*** zhangjn has joined #openstack-cinder04:10
*** IlyaG has joined #openstack-cinder04:18
*** jdurgin1 has quit IRC04:21
*** salv-orlando has joined #openstack-cinder04:29
*** DericHorn-HP has joined #openstack-cinder04:31
jgriffithasselin: ping04:32
jgriffithasselin: did you intentionally remove service logs from your test reports?04:33
jgriffithasselin: I'm looking at a failed run that failed for all of HP and for SolidFire, wanted to see if there was a common issue, but you have no logs :(04:33
jgriffithasselin: and no testr results :(04:34
*** DericHorn-HP has quit IRC04:34
*** BharatK has quit IRC04:38
*** salv-orlando has quit IRC04:38
*** IlyaG has quit IRC04:38
*** harlowja_at_home has quit IRC04:45
*** DericHorn-HP has joined #openstack-cinder04:48
*** zhangjn has quit IRC04:52
*** zhangjn has joined #openstack-cinder04:53
*** DericHorn-HP has quit IRC04:56
*** markvoelker has joined #openstack-cinder04:57
*** sgotliv has joined #openstack-cinder04:59
*** haomaiwang has quit IRC05:01
*** haomaiwang has joined #openstack-cinder05:01
*** markvoelker has quit IRC05:01
*** stevemar has quit IRC05:03
*** nkrinner has joined #openstack-cinder05:03
*** hemna has joined #openstack-cinder05:03
*** sgotliv has quit IRC05:04
*** dims has joined #openstack-cinder05:13
*** IlyaG has joined #openstack-cinder05:16
*** asselin__ has joined #openstack-cinder05:17
*** dims has quit IRC05:18
*** IlyaG has quit IRC05:20
*** asselin_ has quit IRC05:21
*** IlyaG has joined #openstack-cinder05:22
*** krtaylor has quit IRC05:23
*** cFouts has quit IRC05:31
*** deepakcs has joined #openstack-cinder05:31
*** krtaylor has joined #openstack-cinder05:32
*** garthb has quit IRC05:32
*** sileht has quit IRC05:33
*** zhangjn has quit IRC05:36
*** stevemar has joined #openstack-cinder05:36
*** stevemar_ has joined #openstack-cinder05:42
*** stevema__ has joined #openstack-cinder05:43
*** stevemar has quit IRC05:45
*** stevemar_ has quit IRC05:46
*** BharatK has joined #openstack-cinder05:53
*** Lee1092 has quit IRC05:53
*** haomaiwang has quit IRC06:01
*** haomaiwang has joined #openstack-cinder06:01
openstackgerritDave Chen proposed openstack/python-cinderclient: Add commands to show image metadata  https://review.openstack.org/20936006:09
*** jamielennox is now known as jamielennox|away06:12
*** openstackgerrit has quit IRC06:16
*** openstackgerrit has joined #openstack-cinder06:17
*** rajinir has quit IRC06:18
*** vgridnev has joined #openstack-cinder06:18
*** chenying11 has joined #openstack-cinder06:24
*** asselin_ has joined #openstack-cinder06:26
*** e0ne has joined #openstack-cinder06:27
*** stevema__ has quit IRC06:28
*** stevemar has joined #openstack-cinder06:28
*** lpetrut has joined #openstack-cinder06:28
*** asselin__ has quit IRC06:30
*** stevemar has quit IRC06:33
*** e0ne has quit IRC06:34
*** deepakcs has quit IRC06:38
*** deepakcs has joined #openstack-cinder06:42
*** salv-orlando has joined #openstack-cinder06:44
*** markvoelker has joined #openstack-cinder06:45
*** salv-orlando has quit IRC06:47
*** IlyaG has quit IRC06:50
*** markvoelker has quit IRC06:50
*** haomaiwang has quit IRC07:01
*** haomaiwang has joined #openstack-cinder07:01
*** jwcroppe has joined #openstack-cinder07:04
*** e0ne has joined #openstack-cinder07:07
*** e0ne has quit IRC07:09
*** e0ne has joined #openstack-cinder07:21
*** salv-orlando has joined #openstack-cinder07:28
*** alexpilotti has joined #openstack-cinder07:31
*** ronis has joined #openstack-cinder07:32
*** salv-orlando has quit IRC07:32
*** bluex-pl has quit IRC07:44
*** bluex-pl has joined #openstack-cinder07:45
*** bluex-pl has quit IRC07:45
*** bluex-pl has joined #openstack-cinder07:46
*** bluex-pl has quit IRC07:47
*** bluex-pl has joined #openstack-cinder07:47
*** jordanP has joined #openstack-cinder07:49
*** alexpilotti has quit IRC07:52
*** haomaiwang has quit IRC08:01
*** haomaiwang has joined #openstack-cinder08:01
*** vgridnev has quit IRC08:05
*** lpetrut has quit IRC08:06
*** markus_z has joined #openstack-cinder08:07
*** gouthamr_ has quit IRC08:11
*** lpetrut has joined #openstack-cinder08:15
*** tomh__ has quit IRC08:15
*** jistr has joined #openstack-cinder08:16
*** Lee1092 has joined #openstack-cinder08:26
*** bluex-pl has quit IRC08:29
*** bluex-pl has joined #openstack-cinder08:29
*** sgotliv has joined #openstack-cinder08:30
*** bluex-pl has quit IRC08:31
*** bluex-pl has joined #openstack-cinder08:32
*** salv-orlando has joined #openstack-cinder08:35
openstackgerritRajesh Tailor proposed openstack/cinder: Fix unreachable code pylint issues  https://review.openstack.org/22879208:44
*** markvoelker has joined #openstack-cinder08:46
chenke Anybody have the juno version of hyper-v cinder software--CinderVolumeSetup_Beta.msi. The lastest version of cloudbase is kilo.08:47
*** vgridnev has joined #openstack-cinder08:48
*** lpetrut has quit IRC08:48
*** markvoelker has quit IRC08:51
*** haomaiwang has quit IRC09:01
*** edtubill has quit IRC09:01
*** haomaiwang has joined #openstack-cinder09:01
openstackgerritIvan Kolodyazhny proposed openstack/cinder-specs: Attach/detach volumes without Nova (draft)  https://review.openstack.org/22412409:03
*** ndipanov has quit IRC09:07
*** daneyon_ has joined #openstack-cinder09:11
*** daneyon has quit IRC09:11
*** dims has joined #openstack-cinder09:16
*** ndipanov has joined #openstack-cinder09:18
*** salv-orl_ has joined #openstack-cinder09:21
*** dims has quit IRC09:22
*** salv-orlando has quit IRC09:25
*** anshul has joined #openstack-cinder09:38
*** zhenguo has quit IRC09:43
*** lpetrut has joined #openstack-cinder09:44
*** dims has joined #openstack-cinder09:47
*** lpetrut1 has joined #openstack-cinder09:48
*** lpetrut has quit IRC09:48
*** lpetrut1 is now known as lpetrut09:48
*** kbyrne has quit IRC09:58
*** haomaiwang has quit IRC10:01
*** haomaiwa_ has joined #openstack-cinder10:01
*** kbyrne has joined #openstack-cinder10:03
*** chenke has quit IRC10:04
*** IanGovett has joined #openstack-cinder10:04
*** haomaiwa_ has quit IRC10:07
*** chenying11 has quit IRC10:36
*** EinstCrazy has quit IRC10:42
*** markvoelker has joined #openstack-cinder10:47
*** duobei has quit IRC10:47
*** jamielennox|away is now known as jamielennox10:49
*** markvoelker has quit IRC10:52
*** 18VAARFFB has joined #openstack-cinder10:56
*** 18VAARFFB has quit IRC11:01
*** haomaiwang has joined #openstack-cinder11:01
*** EinstCrazy has joined #openstack-cinder11:04
*** haomaiwang has quit IRC11:11
*** svasheka has quit IRC11:19
*** duobei has joined #openstack-cinder11:20
*** delattec has quit IRC11:21
*** salv-orl_ has quit IRC11:22
*** salv-orlando has joined #openstack-cinder11:23
*** rajesht has joined #openstack-cinder11:23
*** svasheka has joined #openstack-cinder11:32
*** timcl has joined #openstack-cinder11:33
asselin_jgriffith, hi, no not intentionally.11:37
asselin_jgriffith, link?11:38
smcginnisasselin_: I noticed one too yesterday. Let me see if I can find which one that was.11:38
asselin_fyi, last night I disabled negative reporting due to devstack failing:
*** stevemar has joined #openstack-cinder11:40
smcginnisasselin_: That might have been related. I think I noticed it because it was a patch where my CI failed to stack as well.11:40
asselin_"publicURL endpoint for volume service in RegionOne region not found"11:41
asselin_smcginnis, oh well then yes for sure, if devstack fails there are no service logs or testr results11:41
smcginnisasselin_: Yep, that's the one. I noticed that message.11:41
smcginnisasselin_: Not sure if that's the on jgriffith was referring to, but that appears to be the one I noticed.11:42
asselin_so perhaps jgriffith is failing to stack as well11:42
smcginnisI don't see SF in the list, so looks likely.11:43
*** delattec has joined #openstack-cinder11:43
smcginnisasselin_: Hey, aren't you west coast?11:43
asselin_yes, aren't you also?11:43
smcginnisasselin_: Central. Shouldn't you be asleep yet? ;)11:44
asselin_smcginnis, 'yet'...just woke up :)11:44
smcginnisasselin_: I thought I got up early.11:44
*** stevemar has quit IRC11:45
asselin_must be the full moon11:45
smcginnisasselin_: You might be on to something there...11:45
*** delattec has quit IRC11:50
*** cdelatte has joined #openstack-cinder11:53
*** delattec has joined #openstack-cinder11:53
asselin_smcginnis, jgriffith did mention this on the mailing list: "It was only in the last milestone of Liberty that folks finally got everything updated and talking V2.  Not to mention the patch to switch the default in devstack just landed (where everything uses it including Nova)."11:55
asselin_smcginnis, jgriffith e0ne https://review.openstack.org/#/c/194726/11:57
e0nesmcginnis, asselin_: yep, it was me:(11:57
asselin_e0ne, what's the fix?11:58
*** jgregor has joined #openstack-cinder11:59
*** timcl has left #openstack-cinder12:00
smcginnise0ne: Think we need to revert that change?12:00
e0nesmcginnis: 5 minutes, please12:00
smcginnise0ne: We might need to do an extra long deprecation on v1 (even longer yet) based on the client library info posted by Mark Voelker.12:01
e0nesmcginnis: I'm trying to understand why asselin_'s CI failed12:01
asselin_e0ne, let me know if I can help12:02
*** markvoelker has joined #openstack-cinder12:03
e0neasselin_: openstackclient still uses api v1 in your case12:03
smcginnise0ne: OK, cool.12:03
*** timcl has joined #openstack-cinder12:03
smcginnise0ne: Even if that's not asselin_'s problem, my impression is we need to extend this until we've reached out to more lib/client consumers.12:04
e0neasselin_: where can I get sources for /opt/stack/new/devstack/extras.d/90-hp.sh?12:04
asselin_e0ne, it should be in the log folder12:05
e0nesmcginnis: that's why I'm proposing to disable v1, NOT to delete12:05
asselin_will check...that was the intent12:05
*** markus_z has quit IRC12:06
*** markvoelker has quit IRC12:07
asselin_e0ne, http://paste.openstack.org/show/474649/12:08
*** sgotliv_ has joined #openstack-cinder12:08
asselin_e0ne, seems it doesn't get saved. we'll fix that12:08
*** aarefiev has quit IRC12:09
*** baumann has joined #openstack-cinder12:09
e0neasselin_: you use too old python-openstackclient
*** jgregor has left #openstack-cinder12:10
*** jgregor has joined #openstack-cinder12:10
asselin_e0ne, where does the version get set?12:10
*** sgotliv has quit IRC12:11
*** baumann has left #openstack-cinder12:11
*** baumann has joined #openstack-cinder12:11
*** aarefiev has joined #openstack-cinder12:12
*** markus_z has joined #openstack-cinder12:13
e0neasselin_: could you please check what exactly version of openstackclient do you use?12:14
asselin_e0ne, python-openstackclient==1.7.0
e0nehm... it should wokr12:16
*** deepakcs has quit IRC12:17
asselin_it should be reproducible outside our ci12:17
e0neasselin_: it works on my devstack env w/o api v212:18
asselin_perhaps there's an issue regarding the devstack phase it's running in?12:18
e0neasselin_: enable_v1_api = true (
asselin_e0ne, I think both v1 and v2 are enabled12:20
e0neasselin_: but keystone has not v1 endpoints12:20
e0neasselin_: i've found only
*** raildo-afk is now known as raildo12:21
asselin_e0ne, ok, I can remove the line that enables cinder v1:
asselin_e0ne, but still not clear why it should break with both enabled and the change in devstack12:23
e0neasselin_: TBH, I don't understans why openstack client still uses v112:23
e0neasselin_: my patch to devstack disables v1 by default. after it, there is no keystone endpoint for v112:24
asselin_e0ne, so then openstack client should be switched to v2 first, then v1 disabled?12:24
*** delatte has joined #openstack-cinder12:24
e0neasselin_: FYI, here is patch for fix rally https://review.openstack.org/#/c/221399/12:24
asselin_e0ne, but we have v1 api enabled12:25
*** kevincarr1991 has joined #openstack-cinder12:26
*** cdelatte has quit IRC12:27
*** delattec has quit IRC12:27
e0neasselin_: you enable it only in cinder.conf. keystone endpoint is not created12:27
*** akerr has joined #openstack-cinder12:28
asselin_e0ne, ok I see, so I should remove both those lines to exports as done in the rally patch12:29
asselin_or just delete the v1 line12:29
asselin_or delete both12:29
asselin_and let devstack do the right thing12:29
e0neasselin_: or fix openstackclient usage12:30
e0neasselin_: I think, I've found a root cause12:30
*** bswartz has quit IRC12:30
*** edmondsw has joined #openstack-cinder12:31
e0neasselin_: https://github.com/openstack/python-openstackclient/blob/master/openstackclient/volume/client.py#L22-L24 - looks like we set devault api version to 2 but didn't change API_NAME :(12:31
asselin_what's the new name?12:32
*** markvoelker has joined #openstack-cinder12:32
e0neasselin_: could you try http://paste.openstack.org/show/474656/?12:33
*** gouthamr has joined #openstack-cinder12:33
e0neasselin_: API_NAME should be 'volumev2"12:34
*** ollie1 has joined #openstack-cinder12:34
asselin_e0ne, ok I will try both and let you know12:34
*** ollie1 has left #openstack-cinder12:34
e0neasselin_:  thanks12:34
*** EinstCrazy has quit IRC12:34
*** amoturi has joined #openstack-cinder12:36
*** dims_ has joined #openstack-cinder12:36
*** aix has quit IRC12:37
asselin_e0ne, are you sure the placement of --os-volume-api-version 2 is correct?12:38
e0neasselin_:  yes http://paste.openstack.org/show/474658/12:39
*** EinstCrazy has joined #openstack-cinder12:40
asselin_so what about the other calls in the file?12:40
*** dims has quit IRC12:40
asselin_e0ne, also regarding the 2nd method, why don't I just delete the 2 enables in local.conf and let devstack set it?12:41
e0neasselin_:  it should be fixed in the same way12:41
e0neasselin_:  do you mean just enable v1 in local.conf for devstack?12:41
asselin_e0ne, no I mean, don't select any to be enabled or disabled. use the devstack default12:42
*** kevincarr1991 has quit IRC12:42
e0neasselin_:  by default, it will disable v112:42
asselin_e0ne, because we use these scripts for kilo and juno also12:43
*** kevincarr1991 has joined #openstack-cinder12:44
e0neasselin_:  ok. just add CINDER_ENABLE_V1_API=True to youd devstack conf12:44
openstackgerritVictor Stinner proposed openstack/cinder: Port test_volume to Python 3  https://review.openstack.org/21572912:45
e0neasselin_:  it will work for current master only and will be ignored for stable branches12:45
*** xyang1 has joined #openstack-cinder12:48
*** diablo_rojo has joined #openstack-cinder12:49
*** merooney has joined #openstack-cinder12:50
*** alejandrito has joined #openstack-cinder12:52
*** dims_ has quit IRC12:53
*** thangp has joined #openstack-cinder12:54
*** breitz has quit IRC12:54
*** breitz has joined #openstack-cinder12:54
*** bswartz has joined #openstack-cinder12:57
*** baumann has left #openstack-cinder12:57
*** baumann has joined #openstack-cinder12:57
*** kevincarr1991 has quit IRC13:00
*** salv-orlando has quit IRC13:01
*** dims has joined #openstack-cinder13:01
*** dave-mccowan has joined #openstack-cinder13:03
*** mriedem_away is now known as mriedem13:07
*** martyturner has joined #openstack-cinder13:10
*** daemontool_ has joined #openstack-cinder13:10
*** amoturi has quit IRC13:10
*** DericHorn-HP has joined #openstack-cinder13:12
*** delatte has quit IRC13:13
*** porrua has joined #openstack-cinder13:16
*** Yogi1 has joined #openstack-cinder13:17
*** dustins has joined #openstack-cinder13:21
*** amoturi has joined #openstack-cinder13:21
*** links has quit IRC13:22
*** jdurgin1 has joined #openstack-cinder13:23
*** bill_az has joined #openstack-cinder13:25
*** alejandrito has quit IRC13:28
openstackgerritThomas Goirand proposed openstack/cinder: Fix typo in cinder-config-generator.conf  https://review.openstack.org/22889713:31
zigoe0ne: This one, really lol ! :)13:31
zigoe0ne: Can you review it?13:31
*** salv-orlando has joined #openstack-cinder13:31
e0nezigo: :)13:31
*** kevincarr1991 has joined #openstack-cinder13:33
zigoGuys, really, do more like this, so I can vote for PTLs on every projects... :D13:33
*** stevemar has joined #openstack-cinder13:34
zigoNow, I get this issue: http://paste.openstack.org/show/474677/13:35
*** skraynev has quit IRC13:35
zigoAnyone has a clue on what's going on ?13:35
*** stevemar_ has joined #openstack-cinder13:36
zigoe0ne: Do you have an idea?13:36
*** haomaiwang has joined #openstack-cinder13:36
smcginnisdiablo_rojo: ^^13:36
*** skraynev has joined #openstack-cinder13:36
openstackgerritAbhishek Kekane proposed openstack/cinder-specs: NFS driver snapshot support  https://review.openstack.org/22890013:37
*** stevemar_ has quit IRC13:37
*** stevemar_ has joined #openstack-cinder13:38
*** stevemar has quit IRC13:39
*** alejandrito has joined #openstack-cinder13:40
*** haomaiwang has quit IRC13:40
*** aix has joined #openstack-cinder13:40
*** haomaiwa_ has joined #openstack-cinder13:40
zigoIf I remove --namespace cinder, then it works ... :/13:41
kevincarr1991Has anyone here every directly attached your compute node to a nfs share. I know this isnt a cinder issue, but this is a good community so i thought i would ask13:42
zigoI kind of believe that's not an option I want to remove when generating the package config file for Debian ...13:42
smcginniszigo: This had been working the last time I tested it. Did you change anything else?13:42
zigosmcginnis: I haven't done anything special to the Cinder package that I can think of.13:43
zigoI just have a disable-zfs-tests.patch which removes tests that I can't run, but that's it.13:43
jgriffithzigo: I just pulled rc branch and it ran fine on my side13:44
*** jaypipes has quit IRC13:44
zigojgriffith: Maybe a missing dependency?13:44
smcginniszigo: I pulled down 228897 and run with that and it ran fine for me.13:45
zigoHow can I trace this?13:46
*** cdelatte has joined #openstack-cinder13:46
*** delattec has joined #openstack-cinder13:46
jgriffithzigo:  need to look in the oslo tool13:46
jgriffithzigo: see what's up with the ep.resolve method I think13:46
*** mdbooth_ has joined #openstack-cinder13:46
zigojgriffith: In the oslo-config-generator?13:47
zigojgriffith: Where's that? I don't see such a thing.13:48
zigogrep -r resolv in the oslo_config folder returns nothing.13:49
*** jaypipes has joined #openstack-cinder13:49
*** mdbooth has quit IRC13:49
*** mdbooth_ is now known as mdbooth13:49
smcginniszigo: From your traceback it looks like it's actually into the stevedore package.13:50
*** erlon has joined #openstack-cinder13:50
zigosmcginnis: Stevedore only shows a missing plugin, IMO.13:50
zigoA trace ending in Stevedore is most of the time not an issue in Stevedore itself.13:50
*** DericHorn-HP has quit IRC13:50
*** DericHorn-HP has joined #openstack-cinder13:52
jgriffithzigo: might want to take a look at the generated cinder/opts.py file13:53
jgriffithzigo: you might have to hack the generate script to keep it around (it create/deletes itself)13:53
*** nkrinner has quit IRC13:53
jgriffithalso maybe paste a git diff of your branch?13:53
zigojgriffith: I'm launching the genration by calling oslo-config-generator directly. Is this bad?13:54
jgriffithzigo: :)13:54
jgriffithI had a feeling :)13:54
jgriffithyeah, that won't work in Cinder13:54
jgriffithwe're *different*13:54
*** stevemar_ is now known as stevemar13:54
jgriffithuse "tox -egenconfig"13:54
zigojgriffith: Ah ok, good to know. FYI, I'm doing this for every other projects, and so far, it's only a problem with Cinder.13:54
zigoNop, can't use tox ... :)13:54
jgriffithzigo: yeah, we're the only ones that are different13:54
zigoThough I can run: tools/config/generate_sample.sh from_tox13:55
jgriffithzigo: no worries, you can13:55
zigo(which is the same)13:55
openstackgerritWilson Liu proposed openstack/cinder: Encode username and password in Huawei driver  https://review.openstack.org/21608513:55
zigojgriffith: Should I ommit the "from_tox" ?13:55
jgriffithzigo: the other project maintain a static opts.py file, we do not13:55
jgriffithzigo: yes13:55
jgriffithhonestly I am not sure why that's there or if it's needed13:56
jgriffithanwho... that should fix you up I think13:56
openstackgerritZhang Jinnan proposed openstack/cinder: not exception error when extend error status volume  https://review.openstack.org/22797113:56
zigoif [ ${1} != "from_tox" ] ; then13:56
zigo    print_error13:56
zigo    exit 113:56
zigoThis doesn't feel right ... :)13:57
*** Yogi11 has joined #openstack-cinder13:57
smcginnisOur meager attempt at protecting from not running this in the expected way.13:57
jgriffithsmcginnis: Oh!13:57
jgriffithsmcginnis: well ther eya go13:57
jgriffithzigo: we're trying to make your life as difficult as possible while making ours easier :)13:58
*** Yogi1 has quit IRC13:58
zigojgriffith: Yeah, thanks so much ! :)13:58
jgriffithzigo: you'll need to edit that line out13:58
smcginnisjgriffith: Now we just need to patch oslo-config-generator: if cinder: do something different. :}13:58
jgriffithzigo: my pleasure :)13:58
* zigo is currently reading the script to avoid calling it ...13:58
jgriffithsmcginnis: ha!13:58
jgriffithzigo: enjoy13:59
zigoIt only exports 2 variables ... :P13:59
jgriffithzigo: don't forget to look at cinder/config/generate_cinder_opts.py13:59
zigojgriffith: So, what you guys is doing, is dynamically generate opts.py, right? Why not sharing this generator directly into oslo.config then?14:00
jgriffithsmcginnis: wonder if we should make that a warning instead of bombing out?14:00
zigoI think it'd be very helpful for other projects too.14:00
*** sgotliv_ has quit IRC14:00
*** haomaiwa_ has quit IRC14:01
jgriffithzigo: indeed14:01
smcginnisjgriffith: Probably safe enough.14:01
zigoI always found having a manual opts.py a shame...14:01
jgriffithzigo: I think once it's "done" that will be the case14:01
zigoAnyway, thanks a lot for the help.14:01
zigoAh ok.14:01
zigoVery good then.14:01
jgriffithI *think*14:01
smcginnisI think the only issue is we had to implement some hacking checks to enforce some of the expected formats to properly parse out config opts.14:01
*** haomaiwa_ has joined #openstack-cinder14:01
jgriffithsmcginnis: +114:01
*** lcurtis has joined #openstack-cinder14:02
jgriffithsmcginnis: but also we need to fix up the script itself I think14:02
jgriffithsmcginnis: and the format of the generated opts file14:02
openstackgerritWilson Liu proposed openstack/cinder: Fix capacity report error in Huawei driver  https://review.openstack.org/22630814:02
jgriffithOn that note... breakfast!!!14:02
smcginnisI do hope it gets to the point where we can share it as a common solution. I definitely don't like having to maintain an opts.py file manually.14:02
smcginnisjgriffith: Enjoy! :)14:03
openstackgerritWilson Liu proposed openstack/cinder: Add hypermetro support for Huawei driver  https://review.openstack.org/20202314:05
tbarronkevincarr1991: when you say 'have you directly attached your compute node to an nfs share?" I would answer "sure", but maybe I don't understand what you are asking :-)14:06
*** eharney has quit IRC14:06
*** mc_nair has joined #openstack-cinder14:07
tbarronkevincarr1991: a compute node can mount an nfs share just like any other machine.  What do you want to do with the nfs share once it is mounted?14:07
kevincarr1991tbarron: I am trying to mount an nfs share to /var/nova/instances . I want to store all instances on the nfs instead of the local disks14:08
*** jungleboyj has joined #openstack-cinder14:08
*** DericHorn-HP has quit IRC14:10
tbarronkevincarr1991: no reason in principle why it won't work.  Whether it's a good idea in a production system depends on the quality of your nfs infra, etc.14:10
* zigo gave up on replacing the upstream tool and just calls it ... :P14:11
kevincarr1991tbarron: the nfs is good quality so it shouldnt bottleneck, but I am getting permission issues when trying to launch an instance14:12
tbarronkevincarr1991: nfs permissions can be fun.  You should be able to run 'ls -l' etc. on the mounted filesystem, see who owns what, check whether root is getting squashed, etc.14:14
kevincarr1991tbarron: when i run "ls-l" it shows that nova owns all of the directories14:15
*** baumann has quit IRC14:15
openstackgerritThang Pham proposed openstack/cinder: Update extend_volume to use versionedobjects  https://review.openstack.org/22869014:16
openstackgerritThang Pham proposed openstack/cinder: Update migrate_volume to use versionedobjects  https://review.openstack.org/22891614:16
tbarronkevincarr1991: I won't be able to debug this with you in detail but I would advise that you check the logs for which commands actually fail and run those manually14:16
tbarronkevincarr1991: as nova, as root, as sudo with rootwrap, whatever.14:17
kevincarr1991tbarron: OK, I will give that a shot.14:17
tbarronkevincarr1991: just like getting cinder driver to start right with nfs, use the logs to see what it does, then manually  get these to work while playing with your exports, mount options, etc.14:18
kevincarr1991tbarron: OK, i guess it is time to fiddle around a lot. Thank you for your advice14:19
*** thangp has quit IRC14:19
tbarronkevincarr1991: good luck!14:19
*** baumann has joined #openstack-cinder14:20
*** thangp has joined #openstack-cinder14:21
*** baumann has quit IRC14:24
*** julim has quit IRC14:25
*** changbl has joined #openstack-cinder14:25
*** baumann has joined #openstack-cinder14:26
*** julim has joined #openstack-cinder14:30
*** julim has quit IRC14:30
*** julim has joined #openstack-cinder14:31
*** julim has quit IRC14:32
*** edtubill has joined #openstack-cinder14:32
*** julim has joined #openstack-cinder14:33
openstackgerritAdriano Freires Rosso proposed openstack/cinder: HNAS iSCSI manage does not work with spaces  https://review.openstack.org/22857514:34
*** baumann has quit IRC14:36
*** juzuluag has joined #openstack-cinder14:37
*** salv-orlando has quit IRC14:39
*** larsks has joined #openstack-cinder14:40
*** crose has joined #openstack-cinder14:47
*** mtanino has joined #openstack-cinder14:47
larsksThe cinder client wants "backup_service" and "backup_url" for the "backup-import" command, but this doesn't seem to match the docs in, e.g., the admin guide.  Are there examples anywhere of how this is supposed to work?14:48
*** duobei has quit IRC14:51
*** vgridnev has quit IRC14:52
*** baumann has joined #openstack-cinder14:52
*** vgridnev has joined #openstack-cinder14:52
*** kevincarr1991 has quit IRC14:52
*** ronis has quit IRC14:53
*** kevincarr1991 has joined #openstack-cinder14:54
*** salv-orlando has joined #openstack-cinder14:54
*** lpetrut has quit IRC14:55
*** merooney has quit IRC14:59
*** martyturner has quit IRC14:59
nikeshmi am trying incremental backup in ubuntu kilo devstack with rest-api, i am following http://developer.openstack.org/api-ref-blockstorage-v2.html, there it is suggesting incremental  value to be True but it is working with true14:59
nikeshmwith rest-api-web-client15:00
*** timcl1 has joined #openstack-cinder15:00
*** haomaiwa_ has quit IRC15:01
nikeshm{"backup": {"description": null, "container": null, "name": null, "volume_id": "a756e67f-367c-489f-96a0-e29faa4479d3", "incremental": True}} was not working15:01
nikeshmbut {"backup": {"description": null, "container": null, "name": null, "volume_id": "a756e67f-367c-489f-96a0-e29faa4479d3", "incremental": true}} worked15:01
nikeshmso is it a bug in manual15:01
nikeshmor normal15:01
*** haomaiwang has joined #openstack-cinder15:01
*** martyturner has joined #openstack-cinder15:02
*** timcl has quit IRC15:02
nikeshmor this is a bug in cinder that it is not taking True15:03
nikeshmor neither because i am missing something :)    ?15:03
*** dims has quit IRC15:04
*** merooney has joined #openstack-cinder15:04
*** crose has quit IRC15:04
*** vgridnev has quit IRC15:05
*** lpetrut has joined #openstack-cinder15:06
*** IlyaG has joined #openstack-cinder15:07
*** kevincarr1991 has quit IRC15:07
*** rushil has joined #openstack-cinder15:09
*** hemnafk is now known as hemna15:10
*** changbl has quit IRC15:11
*** IlyaG has quit IRC15:11
*** vgridnev has joined #openstack-cinder15:12
*** akerr has quit IRC15:13
*** markus_z has quit IRC15:13
xyang1nikeshm: hi, I'll take a look of that15:13
nikeshmxyang1: thanks15:14
*** dims__ has joined #openstack-cinder15:16
*** haomai___ has joined #openstack-cinder15:16
asseline0ne, fyi: just removing enable_v1_api and enable_v2_api from local.conf fails devstack with the same error15:17
e0neasselin: could you please share your local.conf?15:17
*** BharatK has quit IRC15:18
*** haomaiwang has quit IRC15:18
asseline0ne, http://paste.openstack.org/show/474695/15:18
asseline0ne, but removing them and adding to localrc CINDER_ENABLE_V1_API=True works15:20
*** EinstCrazy has quit IRC15:21
e0neasselin: enable_v1_api and enable_v2_api won't work because of keystone endpoint15:21
e0neasselin: there should be a way to add CINDER_ENABLE_V1_API to a local.conf if you don't use localrc15:22
asseline0ne, I think the default (no enable v*_api) should work, but it doesn't currently due to the openstack client bug you mentioned earlier15:22
asseline0ne, we use both local.conf and localrc so it's no issue for us to just enable it there15:23
*** akerr has joined #openstack-cinder15:23
e0neasselin: enable_v1_api enables it only for cinder.15:23
e0neasselin: but to get it working we need to get corresponding keystone endpoint15:24
e0new/o endpoint *client won't work15:24
asseline0ne, I deleted enable_v1_api altogether15:24
asseline0ne, I added CINDER_ENABLE_V1_API to localrc which works15:25
asseline0ne, I bet though that CINDER_ENABLE_V2_API would fail15:25
e0neasselin: AFAIR, there is no CINDER_ENABLE_V2_API option15:26
asseline0ne, ok, so then the default devstack doesn't work today wtih cinder v215:26
e0neasselin: do you mean v1?15:27
*** stevemar has quit IRC15:27
asseline0ne, no, it works wtih v1, does NOT work with v215:27
*** ronis has joined #openstack-cinder15:27
*** stevemar has joined #openstack-cinder15:27
openstackgerritJoel Coffman proposed openstack/cinder: Clean up volume_type_update method  https://review.openstack.org/21382915:28
e0neasselin: hm...15:29
e0neasselin: i'm confused15:29
*** baumann has quit IRC15:29
*** stevemar_ has joined #openstack-cinder15:29
asseline0ne, default devstack does not work today with cinder v2 api because of the openstack client bug you mentioned earlier15:30
openstackgerritJoel Coffman proposed openstack/cinder-specs: Add volume encryption metadata spec  https://review.openstack.org/17605015:30
asseline0ne, asselin_: https://github.com/openstack/python-openstackclient/blob/master/openstackclient/volume/client.py#L22-L24 - looks like we set devault api version to 2 but didn't change API_NAME :(15:30
e0neasselin: actually, after exporting OS_VOLUME_API env variable it works15:31
*** DericHorn-HP has joined #openstack-cinder15:31
e0neI'm sorry, OS_VOLUME_API_VERSION variable15:31
asseline0ne, but that's not happending in devstack today, right?15:31
*** stevemar has quit IRC15:32
xyang1nikeshm: I just tried and True works in the API15:33
*** stevemar_ has quit IRC15:34
e0neasselin: you're right :(15:34
e0neasselin: that's why it works on my env https://github.com/openstack-dev/devstack/blob/71828f756fe3fd28ed9db1ba0812d932a079d782/openrc#L10715:35
openstackgerritTom Swanson proposed openstack/cinder: Dell SC: cgsnapshot-delete doesn't actually delete  https://review.openstack.org/22900715:35
*** haomai___ has quit IRC15:35
e0neasselin: and here is how it works on gates: http://logs.openstack.org/03/221303/5/check/gate-tempest-dsvm-full/775e4cf/logs/devstacklog.txt.gz#_2015-09-28_21_47_33_13115:35
*** ronis has quit IRC15:36
xyang1nikeshm: http://paste.openstack.org/show/474696/15:36
*** rushil has quit IRC15:37
mriedemjgriffith: so is cinder v2 not intended to be backwards compatible with v1?15:38
asseline0ne, we don't have this: --os-volume-api-version 215:39
asseline0ne, why should we need that?15:40
e0neasselin: because of that bug in openstack client:(15:40
*** salv-orl_ has joined #openstack-cinder15:43
asseline0ne, ok is there a bug already, or should I submit one?15:43
e0neasselin: I don't know. anyway, I'll propose a patch in ~1 hour15:43
asseline0ne, ok I'll file a bug15:43
*** larsks has left #openstack-cinder15:44
*** Zhongjun has joined #openstack-cinder15:44
e0neasselin: thanks15:44
jgriffithmriedem: I think people used it as an opportunity to make changes that were not :(15:44
jgriffithmriedem: honestly you probably need to talk to thingee on that whole deal15:44
mriedemjgriffith: ok, i'm just wondering what we should do about that bug fix15:45
mriedemand if it's OK to just accept 200 or 202 in tempest15:45
mriedemmtreinish: ^?15:45
jgriffithmriedem: not sure I folow?15:45
jgriffithmriedem: would you rather see a detection and make sure we match the correct one?15:45
*** salv-orlando has quit IRC15:46
mriedemwell tempest could accept 200 or 202 from volume create15:46
mriedemor cinder v2 could return 200 like v115:46
jgriffithmriedem: right15:46
mriedembut....v2 has been in the wild for awhile now15:46
mtreinishmriedem: no, you have to use the cinder v1 client if you're talking to v1 (with a 200) and the cinder v2 if talking to v215:46
jgriffithmriedem: Mike's argument IIRC was that 200 is bad15:47
jgriffithand abused15:47
mriedemmtreinish: so this is just a bug in tempest?15:47
asseline0ne, https://bugs.launchpad.net/python-openstackclient/+bug/150093815:47
openstackLaunchpad bug 1500938 in python-openstackclient "openstack client doesn't work correctly with cinder v2 api" [Undecided,New]15:47
jgriffithand that 202 was more appropriate15:47
mriedemjgriffith: if it's an rpc cast then yeah 202 is probably correct15:47
jgriffithand a major version rev is "ok" to make such changes, which I kind of agree with15:47
jgriffithmriedem: so the RFC docs state it as "if it's well defined" or some such nonsense :)15:48
*** CBADER_ has joined #openstack-cinder15:48
jgriffith"it" being the API15:48
jgriffiththat 202 is preferred/more-correct15:48
*** eharney has joined #openstack-cinder15:48
*** cbader has quit IRC15:49
mtreinishmriedem: more or less, we need to have tempest default to v2 and fallback to v1 if v2 isn't available15:49
*** jdurgin1 has quit IRC15:49
jgriffithmtreinish: mriedem yeah, I think long term there's a need to have tempest be version aware probably15:50
mtreinishmriedem: the confusion is there isn't a way to tell tempest what version the sc endpoint you told it in config is15:50
jgriffithmriedem: mtreinish whether that's config related, or auto-detect I don't know15:50
e0nemriedem: I fixed such bug in tempest few weeks ago15:51
jgriffithcan't we ask keystone?15:51
jgriffithOh,, no never mind15:51
jgriffithwe can't15:51
mtreinishe0ne: more tests were added15:51
mriedemi will defer any and all service catalog questions to sdague15:51
mtreinishe0ne: also the heat tests had it too15:51
jgriffithmriedem: the problem there is we publish both IIRC15:51
mtreinishwhich I'm not sure why we didn't catch15:51
e0nehttps://review.openstack.org/#/c/209859/ - tempest fix15:52
*** thingee has joined #openstack-cinder15:52
mriedemjgriffith: http://logs.openstack.org/42/227542/7/check/gate-tempest-dsvm-multinode-full/c03f51d/logs/tempest_conf.txt.gz volume-feature-enabled.api_v115:52
mriedemjgriffith: mtreinish: ^ i guess you could just key off that?15:52
thingeelooks like we're good for rc-1. does anyone have anything they want to raise that's pending release blocker that I don't know about?15:52
jgriffiththingee: if the branch is there my migration place holders need pushed15:53
jgriffiththingee: https://review.openstack.org/#/c/228107/15:53
*** merooney has quit IRC15:53
*** martyturner has quit IRC15:53
jgriffiththingee: and the branch is open last I looked15:53
mtreinishmriedem: there is a api_v1/v2 option under volume-feature-enabled15:54
*** martyturner has joined #openstack-cinder15:54
mtreinishand cinder v1 is set to false there15:54
thingeejgriffith: ack15:54
mriedemmtreinish: yeah, i know, see above :)15:54
jgriffiththingee: ahh.. never mind, ttx pointed out waiting til rc2 opens15:54
mtreinishoh, I need to read more15:55
*** martyturner has quit IRC15:55
jgriffithmriedem: mtreinish so I'm happy to change that to read the CONF setting e0ne put in15:57
jgriffithmriedem: mtreinish seems like that would work fine15:57
mtreinishjgriffith: fwiw, I -2d your loosen the response code check patch, that wasn't going to merge :)15:59
jgriffithmtreinish: ?16:00
jgriffithyeah, I don't understand your argument?16:00
jgriffithor "statement"16:00
jgriffithmtreinish: I mean I don't understand how it "breaks"?16:00
mtreinishis 202 a valid response from a v1 request?16:00
jgriffithmtreinish: no16:01
mtreinishthen that patch isn't valid16:01
jgriffithmtreinish: ok, let me try again16:01
mtreinishthose clients are only for v116:01
mtreinishjgriffith: https://github.com/openstack/tempest/blob/master/tempest/services/volume/v2/json/volumes_client.py16:01
jgriffiththen why is it being called here: http://logs.openstack.org/42/227542/7/check/gate-tempest-dsvm-multinode-full/c03f51d/console.html#_2015-09-29_00_36_39_97016:01
mtreinishjgriffith: that's the bug16:02
mtreinishit shouldn't be calling the v1 client if v1 is disabled16:02
jgriffithmtreinish: ahh16:02
openstackgerritTom Swanson proposed openstack/cinder: Dell SC: cgsnapshot-delete doesn't actually delete  https://review.openstack.org/22900716:02
jgriffithmtreinish: ok, now I'm with ya16:02
jgriffithmtreinish: I wasn't aware that there was a seperate v2 client16:02
jgriffithsorry about that16:02
mtreinishno worries16:03
*** kevincarr1991 has joined #openstack-cinder16:05
*** vgridnev has quit IRC16:06
*** mudassirlatif has joined #openstack-cinder16:06
*** martyturner has joined #openstack-cinder16:08
*** ntpttr has joined #openstack-cinder16:08
*** raildo is now known as raildo-afk16:08
*** ronis has joined #openstack-cinder16:10
*** EinstCrazy has joined #openstack-cinder16:10
jordanPjgriffith, mtreinish the interesting bit it here: https://github.com/openstack/tempest/blob/master/tempest/api/compute/base.py#L9216:11
*** alexpilotti has joined #openstack-cinder16:11
jordanPthere's already a switch for v1 vs v216:11
jordanPso I am not sure yet what is going on :)16:12
hemnaanyone seen gorka recently ?16:12
*** leeantho has joined #openstack-cinder16:12
jordanPand the client seems correctly passed to https://github.com/openstack/tempest/blob/master/tempest/common/compute.py#L3016:12
mtreinishjordanP: yeah, that's weird16:12
eharneyhemna: geguileo should be around here somewhere, why?16:20
*** ntpttr has quit IRC16:20
geguileoeharney: What's up?16:20
hemnaoh hey16:21
e0nejordanP: there is no check what api is used for https://github.com/openstack/tempest/blob/master/tempest/common/compute.py#L9216:21
*** EinstCrazy has quit IRC16:21
hemnageguileo, so I'd like to chat about the compare/swap stuff and how it relates (or makes obsolete), my spec on the volume manager lock removal effort16:22
*** IlyaG has joined #openstack-cinder16:22
e0nejordanP: volumes_client or volumes_v2_client should be used depend on config16:22
geguileohemna: ok16:22
*** raildo-afk is now known as raildo16:22
jordanPe0ne, clients is passed as an argument to the create_test_server method16:22
*** pots1 has joined #openstack-cinder16:22
geguileohemna: I remember I looked at those 4 locks we have and thinking that it didn't solve the problem (at least for attach/detach)16:23
geguileohemna: It would for attach/detach from Nova16:23
*** _cjones_ has joined #openstack-cinder16:23
e0nejordanP: yes. it but on line #92 there is clients.volumes_client instead of clients.volumes_v2_client16:23
geguileohemna: Because they use the attaching/detaching intermediary state16:23
hemnageguileo, looks like jgriffith -2'd your patches (just noticed that)16:24
geguileohemna: But we don't use those within cinder16:24
geguileohemna: Yep, I saw that :''-(16:24
*** pots1 has quit IRC16:24
hemnaso, I'm confused16:24
hemnaI'd really like to get this stuff ironed out for M, and early if possible16:25
geguileohemna: I'll have a look at all the comments later, because I think he wants specs or at least clarification on what I'm doing16:25
geguileohemna: Good idea16:25
smcginnisgeguileo: If you could write up something that would be great!16:26
*** jistr has quit IRC16:26
geguileosmcginnis: I was trying to avoid it   XD XD XD16:26
smcginnisgeguileo: Yeah. :)16:26
geguileosmcginnis: But it seems that it is needed after all16:26
smcginnisgeguileo: I do think it will help. I know it's not fun. ;)16:27
geguileoI'll write something16:27
e0neasselin: openstackclient makes me crazy:(16:27
smcginnisgeguileo: Thanks!16:27
geguileosmcginnis: But since you guys think it will help I'll do it  :-)16:27
e0neasselin: it uses cinder api v1 by default because of  https://github.com/openstack/os-client-config/blob/master/os_client_config/defaults.yaml#L1716:27
*** bluex-pl is now known as sw16:27
asseline0ne, yeah....I merged our workarounds. our cis should start passing again, then I'll reenable failure commenting16:28
*** sw has left #openstack-cinder16:28
asseline0ne, ha!16:28
hemnageguileo, so I have a spec that's been up and churned on several times now and I put deps on my spec on your patches :P16:28
geguileohemna: The one about the ing states?16:29
hemnageguileo, https://review.openstack.org/#/c/149894/  (and by mine, I mean my coworkers leeantho, that I've sort of taken over for the time being)16:29
hemnageguileo, yah16:29
hemnaat this point, I'd like to pause and make sure we are all on the same page and get this ironed out.16:29
geguileohemna: Ok, I remember reviewing it long ago, I'll have another look16:29
geguileohemna: I agree16:30
*** baumann has joined #openstack-cinder16:30
hemnageguileo, I do like the idea of the compare/swap for states16:30
hemnabut I do think it could use a spec16:30
geguileo3 cores telling me to write specs, I can't clearly get away without it   XD16:30
hemnafwiw, I was told my spec could just be bugs...so who knows.16:31
hemnaI don't care either way.  What's important is getting everyone on the same page16:31
dulekhemna, geguileo: So from my perspective you two are de-facto working on the same.16:31
hemnadulek, yah kinda16:31
hemnadulek, I'm concerned about the nova <--> cinder interaction16:31
hemnabut it also involves atomic state updates16:32
dulekhemna, geguileo: Atomic state transitions are what makes it possible to remove file locks from c-vol's manager.16:32
hemnait's all related16:32
hemnaand we should all get on the same page, have a unified game plan and see what we can get done and in what order (if any)16:32
geguileodulek: Well, it's related but it doesn't solve the problem16:32
geguileodulek: At least not for attach/detach16:32
hemnadulek, there are other states that are needed16:32
hemnafor 'reading'16:32
dulekgeguileo: Can you provide an example workflow that breaks attach/detach?16:33
hemnato eliminate the delete locks16:33
geguileodulek: Any attach call from inside Cinder16:33
geguileodulek: We don't use the 'attaching' state16:33
dulekhemna: Alternative is checking for volumes being created from the volume we're trying to delete. I think at least…16:33
geguileodulek: Which is used when calling from Nova16:33
hemnadulek, I think that's what I mentioned in my spec16:33
geguileodulek: So the lock does the actual exclusion16:34
hemnait's not clear what's "best"16:34
dulekgeguileo: When do we do attach calls from inside of Cinder?16:34
hemnadulek, create volume from image16:34
hemnanot an actual API call though, it's all internal16:34
dulekShouldn't we just make atomic update here to get that solved?16:35
hemnaso, the confusing part for me is how the atomic updates are done16:35
hemnait looks like the are wrapped in a while16:35
hemnathat seems to go forever ?16:35
hemnawasn't sure if they timeout or whatnot, because we can't just keep trying forever16:36
dulekgeguileo: We're retrying in API? Not returning VolumeIsBusy?16:36
dulekgeguileo: I mean - in your patches.16:36
geguileodulek: In my patches is not really a retry on busy16:37
geguileodulek: It's a retry because there is no longer an issue with performing the operation16:37
e0neasselin: https://review.openstack.org/22903916:37
* geguileo realizes that he really, really, needs a spec for this XD16:37
dulekgeguileo: Yeah, that's what I understood from the code.16:37
* dulek thinks geguileo is right.16:38
dulekhemna: So this doesn't loop forever. Just fails with VolumeIsBusy.16:38
*** apoorvad has joined #openstack-cinder16:38
*** dims__ has quit IRC16:38
*** sghanekar has joined #openstack-cinder16:38
* dulek wrote that as VolumeIsBusty at first…16:38
*** dims__ has joined #openstack-cinder16:39
hemnaI guess it wasn't clear to me16:39
hemnawhile not db.conditional_update()....16:40
geguileoI tried to clarify it with the comments16:40
asseline0ne, thanks16:40
geguileoBut specs will help  :-D16:40
dulekhemna: I was confused at first too.16:40
hemnageguileo, well it simply says "loop until we make a decision"16:41
* hemna is confused16:41
e0neasselin: now, we wait for review, merge and new os-client-config release16:41
e0neit's crazy deps hell16:41
hemnait's not clear when the code decides to bail and give up16:41
geguileohemna: Did I forget to say somewhere that we raise the error/exception?16:42
*** stevemar has joined #openstack-cinder16:42
* geguileo has to review those comments that seem useless XD16:42
hemnamaybe I'm just dumb16:42
geguileoI'm more inclined to think that my communications skills are not that good16:42
asseline0ne, yeah, at least this is the beginning of M16:42
dulekAt least we're confirming that jgriffith was right. ;)16:43
e0neasselin: no-no, I don't want to get such issues during the all M development cycle:)16:43
hemnadulek, :)16:43
*** alexpilotti has quit IRC16:44
*** baumann has quit IRC16:45
*** merooney has joined #openstack-cinder16:46
*** baumann has joined #openstack-cinder16:47
jbernardwinston-d: around?16:53
hemnageguileo, so maybe throw together a spec if you can prior to the meeting16:54
hemnaI'm sure that's what jgriffith is going to ask for anyway.16:54
geguileohemna: Meeting as in tomorrow?16:54
hemna(if possible)16:54
geguileoPuff, I'll do my best16:54
hemnageguileo, https://wiki.openstack.org/wiki/CinderMeetings16:54
scottdageguileo: What about this spec: https://review.openstack.org/#/c/207101/16:54
*** jordanP has quit IRC16:55
scottdaSome smart guy has already written one.16:55
hemnascottda, forgot about that one...so...I'm unclear what the issue is then.16:56
scottdaWell, spec does need to be moved to Mitaka...16:57
*** IlyaG has quit IRC16:57
geguileoscottda: XD XD XD16:57
*** timcl1 has quit IRC16:58
geguileoscottda: I'll update that one and move it to Mitaka16:58
geguileoscottda: Thanks for the reminder, I'm so focussed right now on a SQLAlchemy patch for a bug/feature that is a blocker for my patches that I've forgotten almost everything  %D16:59
*** baumann has quit IRC16:59
*** jgregor has quit IRC16:59
*** garthb has joined #openstack-cinder17:00
scottdaYeah, and I reckon it is the New PTL's job to prioritize the New specs? smcginnis ?17:00
openstackgerritEric Harney proposed openstack/cinder-specs: Support storing volume format info  https://review.openstack.org/16539317:02
*** gouthamr has quit IRC17:04
*** gouthamr has joined #openstack-cinder17:04
*** apoorvad_ has joined #openstack-cinder17:08
*** apoorvad has quit IRC17:12
openstackgerritTom Swanson proposed openstack/cinder: Dell SC: cgsnapshot-delete doesn't actually delete  https://review.openstack.org/22900717:13
*** timcl has joined #openstack-cinder17:13
*** IlyaG has joined #openstack-cinder17:15
*** salv-orl_ has quit IRC17:18
*** timcl1 has joined #openstack-cinder17:18
*** timcl has quit IRC17:19
*** IlyaG has quit IRC17:20
*** asselin has quit IRC17:27
*** leeantho has quit IRC17:27
*** garthb has quit IRC17:27
*** sseverson has quit IRC17:27
*** CBADER_ has quit IRC17:27
*** asselin has joined #openstack-cinder17:28
*** CBADER has joined #openstack-cinder17:28
*** leeantho has joined #openstack-cinder17:28
*** sseverson has joined #openstack-cinder17:28
*** garthb has joined #openstack-cinder17:28
*** sgotliv_ has joined #openstack-cinder17:29
*** sgotliv_ has quit IRC17:30
*** sgotliv__ has joined #openstack-cinder17:30
*** coolsvap has quit IRC17:33
nikeshmxyang1: thanks, this is my observation http://paste.openstack.org/show/474718/17:35
*** julim has quit IRC17:36
*** lpetrut has quit IRC17:36
*** julim has joined #openstack-cinder17:37
*** coolsvap has joined #openstack-cinder17:37
*** mriedem has quit IRC17:38
*** salv-orlando has joined #openstack-cinder17:38
*** krtaylor has quit IRC17:42
nikeshmxyang1 : in openstack manual example, "incremental": "True" is given which is working for me, but "incremental": True did not work for me and "incremental": true worked for me, i am using google chrome advanced rest client17:42
*** mriedem has joined #openstack-cinder17:43
nikeshmsame thing for fasllse17:43
*** sileht_ has joined #openstack-cinder17:44
*** sileht_ is now known as sileht17:45
*** sileht has quit IRC17:45
*** sileht has joined #openstack-cinder17:45
*** rlucio has joined #openstack-cinder17:47
*** sgotliv__ has quit IRC17:51
*** krtaylor has joined #openstack-cinder17:55
xyang1nikeshm: the rest client maybe doing some conversation before sending out the request, I haven't played with it,17:59
xyang1nikeshm: the api is expecting a boolean18:00
*** dims has joined #openstack-cinder18:00
nikeshmxyang1: ok thanks for looking into this18:01
*** dims__ has quit IRC18:02
*** apoorvad_ has quit IRC18:04
*** apoorvad has joined #openstack-cinder18:04
*** Zhongjun has quit IRC18:07
*** delatte has joined #openstack-cinder18:09
*** delattec has quit IRC18:09
*** cdelatte has quit IRC18:09
*** sghanekar_ has joined #openstack-cinder18:17
*** sghanekar has quit IRC18:20
*** delatte has quit IRC18:26
hemnaasselin_, https://review.openstack.org/#/c/22325818:27
asselinhemna, thanks will review18:28
*** boris-42 has quit IRC18:29
*** delatte has joined #openstack-cinder18:29
*** delatte has quit IRC18:30
openstackgerritWalter A. Boring IV (hemna) proposed openstack/os-brick: Add new Connector APIs for path validation  https://review.openstack.org/19976418:30
*** martyturner has quit IRC18:33
*** 32NAAGPES has joined #openstack-cinder18:35
*** cdelatte has joined #openstack-cinder18:35
*** aix has quit IRC18:38
*** sghanekar_ has quit IRC18:38
*** sghanekar_ has joined #openstack-cinder18:38
*** martyturner has joined #openstack-cinder18:41
smcginnisscottda, geguileo: Sure, get it done. :)18:41
smcginnisI'd love to see that work move ahead.18:41
geguileosmcginnis: I'll update the specs tomorrow morning18:41
smcginnisgeguileo: Awesome. Thanks for working on that.18:42
geguileosmcginnis: Congrats!!18:42
smcginnisgeguileo: Thanks!18:42
* geguileo had forgotten to congratulate18:43
*** baumann has joined #openstack-cinder18:43
*** baumann has quit IRC18:45
xyang1smcginnis: congrats!18:46
*** jasondotstar is now known as jasondotstar_afk18:47
*** baumann has joined #openstack-cinder18:48
smcginnisxyang1: Thank you!18:49
*** uberjay has quit IRC18:51
*** e0ne has quit IRC18:54
*** baumann has quit IRC18:55
*** eharney has quit IRC18:55
*** uberjay has joined #openstack-cinder18:59
*** eharney has joined #openstack-cinder19:01
*** stevemar has quit IRC19:01
*** DericHorn-HP has quit IRC19:01
*** DericHorn-HP has joined #openstack-cinder19:02
*** stevemar has joined #openstack-cinder19:08
hemnapatrickeast, ping19:09
*** baumann has joined #openstack-cinder19:11
*** baumann has left #openstack-cinder19:11
*** juzuluag has quit IRC19:15
*** martyturner has quit IRC19:19
*** tsekiyama has joined #openstack-cinder19:22
*** apoorvad has quit IRC19:25
*** kevincarr1991 has quit IRC19:26
*** diegows has joined #openstack-cinder19:31
*** eharney has quit IRC19:32
*** martyturner has joined #openstack-cinder19:33
*** lprice1 has joined #openstack-cinder19:35
*** vgridnev has joined #openstack-cinder19:38
*** sileht has quit IRC19:39
*** krtaylor has quit IRC19:40
*** krtaylor has joined #openstack-cinder19:43
*** eharney has joined #openstack-cinder19:45
*** lprice has joined #openstack-cinder19:45
*** sileht has joined #openstack-cinder19:45
*** lprice1 has left #openstack-cinder19:46
*** sileht has quit IRC19:47
*** sileht has joined #openstack-cinder19:48
*** akerr has quit IRC19:53
*** erlon has quit IRC19:58
*** earlephilhower has quit IRC20:01
nikeshmin liberty just noticed force param in backup create20:07
nikeshm:param force: If True, allows an in-use volume to be backed up.20:08
nikeshmwill it work for incremental too?20:08
*** diegows has quit IRC20:09
*** merooney has quit IRC20:11
nikeshmbetter i should try first20:12
*** e0ne has joined #openstack-cinder20:14
xyang1nikeshm: yes20:20
openstackgerritWalter A. Boring IV (hemna) proposed openstack/os-brick: Add new Connector APIs for path validation  https://review.openstack.org/19976420:20
e0ne"Several vendors are still running cinder api v1. The current vendors we currently have defined need to have default api version set to 1." :(20:20
e0nethey will use it until it will be removed at all20:21
smcginnise0ne: Where's that from?20:21
hemnaML post thread I thinks20:21
e0nesmcginnis, hemna: https://review.openstack.org/22903920:22
*** diegows has joined #openstack-cinder20:22
smcginnisAh, haven't looked lately.20:22
smcginnise0ne: OK, thanks.20:22
patrickeasthemna: pong20:22
hemnathis whole v1 v v2 thing is a bit frustrating20:22
hemnav1 isn't being removed20:22
hemnawe are just trying to change the default to v220:23
e0ne hemna: +220:23
hemnaif we can never do that until the last vendor out in the middle of the pacific ocean forcible updates to v2, then we might as well remove v2.20:23
hemnait's like one extreme or the other.20:23
*** akerr has joined #openstack-cinder20:23
e0nethere are a lot of hardcoded things like os_vol_api=1 in many projects20:24
hemnapatrickeast, https://review.openstack.org/#/c/199764/20:24
hemnapatrickeast, the pure FC CI was puking on my last revision20:24
hemnabut seemed unrelated20:24
e0neI didn't know anything about os-client-config project until last night20:24
*** stevemar has quit IRC20:25
patrickeasthemna: lemmie check the logs, odd are good that its this bug https://bugs.launchpad.net/cinder/+bug/149570120:26
openstackLaunchpad bug 1495701 in OpenStack Compute (nova) "Sometimes Cinder volumes fail to attach with error "The device is not writable: Permission denied"" [Undecided,New]20:26
patrickeastmy FC CI has been failing lots from it20:26
*** apoorvad has joined #openstack-cinder20:26
patrickeastand i've been side tracked (again) from finding a fix20:26
e0neTBH, I didn' fixed all api v2 related issues in our distro. we still use v1 by default20:26
*** salv-orlando has quit IRC20:27
* hemna wonders if we are hit by the same bug20:27
patrickeastand my log server isn't working again20:27
patrickeastoh wait, no thats zuul being dumb20:27
patrickeastnew log server is ok20:28
patrickeastjust posting wrong url20:28
*** dims has quit IRC20:28
*** stevemar has joined #openstack-cinder20:29
*** dims has joined #openstack-cinder20:29
hemnapatrickeast, so that bug shows up in the n-cpu log ?20:29
patrickeastn-cpu, c-bak, or c-vol20:29
patrickeastanywhere we attach a volume and try to do something with it20:29
hemnatrying to find an example of it, so I can look for it in our CI logs20:30
*** akerr has quit IRC20:30
*** svasheka has quit IRC20:30
*** akerr has joined #openstack-cinder20:30
*** martyturner has quit IRC20:30
hemnapatrickeast, does this happen only w/ multipath ?20:31
*** apoorvad has quit IRC20:31
patrickeasthemna: nope, it happens more frequently if i disable multipath, but there is an example here from a single path one in gate http://logs.openstack.org/22/222922/2/check/gate-tempest-dsvm-full-lio/550be5e/logs/screen-n-cpu.txt.gz?level=DEBUG#_2015-09-13_17_34_07_78720:31
patrickeasterr less frequently**20:31
patrickeastso somehow related (maybe extra time to do mpath stuff?)20:32
hemnaencrypted volume stuffs in that one ?20:32
patrickeastyea, for that example20:33
patrickeastbut its not specific to them20:33
hemnadoes it happen outside of the encrypted tests ?20:33
*** salv-orlando has joined #openstack-cinder20:33
hemnaiSCSI vs FC ?20:33
patrickeasti only see it in my FC CI, but in the gate they are only iscsi20:34
patrickeasttheres one in a non-encryption test20:34
*** diegows has quit IRC20:34
*** svasheka has joined #openstack-cinder20:34
* patrickeast pulls up my notes20:34
patrickeasti've got links for ones in c-vol and c-bak too20:34
hemnaso it's hitting in the gate w/ lvm ?20:35
patrickeastyea, very rarely20:35
patrickeastthat logstash query in the bug usually shows like 1-3 in the last 7 days in gate with the n-cpu problem20:35
*** DericHorn-HP has quit IRC20:35
*** bswartz has quit IRC20:37
patrickeasthemna: heres a recent c-vol example http://ec2-54-193-96-24.us-west-1.compute.amazonaws.com/29/213829/5/check/PureFCDriver-tempest-dsvm-volume-multipath/f8d7894/logs/screen-c-vol.txt.gz?level=TRACE20:38
patrickeasthemna: and last but not least, a c-bak one20:39
hemnaso that one looks like a mpath enabled fail20:39
hemnaas the path is /dev/disk/by-id/dm-uuid-mpath-<>20:39
hemnadevice is not writeable20:40
patrickeastyea they all end up getting that20:40
patrickeastfor the n-cpu it says busy or something20:40
*** cdelatte has quit IRC20:40
*** 32NAAGPES has quit IRC20:40
patrickeastbut if you look in the libvirt log file you see the permission denied20:40
*** thangp has quit IRC20:41
patrickeasthemna: like so http://ec2-54-193-96-24.us-west-1.compute.amazonaws.com/28/216728/5/check/PureFCDriver-tempest-dsvm-volume-multipath/bd3618d/logs/libvirt/libvirtd.txt.gz#_2015-09-14_09_00_44_82920:42
patrickeast(the file takes a long time to load)20:42
*** martyturner has joined #openstack-cinder20:42
hemnaso I wonder if I could put a test patch against os-brick to debug log the file perms after the file is discovered20:43
patrickeasti tried that maybe a week ago and beat on it with my local setup20:43
*** Lee1092 has quit IRC20:43
patrickeasttook a few days of the 'run until failure' mode to repro, and the os.stat() output was exactly the same as volumes that work20:43
*** vgridnev has quit IRC20:44
hemnaso fwiw, the flow does some dd's to the volume20:44
hemnawhich work20:44
patrickeastyea read seems ok20:44
hemnaand then another dd right after it doesn't work.20:44
hemnaI think some of those are tests20:46
hemnato see if it can do direct io against the volume or not20:46
patrickeastso that looks like it should have worked then, right?20:46
hemnabut it vomits on the qemu-img convert  immediately after that dd test20:46
patrickeastmaybe a dd with 0 size isn't enough to see the issue?20:47
patrickeastor it changed state?20:47
hemnacould be20:47
hemnapatrickeast, https://github.com/openstack/cinder/blob/master/cinder/image/image_utils.py#L11420:49
hemnaman I don't get it20:52
hemnadoes the test of dd count=0 not test for being able to open and write a byte ?20:53
patrickeasti'm not sure, afaik it does20:53
patrickeastbut maybe not20:53
hemnait's very odd that the dd returns 0 (success) and then immediately qemu-img fails20:53
patrickeasti was thinking as a test to maybe just write some actual data with length > 020:53
patrickeastin theory we overwrite it with image data anyway20:54
patrickeastbut haven't had a chance to try it... busy with that kilo orphan iscsi session thing : /20:54
jungleboyjsmcginnis: Did we get a Tokyo summit session set up for discussing Nova/Cinder cross project issues?20:55
scottdajungleboyj: Yes20:56
scottdaSee row #15:20:56
smcginnisjungleboyj: Not yet I believe. I still need to finalize things. Cross-project meeting is starting soon, so I'll bring that up there.20:56
smcginnisscottda: Oh, well there you go. :)20:56
hemnapatrickeast, I'm at a loss here20:57
jungleboyjCan I add an item there?20:57
scottdaSure, but mriedem gave the caveat that we should be focused for the session, else we'll just bike-shed...20:58
scottdaNote I've already planned on adding a separate session for the bike-shedding20:58
jungleboyjscottda: This was a suggestion from mriedem to discuss the item.20:59
mriedemwhich item?20:59
mriedemthere are many items20:59
jungleboyjVolume extension.21:00
*** earlephilhower has joined #openstack-cinder21:01
*** apoorvad has joined #openstack-cinder21:02
hemnapatrickeast, doing a grep on our log server21:02
hemnalooking for that failure21:02
*** hemna is now known as hemnafk21:03
*** amoturi has left #openstack-cinder21:03
*** ronis has quit IRC21:07
*** mudassirlatif_ has joined #openstack-cinder21:08
*** hemna has joined #openstack-cinder21:09
*** raildo is now known as raildo-afk21:10
*** ociuhandu has joined #openstack-cinder21:10
*** mudassirlatif has quit IRC21:11
*** mudassirlatif_ is now known as mudassirlatif21:11
Swansonpatrickeast: "kilo orphan iscsi session thing"?21:11
patrickeastSwanson: yea i'm seeing this bug in kilo https://bugs.launchpad.net/nova/+bug/138579821:13
openstackLaunchpad bug 1385798 in OpenStack Compute (nova) "Multipath ISCSI connections left open after disconnecting volume with libvirt" [Undecided,Confirmed] - Assigned to Danny Wilson (daniel-wilson)21:13
*** porrua has quit IRC21:13
patrickeastfiguring out how much of the latest os-brick connector code i need to backport to nova to make it go away21:13
*** alexpilotti has joined #openstack-cinder21:14
Swansonpatrickeast: Trying to confirm whether or not we're seeing that now.21:15
hemnapatrickeast, the connect_volume logic changed a bit21:15
patrickeasthemna: yea i spent the morning going through the two line by line to see what changed21:16
patrickeasti got a patch that looks ok21:16
patrickeastbut i'm beating on it a bunch before i write unit tests and try to post a review... i'm not super confident it will be allowed since its such a big change21:16
hemnayah there was quite a bit of changes from the old libvirt code to os-brick21:19
patrickeastluckily it seems that the changes are for the better since this issue doesn't repo on liberty :D21:20
*** vgridnev has joined #openstack-cinder21:24
*** mudassirlatif has quit IRC21:27
*** mudassirlatif has joined #openstack-cinder21:28
patrickeasthemna: Swanson: posted the WIP of it so far https://review.openstack.org/#/c/229152/21:28
patrickeastalso, i apparently can't workflow -1 that change?21:29
patrickeastis that new?21:29
*** mudassirlatif has quit IRC21:30
*** martyturner has quit IRC21:30
Swansonpatrickeast: Test dude just reported something similar.  Trying to figure out what they are running and if it looks the same.  Should be liberty for this.21:32
*** mudassirlatif has joined #openstack-cinder21:33
hemnapatrickeast, updated the review21:34
hemnamissing the flush21:35
*** mudassirlatif has quit IRC21:35
patrickeastSweet thanks21:36
*** e0ne has quit IRC21:38
*** harlowja has quit IRC21:40
*** harlowja has joined #openstack-cinder21:40
*** martyturner has joined #openstack-cinder21:42
*** salv-orl_ has joined #openstack-cinder21:43
*** mudassirlatif has joined #openstack-cinder21:45
*** garthb has quit IRC21:45
*** salv-orlando has quit IRC21:46
*** garthb has joined #openstack-cinder21:46
*** edtubill has quit IRC21:49
*** mudassirlatif has quit IRC21:49
*** devlaps has joined #openstack-cinder21:52
*** mudassirlatif has joined #openstack-cinder21:53
*** hemna has quit IRC21:56
*** hemna has joined #openstack-cinder22:01
*** hemna has quit IRC22:04
*** hemnafk is now known as hemna22:04
*** martyturner has quit IRC22:06
*** ociuhandu has quit IRC22:06
hemnapatrickeast, so my grep in our log server returned 0 results looking for "The device is not writable"22:07
*** jungleboyj has quit IRC22:08
*** mriedem is now known as mriedem_away22:09
*** mriedem_away has quit IRC22:09
hemnapatrickeast, fwiw, my os-brick patch has a failure on Pure FC CI22:11
hemnapatrickeast, our friend....Permission denied22:11
*** boris-42 has joined #openstack-cinder22:12
*** markstur has joined #openstack-cinder22:12
*** markstur has left #openstack-cinder22:14
*** marcusvrn_ has quit IRC22:15
hemnapatrickeast, http://ec2-54-193-96-24.us-west-1.compute.amazonaws.com/64/199764/11/check/PureFCDriver-tempest-dsvm-volume-multipath-brick/7f0e88c/logs/screen-c-vol.txt.gz?#_2015-09-29_20_57_18_44022:17
hemnathat test tries to read a byte22:17
hemnawhich works22:17
hemnathen the same story happens later22:17
hemnapatrickeast, http://ec2-54-193-96-24.us-west-1.compute.amazonaws.com/64/199764/11/check/PureFCDriver-tempest-dsvm-volume-multipath-brick/7f0e88c/logs/screen-c-vol.txt.gz?#_2015-09-29_20_57_18_85822:17
hemnathat works, then the next line is the fail22:18
*** gouthamr has quit IRC22:20
*** Yogi11 has quit IRC22:23
hemnapatrickeast, https://review.openstack.org/#/c/199764/22:29
hemnaso the failure on that guy happens when multipath is enabled in cinder.conf22:29
*** vgridnev has quit IRC22:30
hemnaalso enabled in nova.conf22:30
*** edtubill has joined #openstack-cinder22:32
*** mc_nair has quit IRC22:34
*** stevemar has quit IRC22:38
*** lcurtis has quit IRC22:41
*** jungleboyj has joined #openstack-cinder22:41
*** mudassirlatif has quit IRC22:47
*** alejandrito has quit IRC22:56
*** gouthamr has joined #openstack-cinder22:58
*** mudassirlatif has joined #openstack-cinder22:59
patrickeasthemna: sry was in a meeting23:09
patrickeastyea all of my ci jobs are running multipath enabled for cinder and nova23:10
patrickeastif i turn it off the FC ci will pass like 90% of the time instead of <50% with multipath on23:11
*** cdelatte has joined #openstack-cinder23:15
*** salv-orl_ has quit IRC23:19
*** xyang1 has quit IRC23:19
*** dims_ has joined #openstack-cinder23:19
*** mudassirlatif has quit IRC23:21
*** dims has quit IRC23:22
*** eharney has quit IRC23:23
*** julim has quit IRC23:23
*** mudassirlatif has joined #openstack-cinder23:23
*** apoorvad_ has joined #openstack-cinder23:25
*** apoorvad_ has quit IRC23:26
*** apoorvad has quit IRC23:26
*** apoorvad has joined #openstack-cinder23:26
*** apoorvad has quit IRC23:27
*** Apoorva has joined #openstack-cinder23:30
*** cvstealt1 has joined #openstack-cinder23:31
*** mdbooth_ has joined #openstack-cinder23:31
*** dims_ has quit IRC23:31
*** dims_ has joined #openstack-cinder23:32
*** leeantho_ has joined #openstack-cinder23:32
*** bauzas_ has joined #openstack-cinder23:32
*** mdbooth has quit IRC23:32
*** bauzas has quit IRC23:32
*** johnthetubaguy has quit IRC23:32
*** tristanC has quit IRC23:32
*** bswartz has joined #openstack-cinder23:32
*** mdbooth_ is now known as mdbooth23:32
*** bauzas_ is now known as bauzas23:32
*** tristanC has joined #openstack-cinder23:33
*** leeantho has quit IRC23:33
*** _cjones_ has quit IRC23:33
*** albertom has quit IRC23:33
*** cvstealth has quit IRC23:33
*** brad[] has quit IRC23:33
*** Apoorva is now known as apoorvad23:33
*** jungleboyj has quit IRC23:33
*** albertom has joined #openstack-cinder23:33
*** jungleboyj has joined #openstack-cinder23:33
*** johnthetubaguy has joined #openstack-cinder23:34
*** _cjones_ has joined #openstack-cinder23:35
*** akerr is now known as akerr_away23:37
openstackgerritSonia Ghanekar proposed openstack/cinder: Add multi-initiator extra-spec for Nimble driver  https://review.openstack.org/22918523:37
*** edmondsw has quit IRC23:41
*** julim has joined #openstack-cinder23:42
*** mudassirlatif has quit IRC23:43
*** hemna is now known as hemnafk23:43
*** garthb has quit IRC23:44
*** julim has quit IRC23:47
*** julim has joined #openstack-cinder23:48
*** mudassirlatif has joined #openstack-cinder23:51
*** julim has quit IRC23:52

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