Monday, 2015-02-23

*** scottda_ has quit IRC00:04
*** Manshoon has joined #openstack-cinder00:04
*** scottda_ has joined #openstack-cinder00:08
*** ebalduf has joined #openstack-cinder00:09
*** diegows has joined #openstack-cinder00:12
*** Manshoon has quit IRC00:13
*** ebalduf has quit IRC00:13
*** markvoelker has joined #openstack-cinder00:16
*** Manshoon has joined #openstack-cinder00:19
*** scottda_ has quit IRC00:20
*** markvoelker has quit IRC00:20
*** scottda_ has joined #openstack-cinder00:23
*** scottda_ has quit IRC00:27
*** Manshoon has quit IRC00:32
*** Yogi1 has joined #openstack-cinder00:48
*** markvoelker has joined #openstack-cinder00:50
*** markvoelker has quit IRC00:55
*** scottda_ has joined #openstack-cinder01:01
*** scottda_ has quit IRC01:04
*** harlowja_at_home has joined #openstack-cinder01:06
*** ebalduf has joined #openstack-cinder01:10
*** boris-42 has joined #openstack-cinder01:13
*** ebalduf has quit IRC01:14
*** kbyrne has quit IRC01:23
*** kbyrne has joined #openstack-cinder01:25
*** ho has joined #openstack-cinder01:25
*** markvoelker has joined #openstack-cinder01:51
*** Yogi1 has quit IRC01:55
*** markvoelker has quit IRC01:57
*** scottda_ has joined #openstack-cinder01:59
*** markvoelker has joined #openstack-cinder02:00
*** diegows has quit IRC02:04
*** ebalduf has joined #openstack-cinder02:11
*** ebalduf has quit IRC02:15
*** markvoelker has quit IRC02:26
*** scottda_ has quit IRC02:26
*** markvoelker has joined #openstack-cinder02:27
*** scottda_ has joined #openstack-cinder02:29
*** markvoelker has quit IRC02:31
*** vilobhmm has quit IRC02:36
*** kaisers1 has joined #openstack-cinder02:47
*** kaisers has quit IRC02:48
*** scottda_ has quit IRC02:58
*** scottda_ has joined #openstack-cinder03:02
*** scottda_ has quit IRC03:04
*** mriedem has left #openstack-cinder03:06
*** jcru has joined #openstack-cinder03:10
*** ebalduf has joined #openstack-cinder03:11
*** ebalduf has quit IRC03:16
*** tellesnobrega_ has joined #openstack-cinder03:31
*** tellesnobrega_ has quit IRC03:31
*** annashen has joined #openstack-cinder03:47
*** pdeore_ has joined #openstack-cinder03:52
*** harlowja_at_home has quit IRC03:53
*** pdeore_ has quit IRC04:00
*** garthb has joined #openstack-cinder04:11
*** garthb_ has joined #openstack-cinder04:11
*** garthb_ has quit IRC04:12
*** garthb has quit IRC04:12
*** ebalduf has joined #openstack-cinder04:12
*** vilobhmm has joined #openstack-cinder04:12
*** Mandell has joined #openstack-cinder04:15
*** ebalduf has quit IRC04:16
*** deepakcs has joined #openstack-cinder04:19
*** rwsu-afk has quit IRC04:19
*** annashen has quit IRC04:22
*** annashen has joined #openstack-cinder04:24
*** annashen has quit IRC04:24
*** annashen has joined #openstack-cinder04:27
*** vilobhmm1 has joined #openstack-cinder04:28
*** BharatK has joined #openstack-cinder04:28
*** vilobhmm has quit IRC04:30
*** bswartz has quit IRC04:34
*** bswartz has joined #openstack-cinder04:35
*** rushiagr_away is now known as rushiagr04:56
*** deepakcs has quit IRC05:00
*** pradipta has joined #openstack-cinder05:12
*** ebalduf has joined #openstack-cinder05:13
*** ebalduf has quit IRC05:17
*** jcru has quit IRC05:18
*** annashen has quit IRC05:31
*** anshul has joined #openstack-cinder05:41
*** anshul has quit IRC05:42
*** anshul has joined #openstack-cinder05:43
*** pcaruana has quit IRC05:45
*** Ilja has joined #openstack-cinder05:57
*** bkopilov has quit IRC06:06
openstackgerritOpenStack Proposal Bot proposed openstack/cinder: Imported Translations from Transifex
*** ankit_ag has joined #openstack-cinder06:08
*** Tross has joined #openstack-cinder06:10
*** ebalduf has joined #openstack-cinder06:14
*** ebalduf has quit IRC06:18
*** bkopilov has joined #openstack-cinder06:19
*** rushiagr is now known as rushiagr_away06:20
*** coolsvap_ is now known as coolsvap06:27
*** Lee1092 has joined #openstack-cinder06:27
openstackgerritAnish Bhatt proposed openstack/cinder: Improve error handling in refactored Tgt driver
*** nshaikh has joined #openstack-cinder06:41
*** nshaikh has quit IRC06:44
*** nshaikh has joined #openstack-cinder06:44
*** deepakcs has joined #openstack-cinder06:45
*** navid__ has joined #openstack-cinder06:46
*** nshaikh has left #openstack-cinder06:47
*** nshaikh has joined #openstack-cinder06:47
*** navid__ has quit IRC06:48
*** cnesa has joined #openstack-cinder07:00
*** vilobhmm1 has quit IRC07:01
*** afazekas_ has joined #openstack-cinder07:06
*** vilobhmm has joined #openstack-cinder07:14
*** ebalduf has joined #openstack-cinder07:14
*** ebalduf has quit IRC07:19
*** rushiagr_away is now known as rushiagr07:34
*** vilobhmm1 has joined #openstack-cinder07:42
*** alecv has joined #openstack-cinder07:44
*** vilobhmm has quit IRC07:44
*** Ilja has quit IRC07:46
*** bkopilov has quit IRC07:49
*** bkopilov has joined #openstack-cinder07:49
*** e0ne has joined #openstack-cinder07:49
*** e0ne is now known as e0ne_07:49
*** Miouge has joined #openstack-cinder07:51
*** e0ne_ is now known as e0ne07:53
*** Ilja has joined #openstack-cinder07:57
*** kaisers1 has quit IRC07:58
*** Mandell has quit IRC08:00
*** kaisers has joined #openstack-cinder08:05
*** kaisers has quit IRC08:10
*** theanalyst has quit IRC08:14
*** kaisers has joined #openstack-cinder08:14
*** theanalyst has joined #openstack-cinder08:15
*** ebalduf has joined #openstack-cinder08:15
*** vilobhmm1 has quit IRC08:16
*** jcru has joined #openstack-cinder08:19
*** ebalduf has quit IRC08:19
*** jcru has quit IRC08:24
*** ndipanov has joined #openstack-cinder08:24
*** enterprisedc has quit IRC08:31
*** Miouge_ has joined #openstack-cinder08:37
*** Miouge has quit IRC08:38
*** Miouge_ is now known as Miouge08:38
*** ronis has joined #openstack-cinder08:42
*** sgotliv has joined #openstack-cinder08:45
*** TobiasE has joined #openstack-cinder08:49
*** emaisne has joined #openstack-cinder08:54
*** rushiagr is now known as rushiagr_away08:54
*** jistr has joined #openstack-cinder08:55
*** jordanP has joined #openstack-cinder08:56
*** e0ne has quit IRC08:57
*** dulek has joined #openstack-cinder08:59
*** aarefiev has quit IRC09:06
*** aarefiev has joined #openstack-cinder09:11
*** emaisne_ has joined #openstack-cinder09:16
*** ebalduf has joined #openstack-cinder09:16
openstackgerritMichal Dulko proposed openstack/cinder: Backup object
*** emaisne has quit IRC09:19
*** emaisne_ has quit IRC09:20
*** ebalduf has quit IRC09:20
*** emaisne has joined #openstack-cinder09:23
openstackgerritAnish Bhatt proposed openstack/cinder: Improve error handling in refactored Tgt driver
openstackgerritAnish Bhatt proposed openstack/cinder: Improve error handling in refactored Tgt driver
openstackgerritAnton Arefiev proposed openstack/cinder: Import only modules: H302
*** sgotliv has quit IRC09:47
*** emaisne has quit IRC09:51
*** rushiagr_away is now known as rushiagr09:58
*** e0ne has joined #openstack-cinder09:59
openstackgerritAnton Arefiev proposed openstack/cinder: Add entry create and cast tasks to manage workflow
openstackgerritAnton Arefiev proposed openstack/cinder: Fix lvm manage existing volume
*** sgotliv has joined #openstack-cinder10:12
*** sgotliv has quit IRC10:12
*** sgotliv has joined #openstack-cinder10:12
*** pdeore_ has joined #openstack-cinder10:16
*** ebalduf has joined #openstack-cinder10:17
*** marcusvrn has joined #openstack-cinder10:18
*** e0ne is now known as e0ne_10:21
*** ebalduf has quit IRC10:21
*** marcusvrn1 has joined #openstack-cinder10:23
*** marcusvrn has quit IRC10:25
*** marcusvrn1 has quit IRC10:25
*** e0ne_ is now known as e0ne10:28
*** marcusvrn has joined #openstack-cinder10:29
*** sgotliv has quit IRC10:34
*** kbyrne has quit IRC10:49
*** kbyrne has joined #openstack-cinder10:49
*** sgotliv has joined #openstack-cinder10:49
*** lpetrut has joined #openstack-cinder10:53
openstackgerritAnton Arefiev proposed openstack/cinder: Fix lvm manage existing volume
*** dulek_ has joined #openstack-cinder10:57
*** dulek has quit IRC11:00
*** kallebe has joined #openstack-cinder11:05
*** jaypipes has joined #openstack-cinder11:05
*** sgotliv has quit IRC11:09
openstackgerritVipin Balachandran proposed openstack/cinder: VMware: Skip vSAN for preallocated image download
*** ebalduf has joined #openstack-cinder11:17
*** ebalduf has quit IRC11:22
*** cnesa has quit IRC11:27
*** aix has joined #openstack-cinder11:28
*** nellysmitt has joined #openstack-cinder11:29
*** diegows has joined #openstack-cinder11:35
*** ho has quit IRC11:36
openstackgerritnikeshmahalka proposed openstack/cinder: Add CHAP persistence to SCST target helper
*** pradipta has quit IRC11:54
openstackgerritKanchan Gupta proposed openstack/cinder: Checking name for volume creation in cinder.
*** TobiasE1 has joined #openstack-cinder12:00
*** TobiasE has quit IRC12:00
*** Mandell has joined #openstack-cinder12:01
*** ganso_ has joined #openstack-cinder12:05
*** Mandell has quit IRC12:05
*** Miouge_ has joined #openstack-cinder12:06
*** Miouge has quit IRC12:06
*** Miouge_ is now known as Miouge12:06
*** Ilja has quit IRC12:07
*** Miouge has quit IRC12:09
*** emaisne has joined #openstack-cinder12:09
*** Ilja has joined #openstack-cinder12:09
*** Miouge has joined #openstack-cinder12:10
*** aviramb has quit IRC12:17
*** sgotliv has joined #openstack-cinder12:18
*** ebalduf has joined #openstack-cinder12:18
*** sgotliv has quit IRC12:18
*** sgotliv has joined #openstack-cinder12:19
*** Miouge has quit IRC12:20
*** Miouge has joined #openstack-cinder12:21
*** ebalduf has quit IRC12:23
*** cdelatte has joined #openstack-cinder12:25
*** Miouge has quit IRC12:27
*** lpetrut has quit IRC12:31
*** dulek_ has quit IRC12:43
*** TobiasE has joined #openstack-cinder12:43
*** Miouge has joined #openstack-cinder12:44
*** IanGovett has joined #openstack-cinder12:45
*** zigo has quit IRC12:52
*** marcusvrn1 has joined #openstack-cinder12:53
*** diegows has quit IRC12:53
*** zigo has joined #openstack-cinder12:56
*** marcusvrn has quit IRC12:57
*** lpabon has joined #openstack-cinder12:59
*** deepakcs has quit IRC13:00
*** BharatK has quit IRC13:01
*** dulek_ has joined #openstack-cinder13:07
*** dulek_ is now known as dulek13:18
*** akerr has joined #openstack-cinder13:19
*** ebalduf has joined #openstack-cinder13:19
*** bswartz has quit IRC13:20
*** ebalduf has quit IRC13:23
*** Ilja has quit IRC13:25
*** marcusvrn has joined #openstack-cinder13:28
openstackgerritOpenStack Proposal Bot proposed openstack/cinder: Updated from global requirements
*** marcusvrn1 has quit IRC13:31
*** bill_az has joined #openstack-cinder13:40
*** sgotliv has quit IRC13:43
*** sgotliv has joined #openstack-cinder13:43
*** timcl has joined #openstack-cinder13:44
*** Guest30952 has joined #openstack-cinder13:48
*** bswartz has joined #openstack-cinder13:50
*** jungleboyj_ has quit IRC13:53
*** dulek_ has joined #openstack-cinder13:59
*** dulek has quit IRC14:01
*** cdelatte has quit IRC14:04
*** ankit_ag has quit IRC14:08
*** kaufer has joined #openstack-cinder14:09
*** lpabon has quit IRC14:10
openstackgerritVipin Balachandran proposed openstack/cinder: VMware: Refactor initialize_connection unit tests
*** diegows has joined #openstack-cinder14:12
*** Guest28285 is now known as mgagne14:13
*** mgagne has joined #openstack-cinder14:13
*** tsekiyama has joined #openstack-cinder14:13
*** dulek_ has quit IRC14:14
*** Yogi1 has joined #openstack-cinder14:16
*** dustins has joined #openstack-cinder14:19
openstackgerritVipin Balachandran proposed openstack/cinder: VMware:Use datastore selection logic in new module
*** ebalduf has joined #openstack-cinder14:20
*** mtanino has quit IRC14:22
openstackgerritVipin Balachandran proposed openstack/cinder: VMware: Relocate volume to compliant datastore
*** ebalduf has quit IRC14:24
*** primechuck has joined #openstack-cinder14:25
*** Tross has quit IRC14:27
*** anshul has quit IRC14:29
*** e0ne is now known as e0ne_14:30
*** e0ne_ is now known as e0ne14:31
*** nshaikh has quit IRC14:32
openstackgerritVipin Balachandran proposed openstack/cinder: VMware: Skip vSAN for preallocated image download
*** nlevinki has joined #openstack-cinder14:38
*** dulek has joined #openstack-cinder14:38
*** mriedem has joined #openstack-cinder14:40
*** Guest30952 is now known as annegentle14:42
*** rushiagr is now known as rushiagr_away14:42
*** timcl has quit IRC14:43
*** tsekiyama has quit IRC14:43
*** timcl has joined #openstack-cinder14:43
*** e0ne is now known as e0ne_14:44
*** thangp has joined #openstack-cinder14:47
*** beekneemech is now known as bnemec14:48
*** superdan is now known as dansmith14:53
*** e0ne_ has quit IRC14:54
*** e0ne has joined #openstack-cinder14:56
*** deepakcs has joined #openstack-cinder14:57
*** patrickeast has joined #openstack-cinder14:58
*** rushil has joined #openstack-cinder14:59
*** pdeore_ has quit IRC15:02
*** dulek has quit IRC15:03
*** BharatK has joined #openstack-cinder15:06
*** patrickeast_ has joined #openstack-cinder15:06
*** patrickeast has quit IRC15:07
*** patrickeast_ is now known as patrickeast15:07
*** tbarron has joined #openstack-cinder15:11
*** amoturi has quit IRC15:11
*** amoturi has joined #openstack-cinder15:11
*** Miouge has quit IRC15:15
*** nlevinki has quit IRC15:16
*** sigmavirus24_awa is now known as sigmavirus2415:16
*** rushiagr_away is now known as rushiagr15:18
*** enterprisedc has joined #openstack-cinder15:18
*** Miouge has joined #openstack-cinder15:20
*** ebalduf has joined #openstack-cinder15:21
*** Miouge has quit IRC15:23
*** ebalduf has quit IRC15:25
*** Miouge has joined #openstack-cinder15:26
openstackgerritMatthew Treinish proposed openstack/cinder: Use subunit-trace to enable output during unit test runs
*** kallebe has quit IRC15:31
*** nlevinki has joined #openstack-cinder15:32
openstackgerritSteven Kaufer proposed openstack/cinder: Generic filter support for volume queries
*** zigo_ has joined #openstack-cinder15:35
*** jcru has joined #openstack-cinder15:36
*** jcru has quit IRC15:37
*** alexpilotti has joined #openstack-cinder15:37
*** jcru has joined #openstack-cinder15:37
*** jcru has quit IRC15:38
*** jcru has joined #openstack-cinder15:38
*** Miouge has quit IRC15:39
*** emaisne_ has joined #openstack-cinder15:40
*** asselin_ has joined #openstack-cinder15:40
*** nlevinki has quit IRC15:42
*** jungleboyj_ has joined #openstack-cinder15:43
*** emaisne has quit IRC15:44
*** emaisne_ has quit IRC15:44
*** hemnafk is now known as hemna15:44
*** Miouge has joined #openstack-cinder15:45
*** pschaef has joined #openstack-cinder15:45
*** tsekiyama has joined #openstack-cinder15:45
*** aix has quit IRC15:47
*** Miouge has quit IRC15:50
*** rushiagr is now known as rushiagr_away15:55
openstackgerritJohn Griffith proposed openstack/cinder: Remove warnings for long vgs and lvs calls
*** lpetrut has joined #openstack-cinder15:56
*** xyang1 has joined #openstack-cinder15:57
guitarzanjgriffith: did you get that figured out?15:57
jgriffithguitarzan: which?15:57
jgriffithguitarzan: the DB thing?15:57
guitarzanthe lvm filter thing15:57
jgriffithguitarzan: haha... kinda15:57
jgriffithguitarzan: I figured out that the rescue stuff is an absolute PITA15:58
jgriffithguitarzan: and that negative tests in tempest are even more so :)15:58
*** Mandell has joined #openstack-cinder15:59
*** kallebe has joined #openstack-cinder16:00
*** patrickeast has quit IRC16:00
jgriffithanybody running kernel 3.16 on 14.04?16:02
*** jungleboyj_ is now known as jungleboyj16:02
*** ndipanov has quit IRC16:04
jgriffithahh well.. how bad can it be :)16:04
openstackgerritSteven Kaufer proposed openstack/cinder: Generic filter support for volume queries
openstackgerritSteven Kaufer proposed openstack/cinder: Replication status periodic task optimization
hemnawhat's blowing up this fine Monday?16:06
hemna3.13 here16:07
jgriffithhemna: pishh... that's ancient :)16:08
*** TobiasE has quit IRC16:09
*** Miouge has joined #openstack-cinder16:10
Swansonjgriffith: I was going to mention that the only way I could get multibackend to install correctly was if I put the CINDER_ENABLED_BACKENDS=driver:backend,driver2:backend2 line above the post-config bit.16:14
*** Miouge has quit IRC16:14
jgriffithSwanson: yep....16:14
jgriffithSwanson: my snippet only did the names, not the types :(16:14
jgriffithbahhhh... I messed you up16:15
Swansonjgriffith:  Not an issue as you posted
openstackLaunchpad bug 1424104 in Cinder "Unable to restart c-vol services when using multi-backend due to set_voldb_empty_at_startup_indicator" [Undecided,New]16:16
Swansonjgriffith: I've seen that one on and off.16:17
jgriffithSwanson: yeah, that one is a big scary IMO16:17
jungleboyjhemna: 3.13 ?16:19
hemnaLinux devstack-master 3.13.0-32-generic16:19
*** dannywilson has joined #openstack-cinder16:20
jungleboyjhemna: :-)16:21
*** devlaps has joined #openstack-cinder16:21
*** mtanino has joined #openstack-cinder16:21
*** rushiagr_away is now known as rushiagr16:21
*** ebalduf has joined #openstack-cinder16:21
*** dannywil_ has joined #openstack-cinder16:23
*** dannywilson has quit IRC16:24
*** rmesta has joined #openstack-cinder16:25
*** ebalduf has quit IRC16:25
*** jdandrea has quit IRC16:28
*** jdandrea has joined #openstack-cinder16:29
*** pschaef has quit IRC16:32
*** e0ne has quit IRC16:32
*** nkrinner has quit IRC16:33
*** Mandell has quit IRC16:33
*** e0ne has joined #openstack-cinder16:33
*** ronis has quit IRC16:34
*** lpetrut has quit IRC16:35
hemnawhen you have time16:35
hemnaI'm trying to get all the brick related things to land16:36
hemnaso we can cut a build to pypi16:36
jungleboyjhemna: Ok, will take a look.16:37
*** deepakcs has quit IRC16:39
hemnathanks man16:40
*** devlaps has quit IRC16:42
*** Miouge has joined #openstack-cinder16:42
jgriffithhemna: jungleboyj
*** Miouge has quit IRC16:48
*** devlaps has joined #openstack-cinder16:48
hemnak I'll take a look16:48
hemnacurrently reviewing the CG update16:48
jgriffithhemna: jungleboyj the thing that I don't really get is there seems to be a completely different impl for the image-copy stuff that I just noticed this morning16:48
jgriffithbut I could be completely wrong :)16:48
jgriffithhemna: lemme see if I can find it again16:49
*** sgotliv has quit IRC16:50
jgriffithhemna: why not use the same structure?16:50
*** deepakcs has joined #openstack-cinder16:51
hemna2 different things16:51
hemna1 is for multipath support16:51
hemnathe other is for alternative paths to the same volume16:51
jgriffithhehe... but that's my point16:51
jgriffiththey "could" very easily and probably "should" IMHO be the same thing16:52
jgriffithalternative path, vs multi path16:52
jgriffithseems like a distinction without much of a difference16:52
hemnaso there needs to be additional data then passed back to brick/nova to tell it what it needs to do then.16:52
hemnaif it's trying to do a multipath attach vs. a single attach is different16:52
jgriffithhemna: sure16:52
hemnajust because you have a list of target_portals doesn't mean you want multipath16:53
hemnaso I guess that's why there was a distinction16:53
jgriffithhemna: understood16:53
jgriffithhemna: but why not just use the same data structure and a separate 'thing' to specify whihc you want?16:53
hemnayah I suppose that would be ok16:53
jgriffithhemna: if you look at the DS's they seem to duplicate an awful lot of code16:54
openstackgerritDanny Wilson proposed openstack/cinder: Over subscription for Pure Storage iSCSI driver.
hemnamultipath: True16:54
jungleboyjSeems sensible to me.16:54
jgriffithhemna: and like i said in my review, I really don't like the "if this format, else that format" thing16:54
jgriffithThat's a compatability/support nightmare waiting to happen IMHO16:54
hemnaok I'm cool with that16:54
jgriffithhemna: The only other question is should they both be changed :)16:55
jgriffithhemna: to something that plugs seemlessly into the existing model16:55
jgriffithhemna: you're way more involved in this than I so I'll defer to you16:55
hemnaI like target_portals16:55
jgriffithhemna: yeah... I do too16:56
jgriffithhemna: but that means we're "stuck" until a change is made on the Nova side16:56
hemnaand an additional flag telling brick/nova what to do with them.  Multipath: True16:56
hemnaif that's not there, then it's considered alternative portals16:56
nikesh_vedamsi am trying to run cinder tests under /opt/stack/cinder but getting errors
hemnajgriffith, yah I think so16:56
nikesh_vedamsis any facing this issue,i am using ubuntu 14.0416:57
jgriffithnikesh_vedams: try deleting/rebuilding your venv16:57
jgriffithnikesh_vedams: it looks to me like youre venv never actually built correctly16:57
jgriffithhemna: so that's up to you guys I suppose, but I just thought we could make the DS in such a way that the existing code would all still "just work"16:58
nikesh_vedamsokk let me check16:58
hemnajgriffith, I don't think the nova side has landed16:58
smcginnisnikesh_vedams: ./ -f16:58
jgriffithhemna: no it hasn't... and it's not going to based on the looks of it16:58
hemnaI think they were waiting on brick16:58
hemnaso that part is good16:58
jgriffithhemna: hence my concern for having this stuff half in/ half out16:58
hemnaso we can fix this here in K before it's release16:59
*** pdeore_ has joined #openstack-cinder16:59
jgriffithThat's what I'm thinking... if you and Tomoki agree with me16:59
hemnayah I do16:59
hemnafwiw, the spec was approved though.  and you did +2 it :P17:00
xyang1jgriffith: hemna there are already a couple of drivers made changes based on the multipath enhancement17:00
jgriffithhemna: sure17:00
hemnaxyang1, which ?17:00
jgriffithhemna: but also TBF that was two specs17:01
xyang1jgriffith: hemna emc vnx and dell.  It is backward compatible17:01
jgriffithhemna: and I'm more concerned with impl than with the concept17:01
xyang1So it works with current nova code too17:01
hemnaxyang1, well, the current nova code ignores the new entries17:01
hemnatarget_portals, target_alterative_portals17:01
xyang1I know17:01
hemnaso yes, it 'works'17:01
xyang1I mean the driver still keeps the existing behavior17:02
xyang1If nova side gets in, it will work with the new cide17:02
jgriffithhemna: xyang1 so the one wth the portal lists etc that already landed I didn't really have an issue with17:02
hemnaI don't think we want the nova side to land though17:02
jgriffithxyang1: hemna the concern for me was when I saw this one:
jgriffithxyang1: hemna that IMO could be consolidated17:03
*** pdeore_ has quit IRC17:03
jgriffithI think hemna and I are on the same page17:03
nikesh_vedamsjgriffith smcginnis : rm -rf .venv    ./ targets  and selected yes for virtual environment and getting this errror
hemnato just use target_portals17:03
hemnawith the multipath: false17:03
xyang1Ok, i am fine with that17:03
jgriffithnikesh_vedams: yeah, you need the pre-reqs17:04
jgriffithnikesh_vedams: "sudo apt-get install python-dev libssl-dev python-pip git-core libxml2-dev libxslt-dev pkg-config libffi-dev libpq-dev libmysqlclient-dev libvirt-dev graphviz libsqlite3-dev python-tox"17:04
hemnaxyang1, do you have the EMC review that uses this change ?17:04
Swansonjgriffith: One other multibackend thing.  I've had a failure with a DB deadlock message in addition to the failure you posted.  (Haven't seen it for a while.)  I think that is just a different flavor of the same bug.17:05
*** rwsu has joined #openstack-cinder17:05
* tsekiyama is scrolling back17:05
jungleboyjhemna: jgriffith I think it makes sense to use the same data structure.17:05
hemnaxyang1, ?17:06
jgriffithand by the way, I'm NEVER approving another spec or review if people keep coming back and saying "you approved it" any time I log a bug or raise a suggestion17:06
hemnathe other 417:06
xyang1Sorry, wrong one17:06
tsekiyamahmm, using target_portals/iqns/luns even when multipath is False ?17:07
jgriffithI've had that thrown in my face way too many times the past couple monts17:07
jgriffithtsekiyama: oh... hellooooo!17:07
nikesh_vedamsjgriffith : thanks17:07
jgriffithfor some reason I thought you'd be in a "sleeping" Time zone17:07
tsekiyamajgriffith: :)17:08
hemnaxyang1, so it looks like the emc code could be changed to simply return multipath: True when it populates target_portals.17:08
jgriffithtsekiyama: so did you see my thoughts on the review?  And the conversation with hemna ?17:08
hemnaxyang1, that patch doesn't include support for target_alternative_portals17:08
tsekiyamaWhen multipath = True, it omits target_portal/iqn/lun and add target_portals/iqns/luns. it is ok because multipath=True means the connector knows how to deal with new multipath structure.17:08
tsekiyamaBut when mutlipath = False, we need to think about backward compatibility ..17:08
xyang1hemna: That is right. We don't support it17:08
hemnaxyang1, so the EMC patch is ok.17:08
*** alecv has quit IRC17:08
xyang1hemna: Ok. Good17:09
tsekiyamaso we can't omit target_poratl/iqn/lun17:09
jgriffithtsekiyama: yeah, but I'm wondering if we could just consolidate on a single supported data-structure with optional keys for the multi-path and fail over17:09
hemnatsekiyama, so the change we are thinking of here is to eliminate the target_alternative_portals17:09
*** Miouge has joined #openstack-cinder17:09
*** Saba has joined #openstack-cinder17:09
hemnaand simply reuse target_portals17:09
hemnaas a list17:09
hemnaand the driver says, multipath: True or False17:09
jgriffithtsekiyama: yeah... kinda what hemna said :)17:09
*** Saba is now known as Guest8785817:09
hemnaif the driver returns multipath: False,   then target_portals is the same logic as target_alternative_portals was meant to be.17:10
jgriffithtsekiyama: my main point was not proliferating data structures, but using them more intelligently17:10
Guest87858Hi , I am wondering if get_volume_stats is exposed as metric in cinder  for reading free capacity17:10
hemnajust a list to iterate over to try and get a connection to work at all17:10
tsekiyamahemna: I'm ok with reusing traget_portals17:10
jgriffithGuest87858: nope17:10
Guest87858Hi , I am wondering if get_volume_stats is exposed as metric in cinder  for reading free capacity17:10
jgriffithGuest87858: and again... "nope" :)17:10
jgriffithxyang1: I think it's ok if we need to fix drivers that implemented it already17:11
jgriffithxyang1: it's all the same 'stuff' just presented slightly differently17:11
tsekiyamahemna: jgriffith: just concidering we should have both 'target_portals' and 'target_portal' when multipath=False, for old nova17:11
xyang1jgriffith: That's fine.  We just need to know17:11
smcginnis+1 on consolidating structures. That seems simpler IMO.17:11
hemnatsekiyama, yah17:11
hemnatsekiyama, we'll land this in brick17:11
jgriffithtsekiyama: so that's kinda my original thought17:11
hemnanova won't get anything17:12
hemnafor L17:12
jgriffithbut not sure how anybody else feels17:12
smcginnistsekiyama: I think that part is good. You are correct, we definitely need backward support.17:12
hemnaerr K17:12
jgriffithtsekiyama: if you look at my comments in the review17:12
hemnaK nova will ignore target_portals17:12
jgriffithtsekiyama: that was before I realized that your other patch had already merged17:12
hemnaand L nova will get brick (with any luck)17:12
Guest87858Hi , I am wondering if get_volume_stats is exposed as metric in cinder  for reading free capacity17:12
hemnaGuest87858, nope17:12
jgriffithGuest87858: you're going to get booted in a second17:12
hemnaGuest87858, 3rd time we said nope.17:12
jgriffithYou've asked the same question 4 times and received a prompt response every time17:13
hemnalag ?17:13
jgriffithhemna: nah...  we wouldn't see it out of order on our side I don't think17:13
jgriffithhemna: but maybe17:13
jgriffithI dunno17:13
SwansonPerhaps they aren't seeing anything?17:14
*** Miouge has quit IRC17:14
jgriffithSwanson: maybe not17:14
Guest87858I thought it was related to other conversation . Thanks. Why are we not exposing this metric ? Is is a good idea to expose free capacity gb as metric for monitoring ?17:14
jgriffithhemna: xyang1 tsekiyama Just some ideas17:14
jgriffithGuest87858: no worries17:15
jgriffithGuest87858: Well.. I believe it is sent up to Ceilometer for example17:15
*** Miouge has joined #openstack-cinder17:15
jgriffithGuest87858: but we typically don't want to expose details like that to end user17:15
Guest87858But no metric of ceilometer exposes it too17:15
Guest87858If a system wants to use this for metering then is there any way ?17:16
jgriffithGuest87858: I believe so yes, that's what I was getting at17:16
jgriffithGuest87858: we emit that information in various places17:16
*** Miouge has quit IRC17:16
Guest87858Can you tell me which meters ?17:17
jgriffithGuest87858: nope :(17:17
jgriffithGuest87858: I don't mess with that much, but maybe others here do17:17
*** JayJ has joined #openstack-cinder17:17
*** _cjones_ has joined #openstack-cinder17:18
*** lpetrut has joined #openstack-cinder17:21
guitarzanjgriffith: is it sent to ceilometer?  I'm not sure17:21
*** ebalduf has joined #openstack-cinder17:22
*** bkopilov has quit IRC17:26
*** ebalduf has quit IRC17:26
*** leeantho has joined #openstack-cinder17:27
jgriffithguitarzan: yeah, I'm not certain either17:27
*** jistr has quit IRC17:29
jgriffithguitarzan: seems like it would be of key interest, but who konws17:29
guitarzanoh, I'm sure it's of interest, but I don't think it's currently sent17:30
jgriffithguitarzan: so what "is" sent... just usage I suppose?17:30
jgriffithand not even based on backend based on what I can see17:30
guitarzanI think the only thing sent is our explicit notifications17:30
jgriffithguitarzan: yeah, the "emitters"17:31
jgriffithguitarzan: suppose we could easily add a bunch o'stuff17:31
guitarzanso unless there's one for get_volume_stats or maybe a periodic one in the scheduler or something17:31
*** emaisne has joined #openstack-cinder17:35
*** e0ne has quit IRC17:38
*** emaisne has quit IRC17:39
*** emaisne_ has joined #openstack-cinder17:39
*** bkopilov has joined #openstack-cinder17:41
*** jordanP has quit IRC17:43
*** yuriy_n17 has quit IRC17:45
*** tbarron has quit IRC17:46
*** tbarron has joined #openstack-cinder17:47
*** ronis has joined #openstack-cinder17:48
*** Apoorva has joined #openstack-cinder17:49
*** coolsvap is now known as coolsvap_17:52
*** kallebe has quit IRC17:55
*** kallebe has joined #openstack-cinder17:59
*** annashen has joined #openstack-cinder18:00
*** Mandell has joined #openstack-cinder18:05
*** beecee has joined #openstack-cinder18:07
*** afazekas_ has quit IRC18:07
*** dustins_ has joined #openstack-cinder18:09
*** david-lyle_afk is now known as david-lyle18:09
*** _cjones_ has quit IRC18:10
*** dustins has quit IRC18:12
*** _cjones_ has joined #openstack-cinder18:15
*** deepakcs has quit IRC18:16
*** garthb_ has joined #openstack-cinder18:17
*** garthb has joined #openstack-cinder18:17
*** ebalduf has joined #openstack-cinder18:23
*** patrickeast has joined #openstack-cinder18:25
*** vilobhmm has joined #openstack-cinder18:27
*** rushiagr is now known as rushiagr_away18:27
*** ebalduf has quit IRC18:27
*** Guest87858 has quit IRC18:28
*** vilobhmm has quit IRC18:29
*** vilobhmm has joined #openstack-cinder18:31
*** dustins_ has quit IRC18:31
*** dustins has joined #openstack-cinder18:32
*** vilobhmm1 has joined #openstack-cinder18:32
*** vilobhmm_ has joined #openstack-cinder18:32
*** harlowja has joined #openstack-cinder18:33
*** diemt has joined #openstack-cinder18:34
*** dustins has quit IRC18:35
*** vilobhmm has quit IRC18:35
*** vilobhmm_ is now known as vilobhmm18:35
*** dustins has joined #openstack-cinder18:36
*** casusbelli has joined #openstack-cinder18:39
*** thingee has joined #openstack-cinder18:44
mkodererthingee: ping18:46
mkodererthingee: could you also have a look to
*** asselin_ has quit IRC18:47
thingeemkoderer: maybe today. I was absent friday because of a conference. Definitely tomorrow though. I got a lot of emails of "please look at this" :)18:47
casusbellijgriffith: Hi! Can i take a few mins regarding sos-ci, again? :)18:53
mkodererthingee: sure sure :) I just want to have the ABC stuff somehow delivered18:54
mkodererthingee: since jgriffith already ask for the next steps :)18:55
thingeexyang1: are we going to see an update this week with ?18:55
thingeejgriffith: happy to see this18:56
thingeejgriffith: thanks again for your work on v218:56
xyang1thingee: I hope so. He got one problem overriding the existing max over sub ratio18:56
xyang1thingee: I need to take a look, not sure why it does not work18:57
xyang1thingee: So he wants to override the default value18:57
thingeeI know, we talked about this last week :)18:58
xyang1thingee: Yes, i sent him examples. He tried but said it does not work18:58
kallebeJust a quick question regarding pep8 style: do you think we should enforce the pep8 recommendation of newline >after< the binary operator? I mean, should it be "foo +\nbar" or "foo\n+bar"?18:59
thingeekallebe: space is sufficient to me.19:00
kallebethingee I don't know if I understand what you mean. I am asking this because there are many cases of both coexisting19:01
guitarzankallebe: if you want to add a hacking check, see if people like it19:02
kallebewhen the code would pass 80 characters, it would start a newline and place the operator19:02
thingee1 + 119:03
thingeethere are spaces in between19:03
thingeethat's sufficient to me19:03
guitarzanthingee: he's talking about19:03
guitarzan+ 119:03
kallebethingee no, that's right. I agree, but my question is about \n19:03
*** Apoorva_ has joined #openstack-cinder19:03
guitarzani +19:03
* guitarzan didn't realize pep8 even cared19:03
*** hemna has quit IRC19:03
thingeeI understand that. kallebe asked for an opinion. My opinion is a space is sufficient. I don't see the appeal with a new line.19:03
guitarzanthingee: 80 character limit...19:04
thingeeguitarzan: got an example where we hit that case?19:04
thingeethat might help me see ugly :)19:04
guitarzanI'm trying to find it in pep8...19:04
kallebeI will fetch one here19:04
*** dustins has quit IRC19:04
thingeekallebe: thanks19:05
kallebehere is an example:19:06
*** Apoorva has quit IRC19:06
kallebeimagine it passes 80 characters. Should we put a + at the end or at the beginning?19:06
*** dustins has joined #openstack-cinder19:07
thingeexyang1: are these bugs still happening for kilo? ... I noticed we finally got some patches for bugs in 2/20 :)19:07
openstackLaunchpad bug 1385450 in Cinder "EMC volume create_cloned_volume failure leaves target volume" [Medium,Triaged] - Assigned to Xing Yang (xing-yang)19:07
openstackLaunchpad bug 1372635 in Cinder "MITM vulnerability with EMC VMAX driver" [High,In progress] - Assigned to Xing Yang (xing-yang)19:07
thingeekallebe: Oh I see, my mistake. umm front?19:08
kallebeAnd the pep8 section is this one:
thingeethat's my immediate thought anyways19:08
kallebesee the snippet for that class Rectangle there19:09
thingeekallebe: yeah I don't care one way or another then. :)19:09
guitarzan"break around a binary operator is after the "19:09
xyang1thingee: One has patch submitted, the other fixed already.  I'll update them19:09
thingeeif that's what people do, sure19:09
guitarzantook a lot for me to see it19:09
thingeexyang1: thanks!19:09
guitarzanpersonally I like it the other way, but *shrug*19:09
thingeexyang1: any luck with devstack to test cinder objects?19:09
xyang1thingee: Yes, I did some testing and it worked for me19:10
kallebethingee ok. I was wondering if a general pep8 cleanup would be accepted :)19:10
kallebeI would like to work on it19:10
guitarzankallebe: if it isn't in hacking, it won't be accepted19:10
kallebeI have seen hacking, but it was more about pep819:10
guitarzanright, but optional "I like this better" stuff doesn't usually make it19:11
kallebethe specific recommendations did not mention that case, for example19:11
guitarzanif it isn't enforced, it shouldn't be changed19:11
thingeeso much green "implemented" ... happy right now :)
xyang1thingee: So I don't see big problems with it.  Just want to go thru the code more to make sure19:11
kallebeguitarzan yes, so I was here asking how to know if it is enforced.19:11
thingeexyang1: ok, but did you try it out? schema changes while things are in progress?19:11
kallebeis there any place where I can see the enforced ones?19:11
kallebethe ones jenkins tests, for example19:12
guitarzankallebe: in here maybe?
xyang1thingee: I tried migrate down and also add a column in snapshot table19:12
guitarzankallebe: or better:
*** hemna has joined #openstack-cinder19:12
thingeeleeantho: client landing soon or did? so that we can get an update on this?
thingeehemna: ^19:13
leeanthothingee, yes, it is being released to pypi this week19:13
kallebeguitarzan ok, thanks. I will take a look. Just one more question, again related to pep8 haha About keyword arguments.19:13
thingeewinston-d: last week I asked for an update on this: it's block a couple of bps at the moment.19:13
kallebewhat about that whitespace between keyword arguments?19:14
patrickeastthingee: oh hey, sorry to change direction after asking you to approve it right at the deadline, but can you move out of k-3? We’ve decided to do it in L once the nova side of those iscsi changes can get in (otherwise right now we don’t really get any benefit from it)19:14
patrickeasti mistakenly thought the nova changes were going to make it in K19:14
thingeepatrickeast: making my day :)19:15
thingeepatrickeast: marked it for next19:15
thingeepatrickeast: thanks for the update19:15
patrickeastthingee: yay, thanks19:15
*** hemna has quit IRC19:16
thingeedon't forget everyone, summit sessions proposals have started
thingeeleeantho: thanks, please don't forget about
thingeeleeantho: march 1st for jenkins passing patches19:19
*** Mandell has quit IRC19:22
*** hemna has joined #openstack-cinder19:22
leeanthothingee, thanks, hemna and kmartin are finishing a few things for it.  it should be up before end of the week so there is time for corrections / reviews to the patch19:22
*** ebalduf has joined #openstack-cinder19:24
hemnathingee, we are working on finalizing the client release19:24
*** alexpilotti has quit IRC19:24
thingeeleeantho, hemna: excellent thanks! just wanted to communicate I'm watching the patch :)19:24
hemnaok coolio19:24
thingeeseriously though, thank you everyone for helping with reviews. 28/46 targeted things is still a bit to do, but progress that's nothing to sneeze at.19:27
*** Mandell has joined #openstack-cinder19:28
*** kvidvans has quit IRC19:28
*** ebalduf has quit IRC19:28
*** sgotliv has joined #openstack-cinder19:30
openstackgerritBob Callaway proposed openstack/cinder: NetApp eseries implementation for manage/unmanage
*** Tross has joined #openstack-cinder19:45
*** kallebe has left #openstack-cinder19:47
openstackgerritPatrick East proposed openstack/cinder-specs: Add DB table for driver private data
*** e0ne has joined #openstack-cinder19:51
*** openstackgerrit has quit IRC19:51
*** openstackgerrit has joined #openstack-cinder19:52
*** ronis has quit IRC19:53
patrickeastjgriffith: hemna: ^ I’ve updated that spec to where I’m pretty happy with it after testing it out a bit with some POC code and things seem to work correctly so hopefully this thing should be ready for some final reviews19:53
*** Tross has quit IRC19:54
*** Tross has joined #openstack-cinder19:54
*** Mandell has quit IRC19:54
*** Mandell has joined #openstack-cinder19:56
*** Mandell has quit IRC19:57
*** markvoelker has joined #openstack-cinder20:07
*** drjones has joined #openstack-cinder20:10
*** _cjones_ has quit IRC20:10
*** BharatK has quit IRC20:11
e0nejungleboyj: hi! are you around?20:12
jungleboyje0ne: Yep.20:13
e0nejungleboyj: few questions from me20:13
openstackgerritDanny Wilson proposed openstack/cinder: Over subscription for Pure Storage iSCSI driver.
e0nejungleboyj: did you se my comment for not sure now is the best way to do20:14
e0neit's about volume copy bandwidth20:14
*** marcusvrn has quit IRC20:15
jungleboyje0ne: Looking.20:16
e0nejungleboyj: thank you20:16
jungleboyje0ne: Oh, you think it should be renamed 'throttling' so you don't have to do 'import throttle as throttling'?20:17
e0nejungleboyj: yes, because it is used as 'import throttle as throttling' everywhere20:18
jungleboyjI guess I can see that argument.20:20
e0nei won't -1 it if you thinl it's not important20:21
jungleboyjI just updated and asked Tomoki's input.  See what they say.20:22
jungleboyje0ne: No problem.  Thanks for looking through the incubator patches.  I am getting close to getting the scheduler patch fixed up.  That that point I can do the rest of the cleanup in cinder/openstack/common and then get the move to oslo.log done.20:23
jungleboyje0ne: Will be excited to have all this done.20:24
*** ebalduf has joined #openstack-cinder20:24
e0nejungleboyj: great! it was the second question for you:)20:26
jungleboyj:-)  Yeah, I hope we can get this wrapped up this week.20:27
jungleboyjpatrickeast: I owe you one.  I got Edwin to dig into that issue with the flashsystem CI and he discovered he had a downlevel version of the CI code that wasn't properly pulling the patches down.20:28
patrickeastjungleboyj: ahh nice, that would explain it20:28
jungleboyjpatrickeast: Thanks for catching that.  Still waiting for the storwize CI contact to return from Chinese holiday.  I am thinking she has the same issue though.20:28
*** Mandell has joined #openstack-cinder20:28
patrickeastjungleboyj: glad it wasn’t anything to crazy20:29
jungleboyjpatrickeast: Yeah, something that asselin had fixed a while ago and we missed it.20:29
*** ebalduf has quit IRC20:29
jungleboyjpatrickeast: Yep, thanks for calling it out!20:29
patrickeastjungleboyj: np20:30
*** erlon has joined #openstack-cinder20:30
*** Apoorva_ has quit IRC20:33
*** Apoorva has joined #openstack-cinder20:33
*** harlowja has quit IRC20:34
*** harlowja_ has joined #openstack-cinder20:34
*** annashen has quit IRC20:35
*** mriedem has quit IRC20:35
*** rcallawa_ has joined #openstack-cinder20:36
jgriffithany core want to help a brother out and merge this:
jgriffiththink I've iterated on spelling issues enough :)20:40
jungleboyjWheee. Unit tests now pass with the new scheduler code.20:41
*** markvoelker has quit IRC20:51
*** lpetrut has quit IRC20:59
*** mdenny has quit IRC21:01
*** sandywalsh_ has joined #openstack-cinder21:02
*** sandywalsh has quit IRC21:05
openstackgerritTom Barron proposed openstack/cinder: NFS backup driver
*** dustins has quit IRC21:14
*** markvoelker has joined #openstack-cinder21:15
*** markvoelker has quit IRC21:15
*** markvoelker has joined #openstack-cinder21:16
e0nejgriffith: it was too quick for me to vote for your change:(21:16
*** dustins has joined #openstack-cinder21:16
*** jwang_ has joined #openstack-cinder21:18
*** jwang has quit IRC21:20
*** jwang_ has quit IRC21:20
*** jwang has joined #openstack-cinder21:21
*** ebalduf has joined #openstack-cinder21:25
*** vilobhmm2 has joined #openstack-cinder21:25
*** vilobhmm has quit IRC21:25
*** vilobhmm1 has quit IRC21:25
anishe0ne: your comment is in direct contradiction to what jgriffith xyang1 said on
anishtechnically all 3 of you have different suggestions.21:26
anishjgriffith wants no six., xyang1 wants it only for debug, e0ne wants it for all21:27
anishwelp ?21:27
*** rushiagr_away is now known as rushiagr21:27
e0neanish: hm...21:28
*** jungleboyj has quit IRC21:29
*** ebalduf has quit IRC21:29
jgriffithanish: :)21:30
jgriffithanish: I'm easy, I'll got with what others say21:30
jgriffithanish: admittedly it's a messy thing21:30
jgriffithanish: and I was just reading the doc, which may be wrong... or perhaps I interpretted it wrong21:30
jgriffithanish: I'll be away for a bit, but send me a note and I'll do whatever you all think is right in this case21:31
xyang1anish: If you follow the example in the guideline, which is an info, then you are fine:)21:31
*** annashen has joined #openstack-cinder21:32
xyang1jgriffith: You discovered that bug? If you try to print ex without _ or six, it throws unicode error21:33
e0neanish, jgriffith, xyang1: sometimes it's not easy for me to make right decision. need to know what error message will be21:34
xyang1anish: If you use _, then no need for six21:34
thingeejgriffith: sorry, would've helped. looks ninja'd :)21:35
thingeeI approved it earlier. I blame jenkins21:35
anishe0ne: xyang1 jgriffith sounds good, thanks !21:35
openstackgerritTomoki Sekiyama proposed openstack/cinder: Failover to alternative iSCSI portals on login failure
openstackgerritTomoki Sekiyama proposed openstack/os-brick: Use target_portals/iqns/luns for alternative target information
anishe0ne: can I convince you for something better than 0 maybe :) ?21:37
e0nexyang1: do you mean something like _('err: %s' % exc)?21:37
xyang1e0ne: Yes, _LI,etc, according to guideline21:37
xyang1e0ne: It uses six underneath21:38
smcginnise0ne: I don't think around the formatting??21:38
anishhemna: os-brick doesn't get it's own -specs ?21:38
hemnaanish, not yet21:38
anishhemna: I have a WIP spec for brick made in cinder-specs. I should just let it be for now then ?21:39
*** annashen has quit IRC21:39
*** annashen has joined #openstack-cinder21:40
e0nexyang1: i read guidelines once more, thanks21:41
hemnaso, I don't think we have specs for cinderclient either21:42
xyang1e0ne: It doesn't cover the debug msg case though. That's the problem21:42
hemnaso since os-brick is a subproject of cinder, I think cinder-specs is fine for now21:42
hemnathingee, may want to revisit that later.21:42
anishko !21:42
*** akerr has quit IRC21:42
*** dustins has quit IRC21:43
e0nexyang1: agree, it doesn't cover all cases21:45
*** karimb has joined #openstack-cinder21:45
*** btran has joined #openstack-cinder21:45
*** dustins has joined #openstack-cinder21:46
thingeehemna: ok21:46
* thingee heads out to lunch21:46
*** jwang has quit IRC21:48
*** karimb has quit IRC21:48
*** dustins has quit IRC21:48
anishe0ne: xyang1 documentation needs to be updated to remove ambiguities wrt six+debug ?21:48
*** bswartz has quit IRC21:53
*** jwang has joined #openstack-cinder21:55
*** annashen has quit IRC21:56
smcginnisanish, e0ne, xyang1: Tested exception logging:
smcginnisNot what I expected.21:59
smcginnisUnless I'm doing something different here.21:59
*** annashen has joined #openstack-cinder21:59
xyang1smcginnis: I think it has to do with the exception itself22:00
xyang1smcginnis: I tested it a couple of months ago22:00
smcginnisxyang1: Depending on the type of exception?22:00
*** Yogi1 has quit IRC22:00
xyang1smcginnis: Yes22:00
smcginnisxyang1: Or if the exception contains unicode?22:00
xyang1smcginnis: Correct22:01
xyang1smcginnis: Can you check the patch i referenced there22:01
smcginnisxyang1: Sorry, can you post again?22:01
*** emaisne_ has quit IRC22:02
*** Apoorva_ has joined #openstack-cinder22:03
anishsmcginnis: my understanding was the correct way is this
smcginnisanish: That looks like what I would expect.22:03
smcginnisThough not sure why it would be different between debug and info.22:04
smcginnisI would think either the logger handles unicode or it doesn't.22:04
*** Apoorva has quit IRC22:06
anisherm, the documentation seems way off. Uses % in logging. From a review that I cannot remember, someone mentioned that % + dict can have issues in logging as well22:06
smcginnisLine 325 would support anish's understanding, at least for info level:
anishhopefully same for all non-debug levels22:08
smcginnisanish: It does appear so:
smcginnisThe i18m guidelines do state: "Whenever possible translation should not be forced by use of str(), unicode(), or six.text_type() on a message being used with a format string."22:09
anishsmcginnis++ thank you for the research22:10
smcginnisanish: No problem. I wanted to make sure I understood this. Been hearing conflicting information.22:10
*** thangp has quit IRC22:11
smcginnisxyang1, e0ne, jgriffith: ^^ Making sure you saw
*** primechuck has quit IRC22:15
xyang1smcginnis: Unless if they wrap everything by default now.  It threw unicode error for me last time22:16
*** JayJ has quit IRC22:16
*** nellysmitt has quit IRC22:16
*** nellysmitt has joined #openstack-cinder22:16
xyang1smcginnis: I tried one of the command line in brick22:16
smcginnisxyang1: Hmm... I hate inconsistent results.22:17
*** nellysmitt has quit IRC22:17
*** JayJ has joined #openstack-cinder22:17
smcginnisxyang1: Are you sure it was passed as an arg to the log command and not formatted as a string first?22:17
xyang1smcginnis: That's why i prefer it is wrapped22:17
smcginnisi.e.'Test: %s', e) vs'TEst: %s' % e)22:17
*** vilobhmm1 has joined #openstack-cinder22:18
xyang1smcginnis: Yes22:18
*** vilobhmm2 has quit IRC22:18
xyang1smcginnis: Tried some command line in brick and print that22:18
xyang1smcginnis: Make sure it fails22:19
smcginnisxyang1: Can you give me a pointer to where you tested that? I just want to make sure I'm looking at the same thing.22:20
*** Mandell has quit IRC22:20
xyang1smcginnis: I don't remember exactly which command I tried.  It is in that patch I sent earlier.  Maybe revert it back?  Because jgriffith ran into the problem first and fixed it22:21
*** JayJ has quit IRC22:23
*** JayJ has joined #openstack-cinder22:24
smcginnisxyang1: OK, in that patch exceptions are just being passed in to LOG.error, but six.text_type(e) is used for debug.22:25
smcginnisBut that actually didn't change with his changes.22:25
smcginnisThose appear to only have been changed to go from % to ,22:26
smcginnisI'll have to set up and see.22:26
*** ebalduf has joined #openstack-cinder22:26
*** aix has joined #openstack-cinder22:26
xyang1smcginnis: Some are removed.  Look at code on the left22:26
*** timcl has quit IRC22:27
smcginnisxyang1: I just saw one changed from e to e.stdout.22:27
smcginnisor rather e.stderr.22:28
*** JayJ has quit IRC22:28
xyang1smcginnis: Some debug on the left are removed22:28
*** rushiagr is now known as rushiagr_away22:29
*** harlowja_ has quit IRC22:29
*** JayJ has joined #openstack-cinder22:29
smcginnisxyang1: The logging of "out" and "err" from the _run() command?22:29
*** harlowja has joined #openstack-cinder22:30
*** ebalduf has quit IRC22:30
smcginnisThat doesn't look directly related. At least from the commit message.22:30
smcginnisCommit message actually gives "LOG.error("%s" % e)" as the offending format.22:30
smcginnisNothing about using six.text_type().22:31
smcginnisSo I think it probably comes down to doing LOG.debug('Test: %s', e) vs LOG.debug('Test: %s' % e)22:31
smcginnisWith the first being the correct way.22:31
smcginnisIt appears (and I would expect) the logger to handle it correctly.22:32
smcginnisAlthough this is all speculation. I'll spin it up and see if I can verify.22:32
*** JayJ has quit IRC22:33
*** JayJ has joined #openstack-cinder22:34
smcginnisAh, and of course it's gone in my current stacked view.22:34
*** JayJ has quit IRC22:34
* smcginnis shakes fist at hemna22:34
*** jwang_ has joined #openstack-cinder22:34
hemnawhat did I do ?22:35
*** rcallawa_ has quit IRC22:36
smcginnishemna: Ah, nothing. Testing something that was in brick. :)22:36
*** jwang has quit IRC22:38
*** Apoorva_ has quit IRC22:39
*** Apoorva has joined #openstack-cinder22:39
*** Mandell has joined #openstack-cinder22:40
*** mtanino has quit IRC22:40
*** Mandell has quit IRC22:47
*** asmith_brcd has joined #openstack-cinder22:52
*** xyang has joined #openstack-cinder22:53
xyangsmcginnis: I got some results22:54
smcginnisxyang: Oh!22:54
smcginnisxyang: What did you see?22:54
smcginnisxyang: I'm still trying to sort my devstack setup. :\22:55
xyangsmcginnis: I think you are right. If we use %, we'll get unicode error right away from the oslo i18n library.  If we use ",", it will be handled correctly22:55
smcginnisxyang: Oh good, that's what I would expect. And want.22:55
smcginnisxyang: We still need to watch for uses of msg = 'Unicode %s' % blah,, raise exception(msg).22:56
smcginnisxyang: But that's better than every logging call.22:56
smcginnisxyang: Thanks for checking!22:57
xyangsmcginnis: no problem. I want to get it right as well22:57
*** kaufer has quit IRC22:59
*** e0ne has quit IRC23:00
*** Apoorva_ has joined #openstack-cinder23:02
*** harlowja has quit IRC23:04
*** rwsu is now known as rwsu-afk23:04
*** Mandell has joined #openstack-cinder23:05
*** Apoorva has quit IRC23:06
*** devlaps has quit IRC23:14
hemnasmcginnis, broken ?23:14
*** devlaps has joined #openstack-cinder23:14
smcginnishemna: Nothings broken, don't worry.23:14
smcginnishemna: Just having some fun. ;)23:14
hemnaheh ok23:15
hemnaI'm fixing a fczm issue at the moment23:15
smcginnishemna: Sorry to distract you.23:15
hemnacan't load the cisco driver23:15
hemnathe config is borked23:15
smcginnishemna: Ooh, not good.23:15
hemnalooks like this wasn't tested by the cisco folks23:15
smcginnishemna: Something in the fczm code itself, or specific to cisco?23:15
hemnafczm code23:15
*** harlowja has joined #openstack-cinder23:16
hemnait tries to load the config for the fc-zone-manager group23:16
hemnaand it never sees it23:16
asmith_brcdhemna, what's the fczm issue?23:16
hemnabecause it's only looking at the volume manager group23:16
smcginnishemna: None of that ends up getting covered by CI unless the block vendors happen to use it, is it23:16
asmith_brcdcan i help?23:16
hemnasmcginnis, yah23:16
hemnaasmith_brcd, I'm working on a fix at the moment.23:16
openstackLaunchpad bug 1424852 in Cinder "Cinder fc_zone_manger with Cisco configured" [High,Confirmed]23:16
hemnafor that23:16
*** ganso_ has quit IRC23:17
*** mfisch has joined #openstack-cinder23:19
asmith_brcdhemna, glad you are fixing it, thank you.  FYI, any issues in fczm, i'm available23:19
*** tbarron has quit IRC23:20
*** tbarron has joined #openstack-cinder23:20
hemnaasmith_brcd, sure23:21
*** tbarron has quit IRC23:21
*** Mandell_ has joined #openstack-cinder23:26
*** ebalduf has joined #openstack-cinder23:27
*** Mandell has quit IRC23:29
*** ebalduf has quit IRC23:31
openstackgerritAnthony Lee proposed openstack/cinder: HP 3par driver filter and evaluator function
*** krtaylor has quit IRC23:35
*** mfisch has quit IRC23:36
*** alexpilotti has joined #openstack-cinder23:36
*** jcru has quit IRC23:40
*** krtaylor has joined #openstack-cinder23:47
*** MasterPiece has joined #openstack-cinder23:53
*** mfisch has joined #openstack-cinder23:54
*** mfisch has quit IRC23:54
*** mfisch has joined #openstack-cinder23:54
*** marcusvrn has joined #openstack-cinder23:57

Generated by 2.14.0 by Marius Gedminas - find it at!