Tuesday, 2015-03-24

*** dims has quit IRC00:00
*** markvoelker has joined #openstack-cinder00:02
*** jcru has joined #openstack-cinder00:04
*** jcru has quit IRC00:04
*** _cjones_ has quit IRC00:07
*** thingee has joined #openstack-cinder00:08
*** annashen has quit IRC00:13
*** annegentle has quit IRC00:14
*** emagana has quit IRC00:14
*** _cjones_ has joined #openstack-cinder00:17
*** ho has joined #openstack-cinder00:19
*** melwitt has quit IRC00:19
*** e0ne has quit IRC00:23
*** thingee has quit IRC00:31
*** _cjones_ has quit IRC00:31
*** tsekiyama has joined #openstack-cinder00:32
*** asselin_ has quit IRC00:36
*** Mandell has quit IRC00:37
openstackgerritVilobh Meshram proposed openstack/cinder: WIP - ConsistencyGroup Object  https://review.openstack.org/16364100:43
*** appsdesh has quit IRC00:43
*** dflorea has quit IRC00:44
*** dflorea has joined #openstack-cinder00:45
*** lpetrut has joined #openstack-cinder00:46
*** dflorea has quit IRC00:49
*** lpetrut has quit IRC00:50
*** coolsvap|afk is now known as coolsvap00:50
*** Longgeek has joined #openstack-cinder00:51
*** asselin_ has joined #openstack-cinder00:53
*** tsekiyam_ has joined #openstack-cinder00:56
*** tsekiyama has quit IRC00:57
*** leeantho has quit IRC00:59
*** vilobhmm1 has quit IRC01:01
*** topshare has joined #openstack-cinder01:02
openstackgerritDavanum Srinivas (dims) proposed openstack/cinder: Use oslo_context's overwrite capability for better logs  https://review.openstack.org/16706401:02
*** akerr has joined #openstack-cinder01:03
*** jwang_ has joined #openstack-cinder01:06
*** akerr has quit IRC01:07
*** jwang__ has quit IRC01:09
*** mtanino has joined #openstack-cinder01:09
*** emagana has joined #openstack-cinder01:11
*** dannywilson has quit IRC01:12
*** annashen has joined #openstack-cinder01:13
*** annegentle has joined #openstack-cinder01:14
*** marcusvrn has joined #openstack-cinder01:15
*** liusheng has quit IRC01:16
*** emagana has quit IRC01:17
*** emagana has joined #openstack-cinder01:18
*** annashen has quit IRC01:19
*** akerr has joined #openstack-cinder01:20
*** annegentle has quit IRC01:20
*** emagana has quit IRC01:23
*** topshare_ has joined #openstack-cinder01:24
*** junhongl has joined #openstack-cinder01:24
*** Lee1092 has joined #openstack-cinder01:26
*** topshare has quit IRC01:26
*** junhongl has quit IRC01:30
*** tsekiyam_ has quit IRC01:30
*** mtanino has quit IRC01:32
*** junhongl has joined #openstack-cinder01:32
*** gary-smith__ has quit IRC01:37
*** junhongl has quit IRC01:38
*** marcusvrn has quit IRC01:40
*** marcusvrn has joined #openstack-cinder01:40
*** dims has joined #openstack-cinder01:44
*** dims_ has quit IRC01:46
*** akerr has quit IRC01:47
*** marcusvrn has quit IRC01:50
*** jcru has joined #openstack-cinder01:50
*** jcru has quit IRC01:51
*** mriedem1 has quit IRC01:51
*** marcusvrn1 has joined #openstack-cinder01:51
*** vilobhmm1 has joined #openstack-cinder01:53
*** marcusvrn1 has quit IRC01:56
*** jcru has joined #openstack-cinder01:56
*** hflai_ has quit IRC01:59
*** takedakn has joined #openstack-cinder01:59
*** marcusvrn has joined #openstack-cinder01:59
*** r-daneel has quit IRC02:02
*** _cjones_ has joined #openstack-cinder02:04
*** _cjones_ has joined #openstack-cinder02:05
*** marcusvrn1 has joined #openstack-cinder02:06
*** marcusvrn has quit IRC02:08
*** takedakn has quit IRC02:11
*** marcusvrn1 has quit IRC02:11
*** annegentle has joined #openstack-cinder02:16
*** junhongl has joined #openstack-cinder02:21
*** annegentle has quit IRC02:21
*** scottda has joined #openstack-cinder02:23
*** topshare_ has quit IRC02:25
*** kaisers1 has joined #openstack-cinder02:26
*** topshare has joined #openstack-cinder02:27
*** kaisers has quit IRC02:28
*** haomaiwang has joined #openstack-cinder02:28
*** IanGovett has quit IRC02:29
*** junhongl has quit IRC02:29
*** Maike has quit IRC02:33
*** Maike has joined #openstack-cinder02:33
*** topshare has quit IRC02:38
*** topshare has joined #openstack-cinder02:41
*** topshare has quit IRC02:43
*** hemna has joined #openstack-cinder02:43
*** scottda has quit IRC02:46
*** dims has quit IRC02:51
*** patrickeast has quit IRC02:56
*** junhongl has joined #openstack-cinder03:07
anishhemna: are you still -1 on this ? https://review.openstack.org/#/c/158926/03:08
hemnalet me check03:08
hemnaI dunno man03:09
hemnamaybe I'm too hard on the milestones and such03:09
hemnait's not a bug03:10
hemnait's just a nice to have03:10
anishyes. hence code_cleanup_batching03:10
hemnaI'd rather not see a lot of churn now unless it's fixing a problem03:10
anishunless code_cleanup is not meant for k303:10
hemnaI think we are in bugs only at this point03:10
openstackgerritJohn Griffith proposed openstack/cinder: Enable request-id in cinder API logs  https://review.openstack.org/16708003:10
hemnaif you can get thingee to +2 it, then I'll +A it03:11
hemnabut at this point I have to defer to thingee03:11
anishaite. this should fun03:11
hemnait looks minor03:11
*** junhongl has quit IRC03:11
*** jcru has quit IRC03:13
*** marcusvrn has joined #openstack-cinder03:13
asselin_jgriffith, sorry, just saw your ping now (home computer). Officially infra wants to run all all rechecks, so they want to run on all rechecks.03:15
asselin_jgriffith, so recheck should recheck everything, and that's ok. recheck <vendor> is only really used for those with limited resources03:16
asselin_but infra runs on that query as well (and any other ci with enough resources)03:17
asselin_jgriffith, it's a topic of big debate as you can imagine. I'll have to find some links with discussion on this topic.03:17
openstackgerritJohn Griffith proposed openstack/python-cinderclient: Move unit tests into test directory  https://review.openstack.org/16223003:20
openstackgerritJohn Griffith proposed openstack/python-cinderclient: Add ability to specify path var to testr  https://review.openstack.org/16226203:20
*** Longgeek_ has joined #openstack-cinder03:22
*** jcru has joined #openstack-cinder03:24
*** Longgeek has quit IRC03:25
*** _cjones_ has quit IRC03:26
*** bkopilov has quit IRC03:29
*** _cjones_ has joined #openstack-cinder03:30
*** hemna has quit IRC03:37
*** topshare has joined #openstack-cinder03:39
*** jcru has quit IRC03:42
*** topshare has quit IRC03:43
*** jcru has joined #openstack-cinder03:43
*** _cjones_ has quit IRC03:44
*** _cjones_ has joined #openstack-cinder03:45
*** _cjones_ has quit IRC03:45
*** topshare has joined #openstack-cinder03:48
*** jcru has quit IRC03:48
openstackgerritAnkit Agrawal proposed openstack/cinder: Fix missing translations for log messages  https://review.openstack.org/16469703:50
*** topshare has quit IRC03:50
*** Mandell has joined #openstack-cinder03:55
*** _cjones_ has joined #openstack-cinder03:58
*** dims has joined #openstack-cinder04:12
*** dannywilson has joined #openstack-cinder04:12
*** dannywilson has quit IRC04:17
*** annegentle has joined #openstack-cinder04:18
*** dims has quit IRC04:19
*** annegentle has quit IRC04:23
*** primechuck has joined #openstack-cinder04:26
*** primechuck has quit IRC04:31
*** primechuck has joined #openstack-cinder04:31
*** xyang1 has quit IRC04:35
*** junhongl has joined #openstack-cinder04:39
*** xyang1 has joined #openstack-cinder04:41
*** xyang1 has quit IRC04:43
*** coolsvap is now known as coolsvap|afk04:44
*** junhongl has quit IRC04:44
*** avishay has joined #openstack-cinder04:47
*** ishant has joined #openstack-cinder04:48
*** bkopilov has joined #openstack-cinder04:57
*** hflai_ has joined #openstack-cinder05:02
*** hflai_ has quit IRC05:02
openstackgerritVincent Hou proposed openstack/cinder: Delete the temporary volume if migration fails  https://review.openstack.org/16068205:12
*** junhongl has joined #openstack-cinder05:18
*** annegentle has joined #openstack-cinder05:19
*** shilpa_ has joined #openstack-cinder05:19
*** harlowja is now known as harlowja_away05:22
*** anshul has joined #openstack-cinder05:23
*** annegentle has quit IRC05:23
*** junhongl has quit IRC05:23
*** reed has quit IRC05:26
*** boris-42 has quit IRC05:28
*** BharatK has joined #openstack-cinder05:32
*** junhongl has joined #openstack-cinder05:36
*** junhongl has quit IRC05:41
*** primechuck has quit IRC05:43
*** emagana has joined #openstack-cinder05:45
*** jcru has joined #openstack-cinder05:45
*** jcru has quit IRC05:50
*** nshaikh has joined #openstack-cinder05:58
*** Maike has quit IRC06:03
*** dims has joined #openstack-cinder06:04
openstackgerritOpenStack Proposal Bot proposed openstack/cinder: Imported Translations from Transifex  https://review.openstack.org/15240106:07
*** dims has quit IRC06:10
*** annashen has joined #openstack-cinder06:18
*** _cjones_ has quit IRC06:20
*** annashen has quit IRC06:24
*** ankit_ag has joined #openstack-cinder06:27
*** junhongl has joined #openstack-cinder06:33
*** vilobhmm1 has quit IRC06:34
*** nkrinner has joined #openstack-cinder06:35
*** deepakcs has joined #openstack-cinder06:36
*** pcaruana has quit IRC06:41
*** chlong has quit IRC06:42
*** anuragpalsule has joined #openstack-cinder06:46
*** _cjones_ has joined #openstack-cinder06:47
*** dulek_ has joined #openstack-cinder06:55
*** dulek_ is now known as dulek06:55
*** Bsony has joined #openstack-cinder06:57
*** dulek_ has joined #openstack-cinder07:01
*** chlong has joined #openstack-cinder07:03
*** dulek has quit IRC07:04
*** _cjones_ has quit IRC07:13
*** _cjones_ has joined #openstack-cinder07:18
*** annegentle has joined #openstack-cinder07:20
*** emagana has quit IRC07:22
*** emagana has joined #openstack-cinder07:22
*** annegentle has quit IRC07:25
*** emagana has quit IRC07:27
flip214asselin: ping?07:27
*** navid__ has joined #openstack-cinder07:38
*** dulek_ has quit IRC07:43
*** dulek has joined #openstack-cinder07:43
*** ronis has joined #openstack-cinder07:48
*** alecv has joined #openstack-cinder07:49
*** dims has joined #openstack-cinder07:55
*** markus_z has joined #openstack-cinder07:56
*** TobiasE has joined #openstack-cinder07:57
*** Longgeek_ has quit IRC07:58
*** nshaikh has quit IRC07:59
*** navid__ has quit IRC07:59
*** nshaikh has joined #openstack-cinder08:00
*** dims has quit IRC08:02
*** jordanP has joined #openstack-cinder08:07
*** liusheng has joined #openstack-cinder08:10
*** chlong has quit IRC08:12
*** _cjones_ has quit IRC08:15
*** annegentle has joined #openstack-cinder08:21
*** annegentle has quit IRC08:26
*** ndipanov has joined #openstack-cinder08:28
*** tellesnobrega has quit IRC08:28
*** nlevinki has joined #openstack-cinder08:35
openstackgerritVincent Hou proposed openstack/cinder: Set volume_attachment to [] for the temporary volume creation  https://review.openstack.org/16713408:39
*** tellesnobrega has joined #openstack-cinder08:40
*** Bsony has quit IRC08:41
openstackgerritVincent Hou proposed openstack/cinder: Set volume_attachment to [] for the temporary volume creation  https://review.openstack.org/16713408:44
*** liusheng has quit IRC08:52
*** avishay_ has joined #openstack-cinder09:01
*** sgotliv has joined #openstack-cinder09:01
*** avishay has quit IRC09:02
*** jistr has joined #openstack-cinder09:04
*** dims has joined #openstack-cinder09:15
*** _cjones_ has joined #openstack-cinder09:16
*** dims has quit IRC09:20
*** _cjones_ has quit IRC09:20
openstackgerritVincent Hou proposed openstack/cinder: Set volume_attachment to [] for the temporary volume creation  https://review.openstack.org/16713409:21
openstackgerritAnkit Agrawal proposed openstack/cinder: Fix missing translations for log messages  https://review.openstack.org/16469709:21
*** e0ne has joined #openstack-cinder09:31
*** nshaikh has quit IRC09:31
*** yuriy_n17 has joined #openstack-cinder09:32
*** jcru has joined #openstack-cinder09:37
*** Bsony has joined #openstack-cinder09:38
*** jcru has quit IRC09:38
*** e0ne has quit IRC09:40
*** lpetrut has joined #openstack-cinder09:40
*** TobiasE has quit IRC09:59
*** nshaikh has joined #openstack-cinder09:59
*** nshaikh has quit IRC09:59
openstackgerritMartin Kletzander proposed openstack/cinder: Fix common misspellings  https://review.openstack.org/16682310:01
*** marcusvrn has quit IRC10:01
*** nshaikh has joined #openstack-cinder10:10
*** _cjones_ has joined #openstack-cinder10:17
*** btran has quit IRC10:19
*** dims has joined #openstack-cinder10:19
*** haomaiwang has quit IRC10:21
*** _cjones_ has quit IRC10:22
*** markvoelker has quit IRC10:26
*** e0ne has joined #openstack-cinder10:29
*** Bsony_ has joined #openstack-cinder10:34
*** Bsony has quit IRC10:35
*** Bsony has joined #openstack-cinder10:46
*** Bsony has quit IRC10:47
*** Bsony_ has quit IRC10:48
*** Bsony has joined #openstack-cinder10:52
*** ho has quit IRC10:57
*** Bsony_ has joined #openstack-cinder10:57
*** TobiasE has joined #openstack-cinder10:58
*** Bsony has quit IRC10:59
*** timcl has joined #openstack-cinder11:00
*** Bsony_ has quit IRC11:02
*** Bsony has joined #openstack-cinder11:05
*** e0ne is now known as e0ne_11:09
*** Bsony_ has joined #openstack-cinder11:13
*** Bsony has quit IRC11:16
*** Bsony_ has quit IRC11:18
*** _cjones_ has joined #openstack-cinder11:18
*** e0ne_ has quit IRC11:19
*** e0ne has joined #openstack-cinder11:20
*** _cjones_ has quit IRC11:23
*** annashen has joined #openstack-cinder11:23
*** IanGovett has joined #openstack-cinder11:24
*** markvoelker has joined #openstack-cinder11:26
*** delattec has quit IRC11:28
*** cdelatte has quit IRC11:28
*** annashen has quit IRC11:28
*** Bsony has joined #openstack-cinder11:31
*** markvoelker has quit IRC11:31
*** Bsony has quit IRC11:36
*** e0ne is now known as e0ne_11:36
*** jamielennox is now known as jamielennox|away11:45
*** e0ne_ has quit IRC11:46
*** ociuhandu has joined #openstack-cinder11:47
openstackgerritYasuaki Nagata proposed openstack/cinder: Do 'detach' if a timeout happened in the process of attach_volume  https://review.openstack.org/16718511:48
*** e0ne has joined #openstack-cinder11:53
*** dulek_ has joined #openstack-cinder11:55
*** ronis has quit IRC11:55
*** ronis has joined #openstack-cinder11:55
*** dalgaaf has quit IRC11:57
*** dulek has quit IRC11:58
*** markvoelker has joined #openstack-cinder11:58
*** cdelatte has joined #openstack-cinder12:03
*** dulek_ has quit IRC12:04
*** jaypipes has quit IRC12:06
dimsjgriffith: ping looks like i ended up dup'ing what you had https://review.openstack.org/#/c/167064/ https://review.openstack.org/#/c/167080/ - so with either of those changes, is logging any better? (cc jogo)12:06
*** jaypipes has joined #openstack-cinder12:07
*** boris-42 has joined #openstack-cinder12:07
*** Bsony has joined #openstack-cinder12:09
*** chlong has joined #openstack-cinder12:12
*** akerr has joined #openstack-cinder12:14
*** BharatK has quit IRC12:14
*** chlong has quit IRC12:15
*** dims has quit IRC12:15
*** dims has joined #openstack-cinder12:16
*** _cjones_ has joined #openstack-cinder12:19
*** Bsony has quit IRC12:20
*** _cjones_ has quit IRC12:23
*** shilpa_ has quit IRC12:25
*** deepakcs has quit IRC12:26
*** dulek has joined #openstack-cinder12:30
*** jistr is now known as jistr|class12:30
*** chlong has joined #openstack-cinder12:32
*** chlong has quit IRC12:35
*** ishant has quit IRC12:36
*** chlong has joined #openstack-cinder12:36
*** rushiagr_away is now known as rushiagr12:37
*** chlong has quit IRC12:39
*** Miouge has joined #openstack-cinder12:44
*** chlong has joined #openstack-cinder12:56
*** rushiagr is now known as rushiagr_away13:00
*** scottda has joined #openstack-cinder13:01
*** julim has joined #openstack-cinder13:01
*** scottda has quit IRC13:02
*** scottda has joined #openstack-cinder13:03
*** Miouge has quit IRC13:04
*** Miouge has joined #openstack-cinder13:05
*** Miouge has quit IRC13:05
*** changbl has quit IRC13:06
jgriffithdims: haven't looked at yours yet, but the change I pused last nigt fixed it up13:12
dimsjgriffith: i had the same thing as yours :)13:12
jgriffithdims: haha, I see13:12
jgriffithdims: and it looks like Angus did the same ting in Heat13:13
jgriffithdims: We must be on to something :)13:13
*** diegows has joined #openstack-cinder13:13
dimswe should write this down :)13:13
jgriffithdims: probably13:13
jgriffithdims: it's things like this that I find so annoying13:13
jgriffithe0ne: I'm confused by your comment... I did remove cinder/openstack/common/local ?13:14
jgriffithe0ne: OH you mean the conf entry13:15
e0nejgriffith: hi!13:15
e0nejgriffith: yes, it should be removed from openstack-common.conf too13:15
e0nei'm almost ready with patch13:15
*** dustins has joined #openstack-cinder13:16
*** eharney has quit IRC13:16
e0nejgriffith: i can update yours patch in 3minutes13:16
jgriffithe0ne: why not, everyone else does13:17
*** Yogi1 has joined #openstack-cinder13:18
openstackgerritIvan Kolodyazhny proposed openstack/cinder: Enable request-id in cinder API logs  https://review.openstack.org/16708013:20
*** _cjones_ has joined #openstack-cinder13:20
avishay_jgriffith: what's your address?  i want to send you a box of these: http://www.photo4design.com/files/165/82494-tilted-keyboard-key-letter-h.jpg13:23
*** avishay_ is now known as avishay13:23
*** _cjones_ has quit IRC13:25
dimsjgriffith: does this sound about right? https://etherpad.openstack.org/p/help-cant-find-request-id-in-my-logs13:26
*** lpetrut has quit IRC13:26
dimshaha "everyone else does"13:26
* dims rushes to find one of john's patches13:26
e0nedims: need to add notes about openstack/common/local.py13:27
dimse0ne: go for it? :)13:27
*** timcl has quit IRC13:27
*** lpetrut has joined #openstack-cinder13:27
e0nedims: done13:28
e0nedims: may be it's a good idea to move #10 to the first place? :)13:29
dimse0ne: haha13:29
jgriffithdims: nice!13:32
jgriffithavishay: LOL!!13:32
jgriffithavishay: ya know I still haven't replaced this sucker!13:32
*** timcl has joined #openstack-cinder13:33
*** rushiagr_away is now known as rushiagr13:33
*** Bsony has joined #openstack-cinder13:35
ankit_agHi all, can someone please review https://review.openstack.org/#/c/165057/13:38
*** primechuck has joined #openstack-cinder13:39
openstackgerritwuyuting proposed openstack/cinder: Make backup-list support all-tenants options  https://review.openstack.org/16672713:41
*** mriedem has joined #openstack-cinder13:44
*** r-daneel has joined #openstack-cinder13:44
*** mtanino has joined #openstack-cinder13:46
*** jcru has joined #openstack-cinder13:47
*** anuragpalsule has left #openstack-cinder13:48
*** nshaikh has quit IRC13:48
*** jcru has quit IRC13:49
*** jcru has joined #openstack-cinder13:50
*** jistr|class is now known as jistr13:53
openstackgerritTushar Patil proposed openstack/python-cinderclient: Allow cinderclient to handle exception response  https://review.openstack.org/16635113:53
openstackgerritPetrut Lucian proposed openstack/cinder: Windows iSCSI: fix volume clone  https://review.openstack.org/16722813:54
*** thangp has joined #openstack-cinder13:59
*** obutenko has quit IRC13:59
*** Adriano_ has joined #openstack-cinder14:00
*** obutenko has joined #openstack-cinder14:02
openstackgerritwuyuting proposed openstack/cinder: Add filtering snapshots by tenant ID  https://review.openstack.org/16668914:03
*** mriedem has quit IRC14:03
*** mriedem has joined #openstack-cinder14:03
*** julim has quit IRC14:03
*** eharney has joined #openstack-cinder14:03
*** lpetrut has quit IRC14:03
*** obutenko has quit IRC14:05
*** porrua has joined #openstack-cinder14:06
*** julim has joined #openstack-cinder14:06
*** timcl has quit IRC14:07
*** zhipeng has joined #openstack-cinder14:07
*** tshefi has joined #openstack-cinder14:12
*** ank2090 has quit IRC14:13
*** Yogi11 has joined #openstack-cinder14:13
*** dustins_ has joined #openstack-cinder14:13
*** thingee has joined #openstack-cinder14:14
*** dustins has quit IRC14:16
*** Yogi1 has quit IRC14:16
*** git-harry has quit IRC14:21
*** esker has joined #openstack-cinder14:21
*** git-harry has joined #openstack-cinder14:21
*** _cjones_ has joined #openstack-cinder14:21
*** esker has quit IRC14:22
*** lpetrut has joined #openstack-cinder14:24
*** _cjones_ has quit IRC14:25
*** annashen has joined #openstack-cinder14:26
*** liu has joined #openstack-cinder14:26
*** Mandell has quit IRC14:30
*** annashen has quit IRC14:30
*** e0ne is now known as e0ne_14:31
*** git-harry has quit IRC14:31
kaisers1Q: Is there a defined process how bugfixes are prioritized currently? My CI found a bug in our driver and I've a (small and easy) fix for this. Since the CI keeps on reporting false negatives because of the bug, how can i bring this forward? Or simply shut up & wait? :-D14:36
*** e0ne_ has quit IRC14:36
thingeekaisers1: feel free to ping me with the bug and I can target it to the RC14:37
thingeeI need to hop off, but still post it and I'll pick it up in the logs later.14:37
kaisers1thingee: ok, thanks!14:37
thingeeyou're welcome14:37
*** thingee has quit IRC14:37
kaisers1thingee: ping https://review.openstack.org/#/c/166224/ :)14:37
*** git-harry has joined #openstack-cinder14:38
*** e0ne has joined #openstack-cinder14:38
*** crose has joined #openstack-cinder14:38
*** lpetrut has quit IRC14:38
openstackgerritRichard Hedlind proposed openstack/cinder: Properly remove host object from ISE  https://review.openstack.org/16607914:41
*** reed has joined #openstack-cinder14:42
*** lpetrut has joined #openstack-cinder14:45
*** git-harry has quit IRC14:47
*** tsekiyama has joined #openstack-cinder14:47
*** git-harry has joined #openstack-cinder14:47
*** timcl has joined #openstack-cinder14:50
*** git-harry has quit IRC14:50
*** git-harry has joined #openstack-cinder14:50
*** EmilienM|PTO is now known as EmilienM14:51
*** annegentle has joined #openstack-cinder14:51
*** ankit_ag has quit IRC14:52
*** Yogi11 has quit IRC14:53
*** lpetrut has quit IRC14:54
*** hemnafk is now known as hemna14:55
DuncanTkaisers1: Always worth pinging the channel (not just Mike) for such bugs, since any two cores can merge the fix14:57
*** timcl has left #openstack-cinder15:00
*** git-harry has quit IRC15:00
*** git-harry has joined #openstack-cinder15:01
*** Yogi1 has joined #openstack-cinder15:01
kaisers1DuncanT: Yep. I asked in general and he told me to ping him in reply, that's why i addressed him. Will ping more general next time! :)15:02
*** thingee has joined #openstack-cinder15:02
thingeekaisers1: targeted15:03
*** mriedem has quit IRC15:03
kaisers1thingee: yeeha!15:03
*** mriedem has joined #openstack-cinder15:04
*** jaypipes has quit IRC15:05
thingeeand you're right DuncanT. I was just saying if I know what is a priority to people at some point I'm going to have an etherpad to help reviewers focus, just like I've done with the milestone bps15:05
*** sgotliv has quit IRC15:06
DuncanTkaisers1: Many of us nose at reviews pinged to specific people anyway, so as long as it happens on the channel you're probably ok.15:06
kaisers1DuncanT: ok15:07
thingeeDuncanT: what I love about kaisers1's review is a CI telling me that this patch works. :)15:10
thingeethis is a beautiful thing15:10
avishayCI is nice...  i can go to a customer site and be rather sure that the driver works without having the storage to test against, rather than being ready to debug some random vendor driver15:11
hemnaavishay, we'll just debug a random oslo bug instead.15:13
avishayhemna: when i have a product based on kilo in my hand, i know that everything will work on all drivers (well at least what's covered in the tests)15:14
avishayhemna: that's pretty friggin good15:14
*** lpetrut has joined #openstack-cinder15:19
*** wN has joined #openstack-cinder15:20
*** thingee has quit IRC15:20
*** nlevinki has quit IRC15:26
*** emagana has joined #openstack-cinder15:28
*** bkopilov has quit IRC15:31
*** jproulx has joined #openstack-cinder15:31
*** rushil_ has quit IRC15:32
jproulxHate to ask a usage question in a dev channel, but coming up empty elsewhere15:32
jproulxHow do you apply a quota_class to a project? defining them is easy but can't find how to actually use them...15:32
*** harlowja_at_home has joined #openstack-cinder15:32
*** annegentle has quit IRC15:33
*** dannywilson has joined #openstack-cinder15:35
*** markstur has quit IRC15:35
hemnaavishay, if you are still around, can you take a look at this one for me?  https://review.openstack.org/#/c/167045/15:36
hemnaI'd like to get that in15:36
*** nlevinki has joined #openstack-cinder15:41
avishayhemna: how do we get to the point where detach is called on a volume with no attachments?15:42
*** _cjones_ has joined #openstack-cinder15:42
hemnaavishay, tempest doing bad things15:42
*** ronis has quit IRC15:42
avishayis that something a user can do?15:43
hemnawe had a tempest test yesterday that was calling detach on a volume it had already detached.15:43
avishayhemna: tempest calls manager directly, and not cinder-api?15:43
hemnathere is nothing that prevents folks from calling the detach API in dumb ways15:43
hemnaavishay, nah it goes through the API15:43
*** annegentle has joined #openstack-cinder15:43
avishayhemna: so there is a race and this makes the race smaller?15:44
*** anshul has quit IRC15:44
hemnanot sure there is a race really15:44
hemnajust being safe in the volume manager.15:44
*** _cjones_ has quit IRC15:44
*** _cjones_ has joined #openstack-cinder15:44
avishayhemna: don't we switch to detaching state, and not allow detach while in that state?15:44
hemnabecause someone can call the detach API multiple times for no reason.15:44
hemnathis is way after the volume has been detached15:45
hemnaand it's available again15:45
avishayhemna: this should all be checked atomically in the API, this patch is wrong IMO15:45
hemnawell I can add that as well15:46
*** changbl has joined #openstack-cinder15:46
hemnaI still think it's better to be safe here15:46
*** asselin_ has quit IRC15:46
hemnathe API does no checks really.15:46
avishayso here still if two instances get to "elif len(attachments) == 1" at the same time, they will both try to detach the same attachment15:47
hemnaother than if the volume exists.15:47
avishayhemna: that's the point of states, to avoid these things15:47
hemnathat race can happen in the API as well though15:47
avishayhemna: otherwise we would have to double and triple check everything everywhere, and still have races, like we do here15:47
*** appsdesh has joined #openstack-cinder15:47
avishayhemna: not if you atomically update the state15:48
avishayhemna: if it atomically updates from 'in-use' to 'detaching', then it should be safe15:48
hemnathere is no state checking in the API currently15:48
hemnathat's a much bigger change15:48
hemnaand kinda out of the scope of this small fix15:48
avishayhemna: don't you regret asking for my review? :)15:49
hemnahehe no15:49
*** bkopilov has joined #openstack-cinder15:49
hemnathat's the current API entry point for detach15:49
hemnano state checks at all.15:49
avishayhemna: https://github.com/openstack/cinder/blob/master/cinder/volume/api.py#L51515:49
hemnaonly a simple, does the volume even exist15:50
hemnathis is outside of that15:50
*** lpetrut has quit IRC15:50
hemnathis is a direct call to detach15:50
*** annegentle has quit IRC15:50
hemnaw/o calling begin detaching15:50
avishayhemna: well calling detach without begin_detaching is wrong i think15:50
hemnathe touches on the PAIN of the Nova -> cinder coordination15:51
hemnawhich completely sucks IMHO15:51
avishayhemna: the test should model how nova attaches/detaches15:51
hemnathis is true, but15:51
hemnawe still need to be safe in cinder15:51
hemnabecause people do dumb things15:51
hemnaanyway, I still think this patch is good as it is.15:51
hemnathe bigger problem is 2 things really15:51
avishaythe patch just turns a problem into a race15:52
hemna1) the nova -> cinder API (Which we are going to talk about in Vancouver)15:52
hemna2) the state management of volumes, which has been a topic forever :(15:52
avishayit will fail sporadically instead of consistently15:52
*** lpetrut has joined #openstack-cinder15:52
*** Miouge has joined #openstack-cinder15:53
*** wN has left #openstack-cinder15:53
hemnathe only way to really fix that though15:53
hemnais to collapse begin_detaching and detach_volume into 115:54
hemnaatomic action.15:54
hemnabasically fixing part 1) of the issues I listed above15:54
hemnawhich isn't going to happen for K.15:54
avishayhemna: well that's a good thing to do for liberty, but for now, i would say fix the test15:54
*** markus_z has quit IRC15:55
hemnawhat do you suggest ?15:55
avishayhemna: have the test call the proper APIs15:57
hemnaI already got the tempest test updated15:57
hemnato stop calling detach twice.15:57
avishayhemna: and maybe assert in the 'detach' API that we are in 'detaching' state15:57
hemnabut the volume manager still needs to be safe in trying to pull the attachment out of the list.15:58
*** rushiagr is now known as rushiagr_away15:58
hemnathat was the offending tempest test15:58
hemnait was shutting down the VMs (boot from cinder volume), and then calling detach directly after.15:59
avishayhemna: it's redundant. the fact that the volume was 'in-use' and went to 'detaching' must ensure that there is an attachment there waiting to be attached.15:59
hemnait was long after that was over though16:00
hemnait was available16:00
hemnaand they called detach16:00
avishayhemna: are you going to check in every manager call that assumes 'available' that it has no attachments?16:00
*** emagana has quit IRC16:00
avishaythe test is wrong, and i don't agree with adding code to the product to make tests run, it should be the other way around16:00
hemnathe tests were wrong16:01
hemnaand I had them fixed16:01
hemnaI thought I covered that ?16:01
*** jdurgin1 has joined #openstack-cinder16:01
hemnaall this does is ensure that if someone calls us incorrectly, that we give a meaningful error instead of raising an invalid index exception.16:01
hemnait doesn't solve anything else.16:02
hemnait isn't meant to.16:02
*** dflorea has joined #openstack-cinder16:02
hemnathis patch can't fix the nova -> cinder api issues.16:02
avishayhemna: no, if they are calling detach without begin_detaching, it is wrong16:02
hemnaok nevermind.16:02
hemnaI give up.16:02
*** EmilienM is now known as EmilienM|PTO16:02
*** emagana has joined #openstack-cinder16:03
avishayhemna: sorry, i just don't think it's right.  others can override my decision of course.16:03
hemnaI'm just not sure how many times I need to state the same thing.16:03
hemna<hemna> the tests were wrong16:03
hemna<hemna> and I had them fixed16:03
hemna<hemna> I thought I covered that ?16:03
avishayhemna: no, the tests were wrong and are still wrong16:04
*** melwitt has joined #openstack-cinder16:04
smcginnishemna: I agree the change is a good protection.16:04
* smcginnis dons flak jacket16:05
avishayhemna: they need to call detach after begin_detaching16:05
hemnasmcginnis, +116:05
*** zhipeng has quit IRC16:05
avishayhemna: otherwise, add checks in manager.py for everything that the API should check for16:05
*** jistr has quit IRC16:05
avishayhemna: you can't rely on state anymore.  want to extend a volume? check that there are no attachments.  want to attach with no multi-attach?  check no attachments.16:06
avishayhemna: it will never end16:06
*** markstur has joined #openstack-cinder16:06
avishayhemna: that's why the API does checks and puts the volume into a state that says "the volume is in state foo, you can rely on that"16:06
avishayhemna: otherwise just reimplement all the checks in API in the manager16:07
*** btran has joined #openstack-cinder16:08
avishayhemna: all of those detach and attach APIs are a protocol with Nova and must be followed for correct behavior. if not, the test is wrong.16:08
hemnaavishay, the API doesn't do checks.16:08
hemnathere are no checks, other than the volume exists.16:08
hemnahence back to my original issue of 1) the nova -> cinder API is broken.16:09
avishayhemna: if the test would call the correct APIs (begin_detaching), checks would be called16:09
hemnaand is out of the scope of this fix.16:09
hemna<hemna> <hemna> the tests were wrong16:09
hemna<hemna> <hemna> and I had them fixed16:09
asselinflip214, hi16:09
avishayhemna: dude do whatever you want, there is no way you can convince me that this fix is the correct solution16:09
*** leeantho has joined #openstack-cinder16:09
eharneyif Cinder is throwing IndexError, i think we need to check enough to not do that regardless of what is supposed to happen...16:09
avishayhemna: it's just a hack to fix a bad test16:09
hemnait's not a solution, it's a messaging fix really.16:10
hemnathe entire point of this patch is to raise a meaningful message16:10
hemnainstead of throwing an invalid index error16:10
hemnawhich is completely useles.16:10
hemnaeven useless.16:10
avishaywe should never ever ever get to this point16:10
hemnaback to 1)16:10
hemnathe nova -> cinder api is broken.16:10
hemnahas been for ages16:10
hemnaand is outside the scope of this patch.16:11
*** harlowja_at_home has quit IRC16:11
hemnathere is a lot of cleanup I'd like to see happen for L16:12
hemnabut I'm afraid that we won't get it done in time for the Nova feature freeze cutoff16:12
hemnaand it probably breaks backwards compatibility.16:12
hemnaNova(Kilo) -> Cinder(Liberty)16:12
*** jungleboyj has quit IRC16:13
avishayhemna: OK, fine with me.  Please just change the message to error instead of warning.16:13
*** e0ne is now known as e0ne_16:13
avishayhemna: what a mess16:13
hemnatotal mess16:13
hemnahas been for a long time, which is why I've wanted to raise the subject of the nova -> cinder api16:13
hemnaI'll update the warning -> error16:14
avishayhemna: the test is broken, but users can still call this directly i guess16:14
*** kallebe has joined #openstack-cinder16:14
hemna<hemna> <hemna> <hemna> the tests were wrong16:14
hemna<hemna> <hemna> <hemna> and I had them fixed16:14
hemnabut yes, users can be dumb16:14
avishayhemna: thanks, please do it soon so i can +216:14
jgriffithhemna: I'm still kinda confused on this... sorry16:14
hemnaok, lets party16:14
avishayhemna: again, i know you fixed the test to not call it twice, but any code that calls detach without begin_detach is wrong16:15
jgriffithhemna: I explicitly stated in the code you merged that this needed to handle the existing case16:15
*** dulek has quit IRC16:15
jgriffithhemna: also, the manager is the wrong place for this check16:15
jgriffithhemna: there should be an early check in cinder/volume/api16:15
hemnayah, I mentioned that I could add that16:15
jgriffithhemna: check the state of the volume... if it's available noop16:15
hemnathe existing API didn't do any of those checks FWIW16:15
jgriffithhemna: it didn't need to16:16
hemnasure it did16:16
hemnafor all the same problems16:16
*** chlong has quit IRC16:16
hemnasomeone could call it directly w/o calling begin_detaching16:16
jgriffithhemna: so I'm not looking to have a big debate here :)16:16
jgriffithhemna: I'm just saying the appropriate place to check this IMO is in the API16:16
jgriffithbefore you even get here16:16
jgriffithhemna: it's even fine/good to have both16:17
hemnathat's fine, like I said I could add that16:17
*** shilpa_ has joined #openstack-cinder16:17
jgriffithhemna: but we shouldn't go past the API layer IMO16:17
hemnayah that's cool16:17
hemnaI agree16:17
jgriffithhemna: awesome16:17
hemnaI'm sorry16:17
jgriffithhemna: for?16:17
hemnaI wasn't trying to argue against it16:17
jgriffithhemna: oh... LOL16:17
*** melwitt has quit IRC16:17
jgriffithno worries16:17
jgriffithhemna: I just know how much you hate that whole mechanism between nova and cinder16:18
hemnaI was just trying to relay information about the fact that the API didn't do this already, when in fact it should have been.16:18
jgriffithhemna: you don't need to tell me again :)16:18
jgriffithhemna: yeah, I thought there were some things in there... you're right they should've been16:18
jgriffithif they weren't... well then they weren't16:18
*** thingee has joined #openstack-cinder16:18
jgriffithwe should look at Juno and open a bug if we don't do any checks16:19
hemnayah no big.16:19
*** liu has quit IRC16:19
hemnasame thing in Juno16:19
hemnano checks really16:19
hemnaother than the volume exists.16:19
*** junhongl has quit IRC16:20
*** Miouge has quit IRC16:20
*** melwitt has joined #openstack-cinder16:20
jgriffithhemna: http://goo.gl/bEk8IW16:20
*** Miouge has joined #openstack-cinder16:20
hemnayah that's assuming you call begin_detaching first16:20
jgriffithand http://goo.gl/DKwjSo16:21
jgriffithhemna: right16:21
hemnathe problem is, that there is no guarantee that anyone does16:21
jgriffithhemna: I get ya16:21
hemnawhich the tempest tests weren't16:21
*** Mandell has joined #openstack-cinder16:21
jgriffithhemna: I'm just saying IMO those are the places we should have the check16:21
jgriffithhemna: 504 should have the same call16:21
jgriffithhemna: and we shouldn't monkey with extensions going directly to manager or anything goofy like that :)16:22
jgriffithhemna: if we are, we need to fix that up as well16:22
hemnaso I was going to add the check in the volume_actions.py _detach16:22
*** sgotliv has joined #openstack-cinder16:22
hemnawhich is the entry point for detach16:22
jgriffithhemna: I'd prefer it be in volume/api personally16:22
jgriffithhemna: just because it's a more common point16:23
avishayjgriffith: +116:23
hemnaok that sounds good.16:23
hemnaI'll do that then.16:23
jgriffithhemna: there can be other contrib/extensions etc that call detac16:23
jgriffithhemna: it's more to "manage" that way16:23
hemnaok gotcha16:23
jgriffithhemna: this way you just squash it in one place16:23
hemnayah I keep forgetting that other folks can drop in extensions as well16:24
jgriffithassuming we are sure to enforce that everything goes through api.py as it should16:24
jgriffithhemna: thank you!16:24
hemnathanks for the help guys16:24
hemnaI have to run to a beating....16:24
hemnaI'll post an updated patch in a bit.16:24
jgriffithhemna: FWIW, the only thing that I see things under cinder/api/xxxx should ever go direct to anything is maybe driver extensions16:24
*** bkopilov has quit IRC16:25
jgriffithanyway... thanks for fixing that up16:25
*** Miouge has quit IRC16:26
hemnaavishay, thanks for the help16:26
avishayhemna: sure, sorry for the "argument".  this check should be an assert at best, and the test should call the proper apis, and and and and and and, but that's a lot of ands16:27
*** e0ne_ is now known as e0ne16:27
avishayhemna: so for now, i guess this is as good as it gets :)16:27
*** Bsony has quit IRC16:30
*** Miouge has joined #openstack-cinder16:32
*** Adriano_ has quit IRC16:36
*** nkrinner has quit IRC16:37
*** annegentle has joined #openstack-cinder16:37
*** annegent_ has joined #openstack-cinder16:38
*** rushil has joined #openstack-cinder16:39
*** nlevinki has quit IRC16:39
*** jaypipes_ has joined #openstack-cinder16:41
*** junhongl has joined #openstack-cinder16:42
*** annegent_ has quit IRC16:42
*** annegentle has quit IRC16:43
*** leeantho has quit IRC16:47
*** TobiasE has quit IRC16:48
*** alecv has quit IRC16:49
*** shilpa_ has quit IRC16:57
*** dflorea has quit IRC16:59
*** dflorea has joined #openstack-cinder16:59
*** emagana has quit IRC17:02
*** dflorea has quit IRC17:04
*** dflorea has joined #openstack-cinder17:05
*** ronis has joined #openstack-cinder17:05
*** avishay has quit IRC17:06
*** thingee has quit IRC17:11
*** annegentle has joined #openstack-cinder17:14
*** Adriano_ has joined #openstack-cinder17:16
*** harlowja_away is now known as harlowja17:18
*** e0ne is now known as e0ne_17:19
*** vilobhmm1 has joined #openstack-cinder17:20
*** vilobhmm1 has joined #openstack-cinder17:21
*** e0ne_ is now known as e0ne17:22
*** vilobhmm11 has joined #openstack-cinder17:24
*** vilobhmm1 has quit IRC17:25
*** rushil has quit IRC17:28
*** annashen has joined #openstack-cinder17:28
*** Mandell has quit IRC17:28
*** patrickeast has joined #openstack-cinder17:29
*** dflorea has quit IRC17:32
*** annashen has quit IRC17:33
*** annegentle has quit IRC17:34
*** Mandell has joined #openstack-cinder17:35
*** tbarron has joined #openstack-cinder17:35
*** Bsony has joined #openstack-cinder17:36
*** thingee has joined #openstack-cinder17:36
*** dflorea has joined #openstack-cinder17:37
*** Bsony has quit IRC17:41
*** annegentle has joined #openstack-cinder17:43
*** lpetrut has quit IRC17:44
*** lpetrut has joined #openstack-cinder17:44
*** jdurgin1 has quit IRC17:44
*** jproulx has left #openstack-cinder17:46
*** dflorea has quit IRC17:53
*** dflorea has joined #openstack-cinder17:53
*** dflorea has quit IRC17:58
*** sgotliv has quit IRC17:58
*** emagana has joined #openstack-cinder18:03
*** jamielennox|away is now known as jamielennox18:04
*** tbarron1 has joined #openstack-cinder18:06
*** emagana has quit IRC18:08
*** tbarron has quit IRC18:09
*** jordanP has quit IRC18:10
*** leeantho has joined #openstack-cinder18:15
*** annegentle has quit IRC18:16
*** e0ne has quit IRC18:21
*** yuriy_n17 has quit IRC18:23
*** reed has quit IRC18:24
tbarron1hema, tsekiyama: you two know about the status of multiple iscsi portals and nova volume-attach, right?18:27
tbarron1hemna: ^^18:27
tbarron1sp problem today18:28
*** emagana has joined #openstack-cinder18:29
tsekiyamatbarron1: cinder part is in, but nova side is not yet merged18:29
tbarron1tsekiyama: so will it get merged by nova adopting hemna's brick work?18:29
tsekiyamatbarron1: it'll be after Liberty. maybe by using os-brick.18:30
tbarron1or by another path?18:30
tbarron1tsekiyama: when I fail over one array controller to another,  nova attaches fail18:30
*** annegentle has joined #openstack-cinder18:30
tbarron1even though I/O works fine with attaches done prior to the failover18:31
tbarron1because multipath is working (DM-MPIO) correctly.18:31
tbarron1tsekiyama: your patch will fix that I think?18:31
tsekiyamatbarron1: Right.18:31
tsekiyamatbarron1: Currently nova needs main path alive to attach volumes.18:32
*** lpetrut has quit IRC18:32
*** Maike has joined #openstack-cinder18:33
tbarron1tsekiyama: would I also need to make changes in my backend driver to pass connection details to nova differently?18:33
*** emagana has quit IRC18:33
tsekiyamatbarron1: My nova patch is applied, or if nova moves to os-brick, and the cinder backend are configured to provide multiple paths information to nova, it will be succeed to attach.18:34
tsekiyamatbarron1: yeah, backend driver also need to be modified so that it can multiple paths information in new style in "initialize_connection" method.18:34
tbarron1tsekiyama: I see what you did with LVM, so that's a good model.18:35
*** reed has joined #openstack-cinder18:35
tbarron1tsekiyama: so on the Nova side, just https://review.openstack.org/#/c/140878/, right?18:35
tsekiyamatbarron1: Right. but it is now conflicting with anish's patch..18:36
tbarron1tsekiyama: looking ...18:36
tsekiyamatbarron1: oops sory18:36
tsekiyamatbarron1: that's not right one. it is old18:36
tsekiyamatbarron1: https://review.openstack.org/#/c/134683/18:37
tbarron1tsekiyama: yeah, I need to get up to date :-)18:37
tbarron1tsekiyama: th18:37
tsekiyamatbarron1: ^ is the right one. for multipath enhancement18:37
*** anshul has joined #openstack-cinder18:37
tsekiyamatbarron1: https://review.openstack.org/#/c/140878/ is for single path failover, but data model is changed, so it doesn't work.18:38
tbarron1tsekiyama: thanks.  I am adding myself to the review for  https://review.openstack.org/#/c/134683/ so that I can stay up to date.18:39
tbarron1tsekiyama: is resolving the merge conflict hard?  I want to try it end-to-end with your cinder fix and a fix I will do to our drivers.18:40
tbarron1tsekiyama: I don't mean to be rushing you with that question.  I know it can't be targeted till Liberty.18:41
tsekiyamaHmm, i don't believe it's that hard, but I need more look into it..18:41
tbarron1tsekiyama: understood.  Thank you very much for your help now!18:42
*** dannywilson has quit IRC18:42
tsekiyamatbarron1: no problem18:42
*** dannywilson has joined #openstack-cinder18:42
*** annegentle has quit IRC18:45
*** dflorea has joined #openstack-cinder18:48
*** lpabon has joined #openstack-cinder18:48
openstackgerritTom Swanson proposed openstack/cinder: Dell Storage Center API change fails init_conn  https://review.openstack.org/16736618:48
*** anshul has quit IRC18:52
*** anshul has joined #openstack-cinder18:52
*** dflorea has quit IRC18:53
*** dflorea has joined #openstack-cinder18:55
*** dflorea has quit IRC18:57
*** vilobhmm11 has quit IRC18:59
*** jistr has joined #openstack-cinder19:02
*** vilobhmm1 has joined #openstack-cinder19:02
*** vilobhmm1 has quit IRC19:02
*** vilobhmm11 has joined #openstack-cinder19:02
*** dflorea has joined #openstack-cinder19:03
*** Rockyg has joined #openstack-cinder19:03
*** e0ne has joined #openstack-cinder19:11
*** eharney has quit IRC19:20
nikesh_vedamsjgriffith : hi yesterday i was trying for this http://docs.openstack.org/admin-guide-cloud/content/volume-migration.html which have a example of  migration of attached volume and which is being addressed by https://review.openstack.org/#/c/167134/3/cinder/volume/manager.py19:20
*** diemt has joined #openstack-cinder19:21
*** guest1 has joined #openstack-cinder19:21
*** tbarron1 has quit IRC19:22
*** tbarron has joined #openstack-cinder19:23
*** tellesnobrega has quit IRC19:29
*** emagana has joined #openstack-cinder19:30
*** jistr has quit IRC19:31
*** tellesnobrega has joined #openstack-cinder19:33
*** emagana has quit IRC19:34
*** dustins_ is now known as dustins19:34
*** vilobhmm11 has quit IRC19:37
*** emagana has joined #openstack-cinder19:37
*** vilobhmm1 has joined #openstack-cinder19:39
*** vilobhmm1 has quit IRC19:39
*** morganfainberg is now known as needslesscoffee19:39
*** guest1 has quit IRC19:43
*** _cjones_ has quit IRC19:45
*** _cjones_ has joined #openstack-cinder19:46
*** needslesscoffee is now known as morganfainberg19:50
*** _cjones_ has quit IRC19:50
*** _cjones_ has joined #openstack-cinder19:52
e0nejgriffith, boris-42: ping. am i right that we don't test cinder api v2 with rally at all?19:53
*** eharney has joined #openstack-cinder19:53
*** kallebe has left #openstack-cinder19:53
*** lpabon has quit IRC19:56
boris-42e0ne: yep we are not testing it yet19:56
anishtsekiyama: I'm blanking out on this, where was the conflict ?19:56
e0nebut the good news: no need to investigate performance impact for api v119:57
SwansonIf any cores are around could you take a peek at https://review.openstack.org/#/c/167366/119:57
SwansonReally simple change.19:57
anishe0ne: ping, you wanted this held off till code_cleanup_batching https://review.openstack.org/#/c/158926/19:57
anishalso thingee, if you can take a look. hemna's signoff is contingent on yours19:58
*** porrua has quit IRC19:59
boris-42e0ne: but tests for v2 api can be done20:00
anishe0ne: no +2 ? :(20:00
anishI just realised I've always assumed core reviewers only do +2, -1 or -220:01
e0neanish: i'm waiting for hemna20:01
anishlol, henma is waiting for thingee20:02
e0neboris-42: i want try run current scenarios with api v220:02
anishslowly turning into #openstack-gentoo20:02
*** emagana has quit IRC20:04
*** agentle has joined #openstack-cinder20:06
tsekiyamaanish: ah, don't worry, my nova multipath enhancement patch (which is not targeted to Kilo now) conflicted due to auto-merge failure with your changes (already merged) for --interface option support in iscsiadm.20:10
*** xyang1 has joined #openstack-cinder20:13
anishtsekiyama: please ping me when you retarget it again. I would like to get multipath working simultaneously with interface support at some point, something that can show weird behaviour right now20:13
anish(I really can't think of a good way that these two can co-exist)20:14
tsekiyamaanish: It might be nice if you could apply your hardware iscsi transport support into os-brick20:15
anishtsekiyama: https://review.openstack.org/#/c/152740/20:15
anishI had held that off for liberty specs being open, haven't followed up20:16
Swansonhemna: Thanks for the review.20:16
*** cdelatte has quit IRC20:16
anishhemna: was there a  final decision on a specs repo for brick btw ? or is it supposed to go under cinder20:17
*** vilobhmm1 has joined #openstack-cinder20:17
*** vilobhmm1 has quit IRC20:17
*** anshul has quit IRC20:17
hemnait's supposed to be under Cinder afaik20:17
hemnawe can revisit that in L20:17
*** vilobhmm1 has joined #openstack-cinder20:17
anishcool to remove the WIP from this then, now that specs are open ? https://review.openstack.org/#/c/152740/20:18
hemnaare specs open?20:18
hemnawe haven't even finished K yet20:18
hemnaanish, have to ask thingee20:18
anishoh my bad. nova/keystone specs are open, I assumed everyone is roughly on the same timeline20:19
*** xyang has joined #openstack-cinder20:19
DuncanTWe were planning on not opening specs until RC time or so, to encourage people to work on bug fixes and testing20:19
anishgotcha, misunderstanding on my part20:19
*** vilobhmm1 has quit IRC20:22
*** _cjones_ has quit IRC20:23
*** _cjones_ has joined #openstack-cinder20:23
*** vilobhmm1 has joined #openstack-cinder20:23
*** enterprisedc has joined #openstack-cinder20:25
*** dflorea has quit IRC20:28
*** dflorea has joined #openstack-cinder20:28
*** dflorea has quit IRC20:33
*** dflorea has joined #openstack-cinder20:34
thingeehemna, anish fine with keeping brick related specs under Cinder. As DuncanT mentioned, we talked about opening for L specs after RC focus. http://eavesdrop.openstack.org/meetings/cinder/2015/cinder.2015-03-18-16.00.log.html#l-33720:35
*** jcru has quit IRC20:35
*** rushil has joined #openstack-cinder20:35
*** Adriano_ has quit IRC20:36
*** reed has quit IRC20:42
boris-42e0ne: so we were working on that feature20:43
hemnaman this is painful20:46
hemnamoving the checks into the rpcapi has turned into a CF20:46
e0neboris-42: great! are any estimates?20:46
*** markstur has left #openstack-cinder20:46
hemnaa CF of unit test changes/failures.20:47
*** nellysmitt has joined #openstack-cinder20:47
*** nellysmitt has quit IRC20:47
smcginnishemna: Personally I would be all for your simple protection now, bigger refactoring/protection later.20:49
smcginnishemna: Just my opinion.20:49
hemnayah, i'm seriously leaning that way20:50
hemnathis is requiring me to change the rpcapi20:50
smcginnishemna: It's still better than it was. No worse.20:50
hemnaand a boatload of unit tests20:50
*** angela-s has quit IRC20:51
boris-42e0ne: so for now thre is no estimates20:51
e0neboris-42: are you going to do it for the next release or next+n?20:52
e0nethingee: hi! any updates about cinderclient release date?20:54
e0nethingee: are you going to do it after RC or after Kilo released?20:54
boris-42e0ne: so not in next 100%20:56
boris-42e0ne: next release will be on next week20:56
*** Maike has quit IRC20:56
*** dflorea has quit IRC20:58
*** dflorea has joined #openstack-cinder20:59
thingeee0ne: need this to land..but need to do some rework still https://review.openstack.org/#/c/145613/21:00
thingeebut yeah at this point it will be near rc cut21:00
thingeebeen having operators ask about this on the ops ML21:00
*** Yogi1 has quit IRC21:00
e0nethingee: agree. it's very important feature. i was asked few times about it from our support21:01
anishthingee: thanks for hte link21:02
thingeee0ne: understood. I will make announcements about it. Need to also see how much trouble I'm going to be for being this late with the release21:02
thingeeof the client21:02
thingeeanish: np21:02
*** tbarron has quit IRC21:02
e0nethingee: thank you for update21:03
*** dflorea has quit IRC21:03
anishhemna: I have 2 cores signed off on https://review.openstack.org/#/c/158926/, do you still want thingee to look at it ?21:03
*** vilobhmm1 has quit IRC21:03
* thingee looks21:04
hemnaanish, yes21:04
hemnasince it's not a bug fix21:04
hemnaI don't want to +A it.21:04
anishko !21:04
hemnaleave it to the boss.21:04
*** vilobhmm1 has joined #openstack-cinder21:04
* thingee shakes fist at jungleboy for starting this boss nonsense21:04
*** dflorea has joined #openstack-cinder21:04
thingeedavid wang is our boss. end of story21:05
*** melwitt has quit IRC21:05
anishdo you prefer "everyone core reviewer is equal, some are just more equal" ?21:05
*** melwitt_ has joined #openstack-cinder21:05
* hemna hides.21:05
*** vilobhmm1 has quit IRC21:05
*** vilobhmm11 has joined #openstack-cinder21:05
*** melwitt_ has quit IRC21:05
*** emagana has joined #openstack-cinder21:05
thingeeanish: what?21:05
*** melwitt has joined #openstack-cinder21:05
anishthingee: it's an animal farm reference, ignore me21:06
*** tbarron has joined #openstack-cinder21:06
* thingee needs to read that still and ashamed to report that21:06
e0neanish, hemna: TBH, i prefer to keep unit tests as simple, as possible and fix them w/o cleanup batch21:06
*** openstackgerrit has quit IRC21:07
hemnait's impossible to do that with what I'm doing on this patch21:07
hemnaas it changes the rpcapi21:07
anishthingee: heh, the whole book is like 90 pages long. you could finish it today21:07
*** openstackgerrit has joined #openstack-cinder21:07
e0nea simple 'equals order' fix could save few hours:)21:08
e0nehemna: what rpc changes are you talking about?21:08
hemnaon this bugfix I'm working on21:08
hemnaavishay and jgriffith wanted me to put the attachment checks into the api instead of the manager21:08
hemnasince I'm loading the attachment, it doesn't make sense to pass an attachment_id into the manager, when I already have the attachment object21:09
hemnaand that causes unit test changes21:09
hemnamore changes than I wanted for this simple fix21:09
hemnahonestly, I think this is a bigger problem that's outside of the scope of this issue I'm trying to resolve21:09
hemnae0ne, https://review.openstack.org/#/c/167045/21:10
*** emagana has quit IRC21:10
*** jcru has joined #openstack-cinder21:13
e0nehemna: i need to go deeper to this bug and patch to be on the same page with you21:13
jgriffithhemna: hate to say it but maybe it is indicatvie that you're model of using attachment ID's is flawed21:17
jgriffithhemna: :)21:17
hemnamoving the checks into the api is a bigger change21:17
hemnait just propogates to unit test changes, etc.21:17
jgriffithhemna: but that's where they belong21:17
* e0ne bb tomorrow (in ~10 hours)21:18
hemnayah I agree21:18
*** e0ne is now known as e0ne_21:18
jgriffithhemna: the only change that should be there IMO anyway is "volume['status'] == 'in-use'"21:18
*** akerr has quit IRC21:18
jgriffithhemna: I'll be curious to see why that causes so much problem in tests etc21:19
hemnaI guess I was going a different route21:19
hemnawhich is a bit harder I guess21:19
jgriffithhemna: oh?21:19
hemnaI was moving the existing checks in the manager21:19
hemnainto the API21:19
jgriffithhemna: hmmm...  how come?21:20
jgriffithhemna: as opposed to a one line add in volume/api.py "if volume['status'] != 'in-use': return"21:20
*** reed has joined #openstack-cinder21:20
*** Lee1092 has quit IRC21:20
jgriffithhemna: in volume.api:detach_volume()21:20
jgriffithhemna: fixed, done21:21
hemnathose checks21:21
hemnaas I thought that's what you guys wanted21:21
jgriffithhemna: not me21:21
jgriffithin fact I hate those checks anyway, and I certainly wouldn't put those in volume.api21:21
hemnaheh man21:21
jgriffithhemna: just add the one liner check and be done21:22
jgriffithhemna: or make your life as difficult as you can... your call21:22
*** ndipanov has quit IRC21:22
hemnaI get so many different opinions at times21:22
hemnait's impossible to get this stuff right.21:22
hemnaok I'll scratch my current path21:22
jgriffithhemna: well then ignore me and i'll leave you alone :)21:22
jgriffithhemna: but that's the exact same thing I said this morning21:22
hemnaand I'll add your check and be done with it21:22
jgriffithno different21:22
hemnayah I guess I got a different view from avishay, so I was trying to accomodate21:23
hemnableh, this will be easier21:23
jgriffithOk, then count me out21:23
jgriffithdo what you and Avishay agreed on, I wasn't present for all of that21:23
jgriffithhemna: I guess harder and complex is always better than easy and simple21:24
hemnaheh no21:24
anishaww yisss ! my work here is done21:24
hemnathat's what I'm abandoning it21:24
* anish disappears till RC21:24
jgriffithhemna: I think it's crazy to have all that logic in api.py personally21:25
* hemna hits the reset button21:26
*** julim has quit IRC21:26
hemnaso the api has lots of checks in other methods21:28
hemnasuch as delete()21:28
hemnaso it's kinda confusing really where the checks should live21:29
jgriffithhemna: ok, ignore me then21:29
*** emagana has joined #openstack-cinder21:29
hemnaputting them in the api helps prevent rpc calls that will just fail otherwise21:29
jgriffithhemna: I don't understand what's hard or confusing?21:30
jgriffithhemna: we do this for things like delete, delete-snapshot etc etc21:30
hemnagoing your route, simply adding an check on 'in-use' is easy21:30
hemnaI'm just trying to understand in general, where the checks should live.21:31
jgriffithhemna: it should live in volume/api !21:31
jgriffithhemna: just like "begin_detaching"21:31
jgriffiththe fact that it's not there is an oversight/bug21:31
hemnaok, that's what I thought, but was confused by what you said a bit earlier21:31
*** annashen has joined #openstack-cinder21:32
jgriffithhemna: what did I say that was confusing?21:32
hemna"jgriffith> hemna: I think it's crazy to have all that logic in api.py personally"21:32
jgriffithhemna: YES... all the logic that you have for that call in manager!21:32
jgriffithhemna: That would be crazy to put all of those DB calls and checks in volume/api IMO21:32
hemnashould be in api.py ?21:32
hemnaoh ok, kinda like delete()21:32
*** harlowja is now known as harlowja_away21:32
hemnahas lots of db calls :P21:33
jgriffithI'm saying that the level of complexity in manager should STAY in manager and NOT be in api.py21:33
jgriffithapi.py should be quick/dirty and simple21:33
thingeejgriffith, hemna: I've always said that validation checks should be done in the underlying api.py file. The files in cinder.api should be what's just communicating back.21:33
jgriffiththingee: what?21:33
*** emagana has quit IRC21:33
thingeethat way if other methods or modules call that method, they benefit from those validation checks too.21:34
jgriffiththingee: don't start that one again please :)21:34
hemnathingee, I tend to agree.  it helps prevent rpc calls that would eventually fail21:34
jgriffithI already did the whole argument on why it should no tbe in cinder.api.contrib.xxxxxx21:34
thingeeI want to keep the cinder.api stuff purely for routing and communicating.21:34
jgriffiththingee: that's EXACTLY what I was saying this morning!!!21:34
thingeejgriffith: and I agree21:34
hemnacoolsvap|afk, we agree :)21:35
thingeejgriffith: right, just saying my position on the whole thing21:35
hemnasmh xchat21:35
jgriffiththingee: got ya21:35
hemnaso they should live in cinder.volume.api21:35
hemnaright ?21:35
jgriffiththingee: the problem is you're going to confuse hemna again and he'll say I "changed my mind" :(21:35
jgriffithhemna: I'll write the patch21:35
thingeejgriffith: I thought I was agreeing with you.21:35
thingeejgriffith: sorry if it came off as confusing hemna21:36
openstackgerritWalter A. Boring IV (hemna) proposed openstack/cinder: Be safe with getting attachment  https://review.openstack.org/16704521:36
*** IanGovett1 has joined #openstack-cinder21:36
hemnajgriffith, ^^21:36
*** annashen has quit IRC21:36
* hemna hopes that's what jgriffith meant.21:37
jgriffithhemna: that's EXACTLY what I meant!!!21:39
*** IanGovett has quit IRC21:39
jgriffithhemna: thank you!!!!!21:39
*** thangp has quit IRC21:39
jgriffithhemna: tell me about it :)21:39
openstackgerritVilobh Meshram proposed openstack/cinder: WIP - ConsistencyGroup Object  https://review.openstack.org/16364121:39
jgriffithon that note, I gotta jet for a bit, catch ya all later21:40
thingeejgriffith: seeya. thanks for the log patch21:40
*** asselin_ has joined #openstack-cinder21:42
*** tbarron has quit IRC21:44
*** asselin_ has quit IRC21:47
*** tbarron has joined #openstack-cinder21:47
*** agentle has quit IRC21:48
*** mdbooth has quit IRC21:48
*** wolsen has quit IRC21:48
*** flip214 has quit IRC21:48
*** tristanC has quit IRC21:48
*** angela-s has joined #openstack-cinder21:48
*** agentle has joined #openstack-cinder21:49
*** mdbooth has joined #openstack-cinder21:49
*** wolsen has joined #openstack-cinder21:49
*** flip214 has joined #openstack-cinder21:49
*** tristanC has joined #openstack-cinder21:49
*** sc has quit IRC21:50
*** agentle has quit IRC21:50
*** sc has joined #openstack-cinder21:52
*** mriedem is now known as mriedem_away21:53
*** emagana has joined #openstack-cinder21:53
*** primechuck has quit IRC21:54
*** dflorea has quit IRC21:57
*** ronis has quit IRC21:57
*** thingee has quit IRC22:00
*** tbarron has quit IRC22:00
*** tbarron has joined #openstack-cinder22:00
*** ronis has joined #openstack-cinder22:00
*** dflorea_ has joined #openstack-cinder22:00
*** akerr has joined #openstack-cinder22:03
*** dflorea_ has quit IRC22:05
*** harlowja_away is now known as harlowja22:05
openstackgerritRichard Hedlind proposed openstack/cinder: Properly remove host object from ISE  https://review.openstack.org/16607922:06
*** akerr has quit IRC22:10
*** dflorea has joined #openstack-cinder22:11
*** ociuhandu has quit IRC22:13
*** thingee has joined #openstack-cinder22:14
*** changbl has quit IRC22:14
*** vilobhmm11 has quit IRC22:15
*** vilobhmm1 has joined #openstack-cinder22:15
*** ronis has quit IRC22:16
*** vilobhmm1 has quit IRC22:20
*** vilobhmm1 has joined #openstack-cinder22:21
*** tshefi has quit IRC22:24
*** e0ne_ has quit IRC22:25
anishwhy does zuul show some tests in red even when they succeed ? is it supposed to indicate slow sections ?22:25
SwansonI've never seen that.22:27
anishhttp://status.openstack.org/zuul/ 158926/3 in gate22:28
*** emagana has quit IRC22:29
*** dflorea has quit IRC22:30
SwansonThat's solid green here.22:30
anishhrm gate-dsvm-neutron-full is seen as red for me even though it says SUCCESS22:30
*** dflorea has joined #openstack-cinder22:30
*** Rockyg has quit IRC22:30
SwansonAll green.  Unless you clicked that link in which case the link goes red.22:32
hemnaFunny for the day -- http://goo.gl/1XtzIS22:32
openstackgerritJulie Gravel proposed openstack/cinder: Validate volume type extra specs  https://review.openstack.org/16148022:33
*** dflorea has quit IRC22:35
*** dflorea has joined #openstack-cinder22:37
vilobhmm1thangp : ping22:46
openstackgerritJulie Gravel proposed openstack/cinder: Add HP3PAR validate volume type extra spec  https://review.openstack.org/16423822:50
*** dflorea has quit IRC22:51
*** dflorea has joined #openstack-cinder22:51
*** dustins has quit IRC22:54
*** dflorea has quit IRC22:56
*** rmesta has quit IRC22:56
tbarronhemna: that's spot on22:57
hemnatbarron, :)22:57
*** vilobhmm1 has quit IRC22:58
*** vilobhmm1 has joined #openstack-cinder22:58
*** vilobhmm1 has quit IRC22:59
*** vilobhmm1 has joined #openstack-cinder22:59
*** dannywilson has quit IRC22:59
*** dflorea has joined #openstack-cinder23:01
*** vilobhmm11 has joined #openstack-cinder23:02
*** vilobhmm11 has quit IRC23:02
*** vilobhmm11 has joined #openstack-cinder23:02
*** vilobhmm1 has quit IRC23:03
*** markvoelker has quit IRC23:06
*** dims has quit IRC23:14
openstackgerritJoe Gordon proposed openstack/cinder: Move to hacking 0.10  https://review.openstack.org/16743823:16
*** dims has joined #openstack-cinder23:16
*** dims has quit IRC23:17
*** dims has joined #openstack-cinder23:17
*** chlong has joined #openstack-cinder23:20
*** btran has quit IRC23:20
*** vilobhmm11 has quit IRC23:21
*** vilobhmm1 has joined #openstack-cinder23:22
*** vilobhmm1 has quit IRC23:22
*** Mandell has quit IRC23:22
*** vilobhmm1 has joined #openstack-cinder23:22
*** vilobhmm1 has quit IRC23:23
*** vilobhmm1 has joined #openstack-cinder23:23
*** vilobhmm1 has quit IRC23:23
*** vilobhmm1 has joined #openstack-cinder23:24
*** ociuhandu has joined #openstack-cinder23:25
*** vilobhmm11 has joined #openstack-cinder23:27
*** vilobhmm1 has quit IRC23:27
*** dflorea has quit IRC23:32
*** dflorea has joined #openstack-cinder23:33
*** annashen has joined #openstack-cinder23:34
*** chlong has quit IRC23:35
*** chlong has joined #openstack-cinder23:35
*** dflorea has quit IRC23:37
*** dflorea has joined #openstack-cinder23:37
*** annashen has quit IRC23:38
*** markvoelker has joined #openstack-cinder23:47
*** diemt has quit IRC23:47
*** ho has joined #openstack-cinder23:49
*** markvoelker has quit IRC23:52

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