| *** drjones has joined #openstack-cinder | 00:00 | |
| *** _cjones_ has quit IRC | 00:01 | |
| *** patrickeast_ has joined #openstack-cinder | 00:02 | |
| *** patrickeast has quit IRC | 00:03 | |
| *** patrickeast_ is now known as patrickeast | 00:03 | |
| *** ebalduf has joined #openstack-cinder | 00:03 | |
| thingee | xyang1: I'm working on this issue now https://bugs.launchpad.net/cinder/+bug/1413427 ... didn't see a bug already reported from our earlier conversation | 00:04 |
|---|---|---|
| *** _cjones_ has joined #openstack-cinder | 00:05 | |
| *** drjones has quit IRC | 00:05 | |
| *** ebalduf has quit IRC | 00:05 | |
| *** ho has joined #openstack-cinder | 00:07 | |
| xyang1 | thingee: So you'll fix it? I did not open a bug yet. Thanks! | 00:08 |
| thingee | xyang1: yup almost done | 00:09 |
| thingee | I'm just checking for consistencygroup_id and raising invalid volume | 00:09 |
| xyang1 | thingee: Cool! | 00:09 |
| xyang1 | Ya, that should be enough | 00:09 |
| *** rwsu has quit IRC | 00:10 | |
| *** alexpilotti has quit IRC | 00:11 | |
| *** annashen has quit IRC | 00:14 | |
| *** lcurtis has quit IRC | 00:19 | |
| thingee | jgriffith: LOL if you have to update a unit test with its own fixture of volume. | 00:20 |
| thingee | jgriffith: I just got bit by that | 00:20 |
| *** rwsu has joined #openstack-cinder | 00:21 | |
| *** rilango has joined #openstack-cinder | 00:23 | |
| *** haomaiwa_ has joined #openstack-cinder | 00:26 | |
| *** k69 has joined #openstack-cinder | 00:27 | |
| k69 | Hi, would somebody please help me with this "cinder list" commands ? https://ask.openstack.org/en/question/58869/cinder-list-command-output-status-section-remains-creating-rather-than-availabe-and-stucks/ | 00:28 |
| *** smoriya has joined #openstack-cinder | 00:29 | |
| *** tellesnobrega_ has joined #openstack-cinder | 00:31 | |
| openstackgerrit | Mike Perez proposed openstack/cinder: Prevent deleting volumes in a consistency group https://review.openstack.org/149095 | 00:34 |
| thingee | xyang1: ^ | 00:34 |
| thingee | jgriffith: test_volume has some really odd unit tests. | 00:35 |
| xyang1 | thingee: Docstring needs to be updated | 00:39 |
| xyang1 | thingee: Looks good. Thanks for fixing this! | 00:39 |
| thingee | xyang1: k | 00:39 |
| boris-42 | jgriffith: lol | 00:39 |
| boris-42 | jgriffith: seems like running 50 benchmarks was not a good idea | 00:40 |
| boris-42 | jgriffith: rally gates are limited by 2 hrs so not sure that it will have enough time.. | 00:40 |
| *** sgotliv has quit IRC | 00:44 | |
| openstackgerrit | Mike Perez proposed openstack/cinder: Prevent deleting volumes in a consistency group https://review.openstack.org/149095 | 00:45 |
| *** kurtmartin has quit IRC | 00:45 | |
| *** fischerw has quit IRC | 00:45 | |
| *** kurtmartin has joined #openstack-cinder | 00:52 | |
| *** changbl has joined #openstack-cinder | 00:56 | |
| *** nosleep77 has quit IRC | 00:59 | |
| openstackgerrit | Joshua Harlow proposed openstack/cinder: Shrink down customized logging listener https://review.openstack.org/149102 | 01:02 |
| *** hemna is now known as hemnafk | 01:03 | |
| *** jasondotstar has joined #openstack-cinder | 01:07 | |
| *** rushiagr_away has quit IRC | 01:13 | |
| *** theanalyst has quit IRC | 01:13 | |
| *** marcusvrn has quit IRC | 01:14 | |
| *** rushiagr_away has joined #openstack-cinder | 01:14 | |
| *** marcusvrn has joined #openstack-cinder | 01:15 | |
| boris-42 | jgriffith: here? | 01:15 |
| *** tsekiyam_ has joined #openstack-cinder | 01:15 | |
| boris-42 | jgriffith: http://logs.openstack.org/77/149077/1/check/gate-rally-dsvm-cinder/b251fb5/rally-plot/results.html.gz#/ | 01:18 |
| *** lpabon has quit IRC | 01:18 | |
| boris-42 | jgriffith: seems like attach works better | 01:18 |
| *** tsekiyama has quit IRC | 01:19 | |
| *** mtanino has quit IRC | 01:19 | |
| *** tsekiyam_ has quit IRC | 01:20 | |
| *** k69 has quit IRC | 01:21 | |
| *** k69 has joined #openstack-cinder | 01:22 | |
| *** k69 has quit IRC | 01:22 | |
| *** k69 has joined #openstack-cinder | 01:22 | |
| k69 | hi, could this cause problem for cinder compatibility ? http://paste.openstack.org/show/160008/ | 01:23 |
| *** emagana has quit IRC | 01:26 | |
| boris-42 | jgriffith: Not going to have enough space on the instance for this distribution of tests ?) | 01:26 |
| *** emagana has joined #openstack-cinder | 01:26 | |
| *** tellesnobrega_ has quit IRC | 01:29 | |
| *** ryanpetrello has joined #openstack-cinder | 01:29 | |
| *** IanGovett has joined #openstack-cinder | 01:31 | |
| *** emagana has quit IRC | 01:31 | |
| *** IanGovett has quit IRC | 01:32 | |
| *** Mandell has quit IRC | 01:32 | |
| *** ryanpetrello has quit IRC | 01:35 | |
| *** Mandell has joined #openstack-cinder | 01:37 | |
| *** scottda_ has joined #openstack-cinder | 01:38 | |
| *** ryanpetrello has joined #openstack-cinder | 01:38 | |
| openstackgerrit | Danny Al-Gaaf proposed openstack/cinder: fix typo in config.py https://review.openstack.org/149113 | 01:39 |
| *** esker has joined #openstack-cinder | 01:39 | |
| *** scottda_ has quit IRC | 01:39 | |
| *** scottda_ has joined #openstack-cinder | 01:41 | |
| *** kurtmartin has quit IRC | 01:41 | |
| *** scottda_ has quit IRC | 01:42 | |
| *** esker has quit IRC | 01:44 | |
| *** leeantho has quit IRC | 01:44 | |
| thingee | k69: I think the only issue will be consistency groups not working. Since the node "block1" running your volume service won't be able to understand the request. | 01:46 |
| thingee | k69: don't see any other changes with rpc calls or database model changes. | 01:47 |
| *** ryanpetrello has quit IRC | 01:47 | |
| thingee | oh and block1 won't be able to understand volume replication requests. | 01:48 |
| *** takedakn has joined #openstack-cinder | 01:48 | |
| k69 | thingee, thanks for the reply, well i did all the official manual chapter 8 said (http://docs.openstack.org/juno/install-guide/install/apt/content/ch_cinder.html), and when i use vgdisplay and lvdisplay commands, "cinder-volumes" are displayed | 01:50 |
| k69 | thingee, though i have not set cloud repositories for the block1 node, might it make the problem u mentioned? | 01:51 |
| k69 | http://paste.openstack.org/show/160008/ | 01:51 |
| *** drjones has joined #openstack-cinder | 01:55 | |
| thingee | k69: yeah, if you don't setup the cloud repositories on block1, it'll fetch from another repo that has outdated packages. | 01:55 |
| thingee | k69: http://docs.openstack.org/juno/install-guide/install/apt/content/ch_basic_environment.html | 01:55 |
| *** _cjones_ has quit IRC | 01:55 | |
| *** ryanpetrello has joined #openstack-cinder | 01:55 | |
| *** drjones has quit IRC | 01:56 | |
| *** _cjones_ has joined #openstack-cinder | 01:56 | |
| k69 | thingee, ty...ill try that :) | 01:57 |
| *** leseb- has joined #openstack-cinder | 01:57 | |
| *** ho__ has joined #openstack-cinder | 01:58 | |
| thingee | k69: you're welcome! | 01:58 |
| *** tristanC_ has joined #openstack-cinder | 01:58 | |
| *** kragniz_ has joined #openstack-cinder | 01:59 | |
| *** julim_ has joined #openstack-cinder | 01:59 | |
| *** drjones has joined #openstack-cinder | 01:59 | |
| *** _cjones_ has quit IRC | 02:00 | |
| *** Mandell has quit IRC | 02:00 | |
| *** drjones has quit IRC | 02:01 | |
| *** _cjones_ has joined #openstack-cinder | 02:02 | |
| *** lan_ has quit IRC | 02:03 | |
| *** timbyr_ has joined #openstack-cinder | 02:05 | |
| *** mberlin has joined #openstack-cinder | 02:05 | |
| *** leseb has quit IRC | 02:06 | |
| *** ho has quit IRC | 02:06 | |
| *** tristanC has quit IRC | 02:06 | |
| *** mberlin1 has quit IRC | 02:06 | |
| *** kragniz has quit IRC | 02:06 | |
| *** vukcrni has quit IRC | 02:06 | |
| *** julim has quit IRC | 02:06 | |
| *** kbyrne has quit IRC | 02:06 | |
| *** takedakn has quit IRC | 02:06 | |
| *** vukcrni has joined #openstack-cinder | 02:06 | |
| *** _cjones_ has quit IRC | 02:07 | |
| *** drjones has joined #openstack-cinder | 02:07 | |
| thingee | smcginnis: reply to version discovery: https://review.openstack.org/#/c/145613/ | 02:07 |
| *** lan has joined #openstack-cinder | 02:07 | |
| *** mudassirlatif has quit IRC | 02:07 | |
| *** ryanpetrello has quit IRC | 02:08 | |
| thingee | smcginnis: it's hacks all the way down unfortunately. I'm going to try what jamielennox has suggested, but otherwise those are my reasons :). | 02:09 |
| *** _cjones_ has joined #openstack-cinder | 02:09 | |
| *** drjones has quit IRC | 02:10 | |
| thingee | winston-d_zzZ: ^ | 02:11 |
| *** annashen has joined #openstack-cinder | 02:12 | |
| *** drjones has joined #openstack-cinder | 02:12 | |
| *** _cjones_ has quit IRC | 02:12 | |
| *** drjones has quit IRC | 02:15 | |
| *** ryanpetrello has joined #openstack-cinder | 02:15 | |
| *** _cjones_ has joined #openstack-cinder | 02:15 | |
| *** _cjones_ has quit IRC | 02:17 | |
| *** annashen_ has joined #openstack-cinder | 02:17 | |
| *** _cjones_ has joined #openstack-cinder | 02:17 | |
| *** annashen has quit IRC | 02:17 | |
| *** annashen_ has quit IRC | 02:18 | |
| *** annashen has joined #openstack-cinder | 02:19 | |
| *** tellesnobrega_ has joined #openstack-cinder | 02:20 | |
| *** drjones has joined #openstack-cinder | 02:20 | |
| *** drjones has quit IRC | 02:20 | |
| *** _cjones_ has quit IRC | 02:20 | |
| *** _cjones_ has joined #openstack-cinder | 02:20 | |
| *** ryanpetrello has quit IRC | 02:23 | |
| *** annashen has quit IRC | 02:23 | |
| *** _cjones_ has quit IRC | 02:25 | |
| *** kaisers1 has joined #openstack-cinder | 02:25 | |
| *** mberlin1 has joined #openstack-cinder | 02:25 | |
| *** kaisers has quit IRC | 02:26 | |
| *** mberlin has quit IRC | 02:27 | |
| *** scottda_ has joined #openstack-cinder | 02:33 | |
| *** lifeless has quit IRC | 02:41 | |
| openstackgerrit | wuyuting proposed openstack/cinder: Fix bug in rbd driver: the cloned volume size is wrong https://review.openstack.org/148185 | 02:41 |
| *** lifeless has joined #openstack-cinder | 02:42 | |
| *** lcurtis has joined #openstack-cinder | 02:43 | |
| *** scottda_ has quit IRC | 02:44 | |
| *** scottda_ has joined #openstack-cinder | 02:46 | |
| *** scottda_ has quit IRC | 02:47 | |
| *** ryanpetrello has joined #openstack-cinder | 02:47 | |
| *** tbarron has quit IRC | 02:48 | |
| *** tbarron has joined #openstack-cinder | 02:52 | |
| *** lpabon has joined #openstack-cinder | 02:55 | |
| *** lpabon has quit IRC | 02:55 | |
| *** tbarron has quit IRC | 02:57 | |
| *** ryanpetrello has quit IRC | 02:57 | |
| *** esker has joined #openstack-cinder | 03:04 | |
| openstackgerrit | Rajini Ram proposed openstack/cinder: Fix the eqlx driver to retry on ssh timeout https://review.openstack.org/148715 | 03:07 |
| *** rwsu has quit IRC | 03:10 | |
| openstackgerrit | Rajini Ram proposed openstack/cinder: Fix the eqlx driver to retry on ssh timeout https://review.openstack.org/148715 | 03:10 |
| *** BharatK has joined #openstack-cinder | 03:15 | |
| *** ks_ne has joined #openstack-cinder | 03:16 | |
| *** mtanino has joined #openstack-cinder | 03:17 | |
| *** ks_ne has left #openstack-cinder | 03:18 | |
| *** bill_az has quit IRC | 03:18 | |
| *** scottda_ has joined #openstack-cinder | 03:22 | |
| *** rongze has joined #openstack-cinder | 03:26 | |
| *** fandi has quit IRC | 03:28 | |
| *** bkopilov has quit IRC | 03:32 | |
| *** k69 has quit IRC | 03:37 | |
| *** Longgeek has joined #openstack-cinder | 03:41 | |
| *** tsekiyama has joined #openstack-cinder | 03:42 | |
| *** tsekiyama has quit IRC | 03:42 | |
| *** jasondotstar has quit IRC | 03:47 | |
| *** zhiyan has quit IRC | 03:48 | |
| *** rmesta has quit IRC | 03:48 | |
| *** zhiyan has joined #openstack-cinder | 03:49 | |
| *** Mandell has joined #openstack-cinder | 03:51 | |
| *** Mandell has quit IRC | 03:53 | |
| *** annashen has joined #openstack-cinder | 03:55 | |
| *** patrickeast has quit IRC | 03:57 | |
| *** tellesnobrega_ has quit IRC | 04:01 | |
| *** xyang1 has quit IRC | 04:03 | |
| *** eharney has quit IRC | 04:05 | |
| *** scottda_ has quit IRC | 04:11 | |
| *** dalgaaf has quit IRC | 04:13 | |
| *** eharney has joined #openstack-cinder | 04:21 | |
| *** e0ne has joined #openstack-cinder | 04:24 | |
| *** annashen has quit IRC | 04:29 | |
| *** annashen has joined #openstack-cinder | 04:29 | |
| *** Longgeek has quit IRC | 04:34 | |
| *** annashen has quit IRC | 04:34 | |
| *** Longgeek has joined #openstack-cinder | 04:34 | |
| *** bkopilov has joined #openstack-cinder | 04:34 | |
| *** bkopilov has quit IRC | 04:39 | |
| *** changbl has quit IRC | 04:44 | |
| openstackgerrit | Li Min Liu proposed openstack/cinder: Add secondary sshpool for san storage drivers https://review.openstack.org/146363 | 04:47 |
| *** harlowja is now known as harlowja_away | 04:51 | |
| *** Longgeek_ has joined #openstack-cinder | 04:53 | |
| *** Longgeek_ has quit IRC | 04:56 | |
| *** Longgeek has quit IRC | 04:56 | |
| *** Longgeek has joined #openstack-cinder | 04:56 | |
| openstackgerrit | Li Min Liu proposed openstack/cinder: Add secondary sshpool for san storage drivers https://review.openstack.org/146363 | 04:56 |
| *** Longgeek_ has joined #openstack-cinder | 04:59 | |
| *** Longgeek has quit IRC | 04:59 | |
| *** Longgeek has joined #openstack-cinder | 05:00 | |
| *** Longgeek has quit IRC | 05:01 | |
| *** Longgeek has joined #openstack-cinder | 05:02 | |
| *** Longgeek_ has quit IRC | 05:04 | |
| *** Longgeek has quit IRC | 05:12 | |
| *** Longgeek has joined #openstack-cinder | 05:12 | |
| *** eharney has quit IRC | 05:13 | |
| *** Longgeek has quit IRC | 05:16 | |
| *** esker has quit IRC | 05:17 | |
| *** _cjones_ has joined #openstack-cinder | 05:20 | |
| *** _cjones_ has quit IRC | 05:25 | |
| openstackgerrit | David J Hu proposed openstack/cinder: Support cinder_admin in policy.json https://review.openstack.org/149148 | 05:26 |
| *** RJ has joined #openstack-cinder | 05:30 | |
| RJ | Hi | 05:31 |
| RJ | if i want to add new variable in cinder.conf file and want to use it ..?? is it possible for me ..? | 05:31 |
| *** sgotliv has joined #openstack-cinder | 05:32 | |
| *** e0ne has quit IRC | 05:33 | |
| *** Longgeek has joined #openstack-cinder | 05:35 | |
| RJ | Longgeek: if i want to add new variable in cinder.conf file and want to use it ..?? is it possible for me ..? | 05:37 |
| *** Mandell has joined #openstack-cinder | 05:40 | |
| *** Longgeek has quit IRC | 05:40 | |
| *** e0ne has joined #openstack-cinder | 05:43 | |
| *** nuritv has quit IRC | 05:46 | |
| *** harlowja_at_home has joined #openstack-cinder | 05:47 | |
| e0ne | RJ: sure. there are a lot of samples in the code: https://github.com/openstack/cinder/search?utf8=%E2%9C%93&q=register_opts | 05:49 |
| *** Longgeek has joined #openstack-cinder | 05:50 | |
| *** sgotliv has quit IRC | 05:51 | |
| *** mudassirlatif has joined #openstack-cinder | 05:51 | |
| RJ | e0ne: got that now | 05:52 |
| *** BharatK has quit IRC | 05:54 | |
| *** lcurtis has quit IRC | 05:55 | |
| *** e0ne has quit IRC | 06:00 | |
| *** Longgeek has quit IRC | 06:01 | |
| openstackgerrit | Tiantian Gao proposed openstack/cinder: Fix cinder-manage shell ipython https://review.openstack.org/149155 | 06:03 |
| *** e0ne has joined #openstack-cinder | 06:08 | |
| *** Longgeek has joined #openstack-cinder | 06:09 | |
| anish | e0ne: holy hell, did not know you could search github like that | 06:10 |
| anish | goodbye pycscope ! | 06:10 |
| e0ne | anish: i need to look on pycscope:) | 06:11 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/cinder: Imported Translations from Transifex https://review.openstack.org/148044 | 06:12 |
| *** BharatK has joined #openstack-cinder | 06:12 | |
| *** BharatK has quit IRC | 06:16 | |
| *** harlowja_at_home has quit IRC | 06:18 | |
| *** davidb_ has joined #openstack-cinder | 06:25 | |
| *** Longgeek has quit IRC | 06:28 | |
| *** Longgeek has joined #openstack-cinder | 06:28 | |
| openstackgerrit | Dave Chen proposed openstack/cinder: set/unset volume image metadata https://review.openstack.org/147726 | 06:29 |
| openstackgerrit | Dave Chen proposed openstack/cinder: set/unset snapshot image metadata https://review.openstack.org/149163 | 06:29 |
| *** dannywilson has joined #openstack-cinder | 06:34 | |
| *** Longgeek has quit IRC | 06:38 | |
| *** dannywilson has quit IRC | 06:41 | |
| *** dannywilson has joined #openstack-cinder | 06:42 | |
| *** nikesh_vedams has quit IRC | 06:43 | |
| *** krishna_vedams has quit IRC | 06:43 | |
| *** lpetrut has joined #openstack-cinder | 06:45 | |
| *** dannywilson has quit IRC | 06:46 | |
| *** coolsvap|afk is now known as coolsvap | 06:46 | |
| *** davidb_ has quit IRC | 06:48 | |
| *** BharatK has joined #openstack-cinder | 06:48 | |
| *** davidb_ has joined #openstack-cinder | 06:50 | |
| *** e0ne has quit IRC | 06:51 | |
| *** nlevinki has joined #openstack-cinder | 06:53 | |
| *** tshefi has joined #openstack-cinder | 06:55 | |
| kaisers1 | jgriffith: Hi! still online? | 06:58 |
| *** anshul has joined #openstack-cinder | 06:58 | |
| kaisers1 | Das anybody here have experience with jgriffiths sos-ci? | 06:59 |
| openstackgerrit | Vincent Hou proposed openstack/cinder-specs: Enhance list operations with the additional keys and next link https://review.openstack.org/148181 | 07:01 |
| openstackgerrit | Vincent Hou proposed openstack/cinder-specs: Enhance list operations with the additional keys and next link https://review.openstack.org/148181 | 07:05 |
| openstackgerrit | Vincent Hou proposed openstack/cinder-specs: Enhance list operations with the additional keys and next link https://review.openstack.org/148181 | 07:08 |
| openstackgerrit | Vipin Balachandran proposed openstack/cinder: VMware: Support for non-default port https://review.openstack.org/146384 | 07:08 |
| *** thingee has quit IRC | 07:10 | |
| *** lpetrut has quit IRC | 07:19 | |
| *** davidb_ has quit IRC | 07:19 | |
| *** davidb_ has joined #openstack-cinder | 07:22 | |
| *** nshaikh has joined #openstack-cinder | 07:26 | |
| *** tshefi has quit IRC | 07:29 | |
| *** avishay has joined #openstack-cinder | 07:31 | |
| *** davidb_ has quit IRC | 07:34 | |
| openstackgerrit | Eduard Biceri-Matei proposed openstack/cinder: New Cinder volume driver for openvstorage https://review.openstack.org/130733 | 07:37 |
| *** Longgeek has joined #openstack-cinder | 07:37 | |
| *** chlong has quit IRC | 07:38 | |
| openstackgerrit | Eduard Biceri-Matei proposed openstack/cinder: New Cinder volume driver for openvstorage https://review.openstack.org/130733 | 07:38 |
| openstackgerrit | Dave Chen proposed openstack/python-cinderclient: Add commands for modifying image metadata https://review.openstack.org/147077 | 07:38 |
| *** e0ne has joined #openstack-cinder | 07:39 | |
| *** tshefi has joined #openstack-cinder | 07:44 | |
| *** sgotliv has joined #openstack-cinder | 07:49 | |
| anish | kaisers1: there was someone working with it, but he's not online right now | 07:51 |
| *** davidb_ has joined #openstack-cinder | 07:51 | |
| *** TobiasE has joined #openstack-cinder | 07:57 | |
| *** davidb_ has quit IRC | 08:01 | |
| *** jamielennox is now known as jamielennox|away | 08:01 | |
| *** davidb_ has joined #openstack-cinder | 08:02 | |
| *** lpetrut has joined #openstack-cinder | 08:03 | |
| *** Longgeek has quit IRC | 08:04 | |
| *** takedakn has joined #openstack-cinder | 08:18 | |
| *** takedakn has quit IRC | 08:19 | |
| *** lpetrut has quit IRC | 08:20 | |
| *** mudassirlatif has quit IRC | 08:20 | |
| *** dulek has joined #openstack-cinder | 08:21 | |
| openstackgerrit | Silvan Kaiser proposed openstack/cinder: Update tests for Quobyte Cinder drv from mox->mock. https://review.openstack.org/149181 | 08:23 |
| kaisers1 | anish: ah, ok. Will go hunting for him... ;-) | 08:23 |
| kaisers1 | anish: It's just a config parameter named "key_name" i'm not sure what key is meant. | 08:24 |
| *** mudassirlatif has joined #openstack-cinder | 08:25 | |
| kaisers1 | I'd like to get rid of old mox code, blueprint: https://blueprints.launchpad.net/cinder/+spec/qb-driver-mox2mock and patch: https://review.openstack.org/#/c/149181/ | 08:26 |
| kaisers1 | If anyone would like to review... :) | 08:26 |
| *** Longgeek has joined #openstack-cinder | 08:28 | |
| *** oomichi_ has quit IRC | 08:40 | |
| openstackgerrit | Silvan Kaiser proposed openstack/cinder: Update tests for Quobyte Cinder drv from mox->mock. https://review.openstack.org/149181 | 08:40 |
| *** Longgeek has quit IRC | 08:42 | |
| *** Longgeek_ has joined #openstack-cinder | 08:44 | |
| *** e0ne is now known as e0ne_ | 08:44 | |
| openstackgerrit | Eduard Biceri-Matei proposed openstack/cinder: New Cinder volume driver for openvstorage https://review.openstack.org/130733 | 08:49 |
| *** ameade_ has quit IRC | 08:52 | |
| *** davidb_ has quit IRC | 08:54 | |
| *** e0ne_ is now known as e0ne | 08:54 | |
| *** zhiyan has quit IRC | 08:55 | |
| openstackgerrit | Yasuaki Nagata proposed openstack/cinder: Change volume and snapshot stuck creating to error https://review.openstack.org/149188 | 08:58 |
| *** ameade_ has joined #openstack-cinder | 09:01 | |
| *** zhiyan has joined #openstack-cinder | 09:02 | |
| *** jpich has joined #openstack-cinder | 09:11 | |
| openstackgerrit | Vipin Balachandran proposed openstack/cinder: VMware: Integrate VMDK driver with oslo.vmware https://review.openstack.org/137324 | 09:15 |
| *** mudassirlatif has quit IRC | 09:16 | |
| openstackgerrit | Vipin Balachandran proposed openstack/cinder: VMware: Support for non-default port https://review.openstack.org/146384 | 09:17 |
| *** lpetrut has joined #openstack-cinder | 09:19 | |
| *** jordanP has joined #openstack-cinder | 09:21 | |
| *** jistr has joined #openstack-cinder | 09:24 | |
| *** mudassirlatif has joined #openstack-cinder | 09:25 | |
| *** Mandell has quit IRC | 09:27 | |
| *** nshaikh has quit IRC | 09:30 | |
| *** afazekas has joined #openstack-cinder | 09:47 | |
| openstackgerrit | Silvan Kaiser proposed openstack/cinder: Update tests for Quobyte Cinder drv from mox->mock. https://review.openstack.org/149181 | 09:47 |
| *** Miouge_ has joined #openstack-cinder | 09:48 | |
| *** Miouge has quit IRC | 09:49 | |
| *** Miouge_ is now known as Miouge | 09:49 | |
| *** tbarron has joined #openstack-cinder | 09:49 | |
| *** nellysmitt has joined #openstack-cinder | 09:50 | |
| *** tbarron has quit IRC | 09:51 | |
| *** bkopilov has joined #openstack-cinder | 09:57 | |
| *** haomaiwa_ has quit IRC | 10:01 | |
| *** fandi has joined #openstack-cinder | 10:02 | |
| *** e0ne is now known as e0ne_ | 10:04 | |
| *** e0ne_ is now known as e0ne | 10:04 | |
| *** RJ has left #openstack-cinder | 10:05 | |
| *** tbarron has joined #openstack-cinder | 10:05 | |
| openstackgerrit | Vipin Balachandran proposed openstack/cinder: VMware:Fix error creating vCenter inventory folder https://review.openstack.org/149202 | 10:09 |
| *** mudassirlatif has quit IRC | 10:09 | |
| openstackgerrit | Vipin Balachandran proposed openstack/cinder: VMware: Integrate VMDK driver with oslo.vmware https://review.openstack.org/137324 | 10:13 |
| openstackgerrit | Vipin Balachandran proposed openstack/cinder: VMware: Integrate VMDK driver with oslo.vmware https://review.openstack.org/137324 | 10:13 |
| openstackgerrit | Vipin Balachandran proposed openstack/cinder: VMware: Support for non-default port https://review.openstack.org/146384 | 10:15 |
| *** chlong has joined #openstack-cinder | 10:15 | |
| *** alexpilotti has joined #openstack-cinder | 10:19 | |
| *** aix has joined #openstack-cinder | 10:20 | |
| *** chlong has quit IRC | 10:20 | |
| openstackgerrit | Silvan Kaiser proposed openstack/cinder: Update tests for Quobyte Cinder drv from mox->mock. https://review.openstack.org/149181 | 10:23 |
| *** jasondotstar has joined #openstack-cinder | 10:30 | |
| *** jasondotstar has quit IRC | 10:30 | |
| *** BharatK has quit IRC | 10:31 | |
| *** Longgeek_ has quit IRC | 10:32 | |
| *** chlong has joined #openstack-cinder | 10:33 | |
| *** tbarron has quit IRC | 10:36 | |
| *** tbarron has joined #openstack-cinder | 10:37 | |
| *** dalgaaf has joined #openstack-cinder | 10:37 | |
| *** chlong has quit IRC | 10:37 | |
| *** nikesh_vedams has joined #openstack-cinder | 10:40 | |
| *** BharatK has joined #openstack-cinder | 10:41 | |
| *** EduardBiceriMate has joined #openstack-cinder | 10:42 | |
| *** rongze has quit IRC | 10:46 | |
| *** e0ne is now known as e0ne_ | 10:47 | |
| *** chlong has joined #openstack-cinder | 10:50 | |
| *** nikesh_vedams has quit IRC | 10:50 | |
| *** nikesh_vedams has joined #openstack-cinder | 10:51 | |
| nikesh_vedams | hi | 10:51 |
| nikesh_vedams | stack.sh is failing in kilo devstack | 10:51 |
| nikesh_vedams | 2015-01-22 10:44:50.838 | ERROR: The service catalog is empty. | 10:51 |
| *** e0ne_ has quit IRC | 10:57 | |
| *** TobiasE1 has joined #openstack-cinder | 10:59 | |
| *** TobiasE has quit IRC | 11:01 | |
| flip214 | nikesh_vedams: look at the mailing list, this is just being discussed. | 11:03 |
| *** e0ne has joined #openstack-cinder | 11:06 | |
| openstackgerrit | TaoBai proposed openstack/cinder: IBM Storwize driver Consistency Group Implementation https://review.openstack.org/142683 | 11:08 |
| nikesh_vedams | flip214: do you have link for that mail | 11:14 |
| nikesh_vedams | flip214: got it | 11:15 |
| *** nshaikh has joined #openstack-cinder | 11:21 | |
| *** e0ne is now known as e0ne_ | 11:22 | |
| openstackgerrit | Michal Dulko proposed openstack/cinder: Fix deprecation warnings from TaskFlow 0.6 https://review.openstack.org/149217 | 11:30 |
| *** wpf1 has quit IRC | 11:31 | |
| *** e0ne_ has quit IRC | 11:33 | |
| *** BharatK has quit IRC | 11:36 | |
| *** aix has quit IRC | 11:38 | |
| *** nikesh_vedams has quit IRC | 11:38 | |
| *** rongze has joined #openstack-cinder | 11:40 | |
| *** wpf1 has joined #openstack-cinder | 11:44 | |
| *** kragniz_ is now known as kragniz | 11:45 | |
| *** BharatK has joined #openstack-cinder | 11:52 | |
| *** takedakn has joined #openstack-cinder | 11:52 | |
| *** takedakn has quit IRC | 11:52 | |
| *** BharatK has quit IRC | 11:57 | |
| *** TobiasE has joined #openstack-cinder | 11:58 | |
| *** TobiasE1 has quit IRC | 12:00 | |
| *** wolsen_ has quit IRC | 12:00 | |
| *** dosaboy has quit IRC | 12:02 | |
| *** dosaboy has joined #openstack-cinder | 12:04 | |
| *** wolsen has joined #openstack-cinder | 12:05 | |
| *** BharatK has joined #openstack-cinder | 12:10 | |
| *** e0ne has joined #openstack-cinder | 12:27 | |
| openstackgerrit | Boris Pavlovic proposed openstack/cinder: (do not merge) Stress a bit attach https://review.openstack.org/149077 | 12:28 |
| *** haomaiwang has joined #openstack-cinder | 12:28 | |
| *** timcl has joined #openstack-cinder | 12:28 | |
| *** haomai___ has joined #openstack-cinder | 12:32 | |
| *** Longgeek has joined #openstack-cinder | 12:33 | |
| *** haomaiwang has quit IRC | 12:35 | |
| *** e0ne is now known as e0ne_ | 12:38 | |
| *** e0ne_ is now known as e0ne | 12:40 | |
| dulek | Had you a problem like that with devstack: | 12:40 |
| dulek | pkg_resources.DistributionNotFound: cinder==2015.1.dev215 | 12:40 |
| dulek | Last time unstack.sh && stack.sh solved it, but maybe there's simple solution? | 12:41 |
| *** chlong has quit IRC | 12:44 | |
| *** IanGovett has joined #openstack-cinder | 12:44 | |
| *** nikesh_vedams has joined #openstack-cinder | 12:46 | |
| *** ndipanov has quit IRC | 12:53 | |
| *** ho__ has quit IRC | 12:54 | |
| *** juzuluag has joined #openstack-cinder | 12:56 | |
| *** ndipanov has joined #openstack-cinder | 12:59 | |
| *** takedakn has joined #openstack-cinder | 12:59 | |
| *** smoriya has quit IRC | 13:00 | |
| *** IanGovett has quit IRC | 13:03 | |
| *** takedakn has quit IRC | 13:03 | |
| openstackgerrit | Rajesh Tailor proposed openstack/cinder: Refactoring initialize connection api to use taskflow https://review.openstack.org/148920 | 13:04 |
| *** avishay has quit IRC | 13:08 | |
| *** tbarron1 has joined #openstack-cinder | 13:12 | |
| *** e0ne is now known as e0ne_ | 13:12 | |
| *** avishay has joined #openstack-cinder | 13:13 | |
| *** bswartz has quit IRC | 13:13 | |
| *** tshefi has quit IRC | 13:14 | |
| *** tshefi has joined #openstack-cinder | 13:14 | |
| *** tbarron has quit IRC | 13:15 | |
| *** tbarron1 has quit IRC | 13:16 | |
| *** avishay has quit IRC | 13:17 | |
| *** markvoelker has joined #openstack-cinder | 13:19 | |
| *** e0ne_ has quit IRC | 13:22 | |
| *** avishay has joined #openstack-cinder | 13:24 | |
| openstackgerrit | Rushi Agrawal proposed openstack/cinder: List quota defaults without passing tenant ID https://review.openstack.org/139610 | 13:24 |
| *** Mandell has joined #openstack-cinder | 13:28 | |
| *** jecarey has quit IRC | 13:30 | |
| *** Yogi1 has joined #openstack-cinder | 13:31 | |
| *** Mandell has quit IRC | 13:32 | |
| *** marcusvrn_ has joined #openstack-cinder | 13:33 | |
| openstackgerrit | Stefan Amann proposed openstack/cinder: Adjust Cinder to support FCP on System z systems https://review.openstack.org/149256 | 13:35 |
| *** bkopilov has quit IRC | 13:39 | |
| *** rushiagr_away is now known as rushiagr | 13:40 | |
| *** aix has joined #openstack-cinder | 13:40 | |
| *** eharney has joined #openstack-cinder | 13:40 | |
| *** mtanino has quit IRC | 13:42 | |
| *** amoturi has joined #openstack-cinder | 13:42 | |
| *** ndipanov has quit IRC | 13:43 | |
| *** avishay has quit IRC | 13:45 | |
| *** kaufer has joined #openstack-cinder | 13:45 | |
| *** adam_g is now known as adam_g_out | 13:46 | |
| *** tristanC_ is now known as tristanC | 13:47 | |
| flip214 | hemnafk: please see https://github.com/LINBIT/cinder/tree/drbd-for-nova for the mess re brick/connector/target, please tell me what I can remove. | 13:48 |
| *** akerr has joined #openstack-cinder | 13:48 | |
| *** nikesh_vedams has quit IRC | 13:50 | |
| openstackgerrit | Stefan Amann proposed openstack/cinder: Adjust Cinder to support FCP on System z systems https://review.openstack.org/149256 | 13:51 |
| *** bswartz has joined #openstack-cinder | 13:51 | |
| *** e0ne has joined #openstack-cinder | 13:53 | |
| *** Yogi11 has joined #openstack-cinder | 13:53 | |
| *** ndipanov has joined #openstack-cinder | 13:56 | |
| *** Yogi1 has quit IRC | 13:57 | |
| *** bswartz has quit IRC | 13:58 | |
| openstackgerrit | Ilya Tyaptin proposed openstack/cinder: Fix _usage_from_snapshot in volume.utils https://review.openstack.org/133031 | 13:59 |
| *** tbarron has joined #openstack-cinder | 14:02 | |
| *** fandi has quit IRC | 14:02 | |
| *** zhipeng has joined #openstack-cinder | 14:03 | |
| *** nosleep77 has joined #openstack-cinder | 14:10 | |
| *** dustins has joined #openstack-cinder | 14:10 | |
| *** avishay has joined #openstack-cinder | 14:11 | |
| openstackgerrit | Rushi Agrawal proposed openstack/cinder: EQLX: Consolidate CHAP config options https://review.openstack.org/148091 | 14:20 |
| *** gtt116_ has quit IRC | 14:32 | |
| *** gtt116_ has joined #openstack-cinder | 14:32 | |
| *** xyang1 has joined #openstack-cinder | 14:33 | |
| *** jcru has joined #openstack-cinder | 14:33 | |
| *** nshaikh has left #openstack-cinder | 14:34 | |
| *** haomai___ has quit IRC | 14:36 | |
| *** takedakn has joined #openstack-cinder | 14:38 | |
| *** takedakn has quit IRC | 14:38 | |
| *** dustins has quit IRC | 14:44 | |
| *** thangp has joined #openstack-cinder | 14:46 | |
| *** bswartz has joined #openstack-cinder | 14:47 | |
| *** jecarey has joined #openstack-cinder | 14:50 | |
| *** fandi has joined #openstack-cinder | 14:51 | |
| *** fandi has quit IRC | 14:53 | |
| *** fandi has joined #openstack-cinder | 14:53 | |
| *** Mandell has joined #openstack-cinder | 14:53 | |
| *** fandi has quit IRC | 14:53 | |
| * DuncanT is just catching up on the meeting from yesterday. Looks like public snapshots are already trying to grow ACLs of sorts (sharing with a bunch of tenants *is* acls). I still think glance is the place for this, not cinder | 14:54 | |
| *** rwsu has joined #openstack-cinder | 14:56 | |
| openstackgerrit | Steven Kaufer proposed openstack/python-cinderclient: cinder list fails with 'name' sort key https://review.openstack.org/142964 | 14:56 |
| *** markstur_ has joined #openstack-cinder | 14:57 | |
| *** markstur_ has left #openstack-cinder | 14:57 | |
| *** EduardBiceriMate has quit IRC | 14:58 | |
| *** xyang has joined #openstack-cinder | 15:00 | |
| *** esker has joined #openstack-cinder | 15:00 | |
| *** Yogi11 has quit IRC | 15:02 | |
| *** fandi has joined #openstack-cinder | 15:05 | |
| *** nellysmitt has quit IRC | 15:05 | |
| dulek | DuncanT: I think the decision was that rushiagr will provide better use cases. | 15:05 |
| dulek | So I ACLs are far from accepting IMO. | 15:06 |
| DuncanT | dulek: Yeah, it looked like it was left open for more info | 15:06 |
| DuncanT | dulek: Just wanted to note that the initial response was 'we'll never want ACLs' and then it changed to 'we want ACLs' | 15:07 |
| DuncanT | dulek: Use cases would be good | 15:07 |
| *** jungleboyj has joined #openstack-cinder | 15:07 | |
| dulek | DuncanT: I haven't had that impression. Anyway we'll wait for use cases. :) | 15:08 |
| *** yrabl has joined #openstack-cinder | 15:10 | |
| *** Mandell has quit IRC | 15:11 | |
| openstackgerrit | Michal Dulko proposed openstack/python-cinderclient: Add tests for consistency groups and cgsnapshots https://review.openstack.org/138410 | 15:11 |
| *** nikesh_vedams has joined #openstack-cinder | 15:11 | |
| *** lpabon has joined #openstack-cinder | 15:13 | |
| *** mtanino has joined #openstack-cinder | 15:14 | |
| *** Mandell has joined #openstack-cinder | 15:15 | |
| openstackgerrit | Rushi Agrawal proposed openstack/cinder: EQLX: Consolidate CHAP config options https://review.openstack.org/148091 | 15:15 |
| *** fischerw has joined #openstack-cinder | 15:15 | |
| rushiagr | DuncanT: I don't think sharing with a particular tenants is ACL | 15:17 |
| rushiagr | DuncanT: The snapshots are never going to be writeable. Not even for the original owners, let alone the ones with whom they're shared.. | 15:18 |
| *** fandi has quit IRC | 15:19 | |
| rushiagr | DuncanT: nor would anybody ever have any way to change permission of a snapshot | 15:19 |
| *** dustins has joined #openstack-cinder | 15:19 | |
| *** dustins has quit IRC | 15:19 | |
| DuncanT | Adding more users to it *is* changing the permissions | 15:19 |
| *** Yogi1 has joined #openstack-cinder | 15:19 | |
| *** coolsvap is now known as coolsvap|afk | 15:19 | |
| DuncanT | As is unsharing | 15:19 |
| rushiagr | DuncanT: that part I agree. But it's not what people undestand by ACL. For people ACL almost always mean full ACL, as far as I have seen.. | 15:20 |
| *** fandi has joined #openstack-cinder | 15:21 | |
| *** dustins has joined #openstack-cinder | 15:22 | |
| *** fandi has quit IRC | 15:22 | |
| rushiagr | dulek: right. I'm going to write a spec soon | 15:22 |
| *** anshul has quit IRC | 15:22 | |
| *** bkopilov has joined #openstack-cinder | 15:23 | |
| *** amoturi has quit IRC | 15:23 | |
| dulek | rushiagr: Why don't you share use cases you have in mind here and we can discuss them in non-meeting form? Maybe it will help us to understand the need better? | 15:23 |
| rushiagr | dulek: okay, that's fine with me | 15:23 |
| *** Miouge has quit IRC | 15:24 | |
| rushiagr | dulek: so ours is a public cloud. And a customer of us already wants to have the option of sharing volume snapshots | 15:24 |
| *** dannywilson has joined #openstack-cinder | 15:25 | |
| *** amoturi has joined #openstack-cinder | 15:26 | |
| smcginnis | I could see that capability being very useful for providers that have storage that supports thin clones. | 15:26 |
| rushiagr | dulek: and they want to share a snapshot with a different tenant. The snapshot contains some data (maybe that volume was used as hdfs for some big data analytics) | 15:27 |
| dulek | rushiagr: Okay, so he want's to have multiple tenants with access to snapshots. | 15:27 |
| rushiagr | dulek: yes | 15:28 |
| rushiagr | dulek: the current way is convert snapshot to a volume, and then transfer that volume to the other tenant | 15:28 |
| dulek | So in case of public cloud is isn't cool to have public snapshot | 15:28 |
| rushiagr | dulek: exactly. That's why public snapshots aren't going to cut at all | 15:29 |
| dulek | Okay, I get it. | 15:30 |
| rushiagr | smcginnis: yeah | 15:30 |
| *** rmesta has joined #openstack-cinder | 15:30 | |
| dulek | So now question for DuncanT if he sees another solution for that. | 15:30 |
| *** yrabl has quit IRC | 15:31 | |
| *** Miouge has joined #openstack-cinder | 15:31 | |
| DuncanT | dulek: Upload image to glance. It already has a rich sharing/acl API | 15:31 |
| rushiagr | also I think it's a bit of an overhead to tell the customer to upload it to glance and download it from there. It's not an easy task for the end user. Moreover, it means data transfer (assuming glance and cinder have different backends) | 15:32 |
| DuncanT | Make cinder a proper glance backend if you want to improve preformance | 15:32 |
| DuncanT | Thinks should fall back to the slow path if necessary, but still work, just like the rest of cinder | 15:32 |
| rushiagr | In our case, glance and cinder ARE different backends :) | 15:33 |
| DuncanT | Glance happily supports multiple backends | 15:33 |
| dulek | DuncanT: This makes sense. In this view ACLs in Cinder are doubled feature. | 15:35 |
| rushiagr | if there's a use case with cinder, and an easy way to do stuff, why would we force them to go to the longer path (cinder snapshot->glance image->cinder volume) is what I'm not understanding | 15:36 |
| rushiagr | we=our users | 15:36 |
| smcginnis | Cinder as Glance backend makes sense to me, as long as there is the ability to thin clone rather than duplicating and consuming extra space. | 15:36 |
| rushiagr | and I'm talking in general. We as the openstack community.. | 15:37 |
| DuncanT | rushiagr: Because it solves a tiny usecase, where as getting cinder<-> optimisation right not only solves the same usecase, but also massively improves the boot-from-volume experience | 15:37 |
| rushiagr | Are we avoiding that feature in Cinder just because it will duplicate functionality? Or it will bring more code which more ways in which cinder can break? | 15:38 |
| DuncanT | For me, both | 15:38 |
| rushiagr | s/which/which means/ | 15:38 |
| DuncanT | Some backends are tenant aware. I have no idea which might break / need reworking to get shared snapshots to work | 15:38 |
| DuncanT | At the same time, glance has good sharing semantics that we should duplicate | 15:39 |
| avishay | Why not clone the original, transfer the clone, and then create more clones from there? | 15:39 |
| rushiagr | avishay: DuncanT: the major problem in my eyes is usability | 15:41 |
| rushiagr | avishay: a user would at one point say: wouldn't it have been easy if we could just share snapshots | 15:41 |
| *** dannywilson has quit IRC | 15:41 | |
| avishay | rushiagr: or a user might say "OMG there are a billion APIs - how do I do what I want?" :) | 15:42 |
| rushiagr | DuncanT: tenant-aware backends seem like a serious problem. I wasn't aware of that until now | 15:42 |
| DuncanT | rushiagr: usability of glance and it's sharing semantics have already had lots of work and thought gone into them. | 15:42 |
| avishay | rushiagr: and developers say "OMG there are a billion APIs - how do we make this all recoverable and sane"? | 15:42 |
| rushiagr | avishay: that's a good thing no? :) | 15:42 |
| DuncanT | rushiagr: It is already used to share bootable images, right now, in the exact way you want to share non-bootable. | 15:42 |
| DuncanT | rushiagr: I think it is far better to re-use existing paradigms than invent new ones | 15:44 |
| rushiagr | avishay: i'm not sure if adding more api's will make developers write less recoverable modules, or make them less saner.. | 15:45 |
| *** bswartz1 has joined #openstack-cinder | 15:45 | |
| DuncanT | rushiagr: Pretty soon somebody realises shared snapshots are a great way to share boot images, but want a catalogue of them... maybe some editable metadata.... maybe some billing metadata too.... pretty soon you've got a bad reimplemention of glance inside cinder | 15:45 |
| rushiagr | DuncanT: I was checking yesterday after the meeting, and found out AWS has the exact same feature already | 15:46 |
| *** bswartz has quit IRC | 15:47 | |
| *** bswartz1 is now known as bswartz | 15:47 | |
| *** e0ne is now known as e0ne_ | 15:47 | |
| *** ebalduf has joined #openstack-cinder | 15:48 | |
| DuncanT | rushiagr: Lots of AWS features are badly designed. I want to hear what can't be done with glance if you want to convince me | 15:48 |
| rushiagr | DuncanT: I'm not aware of catalogue concept sorry.. Yes, metadata stuff will be tricky if somebody wants it.. I don't want it.. | 15:48 |
| *** e0ne_ is now known as e0ne | 15:49 | |
| DuncanT | rushiagr: Catalogue just lets you name, tag and search images in various ways. Pretty much the original mission for glance. Glance doesn't actually care that much where the bits live, other than having to write the correct glance driver. Not trivial but I think a better route to explore. I *think* John is looking at what is involved again, I'll catch up with | 15:50 |
| DuncanT | him next week in Texas | 15:50 |
| openstackgerrit | Xiaolei Hu proposed openstack/cinder: Add iSCSI protocol support for IBM FlashSystem. https://review.openstack.org/144627 | 15:51 |
| *** jecarey has quit IRC | 15:53 | |
| rushiagr | DuncanT: Glance already does what I want. No questions there. I'm saying we're not caring about usability. I'll have to settle for not seeing this feature in Cinder if more people disagree than agree, but that's okay... | 15:53 |
| *** esker has quit IRC | 15:53 | |
| *** resker has joined #openstack-cinder | 15:53 | |
| rushiagr | I'll move on :) | 15:54 |
| DuncanT | rushiagr: Ok, I'll get tighter glance <-> cinder integration on the agenda for next week, see if we can scope the work out and get an idea of how difficult it is to get what you want performance / data-transfer-wise via glance (i.e. zero copy thin snaps being used behind the scenes) and revisit in a fortnight, if you want? | 15:55 |
| rushiagr | DuncanT: that would be good | 15:57 |
| rushiagr | DuncanT: thanks.. | 15:57 |
| DuncanT | rushiagr: I'd like to see it done for faster BfV, but it should give you your usecase too I hope. More news Wednesday week I guess, if not before | 15:58 |
| rushiagr | DuncanT: sure.. thanks | 15:59 |
| *** lcurtis has joined #openstack-cinder | 15:59 | |
| *** BharatK has quit IRC | 16:01 | |
| avishay | DuncanT: I think the Cinder driver for Glance was waiting on the brick initiator library | 16:01 |
| *** amoturi has quit IRC | 16:02 | |
| DuncanT | avishay: Yeah, not sure if that approach is going to work for all possible cinder backends... might be better to add a data transfer http endpoint to cinder, and use that - should provide volume upload/download possibilites too. Not sure if it is the best idea, but I at least want to float it | 16:02 |
| avishay | DuncanT: IMO if someone wants to import a 50GB image, HTTP would be my last choice | 16:04 |
| DuncanT | People already put bigger things than that into swift... | 16:04 |
| *** lpabon has quit IRC | 16:04 | |
| DuncanT | Segmented support etc help | 16:05 |
| DuncanT | But upload to glance is currently http only AFAIK | 16:05 |
| avishay | DuncanT: Yes | 16:05 |
| avishay | DuncanT: think so too | 16:05 |
| *** nlevinki has quit IRC | 16:06 | |
| DuncanT | avishay: So in principle we've making upload easier (get it onto a volume via whatever magic you want, then feed the volume into glance) | 16:07 |
| avishay | DuncanT: by adding HTTP to cinder? | 16:08 |
| DuncanT | avishay: By implementing zero-copy cinder<->glance | 16:09 |
| *** tshefi has quit IRC | 16:09 | |
| DuncanT | avishay: http is just a way of getting old-style-glance compatibility | 16:09 |
| e0ne | hi all | 16:09 |
| DuncanT | avishay: I'm not 100% sure I understand all of the issues, which is why I'd like to bring it up next week | 16:10 |
| DuncanT | Hey e0ne | 16:10 |
| e0ne | here (https://github.com/openstack/cinder/blob/ae5a7d5ea31fa14317696859b1e6e900ff740f26/cinder/volume/manager.py#L612) is something strange or i can't understand | 16:10 |
| e0ne | we change shapshot status to 'available' if it is busy during deleting and return True | 16:11 |
| e0ne | this code is old and forked from nova | 16:11 |
| *** Mandell has quit IRC | 16:11 | |
| e0ne | so, it could be an issue, when snapshot could'n be deleted and user can't see and understand why | 16:12 |
| avishay | DuncanT: I think finishing the Cinder driver in Glance is the way to go, but I've been wrong once before | 16:12 |
| e0ne | only one record in logs... | 16:12 |
| DuncanT | e0ne: We should probably return an error by the sound of it | 16:13 |
| *** bill_az has joined #openstack-cinder | 16:13 | |
| avishay | e0ne: same when attach/detach/delete volume fails IMO | 16:13 |
| e0ne | DuncanT: +1 | 16:13 |
| e0ne | avishay: oh no... | 16:13 |
| e0ne | avishay: i don't want believe in it | 16:14 |
| avishay | https://github.com/openstack/cinder/blob/ae5a7d5ea31fa14317696859b1e6e900ff740f26/cinder/volume/manager.py#L469 | 16:14 |
| e0ne | :( | 16:14 |
| avishay | e0ne: with today's framework you have two options - put the volume in error state, or put it back the way it was | 16:15 |
| e0ne | avishay: i'm voting for error state | 16:15 |
| e0ne | avishay: it's really confusing | 16:15 |
| avishay | e0ne: neither one tells you what went wrong, and in some deployments (public cloud) you may not want the user to know | 16:15 |
| *** zhipeng has quit IRC | 16:16 | |
| *** Yogi1 has quit IRC | 16:16 | |
| *** Mandell has joined #openstack-cinder | 16:16 | |
| xyang | DuncanT: can you take a look of the differential backup patch? https://review.openstack.org/#/c/110068/ | 16:16 |
| *** nshaikh has joined #openstack-cinder | 16:16 | |
| e0ne | avishay: why? it's not user-friendly to set volume/snapshot back after deletion failed | 16:17 |
| e0ne | avishay: it's very confusing | 16:17 |
| *** jecarey has joined #openstack-cinder | 16:18 | |
| e0ne | avishay: i'm trying to delete snapshot and it is in 'available' status w/o changes. only 'updated_at' field could say that something was happend with it | 16:18 |
| avishay | e0ne: error state is less confusing? the volume is perfectly fine - why is it in error state? now i have to manually change the state | 16:18 |
| avishay | e0ne: both solutions are bad | 16:19 |
| *** Mandell has quit IRC | 16:19 | |
| e0ne | avishay: error state shows me that delete action _was_ | 16:19 |
| *** ebalduf has quit IRC | 16:19 | |
| e0ne | available status - looks like 'no delete command was send' | 16:19 |
| *** julim_ has quit IRC | 16:19 | |
| dulek | e0ne: I agree. | 16:20 |
| e0ne | avishay: maybe the error state it not the best solution, maybe we should try once more | 16:20 |
| DuncanT | xyang: Looking now | 16:20 |
| e0ne | anyway, i'll file a bug to discuss it | 16:21 |
| dulek | e0ne: rewriting delete to taskflow will make retries possible | 16:21 |
| xyang | DuncanT: thanks! | 16:21 |
| dulek | I think there were a blueprint for that | 16:21 |
| e0ne | now, i know that this it not bad only for me | 16:21 |
| *** jaypipes has joined #openstack-cinder | 16:21 | |
| e0ne | dulek: taskflow it a tool, not solution | 16:22 |
| dulek | e0ne: true, but retry and then set to error state seems like a fair solution for me | 16:22 |
| e0ne | dulek: +1 | 16:22 |
| dulek | and rewriting it to taskflow would give us retries for free | 16:23 |
| dulek | quick question - should I worry if 3rd party CI build failed on my patch? | 16:24 |
| avishay | dulek: depends on why it failed :) | 16:24 |
| e0ne | :) | 16:25 |
| *** crose has joined #openstack-cinder | 16:25 | |
| dulek | avishay: probably random fail - my change don't touch drivers | 16:26 |
| avishay | dulek: but may change something in the generic code that a driver relied on (whether explicitly or not) | 16:26 |
| avishay | dulek: worth a quick grep in the logs IMO | 16:26 |
| openstackgerrit | Stefan Amann proposed openstack/cinder: Adjust Cinder to support FCP on System z systems https://review.openstack.org/149256 | 16:27 |
| *** hemnafk is now known as hemna | 16:27 | |
| *** resker has quit IRC | 16:28 | |
| *** esker has joined #openstack-cinder | 16:28 | |
| dulek | avishay: Such change should be caught by unit tests probably. | 16:28 |
| dulek | Anyway as I'm looking "IBM XIV CI check" is failing constantly today | 16:29 |
| avishay | dulek: probably not caught by unit tests | 16:29 |
| avishay | dulek: if you don't want to look at why it's failing, nobody can force you i guess | 16:29 |
| dulek | avishay: I'm already looking actually. :) As my original question I was meaning - are these voting? Now I concluding that these are not. ;) | 16:31 |
| *** dannywilson has joined #openstack-cinder | 16:33 | |
| *** dalgaaf has quit IRC | 16:33 | |
| *** fandi has joined #openstack-cinder | 16:36 | |
| *** e0ne is now known as e0ne_ | 16:37 | |
| *** _cjones_ has joined #openstack-cinder | 16:37 | |
| *** ndipanov is now known as ndipanov_gone | 16:40 | |
| jgriffith | dulek: avishay the problem with those two cases are that they're "device" errors that don't allow that specific command | 16:40 |
| *** dulek has quit IRC | 16:40 | |
| *** dulek_ has joined #openstack-cinder | 16:40 | |
| jgriffith | dulek: avishay e0ne_ as far as taskflow, you know we can do retries without taskflow :) | 16:40 |
| *** ryanpetrello has joined #openstack-cinder | 16:41 | |
| jgriffith | if that's really the only thing that people are looking for, why not use a good retry decorator on things | 16:41 |
| jgriffith | particularly since IMO the way taskflow does retries isn't really a good thing | 16:41 |
| *** ebalduf has joined #openstack-cinder | 16:41 | |
| jgriffith | ie; doing the retry at from outside of the manager creates all sorts of "problems" | 16:41 |
| *** mdulko__ has joined #openstack-cinder | 16:41 | |
| jgriffith | it makes it pretty much impossible to keep any knowledge of the state and what happened and how to react | 16:42 |
| *** ebalduf has quit IRC | 16:42 | |
| *** mdulko__ is now known as dulek | 16:42 | |
| jgriffith | unless you implement a metric s-ton of logic back up in taskflow | 16:42 |
| *** amoturi has joined #openstack-cinder | 16:43 | |
| jgriffith | e0ne_: dulek avishay for that matter..... we could just put retry loops in cinder/volume/api at each call and have the same thing as taskflow without extra code :) | 16:43 |
| jgriffith | and FWIW... long long ago we did wrap commands in manager with retries (particularly the LVM calls) | 16:44 |
| *** dulek_ has quit IRC | 16:44 | |
| *** e0ne_ is now known as e0ne | 16:44 | |
| dulek | jgriffith: Oh, you can even write a decorator to trigger retries on particular methods ;) there's many possibilities | 16:44 |
| jgriffith | dulek: right, that's what I started this with :) | 16:45 |
| *** dannywilson has quit IRC | 16:45 | |
| *** diemt has joined #openstack-cinder | 16:45 | |
| jgriffith | dulek: and the thing about that is that you actually have knowledge of state and what's going on right there where it matters | 16:45 |
| openstackgerrit | Rushi Agrawal proposed openstack/cinder: EQLX: Consolidate CHAP config options https://review.openstack.org/148091 | 16:46 |
| jgriffith | dulek: I've yet to have anybody show me the advantage of just hammering the manager call repeatedly with the same command | 16:46 |
| jgriffith | and I've yet to see it "fix" anything TBH | 16:46 |
| jgriffith | but I'm hoping to do some testing on that attach patch and see if it in fact 'fixes' anything | 16:46 |
| dulek | jgriffith: had hard time with TaskFlow, didn't you? ;) Maybe you're right, we'll see | 16:47 |
| jgriffith | but I'll down vote it unti there's some data that shows it actually fixes something | 16:47 |
| jgriffith | dulek: I did, but that's not why I'm pointing this out to be honest | 16:47 |
| e0ne | jgriffith: sure, retry could helps us in a short-term | 16:47 |
| jgriffith | dulek: I've always tried to keep an open mind on this, but I've yet to have any "data" to show me how it's an improvement on anything | 16:48 |
| jgriffith | dulek: sadly, create was the worst thing for us to start with in hindsight | 16:48 |
| e0ne | dulek: there are a list of issues with implementing common-retry decorator:( | 16:48 |
| jgriffith | dulek: we should've picked something like delete or attach (even detach) | 16:48 |
| jgriffith | those would've been much more informitive cases to learn from | 16:49 |
| dulek | jgriffith: maybe we should discuss it more later? Got to leave work now, if I'll have time I'll message you from home | 16:49 |
| avishay | jgriffith: and we should have split the create code into totally separate paths with reusable functions | 16:49 |
| jgriffith | avishay: agreed | 16:49 |
| jgriffith | avishay: it's a bit of a frankenstein | 16:50 |
| jgriffith | avishay: live and learn eh :) | 16:50 |
| hemna | mornin | 16:50 |
| jgriffith | dulek: sure | 16:50 |
| avishay | jgriffith: yep :) | 16:50 |
| jgriffith | hemna: howdy | 16:50 |
| jgriffith | brb | 16:50 |
| avishay | hemna: yo | 16:50 |
| e0ne | avishay: +1 and same to unit-tests as jgriffith mentioned one meeting ago | 16:50 |
| hemna | so I think I might have a working brick lib | 16:50 |
| e0ne | hemna: evening:) | 16:51 |
| hemna | so, I'm not sure what the next steps are | 16:51 |
| hemna | do I post it as a personal repo for the time being in github? | 16:51 |
| hemna | but I need to get it into the openstack/brick github to make it part of a subproject for cinder that cinder owns | 16:52 |
| hemna | chicken and egg a bit | 16:52 |
| *** dannywilson has joined #openstack-cinder | 16:53 | |
| e0ne | hemna: i found a bug in my flask-keystone integration durint writing tests. is support authorization only via user input. i'll fix it before weekend | 16:53 |
| hemna | coolio :) | 16:53 |
| *** nshaikh has quit IRC | 16:54 | |
| *** marcusvrn1 has joined #openstack-cinder | 16:54 | |
| *** dannywilson has quit IRC | 16:55 | |
| *** dannywilson has joined #openstack-cinder | 16:56 | |
| *** dulek has quit IRC | 16:56 | |
| e0ne | btw, i'm happy that flask is in global-requirements | 16:56 |
| hemna | e0ne, :) | 16:56 |
| hemna | yah I really like flask | 16:56 |
| e0ne | it depends... | 16:57 |
| e0ne | somtimes i prefer django | 16:57 |
| *** esker has quit IRC | 16:57 | |
| *** marcusvrn has quit IRC | 16:57 | |
| *** marcusvrn_ is now known as marcusvrn | 16:57 | |
| *** resker has joined #openstack-cinder | 16:57 | |
| *** lcurtis has quit IRC | 16:58 | |
| hemna | TobiasE, ok, so it looks like the 2nd attach overwrites the CHAP information for the 1st attach | 16:58 |
| hemna | TobiasE, we'll have to adjust tgt target to account for the same volume being exported already | 16:58 |
| hemna | when jgriffith comes back I can ping him about it. see if we can simply do a follow up patch to fix it. | 16:59 |
| jgriffith | hemna: he's back | 17:03 |
| jgriffith | :) | 17:03 |
| hemna | hey :) | 17:03 |
| jgriffith | yoyos | 17:03 |
| hemna | jgriffith, so TobiasE found an issue with multi-attach and the tgt target object | 17:04 |
| hemna | with CHAP | 17:04 |
| jgriffith | hemna: yeah... reading that | 17:04 |
| hemna | basically, the 2nd attach, overwrites the CHAP credentials for the 1st attach | 17:04 |
| hemna | and then it all goes south from there | 17:04 |
| hemna | the /etc/tgt/stack.d/volume-<UUID> file | 17:04 |
| jgriffith | hemna: so I wonder why the _get_target_chap_auth doesn't pick that up from the persistence file? | 17:05 |
| *** avishay has quit IRC | 17:05 | |
| jgriffith | hemna: know what I mean? | 17:06 |
| jgriffith | hemna: TobiasE We shouldn't actually hit that code | 17:06 |
| hemna | jgriffith, what's your email addy? I can forward you the email with the details that TobiasE sent me this morning | 17:06 |
| jgriffith | hemna: TobiasE if we already have an export created we shouldn't create it again | 17:06 |
| jgriffith | hemna: john.griffith8@gmail.com | 17:06 |
| jgriffith | hemna: sometimes I actually check that one :) | 17:07 |
| hemna | heh | 17:07 |
| hemna | ok sent | 17:07 |
| *** rilango has quit IRC | 17:07 | |
| hemna | I'm assuming this goes through tgt:create_export ? | 17:08 |
| hemna | doesn't look like it looks for an existing target exported | 17:08 |
| *** med_ has joined #openstack-cinder | 17:08 | |
| jgriffith | hemna: gimmie a few, have to finish a conversation :) | 17:08 |
| hemna | ok np | 17:08 |
| *** med_ has quit IRC | 17:08 | |
| *** med_ has joined #openstack-cinder | 17:08 | |
| openstackgerrit | Sean McGinnis proposed openstack/cinder: Improve debug logging of Dell Storage Center driver https://review.openstack.org/148744 | 17:08 |
| jungleboyj | hemna, jgriffith xyang Do you guys mind looking at this today? https://review.openstack.org/148018 I would like to get it merged so I can wrap up the namespace move. | 17:12 |
| hemna | jungleboyj, whoa, you are here today | 17:13 |
| hemna | :P | 17:13 |
| jungleboyj | hemna: :-p | 17:13 |
| jungleboyj | Sorry, been all over the place. | 17:13 |
| xyang | jungleboy: hi, I need to be offline for while, will look at it this afternoon if it is not merged yet | 17:15 |
| *** tsekiyama has joined #openstack-cinder | 17:15 | |
| jungleboyj | xyang: Thank you! | 17:16 |
| jgriffith | hemna: hmmm... I see | 17:16 |
| jgriffith | hemna: I thought it should infact look for the existing target and just move along | 17:16 |
| jgriffith | hemna: I'll have a look here and see if I can understand why it's barfing here | 17:16 |
| hemna | jgriffith, ok thank you | 17:17 |
| jgriffith | hemna: what all do I need to load to test this out? | 17:17 |
| hemna | heh | 17:17 |
| jgriffith | hemna: don't get me wrong... if you're already on it that's fine by me :) | 17:17 |
| hemna | it's a bit of an chore | 17:17 |
| jgriffith | hemna: just assuming you were asking if I'd look at it | 17:17 |
| jgriffith | hemna: it always is :) | 17:17 |
| hemna | jgriffith, you'll need 3 patch sets, my cinder patch, cinderclient and nova | 17:17 |
| jgriffith | hemna: I thought your Cinder patch landed? | 17:18 |
| hemna | well, kinda 2 questions really | 17:18 |
| hemna | it hasn't | 17:18 |
| hemna | just saw that avishay -1'd it | 17:18 |
| hemna | heh | 17:18 |
| e0ne | who can change channel topic? it would be useful to add link to irc logs: http://eavesdrop.openstack.org/irclogs/%23openstack-cinder/ | 17:18 |
| hemna | so, once we come up with a fix for this, I presume it can be a follow up patch ? | 17:19 |
| jgriffith | e0ne: only thingee | 17:19 |
| hemna | I'd prefer this cinder review to not grow in size | 17:19 |
| e0ne | jgriffith: thanks | 17:19 |
| jgriffith | hemna: for sure... follow up | 17:19 |
| hemna | ok | 17:19 |
| jgriffith | hemna: alright... I'll have a look and see if I can help | 17:19 |
| *** xyang1 has quit IRC | 17:20 | |
| hemna | ok thanks | 17:20 |
| hemna | jgriffith, so https://review.openstack.org/#/c/85847/ https://review.openstack.org/#/c/85856/ and https://review.openstack.org/#/c/143114/ | 17:20 |
| jgriffith | hemna: perfect... thanks! | 17:20 |
| hemna | is what's needed for end to end testing | 17:20 |
| *** jpich has quit IRC | 17:20 | |
| jgriffith | ssh devbox | 17:20 |
| hemna | I've tested it with 3par, and LVM on the same devstack instance. | 17:21 |
| hemna | TobiasE, says he sees this problem when he tests with 2 compute hosts. | 17:21 |
| hemna | not sure why it doesn't happen with a single host actually | 17:21 |
| *** dannywilson has quit IRC | 17:21 | |
| jgriffith | hemna: hmmm.... | 17:22 |
| jgriffith | hemna: what target is TobiasE using? | 17:22 |
| hemna | tgt afaik | 17:22 |
| jgriffith | k | 17:22 |
| jgriffith | I'll add a compute node in and check it out | 17:22 |
| *** xyang has quit IRC | 17:23 | |
| *** ebalduf has joined #openstack-cinder | 17:27 | |
| *** aix has quit IRC | 17:27 | |
| *** mudassirlatif has joined #openstack-cinder | 17:28 | |
| *** emagana has joined #openstack-cinder | 17:32 | |
| *** e0ne has quit IRC | 17:35 | |
| *** julim has joined #openstack-cinder | 17:36 | |
| openstackgerrit | John Griffith proposed openstack/cinder: Add a generic retry decorator to cinder/utils https://review.openstack.org/149341 | 17:37 |
| *** amoturi has quit IRC | 17:40 | |
| *** leeantho has joined #openstack-cinder | 17:40 | |
| *** Longgeek has quit IRC | 17:43 | |
| *** jordanP has quit IRC | 17:44 | |
| *** resker has quit IRC | 17:45 | |
| *** harlowja_at_home has joined #openstack-cinder | 17:45 | |
| *** ryanpetrello has quit IRC | 17:49 | |
| *** patrickeast has joined #openstack-cinder | 17:49 | |
| *** yrabl has joined #openstack-cinder | 17:49 | |
| *** ryanpetrello has joined #openstack-cinder | 17:50 | |
| *** david-lyle has joined #openstack-cinder | 17:50 | |
| *** ryanpetrello has quit IRC | 17:50 | |
| *** ryanpetrello has joined #openstack-cinder | 17:55 | |
| *** lpabon has joined #openstack-cinder | 17:56 | |
| *** annashen has joined #openstack-cinder | 17:57 | |
| *** ryanpetrello has quit IRC | 17:58 | |
| *** harlowja_at_home has quit IRC | 17:59 | |
| *** timcl1 has joined #openstack-cinder | 17:59 | |
| *** nellysmitt has joined #openstack-cinder | 18:00 | |
| *** timcl has quit IRC | 18:02 | |
| *** tbarron has quit IRC | 18:03 | |
| *** tbarron has joined #openstack-cinder | 18:04 | |
| *** akerr has quit IRC | 18:04 | |
| *** ryanpetrello has joined #openstack-cinder | 18:04 | |
| *** akerr has joined #openstack-cinder | 18:04 | |
| *** nellysmitt has quit IRC | 18:04 | |
| akerr | jgriffith, thingee: can I get another +2 for https://review.openstack.org/#/c/145239/ ? | 18:05 |
| *** dustins_ has joined #openstack-cinder | 18:06 | |
| openstackgerrit | John Griffith proposed openstack/cinder: Add a generic retry decorator to cinder/utils https://review.openstack.org/149341 | 18:06 |
| *** vilobhmm has joined #openstack-cinder | 18:07 | |
| *** ryanpetrello_ has joined #openstack-cinder | 18:07 | |
| *** ryanpetrello_ has quit IRC | 18:08 | |
| *** ryanpetrello has quit IRC | 18:08 | |
| akerr | Thanks! :) | 18:08 |
| *** ryanpetrello has joined #openstack-cinder | 18:09 | |
| *** Mandell has joined #openstack-cinder | 18:09 | |
| *** ryanpetrello has quit IRC | 18:09 | |
| *** ryanpetrello has joined #openstack-cinder | 18:10 | |
| *** dustins has quit IRC | 18:10 | |
| *** lcurtis has joined #openstack-cinder | 18:11 | |
| *** marcusvrn has quit IRC | 18:12 | |
| *** rushiagr is now known as rushiagr_away | 18:13 | |
| *** nellysmitt has joined #openstack-cinder | 18:17 | |
| openstackgerrit | John Griffith proposed openstack/cinder: Add a generic retry decorator to cinder/utils https://review.openstack.org/149341 | 18:20 |
| *** e0ne has joined #openstack-cinder | 18:21 | |
| *** harlowja_away is now known as harlowja | 18:24 | |
| openstackgerrit | John Griffith proposed openstack/cinder: Add retry to lvm snapshot create https://review.openstack.org/149360 | 18:25 |
| jgriffith | boris-42: ^^ | 18:25 |
| jgriffith | boris-42: I'd like to see how that does on those Rally checks, assuming we avoid the "no space left" conflict | 18:25 |
| *** timcl1 has quit IRC | 18:25 | |
| *** xyang has joined #openstack-cinder | 18:26 | |
| *** emagana has quit IRC | 18:26 | |
| *** emagana has joined #openstack-cinder | 18:27 | |
| jgriffith | jungleboyj: hemna DuncanT e0ne if we merge https://review.openstack.org/#/c/149341/, I have several places where I think it will be of value | 18:27 |
| *** timcl has joined #openstack-cinder | 18:27 | |
| jgriffith | including some custom retry loops we have scattered about | 18:27 |
| *** emagana has quit IRC | 18:28 | |
| *** emagana has joined #openstack-cinder | 18:28 | |
| hemna | is this in replacement of the taskflow retry mechanism ? | 18:30 |
| jgriffith | hemna: it's not taskflow if that's what you're asking :) | 18:30 |
| jgriffith | hemna: it's a retry decorator that can be used without needing taskflow | 18:30 |
| hemna | heh ok | 18:30 |
| jgriffith | hemna: in other words, it's a single class/decorator that you have control over if all you need is smart consistent retries | 18:31 |
| hemna | pretty cool actually | 18:31 |
| jgriffith | hemna: it doesn't do all the other things taskflow does though, it only does retries | 18:31 |
| jungleboyj | jgriffith: Oooh, your first line has me intrigued. Looking. | 18:32 |
| hemna | brick's connector and nova's libvirt volume drivers basically manually do this | 18:32 |
| hemna | while searching for the volume to show up at attach time | 18:33 |
| jgriffith | hemna: yeah, we have a bunch of places in OpenStack that do this on their own | 18:33 |
| jgriffith | hemna: so I'd like to see how this goes, then I'll propose it to OSLO | 18:33 |
| hemna | kinda sucks to repeat that over and over | 18:33 |
| hemna | nice | 18:33 |
| jgriffith | hemna: and we can get a common base for it | 18:33 |
| hemna | we all love oslo | 18:33 |
| jgriffith | hemna: haha | 18:33 |
| hemna | :) | 18:33 |
| *** ebalduf has quit IRC | 18:38 | |
| *** ryanpetrello has quit IRC | 18:38 | |
| *** ebalduf has joined #openstack-cinder | 18:39 | |
| jungleboyj | hemna: Now, be nice. | 18:42 |
| hemna | uh oh. Captain oslo is back | 18:42 |
| jungleboyj | hemna: :-p | 18:42 |
| *** marcusvrn has joined #openstack-cinder | 18:44 | |
| e0ne | griffith: great! thanks, i'll take a look on it | 18:45 |
| e0ne | vilobhmm: hi | 18:45 |
| openstackgerrit | John Griffith proposed openstack/cinder: Add retry to lvm snapshot create https://review.openstack.org/149360 | 18:46 |
| *** nellysmitt has quit IRC | 18:47 | |
| *** marcusvrn1 has quit IRC | 18:47 | |
| *** esker has joined #openstack-cinder | 18:48 | |
| hemna | jungleboyj, jgriffith so cinder multi-attach | 18:53 |
| jgriffith | hemna: working on that now :) | 18:53 |
| hemna | avishay complained about the word "shareable" as the keyword to use for making a volume multi attachable | 18:53 |
| jgriffith | hemna: did you fix it? | 18:53 |
| jgriffith | hemna: reasonable point | 18:53 |
| hemna | thoughts? | 18:54 |
| jgriffith | hemna: confuses things with those "FS" based shares | 18:54 |
| hemna | he wants "allow_multiattach" | 18:54 |
| hemna | but allow_multiattach sucks as an option to cinder create | 18:54 |
| jgriffith | hemna: agree | 18:54 |
| hemna | so..... | 18:54 |
| hemna | I like shareable | 18:54 |
| hemna | even if it's confusing to some. | 18:54 |
| jgriffith | hemna: yeah | 18:55 |
| hemna | unless we have another short word to use | 18:55 |
| jungleboyj | hemna: Let me think. | 18:55 |
| *** dulek has joined #openstack-cinder | 18:55 | |
| jungleboyj | mutliattachable. ;-) | 18:55 |
| hemna | lol | 18:56 |
| hemna | that'll be fun to type | 18:56 |
| jgriffith | watch out for the SP police | 18:56 |
| *** alwaysatthenoc has joined #openstack-cinder | 18:56 | |
| hemna | --cluster | 18:57 |
| hemna | ? | 18:57 |
| hemna | meh | 18:57 |
| jungleboyj | Yeah, I just tried and I had trouble doing it. | 18:57 |
| jungleboyj | Can we just do multiattach ? | 18:57 |
| *** timcl has quit IRC | 18:57 | |
| hemna | cinder create --allow-multiattach True --name foo 1 | 18:58 |
| jgriffith | hemna: works for me, but just make it a flag | 18:58 |
| jgriffith | hemna: no need for arg IMO | 18:58 |
| *** jistr has quit IRC | 18:58 | |
| jgriffith | hemna: honestly the other thing I haven't figured out is "why" we even need it | 18:58 |
| hemna | the flag ? | 18:58 |
| jgriffith | hemna: well, yeah... but I'm wrong :) | 18:58 |
| *** ndipanov_gone has quit IRC | 18:59 | |
| jgriffith | hemna: i was thinking just make the default multi-attach | 18:59 |
| *** timcl has joined #openstack-cinder | 18:59 | |
| jgriffith | and skip the whole thing | 18:59 |
| hemna | I think the original discussion about not having it, was folks were worried that lame users would try it | 18:59 |
| jgriffith | hemna: yeah | 18:59 |
| hemna | and then bitch when their ext4fs got corrupted | 18:59 |
| jgriffith | hemna: Mark my words, that's going to happen anyway :) | 18:59 |
| hemna | hah yah | 18:59 |
| jgriffith | hemna: :) | 18:59 |
| e0ne | :) | 18:59 |
| hemna | we'll just say.....you enabled multiattach. don't do that. | 19:00 |
| jgriffith | :) agreed | 19:00 |
| jgriffith | just making a prediction :) | 19:00 |
| hemna | yah I can see it happening. | 19:00 |
| hemna | wow cool! Cinder lets us multiattach now. | 19:00 |
| hemna | w00t | 19:00 |
| e0ne | make it only for admnins:) | 19:00 |
| hemna | ext4fs here I come! | 19:00 |
| jgriffith | hemna: anyway... I think your last suggestion is the right way to go | 19:00 |
| jgriffith | e0ne: that's not a good idea I don't think | 19:01 |
| e0ne | jgriffith: sure, it was sarcasm | 19:01 |
| hemna | ok I'll do allow-multiattach | 19:01 |
| jgriffith | e0ne: Yeah, just recognized that :) | 19:01 |
| jgriffith | hemna: I think that's good | 19:01 |
| hemna | and change the table column to multiattach | 19:01 |
| jgriffith | hemna: also though... just make it so you don't need an arg to it | 19:01 |
| akerr | will the default be true or faulse? | 19:01 |
| jgriffith | hemna: ie just a flag | 19:01 |
| akerr | false even | 19:01 |
| jgriffith | akerr: I'm proposing default False | 19:02 |
| hemna | jgriffith, yah that'd be ideal. I think I had looked into that at one point and couldn't figure it out.. | 19:02 |
| hemna | the shell decorator crap | 19:02 |
| hemna | the default is off/false | 19:02 |
| akerr | cool | 19:02 |
| jgriffith | akerr: so you could just say "cinder create --allow-multiattach 5" | 19:02 |
| akerr | jgriffith: ya sounds good | 19:02 |
| jgriffith | akerr: hemna not sure if the patch to fix arg order like that landed or not but you'll find out soon enough :) | 19:02 |
| hemna | hehe | 19:03 |
| hemna | ok I'll hack around and see what I can do | 19:03 |
| *** dulek_ has joined #openstack-cinder | 19:03 | |
| hemna | thanks for the input guys | 19:03 |
| boris-42 | jgriffith: hm | 19:03 |
| boris-42 | jgriffith: I can remove failing benchmarks.. | 19:03 |
| jgriffith | boris-42: wait what? | 19:03 |
| akerr | I'll admit I didn't read the patch yet, but is there a mechanism for an entire backend to reject multiattach or handle not supporting it? | 19:03 |
| boris-42 | jgriffith: and publish new patch if you wish | 19:03 |
| *** dulek has quit IRC | 19:03 | |
| boris-42 | jgriffith: for testing | 19:03 |
| jgriffith | boris-42: you mean on the ones that we're running out of space? | 19:04 |
| boris-42 | jgriffith: Not enough space is not related tocreate-attach | 19:04 |
| jgriffith | boris-42: I'm confused :( | 19:04 |
| jgriffith | boris-42: Ohhhh :) | 19:04 |
| jgriffith | hmmm | 19:04 |
| e0ne | boris-42: which benchmarks are you going to delete? | 19:04 |
| jgriffith | boris-42: ok... so let's hold up | 19:04 |
| jungleboyj | hemna: So is the decision to have --allow-multiattach ? | 19:04 |
| boris-42 | jgriffith: so let me expalin | 19:04 |
| hemna | jungleboyj, yah I think so | 19:04 |
| boris-42 | jgriffith: I run 30 times same benchmark | 19:04 |
| boris-42 | jgriffith: related to attaching stuff | 19:04 |
| jgriffith | boris-42: ok | 19:04 |
| boris-42 | jgriffith: and it passed 30 times | 19:04 |
| boris-42 | jgriffith: that proves that your patch seems like works* | 19:05 |
| hemna | jungleboyj, and the volume table's column name will be changed from 'shareable' to 'multiattach' | 19:05 |
| boris-42 | jgriffith: or at least code after it is for some reason more stable | 19:05 |
| jgriffith | boris-42: ok... so that's good | 19:05 |
| jungleboyj | Ok, I think that is fine. Don't need an arg to it but then we are less likely to have people shoot themselves in the foot. | 19:05 |
| boris-42 | jgriffith: that was my goal =) | 19:05 |
| jungleboyj | I am cool with that. | 19:05 |
| jgriffith | boris-42: but why the no space issues? | 19:05 |
| boris-42 | jgriffith: of pushing that patch | 19:05 |
| jungleboyj | hemna: +2 | 19:05 |
| hemna | ok coolio | 19:05 |
| boris-42 | jgriffith: it's related to the failing benchmarks* | 19:05 |
| boris-42 | jgriffith: so seems like we need to reduce a times | 19:05 |
| boris-42 | jgriffith: in some of benchmarks* | 19:05 |
| boris-42 | jgriffith: or concurrency | 19:05 |
| boris-42 | jgriffith: I will publsih separated patch | 19:05 |
| boris-42 | jgriffith: to fix that | 19:06 |
| jgriffith | boris-42: yeah, I thnk it's concurrency, although I haven't dug in to it | 19:06 |
| boris-42 | jgriffith: I'll take about that a bit latter | 19:06 |
| jgriffith | boris-42: ok... so that all sounds promising and I think the patch I did for the snap create will work too | 19:06 |
| jgriffith | boris-42: I'll rework the attach retry stuff next | 19:06 |
| jungleboyj | hemna: Thanks. | 19:06 |
| boris-42 | jgriffith: ok great | 19:06 |
| jgriffith | boris-42: clean that up and use the retry-decorator | 19:06 |
| jgriffith | boris-42: thanks for all the help on this | 19:07 |
| boris-42 | jgriffith: I be back a bit later | 19:07 |
| boris-42 | jgriffith: no worries=) | 19:07 |
| jgriffith | boris-42: Rally has proved very useful here | 19:07 |
| boris-42 | jgriffith: I hope nova will understand that as well=) | 19:07 |
| *** dulek_ has quit IRC | 19:08 | |
| *** lpetrut has quit IRC | 19:08 | |
| *** jecarey_ has joined #openstack-cinder | 19:08 | |
| *** jecarey has quit IRC | 19:08 | |
| *** dustins_ has quit IRC | 19:10 | |
| jaypipes | hemna: FWIW, I like "--allow-multiattach" better than "shareable" as well. | 19:11 |
| hemna | jaypipes, :) ok thanks | 19:11 |
| hemna | I'm working on it now. | 19:11 |
| jaypipes | cheers | 19:11 |
| *** dustins has joined #openstack-cinder | 19:12 | |
| hemna | jgriffith, jungleboyj e0ne: FWIW, here is my first attempt at the brick lib. https://github.com/WaltHP/cinder-brick | 19:12 |
| *** ndipanov_gone has joined #openstack-cinder | 19:12 | |
| hemna | need to figure out how to migrate that under cinder | 19:12 |
| hemna | and get it gated, etc. | 19:12 |
| hemna | if you guys can pull it down and run tox just to make sure it's all good. | 19:12 |
| harlowja | vilobhmm welcome back! | 19:13 |
| harlowja | e0ne vilobhmm shake hands, ha :-P | 19:13 |
| e0ne | harlowja: thanks | 19:14 |
| jgriffith | Ohhhh jungleboyj... | 19:14 |
| e0ne | hemna: what do you mean under 'migrate that under cinder'? | 19:15 |
| hemna | meaning it needs to be an official repo and owned by the cinder team | 19:15 |
| jgriffith | e0ne: how strong do you feel about the name :) | 19:15 |
| jungleboyj | jgriffith: :-) | 19:15 |
| alwaysatthenoc | Is it possible to cinder migrate an attached root ISCSI volume? It seems to error out in my test. | 19:15 |
| jgriffith | e0ne: I like the fact that it's a decorator being explicit myself | 19:15 |
| hemna | I like the name | 19:16 |
| jgriffith | alwaysatthenoc: well then that would indicate no :) | 19:16 |
| jgriffith | alwaysatthenoc: kidding aside... no | 19:16 |
| jgriffith | alwaysatthenoc: the reason being is you need to do an attach for migrate in most cases | 19:16 |
| jgriffith | alwaysatthenoc: so it needs detached to proceed | 19:17 |
| e0ne | jgriffith: i won't put -1, but i feel not very comfotable to use it: too long code lines, need to use milti-line statement | 19:17 |
| e0ne | jgriffith: but in python explicit better than implicit | 19:17 |
| jgriffith | e0ne: :) | 19:17 |
| alwaysatthenoc | jgriffith: so the best option is to poweroff the VM, detatch the volume, migrate then reattach? | 19:17 |
| jgriffith | e0ne: it's better than: HTTPConnectionDecorator | 19:18 |
| jgriffith | :) | 19:18 |
| jgriffith | alwaysatthenoc: no need to power off... unless your boot from volume? | 19:18 |
| jgriffith | alwaysatthenoc: then yes | 19:18 |
| e0ne | jgriffith: :) | 19:18 |
| alwaysatthenoc | jgriffith: I do boot from the volume. Thank you | 19:18 |
| jgriffith | alwaysatthenoc: got ya.... then yeah. | 19:19 |
| jgriffith | hmmm.... seg fault in Jenkins? | 19:20 |
| jgriffith | or testr | 19:21 |
| openstackgerrit | John Griffith proposed openstack/cinder: Add a generic retry decorator to cinder/utils https://review.openstack.org/149341 | 19:22 |
| openstackgerrit | Richard Hagarty proposed openstack/cinder: HP 3PAR modules do not fillow coding guidelines https://review.openstack.org/149373 | 19:23 |
| openstackgerrit | Joshua Harlow proposed openstack/cinder: Remove usage of taskflow 'utils.misc' module https://review.openstack.org/149374 | 19:24 |
| e0ne | hemna: cinder-brick tests work on my env | 19:27 |
| vilobhmm | harlowja : talking to eone in 1:1 chat…sorry just saw your msg was in a meeting | 19:27 |
| hemna | e0ne, great thanks | 19:27 |
| harlowja | vilobhmm np | 19:27 |
| hemna | e0ne, I was able to great a simple python script that imports the lib and gets the connector info. seemed to work. | 19:28 |
| harlowja | just wanted to introduce vilobhmm but i guess u already got that covered :) | 19:28 |
| vilobhmm | :) | 19:28 |
| hemna | so I wanted to get the lib into check/gate, and then work on the cinder patch that uses it. | 19:28 |
| jgriffith | jungleboyj: added the punctuation mark :) | 19:30 |
| jungleboyj | jgriffith: Thank you. | 19:33 |
| openstackgerrit | Walter A. Boring IV (hemna) proposed openstack/cinder: Add volume multi attach support https://review.openstack.org/85847 | 19:35 |
| hemna | ^^ the shareable => multiattach flag | 19:35 |
| hemna | next up cinderclient | 19:35 |
| jungleboyj | jgriffith: I was a good boy and didn't -1 you for it. | 19:36 |
| *** lpetrut has joined #openstack-cinder | 19:38 | |
| smcginnis | jgriffith: Thank you sir! | 19:39 |
| jgriffith | smcginnis: certainly | 19:39 |
| *** vilobhmm has quit IRC | 19:39 | |
| *** vilobhmm has joined #openstack-cinder | 19:40 | |
| *** nellysmitt has joined #openstack-cinder | 19:42 | |
| *** _cjones_ has quit IRC | 19:43 | |
| smcginnis | jgriffith: Noticed decorator patch issue: https://jenkins06.openstack.org/job/gate-cinder-python27/2128/console | 19:48 |
| *** ebalduf has quit IRC | 19:49 | |
| e0ne | hemna: dou you need any help with gating? i've got some expirience with it | 19:49 |
| hemna | e0ne, at this point I need to figure out how to get my personal repo in github into an official repo for openstack as a subproject of cinder | 19:50 |
| *** kaufer has left #openstack-cinder | 19:50 | |
| e0ne | hemna: it's easy | 19:51 |
| * hemna waves magic wand | 19:51 | |
| e0ne | hemna: create review-request to infra and wait few days:) | 19:51 |
| *** dannywilson has joined #openstack-cinder | 19:51 | |
| vilobhmm | eharney : do you have a repo where you are making the delete flow changes ? | 19:52 |
| openstackgerrit | Mikhail Khodos proposed openstack/cinder: Fix Nexenta NFS driver mounts https://review.openstack.org/149380 | 19:53 |
| vilobhmm | e0ne mentioned that you have started working on delete flow so was curious to know | 19:53 |
| hemna | fwiw, I'm helping a coworker look into the idea of adding API 'ing' checks to see if we can nuke the volume manager local file locks. | 19:54 |
| *** Yogi1 has joined #openstack-cinder | 19:54 | |
| hemna | should have something to talk about in Austin next week. | 19:54 |
| hemna | leeantho, is doing the work | 19:54 |
| eharney | vilobhmm: I'm supposed to be, but I've been buried in other work and it hasn't gone anywhere yet | 19:55 |
| *** rongze has quit IRC | 19:57 | |
| *** dannywilson has quit IRC | 19:57 | |
| vilobhmm | oh okay | 19:58 |
| harlowja | hemna cools; less file locks ftw | 19:58 |
| openstackgerrit | Walter A. Boring IV (hemna) proposed openstack/python-cinderclient: Add volume multi attach support https://review.openstack.org/85856 | 20:00 |
| vilobhmm | is there a hangout link for the mid cycle meetup is anyone is aware of | 20:00 |
| *** Mandell has quit IRC | 20:00 | |
| hemna | cinder create --allow-multiattach --name foo 11 | 20:00 |
| *** emagana has quit IRC | 20:00 | |
| harlowja | vilobhmm there usually is one someone sets up, not sure if its created yet | 20:01 |
| harlowja | i tried it last time, it'd be better if that microsoft hologram tech worked though and u could use that | 20:01 |
| harlowja | virtual hemna and jgriffith and ... in hologrammmmms | 20:01 |
| *** vilobhmm has quit IRC | 20:02 | |
| hemna | jaypipes, so just put APIImpact as a line in the commit message ? | 20:03 |
| *** dannywilson has joined #openstack-cinder | 20:04 | |
| jaypipes | hemna: something like this would work: | 20:04 |
| jaypipes | APIImpact: Admin action XXX {URI} now accepts 'attachment_id' in request body | 20:05 |
| jaypipes | hemna: and replace XXX with whatever the API extension name is, and {URI} withthe URI construct | 20:05 |
| hemna | jaypipes, ok | 20:06 |
| *** vilobhmm has joined #openstack-cinder | 20:06 | |
| hemna | do we have any existing examples in outstanding reviews I can check out ? | 20:06 |
| jgriffith | jungleboyj: is Steven Kaufer on IRC (do you know him)? | 20:06 |
| jgriffith | jungleboyj: wondering if there's any value in a compat flag: https://review.openstack.org/#/c/148373/ | 20:06 |
| jaypipes | annegentle: see hemna's ? above... any links for him to show? | 20:06 |
| jgriffith | jungleboyj: so it would be interesting for changes like this to add the capability to have things work the "old" way | 20:07 |
| jgriffith | but I dunno... maybe silly | 20:07 |
| *** yrabl has quit IRC | 20:07 | |
| openstackgerrit | John Griffith proposed openstack/cinder: Add retry to lvm snapshot create https://review.openstack.org/149360 | 20:09 |
| *** kaufer has joined #openstack-cinder | 20:10 | |
| *** juzuluag has quit IRC | 20:10 | |
| vilobhmm | harlowja : ok | 20:11 |
| vilobhmm | hologram thing looks cool though | 20:11 |
| harlowja | ya, i'm sure the more seedy parts of the internet will take full advantage of that,lol | 20:11 |
| harlowja | *not that i know what those seedy parts are... | 20:12 |
| harlowja | i just heard from a friend man... | 20:12 |
| harlowja | lol | 20:12 |
| *** _cjones_ has joined #openstack-cinder | 20:14 | |
| *** xyang has quit IRC | 20:18 | |
| *** dannywilson has quit IRC | 20:21 | |
| *** emagana has joined #openstack-cinder | 20:21 | |
| *** Mandell has joined #openstack-cinder | 20:25 | |
| *** fischerw has quit IRC | 20:25 | |
| *** fischerw has joined #openstack-cinder | 20:26 | |
| *** dannywilson has joined #openstack-cinder | 20:27 | |
| *** xyang has joined #openstack-cinder | 20:29 | |
| *** dannywilson has quit IRC | 20:31 | |
| *** xyang has quit IRC | 20:31 | |
| *** xyang has joined #openstack-cinder | 20:31 | |
| *** jecarey__ has joined #openstack-cinder | 20:34 | |
| *** junhongl has quit IRC | 20:35 | |
| *** vilobhmm has left #openstack-cinder | 20:37 | |
| *** vilobhmm has joined #openstack-cinder | 20:37 | |
| *** junhongl has joined #openstack-cinder | 20:37 | |
| *** jecarey_ has quit IRC | 20:38 | |
| *** drjones has joined #openstack-cinder | 20:42 | |
| *** _cjones_ has quit IRC | 20:42 | |
| *** lifeless has quit IRC | 20:43 | |
| *** timcl has quit IRC | 20:49 | |
| *** Mandell has quit IRC | 20:51 | |
| *** lifeless has joined #openstack-cinder | 20:51 | |
| *** _cjones_ has joined #openstack-cinder | 20:51 | |
| openstackgerrit | Ivan Kolodyazhny proposed openstack/cinder: Allow archiving deleted rows to shadow tables, for performance. https://review.openstack.org/131182 | 20:52 |
| jgriffith | jungleboyj: I"m seeing all sorts of random failures with oslo_config | 20:52 |
| jgriffith | jungleboyj: this one for example: http://paste.openstack.org/show/160441/ | 20:53 |
| jgriffith | occurs about 50x or so | 20:53 |
| *** drjones has quit IRC | 20:54 | |
| jgriffith | jungleboyj: on master, fresh venv on Ubuntu | 20:54 |
| jgriffith | Ohhhh... hm | 20:54 |
| jgriffith | I thnk I know why.... | 20:55 |
| openstackgerrit | Ivan Kolodyazhny proposed openstack/cinder: Allow archiving deleted rows to shadow tables, for performance. https://review.openstack.org/131182 | 20:55 |
| openstackgerrit | xing-yang proposed openstack/cinder: Add provisioned_capacity https://review.openstack.org/147934 | 20:56 |
| jgriffith | jungleboyj: ok, looks like that's resolved, now I keep hanging on test_storwize_vdisk_copy_ops | 20:56 |
| jgriffith | weird | 20:56 |
| *** _cjones_ has quit IRC | 20:58 | |
| *** _cjones_ has joined #openstack-cinder | 20:58 | |
| *** _cjones_ has quit IRC | 21:00 | |
| *** _cjones_ has joined #openstack-cinder | 21:01 | |
| *** timbyr_ has quit IRC | 21:01 | |
| *** Longgeek has joined #openstack-cinder | 21:02 | |
| *** tbarron has quit IRC | 21:03 | |
| *** Mandell has joined #openstack-cinder | 21:03 | |
| *** dulek has joined #openstack-cinder | 21:05 | |
| *** akerr has quit IRC | 21:06 | |
| *** emagana has quit IRC | 21:06 | |
| *** tbarron has joined #openstack-cinder | 21:06 | |
| *** timbyr_ has joined #openstack-cinder | 21:06 | |
| *** emagana has joined #openstack-cinder | 21:06 | |
| *** dannywilson has joined #openstack-cinder | 21:07 | |
| *** dannywilson has quit IRC | 21:08 | |
| *** _cjones_ has quit IRC | 21:08 | |
| *** dannywilson has joined #openstack-cinder | 21:08 | |
| *** Longgeek has quit IRC | 21:09 | |
| *** _cjones_ has joined #openstack-cinder | 21:10 | |
| *** emagana has quit IRC | 21:11 | |
| *** annashen has quit IRC | 21:17 | |
| *** annashen has joined #openstack-cinder | 21:17 | |
| *** thingee has joined #openstack-cinder | 21:17 | |
| *** yrabl has joined #openstack-cinder | 21:21 | |
| *** annashen has quit IRC | 21:21 | |
| *** emagana has joined #openstack-cinder | 21:24 | |
| *** tellesnobrega_ has joined #openstack-cinder | 21:24 | |
| *** dulek has quit IRC | 21:26 | |
| *** Mandell has quit IRC | 21:27 | |
| e0ne | thingee: hi Mike! what do you think about adding linke to irc logs(http://eavesdrop.openstack.org/irclogs/%23openstack-cinder/) to the channel topic? | 21:29 |
| *** dannywilson has quit IRC | 21:30 | |
| jgriffith | jungleboyj: you around? | 21:31 |
| *** akerr has joined #openstack-cinder | 21:31 | |
| *** fifieldt__ has quit IRC | 21:37 | |
| *** fifieldt__ has joined #openstack-cinder | 21:39 | |
| *** annashen has joined #openstack-cinder | 21:39 | |
| *** HT_sergio has joined #openstack-cinder | 21:45 | |
| thingee | e0ne: sounds good to me | 21:47 |
| e0ne | thingee: :). could you add it, please? | 21:47 |
| jgriffith | ohhh... interesting, the mock out of sleep in my unit tests is breaking the usage in the storwize test. Probably a cleanup issue on my side. | 21:48 |
| *** lcurtis has quit IRC | 21:49 | |
| e0ne | jgriffith: it's not a fair unit test... | 21:49 |
| jgriffith | e0ne: how do you mean? | 21:50 |
| smcginnis | Gate queue is empty? Hmm.. | 21:50 |
| jgriffith | hmm... damn storwize test still hangs | 21:51 |
| jgriffith | oh... no it doesn't, it's just really slow :( | 21:51 |
| e0ne | if unit tests depends on time.sleep - it isn't unit because tested unit is responsible not only for one action | 21:51 |
| jgriffith | e0ne: ahh.... well sometimes | 21:52 |
| *** Mandell has joined #openstack-cinder | 21:52 | |
| e0ne | it's only my imo | 21:52 |
| jgriffith | e0ne: so in this case we're testing logic in methods that use time.sleep | 21:52 |
| jgriffith | e0ne: so in order to test that method; mock out sleep (although no real reason to do so IMO | 21:52 |
| jgriffith | ) | 21:52 |
| e0ne | i see real unit tests in practice only once.. unit+light integration/functional as in cinder and other openstack project - is goog enough | 21:53 |
| e0ne | jgriffith: imho, in theory if you mock sleep - it shouln't break unit tests. | 21:54 |
| jgriffith | e0ne: ahhh... YES I completely agree there | 21:54 |
| jgriffith | e0ne: as you know that's along the lines of my big complaint the last few weeks about our unit tests | 21:54 |
| jgriffith | they're pretty horrible | 21:54 |
| *** lpetrut has quit IRC | 21:55 | |
| jgriffith | by definition of "unit tests" | 21:55 |
| e0ne | jgriffith: absolutely agree with you | 21:55 |
| jgriffith | e0ne: I see now what you're saying is agreement with that :) | 21:55 |
| jgriffith | e0ne: the storwize test inparticular is doing some very *odd* things IMO | 21:55 |
| jgriffith | anyway.... seeing if I can unstick the mess | 21:55 |
| e0ne | jgriffith: also you wanted to create some action items or roadmap how we will make our tests better | 21:56 |
| jgriffith | e0ne: yeah... sadly I haven't done that :( | 21:56 |
| e0ne | :( | 21:56 |
| jgriffith | e0ne: and honestly the best suggestion I have to start is agreed upon methodologies and standards | 21:57 |
| jgriffith | and publish them | 21:57 |
| jgriffith | that's what I'd like to start with | 21:57 |
| jgriffith | I'd like to get some traction on that next week and get something written up | 21:57 |
| e0ne | jgriffith: i like this idea. maybe it should be done cross all openstack projects | 21:57 |
| jgriffith | e0ne: yeah, that would be good; we could start with the work jaypipes sdague and company did on the OpenStack Hour presentation | 21:58 |
| *** jaypipes has quit IRC | 21:59 | |
| e0ne | jgriffith: sounds good to me | 21:59 |
| jgriffith | haha... jay hated the idea so much he just "quit" | 22:00 |
| jgriffith | :) | 22:00 |
| e0ne | :) | 22:00 |
| e0ne | jgriffith: something like it https://etherpad.openstack.org/p/kilo-crossproject-ha-integration but for test would be good | 22:00 |
| smcginnis | jgriffith, e0ne: Want to get an official topic on the agenda for unit test guidelines> | 22:00 |
| smcginnis | https://etherpad.openstack.org/p/cinder-kilo-midcycle-meetup | 22:01 |
| e0ne | jgriffith: and we need to remember, that some projects decided on the last summit to move some tests from tempest inside them | 22:01 |
| *** diemt has quit IRC | 22:01 | |
| e0ne | smcginnis: i want to try steaks in Texas.. but i won't be there next week:( | 22:03 |
| *** dustins has quit IRC | 22:04 | |
| jgriffith | e0ne: yes, I'm planning to get that moving after we get some bugs fixed :) | 22:04 |
| jgriffith | smcginnis: yeah I should do that :) | 22:04 |
| smcginnis | e0ne: Definitely a good place for carnivores. :) | 22:04 |
| smcginnis | Chopped brisket at Rudy's is the best. | 22:05 |
| e0ne | :) | 22:05 |
| jgriffith | mmmm.... brisket!!! | 22:06 |
| *** akerr has quit IRC | 22:07 | |
| *** patrickeast_ has joined #openstack-cinder | 22:07 | |
| *** bswartz has quit IRC | 22:07 | |
| *** emagana has quit IRC | 22:07 | |
| *** patrickeast has quit IRC | 22:08 | |
| *** patrickeast_ is now known as patrickeast | 22:08 | |
| *** emagana has joined #openstack-cinder | 22:08 | |
| *** emagana has quit IRC | 22:09 | |
| *** emagana has joined #openstack-cinder | 22:09 | |
| *** mudassirlatif has quit IRC | 22:10 | |
| *** mudassirlatif has joined #openstack-cinder | 22:10 | |
| *** tbarron has quit IRC | 22:14 | |
| *** emagana has quit IRC | 22:14 | |
| *** emagana has joined #openstack-cinder | 22:15 | |
| *** alwaysatthenoc has quit IRC | 22:16 | |
| *** jungleboyj has quit IRC | 22:16 | |
| *** fischerw has quit IRC | 22:17 | |
| xyang | jgriffith: can you take a look of this again: https://review.openstack.org/#/c/147934/5/cinder/brick/local_dev/lvm.py | 22:17 |
| *** emagana has quit IRC | 22:20 | |
| openstackgerrit | Mikhail Khodos proposed openstack/cinder: Fix Nexenta NFS driver mounts https://review.openstack.org/149380 | 22:25 |
| *** thangp has quit IRC | 22:25 | |
| openstackgerrit | Julie Gravel proposed openstack/cinder-specs: Validate volume type extra specs https://review.openstack.org/131280 | 22:25 |
| *** Yogi1 has quit IRC | 22:28 | |
| *** emagana has joined #openstack-cinder | 22:31 | |
| *** smoriya has joined #openstack-cinder | 22:36 | |
| *** emagana has quit IRC | 22:36 | |
| openstackgerrit | Richard Hagarty proposed openstack/cinder: HP 3PAR modules do not follow coding guidelines https://review.openstack.org/149373 | 22:37 |
| *** jungleboyj has joined #openstack-cinder | 22:39 | |
| *** julim has quit IRC | 22:41 | |
| *** emagana has joined #openstack-cinder | 22:45 | |
| *** xyang has quit IRC | 22:46 | |
| *** thingee has quit IRC | 22:48 | |
| openstackgerrit | Ivan Kolodyazhny proposed openstack/cinder: Allow archiving deleted rows to shadow tables, for performance https://review.openstack.org/131182 | 22:50 |
| *** emagana has quit IRC | 22:50 | |
| *** kaufer has quit IRC | 22:54 | |
| *** Longgeek has joined #openstack-cinder | 22:55 | |
| *** alexpilotti has quit IRC | 22:55 | |
| *** xyang has joined #openstack-cinder | 22:56 | |
| *** xyang has quit IRC | 22:57 | |
| *** smoriya has quit IRC | 23:01 | |
| *** jamielennox|away is now known as jamielennox | 23:01 | |
| *** Longgeek has quit IRC | 23:01 | |
| openstackgerrit | Julie Gravel proposed openstack/cinder-specs: Validate volume type extra specs https://review.openstack.org/131280 | 23:03 |
| *** mudassirlatif has quit IRC | 23:04 | |
| *** annashen has quit IRC | 23:05 | |
| openstackgerrit | Jay Bryant proposed openstack/cinder: Move oslo.serialization to oslo_serialization namespace https://review.openstack.org/149426 | 23:05 |
| *** mudassirlatif has joined #openstack-cinder | 23:05 | |
| *** jaypipes has joined #openstack-cinder | 23:06 | |
| *** annashen has joined #openstack-cinder | 23:06 | |
| *** e0ne has quit IRC | 23:07 | |
| *** annashen has quit IRC | 23:10 | |
| *** Yogi11 has joined #openstack-cinder | 23:10 | |
| *** jaypipes has quit IRC | 23:10 | |
| openstackgerrit | Joshua Harlow proposed openstack/cinder: Remove usage of taskflow 'utils.misc' module https://review.openstack.org/149374 | 23:13 |
| openstackgerrit | John Griffith proposed openstack/cinder: Add a generic retry decorator to cinder/utils https://review.openstack.org/149341 | 23:14 |
| openstackgerrit | John Griffith proposed openstack/cinder: Add retry to lvm snapshot create https://review.openstack.org/149360 | 23:14 |
| *** Yogi11 has quit IRC | 23:16 | |
| *** Yogi11 has joined #openstack-cinder | 23:16 | |
| *** annashen has joined #openstack-cinder | 23:17 | |
| *** jaypipes has joined #openstack-cinder | 23:22 | |
| *** thingee has joined #openstack-cinder | 23:23 | |
| *** emagana has joined #openstack-cinder | 23:25 | |
| anish | if the underlying code that my change is based on is constantly changing, is it expected that I should rebase all the time, or wait for review to finish and then rebase later ? | 23:29 |
| *** tellesnobrega_ has quit IRC | 23:31 | |
| *** jcru has quit IRC | 23:34 | |
| *** jasondotstar has joined #openstack-cinder | 23:34 | |
| *** vilobhmm has quit IRC | 23:35 | |
| *** vilobhmm has joined #openstack-cinder | 23:35 | |
| openstackgerrit | Gary W. Smith proposed openstack/cinder: Add 3PAR extra specs schema into volume stats https://review.openstack.org/147005 | 23:36 |
| *** crose has quit IRC | 23:42 | |
| jgriffith | jungleboyj: any plan to move to oslo.log? | 23:45 |
| *** tbarron has joined #openstack-cinder | 23:45 | |
| jgriffith | jungleboyj: which actually, I guess needs oslo_log conversion :( | 23:45 |
| jgriffith | good thing we're not using it yet I suppose | 23:45 |
| jgriffith | jungleboyj: I pushed changes to olso.log that I want to now use in cinder... | 23:45 |
| jgriffith | jungleboyj: but don't want to rework for incubator if we have any plans to move to the lib any time in the near future | 23:46 |
| jgriffith | ahh... that's just the internal namespace | 23:48 |
| jgriffith | crickets... chirp chirp, chirp chirp | 23:49 |
| *** jecarey__ has quit IRC | 23:49 | |
| jgriffith | anish: rebase depends | 23:49 |
| *** vilobhmm_ has joined #openstack-cinder | 23:49 | |
| jgriffith | anish: rebase will auto work for you in a lot of cases | 23:49 |
| jgriffith | or at some point it's going to fail and you can just fix it then | 23:50 |
| jgriffith | anish: that of course assumes your not doing something like "moving" a file or something like that | 23:50 |
| jgriffith | anish: in other words, most of the time you don't have to worry about it unless Gerrit tells you you do | 23:50 |
| jgriffith | exception cases are moving methods to a new location etc | 23:51 |
| *** thingee has quit IRC | 23:51 | |
| *** vilobhmm has quit IRC | 23:52 | |
| *** vilobhmm_ is now known as vilobhmm | 23:52 | |
| anish | I'm adding a new file, so I really doubt rebase is going to fail there. but I do add unit tests to an existing file, so just checking on that | 23:52 |
| anish | plus the pull I did right now was massive | 23:52 |
| anish | jgriffith: thanks for confirming | 23:53 |
| *** david-lyle is now known as david-lyle_afk | 23:56 | |
| *** IanGovett has joined #openstack-cinder | 23:56 | |
| *** vilobhmm_ has joined #openstack-cinder | 23:57 | |
| *** esker has quit IRC | 23:57 | |
| *** vilobhmm has quit IRC | 23:59 | |
| *** vilobhmm_ is now known as vilobhmm | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!