| *** IanGovett has quit IRC | 00:00 | |
| *** scottda_ has joined #openstack-cinder | 00:06 | |
| *** bswartz has joined #openstack-cinder | 00:08 | |
| *** annashen has quit IRC | 00:14 | |
| *** takedakn has joined #openstack-cinder | 00:14 | |
| *** EmilienM is now known as EmilienM|afk | 00:15 | |
| *** zhithuang has joined #openstack-cinder | 00:16 | |
| *** zhithuang is now known as winston-d_ | 00:16 | |
| *** scottda_ has quit IRC | 00:17 | |
| *** takedakn has quit IRC | 00:25 | |
| *** david-lyle has joined #openstack-cinder | 00:25 | |
| *** tbarron has joined #openstack-cinder | 00:26 | |
| *** esker has joined #openstack-cinder | 00:27 | |
| *** crose has quit IRC | 00:29 | |
| *** scottda_ has joined #openstack-cinder | 00:29 | |
| *** tbarron has quit IRC | 00:29 | |
| *** tbarron has joined #openstack-cinder | 00:31 | |
| *** esker has quit IRC | 00:31 | |
| *** scottda_ has quit IRC | 00:32 | |
| *** takedakn has joined #openstack-cinder | 00:33 | |
| *** annashen has joined #openstack-cinder | 00:37 | |
| *** sbadia has joined #openstack-cinder | 00:37 | |
| *** hypnosb_ has quit IRC | 00:37 | |
| *** ho has joined #openstack-cinder | 00:38 | |
| *** ebalduf has joined #openstack-cinder | 00:40 | |
| *** mwichmann has quit IRC | 00:40 | |
| *** rodrigod` is now known as rodrigods | 00:41 | |
| hemna | http://paste.openstack.org/show/158192/ | 00:44 |
|---|---|---|
| hemna | multi-attach is working again with nova | 00:44 |
| hemna | :) | 00:44 |
| hemna | you just can't detach yet :P | 00:45 |
| hemna | ok time to head home | 00:45 |
| *** hemna is now known as hemnafk | 00:45 | |
| *** takedakn has quit IRC | 00:45 | |
| *** ebalduf has quit IRC | 00:46 | |
| *** Yogi11 has joined #openstack-cinder | 00:46 | |
| thingee | brace yourselves | 00:47 |
| gugl3 | thingee, sorry to bother you again, can you take a look at https://review.openstack.org/#/c/140906/ to be or not be...it is in your hand. Thanks | 00:47 |
| thingee | mass email about CI deadline is going out now | 00:47 |
| *** jcru_ has quit IRC | 00:47 | |
| hemnafk | thingee, w00t! | 00:48 |
| *** cppforlife is now known as cppforlife_ | 00:49 | |
| *** tellesnobrega_ has joined #openstack-cinder | 00:50 | |
| *** emagana has quit IRC | 00:53 | |
| *** emagana has joined #openstack-cinder | 00:53 | |
| *** annashen has quit IRC | 00:54 | |
| *** beecee has joined #openstack-cinder | 00:55 | |
| *** dannywilson has quit IRC | 00:57 | |
| *** emagana has quit IRC | 00:58 | |
| *** anish_ is now known as anish | 01:02 | |
| *** anish has quit IRC | 01:02 | |
| *** anish has joined #openstack-cinder | 01:02 | |
| *** david-lyle has quit IRC | 01:08 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/cinder: Updated from global requirements https://review.openstack.org/147719 | 01:10 |
| *** tsekiyama has joined #openstack-cinder | 01:16 | |
| *** mtanino has quit IRC | 01:16 | |
| *** tsekiyam_ has quit IRC | 01:19 | |
| openstackgerrit | Bob proposed openstack/cinder: Fix product support https://review.openstack.org/147720 | 01:19 |
| *** tsekiyama has quit IRC | 01:20 | |
| *** ryanpetrello has joined #openstack-cinder | 01:22 | |
| *** david-lyle has joined #openstack-cinder | 01:26 | |
| *** DaveChen has joined #openstack-cinder | 01:28 | |
| *** boris-42 has quit IRC | 01:33 | |
| *** rmesta has quit IRC | 01:34 | |
| *** openstack has joined #openstack-cinder | 01:35 | |
| *** ryanpetrello has quit IRC | 01:35 | |
| *** ebalduf has joined #openstack-cinder | 01:41 | |
| openstackgerrit | wanghao proposed openstack/cinder: Implement function of import/export snapshots https://review.openstack.org/144590 | 01:44 |
| *** ebalduf has quit IRC | 01:47 | |
| *** david-lyle has quit IRC | 02:01 | |
| openstackgerrit | Dave Chen proposed openstack/cinder: set/unset volume image metadata https://review.openstack.org/147726 | 02:04 |
| *** mudassirlatif has quit IRC | 02:17 | |
| *** takedakn has joined #openstack-cinder | 02:25 | |
| *** takedakn has quit IRC | 02:29 | |
| *** mberlin1 has joined #openstack-cinder | 02:32 | |
| *** _cjones_ has quit IRC | 02:33 | |
| *** mberlin has quit IRC | 02:35 | |
| *** tellesnobrega_ has quit IRC | 02:38 | |
| *** takedakn has joined #openstack-cinder | 02:39 | |
| *** patrickeast has quit IRC | 02:39 | |
| *** david-lyle has joined #openstack-cinder | 02:42 | |
| *** david-lyle has quit IRC | 02:42 | |
| *** Yogi11 has quit IRC | 02:42 | |
| *** takedakn has quit IRC | 02:44 | |
| *** Yogi1 has joined #openstack-cinder | 02:46 | |
| *** ebalduf has joined #openstack-cinder | 02:47 | |
| *** ebalduf has quit IRC | 02:53 | |
| *** harlowja is now known as harlowja_away | 02:57 | |
| openstackgerrit | Vincent Hou proposed openstack/cinder: Remove the useless next link for volumes, transfers and backups https://review.openstack.org/131915 | 02:58 |
| openstackgerrit | wanghao proposed openstack/cinder: Implement function of import/export snapshots https://review.openstack.org/144590 | 03:07 |
| *** Longgeek has joined #openstack-cinder | 03:08 | |
| *** annashen has joined #openstack-cinder | 03:11 | |
| *** vukcrni has joined #openstack-cinder | 03:12 | |
| *** ebalduf has joined #openstack-cinder | 03:21 | |
| *** scottda_ has joined #openstack-cinder | 03:21 | |
| *** ebalduf has quit IRC | 03:26 | |
| *** thingee has quit IRC | 03:39 | |
| *** Longgeek has quit IRC | 03:41 | |
| *** winston-d_ has quit IRC | 03:42 | |
| *** Longgeek has joined #openstack-cinder | 03:42 | |
| openstackgerrit | wanghao proposed openstack/cinder: query volume detail support volume_glance_metadata https://review.openstack.org/147738 | 03:43 |
| *** winston-d_ has joined #openstack-cinder | 03:45 | |
| *** tbarron has quit IRC | 03:46 | |
| *** scottda_ has quit IRC | 03:50 | |
| *** ebalduf has joined #openstack-cinder | 03:56 | |
| *** Longgeek has quit IRC | 04:05 | |
| *** Longgeek has joined #openstack-cinder | 04:07 | |
| anish | jgriffith: hemnafk I have a quick question or two regarding targets if you have time | 04:09 |
| *** rajinir has joined #openstack-cinder | 04:09 | |
| *** rushiagr_away is now known as rushiagr | 04:11 | |
| *** rajinir has quit IRC | 04:11 | |
| openstackgerrit | xing-yang proposed openstack/cinder: Verify the instance's existance in the VMAX driver https://review.openstack.org/140913 | 04:14 |
| *** fifieldt__ has joined #openstack-cinder | 04:17 | |
| *** fandi_ has quit IRC | 04:21 | |
| *** fandi_ has joined #openstack-cinder | 04:21 | |
| *** winston-d_ has quit IRC | 04:26 | |
| *** winston-d_ has joined #openstack-cinder | 04:29 | |
| *** xyang1_away has quit IRC | 04:37 | |
| openstackgerrit | Vincent Hou proposed openstack/cinder: Remove the useless next link for volumes, transfers and backups https://review.openstack.org/131915 | 04:39 |
| *** annashen has quit IRC | 04:41 | |
| *** harlowja_at_home has joined #openstack-cinder | 04:42 | |
| *** boris-42 has joined #openstack-cinder | 04:43 | |
| rushiagr | Deleting a volume by name, using cinderclient, seems a potentially very expensive operation | 04:51 |
| rushiagr | it first gets a list of ALL the volumes for the current user, then from the response finds the UUID of the vol, and then deletes it | 04:52 |
| *** winston-d_ has quit IRC | 04:52 | |
| rushiagr | is there any interest in supporting accessing volumes directly by their name in the API? | 04:53 |
| *** ebalduf has quit IRC | 04:55 | |
| *** harlowja_at_home has quit IRC | 04:59 | |
| *** Yogi1 has quit IRC | 05:03 | |
| *** ryanpetrello has joined #openstack-cinder | 05:13 | |
| *** annashen has joined #openstack-cinder | 05:17 | |
| *** ryanpetrello has quit IRC | 05:24 | |
| *** vbala has joined #openstack-cinder | 05:51 | |
| *** Mandell has quit IRC | 05:52 | |
| *** Mandell has joined #openstack-cinder | 05:53 | |
| *** takedakn has joined #openstack-cinder | 05:53 | |
| *** ebalduf has joined #openstack-cinder | 05:56 | |
| *** ebalduf has quit IRC | 06:00 | |
| *** thingee has joined #openstack-cinder | 06:03 | |
| *** rushiagr is now known as rushiagr_away | 06:05 | |
| *** takedakn has quit IRC | 06:06 | |
| *** rushiagr_away is now known as rushiagr | 06:08 | |
| *** lpetrut has joined #openstack-cinder | 06:09 | |
| openstackgerrit | Anish Bhatt proposed openstack/cinder: [WIP] Add support for chiscsi iscsi helper https://review.openstack.org/147758 | 06:15 |
| rushiagr | boris-42: hi | 06:16 |
| boris-42 | rushiagr: hi there | 06:17 |
| anish | Anyone know if the CloudFounders OpenvStorage CI is misbehaving ? I got a merge failure barely a second after uploading | 06:17 |
| rushiagr | boris-42: I was curious if this bug is covered in some cinder rally job https://bugs.launchpad.net/cinder/+bug/1250696 | 06:17 |
| rushiagr | boris-42: (if there is one) | 06:17 |
| rushiagr | anish: yes, it is | 06:17 |
| anish | thingee: I belive you were concerned about the implementation of https://review.openstack.org/147758. It is done, only missing tests | 06:17 |
| anish | rushiagr: thanks | 06:17 |
| anish | Anyone know how you are supposed to implement unit tests for optional components ? | 06:19 |
| anish | My code implementation is for an iscsi target implementation that will almost never be installed by default or available via repos | 06:20 |
| anish | not sure how tests for those kinda things are written | 06:20 |
| rushiagr | anish: I think if there is code, there should be unit tests. | 06:21 |
| anish | yes | 06:21 |
| boris-42 | anish: rushiagr so current job | 06:21 |
| boris-42 | that is cinder | 06:21 |
| boris-42 | has only fake virt driver | 06:21 |
| *** anshul has joined #openstack-cinder | 06:21 | |
| boris-42 | so someday we will finish effort related to adding fake driver (that adds 1 MB disks) | 06:22 |
| boris-42 | as far as I remember jgriffith was working on it | 06:22 |
| rushiagr | but if your driver is just calling an external thing, with no conditionals, exceptions etc, then yes, it would be weird to just write something which doesn't actually test anything | 06:22 |
| anish | at best I can verify that the command line is being constructed correctly | 06:22 |
| anish | (which is my plan right now) | 06:22 |
| rushiagr | boris-42: the large-ops equivalent is fake virt (vol) driver I think | 06:23 |
| rushiagr | boris-42: so fake driver (1 MB disks) falls outside the scope of it I guess, no? | 06:23 |
| boris-42 | rushiagr: why so? | 06:24 |
| boris-42 | rushiagr: that is even better | 06:24 |
| boris-42 | rushiagr: cause we can create close to any amount of disks in gates | 06:25 |
| rushiagr | boris-42: yes, obviously better | 06:25 |
| boris-42 | I think 20k volumes is enough for large-ops purpose) | 06:25 |
| boris-42 | but as far as I know it's not set by argument of devstack-gate | 06:25 |
| boris-42 | https://github.com/openstack-infra/project-config/blob/master/jenkins/jobs/rally.yaml#L56 | 06:25 |
| boris-42 | I'll need to test it * | 06:25 |
| *** ebalduf has joined #openstack-cinder | 06:26 | |
| rushiagr | boris-42: I was just talking from the perspective of that bug. It says equivalent of nova large-ops, which I think is completely fake (only db entries, no instances) | 06:26 |
| boris-42 | rushiagr: so I don't think that it was the idea* | 06:27 |
| boris-42 | rushiagr: but maybe* | 06:27 |
| boris-42 | rushiagr: from my perspective it's enough to have 1MB disks=) | 06:28 |
| boris-42 | rushiagr: we need just to create proper job | 06:28 |
| boris-42 | +) | 06:28 |
| boris-42 | rushiagr: and replace current with it | 06:28 |
| rushiagr | boris-42: okay. I was just gathering information. I have no objections to any of what you said :) | 06:28 |
| boris-42 | so much to do... | 06:28 |
| boris-42 | rushiagr: btw what was your interest in this topic? | 06:29 |
| boris-42 | rushiagr: you working on perf stuff? | 06:29 |
| rushiagr | boris-42: I wrote a small patch some time back which creates 1k vols https://github.com/rushiagr/cinder/commit/7be5ae1931dd036419f3620f0b8b246619a59b2f | 06:29 |
| rushiagr | boris-42: no. I was just going through the bug list, found this bug, and I know you work on perf stuff, so wanted to know the state of things.. | 06:30 |
| *** ebalduf has quit IRC | 06:30 | |
| boris-42 | ah ok | 06:31 |
| anish | I'm surprised 1K volumes don't make the system go kaput | 06:31 |
| anish | that's like no space at all | 06:31 |
| boris-42 | anish: nope 1K ok | 06:31 |
| boris-42 | anish: 10k makes bad things | 06:31 |
| rushiagr | anish: ah, sorry. I did 1m :) | 06:31 |
| anish | aah, much more sensible | 06:32 |
| boris-42 | rushiagr: anish this is crazy bug https://review.openstack.org/#/c/147756/ btw | 06:32 |
| rushiagr | anish: I vaguely remember I had some issue creating volume less than 4k. | 06:32 |
| boris-42 | rushiagr: anish https://review.openstack.org/#/c/147751/ =) I got message No valid host after that patch when trying to boot vm from volume | 06:32 |
| boris-42 | rushiagr: why you need that lol?) | 06:33 |
| anish | you should get an award for the smalles possible changeset | 06:33 |
| anish | well, maybe replace a with . | 06:33 |
| rushiagr | boris-42: :) well, I just wanted to create too many volumes too fast. Poor man's perf test :P | 06:33 |
| boris-42 | rushiagr: =) | 06:33 |
| boris-42 | rushiagr: heh I hope someday I will finish rally and start actually working on thing that was the root reason of writing rally=) | 06:35 |
| rushiagr | boris-42: so currently in the gate, there is a job running which creates tons of volumes (which are actually just tons of db entries). Am I right? | 06:35 |
| rushiagr | boris-42: I hope you reach there sooner :) | 06:35 |
| boris-42 | rushiagr: nope there is no such job as far as I know (there is job only that creates tons of vms (actually 150) by one request) | 06:35 |
| boris-42 | rushiagr: so what we need is to make change in devstack-gate to make large-ops set fake volumes as well | 06:36 |
| rushiagr | boris-42: ohh | 06:36 |
| boris-42 | rushiagr: hm heat is using fakevirt (maybe it's fake fully) | 06:36 |
| boris-42 | let me check | 06:36 |
| rushiagr | boris-42: I remember you had some patches in infra regarding this | 06:36 |
| boris-42 | rushiagr: they dislike my names lol | 06:36 |
| rushiagr | boris-42: that was me who started the name thing :/ | 06:37 |
| rushiagr | :) | 06:37 |
| boris-42 | haha | 06:37 |
| boris-42 | it happens=) | 06:37 |
| boris-42 | and then I didn't have enough time for it | 06:37 |
| rushiagr | boris-42: okay. That's exactly what I wanted to ask. If that job is done yet or not. I think I have my answer and clarity now :) | 06:38 |
| *** anshul has quit IRC | 06:38 | |
| rushiagr | boris-42: thanks :) | 06:38 |
| boris-42 | rushiagr: not yet | 06:40 |
| boris-42 | rushiagr: https://review.openstack.org/#/c/147762/ this is test | 06:44 |
| boris-42 | rushiagr: if we can create 1k volumes in fakevirt job | 06:44 |
| *** takedakn has joined #openstack-cinder | 06:44 | |
| boris-42 | rushiagr: if so I need just to create one with proper name "fake" and replace job in heat | 06:45 |
| *** takedakn has quit IRC | 06:45 | |
| *** takedakn has joined #openstack-cinder | 06:46 | |
| rushiagr | boris-42: okay, that makes sense | 06:46 |
| boris-42 | rushiagr: if no then we need to patch devatack-gate large ops value to set fake dirver for cinder* | 06:46 |
| *** takedakn has quit IRC | 06:50 | |
| *** thingee has quit IRC | 06:55 | |
| *** annashen has quit IRC | 06:57 | |
| *** annashen has joined #openstack-cinder | 07:03 | |
| *** lan_ has joined #openstack-cinder | 07:17 | |
| *** Longgeek has quit IRC | 07:18 | |
| *** chlong has quit IRC | 07:20 | |
| *** Longgeek has joined #openstack-cinder | 07:20 | |
| *** ebalduf has joined #openstack-cinder | 07:31 | |
| *** annashen has quit IRC | 07:33 | |
| *** flip214_ is now known as flip214 | 07:35 | |
| *** flip214 has joined #openstack-cinder | 07:35 | |
| *** ebalduf has quit IRC | 07:36 | |
| *** nkrinner has joined #openstack-cinder | 07:38 | |
| *** jamielennox is now known as jamielennox|away | 07:42 | |
| *** vukcrni has quit IRC | 07:44 | |
| openstackgerrit | Dave Chen proposed openstack/cinder: set/unset volume image metadata https://review.openstack.org/147726 | 07:45 |
| *** vukcrni has joined #openstack-cinder | 07:45 | |
| openstackgerrit | TaoBai proposed openstack/cinder: IBM Storwize driver Consistency Group Implementation https://review.openstack.org/142683 | 07:46 |
| openstackgerrit | Eduard Biceri-Matei proposed openstack/cinder: New Cinder volume driver for openvstorage https://review.openstack.org/130733 | 07:49 |
| openstackgerrit | Dave Chen proposed openstack/cinder: set/unset volume image metadata https://review.openstack.org/147726 | 07:50 |
| *** Miouge has joined #openstack-cinder | 07:52 | |
| *** TobiasE has joined #openstack-cinder | 07:59 | |
| *** teju has joined #openstack-cinder | 08:01 | |
| *** oomichi has quit IRC | 08:18 | |
| *** vukcrni has quit IRC | 08:21 | |
| *** lpetrut has quit IRC | 08:21 | |
| *** cburgess_ has quit IRC | 08:21 | |
| *** morganfainberg has quit IRC | 08:21 | |
| *** morganfainberg has joined #openstack-cinder | 08:21 | |
| *** cburgess has joined #openstack-cinder | 08:23 | |
| *** vukcrni has joined #openstack-cinder | 08:25 | |
| *** deepakcs has joined #openstack-cinder | 08:30 | |
| *** ebalduf has joined #openstack-cinder | 08:33 | |
| *** ebalduf has quit IRC | 08:38 | |
| *** rwsu has quit IRC | 08:40 | |
| *** anshul has joined #openstack-cinder | 08:40 | |
| *** afazekas_ has joined #openstack-cinder | 08:40 | |
| anish | anyone know how to force a recheck on the netapp ci ? | 08:44 |
| *** nellysmitt has joined #openstack-cinder | 08:48 | |
| boris-42 | rushiagr: http://logs.openstack.org/62/147762/2/experimental/gate-rally-dsvm-fakevirt-heat/a1f0eda/rally-plot/results.html.gz#/CinderVolumes.create_volume/overview | 08:50 |
| boris-42 | rushiagr: this means that large-ops arguments in devstack-gate doesn't set fake cinder driver | 08:50 |
| boris-42 | rushiagr: heh a lot of work =( | 08:50 |
| anish | am I reading that really wrong, coz it's seems like 2% success | 08:51 |
| *** pradipta_away is now known as pradipta | 08:53 | |
| openstackgerrit | Vipin Balachandran proposed openstack/cinder: VMware: Integrate VMDK driver with oslo.vmware https://review.openstack.org/137324 | 08:55 |
| openstackgerrit | Vipin Balachandran proposed openstack/cinder: VMware: Support for non-default port https://review.openstack.org/146384 | 08:58 |
| *** coolsvap|afk is now known as coolsvap | 09:03 | |
| *** jpich has joined #openstack-cinder | 09:06 | |
| openstackgerrit | Eduard Biceri-Matei proposed openstack/cinder: New Cinder volume driver for openvstorage https://review.openstack.org/130733 | 09:07 |
| *** deepakcs has quit IRC | 09:11 | |
| *** pawels has joined #openstack-cinder | 09:13 | |
| openstackgerrit | Julien Danjou proposed openstack/cinder: tests: replace mox by mox3, clean out mox usage https://review.openstack.org/147476 | 09:16 |
| *** jordanP has joined #openstack-cinder | 09:17 | |
| *** Mandell has quit IRC | 09:18 | |
| *** jistr has joined #openstack-cinder | 09:19 | |
| *** takedakn has joined #openstack-cinder | 09:20 | |
| *** dulek has joined #openstack-cinder | 09:20 | |
| *** pawels1 has joined #openstack-cinder | 09:28 | |
| openstackgerrit | wanghao proposed openstack/cinder: Implement function of import/export snapshots https://review.openstack.org/144590 | 09:28 |
| *** pawels has quit IRC | 09:30 | |
| *** ebalduf has joined #openstack-cinder | 09:34 | |
| openstackgerrit | Julien Danjou proposed openstack/cinder: tests: replace mox by mox3, clean out mox usage https://review.openstack.org/147476 | 09:34 |
| *** ebalduf has quit IRC | 09:38 | |
| *** ndipanov_gone is now known as ndipanov | 09:43 | |
| *** ndipanov has joined #openstack-cinder | 09:43 | |
| *** david-lyle has joined #openstack-cinder | 09:46 | |
| openstackgerrit | Julien Danjou proposed openstack/cinder: tests: replace mox by mox3, clean out mox usage https://review.openstack.org/147476 | 09:47 |
| *** Longgeek has quit IRC | 09:48 | |
| *** Longgeek has joined #openstack-cinder | 09:48 | |
| *** Longgeek_ has joined #openstack-cinder | 09:52 | |
| *** pawels1 has quit IRC | 09:55 | |
| *** Longgeek has quit IRC | 09:55 | |
| *** Longgeek_ has quit IRC | 09:56 | |
| *** Longgeek has joined #openstack-cinder | 09:56 | |
| *** deepakcs has joined #openstack-cinder | 10:00 | |
| *** e0ne has joined #openstack-cinder | 10:07 | |
| *** david-lyle has quit IRC | 10:08 | |
| *** takedakn has quit IRC | 10:08 | |
| openstackgerrit | Eduard Biceri-Matei proposed openstack/cinder: New Cinder volume driver for openvstorage https://review.openstack.org/130733 | 10:16 |
| openstackgerrit | Nilesh Bhosale proposed openstack/cinder: Set 'driver_volume_type' to 'gpfs' https://review.openstack.org/147802 | 10:17 |
| *** avishay_ has joined #openstack-cinder | 10:20 | |
| *** Longgeek_ has joined #openstack-cinder | 10:25 | |
| *** Longgeek_ has quit IRC | 10:27 | |
| openstackgerrit | Trung Trinh proposed openstack/cinder: Delete volume attached to already-deleted VM https://review.openstack.org/142006 | 10:27 |
| *** Longgeek_ has joined #openstack-cinder | 10:27 | |
| *** Longgeek has quit IRC | 10:28 | |
| *** e0ne is now known as e0ne_ | 10:28 | |
| openstackgerrit | Trung Trinh proposed openstack/cinder: Delete volume attached to already-deleted VM https://review.openstack.org/142006 | 10:33 |
| *** ebalduf has joined #openstack-cinder | 10:34 | |
| *** karimb has joined #openstack-cinder | 10:35 | |
| *** Longgeek has joined #openstack-cinder | 10:36 | |
| *** Longgeek_ has quit IRC | 10:38 | |
| *** e0ne_ has quit IRC | 10:38 | |
| *** ebalduf has quit IRC | 10:39 | |
| *** Masahiro has quit IRC | 10:48 | |
| *** avishay_ has quit IRC | 10:50 | |
| openstackgerrit | Julien Danjou proposed openstack/cinder: tests: replace mox by mox3, clean out mox usage https://review.openstack.org/147476 | 10:50 |
| *** aix has joined #openstack-cinder | 10:54 | |
| *** pawels has joined #openstack-cinder | 10:56 | |
| *** pawels has quit IRC | 10:58 | |
| *** avishay_ has joined #openstack-cinder | 11:02 | |
| *** pawels has joined #openstack-cinder | 11:03 | |
| *** lpetrut has joined #openstack-cinder | 11:04 | |
| *** lpetrut has quit IRC | 11:06 | |
| rushiagr | 16:45:47 <DuncanT> If you code the SQL version, I will benchmark it with my million fake volumes setup if you want, that should answer the scalability / performance concerns | 11:08 |
| rushiagr | DuncanT: I want that setup :) | 11:08 |
| *** karimb has quit IRC | 11:09 | |
| openstackgerrit | Eduard Biceri-Matei proposed openstack/cinder: New Cinder volume driver for openvstorage https://review.openstack.org/130733 | 11:09 |
| *** dulek has quit IRC | 11:13 | |
| *** ho has quit IRC | 11:14 | |
| *** e0ne has joined #openstack-cinder | 11:16 | |
| *** david-lyle has joined #openstack-cinder | 11:17 | |
| *** alexpilotti has joined #openstack-cinder | 11:23 | |
| *** karimb has joined #openstack-cinder | 11:24 | |
| *** sgotliv has joined #openstack-cinder | 11:31 | |
| gtt116__ | hi, any one? | 11:32 |
| gtt116__ | Does somebody deploy Cell with cinder? | 11:33 |
| *** tellesnobrega_ has joined #openstack-cinder | 11:33 | |
| *** ebalduf has joined #openstack-cinder | 11:35 | |
| *** teju has quit IRC | 11:35 | |
| *** Longgeek has quit IRC | 11:37 | |
| *** ebalduf has quit IRC | 11:40 | |
| *** avishay__ has joined #openstack-cinder | 11:46 | |
| *** avishay_ has quit IRC | 11:46 | |
| *** nshaikh has joined #openstack-cinder | 11:46 | |
| *** fandi_ has quit IRC | 11:46 | |
| *** fandi has joined #openstack-cinder | 11:46 | |
| *** avishay__ has quit IRC | 11:47 | |
| *** Masahiro has joined #openstack-cinder | 11:48 | |
| *** tellesnobrega_ has quit IRC | 11:52 | |
| *** Masahiro has quit IRC | 11:53 | |
| *** dulek has joined #openstack-cinder | 12:01 | |
| *** kbyrne has quit IRC | 12:06 | |
| DuncanT | rushiagr: I'll polish up the scripts for it, but it is quite trivial, uses the fake driver so no actual storage | 12:06 |
| DuncanT | rushiagr: It isn't entirely reliable, but has got to a million volumes before, at which point I took a snapshot - it is handy for testing huge volume count DB situations | 12:08 |
| *** kbyrne has joined #openstack-cinder | 12:08 | |
| DuncanT | rushiagr: You could probably just poke volumes into the db directly for a similar effect | 12:08 |
| *** IanGovett has joined #openstack-cinder | 12:09 | |
| *** thebloggu has joined #openstack-cinder | 12:11 | |
| thebloggu | I have a packstack (on fedora 19) installation of Openstack Havana. I'm not able to attach or detach volumes. Here's what I think is relevant of /var/log/cinder/api.log[http://paste.openstack.org/show/158239/] the scheduler log is empty and the the volume.log has nothing relevant. can someone help me? | 12:12 |
| *** lpetrut has joined #openstack-cinder | 12:13 | |
| *** pradipta is now known as pradipta_away | 12:14 | |
| DuncanT | rushiagr: Actually, looks like I'm not using the in-tree fake driver, I had to patch it. I'll clean it all up and shove it on github | 12:14 |
| DuncanT | thebloggu: That is a symptom of the problem. Can you paste up the cinder-volume log please? | 12:15 |
| thebloggu | DuncanT, volume.log's last entry is from a month ago. do you still think it could be useful? | 12:17 |
| DuncanT | thebloggu: Then you've got a bigger problem - what that error message you posted up means is that the response the API thinks it got back from the volume service on how to attach was empty | 12:18 |
| DuncanT | thebloggu: If the volume service didn't receive it, then I'm not sure what could have happened | 12:18 |
| DuncanT | Havana is old code from my PoV now | 12:18 |
| *** lan_ has quit IRC | 12:20 | |
| *** lan_ has joined #openstack-cinder | 12:21 | |
| *** deepakcs has quit IRC | 12:21 | |
| thebloggu | DuncanT, while volume.log is empty there are some messages when I do `service openstack-cinder-volume status`. Here's the output: http://paste.openstack.org/show/158244/ | 12:22 |
| *** aix has quit IRC | 12:23 | |
| DuncanT | thebloggu: It should be printing messages about periodic task stat updates every few minutes I think | 12:24 |
| DuncanT | thebloggu: Try restarting the volume service? | 12:24 |
| thebloggu | DuncanT, same error but there was something written in the volume.log twice: 2015-01-16 12:27:00.292 24301 WARNING cinder.brick.iscsi.iscsi [req-ed588134-762b-4b00-88d6-6e8261963aa4 None None] Attempting recreate of backing lun... | 12:29 |
| *** Longgeek has joined #openstack-cinder | 12:29 | |
| DuncanT | thebloggu: Ok, that sounds like your base problem. Is that the full message? | 12:33 |
| *** timcl has joined #openstack-cinder | 12:33 | |
| thebloggu | DuncanT, yes | 12:34 |
| DuncanT | I haven't even got code that old to hand | 12:35 |
| *** ebalduf has joined #openstack-cinder | 12:36 | |
| DuncanT | Give me a sec | 12:36 |
| thebloggu | DuncanT, ok, thank you | 12:37 |
| DuncanT | So you are seeing the output of LOG.warning(_('Attempting recreate of backing lun...')) | 12:37 |
| DuncanT | But not the output of LOG.debug('StdOut from recreate backing lun: %s' % out) | 12:37 |
| DuncanT | ? | 12:37 |
| thebloggu | DuncanT, yes, I don't see that | 12:38 |
| DuncanT | Ok, run 'ps auxx | grep tgt' on the volume node please | 12:38 |
| thebloggu | DuncanT, http://paste.openstack.org/show/158248/ | 12:39 |
| DuncanT | thebloggu: So what gets run between those two outputs is tgtadm --lld iscsi --op new --mode logicalunit --tid <tid> --lun 1 -b <path> | 12:41 |
| *** ebalduf has quit IRC | 12:41 | |
| DuncanT | That should cause StdOut / StdErr to get printed in the debug log | 12:42 |
| DuncanT | Oh, that might be it, do you have debug turned on in cinder.conf? | 12:42 |
| thebloggu | no, I don't think so and was about to ask you if it could be that :) | 12:42 |
| thebloggu | let me check it | 12:42 |
| DuncanT | Turn on verbose and debug both | 12:42 |
| DuncanT | Then restart the service | 12:43 |
| thebloggu | DuncanT, all of the cinder services or just the volume? | 12:44 |
| DuncanT | I'd run them all with debug if I were you, until it is working | 12:44 |
| *** rushiagr is now known as rushiagr_away | 12:48 | |
| thebloggu | DuncanT, there's so much output so i did `cat /var/log/cinder/volume.log` and the output's here: http://paste.openstack.org/show/158250/ | 12:52 |
| DuncanT | Ok, that looks like it is possibly ok, no idea why it is then stopping though | 12:54 |
| *** Masahiro has joined #openstack-cinder | 12:57 | |
| thebloggu | DuncanT, ok, I think I'll experiment a bit more to see if I can find the problem. thank you very much for your help | 13:00 |
| DuncanT | Good luck. Should be some US folks coming on soon who might be able to spot something I've missed | 13:00 |
| *** vbala has quit IRC | 13:00 | |
| *** esker has joined #openstack-cinder | 13:01 | |
| thebloggu | DuncanT, ok, thank you once again | 13:01 |
| *** smoriya has quit IRC | 13:02 | |
| *** Masahiro has quit IRC | 13:02 | |
| flip214 | "Slowest Tests: math domain error" | 13:02 |
| *** bkopilov has quit IRC | 13:02 | |
| *** Longgeek has quit IRC | 13:04 | |
| *** EmilienM|afk is now known as EmilienM | 13:05 | |
| *** juzuluag has joined #openstack-cinder | 13:07 | |
| *** afazekas_ has quit IRC | 13:07 | |
| *** Longgeek has joined #openstack-cinder | 13:11 | |
| *** david-lyle has quit IRC | 13:11 | |
| *** bswartz has quit IRC | 13:14 | |
| *** lpabon has joined #openstack-cinder | 13:17 | |
| *** aix has joined #openstack-cinder | 13:18 | |
| *** Mandell has joined #openstack-cinder | 13:18 | |
| *** afazekas_ has joined #openstack-cinder | 13:21 | |
| *** jcru has joined #openstack-cinder | 13:22 | |
| *** nellysmitt has quit IRC | 13:22 | |
| *** Mandell has quit IRC | 13:23 | |
| *** nellysmitt has joined #openstack-cinder | 13:23 | |
| openstackgerrit | Eduard Biceri-Matei proposed openstack/cinder: New Cinder volume driver for openvstorage https://review.openstack.org/130733 | 13:27 |
| *** emagana has joined #openstack-cinder | 13:28 | |
| *** nellysmitt has quit IRC | 13:29 | |
| *** nellysmitt has joined #openstack-cinder | 13:31 | |
| *** jaypipes has joined #openstack-cinder | 13:35 | |
| *** ebalduf has joined #openstack-cinder | 13:37 | |
| *** Yogi11 has joined #openstack-cinder | 13:41 | |
| *** Yogi12 has joined #openstack-cinder | 13:41 | |
| *** ebalduf has quit IRC | 13:41 | |
| *** Yogi11 has quit IRC | 13:45 | |
| *** rushiagr_away is now known as rushiagr | 13:46 | |
| *** fandi has quit IRC | 13:48 | |
| *** jaypipes is now known as leakypipes | 13:49 | |
| rushiagr | DuncanT: sure, thanks! | 13:50 |
| *** bswartz has joined #openstack-cinder | 13:51 | |
| *** mriedem has joined #openstack-cinder | 13:52 | |
| *** Miouge_ has joined #openstack-cinder | 13:56 | |
| *** kaufer has joined #openstack-cinder | 13:56 | |
| *** Miouge has quit IRC | 13:59 | |
| *** Miouge_ is now known as Miouge | 13:59 | |
| *** fandi has joined #openstack-cinder | 13:59 | |
| *** afazekas_ has quit IRC | 14:04 | |
| *** rushil has joined #openstack-cinder | 14:05 | |
| *** mwichmann has joined #openstack-cinder | 14:06 | |
| *** fandi has quit IRC | 14:06 | |
| openstackgerrit | Jordan Pittier proposed openstack/cinder: Scality: Lock around SOFS mount to avoid a race https://review.openstack.org/147859 | 14:06 |
| *** fandi has joined #openstack-cinder | 14:07 | |
| *** bill_az has joined #openstack-cinder | 14:11 | |
| *** Raj__ has joined #openstack-cinder | 14:12 | |
| *** karimb has quit IRC | 14:14 | |
| *** fandi has quit IRC | 14:14 | |
| *** fandi has joined #openstack-cinder | 14:15 | |
| Raj__ | DuncanT: hi | 14:15 |
| *** afazekas_ has joined #openstack-cinder | 14:19 | |
| *** hypnosb_ has joined #openstack-cinder | 14:19 | |
| hypnosb_ | sup guys, anyone awake yet? (damn east coaster here already one cup in) | 14:20 |
| *** akerr has joined #openstack-cinder | 14:20 | |
| openstackgerrit | Jay Bryant proposed openstack/cinder: Move oslo.config to oslo_config namespace https://review.openstack.org/147656 | 14:22 |
| rushiagr | hypnosb_: few already left the office for the day :) | 14:22 |
| hypnosb_ | :O | 14:22 |
| *** diegows has quit IRC | 14:23 | |
| hypnosb_ | well yeah, just comming back around from that crazy problem i figured out... trying to think how i can recreate it i know jgriffith tried but i probably didnt explain or realize how it actually happened | 14:24 |
| *** xyang1_away has joined #openstack-cinder | 14:25 | |
| *** xyang1_away is now known as xyang1 | 14:25 | |
| *** jdandrea has joined #openstack-cinder | 14:25 | |
| *** dustins has joined #openstack-cinder | 14:26 | |
| *** Yogi12 has quit IRC | 14:26 | |
| *** ryanpetrello has joined #openstack-cinder | 14:29 | |
| *** dansmith is now known as superdan | 14:29 | |
| *** jecarey has joined #openstack-cinder | 14:30 | |
| Raj__ | hey i am trying to attach the | 14:31 |
| Raj__ | replicated volume to the instance again .. | 14:31 |
| Raj__ | but instance giving the error like | 14:31 |
| Raj__ | IO error .. | 14:31 |
| *** Mandell has joined #openstack-cinder | 14:32 | |
| *** Yogi1 has joined #openstack-cinder | 14:32 | |
| *** wpf has quit IRC | 14:34 | |
| hypnosb_ | Raj__: Debug=True verbose=True in cinder.conf , restart cinder-volume, and check the volume.log bruh | 14:35 |
| *** Yogi1 has quit IRC | 14:36 | |
| *** Yogi1 has joined #openstack-cinder | 14:39 | |
| *** wpf has joined #openstack-cinder | 14:39 | |
| *** Mandell has quit IRC | 14:46 | |
| *** Masahiro has joined #openstack-cinder | 14:46 | |
| *** Mandell has joined #openstack-cinder | 14:48 | |
| *** nshaikh has quit IRC | 14:50 | |
| *** Masahiro has quit IRC | 14:50 | |
| *** timcl has quit IRC | 14:53 | |
| *** coolsvap is now known as coolsvap|afk | 14:55 | |
| *** pawels1 has joined #openstack-cinder | 14:57 | |
| *** TobiasE1 has joined #openstack-cinder | 14:58 | |
| *** nellysmitt has quit IRC | 14:58 | |
| *** TobiasE has quit IRC | 14:58 | |
| *** lpabon has quit IRC | 14:59 | |
| *** dustins has quit IRC | 14:59 | |
| *** dustins has joined #openstack-cinder | 15:00 | |
| *** emagana has quit IRC | 15:00 | |
| *** pawels has quit IRC | 15:00 | |
| *** emagana has joined #openstack-cinder | 15:00 | |
| *** thangp has joined #openstack-cinder | 15:03 | |
| *** mtanino has joined #openstack-cinder | 15:03 | |
| *** russellb is now known as rustlebee | 15:04 | |
| *** jcru has quit IRC | 15:04 | |
| *** timcl has joined #openstack-cinder | 15:04 | |
| *** jcru has joined #openstack-cinder | 15:05 | |
| *** emagana has quit IRC | 15:05 | |
| *** Mandell has quit IRC | 15:05 | |
| *** bnemec has quit IRC | 15:06 | |
| Raj__ | hypnosb_: Hi | 15:07 |
| Raj__ | but after the volume is attched to the instace and | 15:08 |
| Raj__ | whe nwe are attaching replica then that is not attaching | 15:08 |
| *** bnemec has joined #openstack-cinder | 15:09 | |
| hypnosb_ | Raj__: Y u no log? | 15:11 |
| hypnosb_ | this is comlicated stuff, idk what you mean without details | 15:12 |
| hypnosb_ | been in the same boat | 15:12 |
| hypnosb_ | smashed my head many times | 15:12 |
| *** Raj__ has quit IRC | 15:12 | |
| openstackgerrit | Michal Dulko proposed openstack/cinder-specs: Integrate persistence support of TaskFlow https://review.openstack.org/147879 | 15:16 |
| *** fandi has quit IRC | 15:19 | |
| openstackgerrit | Michal Dulko proposed openstack/cinder-specs: Integrate TaskFlow's persistence support into API https://review.openstack.org/147881 | 15:21 |
| *** TobiasE has joined #openstack-cinder | 15:21 | |
| openstackgerrit | Michal Dulko proposed openstack/cinder-specs: Integrate TaskFlow's persistence into scheduler https://review.openstack.org/147883 | 15:22 |
| *** TobiasE1 has quit IRC | 15:22 | |
| openstackgerrit | Michal Dulko proposed openstack/cinder-specs: Integrate TaskFlow's persistence support into volume https://review.openstack.org/147884 | 15:23 |
| *** nkrinner has quit IRC | 15:24 | |
| *** bnemec is now known as beekneemech | 15:26 | |
| *** dulek has quit IRC | 15:29 | |
| *** fandi has joined #openstack-cinder | 15:32 | |
| *** krtaylor has quit IRC | 15:33 | |
| *** fandi has quit IRC | 15:36 | |
| *** Raj__ has joined #openstack-cinder | 15:36 | |
| *** fandi has joined #openstack-cinder | 15:36 | |
| *** ebalduf has joined #openstack-cinder | 15:36 | |
| jgriffith | has gerrit been VERY slow for anybody else the past day or so? | 15:37 |
| *** madskier has joined #openstack-cinder | 15:38 | |
| *** e0ne has quit IRC | 15:38 | |
| *** fandi has quit IRC | 15:39 | |
| *** fandi has joined #openstack-cinder | 15:40 | |
| *** ebalduf has quit IRC | 15:40 | |
| *** xyang has joined #openstack-cinder | 15:41 | |
| *** fandi has quit IRC | 15:43 | |
| *** fandi has joined #openstack-cinder | 15:44 | |
| *** fandi has quit IRC | 15:45 | |
| *** e0ne has joined #openstack-cinder | 15:45 | |
| *** tsekiyama has joined #openstack-cinder | 15:45 | |
| *** emagana has joined #openstack-cinder | 15:45 | |
| *** fischerw has joined #openstack-cinder | 15:47 | |
| *** rushiagr is now known as rushiagr_away | 15:47 | |
| *** jpich has quit IRC | 15:51 | |
| *** jpich has joined #openstack-cinder | 15:53 | |
| *** karimb has joined #openstack-cinder | 15:55 | |
| xyang | jgriffith: ping | 15:55 |
| *** rmesta has joined #openstack-cinder | 15:55 | |
| *** emagana has quit IRC | 15:55 | |
| *** hemnafk is now known as hemna | 15:55 | |
| *** emagana has joined #openstack-cinder | 15:56 | |
| *** diemt has joined #openstack-cinder | 15:56 | |
| *** crose has joined #openstack-cinder | 15:56 | |
| *** pawels1 has quit IRC | 15:57 | |
| *** anshul has quit IRC | 15:58 | |
| *** pawels has joined #openstack-cinder | 15:58 | |
| *** pawels has left #openstack-cinder | 15:58 | |
| *** wpf1 has joined #openstack-cinder | 15:59 | |
| *** madskier has quit IRC | 15:59 | |
| *** emagana has quit IRC | 16:00 | |
| jgriffith | xyang: hey | 16:01 |
| *** Mandell has joined #openstack-cinder | 16:01 | |
| xyang | jgriffith: want to talk to you about the comments on the over subscription patch | 16:02 |
| xyang | jgriffith: so do you mean "provisioned_capacity" should be moved to a separate patch? | 16:02 |
| *** wpf has quit IRC | 16:03 | |
| jgriffith | xyang: yeah, that's what I was thinking | 16:03 |
| jgriffith | xyang: it would be nice to have that settled as a seperate interest IMO | 16:03 |
| *** madskier has joined #openstack-cinder | 16:03 | |
| xyang | jgriffith: just that one or do you also want other capacities | 16:03 |
| jgriffith | xyang: so I'd do the "others" seperately as well | 16:04 |
| *** emagana has joined #openstack-cinder | 16:04 | |
| jgriffith | xyang: just the idea of trying to make our patches "smaller" and more focused | 16:04 |
| jgriffith | it seemed like a pretty easy seperation point when I looked at it | 16:04 |
| Raj__ | xyang: hey when disaster happens then | 16:04 |
| jgriffith | xyang: do you see it differently? | 16:04 |
| Raj__ | promote automatically called by cinder or not ..? | 16:04 |
| xyang | jgriffith: so move "provisioned_capacity" to one patch, and the others such as allocated and available-apparent and actual should be in a different patch? | 16:05 |
| Raj__ | xyang: or reenable etc ..?? it is automaticaly called or need to trigger by the admin ..?? | 16:05 |
| *** annashen has joined #openstack-cinder | 16:05 | |
| *** dustins has quit IRC | 16:05 | |
| xyang | jgriffith: I didn't really change allocated and still not clear what we want to change it to | 16:06 |
| jgriffith | xyang: well that's what I was thinking... but maybe it would turn out making sense to have them in the same... like a "capacities" patch | 16:06 |
| jgriffith | xyang: ok... | 16:06 |
| xyang | Raj__:that is a manual step, it is a cinder command, but not auto | 16:06 |
| jgriffith | xyang: so the only one I was really thinking about in the review was the provisioned work | 16:06 |
| jgriffith | xyang: The rest of the capacities stuff I wasn't thinking/considering | 16:07 |
| *** dustins has joined #openstack-cinder | 16:07 | |
| Raj__ | xyang: ohk thanks .. | 16:07 |
| jgriffith | xyang: the allocated and others I had mentioned only because there items that I'm thikning of changing :) | 16:07 |
| jgriffith | xyang: so that would be seperate work completely | 16:07 |
| xyang | jgriffith: I can move "provisioned" to a different patch. so no scheduler change in the first patch which has provisioned? | 16:08 |
| xyang | jgriffith: sure:) | 16:08 |
| jgriffith | xyang: yeah, I think that would be fantastic | 16:08 |
| Raj__ | xyang: one more ..?? when we are creating replica of volume ... and afetr disaster happen then volume lost and 2nd replica will take the charge of it .... | 16:08 |
| xyang | jgriffith: let me work on that and ask you to take a look after I'm done | 16:09 |
| jgriffith | xyang: great! | 16:09 |
| xyang | jgriffith: thanks! | 16:09 |
| Raj__ | xyang: it is necessary to write in driver code detach primary and attach 2nd replica on another node .. | 16:09 |
| Raj__ | xyang: from other node to instance .. | 16:10 |
| xyang | Raj__: seems like that is needed. I didn't implement replication though | 16:11 |
| Raj__ | xyang: thanks for info | 16:12 |
| *** madskier has quit IRC | 16:14 | |
| *** ebalduf has joined #openstack-cinder | 16:18 | |
| *** karimb has quit IRC | 16:19 | |
| *** Miouge has quit IRC | 16:19 | |
| *** Miouge has joined #openstack-cinder | 16:21 | |
| *** rhe00_ has quit IRC | 16:25 | |
| *** rhe00_ has joined #openstack-cinder | 16:29 | |
| *** jdurgin1 has joined #openstack-cinder | 16:30 | |
| *** annashen has quit IRC | 16:33 | |
| *** krtaylor has joined #openstack-cinder | 16:33 | |
| *** Masahiro has joined #openstack-cinder | 16:35 | |
| *** ndipanov has quit IRC | 16:36 | |
| *** timcl has quit IRC | 16:36 | |
| *** e0ne is now known as e0ne_ | 16:39 | |
| *** dustins_ has joined #openstack-cinder | 16:39 | |
| *** Masahiro has quit IRC | 16:39 | |
| *** dustins has quit IRC | 16:39 | |
| *** Longgeek has quit IRC | 16:41 | |
| *** annashen has joined #openstack-cinder | 16:41 | |
| *** karimb has joined #openstack-cinder | 16:41 | |
| thebloggu | I have an Havana deployment where I have some instances with associated volumes running correctly. However every time I try to attach a new volume or detach an existing one I get a "Attaching"/"Detaching" status for a while and then returns to its old status. I looked around the cinder logs and here's the relevant part of the api.log. volume and scheduler logs were either empty or with old entries. After a cinder-volume service restart I g | 16:43 |
| thebloggu | ot the following message on volume.log: "2015-01-16 12:27:00.292 24301 WARNING cinder.brick.iscsi.iscsi [req-ed588134-762b-4b00-88d6-6e8261963aa4 None None] Attempting recreate of backing lun...". Meanwhile I turned cinder debug and verbose mode so I can provide more info. Can someone help me? | 16:43 |
| *** _cjones_ has joined #openstack-cinder | 16:44 | |
| openstackgerrit | TaoBai proposed openstack/cinder: IBM Storwize driver Consistency Group Implementation https://review.openstack.org/142683 | 16:45 |
| *** Longgeek has joined #openstack-cinder | 16:45 | |
| *** tellesnobrega_ has joined #openstack-cinder | 16:46 | |
| tsekiyama | Hmm, ./run_tests.sh fails in cinder.tests.test_volume.VolumeTestCase.test_begin_detaching_fails_available with an exception "StopIteration" at policy_files = next(os.walk(path))[2] in _walk_through_policy_directory()... | 16:46 |
| *** rwsu has joined #openstack-cinder | 16:47 | |
| tsekiyama | Have somebody seen this exception? | 16:47 |
| *** dustins_ has quit IRC | 16:47 | |
| tsekiyama | It seems trying to walk non-existing directory (path is like "/tmp/abcdef123/policy.d", but no such directory). | 16:49 |
| hemna | tsekiyama, yes | 16:49 |
| hemna | tsekiyama, there is a problem with running tests.test_volume standalone like that | 16:50 |
| hemna | tsekiyama, if you run all the tests, it won't happen. | 16:50 |
| hemna | ./run_tests.sh | 16:50 |
| tsekiyama | By some reason, it goes on successfully on Jenkins. | 16:50 |
| hemna | I ran into that a few weeks back | 16:50 |
| *** jordanP has quit IRC | 16:50 | |
| tsekiyama | hemna: Ah.. I see. | 16:50 |
| hemna | well, Jenkins runs all the tests at once | 16:50 |
| hemna | not them individually | 16:50 |
| hemna | there is a problem with the setup of that test file | 16:50 |
| hemna | that creates a policy FILE instead of a directory, which causes the oslo code to puke. | 16:51 |
| hemna | I just haven't spent time to track it further | 16:51 |
| hemna | :( | 16:51 |
| hemna | It's a pain | 16:51 |
| *** lan_ has quit IRC | 16:51 | |
| *** junhongl has quit IRC | 16:51 | |
| *** wpf1 has quit IRC | 16:51 | |
| tsekiyama | hemna: I just trying to run cinder.tests.test_volume | 16:52 |
| *** jistr has quit IRC | 16:52 | |
| tsekiyama | hemna: will retry with whole tests. Thanks! | 16:52 |
| *** bswartz has quit IRC | 16:53 | |
| hemna | tsekiyama, np | 16:53 |
| hemna | I'll see if I can find some more time today and hack on it | 16:53 |
| hemna | I know what the problem is, I just don't know what causes it yet | 16:53 |
| *** Mandell has quit IRC | 16:53 | |
| *** anshul has joined #openstack-cinder | 16:53 | |
| *** Longgeek has quit IRC | 16:54 | |
| *** Longgeek has joined #openstack-cinder | 16:55 | |
| *** Raj__ has quit IRC | 16:55 | |
| *** afazekas_ has quit IRC | 16:56 | |
| hemna | TobiasE, ping | 16:58 |
| *** annashen has quit IRC | 16:58 | |
| *** tellesnobrega_ has quit IRC | 16:59 | |
| *** emagana has quit IRC | 16:59 | |
| *** emagana has joined #openstack-cinder | 17:00 | |
| *** rwsu has quit IRC | 17:00 | |
| *** ebalduf has quit IRC | 17:00 | |
| *** jdurgin1 has quit IRC | 17:01 | |
| *** emagana_ has joined #openstack-cinder | 17:03 | |
| *** ebalduf has joined #openstack-cinder | 17:04 | |
| *** emagana has quit IRC | 17:04 | |
| *** Mandell has joined #openstack-cinder | 17:05 | |
| *** anshul has quit IRC | 17:06 | |
| *** Mandell has quit IRC | 17:07 | |
| *** juzuluag has quit IRC | 17:07 | |
| *** e0ne_ is now known as e0ne | 17:07 | |
| *** juzuluag has joined #openstack-cinder | 17:09 | |
| *** karimb has quit IRC | 17:13 | |
| *** e0ne is now known as e0ne_ | 17:18 | |
| *** rwsu has joined #openstack-cinder | 17:19 | |
| openstackgerrit | Navneet Singh proposed openstack/cinder: NetApp eseries implementation for manage/unmanage https://review.openstack.org/138093 | 17:19 |
| *** TobiasE has left #openstack-cinder | 17:20 | |
| *** e0ne_ is now known as e0ne | 17:21 | |
| *** jpich has quit IRC | 17:22 | |
| *** wpf1 has joined #openstack-cinder | 17:23 | |
| *** junhongl has joined #openstack-cinder | 17:23 | |
| *** lan_ has joined #openstack-cinder | 17:24 | |
| *** leeantho has joined #openstack-cinder | 17:26 | |
| openstackgerrit | Gary W. Smith proposed openstack/python-cinderclient: Expose cinder's scheduler pool API https://review.openstack.org/140142 | 17:26 |
| *** thingee has joined #openstack-cinder | 17:27 | |
| *** dannywilson has joined #openstack-cinder | 17:28 | |
| *** Tross1 has quit IRC | 17:30 | |
| *** Tross has joined #openstack-cinder | 17:32 | |
| *** dustins has joined #openstack-cinder | 17:36 | |
| *** tbarron has joined #openstack-cinder | 17:38 | |
| *** e0ne has quit IRC | 17:39 | |
| *** mudassirlatif has joined #openstack-cinder | 17:41 | |
| *** dannywilson has quit IRC | 17:43 | |
| *** timcl has joined #openstack-cinder | 17:44 | |
| *** leeantho has quit IRC | 17:49 | |
| *** rwsu has quit IRC | 17:51 | |
| *** bswartz has joined #openstack-cinder | 17:52 | |
| *** ebalduf has quit IRC | 17:54 | |
| *** annashen has joined #openstack-cinder | 17:56 | |
| *** rwsu has joined #openstack-cinder | 17:58 | |
| *** ebalduf has joined #openstack-cinder | 17:58 | |
| *** Miouge has quit IRC | 18:00 | |
| *** shakamunyi has joined #openstack-cinder | 18:02 | |
| *** xyang has quit IRC | 18:02 | |
| *** changbl has quit IRC | 18:03 | |
| openstackgerrit | xing-yang proposed openstack/cinder: Add provisioned_capacity https://review.openstack.org/147934 | 18:03 |
| *** patrickeast has joined #openstack-cinder | 18:04 | |
| jgriffith | patrickeast: speak of the devil :) | 18:07 |
| patrickeast | uh oh | 18:07 |
| patrickeast | whats up | 18:07 |
| jgriffith | patrickeast: if you fix the merge conflic up I'll get this merged: https://review.openstack.org/#/c/143726/ | 18:07 |
| patrickeast | jgriffith: sweet | 18:08 |
| *** leeantho has joined #openstack-cinder | 18:08 | |
| jgriffith | jungleboyj: do we even have a bug for this: https://review.openstack.org/#/c/146763/3//COMMIT_MSG | 18:08 |
| *** dannywilson has joined #openstack-cinder | 18:08 | |
| *** xyang has joined #openstack-cinder | 18:10 | |
| *** dannywilson has quit IRC | 18:14 | |
| *** dannywilson has joined #openstack-cinder | 18:14 | |
| jungleboyj | jgriffith: Hmmm, | 18:15 |
| *** bkopilov has joined #openstack-cinder | 18:15 | |
| *** smcginni1 is now known as smcginnis_afk | 18:16 | |
| openstackgerrit | Navneet Singh proposed openstack/cinder: Implement Cinder Manage/Unmanage APIs in NFS driver. https://review.openstack.org/138125 | 18:16 |
| *** tellesnobrega_ has joined #openstack-cinder | 18:17 | |
| jungleboyj | jgriffith: It is related to: https://bugs.launchpad.net/cinder/juno/+bug/1408099 | 18:17 |
| jungleboyj | We don't have a separate bug though. | 18:17 |
| *** kaufer has quit IRC | 18:17 | |
| *** Miouge has joined #openstack-cinder | 18:17 | |
| jgriffith | jungleboyj: good enough, you want to update as per Mikes request? | 18:17 |
| *** harlowja_away is now known as harlowja | 18:18 | |
| jungleboyj | Yep. Will do it now. Thanks for pointing that out jgriffith . | 18:18 |
| jgriffith | hemna: I'm trying to review the multi-attach again | 18:18 |
| hemna | ok thanks | 18:18 |
| jgriffith | hemna: honestly though it's REALLY hard to figure out 1300 line patches :( | 18:18 |
| hemna | yah :( it is | 18:19 |
| *** dannywilson has quit IRC | 18:19 | |
| hemna | I'm testing it with the client, nova and horizon patches | 18:19 |
| hemna | the nova side still has some issues | 18:20 |
| jgriffith | hemna: I still say we shouldn't limit this to being set on create only :) | 18:20 |
| *** mriedem has quit IRC | 18:20 | |
| openstackgerrit | Jay Bryant proposed openstack/cinder: Ensure lazy translation is disabled properly https://review.openstack.org/146763 | 18:20 |
| hemna | because you want more code to review in this patch? :P | 18:20 |
| jgriffith | hemna: but that's certainly somethign that can be added later | 18:21 |
| jgriffith | hemna: LOL | 18:21 |
| jgriffith | EXACTLY | 18:21 |
| hemna | I say the update/add shareable is a good follow on patch :) | 18:21 |
| jgriffith | I was just going to say that actually | 18:21 |
| jgriffith | :) | 18:21 |
| hemna | :) | 18:21 |
| jgriffith | hemna: yes, totally agree | 18:21 |
| jgriffith | hemna: why the lock change: https://review.openstack.org/#/c/85847/19/cinder/volume/manager.py | 18:22 |
| jgriffith | hemna: at line 735 | 18:22 |
| jgriffith | rtt... 738 | 18:22 |
| jgriffith | hemna: NM :) | 18:23 |
| jgriffith | You documented it | 18:23 |
| hemna | :) | 18:23 |
| *** Masahiro has joined #openstack-cinder | 18:23 | |
| jgriffith | hemna: wait.... | 18:25 |
| jgriffith | hemna: so now there's no reference in the Vol object (DB) indicating it has an attachment? | 18:25 |
| hemna | the vol object has attachments list | 18:26 |
| hemna | which is a ref to the entries in the volume_attachment table | 18:26 |
| jgriffith | hemna: I only see attach_status | 18:26 |
| jgriffith | hemna: and shareable column | 18:26 |
| jgriffith | the volume object is the FK in VolumeAttachment table | 18:26 |
| hemna | backref="volume_attachment" | 18:27 |
| jgriffith | hemna: am I missing something obvious (or non-obvious) | 18:27 |
| jgriffith | hemna: BAhhhh | 18:27 |
| hemna | sec | 18:27 |
| hemna | db api.py _volume_get_query adds the joinedload of volume_attachment | 18:28 |
| *** Masahiro has quit IRC | 18:28 | |
| jgriffith | hemna: oh, sorry, yeah I found it | 18:28 |
| jgriffith | hemna: so I'm nervous about the db changes | 18:28 |
| *** e0ne has joined #openstack-cinder | 18:28 | |
| hemna | sure | 18:28 |
| jgriffith | hemna: particularly though when it comes to upgrades and such | 18:29 |
| jgriffith | hemna: ripping those columns out I think is tricky | 18:29 |
| hemna | the upgrade script migrates the data | 18:29 |
| hemna | actually the javelin in grenade found some issues that I fixed | 18:29 |
| hemna | wrt upgrading the data | 18:29 |
| *** kaufer has joined #openstack-cinder | 18:29 | |
| jgriffith | hemna: yeah, I know you handled it in the upgrade script | 18:30 |
| jgriffith | hemna: just nervous trying to think of all the scenarios | 18:30 |
| jgriffith | hemna: and wondering if there's a "safer" way to do this | 18:30 |
| * jgriffith isn't saying there is | 18:30 | |
| jgriffith | just wondering/thinking | 18:31 |
| hemna | yah | 18:31 |
| hemna | well if you think of anything let me know | 18:32 |
| *** aix has quit IRC | 18:32 | |
| jgriffith | haha | 18:32 |
| jgriffith | hemna: well... converting columns to lists instead of new tables came to mind :) | 18:33 |
| hemna | doing queries against custom formatted data inside a column can suck though | 18:33 |
| openstackgerrit | Jay Bryant proposed openstack/cinder: Move oslo.config to oslo_config namespace https://review.openstack.org/147656 | 18:34 |
| jgriffith | hemna: that's an understatement | 18:34 |
| jgriffith | hemna: but not as bad as some things | 18:35 |
| jgriffith | anyway... | 18:35 |
| jgriffith | I'm just wondering if there was a way to chunk this up | 18:35 |
| *** Mandell has joined #openstack-cinder | 18:35 | |
| jgriffith | and stage the adoption of it a little bit | 18:35 |
| hemna | into multiple patches ? | 18:35 |
| jgriffith | rather than just flip everything on it's head at once | 18:35 |
| hemna | I think that might be tough | 18:35 |
| jgriffith | hemna: Not even multiple patches so much as a staged introduction of the feature | 18:36 |
| jgriffith | hemna: well, yeah... it's def tough | 18:36 |
| hemna | I think it's better than it was now though | 18:36 |
| hemna | with backwards compatible detach api | 18:36 |
| hemna | making the attachment_id optional definitely helped | 18:36 |
| *** tellesnobrega_ has quit IRC | 18:43 | |
| *** alexpilotti has quit IRC | 18:52 | |
| thingee | jgriffith: I remember what I wanted to discuss with you earlier after the meeting, this https://bugs.launchpad.net/cinder/+bug/1408763 | 18:56 |
| thingee | jgriffith: not sure what the decided approach is at this time? | 18:56 |
| jgriffith | thingee: nobody seems to have much input to be honest | 18:57 |
| jgriffith | thingee: I proposed the DB option and nobody balked | 18:57 |
| jgriffith | thingee: you however said something about "shouldn't punt on fixing it in taskflow because it's hard" | 18:58 |
| *** tellesnobrega_ has joined #openstack-cinder | 18:58 | |
| *** _cjones_ has quit IRC | 18:59 | |
| jgriffith | thingee: updated... it's now fair game | 19:00 |
| *** _cjones_ has joined #openstack-cinder | 19:01 | |
| thingee | jgriffith: ok thanks | 19:01 |
| jgriffith | thingee: np | 19:02 |
| boris-42 | jgriffith: attach issues strikes again | 19:05 |
| boris-42 | jgriffith: so I'll recehck your patch | 19:05 |
| boris-42 | jgriffith: to try to face them | 19:05 |
| jgriffith | boris-42: well, part of me is glad | 19:05 |
| jgriffith | boris-42: as long as it doesn't show up in the retry patch :) | 19:06 |
| boris-42 | jgriffith: hehe | 19:06 |
| *** openstick has joined #openstack-cinder | 19:09 | |
| *** openstick has quit IRC | 19:09 | |
| *** vzm has joined #openstack-cinder | 19:10 | |
| openstackgerrit | xing-yang proposed openstack/cinder: EMC VMAX driver Kilo update https://review.openstack.org/141729 | 19:12 |
| *** xyang has quit IRC | 19:12 | |
| jgriffith | I'm out | 19:13 |
| vzm | Hello, I have a question regarding the Openstack zone manager... Does it allow both Brocade and Cisco switches(connected to both) to be used for the dual fabrics on zoning? | 19:14 |
| *** rwsu has quit IRC | 19:15 | |
| *** fandi has joined #openstack-cinder | 19:17 | |
| *** dannywilson has joined #openstack-cinder | 19:17 | |
| openstackgerrit | Patrick East proposed openstack/cinder: Add support for manage/unmanage volume commands to PureISCSIDriver https://review.openstack.org/143726 | 19:22 |
| *** xyang has joined #openstack-cinder | 19:23 | |
| *** thebloggu has quit IRC | 19:25 | |
| *** tbarron has quit IRC | 19:25 | |
| *** tbarron has joined #openstack-cinder | 19:26 | |
| *** Mandell has quit IRC | 19:27 | |
| *** kaufer1 has joined #openstack-cinder | 19:29 | |
| *** kaufer has quit IRC | 19:31 | |
| *** Yogi1 has quit IRC | 19:32 | |
| *** rwsu has joined #openstack-cinder | 19:32 | |
| *** lpetrut has quit IRC | 19:35 | |
| *** tellesnobrega_ has quit IRC | 19:35 | |
| *** crose has quit IRC | 19:38 | |
| *** _cjones_ has quit IRC | 19:53 | |
| akerr | slightly selfish request: can a core member abandon change https://review.openstack.org/#/c/114042/? It was -2'd in November and it just clutters my "incoming review" field :) | 19:53 |
| hemna | akerr, done | 19:55 |
| akerr | hemna: thanks | 19:55 |
| hemna | np | 19:55 |
| openstackgerrit | Ivan Kolodyazhny proposed openstack/cinder: Added snapshots list to volume details view https://review.openstack.org/144739 | 19:57 |
| *** mudassirlatif has quit IRC | 19:59 | |
| *** annashen has quit IRC | 20:04 | |
| *** Yogi11 has joined #openstack-cinder | 20:05 | |
| hemna | e0ne, it'll take me a day or so to get the brick lib updated | 20:08 |
| *** bill_az has quit IRC | 20:09 | |
| tsekiyama | hemna: I drilled down the policy_files = next(os.walk(path))[2] StopIteration exception on './run_tests.sh test_volume'. | 20:09 |
| tsekiyama | hemna: It is related to mock of os.path.exists(). | 20:09 |
| tsekiyama | hemna: The policy is initialized when the first context is crated. If the exists() is mocked and always return True at that time, the oslo_config is confused that "policy.d" is existing. | 20:09 |
| tsekiyama | hemna: And it tries to walk non-existing directory, that causes the Exception. | 20:10 |
| *** thangp has quit IRC | 20:10 | |
| hemna | tsekiyama, from what I saw, the policy.d wasn't a directory, but a file | 20:10 |
| hemna | which os.walk() pukes on. | 20:10 |
| hemna | or it was next() | 20:10 |
| hemna | one of the 2 | 20:10 |
| tsekiyama | hemna: Hmm, then maybe I'm experiencing similar but another issue? | 20:12 |
| *** Masahiro has joined #openstack-cinder | 20:12 | |
| hemna | tsekiyama, could be | 20:12 |
| hemna | afk....lunch | 20:13 |
| *** ebalduf has quit IRC | 20:15 | |
| *** lpabon has joined #openstack-cinder | 20:15 | |
| *** kaufer1 has quit IRC | 20:16 | |
| *** Masahiro has quit IRC | 20:17 | |
| *** lpetrut has joined #openstack-cinder | 20:17 | |
| *** e0ne is now known as e0ne_ | 20:18 | |
| *** kaufer has joined #openstack-cinder | 20:18 | |
| *** rushiagr_away is now known as rushiagr | 20:22 | |
| *** diegows has joined #openstack-cinder | 20:23 | |
| *** _cjones_ has joined #openstack-cinder | 20:23 | |
| *** annashen has joined #openstack-cinder | 20:26 | |
| *** xyang has quit IRC | 20:28 | |
| *** xyang has joined #openstack-cinder | 20:28 | |
| xyang | tsekiyama, hemna: I ran into the same problem. I modified the test to get around it: https://review.openstack.org/#/c/110068/18/cinder/tests/test_volume.py line 128, 2416 | 20:28 |
| patrickeast | jgriffith: hey, i rebased that change and fixed the conficts, jenkins just gave it the ok so it should be ready to review whenever you have a chance (https://review.openstack.org/#/c/143726/ ) | 20:29 |
| *** Mandell has joined #openstack-cinder | 20:29 | |
| *** akerr has quit IRC | 20:30 | |
| *** Longgeek has quit IRC | 20:42 | |
| *** Longgeek has joined #openstack-cinder | 20:43 | |
| *** dustins has quit IRC | 20:47 | |
| *** xyang has quit IRC | 20:47 | |
| openstackgerrit | xing-yang proposed openstack/cinder: EMC VMAX driver Kilo update https://review.openstack.org/141729 | 20:48 |
| *** hypnosb_ has quit IRC | 20:50 | |
| *** changbl has joined #openstack-cinder | 20:52 | |
| openstackgerrit | Ivan Kolodyazhny proposed openstack/cinder: Added snapshots list to volume details view https://review.openstack.org/144739 | 20:53 |
| *** e0ne_ has quit IRC | 20:53 | |
| *** _cjones_ has quit IRC | 20:54 | |
| *** _cjones_ has joined #openstack-cinder | 20:54 | |
| *** rushil has quit IRC | 20:54 | |
| *** Yogi11 has quit IRC | 20:56 | |
| *** timcl has left #openstack-cinder | 20:57 | |
| *** rlucio has joined #openstack-cinder | 20:57 | |
| anish | jgriffith: is there a reason why the first parma for this is not a formatting specifier ? https://github.com/openstack/cinder/blob/master/cinder/volume/targets/iscsi.py#L125 | 20:58 |
| anish | in all it's use cases, everyone calls this function and then regex/subs the result later to get intended format for use | 20:58 |
| anish | seems wasteful | 20:58 |
| *** Yogi1 has joined #openstack-cinder | 20:58 | |
| jgriffith | anish: I just carried that over from what already existed | 20:59 |
| jgriffith | anish: certainly could probably be made more efficient | 20:59 |
| jgriffith | anish: I'm not sure about the format difference you mention though | 20:59 |
| anish | called as chap_auth = self._iscsi_authentication('IncomingUser', chap_username, | 21:00 |
| anish | chap_password) | 21:00 |
| anish | later on : chap_str = re.sub('^IncomingUser ', 'incominguser ', chap_auth) | 21:00 |
| anish | I presume it is for tgt/lio compatibility | 21:01 |
| jgriffith | anish: that all was rewritten in a somewhat recent patch | 21:01 |
| jgriffith | anish: suspect there's some mess that could be cleaned up | 21:01 |
| jgriffith | the chap_str changes you mention above were for fixing the persist file | 21:02 |
| jgriffith | and making things work on reinit (yes, tgt inparticular) | 21:02 |
| anish | I figured | 21:02 |
| *** emagana_ has quit IRC | 21:02 | |
| *** emagana has joined #openstack-cinder | 21:03 | |
| anish | it's also partly selfish, the cxt helper I proposed uses a slightly more complicated format that actually can use a format specifier for iscsi_authentication | 21:03 |
| anish | but it overall made more sense than a function that does a+b+c | 21:03 |
| *** xyang has joined #openstack-cinder | 21:04 | |
| *** emagana has quit IRC | 21:04 | |
| *** emagana has joined #openstack-cinder | 21:05 | |
| tsekiyama | xyang: thanks, it worked for me. | 21:05 |
| *** Longgeek_ has joined #openstack-cinder | 21:06 | |
| *** Longgeek has quit IRC | 21:06 | |
| xyang | tsekiyama: great | 21:07 |
| *** emagana has quit IRC | 21:09 | |
| *** emagana has joined #openstack-cinder | 21:15 | |
| *** tellesnobrega has quit IRC | 21:15 | |
| *** ebalduf has joined #openstack-cinder | 21:16 | |
| *** madskier has joined #openstack-cinder | 21:17 | |
| *** juzuluag has quit IRC | 21:19 | |
| *** ebalduf has quit IRC | 21:20 | |
| *** Longgeek_ has quit IRC | 21:22 | |
| *** xyang has quit IRC | 21:24 | |
| openstackgerrit | xing-yang proposed openstack/cinder: Add provisioned_capacity https://review.openstack.org/147934 | 21:24 |
| *** tellesnobrega has joined #openstack-cinder | 21:28 | |
| *** e0ne has joined #openstack-cinder | 21:28 | |
| *** e0ne has quit IRC | 21:29 | |
| *** elazro has quit IRC | 21:31 | |
| openstackgerrit | Joe D'Andrea proposed openstack/cinder: Fixes-Bug: pvs-separator-bug https://review.openstack.org/148002 | 21:32 |
| *** fandi has quit IRC | 21:32 | |
| openstackgerrit | xing-yang proposed openstack/cinder: Support over subscription in thin provisioning https://review.openstack.org/142171 | 21:34 |
| *** mudassirlatif has joined #openstack-cinder | 21:36 | |
| *** leakypipes has quit IRC | 21:38 | |
| * jdandrea does hopeful happy dance upon submitting first (tiny) patch. | 21:41 | |
| * anish grabs jdandrea | 21:42 | |
| anish | hold your dancing till jenkins tells you so | 21:42 |
| *** fandi has joined #openstack-cinder | 21:44 | |
| *** changbl has quit IRC | 21:44 | |
| *** Mandell has quit IRC | 21:48 | |
| *** tellesnobrega has quit IRC | 21:50 | |
| *** Mandell has joined #openstack-cinder | 21:50 | |
| *** kaufer has quit IRC | 21:54 | |
| *** xyang has joined #openstack-cinder | 21:54 | |
| jdandrea | anish: That's why I said it was hopeful. ;) | 21:57 |
| jdandrea | But yes, jenkins said BZZZZT. | 21:57 |
| * jdandrea goes to look | 21:58 | |
| jdandrea | Hmm. tgt failed to restart. | 21:58 |
| eharney | jdandrea: glad to see it, i added a question | 21:59 |
| * jdandrea scratches head | 21:59 | |
| jdandrea | Thanks, eharney | 21:59 |
| *** tellesnobrega has joined #openstack-cinder | 22:01 | |
| *** Masahiro has joined #openstack-cinder | 22:01 | |
| jdandrea | eharney: Naive question. By "Summary line" do you mean I add a line at the top of the commit beginning with "Summary: " ? | 22:01 |
| *** junhongl has quit IRC | 22:01 | |
| eharney | jdandrea: not to add "Summary:", just that that's what the first line is called and is usually something easily understandable | 22:02 |
| eharney | jdandrea: you may want to look over https://wiki.openstack.org/wiki/GitCommitMessages for general ideas of what commit messages should look like | 22:02 |
| *** lan_ has quit IRC | 22:02 | |
| jdandrea | eharney: Ah, ok. So instead of Fixes-Bug: just put ... yep, I read that. I don't think I read it well enough. :) | 22:02 |
| *** wpf1 has quit IRC | 22:02 | |
| jdandrea | Also: http://paste.openstack.org/show/158024/ | 22:02 |
| jdandrea | There are cases where the underlying disk is remotely mounted (if I understand correctly - we use iSCSI f'rinstance) and the ':' can enter into it, as seen in that paste. | 22:03 |
| eharney | jdandrea: well that's neat :) | 22:03 |
| eharney | i figured you hit some case where a ':' showed up but i couldn't figure out how, was just curious | 22:04 |
| jdandrea | eharney: np - and actually this has been dogging us for a while. Once we spotted that and tried another separator, things worked MUCH better! | 22:04 |
| eharney | i'd bet so | 22:05 |
| jdandrea | I also see ':' used in get_all_volume_groups, but that's not where I saw our issue, and so I left it alone for now. | 22:05 |
| jdandrea | Don't want to break something if it's already working. | 22:05 |
| *** Masahiro has quit IRC | 22:06 | |
| jdandrea | Ok, so I will ditch the trailing space, I'll mention the paste in my reply to your comment, and fix up the summary. | 22:06 |
| eharney | jdandrea: well, LVM doesn't consider ':' to be valid in a volume group name | 22:06 |
| eharney | so we're probably ok on that one | 22:06 |
| jdandrea | eharney: Excellent. | 22:06 |
| jdandrea | Thx! | 22:06 |
| *** wpf1 has joined #openstack-cinder | 22:07 | |
| jdandrea | This is a good intro for me too. It's a small, understandable change. | 22:07 |
| anish | jdandrea: some more nits | 22:07 |
| *** lan_ has joined #openstack-cinder | 22:07 | |
| jdandrea | anish: Go! | 22:07 |
| anish | I put them in the spec as comments | 22:07 |
| jdandrea | Good, tx - will look. | 22:07 |
| jdandrea | Appreciated! | 22:07 |
| *** junhongl has joined #openstack-cinder | 22:11 | |
| *** lpabon has quit IRC | 22:12 | |
| jdandrea | Yes, I was thinking LVM.FIELD_SEP was the way to go, no? | 22:13 |
| anish | eharney: my python-fu is not the greatest, but I assumed if someday someone inherited from the LVM class, they would be able to overwrite FIELD_SEP via self, no ? | 22:14 |
| jdandrea | Oooh, interesting. | 22:14 |
| eharney | how about we just make it a local var since it's only used in one method anyway :) | 22:15 |
| jdandrea | eharney: Yeah, good point. | 22:15 |
| jdandrea | field_sep = '|' and go to town. | 22:15 |
| jdandrea | I added it up top thinking I might someday use it for both methods, but of course if we ever had to I could refactor it *then*. | 22:16 |
| jdandrea | Since LVM says : is verboten in a vg name ... :) | 22:16 |
| eharney | right | 22:16 |
| jdandrea | Now, I wonder what's up with all those other tests failing. Do those normally fail? (Suspecting no.) :( | 22:17 |
| * jdandrea looks at logs ... tgt restart fails? Ugh. | 22:17 | |
| eharney | from the various CIs? | 22:17 |
| jdandrea | eharney: Yeah. | 22:17 |
| anish | the 3par-fc has been failing as far as I know | 22:17 |
| anish | the cloudbyte CI fails in like 1 sec | 22:18 |
| eharney | cloudbyte's is broken, it's been posting merge failures, which is just bad | 22:18 |
| *** Mandell has quit IRC | 22:18 | |
| anish | the hyper-v ci has been failing for the last 2 weeks afaik | 22:18 |
| jdandrea | :( | 22:18 |
| eharney | these are all new setups, mainly need to check Jenkins for a patch like this | 22:18 |
| jdandrea | When I post my comments, I'll set Workflow to -1, ya? | 22:18 |
| anish | if you think code is not complete | 22:19 |
| anish | -1 implies WIP I believe | 22:19 |
| eharney | that just means the code is a work in progress, which is probably not the case here | 22:19 |
| jdandrea | anish: Bingo. Ok then. | 22:19 |
| anish | a lot of the CIs seem to fail at (what looks like) the devstack setup phase | 22:20 |
| anish | which stuck me as really weird | 22:20 |
| jdandrea | anish: agreed. | 22:21 |
| jdandrea | eharney: Is it advisable to reference the pastebin entry as supplemental info in the commit message? TMI? | 22:21 |
| eharney | jdandrea: i don't really like links to things like pastebins in commit messages, since commit messages last forever and the pastebin might not.. | 22:22 |
| jdandrea | eharney: Good point. | 22:22 |
| jdandrea | I knew I felt odd about that for a reason. | 22:22 |
| eharney | jdandrea: you can add an example of the pvs output that breaks it in the message maybe | 22:23 |
| jdandrea | eharney: Sounds good. | 22:23 |
| *** tbarron has quit IRC | 22:26 | |
| *** diemt has quit IRC | 22:26 | |
| *** EmilienM is now known as EmilienM|afk | 22:26 | |
| jdandrea | eharney: the commit message lines should be no more than 79 characters, yes? (The pvs output is > 80.) | 22:30 |
| jdandrea | Though I suppose I could break it. Carefully. :) | 22:30 |
| anish | 70 | 22:30 |
| jdandrea | anish: tx | 22:30 |
| jdandrea | eharney, anish: Just to make extra sure I'm using good form ... http://paste.openstack.org/show/158323/ | 22:33 |
| jdandrea | I may have to shorten my summary. | 22:33 |
| jdandrea | Will make it "Changed pvs separator from ':' to '|' to support names with ':'." | 22:34 |
| anish | if you really want to make sure, https://wiki.openstack.org/wiki/GitCommitMessages | 22:34 |
| jdandrea | anish: Yes, read that. | 22:35 |
| jdandrea | Just being mildly paranoid. :) | 22:35 |
| jdandrea | Ok, removing leading spaces in the supplemental and made the summary one line, all under 70 chars. | 22:35 |
| jdandrea | Arrgh, I did a "git commit" and didn't squash/amend. | 22:38 |
| jdandrea | Aborted the review. | 22:38 |
| jdandrea | Not sure what to do now. | 22:39 |
| eharney | jdandrea: you have two commits? | 22:40 |
| jdandrea | eharney: Apparently. :/ | 22:40 |
| jdandrea | Unsure if I should rebase. | 22:40 |
| eharney | i usually do git rebase -i HEAD~2 and squash them | 22:40 |
| *** shakamunyi has quit IRC | 22:40 | |
| jdandrea | eharney: Ok. Not HEAD~1? | 22:41 |
| anish | that would be just the first commit on top | 22:41 |
| jdandrea | Ah. >.< ok | 22:41 |
| jdandrea | heh | 22:41 |
| eharney | well, that might work, i'm not really sure | 22:41 |
| jdandrea | I don't normally use rebase. This is new territory. I've heard horror stories about misuse of rebase. | 22:41 |
| jdandrea | Just want to be sure I do it properly. | 22:42 |
| jdandrea | Ok, then I change both 'pick's to 'squash' instead. | 22:42 |
| anish | eharney: no, you'd need at least ~2 to squash | 22:42 |
| jdandrea | I used ~2. I see both commits with pick. | 22:42 |
| eharney | no, one pick and one squash | 22:42 |
| anish | no, pick one, squash one | 22:42 |
| jdandrea | So I could leave the earlier one as pick and make the new one squash. | 22:43 |
| jdandrea | I think. | 22:43 |
| eharney | yes | 22:43 |
| anish | yes | 22:43 |
| jdandrea | :) | 22:43 |
| *** fischerw has quit IRC | 22:43 | |
| eharney | (this is a routine part of my development workflow actually) | 22:43 |
| anish | you're like always half a second faster eharney | 22:43 |
| eharney | haha | 22:43 |
| jdandrea | lol | 22:43 |
| jdandrea | For the commit message. Do I get rid of "This is a combination of 2 commits" and just use the latest commit? | 22:43 |
| jdandrea | That may be less confusing. | 22:43 |
| eharney | i like to stack up four or five intermediate commits as i code and squash them once things look good | 22:43 |
| eharney | yes, get rid of that message and just stick with... whichever one has what you want | 22:44 |
| * jdandrea keeps that in mind. | 22:44 | |
| jdandrea | eharney: Good. | 22:44 |
| *** lpetrut has quit IRC | 22:44 | |
| eharney | jdandrea: but | 22:44 |
| eharney | make sure you keep the same Change-Id | 22:44 |
| jdandrea | Arrrrgh. I swear, I still messed it up. | 22:44 |
| eharney | or you'll get a new gerrit review | 22:44 |
| eharney | instead of updating the same one | 22:44 |
| jdandrea | I went to delete a line and actually used Cmd-K instead of Ctrl-K. | 22:45 |
| jdandrea | It rebased/updated without changing the commit msg. >.< | 22:45 |
| jdandrea | Sorry. | 22:45 |
| jdandrea | Sig. | 22:45 |
| jdandrea | sigh | 22:45 |
| jdandrea | Maybe I can edit the commit msg. | 22:46 |
| * jdandrea he asks hopefully - checks git docs ... and I have the change id. | 22:46 | |
| anish | hrm, anone know of a way to check if an ip is v6 in python2 ? ipaddr seems to be python3 only, and does not seem to be used in cinder anywhere | 22:46 |
| anish | jdandrea: super easy. git commit --amend | 22:46 |
| eharney | anish: don't we have code doing that in Cinder now? | 22:46 |
| jdandrea | anish: Got it - tx! I am using the change id from my LATEST update. It's at the very end. | 22:47 |
| jdandrea | Whew. | 22:47 |
| *** Yogi1 has quit IRC | 22:47 | |
| eharney | jdandrea: i think you want the original ChangeId | 22:47 |
| jdandrea | eharney: :-o | 22:47 |
| jdandrea | ok, I still have it: Change-Id: Ia5442af2a06a004ff1af58ce08b816506b369f22 | 22:48 |
| anish | eharney: do you happen to know the fn call ? I see it in nova/utils but not cinder | 22:48 |
| *** vzm has quit IRC | 22:48 | |
| eharney | anish: i see netaddr.valid_ipv6() in our tests | 22:48 |
| openstackgerrit | Joe D'Andrea proposed openstack/cinder: Changed pvs separator from ':' to '|' to support names with ':'. https://review.openstack.org/148002 | 22:49 |
| anish | I see this : netutils.is_valid_ipv6 | 22:49 |
| jdandrea | Huzzah! | 22:49 |
| eharney | anish: that's in oslo_utils, which would be the thing to use | 22:49 |
| jdandrea | This looks much better. Crossing fingers. | 22:49 |
| jdandrea | Thanks everyone. I appreciate the nits and notes. | 22:49 |
| anish | eharney: aite, will use that. thanks ! | 22:50 |
| eharney | jdandrea: does this happen on an existing driver? | 22:50 |
| jdandrea | eharney: It's happening on at least one of our clusters, yes. | 22:51 |
| eharney | jdandrea: if it's something you'll want backported to stable branches, you'll probably need a launchpad bug for it | 22:51 |
| jdandrea | it was explained to me as iSCSI vols showing up as phys vols. | 22:51 |
| jdandrea | eharney: Ah, ok. | 22:51 |
| jdandrea | If I should need to do that (though I think it would probably be appreciated regardless), can I do that and then reference this? | 22:52 |
| eharney | yes, basically you just file a bug and then put "Closes-Bug: #1234" in the commit message to link them together | 22:53 |
| *** Mandell has joined #openstack-cinder | 22:53 | |
| *** Mandell has quit IRC | 22:55 | |
| *** Mandell has joined #openstack-cinder | 22:55 | |
| openstackgerrit | Jay Bryant proposed openstack/cinder: Move oslo.config to oslo_config namespace https://review.openstack.org/147656 | 22:58 |
| openstackgerrit | Jay Bryant proposed openstack/cinder: Move oslo.utils to oslo_utils namespace https://review.openstack.org/148018 | 22:58 |
| openstackgerrit | Jay Bryant proposed openstack/cinder: Move oslo.config to oslo_config namespace https://review.openstack.org/147656 | 22:59 |
| *** ryanpetrello has quit IRC | 22:59 | |
| jdandrea | eharney: Ah, but I already submitted this for review. | 23:01 |
| jdandrea | So, I guess worst case, I can post another commit for review, closing the bug. | 23:01 |
| jdandrea | Though the commit won't have any changes, hm. :/ | 23:01 |
| eharney | it'll just update the message, that is fine to do | 23:01 |
| openstackgerrit | Jay Bryant proposed openstack/cinder: Move oslo.utils to oslo_utils namespace https://review.openstack.org/148018 | 23:02 |
| openstackgerrit | Jay Bryant proposed openstack/cinder: Move oslo.config to oslo_config namespace https://review.openstack.org/147656 | 23:02 |
| jdandrea | Ok, so if I understand I'll just do another commit (--amend?) and then add Closes-Bug: xxxx to match the bug I'll eventually file, then another git-review. | 23:02 |
| eharney | correct | 23:02 |
| jdandrea | But I won't do that of course until I have the bug filed, got it. | 23:02 |
| jdandrea | ok | 23:02 |
| eharney | see Jay just did it three times ^ :) | 23:03 |
| *** ryanpetrello has joined #openstack-cinder | 23:03 | |
| jdandrea | Jay's on a roll. ;) | 23:03 |
| jdandrea | I'm sure, before long, this will be second nature! | 23:03 |
| eharney | then you will begin to notice other things like... the fact that our entire CI infrastructure is failing patchsets right now | 23:04 |
| *** emagana has quit IRC | 23:04 | |
| *** emagana has joined #openstack-cinder | 23:04 | |
| *** emagana has quit IRC | 23:09 | |
| *** emagana has joined #openstack-cinder | 23:11 | |
| jdandrea | eharney: Well, yes, there's that. :) | 23:11 |
| *** marcusvrn has quit IRC | 23:15 | |
| *** marcusvrn has joined #openstack-cinder | 23:15 | |
| *** alexpilotti has joined #openstack-cinder | 23:17 | |
| *** openstackgerrit has quit IRC | 23:19 | |
| *** openstackgerrit has joined #openstack-cinder | 23:19 | |
| *** 17SABIH3M is now known as annegentle | 23:26 | |
| *** scottda_ has joined #openstack-cinder | 23:29 | |
| *** scottda_ has quit IRC | 23:34 | |
| openstackgerrit | Gloria Gu proposed openstack/cinder: Update volume type name for volume type API https://review.openstack.org/140906 | 23:34 |
| *** mwichmann has quit IRC | 23:36 | |
| anish | eharney: any idea what causes this ? imports not in alphabetical order (oslo_utils.netutils, oslo_concurrency.processutils) | 23:38 |
| anish | there are only two imports, and flipping their order makes zero difference | 23:38 |
| *** emagana has quit IRC | 23:39 | |
| *** emagana has joined #openstack-cinder | 23:40 | |
| jdandrea | Interesting. Storwise succeeded this time! | 23:42 |
| *** emagana has quit IRC | 23:44 | |
| *** IanGovett has quit IRC | 23:48 | |
| *** Masahiro has joined #openstack-cinder | 23:50 | |
| *** Masahiro has quit IRC | 23:54 | |
| *** ryanpetrello has quit IRC | 23:55 | |
| *** ryanpetrello has joined #openstack-cinder | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!