Friday, 2016-03-04

« Thursday, 2016-03-03 Index Saturday, 2016-03-05 »
*** a_ta has joined #openstack-manila00:05
*** yangyapeng has quit IRC00:08
*** a_ta has quit IRC00:10
*** gouthamr has quit IRC00:51
*** akshai has quit IRC00:58
*** akshai has joined #openstack-manila01:02
*** akshai has quit IRC01:08
*** yangyapeng has joined #openstack-manila01:11
*** yangyapeng has quit IRC01:17
*** gouthamr has joined #openstack-manila01:25
*** gouthamr_ has joined #openstack-manila01:28
*** akshai has joined #openstack-manila01:31
*** gouthamr has quit IRC01:31
*** mtanino has quit IRC01:37
*** akshai_ has joined #openstack-manila01:39
*** akshai has quit IRC01:42
gouthamr_bswartz: there?01:58
*** esker has joined #openstack-manila02:01
*** rhagarty_ has quit IRC02:04
*** esker has quit IRC02:07
*** a_ta has joined #openstack-manila02:07
*** a_ta has quit IRC02:11
*** rhagarty has joined #openstack-manila02:12
*** baojg has joined #openstack-manila02:17
*** rhagarty has quit IRC02:36
*** esker has joined #openstack-manila02:43
*** houming has joined #openstack-manila02:44
*** esker has quit IRC02:48
*** docaedo has joined #openstack-manila02:50
*** akshai_ has quit IRC02:51
*** akshai has joined #openstack-manila02:56
openstackgerritMerged openstack/manila: Remove unintended exposure of private attribute  https://review.openstack.org/28750602:58
*** akshai has quit IRC03:00
*** amrith is now known as _amrith_03:15
*** esker has joined #openstack-manila03:18
bswartzgouthamr_: sup?03:27
gouthamr_bswartz: hey! was thinking about deletion of snapshots for replicas and thought you'd be able to help me out...03:28
bswartzok03:28
gouthamr_bswartz: if i send all the snapshot instances to the active replica's host.. what does the driver return?03:30
gouthamr_bswartz: 1) return immediately 2) take its time for cleanup and return when done, or 3) return a list of snapshot instances it was able to clean up03:30
gouthamr_we discussed case 1 and we said we will poll for 'deleting' snapshots.. but isn't 3 a good thing too?03:31
gouthamr_bswartz: no hurry in letting me know; we can speak about this tomorrow too.03:33
*** gouthamr_ has quit IRC03:48
*** esker has quit IRC03:49
*** yangyapeng has joined #openstack-manila03:50
*** esker has joined #openstack-manila04:27
*** yangyape_ has joined #openstack-manila04:27
*** openstack has joined #openstack-manila14:09
*** dustins has joined #openstack-manila14:25
tbarronvponomaryov: jcsp: each instance of a driver (linux process) will have its own discrete set of shares, but can there be greenthread concurrent access to a single share within a given driver instance?14:26
tbarronvponomaryov: jcsp: you could outlaw this and enforce it in manila api14:26
tbarronbut in cinder there has been discussion of allowing some concurrent operations on a given volume14:27
tbarronactually more than discussion14:27
bswartztbarron: I think we want to move the opposite direction14:29
bswartzright now we don't enforce mutual exclusion as well as we should14:29
bswartzso any concurrent operations that we allow might be by accident14:29
tbarronbswartz: i'm not disagreeing, just posing the question given the ML discussion14:30
tbarronlife will certainly be simpler and it's complex enough14:31
tbarronanyways :)14:31
*** merooney has quit IRC14:32
*** merooney has joined #openstack-manila14:35
*** openstackgerrit_ has quit IRC14:36
openstackgerritOpenStack Proposal Bot proposed openstack/manila-ui: Updated from global requirements  https://review.openstack.org/28848414:37
*** openstackgerrit_ has joined #openstack-manila14:37
*** eharney has joined #openstack-manila14:39
*** rhagarty has joined #openstack-manila14:45
*** rhagarty_ has joined #openstack-manila14:46
openstackgerritAlexey Ovchinnikov proposed openstack/manila-image-elements: Add tox job for building LXD image  https://review.openstack.org/23214214:46
*** rhagarty has quit IRC14:48
jcspbswartz, tbarron: the #1 thing for me is just that the behaviour is defined; if it's more concurrent than we like then we can always add locks, but if we're not sure then it's hard14:50
bswartzdoh! why is proposal bot proposing g-r update NOW of all times?14:51
bswartzseems like the day after feature freeze isn't the smartest time to change requirements14:51
bswartzmaybe I'm missing something...14:51
openstackgerritDustin Schoenbrun proposed openstack/manila: Add thin provisioning column to support mapping  https://review.openstack.org/28850414:53
bswartztbarron, jcsp: I'm lacking context about your concern14:53
bswartzis there a specific operation you're concerned about?14:53
bswartzor just general angst about parallelism14:53
jcspbswartz: so I think we kind of need to know in general because otherwise it's hard to write correct code.  The specific case here is that to implement update_access() (and do some other stuff) we have to start persisting more metadata about our auth stuff14:54
jcspand when I'm doing that from the manila side, I'm writing some files to cephfs, and I'm also updating our authorization keys14:54
jcspand that has natural two phase commit type issues that you might imagine14:54
*** cknight has joined #openstack-manila14:55
bswartzjcsp: fwiw I agree that defined behavior is better than undefined behavior -- and we have a lot of poorly documented behaviors14:55
bswartzokay14:55
jcspfor most stuff we use existing drivers to cover the gaps in docs, but for concurrency stuff that's not possible14:55
bswartzwell during newton, address concurrency issues at all level is likely to be a top priority14:55
jcspcool14:55
jcspnot trying to beat anyone up about it, just gathering what info I can14:55
bswartzpart of that must be writing down more detail definitions of how stuff *should* work14:56
vponomaryov1jcsp: note about concurrency: ML answer related to several "ensure_share" calls, but if you do some other call, i.e. deletion it may be concurrent14:56
*** a_ta has joined #openstack-manila14:57
vponomaryov1jcsp: so, from driver point of view, you can set locks based on share ID14:57
vponomaryov1jcsp: and it should be enough14:57
*** martyturner has quit IRC14:57
vponomaryov1mkoderer: ping14:57
*** merooney has quit IRC14:58
vponomaryov1mkoderer__: ping14:58
vponomaryov1mkoderer, mkoderer__: fix of bug filed by you - https://bugs.launchpad.net/manila/+bug/1553082 will require fix on nova and cinder clients that are used in Generic driver14:59
openstackLaunchpad bug 1553082 in Manila "Keystone v3 support broken" [Undecided,New] - Assigned to Marc Koderer (m-koderer)14:59
bswartzmarkstur_ toabctl xyang1: I'd like to speed this change through the gate in order to test a fix to the "post" job for manila-image-elements: https://review.openstack.org/#/c/28837814:59
* bswartz wishes it were easier to test post jobs15:00
xyang1bswartz: ok, looking15:00
toabctldone :)15:00
bswartzty15:00
vponomaryov1yeah, it takes for 5 minutes against 9,5 as it is now15:00
xyang1toabctl: you beat me:)15:01
toabctl:)15:01
bswartzvponomaryov1: regarding your comment about PS5 failure on my bugfix https://review.openstack.org/#/c/287316 that failure was because my fix was incorrect, and I added PS615:02
*** timcl has quit IRC15:02
*** timcl has joined #openstack-manila15:03
*** rraja has quit IRC15:03
vponomaryov1bswartz: even so, itis safer to keep retry15:03
bswartzactually no, the incorrect fix was in 415:04
bswartzso 5 shold have worked15:04
bswartzI'm going to look again15:04
*** martyturner has joined #openstack-manila15:04
*** merooney has joined #openstack-manila15:04
bswartzokay you're right15:06
bswartzI can see failures even with patchset 515:06
bswartzthis makes me worry about how correct the fix is15:06
bswartzI wonder what could cause the dataset to remain busy if not another namespace15:07
bswartzvponomaryov1: when you were debugging busy dataset problem initially, what tools did you use to track down the cause of the problem?15:07
vponomaryov1bswartz: I did search for mounts in "/proc/*/mounts" restarting neutron dhcp service after dataset creation15:09
vponomaryov1bswartz: also, sometimes it returns busy just because some other dataset being created15:10
vponomaryov1bswartz: so, I would say absence of "retry" will make "deletion of dataset" very fragile15:10
*** esker has joined #openstack-manila15:14
openstackgerritMerged openstack/manila-image-elements: Disable ZFS element by default  https://review.openstack.org/28837815:14
*** esker has quit IRC15:15
*** esker has joined #openstack-manila15:15
*** jasonsb has joined #openstack-manila15:16
*** timcl has quit IRC15:17
vponomaryov1bswartz: after merge of commit to manila-image-elements, previously existed image gone from http://tarballs.openstack.org/manila-image-elements/images/15:21
vponomaryov1bswartz: there was 'manila-service-image-master.qcow2'15:21
vponomaryov1bswartz: and for this last commit job "gate-manila-publishimage-generic" has not started yet15:22
*** david_lyle__ has joined #openstack-manila15:34
*** rhagarty_ has quit IRC15:36
*** david-lyle has quit IRC15:37
*** yangyapeng has joined #openstack-manila15:38
*** yangyapeng has quit IRC15:39
*** lpetrut has quit IRC15:39
*** lpetrut has joined #openstack-manila15:39
*** lpetrut has quit IRC15:47
*** mtanino has joined #openstack-manila15:49
*** rhagarty has joined #openstack-manila15:51
*** rhagarty_ has joined #openstack-manila15:52
*** merooney has quit IRC15:54
*** martyturner has quit IRC15:54
*** rhagarty has quit IRC15:56
*** timcl has joined #openstack-manila15:56
*** martyturner has joined #openstack-manila15:58
*** merooney has joined #openstack-manila15:58
*** merooney has quit IRC16:13
*** david_lyle__ is now known as david_lyle16:17
openstackgerritBen Swartzlander proposed openstack/manila: Use official location for service image  https://review.openstack.org/28856416:21
bswartzvponomaryov1: ?16:21
bswartzvponomaryov1: I don't see any problems16:21
*** timcl has quit IRC16:22
bswartzmy commit above tests the new service image16:22
bswartzvponomaryov: ^16:22
vponomaryov1bswartz: problem is in absense of the image while "publish" job is not finished16:25
vponomaryov1bswartz: it is "hours"16:25
vponomaryov1bswartz: and generic driver jobs will fail a lot16:26
vponomaryov1bswartz: after each new commit merged to elements project16:26
bswartzokay, so we need a way to atomically swap in new images16:26
bswartzor at least in a small window16:26
bswartzI suppose it's okay if 1 or 2 jobs fail during a job upload16:27
bswartzhow can it take hours to scp a 300MB file?16:27
bswartzthat seems broken16:27
bswartzone moment....16:27
vponomaryov1bswartz: not copy - "waiting in queue"16:27
bswartzyeah but the old image doesn't disappear while it's in the queue16:28
vponomaryov1bswartz: I observed it16:28
*** rhagarty has joined #openstack-manila16:28
bswartzwhat did you observe? no manila-service-image-master.qcow2 file in directory? or whole directory missing?16:29
vponomaryov1yes, no file16:29
bswartzI *just* fixed the job that updates the image16:29
bswartzso that was the first upload ever16:29
bswartznext upload should simply overwrite file16:29
vponomaryov1this image was before first merge16:29
vponomaryov1that I observed too16:29
bswartzhttp://logs.openstack.org/92/9291b4781e4a10e9343528925fb120844f496925/post/gate-manila-publishimage-generic/8d5b24c/console.html#_2016-03-04_15_44_15_96216:30
vponomaryov1it disappeared after merge and till end of publish job16:30
*** rhagarty_ has quit IRC16:30
bswartzupload appears to be 40 seconds16:30
vponomaryov1bswartz: but this job ran after hour+16:30
vponomaryov1bswartz: waiting in queue16:30
bswartzyes but there was no file there before the job ran16:30
bswartzlet's merge another change to tests this16:31
bswartzI think it's just a fluke because this is the first patch to merge since my change to the job16:31
vponomaryov1bswartz: could be someone copied it and renamed to second image that is there16:31
bswartzthe other file there is very old -- last november16:31
bswartzI'm not sure where that came from16:32
vponomaryov1bswartz: ok, let's think it was some manual change16:32
bswartzI think it must have been16:32
bswartzstill, let's merge a dummy change16:32
bswartzor fix some small bug in manila-image-elements16:32
bswartzand watch what happens16:32
bswartzdo you have any other small bugs to fix?16:33
vponomaryov1me?16:34
bswartzyeah16:34
bswartzor anyone16:34
vponomaryov1yes16:34
vponomaryov1https://bugs.launchpad.net/manila/+bugs?field.tag=zfsonlinux16:34
bswartzI meant bugs for manila-image-elements16:35
vponomaryov1oh ))16:35
bswartzI want to push another commit and observe the tarballs.o.o site16:35
vponomaryov1bswartz: for example: add info about target palce where newest version is stored16:37
vponomaryov1to https://github.com/openstack/manila-image-elements/blob/master/README.rst16:37
bswartzgood idea16:37
bswartzokay 1 sec16:37
openstackgerritBen Swartzlander proposed openstack/manila-image-elements: Provide link to where images are uploaded  https://review.openstack.org/28857316:40
bswartzvponomaryov1: ^16:40
bswartzpls don't workflow until I'm back from lunch16:41
bswartzor unless you have time to watch the server16:41
bswartzsomeone needs to watch zuul while the job is traveling through the gate->post pipelines16:41
bswartzI'll bbiab16:41
*** ayma has joined #openstack-manila16:43
*** rhagarty_ has joined #openstack-manila16:45
*** mlima has joined #openstack-manila16:45
*** rhagarty has quit IRC16:47
*** fthiagogv__ is now known as fthiagogv16:47
*** ayma1 has joined #openstack-manila17:00
*** ayma has quit IRC17:00
*** ayma has joined #openstack-manila17:03
*** ayma1 has quit IRC17:06
*** absubram has joined #openstack-manila17:13
*** absubram_ has joined #openstack-manila17:14
*** absubram has quit IRC17:18
*** absubram_ is now known as absubram17:18
*** martyturner has quit IRC17:22
*** martyturner has joined #openstack-manila17:24
*** jasonsb has quit IRC17:26
openstackgerritValeriy Ponomaryov proposed openstack/manila: [DEBUG] do not merge 1  https://review.openstack.org/28655817:31
bswartzI'm back17:42
*** merooney has joined #openstack-manila17:46
bswartzvponomaryov, vponomaryov1: my manila-image-elements patch is in the gate17:46
bswartzI'll be watching tarballs site for modifications17:46
tpsilvacknight: ping17:48
cknighttpsilva: pong17:48
*** Yogi1 has joined #openstack-manila17:49
*** mlima has quit IRC17:50
tpsilvacknight: hello! sorry to bother, it's kinda offtopic actually... I see that you always complain about the code styles on the reviews, so maybe you'll know this. Why do we always use db models as dictionaries instead of objects? Is this a guideline or something will break if we don't use them like this?17:51
cknighttpsilva: Hi, Tiago.  Honestly I have no idea apart from inertia and consistency.  Perhaps bswartz or ameade could provide some historical insight.17:54
gouthamrtpsilva: i might have a point to add.. *never* send drivers SQLAlchemy objects17:54
*** akerr_ has quit IRC17:54
*** timcl has joined #openstack-manila17:55
vponomaryov1tpsilva: first of all, it does not pick up all the data from DB for all "relationships" and will try to get info when you "try to use it". BUT, in this cae, when ou provide it somewhere else, you get "detached" state and cannot get info anymore17:55
tpsilvacknight: consistency is a great reason by itself :)17:55
vponomaryov1tpsilva: so, you just fail to get required info when you send DB object17:55
vponomaryov1tpsilva: use always primitives17:55
openstackgerritTom Barron proposed openstack/manila: Fix unconditional pylxd unit test dependency  https://review.openstack.org/28861317:56
ameadetpsilva: there is a TODO from the very first impl in nova to have the db layer only return dictionaries17:57
ameadethat work has just never been done17:57
ameadeimo, returning sqlalchemy objects has caused a lot of pain17:57
ameadevponomaryov1: +117:58
ameadeonly reasoning i can imagine would be lazy loading but vponomaryov1's point is why that doesnt work anyways17:58
tpsilvaawesome, lots of info... was not expecting that17:58
vponomaryov1tpsilva: and second - easy to log/handle instead of using "to_dict()" explciit translations17:58
tpsilvavponomaryov1: yep, in fact is really easier to log with dicts17:59
tpsilvagouthamr: the db layer already does this work for me, right?17:59
ameadetpsilva: no unfortunately17:59
gouthamrtpsilva: as of today, no.. there's a TODO kinda comment in db/api.py17:59
gouthamrtpsilva: https://github.com/openstack/manila/blob/master/manila/db/api.py#L2518:00
tpsilvaameade, gouthamr: oh wow... so the update_access is wrong? https://github.com/openstack/manila/blob/master/manila/share/access.py#L7418:00
bswartzthis discussion overlaps with the versioned objects project18:00
tpsilvarules is retrieved from the DB on LN 6018:01
gouthamrtpsilva: not wrong.. it is sending SQLAlchemy objects to the driver... as is the case with a bunch of methods18:01
bswartzversioned objects are better than dicts because they can emulate both new and old behaviors while both are needed during an upgrade18:01
*** timcl has quit IRC18:02
*** timcl1 has joined #openstack-manila18:02
*** esker has quit IRC18:02
bswartzvponomaryov1: jobs is still queued in post pipe and no change to tarballs site yet18:02
gouthamrtpsilva: look at https://github.com/openstack/manila/blob/master/manila/share/manager.py#L2454 -> not the grosest thing.. but a workaround for now.. we should solve this problem soon..18:03
bswartzgouthamr: there's no time like today...18:05
tpsilvagouthamr: I see... looks easy to be implemented on the DB layer18:05
tpsilvaawesome, thank you guys!18:06
vponomaryov1bswartz: then it was manual intruding18:06
* bswartz is still watching18:07
gouthamrbswartz: i've a post it note.. :P18:08
* bswartz wonders how long jobs remain queued in post pipe....18:10
bswartzso vponomaryov's patch merged 7 hours ago, but the last update to tarballs was just 2.25 hours ago18:11
*** merooney has quit IRC18:12
bswartzthat implies that job waited ~5 hours18:12
bswartzmaybe I won't watch that closely18:12
bswartzI have a 10 second cron job watching the directory on the server, I'll just let that run18:12
* gouthamr zuul wants some space18:13
tpsilvaanother thing: manila doesn't use versioned objects, right?18:14
gouthamrno tpsilva...18:16
dustinscknight: thanks for the feedback, I think I'm going to create a new table at the end of the feature support mapping18:21
*** merooney has joined #openstack-manila18:21
tpsilvagouthamr: thanks!18:22
cknightdustins: ok18:24
*** Yogi1 has quit IRC18:29
*** JoseMello has joined #openstack-manila18:34
*** akerr has joined #openstack-manila18:35
*** eharney has quit IRC18:35
gouthamrtpsilva: np!18:36
*** mlima has joined #openstack-manila18:39
*** yangyapeng has joined #openstack-manila18:40
*** pcaruana has quit IRC18:41
*** yangyapeng has quit IRC18:45
*** merooney has quit IRC18:46
*** merooney has joined #openstack-manila18:47
*** akerr has quit IRC18:57
*** akerr has joined #openstack-manila18:58
*** martyturner has quit IRC19:09
openstackgerritValeriy Ponomaryov proposed openstack/manila: [DEBUG] do not merge 1  https://review.openstack.org/28655819:12
*** a_ta has quit IRC19:12
*** ociuhandu has quit IRC19:13
*** a_ta has joined #openstack-manila19:13
*** eharney has joined #openstack-manila19:13
*** timcl1 has quit IRC19:17
*** a_ta has quit IRC19:17
*** fthiagogv has quit IRC19:22
openstackgerritAndrea Ma proposed openstack/manila: glusterManager instantiation regexp validation  https://review.openstack.org/28377719:23
bswartzvponomaryov1: it finally ran the job19:25
bswartzthere was a ~60 second window when the qcow2 file was being updated19:25
*** a_ta has joined #openstack-manila19:25
bswartzI've got logs to show it19:25
bswartzI think a 60 second window is small enough that problems are extremely unlikely, nevertheless I'll talk to the infra folks and see if we can get it down to zero19:26
vponomaryov1bswartz: whole 60 seconds19:27
*** ociuhandu has joined #openstack-manila19:31
*** Yogi1 has joined #openstack-manila19:32
*** merooney has quit IRC19:32
docaedobswartz: regarding linking to the image in https://apps.openstack.org, if you know what the URL is going to be now19:32
docaedobswartz: you could add it to assets.yaml and we'd be ready to merge when you're ready19:33
*** timcl has joined #openstack-manila19:33
docaedobswartz: also just waving my hand, I'm around for another 30, then teaching until 2pm PST after which time I'm back around for the day19:33
*** merooney has joined #openstack-manila19:34
*** martyturner has joined #openstack-manila19:41
openstackgerritValeriy Ponomaryov proposed openstack/manila: Download service image only when needed  https://review.openstack.org/28870519:49
*** martyturner has quit IRC19:54
*** Yogi1 has quit IRC19:55
*** dencaval has quit IRC19:56
openstackgerritValeriy Ponomaryov proposed openstack/manila: [DEBUG] do not merge 1  https://review.openstack.org/28655820:01
*** mlima has quit IRC20:02
*** timcl has quit IRC20:11
*** merooney has quit IRC20:19
*** merooney has joined #openstack-manila20:23
openstackgerritTom Barron proposed openstack/manila: Remove config generation pylxd hard dependency  https://review.openstack.org/28872120:25
bswartzdocaedo: yes I plan to push a change tehre20:27
*** lpetrut has joined #openstack-manila20:27
openstackgerritBen Swartzlander proposed openstack/manila: Allow devstack plugin to work without Cinder  https://review.openstack.org/27856220:32
openstackgerritBen Swartzlander proposed openstack/manila: Allow devstack plugin to work without Cinder  https://review.openstack.org/27856220:33
*** timcl has joined #openstack-manila20:35
bswartztbarron: ping re https://review.openstack.org/#/c/288721/20:36
tbarronbswartz: pong20:36
bswartztbarron: why does change need to be in opts.py and not down inside lxd driver?20:36
tbarronmaybe that would work, this is one way to do it20:37
bswartzwe've seen drivers before with "soft" dependencies20:37
bswartzperhaps LXD should be one of those20:37
tbarronthat might be cleaner20:37
bswartzI'm curious what the issue is with the pylxd lib -- is there a license problem? is redhat unable to distribute it due to patents?20:38
tbarroni'll mark it wip till i get a firmer opinion20:38
bswartzif there's a good reason redhat cannot distribute it, then I expect the TC would support not allowing it in reqs.txt20:38
tbarronit broke RDO nightly build20:38
tbarronpackaging people don't want to add rpms for stuff we aren't actually using20:39
tbarronor generate options for stuff we won't actually ship20:39
bswartzredhat won't ship the lxd driver?20:39
bswartzwill you delete it from your downstream branch?20:39
tbarrononly canonical has support for lxd20:39
tbarronnot even debian20:40
tbarronthat could change, but that's the way it is today20:40
tbarroni don't think we'll remove the driver20:41
tbarronbut it won't be "first party" in our distro20:41
bswartzI know the project was started by canonical -- does redhat have something that competes?20:41
tbarroni don't know the answers to the product manager questions :-)20:41
tbarronmy only role in this is to get our build working20:42
tbarronalso, rdo will be shipping unit tests soon. hence https://review.openstack.org/#/c/288613/20:44
bswartzwell it matters because the community tends to push back on requirements that are not widely available20:44
bswartzmy personal tendency is to go with whatever works20:44
bswartzbut we need to be careful and make sure everyone is happy20:44
tbarronbswartz: I agree with that perception, am just trying to be neutral/technical on the actual concrete problems.20:45
bswartzany dependencies we're invoking that aren't widely available should be turned to soft dependencies20:45
bswartzto avoid exactly this kind of technical issue20:45
*** merooney has quit IRC20:55
openstackgerritValeriy Ponomaryov proposed openstack/manila: [DEBUG] do not merge 1  https://review.openstack.org/28655820:56
*** timcl has quit IRC21:02
*** rhagarty_ has quit IRC21:07
*** rhagarty_ has joined #openstack-manila21:07
*** JoseMello has quit IRC21:08
*** vponomaryov1 has left #openstack-manila21:11
*** akerr has quit IRC21:34
*** porrua has quit IRC21:39
*** gouthamr has quit IRC21:43
openstackgerritDustin Schoenbrun proposed openstack/manila: Add thin provisioning column to support mapping  https://review.openstack.org/28850421:51
*** pcaruana has joined #openstack-manila21:58
*** dustins has quit IRC22:10
*** gouthamr has joined #openstack-manila22:23
*** akshai has quit IRC22:25
*** tpsilva has quit IRC22:27
*** erlon has quit IRC22:46
*** sbjorn has quit IRC22:48
*** a_ta has quit IRC22:54
*** lpetrut has quit IRC23:03
*** sedwards has joined #openstack-manila23:07
*** pgbridge has quit IRC23:26
*** xyang1 has quit IRC23:29
aymaseems like gate-manila-tempest-dsvm-neutron-multibackend is failing?  looking at the logs it seems to say23:50
ayma manila_tempest_tests.share_exceptions.ShareBuildErrorException23:51
aymawas something changed with the gate?23:51
openstackgerritTom Barron proposed openstack/manila: Fix pylxd hard dependencies  https://review.openstack.org/28880023:56
tbarronbswartz: not sure how you want to do soft depdencies, but here's a strawman23:57
tbarron^^23:57
« Thursday, 2016-03-03 Index Saturday, 2016-03-05 »

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