Tuesday, 2016-03-29

*** resker has joined #openstack-manila00:08
*** ociuhandu has quit IRC00:10
*** vbellur has quit IRC00:12
*** esker has quit IRC00:12
*** ociuhandu has joined #openstack-manila00:12
*** vbellur has joined #openstack-manila00:13
cfoutsscottda: hi00:16
cfoutsscottda: because that is what novaclient had used at the time. I see that they replaced it though: https://github.com/openstack/python-novaclient/pull/1700:17
cfoutsscottda: looks like I have a new bug to open00:17
*** a_ta has joined #openstack-manila00:29
*** mtanino_ has joined #openstack-manila00:29
*** mtanino has quit IRC00:31
*** a_ta has quit IRC00:34
*** esker has joined #openstack-manila00:47
*** swamireddy1 has joined #openstack-manila00:47
*** qeelee has joined #openstack-manila00:49
*** akshai has joined #openstack-manila00:50
*** resker has quit IRC00:51
*** akshai_ has joined #openstack-manila00:58
*** akshai has quit IRC01:01
*** esker has quit IRC01:14
*** qeelee has quit IRC01:17
*** qeelee has joined #openstack-manila01:19
*** bill_az has quit IRC01:34
*** swamireddy1 has quit IRC01:36
*** mtanino_ has quit IRC01:43
*** bill_az has joined #openstack-manila01:52
*** qeelee has quit IRC01:55
*** cknight has joined #openstack-manila02:11
*** qeelee has joined #openstack-manila02:15
*** houming has joined #openstack-manila02:23
*** bill_az has quit IRC02:23
*** baojg has joined #openstack-manila02:29
*** cknight has quit IRC02:42
*** baojg has quit IRC02:45
*** baojg has joined #openstack-manila02:46
*** mtanino has joined #openstack-manila02:51
*** qeelee has quit IRC02:52
*** gouthamr has quit IRC02:52
*** qeelee has joined #openstack-manila02:56
*** erlon has quit IRC03:07
*** a_ta has joined #openstack-manila03:31
*** a_ta has quit IRC03:36
*** qeelee has quit IRC03:52
*** sgotliv has quit IRC04:19
*** sgotliv has joined #openstack-manila04:20
*** swamireddy1 has joined #openstack-manila04:28
*** akshai_ has quit IRC04:29
*** swamireddy2 has joined #openstack-manila04:30
*** swamireddy1 has quit IRC04:33
*** pcaruana has quit IRC05:09
*** lpetrut has joined #openstack-manila05:10
*** shausy has joined #openstack-manila05:15
*** baojg has quit IRC05:24
*** baojg has joined #openstack-manila05:25
*** lpetrut has quit IRC05:57
*** nkrinner has joined #openstack-manila06:02
*** lpetrut has joined #openstack-manila06:06
*** sgotliv has quit IRC06:17
*** mkoderer has quit IRC06:46
*** tpatzig has quit IRC06:46
*** mkoderer has joined #openstack-manila06:49
*** tpatzig has joined #openstack-manila06:50
openstackgerritAnh Tran proposed openstack/manila: Removing some redundant words  https://review.openstack.org/29857207:20
*** lpetrut has quit IRC07:34
*** pcaruana has joined #openstack-manila07:48
*** sgotliv has joined #openstack-manila07:53
*** lpetrut has joined #openstack-manila08:02
openstackgerritAnh Tran proposed openstack/manila: Removing some redundant words  https://review.openstack.org/29857208:10
*** digvijay has joined #openstack-manila08:24
*** mtanino has quit IRC08:24
*** mtanino has joined #openstack-manila08:25
*** rraja has joined #openstack-manila08:27
*** lpetrut has quit IRC08:29
digvijayfailed: driver does not expect share-network to be provided with current configuration.08:31
digvijayanyone knows about this error?08:31
*** jcsp has joined #openstack-manila08:33
*** jcsp has quit IRC08:33
*** jcsp has joined #openstack-manila08:33
openstackgerritFelix Ma proposed openstack/manila: Generic dirver:ignore VolumeNotFound in deleting  https://review.openstack.org/29860708:57
openstackgerritzhongjun proposed openstack/manila: Add user_id echo in manila show api  https://review.openstack.org/29811509:07
*** lpetrut has joined #openstack-manila09:18
*** johnthetubaguy has left #openstack-manila09:23
vponomaryovdigvijay: then do not provide share-network09:29
*** johnthetubaguy has joined #openstack-manila09:41
*** baojg has quit IRC09:42
*** baojg has joined #openstack-manila09:42
*** baojg has quit IRC09:44
*** baojg has joined #openstack-manila09:45
openstackgerritMarc Koderer proposed openstack/manila: [WIP] Add multi-segment support  https://review.openstack.org/27773109:49
openstackgerritMarc Koderer proposed openstack/manila: [WIP] Add multi-segment support  https://review.openstack.org/27773110:00
*** ganso has joined #openstack-manila10:54
*** ociuhandu has quit IRC10:57
*** baojg has quit IRC10:59
*** tpsilva has joined #openstack-manila11:26
*** ociuhandu has joined #openstack-manila11:26
openstackgerritFelix Ma proposed openstack/manila: Generic dirver: ignore VolumeNotFound in deleting  https://review.openstack.org/29860711:31
openstackgerritFelix Ma proposed openstack/manila: Generic dirver: ignore VolumeNotFound in deleting  https://review.openstack.org/29860711:34
*** digvijay has quit IRC11:46
*** timcl has joined #openstack-manila11:47
openstackgerritRamana Raja proposed openstack/manila: Doc fixes for CephFS native driver  https://review.openstack.org/29544811:52
openstackgerritRamana Raja proposed openstack/manila: Doc fixes for CephFS native driver  https://review.openstack.org/29544812:06
openstackgerritFelix Ma proposed openstack/manila: Generic driver: ignore VolumeNotFound in deleting  https://review.openstack.org/29860712:36
*** baojg has joined #openstack-manila12:36
*** xyang1 has joined #openstack-manila12:40
*** gouthamr has joined #openstack-manila12:45
*** houming has quit IRC12:55
*** akerr has joined #openstack-manila12:59
*** rraja has quit IRC13:01
*** cknight has joined #openstack-manila13:02
*** porrua has joined #openstack-manila13:10
*** dencaval has quit IRC13:11
*** rraja has joined #openstack-manila13:15
*** bill_az has joined #openstack-manila13:19
*** dencaval has joined #openstack-manila13:25
*** absubram has joined #openstack-manila13:34
*** dustins has joined #openstack-manila13:34
*** absubram_ has joined #openstack-manila13:35
*** absubram has quit IRC13:40
*** absubram_ is now known as absubram13:40
openstackgerritMarc Koderer proposed openstack/manila: Add binding_profile option for backends  https://review.openstack.org/28403413:45
openstackgerritMarc Koderer proposed openstack/manila: Add neutron driver for binding  https://review.openstack.org/28349413:45
*** mtanino has quit IRC13:49
*** akshai has joined #openstack-manila13:57
*** akshai_ has joined #openstack-manila13:59
*** mkoderer__ has quit IRC13:59
*** mkoderer__ has joined #openstack-manila14:02
*** akshai has quit IRC14:02
*** Yogi1 has joined #openstack-manila14:02
*** sgotliv has quit IRC14:03
openstackgerritMarc Koderer proposed openstack/manila: [WIP] Add multi-segment support  https://review.openstack.org/27773114:04
*** absubram has quit IRC14:04
*** martyturner has joined #openstack-manila14:07
*** eharney has joined #openstack-manila14:13
*** sgotliv has joined #openstack-manila14:20
*** a_ta has joined #openstack-manila14:27
*** timcl has quit IRC14:28
*** timcl has joined #openstack-manila14:31
openstackgerritChuck Fouts proposed openstack/python-manilaclient: UnsupportedVersion Thrown for Supported Version  https://review.openstack.org/29878014:32
bswartzxyang1 vponomaryov cknight ganso toabctl markstur: ^ this fix is something we should seriously consider for RC2 -- please review it14:33
*** mtanino has joined #openstack-manila14:34
*** baojg has quit IRC14:34
*** rraja has quit IRC14:43
xyang1bswartz: any specific rule on what patch can be merged?14:47
bswartzxyang1: we need to get it merged to newton before we can consider it for mitaka-rc214:48
bswartzso first priority is review it and get it into newton14:49
xyang1bswartz: ok14:49
bswartzpersonally I lean towards cutting an RC2 of the client that includes this bugfix14:49
xyang1bswartz: what about server side14:50
bswartzso far I haven't seen any issues serious enough to consider an RC2 of the server14:52
*** qeelee has joined #openstack-manila14:53
*** openstackgerrit has quit IRC15:06
*** openstackgerrit has joined #openstack-manila15:07
cfoutsxyang1: thanks for the review. I've updated the commit message on 29878015:07
*** nkrinner has quit IRC15:08
*** Yogi1 has quit IRC15:09
openstackgerritFelix Ma proposed openstack/manila: Generic driver: ignore VolumeNotFound in deleting  https://review.openstack.org/29860715:10
*** qeelee has quit IRC15:14
*** ociuhandu_ has joined #openstack-manila15:20
*** ociuhandu has quit IRC15:21
*** ociuhandu_ is now known as ociuhandu15:21
xyang1cfouts: have you published your commit message?  I have not seen a new patch set15:27
*** shausy has quit IRC15:28
cfoutsyes, you have to take the /1 off the URI in order to see it15:29
xyang1cfouts: I removed 1 but still can only see 1 patch set15:31
openstackgerritChuck Fouts proposed openstack/python-manilaclient: UnsupportedVersion Thrown for Supported Version  https://review.openstack.org/29878015:32
cfoutsxyang1: sorry, missed pushing the "publish edit" button :/15:32
cfoutsfixed now15:33
xyang1cfouts: Ok, I can see patch set 2 now. no problem15:33
openstackgerritRodrigo Barbieri proposed openstack/manila: Fix Manage API synchronous call  https://review.openstack.org/29670615:37
bswartzganso: you were looking for me yesterday?15:44
gansobswartz: yes15:45
vponomaryovcfouts: not sure it is good idea to change microversion15:45
vponomaryovcfouts: maybe we should require it be set explicitly?15:45
vponomaryovcfouts: via env var?15:45
vponomaryovcfouts: if not provided15:45
vponomaryovcfouts: via CLI15:46
gansobswartz: I am listing retype scenarios, I am wondering if your backend can successfully move files around and not need to change or update an export path, thus being undisruptive15:46
vponomaryovcfouts: as we do for tenant/project ID, username, etc15:46
bswartzganso: it depends15:46
*** swamireddy2 has left #openstack-manila15:47
bswartzin some cases, the answer is yes15:47
vponomaryovcfouts: and, in this case, upgrade to newer client will not change behaviour (because of another microversion)15:47
bswartzhowever in most cases it would be no15:47
gansobswartz: in which cases would it work?15:47
bswartzganso: if the retype involved just flipping some bits on the storage, then obviously there would be no data copying and the export location could stay the same -- also, even if we need to move the data, if the destination is on the same storage cluster we can copy it internally and the outside world doesn't know it's moving15:49
bswartzhowever there will clearly be cases when a retype will unavoidably involve changing the export location15:50
gansobswartz: yes, I am considering the case where the host and pool is still the same, like, for instance, the share type change is dedupe on/off, so the backend may need to re-create the share, copy the contents, but I am assuming that the export location changes, but how can it not be disruptive?15:50
bswartznetapp could support a case where the pool was different but the export location didn't change15:51
bswartzthat's because pools and "share servers" are orthagonal concepts15:52
bswartza share server can span more than 1 pool so as long as the source and destination pools are both on the same share server, then the transition can be done nondisruptively15:52
gansobswartz: I see, so the export location definition is still the same IP and folder, but you are just pointing to other pool?15:53
bswartzconsider a storage controller with a bunch of spinning disks and a bunch of flash15:53
bswartzyou can move back and forth between the 2 types of storage but from the outside it looks like it never moves (it just gets faster/slower)15:53
bswartzganso: yes15:53
gansobswartz: ok, I will consider this scenario then15:54
gansobswartz: thanks!15:54
bswartzganso: I have a question about the manage share bugfix you've been working on15:56
gansobswartz: sure15:56
bswartzdoes the latest fix return 202 and create the share then let it go into an error state if the share type is invalid?15:56
gansobswartz: yes15:56
bswartzokay that's what I thought and that's what I was hoping for15:57
bswartzjust wanted to check with you15:57
*** esker has joined #openstack-manila16:03
marksturcfouts, ping16:05
cfoutsmarkstur: pong16:07
vponomaryovmarkstur, cfouts: about fix for manilaclient, don't you think, that it is better to require appropriate env var be set as for project ID or username instead of getting another, not requested one microversion?16:07
vponomaryovbswartz: ^16:07
cfoutsvponomaryov: I'm not sure why we wouldn't want the mitaka client to automatically work with the liberty manila server16:08
cfoutsThis patch changes the microversion, used by the client, to the max version supported by the server16:08
cfoutswhich should be what the user wants16:08
cfoutsI would expect that if I where using the latest version of the client with a liberty server16:09
*** esker has quit IRC16:13
*** esker has joined #openstack-manila16:13
*** a_ta has quit IRC16:13
*** a_ta has joined #openstack-manila16:14
marksturvponomaryov, Making the client work seems good. Is there a use case where this is bad behavior?16:15
marksturvponomaryov, I guess the client caller might expect functionality to be a little different?16:16
*** a_ta has quit IRC16:18
*** a_ta has joined #openstack-manila16:20
*** timcl has quit IRC16:29
*** a_ta has quit IRC16:48
*** pcaruana has quit IRC16:48
*** a_ta has joined #openstack-manila16:48
tpsilvabswartz: ping16:50
tpsilvabswartz: quick question: if a patch is merged before a vendor's CI checked it, does the ci still have to check it? even after it got merged?16:51
tpsilvais there any rules or guidelines from openstack for that?16:51
*** a_ta has quit IRC16:52
bswartztpsilva: that's a good question16:54
bswartztpsilva: if your CI has a queuing discipline I'd certainly deprioritize those changes16:55
bswartzbut if you have spare cycles it's better to do the run and vote for a better historical record16:55
*** erlon has joined #openstack-manila16:55
bswartzvponomaryov: my opinion is that we should try to maximize backwards compatibility16:56
tpsilvabswartz: right, thank you. Running all the patches seems like the correct thing to do16:56
bswartzold scripts that work on old servers should ideally continue to work exactly as before even if you swap in newer client version16:56
vponomaryovbswartz: after some thinking I came to idea, that we need such env var in general not as bugfix16:57
vponomaryovbswartz: when we want to use not latest and don't want to provide it via CLI16:57
vponomaryovbswartz: each time16:57
bswartzCLI users should not be specifying version number16:58
bswartzthey should just use the latest16:58
vponomaryovbswartz: why?16:58
bswartzpython code that imports the lib is more likely to care about exact version16:58
bswartzbecause typically a CLI user is a human, who can respond to changed parameters by reading the help text16:58
*** sgotliv has quit IRC16:59
bswartznon-human CLI users (such a bash scripts) do need a way to lock down the version they get16:59
vponomaryovbswartz: so, manilaclient will set this env var if not set by user for itself17:00
*** martyturner has quit IRC17:01
vponomaryovbswartz: to reduce amount of requests to server17:01
*** tpsilva has quit IRC17:07
*** kambiz has quit IRC17:07
*** mtanino has quit IRC17:07
*** docaedo has quit IRC17:07
*** akshai_ has quit IRC17:07
*** vbellur has quit IRC17:07
*** dhellmann has quit IRC17:07
*** darrenc has quit IRC17:07
*** tbarron has quit IRC17:07
*** vkmc has quit IRC17:07
*** cfouts has quit IRC17:07
*** dencaval has quit IRC17:07
*** tbarron has joined #openstack-manila17:07
vponomaryovbswartz: starting from second17:07
bswartzyou're suggesting that the client allows the user to specify the requested microversion 2 ways?17:07
bswartzby CLI arg and by env var?17:07
vponomaryovyes17:07
bswartzthat sounds fine to me17:07
*** martyturner has joined #openstack-manila17:07
bswartzbut if it's not already implemented it should not be added in mitaka -- it should be a feature for newton17:07
vponomaryovconsider all releases ))17:07
vponomaryovin most cases latest client will be used17:07
bswartzI would be nice to have the feature in all releases, but I don't want to bend to rules to backport a feature17:07
*** akshai has joined #openstack-manila17:07
*** mtanino_ has joined #openstack-manila17:07
bswartzso it will have to go into newton17:07
vponomaryovI am ok with it17:07
*** gnarld_ has joined #openstack-manila17:07
*** kambiz_ has joined #openstack-manila17:07
*** kambiz_ is now known as kambiz17:07
*** dhellmann has joined #openstack-manila17:07
*** dencaval has joined #openstack-manila17:08
*** darrenc has joined #openstack-manila17:08
*** vbellur has joined #openstack-manila17:09
*** vkmc has joined #openstack-manila17:09
*** timcl has joined #openstack-manila17:09
*** tpsilva has joined #openstack-manila17:17
*** docaedo_ has joined #openstack-manila17:30
*** docaedo_ is now known as docaedo17:30
*** eharney has quit IRC17:32
*** lpetrut has quit IRC17:43
*** a_ta has joined #openstack-manila17:50
*** martyturner has quit IRC17:52
*** a_ta has quit IRC17:54
*** pcaruana has joined #openstack-manila18:02
*** martyturner has joined #openstack-manila18:03
*** lpetrut has joined #openstack-manila18:15
*** martyturner has quit IRC18:18
*** eharney has joined #openstack-manila18:21
*** a_ta has joined #openstack-manila18:31
*** david-lyle has quit IRC18:45
*** david-lyle has joined #openstack-manila18:46
*** krotscheck is now known as krotscheck_dcm19:07
*** ociuhandu has quit IRC19:21
*** cfey_ has joined #openstack-manila19:33
*** cfey has quit IRC19:33
gouthamrganso: ping19:39
gansogouthamr: pong19:39
gouthamrganso: hey! quick q..19:39
gouthamrganso: https://review.openstack.org/#/c/296706/10/manila_tempest_tests/tests/api/admin/test_share_manage.py19:39
gouthamrganso: it would be nice to split that test into two.. so you can avoid the loop..19:40
gansogouthamr: yes, I first wanted to test if that solution worked19:40
gouthamrganso: it works :)19:40
gansogouthamr: second, it will imply in creating another file, because one of the tests is negative19:41
gansogouthamr: if I create another file, it will have to create another share, and it will take a bit longer, and the no-share-servers CI is already timing out, twice in a row19:41
gouthamrganso: we had a negative test19:41
gansogouthamr: we do?19:42
gouthamrganso: in the same file.. (in the same test)..19:42
gouthamrganso: i guess when you have more negative tests, it makes sense to split them out into separate files.. if you just have one, it wouldn't.19:42
gansogouthamr: which test is that?19:42
*** timcl has quit IRC19:43
gansogouthamr: oh, you mean in other cases?19:43
gouthamrganso: the same test had self.assertRaises(lib_exc.Conflict, ..19:43
*** a_ta has quit IRC19:43
gansogouthamr: ok, if it stays in the same file it is ok19:43
gansogouthamr: I will send update right away19:44
gouthamrganso: guess you added that when the scheduler was being 'call'ed..19:44
gouthamrganso: thank you! :)19:44
*** a_ta has joined #openstack-manila19:44
gansogouthamr: no, I restored the original test19:44
gansogouthamr: and for some reason it is not working19:44
gansogouthamr: it *was19:44
gouthamrganso: that is strange.. but this works?19:45
gansogouthamr: it required changes19:46
gansogouthamr: latest patch works19:46
gansogouthamr: but original one did not19:46
gouthamrganso: yep.. because of the double cleanup..19:46
gansogouthamr: previous one also required double cleanup19:46
*** jcsp has quit IRC19:47
gansogouthamr: API behavior was changed in the bugfix that merged in RC1, but this fix restored original behavior19:47
*** sgotliv has joined #openstack-manila19:48
gouthamrganso: yep.. the behavior's same. i hope this gets back-ported19:48
*** a_ta has quit IRC19:48
gansogouthamr: I hope so too19:49
*** a_ta has joined #openstack-manila19:54
*** ociuhandu has joined #openstack-manila19:57
*** akerr has quit IRC19:57
openstackgerritRodrigo Barbieri proposed openstack/manila: Fix Manage API synchronous call  https://review.openstack.org/29670619:57
gansogouthamr: ^19:58
*** jcsp has joined #openstack-manila20:02
*** porrua has quit IRC20:05
*** esker has quit IRC20:12
*** esker has joined #openstack-manila20:13
*** abhi has quit IRC20:32
*** abhi has joined #openstack-manila20:33
*** cknight has quit IRC20:38
*** lpetrut has quit IRC20:57
*** a_ta has quit IRC21:25
*** a_ta has joined #openstack-manila21:26
*** mtanino__ has joined #openstack-manila21:27
*** a_ta has quit IRC21:31
*** eharney has quit IRC21:31
*** esker has quit IRC21:43
*** gouthamr has quit IRC21:46
*** xyang1 has quit IRC21:54
*** esker has joined #openstack-manila22:02
*** gouthamr has joined #openstack-manila22:03
*** gouthamr_ has joined #openstack-manila22:05
*** tpsilva has quit IRC22:07
*** dustins has quit IRC22:07
*** gouthamr has quit IRC22:08
*** sgotliv has quit IRC22:35
*** esker has quit IRC22:48
*** esker has joined #openstack-manila22:57
*** esker has quit IRC23:11
*** ganso has quit IRC23:20
*** bill_az has quit IRC23:38

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