Thursday, 2016-09-22

*** akerr has joined #openstack-manila00:11
*** akerr has quit IRC00:18
*** kaisers1 has joined #openstack-manila00:23
*** kaisers2 has joined #openstack-manila00:23
*** kaisers has quit IRC00:25
*** kaisers3 has quit IRC00:25
*** gouthamr has quit IRC01:20
openstackgerritMerged openstack/python-manilaclient: Add __ne__ built-in function  https://review.openstack.org/35515701:46
*** StraubTW has quit IRC02:17
*** david-lyle has quit IRC03:03
openstackgerritYingzhe Zeng proposed openstack/manila: Fix huawei driver cannot delete qos while status is idle  https://review.openstack.org/37287103:04
openstackgerritYingzhe Zeng proposed openstack/manila: Fix huawei driver cannot delete qos while status is idle  https://review.openstack.org/37287103:06
openstackgerritHa Van Tu proposed openstack/manila: [api-ref] Refactor Malina availability-zones API  https://review.openstack.org/37451703:11
*** akshai has joined #openstack-manila03:33
*** daidv has joined #openstack-manila03:34
*** akshai has quit IRC03:37
*** rooneym has quit IRC03:39
*** daidv has quit IRC03:48
*** akapil has joined #openstack-manila04:15
*** kberger has quit IRC04:15
*** kberger has joined #openstack-manila04:16
*** StraubTW has joined #openstack-manila04:17
*** akapil has quit IRC04:19
*** StraubTW has quit IRC04:23
*** shausy has joined #openstack-manila04:37
*** kberger has quit IRC04:45
*** kberger has joined #openstack-manila04:46
*** kaisers has joined #openstack-manila05:45
*** lpetrut has joined #openstack-manila06:14
openstackgerritAlexey Ovchinnikov proposed openstack/manila: Bring remote and local executors into accord  https://review.openstack.org/37239606:23
*** aovchinnikov has joined #openstack-manila06:28
*** lpetrut has quit IRC07:00
*** lpetrut has joined #openstack-manila07:12
openstackgerritAlexey Ovchinnikov proposed openstack/manila: Switch scenario tests on for Zfsonlinux driver  https://review.openstack.org/37458107:22
*** nherciu has joined #openstack-manila07:24
*** lpetrut has quit IRC07:56
*** dsariel has joined #openstack-manila07:59
*** StraubTW has joined #openstack-manila08:21
*** narayrak has joined #openstack-manila08:21
*** StraubTW has quit IRC08:26
*** senk has joined #openstack-manila08:35
*** shausy has quit IRC09:01
*** shausy has joined #openstack-manila09:01
*** zhongjun_ has quit IRC09:03
*** nherciu_ has joined #openstack-manila09:03
*** zhongjun_ has joined #openstack-manila09:04
*** nherciu has quit IRC09:04
*** lpetrut has joined #openstack-manila09:11
*** ig0r_ has joined #openstack-manila09:21
*** sticker__ is now known as sticker09:22
*** zengyingzhe has quit IRC09:33
*** zengyingzhe has joined #openstack-manila09:34
*** mkoderer has joined #openstack-manila09:47
openstackgerritAlexey Ovchinnikov proposed openstack/manila: Switch scenario tests on for Zfsonlinux driver  https://review.openstack.org/37458110:25
openstackgerritOpenStack Proposal Bot proposed openstack/python-manilaclient: Updated from global requirements  https://review.openstack.org/37378010:47
*** kberger has quit IRC10:47
*** kberger has joined #openstack-manila10:48
openstackgerritMerged openstack/manila: Add tempest tests for mtu and gateway fields  https://review.openstack.org/37310111:01
*** hoonetorg has joined #openstack-manila11:06
*** aovchinnikov has quit IRC11:08
openstackgerritDigvijay Ukirde proposed openstack/manila: Add support for manage/unmanage in GPFS driver  https://review.openstack.org/37470511:11
*** shausy has quit IRC11:34
openstackgerritYingzhe Zeng proposed openstack/manila: Fix huawei driver cannot delete qos while status is idle  https://review.openstack.org/37287111:36
openstackgerritDigvijay Ukirde proposed openstack/manila: Add support for manage/unmanage in GPFS driver  https://review.openstack.org/37470511:36
openstackgerritValeriy Ponomaryov proposed openstack/manila-specs: Add spec for Scenario tests  https://review.openstack.org/37473111:42
openstackgerritMarc Koderer proposed openstack/manila: Remove tempest.test usage from manila tests  https://review.openstack.org/37471911:46
mkoderervponomaryov: btw I am trying to get rid of as much tempest dependencies as possible - I started with tempest.test11:48
mkodererhttps://review.openstack.org/#/c/374719/11:48
*** zigo has quit IRC11:54
vponomaryovmkoderer: direct call of testtools is too long11:57
vponomaryovmkoderer: better to port it to our base class11:57
vponomaryovthat is imported always11:57
mkoderervponomaryov: you mean the decorator is too long?11:58
vponomaryovmkoderer: yes11:58
*** zigo has joined #openstack-manila11:58
*** zigo is now known as Guest8360111:59
mkoderervponomaryov: I could just import testcase .. would be shorter11:59
mkodereror I'll put it to the base12:00
mkodererlet's see12:00
*** Guest83601 has quit IRC12:03
*** zigo_ has joined #openstack-manila12:12
*** zigo_ has quit IRC12:17
*** zigo_ has joined #openstack-manila12:18
*** vbellur has quit IRC12:18
*** StraubTW has joined #openstack-manila12:22
*** StraubTW has quit IRC12:27
openstackgerritMarc Koderer proposed openstack/manila: Remove tempest.test usage from manila tests  https://review.openstack.org/37471912:28
*** vbellur has joined #openstack-manila12:38
*** surabujin has quit IRC12:48
*** zigo_ has quit IRC12:48
*** zengyingzhe has quit IRC12:49
*** zengyingzhe has joined #openstack-manila12:49
*** zigo_ has joined #openstack-manila12:51
*** david-lyle has joined #openstack-manila12:57
*** alyson_ has joined #openstack-manila13:03
*** rooneym has joined #openstack-manila13:07
*** surabujin has joined #openstack-manila13:09
openstackgerritOpenStack Proposal Bot proposed openstack/manila: Updated from global requirements  https://review.openstack.org/37375413:10
*** eharney has quit IRC13:12
*** StraubTW has joined #openstack-manila13:15
*** gouthamr has joined #openstack-manila13:15
*** dustins has joined #openstack-manila13:20
openstackgerritDigvijay Ukirde proposed openstack/manila: Add support for manage/unmanage in GPFS driver  https://review.openstack.org/37470513:21
*** eharney has joined #openstack-manila13:30
*** akerr has joined #openstack-manila13:33
gansogouthamr: ping13:44
gouthamrganso: hey Rodrigo!13:44
gansogouthamr: Hi Goutham! good morning13:44
gansogouthamr: so you had a question for me yesterday?13:44
gouthamrganso: Goodmorning! yes.. :)13:44
gansogouthamr: np, I have a few for you today :P13:45
gouthamrganso: so, in the migration API, we don't allow shares with snapshots to be migrated13:45
gouthamrganso: haha13:45
gansogouthamr: yes, we don't13:45
*** surabujin has quit IRC13:46
gouthamrganso: if driver assisted migration is chosen, we *should* be able to move snapshots too..13:46
*** mtanino has joined #openstack-manila13:46
gansogouthamr: if all drivers are able to do that, yes...13:46
*** Yogi1 has joined #openstack-manila13:47
mkoderermeans the end-user has to delete all snapshots in case of the generic implementation?13:47
gouthamrmkoderer: yessir13:48
mkodererand also the shares created with --snapshot-id?13:48
gouthamrmkoderer ganso: in case of driver-assisted too, for the moment.. i was hoping we can get that sorted in Ocata :)13:49
gansogouthamr: so, we have the problem of guaranteeing that all drivers are capable. Then, if we want the driver-assisted and host-assisted to have a consistent API, then we cannot have this feature if host-assisted is not capable of, even if driver-assisted is13:49
gouthamrmkoderer: no such restriction applies at the API..13:50
gouthamrmkoderer: you should be able to move data from a share created through a snapshot, with host and driver assisted modes..13:50
gouthamrganso: yeah, this is almost like preserve-metadata, no?13:51
gansoyes, it is a totally new and considered to be independent share13:51
gansogouthamr: yes, could be seen as similar to preserve-metadata13:51
gansogouthamr: but we would need to control it13:51
gansogouthamr: like '--preserve-snapshots'13:52
gouthamrganso: sure i think that's sane.. with replication though, the design states that you need to be able to take snapshots to replicas..13:53
gansogouthamr: replication is solely dependent on drivers13:53
gouthamrganso: true..13:54
bswartzgouthamr: replication doesn't need to support snapshots...13:55
gouthamrganso: thanks.. we can discuss this with any improvements in Ocata... It would be nice to have;13:55
gansogouthamr: we will definitely have a few improvements to do in Ocata13:55
bswartzgouthamr: replication just has to support on the replicas everything the active replica supports13:55
gouthamrbswartz: true.. i mean, if snapshots are supported, they should be replicated..13:55
gansogouthamr: thanks for the feedback :)13:55
bswartzyes13:55
gouthamrganso: np! your turn..13:56
gansogouthamr:  lol ok13:56
gansogouthamr: so I was creating the share migration docs13:56
gansogouthamr: started with the devrefd13:56
gansogouthamr: *devref13:56
gouthamrnice...13:57
gansogouthamr: my understanding is that the devref is aimed at developers which would like to understand how the feature works to implement in their driver or to contribute13:57
gouthamrganso: yes..13:58
gansogouthamr: but most of the devref docs I've seen seems targeted at operators/administrators and don't describe many details that should be more relevant to developers13:58
*** surabujin has joined #openstack-manila13:59
gouthamrganso: true.. as vponomaryov mentioned, we're also using our devref as scratch-space before taking content to user guide/admin guide/config ref in the openstack-manuals project13:59
gansogouthamr: ok so I assume I am in the right direction... and I will prepare a different version aimed at administrators14:01
gouthamrganso: this aims at driver authors: http://docs.openstack.org/developer/manila/devref/share_replication.html14:01
gansogouthamr: I noticed there is a folder admin-ref, I will upload the admin one to that14:01
gouthamrganso: sure.. eventually user/admin facing documentation needs to go into the respective guide in openstack-manuals..14:01
gansogouthamr: weird, where is that document hosted? it isn't https://github.com/openstack/manila/blob/master/doc/source/devref/share_replication.rst, right?14:03
gouthamrganso: it is.. but github's rst rendering can't understand sphinx..14:04
gouthamrganso: look at the raw file for any missing content.14:04
gansogouthamr: hummm, yea I noticed that it is expanded on docs.openstack.org14:04
gansogouthamr: cool14:04
*** cknight has joined #openstack-manila14:04
gansogouthamr: ok, other question14:05
gansogouthamr: there is a very big problem in regards to using update_access only as "sync" as we discussed wrt that access_rules bugfix14:06
gansogouthamr: https://bugs.launchpad.net/manila/+bug/162652314:06
openstackLaunchpad bug 1626523 in Manila "Share is writable during host-assisted migration" [High,New] - Assigned to Rodrigo Barbieri (rodrigo-barbieri2010)14:06
gouthamrganso: i was going to push up an update to that change shortly..14:07
gansogouthamr: I am trying to fix this problem in a proper way14:07
gouthamrganso: "If you are quick enough, try to write some data in the share to confirm it is read only"14:07
gouthamrganso: what does that mean..14:08
gansogouthamr: but I found out that the current design prevents that14:08
gansogouthamr: what happens is that migration-start changes the access_rules to read-only only in the driver, not in the DB14:08
* gouthamr wonders if he can be quick enough14:08
gansogouthamr: this works until a later request runs update_access again14:09
gansogouthamr: update_access gets everything from DB and pass to driver, so it applies the RW rules back14:09
gansogouthamr: at first, there is no problem because the API is blocked during 'migration' status14:09
gansogouthamr: but the Data Service dodges the API and uses update_access to add a rule14:10
gouthamrganso: oh.. so, any existing R/W rules are preserved/reverted if an update_access call comes in14:10
gansogouthamr: yes14:10
gansogouthamr: for drivers that do not use recovery mode all the time, this is ok, since only the new rule will be added14:10
gansogouthamr: I did not detect this problem with the generic driver14:10
gansogouthamr: because it is not affected14:11
gansogouthamr: but other drivers are14:11
gouthamrganso: so, it feels like this needs to be handled in the method that's changing all the access rules to read/only.. i.e, use a db helper to cast any existing r/w rules to read-only before calling update-access ?14:11
gansogouthamr: yes, that was my first thought14:11
gansogouthamr: except for the fact that the call should be async because it is coming from another service (the Data Service)14:12
gansogouthamr: so the Data Service waits until the share_instance_access_rules_status is back to active14:12
gansogouthamr: that's why it is following the approach of creating that rule in DB14:12
gansogouthamr: so, by touching the DB, there are many things that must be done, which are done by access.update_access_rules, that's why it is invoked14:13
gansogouthamr: like removing from db on deny, setting the instance status back to active, etc14:14
gansogouthamr: so I am thinking of a way of fixing this but also not compromising your goal design14:14
gansogouthamr: if the migration helper does all of this, having a lot of duplicated code, it will fix the issue14:15
gansogouthamr: but it will be harder to maintain14:16
gouthamrganso: hmmm, so you want to be able to invoke the API to allow_access_to_instance, and have the read-only rules committed to the database as well?14:17
gansogouthamr: if update_acccess does this, to fix it right now, there is the possibility of adding a parameter that introduces this custom behavior to update_access when needed... it is less code and it is maintained along with other fixes and future designs, although it is a not a good looking approach14:17
gouthamrganso: the parameter would be 'cast-to-readonly'?14:18
gansogouthamr: no, the read-only rules do not need to be committed to the database, only the data service access rule14:18
gansogouthamr: something like that14:18
gouthamrganso: sounds reasonable.. but how will we be able to tell the difference between the data service IP and the rest of the rules?14:19
gouthamrganso: you mean to make two separate RPCs?14:19
gansogouthamr: the Data Service IP can be read only as well, it does not need to be different14:19
gouthamrganso: oh yes, true.. cool, works for me.. i'll push up with the changes we discussed.. we can make a follow on patch for this bugfix dependent on this patch, where we have a new parameter to cast existing rules to read-only..14:20
gansogouthamr: ok, I should have the patch pushed up today as well14:21
*** tovchinnikova has joined #openstack-manila14:22
*** aovchinnikov has joined #openstack-manila14:25
mkodererdustins: dmellado: I started to reduce the tempest dependency in manila tests https://review.openstack.org/#/c/374719/14:30
mkodererdustins: dmellado: let me know if you guys want to help me on that and then we can use the etherpad to coordinate14:30
dustinsmkoderer: This is part of the transition to using tempest.lib as well, yeah?14:31
*** senk has quit IRC14:34
*** xyang_ has joined #openstack-manila14:35
dustinsmkoderer: and I'd love to help out if I can!14:37
gouthamrbswartz: hi! https://review.openstack.org/#/c/355257/ needs your attention..14:43
gouthamrand looking forward for someone to take pity and +W https://review.openstack.org/#/c/359491/ :P14:43
gansogouthamr: I haven't +W'ed so other people could take a look14:45
gouthamrganso: yes :) thanks for the review.. i was hoping i could ask for volunteers to play around with our installation instructions on the ML - the core projects are being covered by the docs team..14:46
*** catintheroof has joined #openstack-manila14:50
openstackgerritBen Swartzlander proposed openstack/manila-specs: Add process for specs and spec deadline  https://review.openstack.org/37488314:50
bswartzvponomaryov, tbarron, cknight, markstur, xyang_, ganso: ^14:51
*** markstur_ has joined #openstack-manila14:53
xyang_bswartz: thanks14:54
*** kaisers has quit IRC14:55
*** zengyingzhe_ has joined #openstack-manila15:06
*** zigo_ is now known as zigo15:14
*** kaisers has joined #openstack-manila15:23
*** kaisers has quit IRC15:31
*** zengyingzhe_ has quit IRC15:39
*** dsariel has quit IRC15:45
*** gouthamr-afk has joined #openstack-manila15:47
*** kaisers has joined #openstack-manila15:50
*** gouthamr_afk has joined #openstack-manila15:54
*** gouthamr-afk has quit IRC15:54
*** kaisers has quit IRC15:54
*** gouthamr-afk has joined #openstack-manila15:55
*** tovchinnikova has quit IRC15:56
*** gouthamr_afk has quit IRC15:58
*** gouthamr_afk has joined #openstack-manila16:01
*** adrianofr_ has quit IRC16:02
*** gouthamr-afk has quit IRC16:03
*** senk has joined #openstack-manila16:19
*** gouthamr-afk has joined #openstack-manila16:28
*** gouthamr-afk has quit IRC16:32
*** gouthamr_afk has quit IRC16:32
*** xyang_ has quit IRC16:33
*** xyang_ has joined #openstack-manila16:34
*** xyang_ has quit IRC16:35
*** lpetrut has quit IRC16:36
*** narayrak has quit IRC16:38
*** aovchinnikov has quit IRC16:38
*** xyang_ has joined #openstack-manila16:40
*** zigo has quit IRC16:41
*** xyang_ has quit IRC16:41
*** markstur_ has quit IRC16:42
*** kaisers has joined #openstack-manila16:44
*** kaisers has quit IRC16:49
*** zigo has joined #openstack-manila16:51
*** zigo is now known as Guest1865616:52
*** senk has quit IRC16:53
*** Guest18656 has quit IRC16:56
*** senk has joined #openstack-manila16:57
*** zigo_ has joined #openstack-manila16:59
*** jcsp has quit IRC17:04
*** zigo_ has quit IRC17:07
*** zigo_ has joined #openstack-manila17:11
*** gouthamr has quit IRC17:15
*** lpetrut has joined #openstack-manila17:26
*** gouthamr has joined #openstack-manila17:31
*** xyang_ has joined #openstack-manila17:35
*** mkoderer has quit IRC17:39
openstackgerritAlyson proposed openstack/manila: Updated Hitachi NAS Platform Driver documentation  https://review.openstack.org/36072217:47
*** jseiler_ has left #openstack-manila17:53
*** jseiler has joined #openstack-manila17:55
*** senk has quit IRC18:03
*** ig0r_ has quit IRC18:09
*** dustins has quit IRC18:09
*** xyang1 has joined #openstack-manila18:13
openstackgerritRodrigo Barbieri proposed openstack/manila: Fix share writable in host-assisted migration  https://review.openstack.org/37504718:21
*** xyang_ has quit IRC18:22
*** vbellur has quit IRC18:25
*** lpetrut has quit IRC18:30
*** kaisers has joined #openstack-manila18:33
*** kaisers has quit IRC18:38
*** dustins has joined #openstack-manila18:51
*** absubram has joined #openstack-manila19:01
*** absubram has quit IRC19:03
*** nherciu_ has quit IRC19:04
*** absubram has joined #openstack-manila19:06
*** vbellur has joined #openstack-manila19:16
dustinsHey all, I finished up the Etherpad for all of the docs that we need to have a look at19:19
dustinshttps://etherpad.openstack.org/p/newton_documentation_blitz19:19
dustinsI'll be sending out an email to the list to get a wider audience, but if you wanna jump on the bandwagon now, all the better :)19:19
dustinsOf top priority are the installation guides for openSuse/SLES, RHEL and CentOS, and Ubuntu, as well as the "Steps to Perform" section of the Administration Reference19:20
dustinsPlease let me know if you have any questions!19:20
*** zigo_ is now known as zigo19:30
openstackgerritRodrigo Barbieri proposed openstack/manila: Fix share writable in host-assisted migration  https://review.openstack.org/37504719:54
*** akerr has quit IRC20:03
*** faiz89 has joined #openstack-manila20:08
*** Yogi1 has quit IRC20:19
*** gouthamr has quit IRC20:20
*** kaisers has joined #openstack-manila20:22
*** kaisers has quit IRC20:27
*** absubram_ has joined #openstack-manila20:32
*** absubram has quit IRC20:33
*** absubram_ is now known as absubram20:33
*** faiz89 has quit IRC20:36
*** senk has joined #openstack-manila20:37
*** cknight has quit IRC20:51
*** eharney has quit IRC20:58
*** kaisers has joined #openstack-manila21:15
*** cknight has joined #openstack-manila21:18
*** kaisers has quit IRC21:19
*** gobbe has joined #openstack-manila21:21
*** senk__ has joined #openstack-manila21:30
*** xyang_ has joined #openstack-manila21:31
*** senk has quit IRC21:33
*** xyang_ has quit IRC21:33
*** senk__ has quit IRC21:35
jay-mehtaxyang_, bswartz , https://review.openstack.org/#/c/369763/ Since this patch does not have string translation now, can it merge in newton21:36
*** senk has joined #openstack-manila21:36
*** xyang_ has joined #openstack-manila21:36
*** amitkqed has quit IRC21:37
*** xyang1 has quit IRC21:37
*** DuncanT has quit IRC21:37
*** markstur has quit IRC21:37
*** madorn has quit IRC21:37
*** tries_ has quit IRC21:37
*** ameade has quit IRC21:37
*** jay-mehta has quit IRC21:37
*** bswartz has quit IRC21:37
*** vbellur has quit IRC21:37
*** rhagarty_ has quit IRC21:37
*** csaba has quit IRC21:37
*** cfey has quit IRC21:37
*** mkoderer__ has quit IRC21:37
*** draynium has quit IRC21:37
*** jseiler has quit IRC21:37
*** StraubTW has quit IRC21:37
*** aorourke has quit IRC21:37
*** cargonza has quit IRC21:37
*** amit213 has quit IRC21:37
*** netapp has quit IRC21:37
*** andreaf has quit IRC21:37
*** markstur has joined #openstack-manila21:38
*** madorn has joined #openstack-manila21:38
*** tries_ has joined #openstack-manila21:38
*** ameade has joined #openstack-manila21:38
*** bswartz has joined #openstack-manila21:38
*** jay-mehta has joined #openstack-manila21:38
*** jseiler has joined #openstack-manila21:40
*** StraubTW has joined #openstack-manila21:40
*** aorourke has joined #openstack-manila21:40
*** amit213 has joined #openstack-manila21:40
*** netapp has joined #openstack-manila21:40
*** andreaf has joined #openstack-manila21:40
*** amitkqed has joined #openstack-manila21:40
*** xyang1 has joined #openstack-manila21:41
*** ameade has quit IRC21:41
*** cknight has quit IRC21:42
*** rhagarty_ has joined #openstack-manila21:42
*** csaba has joined #openstack-manila21:42
*** cfey has joined #openstack-manila21:42
*** mkoderer__ has joined #openstack-manila21:42
*** draynium has joined #openstack-manila21:42
*** amit213 has quit IRC21:44
*** rooneym has quit IRC21:56
jay-mehtaxyang_, thank you for reviewing. I think the patch does not have +1 for workflow and so it getting merged. Am I right?21:58
jay-mehtaso it is NOT *21:58
xyang_jay-mehta: need to ask bswartz about it.  It looks okay to me.21:59
xyang_jay-mehta: bswartz had asked us to be careful on what to be merged, so I need to check with him before merging anything21:59
xyang_jay-mehta: also this will be merged in master.  you'll need to backport to get to newton22:00
*** dustins has quit IRC22:00
*** cargonza has joined #openstack-manila22:04
jay-mehtaxyang_, thanks22:05
*** StraubTW has quit IRC22:07
*** DuncanT has joined #openstack-manila22:08
*** xyang_ has quit IRC22:08
*** amit213 has joined #openstack-manila22:14
*** ameade has joined #openstack-manila22:15
*** xyang_ has joined #openstack-manila22:19
*** xyang_ has quit IRC22:20
*** xyang1 has quit IRC22:20
*** vbellur has joined #openstack-manila22:44
openstackgerritRodrigo Barbieri proposed openstack/manila: Fix share writable in host-assisted migration  https://review.openstack.org/37504722:47
*** StraubTW has joined #openstack-manila22:53
*** StraubTW_ has joined #openstack-manila22:54
*** StraubTW has quit IRC22:58
*** alyson_ has quit IRC22:59
*** StraubTW_ has quit IRC23:01
*** kaisers has joined #openstack-manila23:04
*** kaisers has quit IRC23:09
openstackgerritRodrigo Barbieri proposed openstack/manila-ui: Fix migration-start id-dependent form  https://review.openstack.org/37516423:49

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