| *** dustins has quit IRC | 00:07 | |
| *** tommylikehu_ has joined #openstack-manila | 00:10 | |
| *** tommylikehu_ has quit IRC | 00:16 | |
| *** akerr has quit IRC | 00:30 | |
| *** cknight has joined #openstack-manila | 01:01 | |
| *** a-pugachev has quit IRC | 01:05 | |
| openstackgerrit | TommyLike proposed openstack/manila: Enable IPv6 in manila(network plugins and drivers) https://review.openstack.org/406776 | 01:26 |
|---|---|---|
| *** mtanino has quit IRC | 01:30 | |
| openstackgerrit | TommyLike proposed openstack/manila: Enable IPv6 in manila(allow access) https://review.openstack.org/312321 | 01:30 |
| *** mtanino has joined #openstack-manila | 01:32 | |
| *** cknight has quit IRC | 01:42 | |
| openstackgerrit | Rodrigo Barbieri proposed openstack/manila: Add mountable snapshots support https://review.openstack.org/345526 | 01:52 |
| *** kaisers_ has joined #openstack-manila | 02:05 | |
| *** kaisers has quit IRC | 02:08 | |
| *** akerr has joined #openstack-manila | 02:28 | |
| openstackgerrit | Ha Dung proposed openstack/manila: Fixes some current typo errors on Manila project. https://review.openstack.org/423164 | 02:34 |
| *** mtanino has quit IRC | 02:40 | |
| *** harlowja has quit IRC | 02:44 | |
| openstackgerrit | TommyLike proposed openstack/manila: Enable IPv6 in manila(network plugins and drivers) https://review.openstack.org/406776 | 03:43 |
| *** ociuhandu has quit IRC | 03:57 | |
| *** ganso has quit IRC | 04:14 | |
| *** ociuhandu has joined #openstack-manila | 04:21 | |
| *** tommylikehu_ has joined #openstack-manila | 05:07 | |
| *** tommylikehu_ has quit IRC | 05:11 | |
| *** tommylikehu1 has joined #openstack-manila | 05:20 | |
| *** tommylikehu has quit IRC | 05:20 | |
| *** tommylikehu1 is now known as tommylikehu | 05:20 | |
| *** ianychoi has joined #openstack-manila | 05:26 | |
| *** makowals_ has quit IRC | 05:29 | |
| *** makowals has joined #openstack-manila | 05:33 | |
| *** lpetrut has joined #openstack-manila | 06:15 | |
| *** harlowja has joined #openstack-manila | 06:34 | |
| *** lpetrut has quit IRC | 06:56 | |
| *** nkrinner_afk is now known as nkrinner | 06:59 | |
| *** lpetrut has joined #openstack-manila | 07:16 | |
| *** jprovazn has joined #openstack-manila | 07:17 | |
| *** harlowja has quit IRC | 07:26 | |
| *** lpetrut has quit IRC | 07:50 | |
| *** openstackgerrit has quit IRC | 08:03 | |
| *** lpetrut has joined #openstack-manila | 08:06 | |
| *** zengyingzhe has quit IRC | 08:21 | |
| *** vponomaryov1 has joined #openstack-manila | 08:37 | |
| *** pcaruana has joined #openstack-manila | 08:45 | |
| *** openstackgerrit has joined #openstack-manila | 08:49 | |
| openstackgerrit | Vitaliy Levitski proposed openstack/manila-ui: Add MapRFS protocol https://review.openstack.org/421884 | 08:49 |
| *** lpetrut has quit IRC | 08:56 | |
| *** makowals has quit IRC | 08:56 | |
| *** makowals_ has joined #openstack-manila | 08:59 | |
| *** a-pugachev has joined #openstack-manila | 09:03 | |
| openstackgerrit | Helen Walsh proposed openstack/manila: VMAX manila plugin - Support for VMAX in Manila https://review.openstack.org/404859 | 09:05 |
| *** ganso has joined #openstack-manila | 09:15 | |
| *** dsariel has joined #openstack-manila | 09:48 | |
| *** alyson_ has joined #openstack-manila | 09:59 | |
| ganso | vponomaryov, vponomaryov1: Hi Valeriy. I noticed you're reviewing the mountable patch, I am addressing all your comments right now (have already addressed the existing ones, I am testing the HTTPMethodNotAllowed change) | 10:09 |
| ganso | vponomaryov, vponomaryov1: Please let me know when you're done so I can submit a new patch | 10:11 |
| vponomaryov1 | ganso: Hello, I am in process and publishing comments right away | 10:13 |
| ganso | vponomaryov1: thanks | 10:13 |
| vponomaryov1 | ganso: you are responsible now for "mountable_snapshots" feature? | 10:15 |
| ganso | vponomaryov1: yes | 10:15 |
| ganso | vponomaryov1: well, the whole team here is now, actually | 10:17 |
| vponomaryov1 | ganso: yeah, I remember that you have "BIG" support )) | 10:22 |
| vponomaryov1 | ganso: do you have cheerleaders among them? )) | 10:22 |
| ganso | vponomaryov1: unfortunately no | 10:24 |
| ganso | vponomaryov1: as you can notice, coding style is varying a lot within the same patch | 10:32 |
| vponomaryov1 | ganso: yes, I see | 10:33 |
| vponomaryov1 | ganso: style of "Crouching Tiger" and "Hidden Dragon" =) | 10:34 |
| ganso | vponomaryov1: I haven't watched that movie | 10:35 |
| vponomaryov1 | ganso: in some places - "drunken monkey" )) | 10:35 |
| ganso | vponomaryov1: drunken monkey still would be better than sleepy monkey | 10:35 |
| vponomaryov1 | ganso: have some of you tested this feature with LVM driver? | 10:38 |
| ganso | vponomaryov1: yes | 10:38 |
| ganso | vponomaryov1: it needs the code you mentioned as being separate bugfix | 10:38 |
| vponomaryov1 | ganso: exactly mounting snapshot? | 10:38 |
| ganso | vponomaryov1: else it does not work | 10:38 |
| ganso | vponomaryov1: yes, we tested it, the scenario test patch that is dependent on that one tests that as well | 10:39 |
| *** yumiriam has joined #openstack-manila | 10:41 | |
| *** tommylikehu_ has joined #openstack-manila | 10:52 | |
| *** rraja has joined #openstack-manila | 11:07 | |
| vponomaryov1 | ganso: letting you know I've finished review of server side change | 11:09 |
| ganso | vponomaryov1: thanks | 11:10 |
| vponomaryov1 | ganso: but you have there bugs, that client change discovered -> http://logs.openstack.org/25/345625/23/check/gate-manilaclient-dsvm-neutron-functional-ubuntu-xenial/62024a2/logs/screen-m-shr.txt.gz?level=TRACE | 11:11 |
| ganso | vponomaryov1: I'll take a look | 11:12 |
| vponomaryov1 | gouthamr: for case you read it, change with microversion 2.30 requires commit to client too, at least for having reasenote | 11:16 |
| ganso | vponomaryov1: mapping.soft_delete(session, update_status=True ==> AttributeError: 'NoneType' object has no attribute 'soft_delete' | 11:17 |
| ganso | vponomaryov1: there is a "if not mapping" right above that statement | 11:17 |
| ganso | vponomaryov1: any idea why it would be None and not enter condition? | 11:17 |
| vponomaryov1 | ganso: I guess it is object | 11:22 |
| vponomaryov1 | ganso: sql object | 11:22 |
| vponomaryov1 | ganso: s/sql/sqla/ | 11:23 |
| vponomaryov1 | ganso: so, there should be other kind of check to verify absense of record | 11:23 |
| vponomaryov1 | but I agree, it is VERY strange | 11:24 |
| openstackgerrit | Ramana Raja proposed openstack/manila: WIP ganesha: dynamically update access of share https://review.openstack.org/424598 | 11:33 |
| openstackgerrit | Ramana Raja proposed openstack/manila: WIP: Refactor and rename CephFSNativeDriver https://review.openstack.org/421201 | 11:45 |
| openstackgerrit | Ramana Raja proposed openstack/manila: cephfs: add nfs protocol support https://review.openstack.org/424605 | 11:56 |
| openstackgerrit | Ramana Raja proposed openstack/manila: [WIP] cephfs: add nfs protocol support https://review.openstack.org/424605 | 11:56 |
| openstackgerrit | Ramana Raja proposed openstack/manila: [WIP] cephfs: add nfs protocol support https://review.openstack.org/424605 | 11:59 |
| openstackgerrit | Ramana Raja proposed openstack/manila: [WIP] Refactor and rename CephFSNativeDriver https://review.openstack.org/421201 | 11:59 |
| *** porrua has joined #openstack-manila | 12:04 | |
| openstackgerrit | Rodrigo Barbieri proposed openstack/manila: Add mountable snapshots support https://review.openstack.org/345526 | 12:04 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/manila: Updated from global requirements https://review.openstack.org/424616 | 12:11 |
| *** gouthamr has joined #openstack-manila | 12:14 | |
| *** catintheroof has joined #openstack-manila | 12:16 | |
| *** shausy has joined #openstack-manila | 12:28 | |
| *** lpetrut has joined #openstack-manila | 12:37 | |
| *** mtanino has joined #openstack-manila | 13:00 | |
| *** lpetrut has quit IRC | 13:04 | |
| *** dustins has joined #openstack-manila | 13:05 | |
| openstackgerrit | Alyson proposed openstack/python-manilaclient: Add mountable snapshots support to manila client https://review.openstack.org/345625 | 13:18 |
| *** tommylikehu_ has quit IRC | 13:26 | |
| *** gouthamr has quit IRC | 13:32 | |
| openstackgerrit | Helen Walsh proposed openstack/manila: VMAX manila plugin - Support for VMAX in Manila https://review.openstack.org/404859 | 13:41 |
| *** akerr has quit IRC | 13:52 | |
| *** tommylikehu_ has joined #openstack-manila | 13:59 | |
| *** tommylikehu_ has quit IRC | 13:59 | |
| *** tommylikehu_ has joined #openstack-manila | 13:59 | |
| *** mtanino has quit IRC | 14:13 | |
| *** gouthamr has joined #openstack-manila | 14:16 | |
| *** breitz has quit IRC | 14:30 | |
| *** eharney has joined #openstack-manila | 14:37 | |
| bswartz | tommylikehu tommylikehu_: can you share the devstack config you used for testing of ipv6 with manila? | 14:42 |
| *** lpetrut has joined #openstack-manila | 14:43 | |
| bswartz | tommylikehu tommylikehu_: I have created several different configurations but I haven't yet found one that allows a Nova VM to communicate with the LVM driver over ipv6 | 14:43 |
| tbarron | bswartz: tangentially related to ^^ are you by any chance using bridged KVM ubuntu xenial vm as your devstack host machine? | 14:46 |
| tbarron | bswartz: I am and it auto ipv6 doesn't work on its main external interface | 14:46 |
| tbarron | s/it// | 14:47 |
| *** rraja has quit IRC | 14:47 | |
| tbarron | bswartz: my home network has ipv6 of sorts, i.e. centurylink ipv6 "rd" | 14:48 |
| bswartz | tbarron: I have tried many things | 14:48 |
| bswartz | tbarron: in the process I found 4 bugs in devstack | 14:48 |
| bswartz | >_< | 14:48 |
| bswartz | it's clear nobody has tested devstack with IPv6 and provider networks | 14:48 |
| *** gouthamr_ has joined #openstack-manila | 14:49 | |
| bswartz | tbarron: https://review.openstack.org/424444 | 14:49 |
| tbarron | physical centos and fedora machines get non link local ipv6 addr on main external interface, but my ubuntu xenial machine does not | 14:49 |
| bswartz | tbarron: part of the problem is that there are too many options for networking in devstack | 14:49 |
| bswartz | the number of possible configuration is dizzying and impossible to test | 14:49 |
| tbarron | bswartz: right | 14:49 |
| tbarron | bswartz: I've just been trying to get it to work with the neutron ipv6 cidrs assigned to the "public" and "private" networks, but so far no dice | 14:50 |
| *** gouthamr has quit IRC | 14:50 | |
| tbarron | bswartz: so I thought I'd debug by manually updating exports on my host/hypervisor attempting ipv6 mounts from the guest | 14:51 |
| tbarron | bswartz: or from the host itself | 14:51 |
| *** gouthamr_ is now known as gouthamr | 14:51 | |
| bswartz | tbarron: I'm close to having a working config to share | 14:51 |
| bswartz | tbarron: I need to fix a bug in manila now to get it fully working | 14:52 |
| bswartz | tbarron: I was hoping tommylikehu had some alternative config that would work out of the box | 14:52 |
| tbarron | bswartz: but my hypervisor/devstack host only has a non-link-local ipv6 addr on br-ex, it's not acquiring one on its main external interface | 14:52 |
| bswartz | I bet it's possible to use the l3 agent and create a virtual router that make this "just work" | 14:52 |
| bswartz | however I haven't stumbled across that config yet | 14:53 |
| bswartz | personally I find it easier to debug when there is no l3 agent involved so I'm pushing ahead with that config | 14:53 |
| tbarron | bswartz: you'd think that the default ipv6 cidrs assigned as neutron subnets to the basic admin/public and demo/private subnets would be usable | 14:53 |
| bswartz | tbarron: they *are* usable | 14:53 |
| tbarron | bswartz: routable | 14:54 |
| bswartz | there just aren't routes to them from the hypervisor | 14:54 |
| bswartz | that's by design -- it's a security issue | 14:54 |
| tbarron | bswartz: yeah, I can get a dhcp6 assignment to my guest just fine | 14:54 |
| bswartz | remember there's no NAT for ipv6 and thus no "floating IPs" | 14:54 |
| tbarron | bswartz: yes, that's why I thought it would in a sense be easier | 14:54 |
| bswartz | heh | 14:55 |
| bswartz | we are all so addicted to NAT that we don't realize how hard it will be to give up when we move to ipv6 | 14:55 |
| tbarron | maybe there's some adjustment that needs to be done to that little router that connects public and private networks | 14:55 |
| bswartz | I'm one of the few people on this planet that thinks NAT6 is a good thing | 14:55 |
| *** belmoreira has joined #openstack-manila | 14:56 | |
| tbarron | bswartz: whether good or bad, we should be able to make it work w/o floating ips | 14:56 |
| *** jprovazn has quit IRC | 14:57 | |
| openstackgerrit | Ben Swartzlander proposed openstack/manila: Fix devstack plugin to not depend on private network https://review.openstack.org/424685 | 14:59 |
| bswartz | tbarron: it will work | 14:59 |
| bswartz | tbarron: after I fix all the bugs | 14:59 |
| * bswartz was up late last night fixing bugs | 14:59 | |
| bswartz | vponomaryov1: this is what we talked about -> https://review.openstack.org/424685 | 15:01 |
| vponomaryov1 | bswartz: thanks, LGTM | 15:02 |
| *** tommylikehu_ has quit IRC | 15:09 | |
| openstackgerrit | Merged openstack/manila: Rename consistency group modules to share groups https://review.openstack.org/409864 | 15:09 |
| openstackgerrit | Merged openstack/manila: Manila Share Groups https://review.openstack.org/335093 | 15:10 |
| *** tommylikehu_ has joined #openstack-manila | 15:11 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/manila: Updated from global requirements https://review.openstack.org/424616 | 15:11 |
| *** cknight has joined #openstack-manila | 15:12 | |
| tommylikehu_ | hey bswartz, tbarron does the IPv6 feature have something related to devstack? | 15:12 |
| openstackgerrit | Rodrigo Barbieri proposed openstack/manila-ui: Implement Share Migration Ocata improvements https://review.openstack.org/406307 | 15:13 |
| *** mtanino has joined #openstack-manila | 15:13 | |
| *** gouthamr has quit IRC | 15:13 | |
| *** cdelatte has joined #openstack-manila | 15:13 | |
| vponomaryov1 | bswartz, markstur: after merging server side change for share groups, need to merge client -> https://review.openstack.org/#/c/335120 | 15:14 |
| *** mtanino has quit IRC | 15:15 | |
| bswartz | vponomaryov1: no | 15:15 |
| *** mtanino has joined #openstack-manila | 15:15 | |
| bswartz | err sorry vponomaryov1 | 15:15 |
| vponomaryov1 | bswartz: ? | 15:15 |
| bswartz | nm | 15:15 |
| bswartz | tommylikehu_: obviously your patch doesn't modify the devstack plugin | 15:15 |
| bswartz | tommylikehu_: however properly testing ipv6 will require that I think | 15:16 |
| bswartz | tommylikehu_: we were hoping you could tell us how you did your testing | 15:16 |
| tommylikehu_ | :( I did not realised that we the devstack should be changed. | 15:17 |
| tommylikehu_ | The tested procedure for me about IPv6 is configure backend with lvm which use the host's ipv6 address and filter backend with extra-specs | 15:17 |
| bswartz | tommylikehu_: it doesn't need to be done as part of this patch, we can do it later | 15:17 |
| vponomaryov1 | tommylikehu_: how was your feature tested? | 15:17 |
| tommylikehu_ | then add the access rule for agent with IPv6 address | 15:18 |
| bswartz | tommylikehu_: but those of us testing it want to see a mount from a nova vm to the lvm driver over ipv6 succeed | 15:18 |
| vponomaryov1 | tommylikehu_: have you tested real IPv6 access? | 15:18 |
| bswartz | tommylikehu_: were you able to get a mount to succeed? | 15:18 |
| bswartz | tommylikehu_: and if so, how did you get the routing to work? | 15:18 |
| tommylikehu_ | the agent nodes are not belonged to the openstack | 15:20 |
| bswartz | what agent? | 15:21 |
| bswartz | you mean the client you mounted from was not a nova VM? | 15:21 |
| tommylikehu_ | no | 15:22 |
| bswartz | no it wasn't a nova VM or no it was a nova VM? | 15:22 |
| tommylikehu_ | it was not a nova vm | 15:22 |
| bswartz | k | 15:23 |
| bswartz | that's an easier case to test | 15:23 |
| bswartz | I should have started with that one | 15:23 |
| bswartz | tbarron: ^ | 15:23 |
| bswartz | I'm very close to getting an actual nova VM to work now though so I'm going to pursue that case a bit longer | 15:23 |
| tbarron | yeah, i don't see a lot of point in this unless we know how to mount shares to nova instances | 15:27 |
| tbarron | it's good work, don't get me wrong, just not an exposable feature till then | 15:27 |
| bswartz | tbarron: it has value even without that | 15:27 |
| bswartz | tbarron: because we care about non-nova use cases | 15:27 |
| bswartz | tbarron: however there's no reason we shouldn't be able to get it working with nova too | 15:27 |
| bswartz | the main problems are that devstack has too many networking options and too little documentation | 15:28 |
| tommylikehu_ | bswartz, tbarron if we can not make the generic driver and at least one of the vendor drivers get merged into this version of IPv6, could this be little valued | 15:33 |
| tommylikehu_ | ? | 15:34 |
| bswartz | tommylikehu_: I'm okay with LVM support only | 15:36 |
| bswartz | tommylikehu_: we have to get started somewhere and ipv6 is an essential feature | 15:36 |
| bswartz | assuming we can get ipv6 merged today or tomorrow vendors can start working on driver support in pike | 15:36 |
| bswartz | tommylikehu_: you do have at least 1 merge conflict to resolve | 15:37 |
| tommylikehu_ | bswartz: ok, I will fix that. | 15:37 |
| *** nkrinner is now known as nkrinner_afk | 15:44 | |
| *** tommylikehu_ has quit IRC | 15:46 | |
| *** ociuhandu has quit IRC | 15:48 | |
| *** breitz has joined #openstack-manila | 15:51 | |
| ganso | bswartz, tbarron, vponomaryov1, markstur, cknight: Hello, could we please merge https://review.openstack.org/#/c/406307/ ? | 15:55 |
| openstackgerrit | Rodrigo Barbieri proposed openstack/manila: Add mountable snapshots support https://review.openstack.org/345526 | 15:56 |
| *** shausy has quit IRC | 15:56 | |
| markstur | vponomaryov1: share groups client needs microversion update doesn't it (looking now) | 15:57 |
| vponomaryov1 | markstur: it doesn't | 15:59 |
| openstackgerrit | Alyson proposed openstack/python-manilaclient: Add mountable snapshots support to manila client https://review.openstack.org/345625 | 15:59 |
| markstur | vponomaryov1: saw some suspicious looking 2.29s | 15:59 |
| bswartz | omg devstack completed without errors | 16:00 |
| bswartz | first time this happened since 11pm last night | 16:00 |
| vponomaryov1 | markstur: 2.30 was not merged as it does nto require changes to client except change of version | 16:00 |
| vponomaryov1 | ganso: are you sure that in your change for manila-ui there is no need to change other APIs after bumping used microversion? | 16:03 |
| markstur | vponomaryov1: share_group_snapshots.py wraps("2.29") | 16:03 |
| markstur | max version bumped up to 2.31 | 16:05 |
| vponomaryov1 | markstur: strange, very strange | 16:06 |
| vponomaryov1 | markstur: tought I changed it | 16:06 |
| vponomaryov1 | markstur: thanks | 16:06 |
| markstur | microstrange | 16:06 |
| vponomaryov1 | will update | 16:06 |
| markstur | we should use MACROversions. They are easier to see | 16:06 |
| *** dmellado has quit IRC | 16:07 | |
| bswartz | markstur: nanoversions | 16:07 |
| markstur | bswartz: Ooooh. That sounds interesting. Next thing! | 16:07 |
| bswartz | 2.000000031 | 16:07 |
| *** breitz has quit IRC | 16:07 | |
| ganso | vponomaryov1: share migration is 2.29 | 16:07 |
| bswartz | 2.000000032 | 16:07 |
| *** breitz has joined #openstack-manila | 16:08 | |
| markstur | bswartz: how about we just use timestamps. API as of now() | 16:08 |
| bswartz | markstur: only if you use gettimeofday() and tv_usec | 16:09 |
| *** a-pugachev has quit IRC | 16:10 | |
| bswartz | the API could change every microsecond | 16:10 |
| markstur | bswartz: Not with our gate | 16:10 |
| bswartz | lol | 16:10 |
| bswartz | DuncanT proposed using semver for API versions | 16:11 |
| bswartz | I have reservations about that idea | 16:11 |
| bswartz | okay I'm banged my head against the wall long enough that there's a hole in the wall -- going to get some food and drive into the office | 16:12 |
| *** pcaruana has quit IRC | 16:16 | |
| openstackgerrit | Valeriy Ponomaryov proposed openstack/python-manilaclient: Add share group support to Manila client https://review.openstack.org/335120 | 16:17 |
| markstur | vponomaryov1: Sorry I didn't catch that last night. I thought it was ready to go. | 16:17 |
| *** a-pugachev has joined #openstack-manila | 16:17 | |
| vponomaryov1 | markstur: I thought I changed it )) | 16:17 |
| vponomaryov1 | markstur: anyway, check this out -> https://review.openstack.org/335120 | 16:18 |
| vponomaryov1 | ganso: it is not the answer for question | 16:20 |
| *** ociuhandu has joined #openstack-manila | 16:21 | |
| ganso | vponomaryov1: I haven't noticed anything break | 16:21 |
| vponomaryov1 | markstur: what about https://review.openstack.org/#/c/355264/ ? | 16:21 |
| ganso | vponomaryov1: even if so, a separate patch should be created | 16:21 |
| ganso | vponomaryov1: not in scope of my patch | 16:21 |
| markstur | vponomaryov1: OK. So you are just turning it off to skip redundant tests. Thought that might be the case. | 16:23 |
| vponomaryov1 | ganso: like this -> https://github.com/openstack/manila-ui/blob/7c90c469/manila_ui/api/manila.py#L228 | 16:23 |
| vponomaryov1 | ganso: actually, it is part of the patch, because you change used microversion | 16:24 |
| markstur | vponomaryov1: I thought it was odd you skipped it for those but then turned it on by default for everyone else <-- good explanation though. Thanks. | 16:24 |
| ganso | vponomaryov1: no, a separate patch needs to be create to address that according to its microversion, as we are doing with python-manilaclient | 16:25 |
| ganso | vponomaryov1: but it should come before my patch | 16:25 |
| vponomaryov1 | ganso: bump of microversion in manila-ui should be done with update of APIs we use there, so they are compatible with this microversion or nanoversion | 16:26 |
| ganso | vponomaryov1: yes, should, but we changed server deprecating nova_network but we haven't updated manila-ui | 16:26 |
| vponomaryov1 | markstur: so, I answered your questions there? | 16:27 |
| markstur | vponomaryov1: Yes. Just doing a quick review now. Just in case. | 16:27 |
| vponomaryov1 | ganso: we also haven't bumped used microversion | 16:27 |
| ganso | vponomaryov1: yes, but we should have | 16:28 |
| ganso | vponomaryov1: also, just an observation, some changes are automatically incorporated into manila-ui when we just bump the microversion, because they are done in python-manilaclient | 16:28 |
| ganso | vponomaryov1: I am not sure if the deprecation of nova_network falls into this category | 16:29 |
| vponomaryov1 | ganso: it becomes server error now | 16:29 |
| vponomaryov1 | ganso: with your change | 16:29 |
| vponomaryov1 | ganso: can you create share-network with your patch? | 16:32 |
| ganso | vponomaryov1: no | 16:32 |
| vponomaryov1 | ganso: using manila-ui? | 16:32 |
| vponomaryov1 | ganso: no - you tried and failed, or no - you cannot try? | 16:33 |
| ganso | vponomaryov1: I tried and failed | 16:33 |
| vponomaryov1 | ganso: it is unacceptable, don't you agree? | 16:34 |
| ganso | vponomaryov1: Recoverable error: create() got an unexpected keyword argument 'nova_net_id' | 16:34 |
| ganso | vponomaryov1: I agree, someone should fix it | 16:34 |
| *** ociuhandu has quit IRC | 16:40 | |
| rhagarty | hello... I'm also running into issues using manila-ui. I've started to implement this - https://blueprints.launchpad.net/manila/+spec/manila-share-group-ui | 16:45 |
| rhagarty | in Horizon IRC, it looks like latest pyton-novaclient has issues, and it is recommended to go back to 6.0.0 | 16:47 |
| vponomaryov1 | rhagarty: "forbidden" thing? | 16:47 |
| *** dmellado has joined #openstack-manila | 16:48 | |
| *** a-pugachev has quit IRC | 16:50 | |
| rhagarty | vponomaryov1: the error I was seeing was it couldn't access nova quotas | 16:52 |
| vponomaryov1 | rhagarty: right, "forbidden" error on nova quotas | 16:52 |
| rhagarty | with 6.0.0 it gets past that, but now seeing errors creating shares | 16:53 |
| rhagarty | http://pastebin.com/ZngqsX9s | 16:53 |
| *** dmellado has quit IRC | 16:53 | |
| *** dsariel has quit IRC | 16:53 | |
| vponomaryov1 | rhagarty: I haven't seen this problem, looks like something funny with novaclient again, you can avoid usage of novaclient | 16:56 |
| vponomaryov1 | rhagarty: in your case you can use any other driver except generic | 16:56 |
| vponomaryov1 | rhagarty: also, "dummy" one | 16:56 |
| *** akapil has joined #openstack-manila | 16:56 | |
| vponomaryov1 | rhagarty: it is recommended to use when you care about APIs only | 16:56 |
| vponomaryov1 | rhagarty: like in your case - support in manila UI | 16:57 |
| rhagarty | vponomaryov1: ok, thanks | 16:57 |
| vponomaryov1 | rhagarty: example of its configuration is not in the docs, but in our mainla repo and used in CI | 16:57 |
| vponomaryov1 | rhagarty: see https://github.com/openstack/manila/blob/master/contrib/ci/pre_test_hook.sh and https://github.com/openstack/manila/blob/master/contrib/ci/post_test_hook.sh searching for "dummy" word | 16:59 |
| *** dmellado has joined #openstack-manila | 16:59 | |
| vponomaryov1 | rhagarty: and manila config in CI with this fake driver -> http://logs.openstack.org/59/404859/19/check/gate-manila-tempest-minimal-dsvm-dummy-ubuntu-xenial/bba6028/logs/etc/manila/manila.conf.txt.gz | 17:00 |
| vponomaryov1 | rhagarty: you do not need reinstall devstack, you can just update manila config and restart services adding new backends | 17:01 |
| vponomaryov1 | rhagarty: that use this driver | 17:01 |
| *** a-pugachev has joined #openstack-manila | 17:01 | |
| vponomaryov1 | rhagarty: in this case novaclient will not make problems at all | 17:02 |
| rhagarty | vponomaryov1: thank you. | 17:02 |
| rhagarty | vponomaryov1: btw, will be needing to set up microversion support in manila-ui to access share group APIs. Not sure if anyone else has looked at this yet... | 17:03 |
| openstackgerrit | Rodrigo Barbieri proposed openstack/manila-ui: Bump microversion to 2.26 https://review.openstack.org/424748 | 17:03 |
| ganso | vponomaryov1: ^ happy? | 17:03 |
| bswartz | back | 17:07 |
| vponomaryov1 | ganso: I already have tests for it, don't you mind if I upload it to your patch? | 17:07 |
| ganso | vponomaryov1: go ahead | 17:08 |
| openstackgerrit | Valeriy Ponomaryov proposed openstack/manila-ui: Bump microversion to 2.26 https://review.openstack.org/424748 | 17:08 |
| *** xyang_ has joined #openstack-manila | 17:09 | |
| *** cdelatte has quit IRC | 17:11 | |
| openstackgerrit | Rodrigo Barbieri proposed openstack/manila-ui: Bump microversion to 2.26 https://review.openstack.org/424748 | 17:13 |
| openstackgerrit | Merged openstack/manila-ui: Add MapRFS protocol https://review.openstack.org/421884 | 17:14 |
| ganso | bswartz: https://review.openstack.org/#/c/345526/ is ready | 17:15 |
| bswartz | ganso: ty | 17:15 |
| *** Yogi1 has joined #openstack-manila | 17:15 | |
| rhagarty | microversion question - looks like the manila-ui just passes the a version number to the client. What happens if the server and/or client don't support that version? | 17:21 |
| bswartz | ganso: did you rebase recently? | 17:21 |
| ganso | bswartz: yes, after share groups | 17:21 |
| bswartz | rhagarty: if the client doesn't support that version, then the installation has been screwed up -- dependencies should catch that case | 17:22 |
| bswartz | rhagarty: if the server doesn't support the version, then microversion negotiation should save you, unless you're insanely out of date | 17:23 |
| rhagarty | bswartz: and how would you handle deprecation? Like feature_a works until 2.24 and feature_b workis in > 2.25? | 17:23 |
| bswartz | rhagarty: applications can write 2 code paths if they want to support both feature_a and feature_b, but if they don't then they're stuck with whichever one they choose | 17:24 |
| vponomaryov1 | rhagarty: manila UI, providing only single microversion, is of single version too, so, it either support something or not | 17:24 |
| rhagarty | vponomaryov1: so how do you show that in the UI? things just don't work anymore? | 17:25 |
| vponomaryov1 | rhagarty: the only problem - to update all usages API usages in manila UI bumping microversion | 17:25 |
| bswartz | rhagarty: currently manila-ui is in the latter case as vponomaryov1 says | 17:25 |
| vponomaryov1 | rhagarty: not "doesn't work", "available" only that is supported by current microversion | 17:26 |
| rhagarty | vponomaryov1: but you would only have to bump the API's that care about microversions. Everything else just uses latest | 17:26 |
| vponomaryov1 | rhagarty: yes, update changed ones | 17:26 |
| vponomaryov1 | rhagarty: microversion changes can be tiny | 17:27 |
| vponomaryov1 | rhagarty: like this -> https://review.openstack.org/424748 | 17:27 |
| rhagarty | vponomaryov1: so if feature_a has a tab and requires 2.25, but server is running at 2.24, it would be nice to not show tab | 17:28 |
| vponomaryov1 | rhagarty: manila UI uses same microversion for every call | 17:28 |
| vponomaryov1 | rhagarty: so, if server does not support it, you get nothing at all | 17:28 |
| bswartz | ganso: any other changes in ps61 other than the rebase? | 17:29 |
| ganso | bswartz: reverted the HTTP 405 error to 400 | 17:29 |
| bswartz | ganso: why not 404? | 17:29 |
| ganso | bswartz: IMO 400 is more appropriate than 404, and can be consistent across acess-allow/deny and access-list | 17:30 |
| rhagarty | vponomaryov1: for now, but what happens when you want to replace a feature. Cinder has this issue with going from consistency groups to generic groups. They want to show one or the other depending on current server microversion support | 17:30 |
| bswartz | vponomaryov1: that may be what we do, but we don't have to -- we could make the UI smarter, which is what I think rhagarty is suggesting | 17:30 |
| ganso | bswartz: I don't like the idea of saying HTTP 404 Not found: "Canno control access of snapshot because it is not supported" | 17:30 |
| vponomaryov1 | rhagarty: as bswartz points out, I just described current approach | 17:31 |
| ganso | bswartz: s/Canno/Cannot | 17:31 |
| bswartz | ganso: I agree it's shitty | 17:31 |
| vponomaryov1 | ganso: actually, 404 is the answer you get when URL does not exist | 17:31 |
| ganso | vponomaryov1: yes, but the URL does exist | 17:32 |
| bswartz | ganso: you have to admin that 400 Bad Request is also confusing because it implies that the client send malformed data | 17:32 |
| vponomaryov1 | ganso: but should not | 17:32 |
| bswartz | s/admin/admit/ | 17:32 |
| vponomaryov1 | ganso: it is restriction of API router usage | 17:32 |
| ganso | vponomaryov1: the URL + resource combination is not valid though | 17:32 |
| bswartz | ganso: the resource is part of the URL in REST | 17:33 |
| rhagarty | well, I know there is an effort to standardize on getting back current server/client microversion support. Once this is in place, I suggest manila adopt it, and then use that information to make the UI work "better" | 17:33 |
| vponomaryov1 | ganso: in GET you have only URL | 17:33 |
| bswartz | the deeper problem here is that HTTP 400 is already used in many similar situations | 17:34 |
| vponomaryov1 | rhagarty: but I cannot say I like this idea as a developer )) | 17:34 |
| vponomaryov1 | rhagarty: it will be much more complicated | 17:34 |
| vponomaryov1 | rhagarty: to support each microversion separately, | 17:35 |
| bswartz | vponomaryov1: it's more work for us, but better experience for the user | 17:35 |
| bswartz | I think that's the argument rhagarty is making | 17:35 |
| ganso | bswartz: how about 403? | 17:35 |
| vponomaryov1 | bswartz: actually, we should focus only on latest microversions for each release | 17:35 |
| bswartz | there is no technical reason we can't do it, it's a question of effort | 17:35 |
| rhagarty | vponomaryov1: not that much harder... it just effects those APIs that were implemented after microversions. And you set them once and forget about them until they need to be replaced. | 17:36 |
| bswartz | vponomaryov1: that's the higher priority but if we had more resources we would do what rhagarty suggests too | 17:36 |
| bswartz | ganso: 403 is my 3rd choice | 17:36 |
| bswartz | ganso I prefer 404, then 400, then 403 | 17:36 |
| vponomaryov1 | bswartz: and, since we release manila-ui with all openstack, we really need only single microversion - latest for release | 17:36 |
| rhagarty | I'll sign up to do what I suggested | 17:37 |
| ganso | bswartz: can we settle for 400 then? :P | 17:37 |
| bswartz | ganso: honestly I just hate how REST limits itself to HTTP error codes only | 17:37 |
| vponomaryov1 | rhagarty, bswartz: support of multiple microversions is useful only when we have single version of manila-ui | 17:37 |
| *** shausy has joined #openstack-manila | 17:37 | |
| vponomaryov1 | rhagarty, bswartz: not tied to openstack releases | 17:37 |
| ganso | bswartz: since it will have the advantage of being the same response for allow and deny | 17:37 |
| bswartz | vponomaryov1: it's not unreasonable to say that the ocata version of manila-ui should work with both ocata servers and newton servers | 17:38 |
| vponomaryov1 | ganso: what is the point having the same answer? | 17:38 |
| vponomaryov1 | bswartz: actually, it is question of being compatible with Horizon too | 17:38 |
| *** cdelatte has joined #openstack-manila | 17:38 | |
| ganso | vponomaryov1: just consistency | 17:39 |
| vponomaryov1 | bswartz: it is really possible that you will just fail to use manila-ui with other version of horizon | 17:39 |
| vponomaryov1 | ganso: for consistency we can raise same error in all our APIs | 17:39 |
| bswartz | vponomaryov1: horizon version and manila-ui version must be compatible, as they're installed together | 17:39 |
| vponomaryov1 | ganso: don't see value in this kind of "consistency" | 17:39 |
| bswartz | however the server could run elsewhere and there could be more than 1 server to talk to | 17:39 |
| * bswartz crosses fingers | 17:40 | |
| bswartz | about to boot another ipv6 vm | 17:40 |
| ganso | vponomaryov1: thing is, error is the same for access_allow/deny and access-list, for the exact same error, I think it is preferable if the response code was consistent | 17:41 |
| rhagarty | and there are "real world" reasons to stick with a working version of a server, but want to have the latest UI | 17:41 |
| vponomaryov1 | bswartz: "complex" means "error-prone"r | 17:41 |
| vponomaryov1 | ganso: API URL are different | 17:41 |
| vponomaryov1 | ganso: consistency should be for allow/deny only | 17:41 |
| ganso | vponomaryov1: what code should allow/deny have then? | 17:42 |
| vponomaryov1 | ganso: have you read latest comments in gerrit? | 17:42 |
| *** belmoreira has quit IRC | 17:42 | |
| ganso | bswartz, markstur, tbarron, xyang_: Jenkins has voted https://review.openstack.org/#/c/424748 | 17:43 |
| vponomaryov1 | rhagarty: bugfixes in manila Ui and Horizon? | 17:43 |
| *** xyang_ has quit IRC | 17:43 | |
| rhagarty | vponomaryov1: sure | 17:43 |
| ganso | vponomaryov1: it says should stay 400, but how about this argument: "<bswartz> ganso: you have to admin that 400 Bad Request is also confusing because it implies that the client send malformed data" ?? | 17:44 |
| ganso | vponomaryov1: it is not malformed at all | 17:44 |
| vponomaryov1 | ganso: that is why 400 is not really suitable, it is not malformed | 17:45 |
| ganso | vponomaryov1: then why would we continue using 400 for POST? | 17:45 |
| vponomaryov1 | ganso: it is correct, but server refuses to process it | 17:46 |
| ganso | vponomaryov1: so, what is the correct code for that? | 17:46 |
| vponomaryov1 | ganso: it is debatable, as you can see | 17:46 |
| bswartz | honestly I hate the REST API error code situation so much I'm okay with using 400 as a "catch all" error code | 17:47 |
| ganso | bswartz: +1 | 17:47 |
| bswartz | I think that's how it's done in other projects | 17:47 |
| bswartz | nova and neutron give me HTTP 400 all the time when I made requests that a valid except for some value in the database | 17:47 |
| bswartz | that are* valid | 17:48 |
| vponomaryov1 | bswartz, ganso: we need to bump minimal client version for UI | 17:48 |
| ganso | vponomaryov1: another patch? | 17:49 |
| bswartz | vponomaryov: https://review.openstack.org/#/c/424205/ https://review.openstack.org/#/c/424207/ | 17:49 |
| bswartz | I will do another release later this week but I can do additional ones as needed | 17:49 |
| vponomaryov1 | ganso: to requirements project | 17:49 |
| bswartz | good point | 17:49 |
| ganso | vponomaryov1: oh I see | 17:50 |
| vponomaryov1 | bswartz: to requirements rpoject so we could have automatic requirements update | 17:50 |
| bswartz | yes | 17:52 |
| bswartz | g-r.txt and uc.txt | 17:52 |
| *** akapil has quit IRC | 17:52 | |
| bswartz | vponomaryov1: if you want to push I can +1 | 17:52 |
| bswartz | I have a meeting in a few minutes and I'm still fighting with the ipv6 routing tables | 17:53 |
| vponomaryov1 | bswartz: ok, I will make a change | 17:54 |
| ganso | markstur: thanks | 17:56 |
| *** a-pugachev has quit IRC | 17:57 | |
| markstur | ganso: I had to put an end to it. Couldn't just sit back and watch you and Valeriy +1 your own patches ;) | 18:00 |
| ganso | markstur: lol | 18:00 |
| ganso | markstur: it is sad to not be able to use +2 powers | 18:01 |
| markstur | It is good to pick on someone for voting on their own patch -- but in cases like this it is actually good to know that the contributors both approve of the other | 18:01 |
| *** yumiriam1 has joined #openstack-manila | 18:02 | |
| *** yumiriam has quit IRC | 18:04 | |
| vponomaryov1 | ganso, bswartz: https://review.openstack.org/#/c/424781/ | 18:06 |
| *** akapil has joined #openstack-manila | 18:07 | |
| *** yumiriam1 has quit IRC | 18:07 | |
| ganso | vponomaryov1: won't it need to be updated when we merge the share groups client patch? | 18:07 |
| *** akapil has quit IRC | 18:07 | |
| openstackgerrit | Merged openstack/manila: VMAX manila plugin - Support for VMAX in Manila https://review.openstack.org/404859 | 18:08 |
| *** akapil has joined #openstack-manila | 18:08 | |
| *** yumiriam has joined #openstack-manila | 18:09 | |
| vponomaryov1 | ganso: only when we will add its support to client | 18:10 |
| vponomaryov1 | ganso: 1.12.0 supports 2.29 - it is what you need | 18:10 |
| ganso | vponomaryov1: ok | 18:10 |
| vponomaryov1 | markstur, ganso: temporary amnesia about 24hour rule detected )) | 18:11 |
| *** jprovazn has joined #openstack-manila | 18:11 | |
| ganso | vponomaryov1: does 24 hour apply to FF? | 18:11 |
| ganso | vponomaryov1: hour *rule | 18:11 |
| vponomaryov1 | ganso: I am just kidding )) | 18:12 |
| ganso | vponomaryov1: as you can see, I am not good with jokes | 18:12 |
| vponomaryov1 | ganso: today or in general? )) | 18:12 |
| ganso | vponomaryov1: in general | 18:13 |
| markstur | we can laugh about it tomorrow | 18:13 |
| *** shausy has quit IRC | 18:14 | |
| ganso | vponomaryov1, markstur, bswartz, tbarron, cknight: https://review.openstack.org/#/c/406307 can now be merged without breaking previously microversioned changes | 18:15 |
| openstackgerrit | Rodrigo Barbieri proposed openstack/manila: Add mountable snapshots support to HNAS driver https://review.openstack.org/411474 | 18:25 |
| *** akapil has quit IRC | 18:25 | |
| openstackgerrit | Rodrigo Barbieri proposed openstack/manila: Mountable snapshots scenario tests https://review.openstack.org/412001 | 18:28 |
| * vponomaryov1 thinks that markstur is waiting while ganso sets +1 on https://review.openstack.org/#/c/406307 | 18:42 | |
| markstur | vponomaryov1: Stop spying on me | 18:43 |
| vponomaryov1 | markstur: coincidence | 18:44 |
| ganso | vponomaryov1, markstur: thanks | 18:50 |
| ganso | vponomaryov1: it seems bswartz has agreed with code 400 here https://review.openstack.org/#/c/345526/ | 18:50 |
| ganso | vponomaryov1: is there anything holding you back? | 18:50 |
| vponomaryov1 | ganso: I am looking at diff | 18:51 |
| ganso | vponomaryov1: great, thanks | 18:51 |
| vponomaryov1 | ganso: and making note about that 400 stuff )) | 18:51 |
| vponomaryov1 | "evil" | 18:51 |
| vponomaryov1 | ^_^ | 18:51 |
| ganso | vponomaryov1: ='( | 18:52 |
| vponomaryov1 | ganso: https://review.openstack.org/#/c/345526/59..61/manila/db/sqlalchemy/models.py | 18:55 |
| vponomaryov1 | ganso: first comment half fixed | 18:55 |
| openstackgerrit | Merged openstack/manila-ui: Implement Share Migration Ocata improvements https://review.openstack.org/406307 | 18:55 |
| vponomaryov1 | ganso: line 213 should have been fixed too | 18:56 |
| vponomaryov1 | ganso: "not touched" - more correctly | 18:56 |
| ganso | vponomaryov1: oops :( | 18:56 |
| *** xyang_ has joined #openstack-manila | 18:56 | |
| ganso | vponomaryov1: ok I'll wait for the last set of comments before I upload the updated patch | 18:57 |
| vponomaryov1 | ganso: ok | 18:58 |
| *** xyang_ has quit IRC | 18:59 | |
| *** ianychoi has quit IRC | 19:00 | |
| bswartz | hey I'm back | 19:01 |
| vponomaryov1 | ganso: no nee dto use "six" in https://review.openstack.org/#/c/345526/59..61/manila_tempest_tests/tests/api/test_snapshot_rules.py | 19:04 |
| vponomaryov1 | ganso: just wrapping with "list" would be enough | 19:05 |
| vponomaryov1 | ganso: have you solved problem found by client change? | 19:12 |
| vponomaryov1 | ganso: s/problem/problems/ | 19:12 |
| ganso | vponomaryov1: yes | 19:13 |
| vponomaryov1 | ganso: and what was the problem with "NoneType"? | 19:14 |
| ganso | vponomaryov1: incorrect ID was being passed, but I am unsure about how the object was able to skip the condition | 19:15 |
| *** lpetrut has quit IRC | 19:16 | |
| bswartz | crap | 19:17 |
| bswartz | does anyone know if cirros even has ipv6 support? | 19:17 |
| bswartz | my problem might be that the VM itself doesn't talk ipv6 | 19:17 |
| * bswartz goes to get an image of a real OS | 19:18 | |
| vponomaryov1 | bswartz: our service image should have ipv6 support | 19:18 |
| vponomaryov1 | bswartz: if I am not mistaken | 19:18 |
| vponomaryov1 | ganso: those 5 is all I found | 19:19 |
| bswartz | standard ubuntu cloud images are smaller than our service image | 19:19 |
| vponomaryov1 | bswartz: it does not have samba ) | 19:19 |
| vponomaryov1 | bswartz: and nfs by default | 19:19 |
| bswartz | I don't care just need the client bits | 19:19 |
| bswartz | I can always apt-get install stuff after I'm able to SSH in | 19:20 |
| bswartz | problem with cirros is I can't even ssh or ping6 | 19:20 |
| ganso | vponomaryov1: thanks, working on them | 19:21 |
| *** eharney has quit IRC | 19:22 | |
| openstackgerrit | Rodrigo Barbieri proposed openstack/manila: Add mountable snapshots support https://review.openstack.org/345526 | 19:30 |
| ganso | vponomaryov1: fixed ^ | 19:33 |
| *** yumiriam has quit IRC | 19:36 | |
| *** eharney has joined #openstack-manila | 19:37 | |
| bswartz | okay it wasn't the guest | 19:40 |
| bswartz | still neutron that's screwed up | 19:41 |
| *** harlowja has joined #openstack-manila | 19:41 | |
| cknight | ganso: NetApp CI is green again. https://review.openstack.org/#/c/357545/ | 19:42 |
| cknight | ganso: We enabled snapshot revert tests, so all other patches will fail our CI until this merges. | 19:42 |
| ganso | cknight: give me 10min I'll +2, I am on the road | 19:43 |
| cknight | ganso: No worries, thanks! | 19:44 |
| cknight | ganso: And I hope you're not driving. | 19:44 |
| bswartz | ganso has a neural link to IRC so he can talk on IRC and drive at the same time | 19:45 |
| ganso | bswartz: lol | 19:45 |
| ganso | cknight: no not driving at this moment :) | 19:45 |
| vponomaryov1 | ganso: you are literally on the road? )) | 19:47 |
| vponomaryov1 | ganso: I like fresh air too | 19:47 |
| vponomaryov1 | ganso: but not that much | 19:47 |
| ganso | vponomaryov1: was literally on the road | 19:48 |
| ganso | vponomaryov1: getting close to my building now | 19:48 |
| ganso | cknight: done | 19:53 |
| cknight | ganso: Thanks! | 19:54 |
| *** timcl has quit IRC | 19:56 | |
| *** xyang_ has joined #openstack-manila | 20:13 | |
| *** xyang_ has quit IRC | 20:15 | |
| *** timcl has joined #openstack-manila | 20:23 | |
| *** xyang_ has joined #openstack-manila | 20:33 | |
| *** xyang1 has joined #openstack-manila | 20:36 | |
| openstackgerrit | Merged openstack/manila: NetApp: Support share revert to snapshot https://review.openstack.org/357545 | 20:43 |
| bswartz | vponomaryov1: you still waiting on anything for the mountable snapshot patch? | 20:52 |
| *** jprovazn has quit IRC | 20:52 | |
| ganso | vponomaryov1: server and client patches passed CI | 20:54 |
| *** ChanServ changes topic to "OpenStack Share File Systems | Manila | Review focus https://etherpad.openstack.org/p/manila-ocata-code-review-focus" | 20:54 | |
| bswartz | I should have updated the topic last week | 20:55 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/manila: Updated from global requirements https://review.openstack.org/424616 | 20:55 |
| *** gcb has joined #openstack-manila | 21:01 | |
| *** kaisers_ has quit IRC | 21:01 | |
| *** porrua has quit IRC | 21:01 | |
| *** kaisers has joined #openstack-manila | 21:02 | |
| vponomaryov1 | bswartz: mountable snapshots destiny is in your hands | 21:04 |
| *** alyson_ has quit IRC | 21:07 | |
| *** lpetrut has joined #openstack-manila | 21:10 | |
| tbarron | bswartz: tommylikehu there is an issue with ipv6 exports that breaks lvm quite apart from the routing from hypervisor to host | 21:14 |
| tbarron | bswartz: tommylikehu I put notes in my latest reply on https://review.openstack.org/#/c/406776 | 21:14 |
| tbarron | bswartz: tommylikehu but essentially the 'exportfs ' command is returning shell status zero but failing to actually do an export | 21:15 |
| tbarron | bswartz: tommylikehu appears to be a known nfs-kernel-server bug | 21:15 |
| tbarron | bswartz: just saw your remarks above about cirros | 21:16 |
| tbarron | bswartz: apparently it does, but you need separate interface than for ipv4 | 21:16 |
| tbarron | bswartz: rather than getting cloudinit to make it work for me I moved to using a CentOS image, which was (for me) much easier to set up | 21:17 |
| tbarron | bswartz: I set it up for dhcp6 and dhcp4 on the same interface and it acquires both an addr from the neutron v4 cidr and from the neutron v6 cidr on the subnets associated with the neutron network that is attached to that single nic | 21:18 |
| tbarron | bswartz: I tried acquiring ipv6 addr with stateless slacc and got nothing | 21:19 |
| bswartz | tbarron: thanks | 21:19 |
| bswartz | tbarron: those kinds of bugs should be fixed | 21:19 |
| bswartz | tbarron: I'm about to give up (for now) on trying to get nova vms to have working ipv6 | 21:20 |
| tbarron | bswartz: I'd be interested in your devstack setup in any case | 21:20 |
| bswartz | I was hoping that investing a day in it would pay dividends, but all I'm getting is frustration dividends | 21:21 |
| tbarron | bswartz: you may have one end of this working and I may have another | 21:21 |
| bswartz | tbarron: I tried hard to make provider networks work | 21:21 |
| bswartz | it sort of worked for ipv4, although I somehow broke cloudinit | 21:21 |
| bswartz | for some reason ipv6 packet just go into a black hole | 21:21 |
| bswartz | tbarron: do you happen to know how to change neutron to use the noop firewall driver? | 21:22 |
| bswartz | I'd like to remove neutron's security layers so I can rule them out as the source of problems | 21:22 |
| tbarron | bswartz: no, but if I find out I'll share | 21:22 |
| tbarron | agree, had the same thought | 21:22 |
| tbarron | bswartz: do you have an ipv6 address (non link local) on your hypervisor external interface? | 21:23 |
| tbarron | bswartz: I can't get one there (though I have it on br-ex) | 21:23 |
| tbarron | bswartz: dunno if it's b/c of my virsh definition for the hypervisor or something about the weird ipv6-rd setup I have at home. | 21:24 |
| tbarron | my physical machines get ipv6 addr on their external interface just fine with ipv6-rd on my home router (the way I do IPv6 with CenturyLink ISP) | 21:25 |
| tbarron | bswartz: anyways, we have way too may obstacles in the way of doing end-to-end (actual mount to compute instance) testing of this new feature and I really think we need to see the results of such testing. | 21:27 |
| tbarron | bswartz: there may be some infra here worth merging in ocata anyways so that continues research/progress can be made on this front. | 21:27 |
| bswartz | tbarron: yes I have ipv6 addresses everywhere | 21:28 |
| tbarron | bswartz: but I don't see any end-user-exposable feature in this at the moment, despite my supporting the overall initiative | 21:28 |
| bswartz | and routes are working and slaac is working | 21:28 |
| *** timcl has quit IRC | 21:28 | |
| bswartz | but packets that enter the land of neutron are never seen again | 21:28 |
| tbarron | bswartz: for me too with physical machines but not with my xenial vm/hypervisor | 21:28 |
| bswartz | you use virt-manager/virsh on your hypervisor? | 21:29 |
| *** kaisers has quit IRC | 21:29 | |
| bswartz | or something else? | 21:29 |
| *** kaisers_ has joined #openstack-manila | 21:29 | |
| tbarron | bswartz: but within that vm/hypervisor neutron dhcp works to my compute instance | 21:29 |
| tbarron | :D | 21:29 |
| tbarron | I use virt-manager/virsh | 21:29 |
| tbarron | if you have a chance just dumpxml in a pastebin for me | 21:30 |
| bswartz | okay I have a frontend and a backend network configured in virt-manager | 21:30 |
| bswartz | yeah 1 sec | 21:30 |
| *** catintheroof has quit IRC | 21:35 | |
| *** catintheroof has joined #openstack-manila | 21:35 | |
| *** catintheroof has quit IRC | 21:35 | |
| *** kaisers has joined #openstack-manila | 21:45 | |
| *** kaisers_ has quit IRC | 21:48 | |
| *** Yogi1 has quit IRC | 21:50 | |
| *** cknight has quit IRC | 21:52 | |
| openstackgerrit | Merged openstack/python-manilaclient: Add share group support to Manila client https://review.openstack.org/335120 | 21:58 |
| *** breitz has quit IRC | 21:59 | |
| *** breitz has joined #openstack-manila | 22:00 | |
| *** kaisers_ has joined #openstack-manila | 22:01 | |
| *** kaisers__ has joined #openstack-manila | 22:02 | |
| *** kaisers__ has quit IRC | 22:04 | |
| *** kaisers has quit IRC | 22:04 | |
| *** kaisers has joined #openstack-manila | 22:05 | |
| *** kaisers_ has quit IRC | 22:05 | |
| *** vponomaryov1 has quit IRC | 22:07 | |
| *** eharney has quit IRC | 22:18 | |
| *** dustins has quit IRC | 22:19 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/manila-ui: Updated from global requirements https://review.openstack.org/424884 | 22:26 |
| openstackgerrit | Rodrigo Barbieri proposed openstack/python-manilaclient: Add mountable snapshots support to manila client https://review.openstack.org/345625 | 22:31 |
| *** lpetrut has quit IRC | 22:44 | |
| *** david-lyle has quit IRC | 22:50 | |
| *** david-lyle has joined #openstack-manila | 22:53 | |
| *** xyang_ has quit IRC | 22:57 | |
| *** xyang1 has quit IRC | 22:59 | |
| *** gcb has quit IRC | 23:06 | |
| *** harlowja has quit IRC | 23:08 | |
| *** tommylikehu_ has joined #openstack-manila | 23:23 | |
| *** tommylikehu_ has quit IRC | 23:28 | |
| *** mtanino_ has joined #openstack-manila | 23:29 | |
| *** mtanino has quit IRC | 23:30 | |
| *** harlowja has joined #openstack-manila | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!