| *** barclaac has quit IRC | 00:01 | |
| openstackgerrit | Doug Wiegley proposed openstack/neutron-lbaas: Post-split, get jenkins tests passing https://review.openstack.org/140182 | 00:19 |
|---|---|---|
| *** sbfox has quit IRC | 00:27 | |
| *** crc32 has quit IRC | 00:58 | |
| *** SumitNaiksatam has quit IRC | 01:15 | |
| *** mestery has quit IRC | 01:18 | |
| *** mestery has joined #openstack-lbaas | 01:20 | |
| *** xgerman has quit IRC | 01:21 | |
| *** mestery has quit IRC | 01:23 | |
| openstackgerrit | Doug Wiegley proposed openstack/neutron-lbaas: Post-split, get jenkins tests passing https://review.openstack.org/140182 | 01:24 |
| *** mlavalle has quit IRC | 01:27 | |
| *** SumitNaiksatam has joined #openstack-lbaas | 01:29 | |
| openstackgerrit | Doug Wiegley proposed openstack/neutron-lbaas: Post-split, get jenkins tests passing https://review.openstack.org/140182 | 01:57 |
| openstackgerrit | Doug Wiegley proposed openstack/neutron-lbaas: Post-split, get jenkins tests passing https://review.openstack.org/140182 | 02:02 |
| *** barclaac|2 has quit IRC | 02:21 | |
| *** fnaval has quit IRC | 02:30 | |
| *** SumitNaiksatam has quit IRC | 02:39 | |
| *** ptoohill_ has joined #openstack-lbaas | 02:45 | |
| *** ptoohill_ has quit IRC | 02:46 | |
| *** fnaval has joined #openstack-lbaas | 02:46 | |
| *** ptoohill_ has joined #openstack-lbaas | 02:46 | |
| *** mestery has joined #openstack-lbaas | 02:57 | |
| *** SumitNaiksatam has joined #openstack-lbaas | 03:14 | |
| *** sbalukoff has quit IRC | 03:21 | |
| *** mestery has quit IRC | 04:00 | |
| *** mestery has joined #openstack-lbaas | 04:22 | |
| dougwig | yes, yes, we are guinea pigs. | 04:34 |
| *** sbfox has joined #openstack-lbaas | 04:42 | |
| openstackgerrit | Doug Wiegley proposed openstack/neutron-lbaas: Post-split, get jenkins tests passing https://review.openstack.org/140182 | 04:46 |
| *** sbfox has quit IRC | 05:20 | |
| openstackgerrit | Doug Wiegley proposed openstack/neutron-lbaas: Post-split, get jenkins tests passing https://review.openstack.org/140182 | 05:32 |
| dougwig | aha, 2 out of 3 passing now. stupid environment differences. | 05:37 |
| *** sbfox has joined #openstack-lbaas | 05:38 | |
| openstackgerrit | Doug Wiegley proposed openstack/neutron-lbaas: Post-split, get jenkins tests passing https://review.openstack.org/140182 | 05:45 |
| openstackgerrit | Doug Wiegley proposed openstack/neutron-lbaas: Post-split, get jenkins tests passing https://review.openstack.org/140182 | 05:52 |
| *** jschwarz has joined #openstack-lbaas | 05:54 | |
| *** fnaval has quit IRC | 05:55 | |
| *** fnaval has joined #openstack-lbaas | 05:56 | |
| *** mestery has quit IRC | 05:57 | |
| *** fnaval has quit IRC | 06:00 | |
| dougwig | woohoo, passing. | 06:02 |
| *** ptoohill_ has quit IRC | 07:04 | |
| *** sbalukoff has joined #openstack-lbaas | 07:36 | |
| *** kobis has joined #openstack-lbaas | 08:06 | |
| *** barclaac has joined #openstack-lbaas | 08:18 | |
| *** kobis has left #openstack-lbaas | 08:19 | |
| *** barclaac has quit IRC | 08:22 | |
| *** sbfox has quit IRC | 09:39 | |
| *** woodster_ has quit IRC | 10:30 | |
| *** SumitNaiksatam has quit IRC | 11:22 | |
| *** jschwarz has quit IRC | 11:24 | |
| *** SumitNaiksatam has joined #openstack-lbaas | 11:26 | |
| *** enikanorov__ has joined #openstack-lbaas | 13:18 | |
| *** enikanorov_ has quit IRC | 13:18 | |
| *** woodster_ has joined #openstack-lbaas | 13:48 | |
| *** mestery has joined #openstack-lbaas | 13:59 | |
| *** mestery has quit IRC | 14:29 | |
| *** kobis has joined #openstack-lbaas | 14:45 | |
| *** fnaval has joined #openstack-lbaas | 14:51 | |
| openstackgerrit | Merged openstack/neutron-lbaas: Post-split, get jenkins tests passing https://review.openstack.org/140182 | 15:37 |
| *** ptoohill_ has joined #openstack-lbaas | 15:37 | |
| *** ptoohill_ has quit IRC | 15:37 | |
| *** ptoohill_ has joined #openstack-lbaas | 15:38 | |
| *** fnaval has quit IRC | 15:48 | |
| *** fnaval has joined #openstack-lbaas | 15:48 | |
| rm_work | 8m to meeting, woo | 15:52 |
| *** fnaval has quit IRC | 15:53 | |
| *** mestery has joined #openstack-lbaas | 15:55 | |
| *** vjay7 has joined #openstack-lbaas | 15:57 | |
| *** xgerman has joined #openstack-lbaas | 16:00 | |
| *** jorgem has joined #openstack-lbaas | 16:01 | |
| *** mlavalle has joined #openstack-lbaas | 16:17 | |
| *** fnaval has joined #openstack-lbaas | 16:18 | |
| blogan | dougwig: if no one wants to do teh alembic i will | 16:18 |
| blogan | but if someone has a lot of knowledge in it then i think it would be best if they did it | 16:19 |
| blogan | though i don't foresee it being too difficult | 16:19 |
| dougwig | blogan: hit it. bob chukka said that the group policy stackforge already set up its own alembic chain for neutron, so there is an example. | 16:19 |
| dougwig | you will be putting yourself on the critical path, though. :) | 16:19 |
| xgerman | blogan I think you know the most | 16:19 |
| sballe | blogan: I agree | 16:20 |
| xgerman | let me know if we need to get you an HP Cloud instance | 16:20 |
| xgerman | blogan -- | 16:20 |
| rm_work | T_T | 16:20 |
| dougwig | fyi, it is safe to clone the new repos now. | 16:21 |
| dougwig | i've started adding 'standalone' tasks in the split etherpad, if you're looking for something to do: https://etherpad.openstack.org/p/services_split | 16:26 |
| *** SumitNaiksatam has quit IRC | 16:28 | |
| *** vjay7 has quit IRC | 16:29 | |
| *** enikanorov has quit IRC | 16:30 | |
| *** enikanorov has joined #openstack-lbaas | 16:30 | |
| *** vjay7 has joined #openstack-lbaas | 16:32 | |
| blogan | dougwig: essentially we just need all tables for v1 for the initial alembic right? | 16:39 |
| *** vjay7 has quit IRC | 16:40 | |
| dougwig | how about the upgrade case from icehouse? :) it might need an initial state, which it only runs on fresh install. or it might need to leave those tables to neutron's chains and just start diverging somehow. | 16:41 |
| blogan | you mean if somehow upgrades from icehouse to kilo, they may have data in the tables that exist in neutron, so we need to migrate that data over as well? | 16:43 |
| blogan | s/somehow/someone | 16:43 |
| *** vjay7 has joined #openstack-lbaas | 16:44 | |
| dougwig | it's all one db, we're just taking over the tables. so how do we handle new installs vs upgrades? | 16:47 |
| blogan | oh god i forgot we decided on that | 16:47 |
| blogan | isn't just as simple as checking if the table exists, and if it doesn't we create it? | 16:49 |
| blogan | are we removing foreign keys? | 16:49 |
| blogan | ok so if someone upgrades from icehouse, they also need to get hte juno chagnes | 16:50 |
| blogan | if any | 16:50 |
| blogan | and kilo changes if any | 16:51 |
| blogan | and if they install fresh they should just get fresh tables from kilo state | 16:51 |
| blogan | juno migrations for lbaas remain in neutron, so i think its just a matter of getting the sequence right | 16:52 |
| blogan | dougwig: thinking out loud here so tell me if i've got it wrong | 16:52 |
| *** SumitNaiksatam has joined #openstack-lbaas | 16:52 | |
| dougwig | i think that sounds sane. no reason to rework the big neutron change unless there's a strong reason. | 16:53 |
| blogan | and then of course downgrading will be a pain | 16:53 |
| blogan | not sure how I will be able to tell if a person is downgrading from a fresh install versus an upgrade | 16:53 |
| xgerman | blogan, downgrading is something only jorgem likes :-) | 16:58 |
| dougwig | does alembic have a table for migration state? can you put a flag in it? | 16:58 |
| blogan | xgerman: bc making mistakes is acceptable to him, not for me | 16:58 |
| xgerman | lol | 16:59 |
| blogan | dougwig: yeah ill need to look into that | 16:59 |
| dougwig | be prepared to stamp this out for vpn and fw when you're done, or find some minions to do those two, please. | 16:59 |
| jorgem | xgerman: he yo! | 17:01 |
| blogan | dougwig: that'll be fine, ocne i figure it out here ill have it figured out for all | 17:01 |
| blogan | though will need someone knowledgeable people with those | 17:01 |
| dougwig | paul and sumit are standing by to help. | 17:02 |
| *** barclaac|2 has joined #openstack-lbaas | 17:26 | |
| xgerman | ok, cloned the repos now... | 17:30 |
| kobis | any news about the lbaas mid-cycle meetup? | 17:31 |
| dougwig | last i heard was week of feb 2nd. | 17:35 |
| xgerman | blogan I know this is embarassing but I loast the link to the LBaaS V2 API docs Min created | 17:38 |
| xgerman | I am wondering if you still have that | 17:38 |
| sballe | dougwig: Do you have the install steps for the new LBaaS handy? | 17:43 |
| dougwig | sballe: https://wiki.openstack.org/wiki/Neutron/LBaaS/DeployWithDevstack | 17:45 |
| sballe | thx | 17:45 |
| *** vivek-ebay has joined #openstack-lbaas | 17:46 | |
| *** sbfox has joined #openstack-lbaas | 17:52 | |
| blogan | xgerman: one sec | 18:20 |
| blogan | xgerman: https://review.openstack.org/#/c/110824/ | 18:20 |
| blogan | xgerman: and https://review.openstack.org/#/c/106434/ | 18:20 |
| blogan | not sure which is which | 18:21 |
| xgerman | ok, thanks - also Min came in and gace it to me | 18:21 |
| blogan | ah | 18:21 |
| xgerman | do we want to revive it :-) | 18:21 |
| blogan | at some point, but its not immediately needed | 18:21 |
| xgerman | cool | 18:21 |
| *** rohara2 has quit IRC | 18:22 | |
| *** rohara has joined #openstack-lbaas | 18:23 | |
| *** barclaac|3 has joined #openstack-lbaas | 18:28 | |
| sballe | dougwig: Do we want to continue to use lbaaS v1 in devstack or are we moving it to V2? I am assumign Ve but just wanted to check. | 18:30 |
| sballe | s/Ve/V2 | 18:30 |
| dougwig | for the split, v1. | 18:30 |
| sballe | ok so we leave it the way it is | 18:30 |
| xgerman | we want v2 :-) | 18:30 |
| dougwig | soon, german, soon. | 18:31 |
| sballe | :-) | 18:31 |
| *** barclaac|2 has quit IRC | 18:31 | |
| blogan | dougwig: will the providerresourceassociations table continue to be owned by neutron? | 18:32 |
| blogan | at least for teh split | 18:32 |
| blogan | by own, i mean neutron does the manages the migration chain | 18:32 |
| blogan | since this table is used by all advanced services, at least that is the structure | 18:33 |
| *** kobis has quit IRC | 18:38 | |
| *** sbfox has quit IRC | 18:45 | |
| *** sbfox has joined #openstack-lbaas | 18:55 | |
| *** jorgem has quit IRC | 19:00 | |
| openstackgerrit | German Eichberger proposed openstack/neutron-lbaas: WIP - service insertion https://review.openstack.org/140438 | 19:12 |
| *** jorgem has joined #openstack-lbaas | 19:23 | |
| dougwig | blogan: yes, neutron | 19:25 |
| blogan | dougwig: someone who installs kilo fresh will still need to run through all the neutron migrations right? | 19:26 |
| dougwig | yes | 19:26 |
| blogan | dougwig: and we're not moving any migrations for lbaas out of the icehouse and juno chain | 19:26 |
| dougwig | that seems reasonable to me, assuming we can do future mods to those tables from our alembic chain? | 19:27 |
| blogan | so wouldn't the fresh installs also run through that same migration?, we just need to make sure everything thats been done in kilo gets moved over | 19:27 |
| dougwig | it's an issue of which alembic chain owns the table, which if alembic doesn't track, i'm not concerned about. | 19:27 |
| dougwig | if that made any sense. | 19:27 |
| blogan | if they're using the same database then it shouldn't matter | 19:30 |
| blogan | once we split out into different dbs i see this being more effort | 19:30 |
| *** cipcosma has joined #openstack-lbaas | 19:32 | |
| dougwig | you're *sure* that alembic doesn't track which chain owns which table? | 19:32 |
| blogan | im going to be 100% certain before i commit to my answer, i know you will remember for eternity if im wrong | 19:34 |
| *** barclaac|3 has quit IRC | 19:35 | |
| *** barclaac|2 has joined #openstack-lbaas | 19:40 | |
| blogan | dougwig: so im probably wrong, alembic won't work if the alembic_version table has a migration name it can't find | 19:42 |
| dougwig | which means for us? | 19:43 |
| blogan | dougwig: looking for a workaround, i have none in mind right now, unless alembic has a mechanism to search for migrations in many locations | 19:44 |
| dougwig | did you check out the group based policy example? | 19:44 |
| blogan | yep nothing stood out | 19:45 |
| blogan | though im sure i missed something | 19:45 |
| dougwig | hang on, i know we can do separate chains, handling them independently. are you saying we can't? or something else? | 19:57 |
| *** jorgem has quit IRC | 19:59 | |
| dougwig | blogan: ^^ | 19:59 |
| blogan | dougwig: im saying we can do separate chains, but if we're doing a fork which is what this is, then i dont know if we can, im looking, gbp should hve had the same issue | 20:00 |
| blogan | dougwig: ivar lazzaro wouldn't happen to be there would he? | 20:01 |
| dougwig | no | 20:02 |
| dougwig | ok, what do you recommend? | 20:03 |
| blogan | reading about alembic branches | 20:03 |
| blogan | dougwig: branches require the parent to be in the same directory, so it goes back again to telling alembic to look in multiple locations for migrations | 20:05 |
| dougwig | hmm, neutron-db-manage could copy them all into a tmp directory. (gross) | 20:05 |
| *** mwang2 has joined #openstack-lbaas | 20:05 | |
| blogan | dougwig: yeah, but gbp had to have had this same problem | 20:05 |
| blogan | I've already asked SumitNaiksatam but he's afk | 20:05 |
| dougwig | ping rkukura | 20:05 |
| dougwig | he's sitting next to me if he doesn't reply | 20:06 |
| blogan | he worked on gbp? | 20:08 |
| *** vjay7 has quit IRC | 20:09 | |
| dougwig | yes. and he's the one that brought up their alembic implementation to me. | 20:12 |
| blogan | dougwig: okay think i got it now, one sec | 20:16 |
| dougwig | k | 20:17 |
| blogan | dougwig: we want neutron-db-manage to use this correct? | 20:22 |
| blogan | not have our own client | 20:22 |
| *** BB_ has joined #openstack-lbaas | 20:22 | |
| blogan | i mean all services uses neutron-db-manage | 20:22 |
| dougwig | i think that'd be cleaner, but if separate scripts makes life easier, i'm sure we can doc it. | 20:22 |
| dougwig | clenaer for the user, not cleaner for us. | 20:22 |
| BB_ | hello. I configured LBAAS according to link http://docs.openstack.org/admin-guide-cloud/content/install_neutron-lbaas-agent.html | 20:23 |
| blogan | yeah, its cleaner for the user | 20:23 |
| BB_ | i would like to configure them, like VM-A talks to VM-B using LB. | 20:23 |
| BB_ | Is there a link explaining that configuration ? | 20:23 |
| BB_ | Folks, any help on configurtion ? | 20:33 |
| dougwig | BB_: sorry, we're in the midst of splitting repos from neutron, so are all a bit distracted just now. | 20:39 |
| dougwig | are you around later? | 20:39 |
| openstackgerrit | Brandon Logan proposed openstack/neutron-lbaas: Initializing alembic for separate chain https://review.openstack.org/140472 | 20:41 |
| blogan | dougwig: ^^ won't work without neutron-db-manage fix which I will do after an appointment i have to make | 20:41 |
| blogan | so I'll have it done by tonight | 20:41 |
| blogan | dougwig: however you can test with the cli.py file in that review | 20:42 |
| blogan | whoops didn't mean to keep one of those files | 20:42 |
| *** crc32 has joined #openstack-lbaas | 20:42 | |
| openstackgerrit | Brandon Logan proposed openstack/neutron-lbaas: Initializing alembic for separate chain https://review.openstack.org/140472 | 20:43 |
| openstackgerrit | Brandon Logan proposed openstack/neutron-lbaas: Initializing alembic for separate chain https://review.openstack.org/140472 | 20:44 |
| johnsom | dougwig: For the devstack task, I am guessing we want vpn/fw/lbaas to become "extras", in extras.d? | 20:47 |
| dougwig | i've never modified devstack, so your guess is as good as mine. | 20:47 |
| johnsom | Grin. I see that opendaylight is setup that way. Does anyone else in the room have an opinion? | 20:48 |
| dougwig | my only opinion is that all the existing conf files that specific q-lbaas or whatnot work before and after, just with an extra package installed. | 20:49 |
| sballe | johnsom: I looked at this too. Why wouldn't you just add it to the lib/neutron? and stackrc? | 20:51 |
| sballe | They are core services and not plugins like opendaylight | 20:51 |
| johnsom | Ok. That was what I was trying to figure out. That is a lot easier too. Thanks sballe | 20:52 |
| sballe | johnsom: that was just my gut feeling when I looked at it | 20:52 |
| *** sbfox has quit IRC | 20:59 | |
| xgerman | dougwig, so for the tests we are pulling in the config from neutron -- is this ok or do we want our own config? | 21:25 |
| xgerman | neutron-lbaas/.venv/src/neutron/neutron/tests/etc/neutron.conf.test | 21:26 |
| *** vivek-ebay has quit IRC | 21:31 | |
| *** mestery_ has joined #openstack-lbaas | 21:37 | |
| *** mestery has quit IRC | 21:38 | |
| *** mlavalle has quit IRC | 21:39 | |
| BB_ | dougwig: okay. let me know when you guys are fre | 21:44 |
| *** mestery_ is now known as mestery | 21:51 | |
| *** cipcosma has quit IRC | 21:51 | |
| dougwig | xgerman: fine to pull in neutron config | 21:52 |
| *** mlavalle has joined #openstack-lbaas | 21:53 | |
| *** xgerman_ has joined #openstack-lbaas | 21:58 | |
| SumitNaiksatam | dougwig: blogan: for gbp we have an independent migration script, if that was the question | 22:23 |
| *** vivek-ebay has joined #openstack-lbaas | 22:25 | |
| dougwig | blogan found ivar. review is linked above. | 22:25 |
| blogan | dougwig, SumitNaiksatam: i have a feeling that if we want neutron-db-manage to manage all services, we're going to have to modify neutron's env.py, but im about to start working on that closely | 22:29 |
| SumitNaiksatam | blogan: okay | 22:30 |
| dougwig | neutron changes are relatively easy today and tomorrow. :) | 22:30 |
| blogan | yes i know, which is hwy i want to get this done quickly | 22:31 |
| blogan | as im sure you do too | 22:31 |
| blogan | as well as everyone | 22:31 |
| *** blogan_ has joined #openstack-lbaas | 22:32 | |
| *** SumitNaiksatam has quit IRC | 22:37 | |
| *** sballe is now known as sballe_mobile | 22:38 | |
| *** sballe_mobile is now known as sballe | 22:38 | |
| *** SumitNaiksatam has joined #openstack-lbaas | 22:48 | |
| dougwig | plugin shims - https://review.openstack.org/#/c/140515/ | 22:59 |
| *** mikedillion has joined #openstack-lbaas | 23:16 | |
| blogan_ | sweet, dont have to make changes to neutron migration's env.py | 23:17 |
| blogan_ | dougwig: which do you prefer neutron-db-manage -service lbaas,fwaas,vpnaas runs migrations for all 3 advanced services, but not neutron, or runs for all 3 AND neutron | 23:18 |
| blogan_ | --service | 23:18 |
| blogan_ | first option may as well just be its own client, while the second one seems more in line with a unified client | 23:20 |
| *** mikedillion has quit IRC | 23:23 | |
| dougwig | i don't have a strong opinion. | 23:23 |
| *** mikedillion has joined #openstack-lbaas | 23:24 | |
| blogan_ | im going to do the second option | 23:29 |
| *** sbfox has joined #openstack-lbaas | 23:36 | |
| *** ajmiller has quit IRC | 23:36 | |
| dougwig | ok | 23:37 |
| dougwig | the neutron mass delete is in the merge queue! | 23:37 |
| dougwig | +36, -24982 | 23:37 |
| blogan_ | yowzers | 23:39 |
| blogan_ | gut that pig | 23:39 |
| blogan_ | well im dumb for not thinking of this | 23:40 |
| blogan_ | if someone does a neutron-db-manage upgrade <migration_name> it will only work for one of the four | 23:41 |
| *** mestery has quit IRC | 23:50 | |
| *** ptoohill_ has quit IRC | 23:52 | |
| blogan_ | im just going to do one at a time | 23:53 |
| dougwig | ok | 23:56 |
| dougwig | at some point, they'll have to do the flows separately. and the packager can script doing four as easily as one. | 23:57 |
| *** ptoohill_ has joined #openstack-lbaas | 23:57 | |
| blogan_ | yeah we'll have our own client at some point, and it won't be hard | 23:57 |
| blogan_ | then again this is going to require almost the same amoutn of work as if we had our own client | 23:58 |
| blogan_ | as in now devstack is going to have to run neutron-db-manage with the --service for each service, and without the --service for neutron | 23:59 |
| blogan_ | ill get it up in review though, and people can comment on it | 23:59 |
| *** crc32 has quit IRC | 23:59 | |
| *** mestery has joined #openstack-lbaas | 23:59 | |
| dougwig | ok | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!