| *** amotoki has joined #openstack-lbaas | 00:00 | |
| *** outofmemory is now known as reedip | 00:04 | |
| *** yamamoto_ has quit IRC | 00:06 | |
| *** amotoki has quit IRC | 00:15 | |
| *** chlong has quit IRC | 00:20 | |
| *** chlong has joined #openstack-lbaas | 00:20 | |
| *** diogogmt has quit IRC | 00:28 | |
| *** chlong has quit IRC | 00:41 | |
| openstackgerrit | Doug Wiegley proposed openstack/neutron-lbaas: WIP - Test state of gate https://review.openstack.org/272338 | 00:41 |
|---|---|---|
| *** chlong has joined #openstack-lbaas | 00:41 | |
| *** chlong has quit IRC | 00:48 | |
| *** chlong has joined #openstack-lbaas | 00:49 | |
| xgerman | blogan dougwig ajmiller trevorV https://review.openstack.org/#/c/271544/ | 00:50 |
| ajmiller | hi | 00:50 |
| xgerman | trying to get this keystone stuff sorted out for clouds bigger than devstack | 00:50 |
| xgerman | ajmiller fishing for +2s :-) | 00:51 |
| ajmiller | LOL, yeah, I gathered as much ;) | 00:51 |
| *** fnaval has quit IRC | 00:52 | |
| *** chlong has quit IRC | 00:53 | |
| *** chlong has joined #openstack-lbaas | 00:53 | |
| *** Aish has joined #openstack-lbaas | 00:56 | |
| ajmiller | I have it +2. Aish and I were using that code as a necessary part of some of the things we were working on earlier today, and it works well. | 00:57 |
| ajmiller | Heading for the ferry.... | 00:58 |
| *** ajmiller has quit IRC | 00:58 | |
| xgerman | thanks ajmiller | 00:58 |
| *** bana_k has quit IRC | 01:00 | |
| *** amotoki has joined #openstack-lbaas | 01:04 | |
| *** ducttape_ has joined #openstack-lbaas | 01:08 | |
| *** yuanying has joined #openstack-lbaas | 01:10 | |
| openstackgerrit | min wang proposed openstack/octavia: Add anti-affinity server group feature for octavia https://review.openstack.org/272344 | 01:10 |
| *** yuanying_ has quit IRC | 01:11 | |
| *** minwang2 has quit IRC | 01:12 | |
| *** pc_m has quit IRC | 01:13 | |
| *** madhu_ak has quit IRC | 01:13 | |
| *** pc_m has joined #openstack-lbaas | 01:15 | |
| *** fnaval has joined #openstack-lbaas | 01:15 | |
| *** fnaval has quit IRC | 01:15 | |
| *** _cjones_ has quit IRC | 01:18 | |
| openstackgerrit | Stephen Balukoff proposed openstack/octavia: Add APP_COOKIE session persistence type https://review.openstack.org/272345 | 01:18 |
| *** _cjones_ has joined #openstack-lbaas | 01:19 | |
| *** _cjones_ has quit IRC | 01:23 | |
| *** ducttape_ has quit IRC | 01:24 | |
| *** diogogmt has joined #openstack-lbaas | 01:24 | |
| *** Aish has quit IRC | 01:40 | |
| *** paco20151113 has joined #openstack-lbaas | 01:42 | |
| *** jaff_cheng has joined #openstack-lbaas | 01:44 | |
| *** chlong has quit IRC | 01:55 | |
| *** chlong has joined #openstack-lbaas | 01:55 | |
| *** yuanying has quit IRC | 02:00 | |
| *** yuanying has joined #openstack-lbaas | 02:03 | |
| *** davidlenwell has quit IRC | 02:14 | |
| *** yamamoto_ has joined #openstack-lbaas | 02:16 | |
| *** davidlenwell has joined #openstack-lbaas | 02:16 | |
| sbalukoff | Man, tempest tests take a while to run locally. :P Is there a simple way to skip some of these so I can just concentrate on the one that's failing? | 02:23 |
| a2hill | sbalukoff: You can give it a dotted path to the test file/method of your choosing. | 02:23 |
| sbalukoff | Er.... how do you do that specifically? | 02:24 |
| *** yuanying has quit IRC | 02:25 | |
| a2hill | This might work also | 02:25 |
| a2hill | http://blog.jasonmeridth.com/posts/how-to-run-a-single-test-with-tox/ | 02:25 |
| a2hill | But you can do tox -e apiv2 neutron_lbaas.tests.api.something.testfile.testmethod | 02:25 |
| *** yuanying has joined #openstack-lbaas | 02:27 | |
| *** diogogmt has quit IRC | 02:28 | |
| sbalukoff | Danke schoen! | 02:29 |
| *** bdrich_ has joined #openstack-lbaas | 02:29 | |
| a2hill | ; | 02:30 |
| a2hill | ;) | 02:30 |
| *** yamamoto_ has quit IRC | 02:42 | |
| *** Aish has joined #openstack-lbaas | 02:47 | |
| *** Aish has left #openstack-lbaas | 02:47 | |
| *** chlong has quit IRC | 02:49 | |
| *** chlong has joined #openstack-lbaas | 02:49 | |
| *** superflyy has joined #openstack-lbaas | 02:50 | |
| *** ducttape_ has joined #openstack-lbaas | 02:52 | |
| *** ducttape_ has quit IRC | 03:01 | |
| *** sbalukoff has quit IRC | 03:02 | |
| *** bochi-michael has joined #openstack-lbaas | 03:06 | |
| *** ducttape_ has joined #openstack-lbaas | 03:12 | |
| *** diogogmt has joined #openstack-lbaas | 03:17 | |
| *** chlong has quit IRC | 03:17 | |
| *** chlong has joined #openstack-lbaas | 03:18 | |
| *** chlong has quit IRC | 03:19 | |
| *** yuanying_ has joined #openstack-lbaas | 03:19 | |
| *** chlong has joined #openstack-lbaas | 03:20 | |
| *** yuanyin__ has joined #openstack-lbaas | 03:20 | |
| *** yuanying has quit IRC | 03:22 | |
| *** yuanying_ has quit IRC | 03:23 | |
| *** Aish has joined #openstack-lbaas | 03:26 | |
| *** Aish has quit IRC | 03:26 | |
| *** woodster_ has quit IRC | 03:26 | |
| *** yamamoto_ has joined #openstack-lbaas | 03:28 | |
| *** yuanyin__ has quit IRC | 03:29 | |
| *** yuanying has joined #openstack-lbaas | 03:29 | |
| *** amotoki_ has joined #openstack-lbaas | 03:42 | |
| *** harlowja has quit IRC | 03:54 | |
| *** yuanying has quit IRC | 03:57 | |
| *** yuanying has joined #openstack-lbaas | 03:58 | |
| *** ducttape_ has quit IRC | 03:58 | |
| *** superflyy has quit IRC | 04:01 | |
| *** yuanying has quit IRC | 04:02 | |
| *** yuanying has joined #openstack-lbaas | 04:08 | |
| *** amotoki has quit IRC | 04:08 | |
| *** amotoki_ has quit IRC | 04:20 | |
| *** sbalukoff has joined #openstack-lbaas | 04:22 | |
| *** amotoki has joined #openstack-lbaas | 04:28 | |
| *** bdrich_ has quit IRC | 04:36 | |
| *** minwang2 has joined #openstack-lbaas | 04:44 | |
| *** diogogmt has quit IRC | 04:56 | |
| *** bdrich_ has joined #openstack-lbaas | 04:57 | |
| *** yamamot__ has joined #openstack-lbaas | 05:25 | |
| *** yamamoto_ has quit IRC | 05:26 | |
| *** bana_k has joined #openstack-lbaas | 05:34 | |
| openstackgerrit | caoyue proposed openstack/octavia: Add hacking check to ensure not use xrange() https://review.openstack.org/272393 | 05:49 |
| *** amotoki has quit IRC | 06:19 | |
| openstackgerrit | caoyue proposed openstack/octavia: Add hacking check to ensure not use xrange() https://review.openstack.org/272393 | 06:24 |
| *** redrobot has left #openstack-lbaas | 06:26 | |
| *** redrobot has joined #openstack-lbaas | 06:27 | |
| *** rcernin has joined #openstack-lbaas | 06:53 | |
| *** amotoki has joined #openstack-lbaas | 07:27 | |
| openstackgerrit | caoyue proposed openstack/octavia: Add hacking check to ensure not use xrange() https://review.openstack.org/272393 | 07:51 |
| *** prabampm has quit IRC | 08:08 | |
| openstackgerrit | Reedip proposed openstack/neutron-lbaas: range() is not same in py3.x and py2.x https://review.openstack.org/267963 | 08:11 |
| *** armax has joined #openstack-lbaas | 08:16 | |
| *** nmagnezi_ has joined #openstack-lbaas | 08:26 | |
| *** amotoki has quit IRC | 08:28 | |
| *** amotoki has joined #openstack-lbaas | 08:29 | |
| *** armax has quit IRC | 08:49 | |
| *** bdrich_ has quit IRC | 08:52 | |
| *** armax has joined #openstack-lbaas | 08:55 | |
| openstackgerrit | Stephen Balukoff proposed openstack/octavia: Shared pools support https://review.openstack.org/256369 | 09:25 |
| openstackgerrit | caoyue proposed openstack/octavia: Add hacking check to ensure not use xrange() https://review.openstack.org/272393 | 09:40 |
| *** minwang2 has quit IRC | 09:45 | |
| *** paco20151113 has quit IRC | 10:03 | |
| *** jaff_cheng has quit IRC | 10:25 | |
| *** yamamot__ has quit IRC | 10:27 | |
| openstackgerrit | LIU Yulong proposed openstack/neutron-lbaas: Restrict delete pool if it has binding healthmonitor https://review.openstack.org/179684 | 10:31 |
| *** bochi-michael has quit IRC | 10:44 | |
| *** bdrich_ has joined #openstack-lbaas | 10:49 | |
| *** bdrich_ has quit IRC | 10:54 | |
| *** chlong has quit IRC | 11:05 | |
| *** chlong has joined #openstack-lbaas | 11:10 | |
| *** amotoki has quit IRC | 11:20 | |
| *** amotoki has joined #openstack-lbaas | 11:56 | |
| *** eranra has joined #openstack-lbaas | 12:14 | |
| *** yamamoto has joined #openstack-lbaas | 12:15 | |
| *** doug-fish has joined #openstack-lbaas | 12:21 | |
| *** doug-fish has quit IRC | 12:23 | |
| *** yamamoto has quit IRC | 12:23 | |
| *** doug-fish has joined #openstack-lbaas | 12:24 | |
| *** yamamoto has joined #openstack-lbaas | 12:24 | |
| *** eranra has quit IRC | 12:25 | |
| *** yamamoto has quit IRC | 12:28 | |
| *** doug-fish has quit IRC | 12:28 | |
| *** doug-fish has joined #openstack-lbaas | 12:31 | |
| *** rtheis has joined #openstack-lbaas | 12:35 | |
| *** doug-fish has quit IRC | 12:35 | |
| openstackgerrit | Evgeny Fedoruk proposed openstack/neutron-lbaas: L7 capability extension implementation for lbaas v2 https://review.openstack.org/148232 | 13:05 |
| *** mhayden has joined #openstack-lbaas | 13:24 | |
| mhayden | i'm trying to write up some deployment documentation (and some ansible code) around deploying lbaasv2, but i'm tripped up on how the lbaasv2 agent communicates with octavia | 13:24 |
| mhayden | or if octavia is required | 13:24 |
| *** doug-fish has joined #openstack-lbaas | 13:30 | |
| *** doug-fish has quit IRC | 13:37 | |
| *** doug-fish has joined #openstack-lbaas | 13:38 | |
| *** doug-fish has quit IRC | 13:43 | |
| *** diogogmt has joined #openstack-lbaas | 13:57 | |
| *** amotoki has quit IRC | 14:09 | |
| *** neelashah has joined #openstack-lbaas | 14:12 | |
| *** diogogmt has quit IRC | 14:20 | |
| *** bdrich_ has joined #openstack-lbaas | 14:22 | |
| *** doug-fish has joined #openstack-lbaas | 14:26 | |
| *** doug-fish has quit IRC | 14:26 | |
| *** bdrich_ has quit IRC | 14:27 | |
| *** doug-fish has joined #openstack-lbaas | 14:27 | |
| *** davidlenwell has quit IRC | 14:35 | |
| *** davidlenwell has joined #openstack-lbaas | 14:49 | |
| *** ducttape_ has joined #openstack-lbaas | 14:56 | |
| xgerman | mhayden Octavia is the reference implementation since liberty | 14:57 |
| xgerman | though the old agent based driver should still work | 14:58 |
| mhayden | so i assume that one cannot have lbaasv2 without the octavia daemons? | 14:58 |
| xgerman | not without diverging from our standard setup | 15:02 |
| mhayden | makes sense | 15:04 |
| mhayden | so the neutron-lbaasv2-agent would make requests to the octavia-api for load balancers? | 15:05 |
| mhayden | i've been on a hunt for docs for a while, so feel free to point me there if there's a good explanation i missed | 15:05 |
| *** amotoki has joined #openstack-lbaas | 15:08 | |
| *** neelashah has quit IRC | 15:11 | |
| *** neelashah has joined #openstack-lbaas | 15:12 | |
| *** eranra has joined #openstack-lbaas | 15:14 | |
| *** neelashah1 has joined #openstack-lbaas | 15:15 | |
| *** eranra_ has joined #openstack-lbaas | 15:15 | |
| *** neelashah has quit IRC | 15:16 | |
| xgerman | octavia has a plugin which sends stuff straight to Octavia… so no more lbaasv2 agents | 15:17 |
| *** neelashah has joined #openstack-lbaas | 15:18 | |
| *** eranra has quit IRC | 15:19 | |
| *** neelashah1 has quit IRC | 15:19 | |
| mhayden | that makes things a little easier | 15:21 |
| mhayden | is this still a valid way to test and review working configs? http://docs.openstack.org/developer/devstack/guides/devstack-with-lbaas-v2.html | 15:21 |
| *** nmagnezi_ has quit IRC | 15:22 | |
| *** ajmiller has joined #openstack-lbaas | 15:24 | |
| xgerman | yep, that should still work | 15:25 |
| xgerman | in fact none of the commands change with Octavia - just troubleshooting becomes different | 15:26 |
| *** nmagnezi_ has joined #openstack-lbaas | 15:34 | |
| *** ajmiller has quit IRC | 15:37 | |
| *** ajmiller has joined #openstack-lbaas | 15:38 | |
| *** armax has quit IRC | 15:42 | |
| *** bdrich_ has joined #openstack-lbaas | 15:43 | |
| *** fnaval has joined #openstack-lbaas | 15:57 | |
| *** TrevorV has joined #openstack-lbaas | 15:59 | |
| *** armax has joined #openstack-lbaas | 16:01 | |
| *** prabampm has joined #openstack-lbaas | 16:05 | |
| *** amotoki has quit IRC | 16:16 | |
| *** diogogmt has joined #openstack-lbaas | 16:26 | |
| *** _ducttape_ has joined #openstack-lbaas | 16:27 | |
| *** chlong is now known as chlong_zzz | 16:28 | |
| blogan | sbalukoff: ping | 16:28 |
| blogan | xgerman: didn't we decide to remove the APP_COOKIE session persistence a long time ago? | 16:28 |
| blogan | we just ahven't yet | 16:28 |
| blogan | for some reason I feel like I remember that | 16:28 |
| xgerman | Cookies are just another HEADER entry | 16:29 |
| xgerman | but I forgot that issue... | 16:29 |
| johnsom | I don't remember that. | 16:29 |
| nmagnezi_ | blogan, xgerman, hey guys | 16:30 |
| blogan | hey nmagnezi_ | 16:30 |
| xgerman | hi | 16:30 |
| blogan | xgerman, johnsom: i could be going insane | 16:30 |
| *** ducttape_ has quit IRC | 16:30 | |
| xgerman | blogan you want an answer? :-) | 16:30 |
| johnsom | BTW, question you might know off your head, with LBaaS and third party drivers, the requests don't hit rabbit unless the driver is using it right? | 16:30 |
| xgerman | that is my understanding as well | 16:31 |
| johnsom | Yeah, ok. I was clarifying for someone. Still recovering/catching up from the two weeks on the road and head cold. | 16:32 |
| nmagnezi_ | blogan, xgerman, hi :) having some Octavia related issues here. I'm trying to set-up a multi-node devstack with Octavia and when I ./stack my api node (with: https://paste.fedoraproject.org/314942/45382587/) i get some db related errors, which looks related to Octavia: https://paste.fedoraproject.org/314941/38258621/ | 16:32 |
| nmagnezi_ | blogan, xgerman, have you ever encountered something similar? | 16:33 |
| johnsom | nmagnezi_ I wonder if this recent patch doesn't have an issue: https://review.openstack.org/#/c/267838/ | 16:34 |
| johnsom | nmagnezi_ Yeah, I'm pretty sure that patch is bad based on your logs | 16:35 |
| *** woodster_ has joined #openstack-lbaas | 16:36 | |
| nmagnezi_ | johnsom, I can try to test the same thing without that change | 16:36 |
| johnsom | Ok. Thanks. I can see that the patch doesn't take into account the foreign key. | 16:36 |
| nmagnezi_ | johnsom, how come it breaks only for me? | 16:37 |
| nmagnezi_ | johnsom, shouldn't it break the gates? | 16:37 |
| johnsom | That is the interesting question. It should have failed the scenario test gate | 16:37 |
| nmagnezi_ | johnsom, I will generate a reverted patch, just so I can cherry-pick. | 16:39 |
| nmagnezi_ | johnsom, don't merge it. is that ok? | 16:39 |
| johnsom | Sure, just mark it WIP and put do-not-merge in the commit message | 16:39 |
| nmagnezi_ | aye | 16:39 |
| openstackgerrit | Nir Magnezi proposed openstack/octavia: (WIP) Revert "Make pool.lb_algorithm field larger" https://review.openstack.org/272630 | 16:40 |
| johnsom | I will also get my devstack fired up and try to reproduce it | 16:41 |
| nmagnezi_ | johnsom, ack. you can see my local.conf just in case I did something wrong :-) | 16:41 |
| nmagnezi_ | johnsom, going offline for ~1 hour | 16:42 |
| johnsom | Ok | 16:42 |
| *** _ducttape_ has quit IRC | 16:44 | |
| *** ducttape_ has joined #openstack-lbaas | 16:45 | |
| *** _cjones_ has joined #openstack-lbaas | 16:46 | |
| *** nmagnezi_ has quit IRC | 16:47 | |
| *** _cjones_ has quit IRC | 16:48 | |
| *** _cjones_ has joined #openstack-lbaas | 16:48 | |
| *** neelashah has quit IRC | 16:54 | |
| *** neelashah has joined #openstack-lbaas | 16:54 | |
| *** Aish has joined #openstack-lbaas | 16:55 | |
| *** neelashah has quit IRC | 16:59 | |
| blogan | johnsom: no rabbit unless the drivers are using it correct...well neutron will send out events to rpc when api requests come in | 16:59 |
| blogan | sorry to rabbit | 16:59 |
| *** neelashah has joined #openstack-lbaas | 16:59 | |
| johnsom | What type of events? | 17:00 |
| blogan | like create resource events | 17:01 |
| blogan | the lbaas code doesn't do it, the overarching neutron code does it | 17:02 |
| blogan | basically whenver a create update adn delete method come in it sends out an rpc notifier | 17:02 |
| *** fnaval_ has joined #openstack-lbaas | 17:02 | |
| johnsom | Ok, cool. Thanks! | 17:02 |
| johnsom | Did I get this cold from you? | 17:03 |
| blogan | mine was only bad one day, and just annoying 2 other days | 17:03 |
| johnsom | So, yes.... | 17:03 |
| blogan | no! | 17:03 |
| blogan | i wasnt sick that week | 17:03 |
| johnsom | I wasn't until last Saturday | 17:04 |
| johnsom | Very well could be Seattle sharing too | 17:04 |
| *** neelashah1 has joined #openstack-lbaas | 17:04 | |
| *** neelashah has quit IRC | 17:04 | |
| *** diogogmt_ has joined #openstack-lbaas | 17:04 | |
| blogan | it might be that you got used to not so fresh air, and when you got back to oregon your body couldn't handle it | 17:05 |
| johnsom | Ha | 17:05 |
| *** prabampm1 has joined #openstack-lbaas | 17:06 | |
| *** minwang2 has joined #openstack-lbaas | 17:06 | |
| *** prabampm has quit IRC | 17:07 | |
| *** fnaval has quit IRC | 17:07 | |
| *** diogogmt has quit IRC | 17:07 | |
| *** diogogmt_ is now known as diogogmt | 17:07 | |
| *** neelashah1 has quit IRC | 17:08 | |
| *** neelashah has joined #openstack-lbaas | 17:09 | |
| openstackgerrit | Merged openstack/neutron-lbaas: Updated from global requirements https://review.openstack.org/271648 | 17:13 |
| *** neelashah has quit IRC | 17:13 | |
| *** neelashah1 has joined #openstack-lbaas | 17:13 | |
| *** neelashah has joined #openstack-lbaas | 17:17 | |
| *** openstackgerrit has quit IRC | 17:17 | |
| *** minwang2 has quit IRC | 17:17 | |
| *** openstackgerrit has joined #openstack-lbaas | 17:18 | |
| *** neelashah1 has quit IRC | 17:18 | |
| *** neelashah has quit IRC | 17:22 | |
| *** neelashah1 has joined #openstack-lbaas | 17:22 | |
| *** neelashah has joined #openstack-lbaas | 17:25 | |
| *** neelashah1 has quit IRC | 17:26 | |
| *** neelashah1 has joined #openstack-lbaas | 17:29 | |
| *** neelashah has quit IRC | 17:30 | |
| *** neelashah has joined #openstack-lbaas | 17:32 | |
| *** neelashah1 has quit IRC | 17:33 | |
| *** neelashah1 has joined #openstack-lbaas | 17:35 | |
| *** neelashah has quit IRC | 17:36 | |
| *** neelashah1 has quit IRC | 17:40 | |
| *** neelashah has joined #openstack-lbaas | 17:40 | |
| a2hill | quick win? https://review.openstack.org/#/c/267220/ Then I can worry about a backport | 17:42 |
| *** nmagnezi_ has joined #openstack-lbaas | 17:44 | |
| *** neelashah1 has joined #openstack-lbaas | 17:44 | |
| *** diogogmt has quit IRC | 17:44 | |
| *** neelashah has quit IRC | 17:44 | |
| nmagnezi_ | johnsom, ping re: devstack issue we discussed. re-stacking now. will update you soon | 17:44 |
| johnsom | nmagnezi_ Yeah, I did a reclone restack. Hitting some issues, but still working on i | 17:45 |
| johnsom | it | 17:45 |
| *** neelashah has joined #openstack-lbaas | 17:47 | |
| *** neelashah1 has quit IRC | 17:49 | |
| *** neelashah1 has joined #openstack-lbaas | 17:51 | |
| *** hockeynut is now known as hockeynut_otr | 17:51 | |
| *** hockeynut_otr is now known as hockeynut | 17:52 | |
| *** neelashah has quit IRC | 17:52 | |
| *** bana_k has quit IRC | 17:53 | |
| *** Alex_Stef has joined #openstack-lbaas | 17:54 | |
| nmagnezi_ | johnsom, okay at this point it creates the amphora image, So I guess we're good now | 17:54 |
| johnsom | By reverting? | 17:55 |
| nmagnezi_ | johnsom, anyways I'll let you know when it successfully stacked. | 17:55 |
| nmagnezi_ | yes | 17:55 |
| johnsom | Ok | 17:55 |
| *** neelashah1 has quit IRC | 17:55 | |
| *** neelashah has joined #openstack-lbaas | 17:56 | |
| *** diogogmt has joined #openstack-lbaas | 17:58 | |
| *** minwang2 has joined #openstack-lbaas | 17:58 | |
| *** neelashah1 has joined #openstack-lbaas | 17:59 | |
| *** hockeynut_afk has joined #openstack-lbaas | 18:00 | |
| *** neelashah has quit IRC | 18:00 | |
| *** jschwarz has joined #openstack-lbaas | 18:01 | |
| *** neelashah1 has quit IRC | 18:03 | |
| ajmiller | a2hill +A ^^ | 18:03 |
| a2hill | ty | 18:03 |
| ajmiller | On a somewhat related note, https://review.openstack.org/#/c/271544/ | 18:04 |
| *** neelashah has joined #openstack-lbaas | 18:05 | |
| a2hill | wanted rm_work to take a peak at that one, it looks good to me also | 18:05 |
| *** TrevorV has quit IRC | 18:05 | |
| *** hockeynut_afk has quit IRC | 18:05 | |
| a2hill | rm_work: If you get a sec please take a gander at the patch ajmiller linked. | 18:05 |
| *** TrevorV has joined #openstack-lbaas | 18:06 | |
| *** hockeynut_otr has joined #openstack-lbaas | 18:06 | |
| ajmiller | OK, cool | 18:06 |
| *** Alex_Stef has quit IRC | 18:09 | |
| *** neelashah has quit IRC | 18:09 | |
| *** neelashah has joined #openstack-lbaas | 18:10 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/octavia: Updated from global requirements https://review.openstack.org/270948 | 18:12 |
| rm_work | ok | 18:12 |
| *** neelashah1 has joined #openstack-lbaas | 18:13 | |
| *** neelashah has quit IRC | 18:14 | |
| *** Aish has quit IRC | 18:16 | |
| *** bana_k has joined #openstack-lbaas | 18:16 | |
| *** neelashah has joined #openstack-lbaas | 18:17 | |
| *** madhu_ak has joined #openstack-lbaas | 18:18 | |
| *** neelashah1 has quit IRC | 18:18 | |
| *** harlowja has joined #openstack-lbaas | 18:18 | |
| *** mgarza has joined #openstack-lbaas | 18:19 | |
| nmagnezi_ | johnsom, just to keep you posted, it failed to create the amphora image, not sure if it's related: https://paste.fedoraproject.org/314986/83242014/ | 18:21 |
| *** hockeynut_otr has quit IRC | 18:28 | |
| *** Aish has joined #openstack-lbaas | 18:29 | |
| johnsom | nmagnezi_ My image just built. Those errors don't seem related to the DB issue | 18:29 |
| nmagnezi_ | johnsom, tyup | 18:29 |
| nmagnezi_ | johnsom, yup | 18:29 |
| bana_k | johnsom nmagnezi_ this is the fix | 18:32 |
| bana_k | https://review.openstack.org/#/c/271178/ | 18:32 |
| bana_k | for that prob | 18:32 |
| johnsom | nmagnezi_ Ok, not sure what is up. I can't reproduce the problem with a fresh/stock devstack | 18:32 |
| bana_k | johnsom that happens in the centos env | 18:33 |
| johnsom | Ah | 18:33 |
| nmagnezi_ | interesting. | 18:33 |
| nmagnezi_ | i'm using fedora23 for my devstack here.. | 18:33 |
| johnsom | I use Ubuntu 14.04.3 | 18:33 |
| bana_k | yea https://bugs.launchpad.net/octavia/+bug/1536876 | 18:34 |
| openstack | Launchpad bug 1536876 in octavia "Database migration 'make pool.lb_algorithm larger' fails on DevStack" [Undecided,In progress] - Assigned to Banashankar (bkalebe) | 18:34 |
| bana_k | that was happening in fedora also, according to the bug description I faced the same on cetos | 18:35 |
| bana_k | not happening in Ubuntu | 18:35 |
| rm_work | my question on that was | 18:36 |
| rm_work | can you *patch* an existing migration? | 18:36 |
| rm_work | is that kosher? | 18:36 |
| rm_work | I feel like ideally we should do that, I just didn't know if it was accepted practice | 18:36 |
| bana_k | yea was thinking the same but, I dont know how can we do that, as the migration script ll fail when it hits that patch. | 18:37 |
| rm_work | like, fail & exit? | 18:39 |
| bana_k | yea | 18:39 |
| rm_work | it wouldn't just fail that patch and go to the next? :( | 18:39 |
| rm_work | yeah k | 18:39 |
| rm_work | guess this is necessary then | 18:39 |
| bana_k | I think so. | 18:39 |
| rm_work | so it fails everywhere EXCEPT Ubuntu? | 18:39 |
| bana_k | yes on Centos and fedora | 18:39 |
| rm_work | so weird | 18:40 |
| rm_work | Yeah I don't have any test-beds other than ubuntu 14.04 | 18:40 |
| nmagnezi_ | johnsom, still fails to create the amphora image under f23. non-related, yet strange. | 18:40 |
| bana_k | oh | 18:40 |
| *** neelashah has quit IRC | 18:42 | |
| bana_k | nmagnezi_ whats the error ? | 18:45 |
| openstackgerrit | Merged openstack/neutron-lbaas: Encode certificate data for processing https://review.openstack.org/267220 | 18:45 |
| nmagnezi_ | bana_k, i'm running the image creation script now manually (not via ./stack.sh) I'll share the results as if finish to run | 18:45 |
| bana_k | ok | 18:46 |
| nmagnezi_ | bana_k, until it finish, i had additional problems with that script: https://bugs.launchpad.net/octavia/+bug/1531092 | 18:46 |
| openstack | Launchpad bug 1531092 in octavia "The diskimage-create script fails to build an amphora image for centos and fedora" [Medium,New] | 18:46 |
| *** jschwarz has quit IRC | 18:48 | |
| nmagnezi_ | bana_k, johnsom, http://pastebin.com/ZfBF7fsQ | 18:54 |
| *** neelashah has joined #openstack-lbaas | 18:54 | |
| *** neelashah has quit IRC | 18:59 | |
| *** neelashah has joined #openstack-lbaas | 19:00 | |
| bana_k | nmagnezi_ did you try this after running ./clean.sh | 19:04 |
| bana_k | ? | 19:04 |
| nmagnezi_ | bana_k, clean.sh? where is that script? | 19:09 |
| bana_k | in devstack | 19:09 |
| openstackgerrit | Merged openstack/neutron-lbaas: Adding "region and endpoint_type" parameters to barbican_acl.py https://review.openstack.org/271544 | 19:12 |
| TrevorV | sbalukoff, you around? | 19:17 |
| blogan | bana_k: why do you have to drop the contraint and then add it back? is that mysql just now allowing a fk column to be modified? | 19:22 |
| *** diogogmt has quit IRC | 19:35 | |
| johnsom | blogan They are seeing this on RedHat tribe systems: https://paste.fedoraproject.org/314941/38258621/ | 19:46 |
| *** diogogmt has joined #openstack-lbaas | 19:49 | |
| *** rcernin has quit IRC | 19:58 | |
| *** neelashah has quit IRC | 20:00 | |
| a2hill | Would this still be an issue too johnsom ? | 20:00 |
| a2hill | https://bugs.launchpad.net/octavia/+bug/1531092 | 20:00 |
| openstack | Launchpad bug 1531092 in octavia "The diskimage-create script fails to build an amphora image for centos and fedora" [Medium,New] | 20:00 |
| johnsom | Yes. I think it was mentioned above | 20:01 |
| a2hill | Oh | 20:01 |
| *** neelashah has joined #openstack-lbaas | 20:02 | |
| blogan | johnsom: yeah i saw that, just wondering why that is the case just for that | 20:09 |
| blogan | johnsom: if its a redhat tribe thing then thats over my head | 20:09 |
| johnsom | Yeah, not sure. I suspect it's version issue. I just tried it out stock on my ubuntu devstack and had no issue | 20:10 |
| blogan | i was thinking maybe bc the pools table had previous data and running hte migration on that could maybe cause that but that doesn't make sense either, so ill just accept that its a redhat tribe thing | 20:10 |
| johnsom | Yeah, I haven't done anything with the redhat distro stuff since the original image builder. But I think we have some folks that can figure it out | 20:11 |
| sbalukoff | TrevorV: I am around now. | 20:13 |
| *** neelashah has quit IRC | 20:13 | |
| TrevorV | I've just been digging through your review piece by piece and had a question earlier, though I'm afraid I forgot it o_0 | 20:13 |
| sbalukoff | D'oh! | 20:13 |
| TrevorV | I do have a few comments. Some questions, some distinctions to be made, etc etc. They'll be posted soon | 20:14 |
| TrevorV | Its A LOT to dig through :P | 20:14 |
| sbalukoff | Yeah, I know. | 20:16 |
| sbalukoff | The L7 stuff is going to be even bigger, though split up in to hopefully more digestible chunks. | 20:17 |
| TrevorV | Granted its taking me longer than I want it to to consume... BUT, I'm almost done :D | 20:17 |
| sbalukoff | Yay! | 20:18 |
| sbalukoff | Thank you very much for doing this. | 20:18 |
| TrevorV | Its on our "important things" list so most of us from Racks has been watching it relatively closely :D | 20:26 |
| *** madhu_ak has quit IRC | 20:26 | |
| sbalukoff | Yay! | 20:27 |
| openstackgerrit | Franklin Naval proposed openstack/octavia: Preparing for tempest testing https://review.openstack.org/172199 | 20:30 |
| *** rcernin has joined #openstack-lbaas | 20:33 | |
| openstackgerrit | Franklin Naval proposed openstack/octavia: Preparing for Tempest testing https://review.openstack.org/172199 | 20:35 |
| fnaval_ | please take a look when anyone has as chance ^^ | 20:36 |
| *** harlowja_ has joined #openstack-lbaas | 20:36 | |
| *** fnaval_ is now known as fnava | 20:36 | |
| *** fnava is now known as fnaval | 20:36 | |
| *** harlowja has quit IRC | 20:37 | |
| sbalukoff | Wow, 1200 lines in a conf file? | 20:38 |
| *** madhu_ak has joined #openstack-lbaas | 20:38 | |
| fnaval | hmm... yeah, i guess that could be trimmed | 20:40 |
| *** neelashah has joined #openstack-lbaas | 20:40 | |
| sbalukoff | Well.. | 20:40 |
| fnaval | or introduce it when it's ready | 20:40 |
| fnaval | with actual tests | 20:41 |
| sbalukoff | Actually, I have a question about that: Clearly this conf file is pulled from an example file, the original of which is probably in the tempest project somewhere right? | 20:41 |
| fnaval | I would think so as well | 20:41 |
| TrevorV | sbalukoff, bad-ish news... I made different "replies" to the page. One with "what about these" and one with "here is every file that needs a copyright line from Bluebox" | 20:42 |
| sbalukoff | Would it be better to just make a note at the top pointing the user in the direction of where they can find *all* the knobs they can tweak, and only leave the uncommented stuff (and appropriate comments for the same) in this file | 20:42 |
| sbalukoff | ? | 20:42 |
| openstackgerrit | Michael Johnson proposed openstack/octavia: Make sample web server close connections https://review.openstack.org/272722 | 20:42 |
| fnaval | sbalukoff: good idea - i'll do just that | 20:42 |
| sbalukoff | TrevorV: Er... that's bad news? I mean, I can still find both replies, right? | 20:42 |
| sbalukoff | fnaval: Ok, I'll leave a comment and -1 for now. ;) | 20:43 |
| sbalukoff | You know, because I love you. | 20:43 |
| TrevorV | Yeah, I just feel like its spam ha ha | 20:43 |
| fnaval | yep. =) thanks for the review. | 20:43 |
| openstackgerrit | Michael Johnson proposed openstack/neutron-lbaas: Make sample web server close connections https://review.openstack.org/272725 | 20:43 |
| *** harlowja has joined #openstack-lbaas | 20:45 | |
| *** harlowja_ has quit IRC | 20:45 | |
| sbalukoff | TrevorV: Thanks for the review. I'll probably have a new version addressing most (if not all) your notes later this afternoon. | 20:46 |
| sbalukoff | Gonna spend some time on fnaval's patch for now. | 20:46 |
| TrevorV | sbalukoff, sounds great. | 20:48 |
| TrevorV | I'm going to pay some attention to my reviews a bit now anyway, and then call it a day :D | 20:48 |
| sbalukoff | TrevorV: On the copyright notice stuff. I honestly am not sure what IBM's policy is (which I'm sure they have... buried somewhere in the intranet sites somewhere). However, you're probably right in that it's probably safest to just spam it pretty much everywhere I've made a change. :P | 20:48 |
| TrevorV | that's what I'd go with. | 20:50 |
| *** minwang2 has quit IRC | 20:50 | |
| sbalukoff | So, in a tempest test, the "client" is actually just talking to a REST service directly, right? (As in, we aren't typically pulling in a CLI or anything else one might refer to as a "client" in other contexts, correct?) | 20:53 |
| blogan | sbalukoff: yes its using requests basically | 20:54 |
| sbalukoff | Ok. | 20:54 |
| blogan | sbalukoff: don't want the actual python clients to do their own validation of stuff since you're testing the api | 20:54 |
| sbalukoff | Yep. | 20:55 |
| sbalukoff | Make sense. | 20:55 |
| blogan | though it would probably be smart to have the tempest clients and cli clients use teh same raw clients, if that makes sense | 20:55 |
| sbalukoff | (Especially given there is no direct CLI for Octavia yet.) | 20:55 |
| *** nmagnezi_ has quit IRC | 20:55 | |
| blogan | anyone care to research the feasability of allowing the v1 and v2 neutron-lbaas plugins to work at the same time? | 20:56 |
| johnsom | You thinking of doing flavors? | 20:56 |
| sbalukoff | blogan: Code re-use is generally a good idea. Though with tests you have to be careful specifically *not* to rely too much on code that's actually part of the production code base. (eg. if altering a method somewhere in the production code base breaks things, a test which uses that could potentially not detect the bug such an alteration introduces.) | 20:57 |
| sbalukoff | blogan: No. We should kill v1. | 20:57 |
| sbalukoff | Or it will never die. | 20:57 |
| sbalukoff | ;) | 20:57 |
| blogan | johnsom: well i doubt flavors would work, what i mean is the v1 api and v2 api work at the same time | 20:57 |
| *** minwang2 has joined #openstack-lbaas | 20:57 | |
| johnsom | It's the cold talking, right. Ummm, no. We also think v1 should die | 20:58 |
| blogan | sbalukoff: bc there seem to be more private clouds running v1 than we expected, and if they can't have v1 running at the same time as v2 it makes it much more painful for them | 20:58 |
| blogan | meaning, to migration from v1 to v2 | 20:59 |
| blogan | which is what they want to do | 20:59 |
| blogan | but with the restriction its really painful | 20:59 |
| sbalukoff | blogan: You have paying customers asking for this? | 20:59 |
| sbalukoff | Because... I have been doing my best to tell our internal customers to go to hell when they ask for stupid shit. ;) | 21:00 |
| blogan | sbalukoff: nope, openstack, there's a thread on the ML | 21:00 |
| openstackgerrit | Merged openstack/octavia: Updated from global requirements https://review.openstack.org/270948 | 21:00 |
| sbalukoff | Blah. | 21:00 |
| sbalukoff | See... I just don't get it, I guess: v1 is such a useless, incapable piece of shit that altering your code to talk v2 instead should be a breeze. Especially once those heat templates for deploying lbaasv2 services land. | 21:01 |
| openstackgerrit | Franklin Naval proposed openstack/octavia: Preparing for Tempest testing https://review.openstack.org/172199 | 21:02 |
| blogan | sbalukoff: its not about altering the code, its about deployers who want to move from v1 to v2, the opportunity to move to v2 with less pain instead of just wiping all their v1 lbs out | 21:02 |
| sbalukoff | But making things work on the OpenStack side to "ease a transition" (read: Put it off as long as humanly possible because we haven't budgeted properly for maintaining legacy systems and nobody wants to touch it anymore, even if it's trivial)... Is like heading off in a northerly direction from New York to try to get to Washington... eventually, you can probably get there... but you're going to have a bad time of it g | 21:03 |
| sbalukoff | oing that way. | 21:03 |
| sbalukoff | Anyway... if you have people interested in making that happen... well, more power to you. Right now, my priority is to make L7 happen by Mitaka. :P | 21:05 |
| sbalukoff | blogan: Wait, are you talking more about a tool to help with a v1 -> v2 transition? | 21:06 |
| sbalukoff | Because *that* might be much more doable. | 21:06 |
| a2hill | xgerman: https://review.openstack.org/#/c/272735/ https://review.openstack.org/#/c/272706/1 | 21:06 |
| blogan | if making that transitione easier isn't too much work then to me that makes it worthwhile, if its a huge code work then i'd ahve to be against it though | 21:06 |
| blogan | sicne it is deprecated | 21:06 |
| sbalukoff | Something which does a dump of the v1 database and then parses that to build v2-equivalent configurations out of it... | 21:06 |
| blogan | sbalukoff: no v1 lbs can be spun up separate from v2 lbs | 21:07 |
| sbalukoff | blogan: See, that that's what I don't support. | 21:07 |
| sbalukoff | I'd be more for it if at the end of the day v1 was definitively dead. | 21:07 |
| xgerman | a2hill ack | 21:07 |
| sbalukoff | I don't like keeping that piece of shit plugged into life support. | 21:07 |
| blogan | sbalukoff: well i would hope the operators would put a stop to new ones being spun up on v1, and require just v2 for that | 21:07 |
| * xgerman reading scrollback | 21:07 | |
| blogan | but if the just move from v1 to v2 right now, they will lose all the v1 lbs | 21:08 |
| sbalukoff | blogan: Not if we were to write a tool to create v2 "equivalents" of the v1 load balancers. | 21:08 |
| sbalukoff | (More or less, anyway.) | 21:08 |
| a2hill | Is someone asking for it? | 21:08 |
| sbalukoff | a2hill: On the ML, apparently. | 21:08 |
| a2hill | oh | 21:09 |
| blogan | sbalukoff: i think thats a bit mroe complicated than it sounds | 21:09 |
| sbalukoff | blogan: probably. But I'll bet someone could write a tool that captures the 90% use case. | 21:09 |
| blogan | http://lists.openstack.org/pipermail/openstack-dev/2016-January/085022.html | 21:09 |
| blogan | that thread | 21:09 |
| sbalukoff | Mind you, I'm *not* volunteering to write said tool, of course. | 21:10 |
| blogan | sbalukoff: would you port v1 agent lbs to v2 octavia? | 21:10 |
| xgerman | blogan, sbalukoff HP has a tool to migrate… so if they only would buy our stuff ;-) | 21:10 |
| sbalukoff | This is one area where I think those asking for it ought to do it, eh... | 21:10 |
| blogan | xgerman: lol | 21:10 |
| blogan | xgerman: how does that work? | 21:10 |
| blogan | xgerman: v1 agent to v2 octavia? | 21:10 |
| a2hill | Oh hayden | 21:10 |
| xgerman | you shut down V1, run the tool, fire up V2 | 21:11 |
| sbalukoff | xgerman: Haha! Indeed! you should totally suggest that on the ML. XD | 21:11 |
| blogan | xgerman: what does the tool do! | 21:11 |
| xgerman | it does a DB migration and fixes keys, makes them up | 21:11 |
| xgerman | so once the V2 database is populated you start it and the lbaas v2 namespace comes up | 21:12 |
| openstackgerrit | Franklin Naval proposed openstack/octavia: Preparing for Tempest testing https://review.openstack.org/172199 | 21:12 |
| blogan | xgerman: okay so its agent to agent | 21:12 |
| xgerman | yep | 21:12 |
| blogan | v1 agent driver to v2 agent driver | 21:12 |
| sbalukoff | Yeah, that's all I'd do. | 21:12 |
| xgerman | yeah | 21:12 |
| sbalukoff | Getting them to adopt Octavia at the same time would be nice, but probably much less feasible. | 21:12 |
| xgerman | Octavia is self migrate at the moment on our side | 21:13 |
| blogan | and there were issues with both v1 and v2 agents running at the same time | 21:13 |
| sbalukoff | And going from v2 + agent to v2 + octavia later should, in theory, have minimal impact on the tenant. | 21:13 |
| xgerman | blogan, hence you stop v1 agent first | 21:13 |
| blogan | xgerman: i mean if we allowed v1 and v2 to run at the same time | 21:14 |
| sbalukoff | blogan: And that's what I'm most against here. :) | 21:14 |
| xgerman | yeah, I think that train left the station | 21:14 |
| sbalukoff | Though I've probably made that abundantly clear. :) | 21:14 |
| xgerman | we are removing v1 in like N? | 21:14 |
| sbalukoff | I really, really hope so. | 21:14 |
| blogan | yes its deprecated until N | 21:15 |
| blogan | has the agent driver even been getting love? | 21:20 |
| blogan | sbalukoff: like will it support L7? | 21:20 |
| *** neelashah has quit IRC | 21:21 | |
| sbalukoff | blogan: It's actually pretty simple to add support for L7 to the agent driver. It's mostly about altering jinja templates, which will happen in essentially the same way for Octavia. So i'm not to worried about bringing the agent driver up to speed in that regard once we have it in Octavia. (And I do imagine there will be demand for it in the agent driver.) | 21:22 |
| blogan | sbalukoff: true and really going from v1 to v2 there woudln't be an expectation of L7 right away | 21:22 |
| sbalukoff | Again, the biggest hurdle on both fronts is the shared pools support. Because that touches everything. In contract, L7 is almost purely an extension of existing functionality. | 21:22 |
| sbalukoff | blogan: Yep! v1 has no concept of v2. And even v2 with L7... you don't actually have to use L7 if you don't want. You can ignore it and use your v2 service like you used to use your v1 service. | 21:23 |
| sbalukoff | In constrast. | 21:23 |
| sbalukoff | :P | 21:23 |
| blogan | sbalukoff: so basically what we're saying isn't an easy upgrade path from v1 agent driver to v2 octavia, but there is a simple one from v1 agent driver to v2 agent driver | 21:24 |
| blogan | and would be even simpler if xgerman pushed that migration script upstream | 21:24 |
| * blogan whips xgerman into action | 21:24 | |
| sbalukoff | blogan: Actually, I expect the upgrade from the tenant's perspective to be just easy either way. For an operator, going from agent to octavia takes some planning. | 21:24 |
| sbalukoff | blogan: +1 | 21:25 |
| xgerman | blogan it is proprietary - we need some way to add value ;-) | 21:25 |
| xgerman | but I will check with my overlords | 21:25 |
| sbalukoff | xgerman: Wait, HP adds value? O.O | 21:26 |
| blogan | xgerman: i'm going to proprietarize your feace! | 21:26 |
| sbalukoff | ;) | 21:26 |
| blogan | face! (feace looks too much like feces) | 21:26 |
| johnsom | You all should just adopt Helion | 21:26 |
| sbalukoff | IBM adds obfuscation! :D | 21:26 |
| johnsom | (had to throw it out there) | 21:26 |
| sbalukoff | :) | 21:26 |
| blogan | isn't Helion just a wrapper around rackspace private cloud? | 21:27 |
| xgerman | RAX will support Helion rather sooner than later ;-) | 21:27 |
| blogan | xgerman: we don't support huge coporations with their own cloud! | 21:27 |
| * blogan has been asleep for a year | 21:27 | |
| xgerman | that would be us | 21:27 |
| sbalukoff | Haha | 21:27 |
| *** rcernin has quit IRC | 21:28 | |
| johnsom | https://www.rackspace.com/en-us/microsoft/managed-azure-cloud | 21:28 |
| johnsom | All I have to say.... | 21:28 |
| sbalukoff | XD | 21:28 |
| blogan | johnsom: trojan horse will be known as a racker horse in the future | 21:28 |
| blogan | its all a ruse | 21:28 |
| johnsom | "These aren't the windows servers you were looking for...." | 21:30 |
| blogan | xgerman, johnsom: would yall mind responding to that thread explaining how yall migrate from v1 to v2? (high level of course, no proprietary info) | 21:31 |
| xgerman | k, will do | 21:31 |
| xgerman | blogan is it that unusable thread? | 21:35 |
| xgerman | ansible | 21:35 |
| blogan | lol | 21:35 |
| blogan | yes | 21:35 |
| blogan | xgerman: http://lists.openstack.org/pipermail/openstack-dev/2016-January/085034.html | 21:36 |
| xgerman | found it | 21:36 |
| blogan | thanks | 21:36 |
| *** pai15 has joined #openstack-lbaas | 21:37 | |
| pai15 | Hello, can someone take a peak at https://bugs.launchpad.net/neutron/+bug/1481936 ? do we need to add lbaas v1 support? | 21:38 |
| openstack | Launchpad bug 1481936 in neutron "LBaaS "Configure neutron agents" docs page needs to be updated" [Undecided,New] - Assigned to P Ingle (pingle) | 21:39 |
| blogan | pai15: v1 is deprecated so I'd say only v2, but i honestly dont knwo the official openstack answer on that | 21:40 |
| xgerman | well, we like to discourage the use of v1 — so it not being documented might help there :-) | 21:40 |
| pai15 | in that case, what do we do with the open ticket? | 21:40 |
| johnsom | I opened that a long time ago. Maybe mark it wont-fix due to deprecated. | 21:41 |
| sbalukoff | mark as closed "wont fix" | 21:41 |
| sbalukoff | :) | 21:41 |
| johnsom | Or I can | 21:41 |
| pai15 | johnsom: thanks - I wil unassign myself with a comment | 21:43 |
| johnsom | Ok, after you are done I will mark wont-fix | 21:43 |
| pai15 | done | 21:44 |
| pai15 | thanks | 21:44 |
| *** neelashah has joined #openstack-lbaas | 21:47 | |
| johnsom | done as well | 21:49 |
| blogan | xgerman: nooo, v1 and v2 dont share the same tables | 21:50 |
| xgerman | believe me they do | 21:50 |
| blogan | xgerman: maybe for the agent some do | 21:51 |
| pai15 | but is lbaas v1 going to be supported going forward or removed/cleaned up? | 21:51 |
| xgerman | removed | 21:51 |
| mhayden | blogan: is it possible to get a v6 address automatically added in the amphora VM with SLAAC? i haven't tested it yet, but it's in my list of to-do's | 21:54 |
| johnsom | Yes | 21:55 |
| pai15 | asking in doc channel if there is a recommended way to document deprecated features and policies around it | 21:56 |
| johnsom | pai15 we have already marked them deprecated in the docs | 21:57 |
| johnsom | At least the API sections | 21:57 |
| pai15 | I mean in installation/config guides | 21:57 |
| rm_work | mhayden: "amphora VM" is redundant :P | 21:58 |
| rm_work | ... unless you're trying to differentiate between different amphora types, I guess | 21:58 |
| *** ducttape_ has quit IRC | 21:58 | |
| * mhayden is very very new to this topic :) | 21:58 | |
| mhayden | i was lucky enough to get a brain dump from blogan this morning on a few things | 21:59 |
| xgerman | well, did he initiate you with the towing? | 21:59 |
| rm_work | yeah I would have stopped by for that if i were in the office today | 22:00 |
| johnsom | mhayden it will be on the interface, but haproxy won't be listening on it. | 22:02 |
| *** ducttape_ has joined #openstack-lbaas | 22:04 | |
| *** TrevorV has quit IRC | 22:09 | |
| ajmiller | doug-fish ping | 22:14 |
| doug-fish | ajmiller: hi | 22:21 |
| ajmiller | Hi. I'm looking at the dashboard patches. | 22:21 |
| doug-fish | cool | 22:22 |
| ajmiller | I have an annoyance that isn't specific to these panels. It is pervasive throughout horizon: error messages are basically devioid of useful content. Is there anything we can do to make it better? | 22:22 |
| ajmiller | For example, if creating a loadbalancer fails, we get a generic message, without any reference to the actual error, not even HTTP return code. | 22:23 |
| *** davidlenwell has quit IRC | 22:24 | |
| ajmiller | (that particular patch has already merged, but it would be great if we could improve that) | 22:24 |
| doug-fish | yeah - I hear what you are saying - does the API return anything specific after a create lb fails? | 22:24 |
| doug-fish | that's usually where things fall down | 22:24 |
| blogan | mhayden: i've not tested that out yet, and honestly i hadn't even thought about it, but johnsom is probably right | 22:24 |
| doug-fish | we can make some improvements on messages by interpreting the HTTP error codes a bit | 22:25 |
| johnsom | Yeah, I have my local devstack on an ipv6 enabled network | 22:25 |
| *** davidlenwell has joined #openstack-lbaas | 22:25 | |
| ajmiller | The API isn't often very helpful, but from what i can tell the UI panels just put up a hard-coded fixed message. | 22:26 |
| mhayden | okay | 22:27 |
| ajmiller | doug-fish It would be great if we could report the HTTP code and whatever message text we get back from the API | 22:28 |
| *** rtheis has quit IRC | 22:28 | |
| doug-fish | ajmiller: I think in places where we've done better the model is closer to mapping a message to each HTTP code to suggest what really happened ... | 22:32 |
| doug-fish | "HTTP 409" might not be as helpful as "There was a conflict creating your thing, try a different name" or something like that | 22:33 |
| ajmiller | Yes, I agree. The panels do a pretty good job of filtering out bad input, so it heads off a lot of problems. One of the problems I ran into today was that my devstack wasn't set up (octavia wasn't fully running). The API was returning 500 internal server error. Knowing that would have pointed me in the right direction sooner. | 22:34 |
| doug-fish | that makes sense | 22:35 |
| ajmiller | In any case, I'm going through the listener and member tabs (I just grabbed Justin's most recent update to https://review.openstack.org/#/c/256085/8) | 22:38 |
| doug-fish | I've had the mindset that it would be helpful to quickly get code working for the main paths and there can be iterative improvements from multiple people to fix key issue, like bad error messages ... | 22:39 |
| doug-fish | ajmiller: I think that code works well, but it reminded me we need a couple of more gate jobs to lint + test javascript! | 22:39 |
| ajmiller | Agreed. I think the listener tab is good to go. I will go through and look specifically for error messages, keep track of them and file a bug. | 22:40 |
| ajmiller | Regarding gate jobs, do you have some ready to go? | 22:40 |
| doug-fish | cool, I was thinking a bug might be a good way to track | 22:40 |
| doug-fish | I was just exploring how those work | 22:40 |
| ajmiller | OK. | 22:41 |
| doug-fish | As long as the project has two magically named jobs I think all of the other pieces are in place | 22:41 |
| fnaval | review please; thanks in advance: https://review.openstack.org/#/c/172199/ | 22:53 |
| doug-fish | I think this will make the javascript unit tests/lint checks gating for the dashboard: https://review.openstack.org/272771 | 22:56 |
| *** doug-fish has quit IRC | 23:00 | |
| *** neelashah1 has joined #openstack-lbaas | 23:08 | |
| *** neelashah has quit IRC | 23:12 | |
| ajmiller | doug-fish https://review.openstack.org/#/c/254299/ merged | 23:21 |
| *** bdrich has joined #openstack-lbaas | 23:36 | |
| *** bdrich_ has quit IRC | 23:39 | |
| *** neelashah1 has quit IRC | 23:39 | |
| openstackgerrit | min wang proposed openstack/octavia: Add anti-affinity server group feature for octavia https://review.openstack.org/272344 | 23:41 |
| *** bdrich has quit IRC | 23:47 | |
| *** yamamoto_ has joined #openstack-lbaas | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!