| *** rtjure has quit IRC | 00:00 | |
| *** rtjure has joined #openstack-lbaas | 00:03 | |
| *** yamamoto has joined #openstack-lbaas | 00:14 | |
| *** yamamoto has quit IRC | 00:20 | |
| *** slaweq has joined #openstack-lbaas | 00:30 | |
| *** slaweq has quit IRC | 00:35 | |
| *** fnaval has quit IRC | 00:40 | |
| *** fnaval has joined #openstack-lbaas | 00:46 | |
| *** Swami has quit IRC | 00:56 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: ACTIVE-ACTIVE: Initial distributor driver https://review.openstack.org/519509 | 01:06 |
|---|---|---|
| *** sshank has quit IRC | 01:12 | |
| *** yamamoto has joined #openstack-lbaas | 01:16 | |
| *** yamamoto has quit IRC | 01:21 | |
| *** bzhao has joined #openstack-lbaas | 01:28 | |
| *** bbzhao has joined #openstack-lbaas | 01:28 | |
| *** rcernin has quit IRC | 01:29 | |
| *** rcernin has joined #openstack-lbaas | 01:30 | |
| *** links has joined #openstack-lbaas | 01:44 | |
| *** sanfern has quit IRC | 02:03 | |
| *** sanfern has joined #openstack-lbaas | 02:04 | |
| *** yamamoto has joined #openstack-lbaas | 02:17 | |
| *** yamamoto has quit IRC | 02:22 | |
| *** annp has joined #openstack-lbaas | 02:27 | |
| openstackgerrit | Merged openstack/octavia master: Refine api-ref regarding loadbalancer creation https://review.openstack.org/519370 | 02:41 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Fix a gate breakage for tempest https://review.openstack.org/522082 | 02:52 |
| *** yamamoto has joined #openstack-lbaas | 03:09 | |
| openstackgerrit | Hengqing Hu proposed openstack/octavia-dashboard master: Fix Angular tabs https://review.openstack.org/522094 | 03:13 |
| mnaser | hey folks | 03:16 |
| mnaser | api.log:2017-11-16 20:24:39.609 2244 ERROR octavia.api.v1.controllers.load_balancer [req-b426893b-f66c-4b78-a1c4-908f66a7c491 - - - - -] Cannot delete Load Balancer 1f275fb2-e7ba-41c7-b4a8-0afc39ae29e2 - it has children: None | 03:17 |
| mnaser | anyoen ever ran into that bug? children: None sounds like something that isn't right | 03:17 |
| mnaser | i havent started digging into code yet | 03:17 |
| johnsom | mnaser If you have an LB that has pools or listeners, and you don't use the "cascade" parameter, you will get this error | 03:18 |
| johnsom | I'm not sure what the "None" is all about, but the error is pretty common user error | 03:18 |
| mnaser | johnsom: trying to figure out what the "None" means there though because the source code doesnt even have it :X | 03:19 |
| *** sanfern has quit IRC | 03:19 | |
| mnaser | johnsom: could that result in the load balancer ending up in error state when being deleted though? | 03:19 |
| mnaser | this is obviously going through the v1 api through the lbaas v2 | 03:20 |
| johnsom | No, it won't go to ERROR, it just returns an error code to the user/client. 4xx, I just don't remember which one. | 03:20 |
| mnaser | johnsom: hmm odd, because in neutron the provisioning_status is ERROR in 'neutron lbaas-loadbalancer-list' | 03:21 |
| johnsom | The delete request is rejected in the API before the LB is even touched | 03:21 |
| mnaser | well i guess in that case i have to figure out why it went to ERROR, time to dig up neutron logs | 03:21 |
| johnsom | Ah, well, hmm, neutron, maybe neutron would think that is an "ERROR" LB, but octavia will not | 03:22 |
| mnaser | thing is terraform/k8s still use the lbaas v2 api so i guess we want to get this figured out | 03:22 |
| johnsom | mnaser, You will have no end of failures and errors trying to use neutron-lbaas with terraform/k8s | 03:23 |
| mnaser | oh thats interesting | 03:23 |
| mnaser | johnsom: the first attempt returned a 204 | 03:23 |
| mnaser | 2017-11-16 20:36:31.973 34 INFO neutron.wsgi [req-e82504f1-b87f-4338-9da2-0132cbd0549c 6024403e7f6548f09ff344993a4b7874 534f3632c57f48498e6e3fc8c393e7c2 - default default] 172.21.1.200,172.21.1.30 "DELETE /v2.0/lbaas/loadbalancers/1f275fb2-e7ba-41c7-b4a8-0afc39ae29e2 HTTP/1.1" status: 204 len: 149 time: 0.6442392 | 03:23 |
| mnaser | further attempts returned a 409 | 03:23 |
| mnaser | 2017-11-16 21:03:42.758 35 INFO neutron.wsgi [req-bc3e87b5-768e-4e81-96f3-a116b3486509 6024403e7f6548f09ff344993a4b7874 534f3632c57f48498e6e3fc8c393e7c2 - default default] 73.96.206.126,172.21.1.20 "DELETE /v2.0/lbaas/loadbalancers/1f275fb2-e7ba-41c7-b4a8-0afc39ae29e2 HTTP/1.1" status: 409 len: 361 time: 0.3381560 | 03:23 |
| johnsom | Not sure If my last message went through or not, but I tried to put up some bubble gum patches to make neutron-lbaas less dumb about driver result codes, but in the end I found neutron-lbaas to be too broken. | 03:24 |
| mnaser | nope didnt get that | 03:24 |
| mnaser | shouldnt the first call return 409 from the first place :X | 03:24 |
| johnsom | Yeah, I got a disconnect message, I guess IRC issues | 03:24 |
| mnaser | nothing works today D: | 03:25 |
| johnsom | Seriously though, if your use case is teraform/k8s you really need to switch to Octavia native API and stop using neutron-lbaas | 03:26 |
| mnaser | johnsom: do they support it though? | 03:26 |
| mnaser | johnsom: afaik they talk to neutron-lbaas no? | 03:26 |
| johnsom | Basically neutron-lbaas and the API don't do locking right and have trouble with driver result codes. Any high rate requests through neutron-lbaas is going to get bad results as objects should be locked that aren't | 03:27 |
| johnsom | mnaser The patches are up for review and/or merging now from what I know | 03:27 |
| mnaser | johnsom: yeah the whole poll stuff is not ideal | 03:27 |
| mnaser | if it was up to me id drop the whole poll thing and make *every* request get forwarded to octavia | 03:28 |
| johnsom | Well, it's just unfortunate database work on the neutron-lbaas side. I thought I could bubble gum patch it, but the more I read the code, the more I realized it's an API re-write to fix | 03:28 |
| johnsom | mnaser https://review.openstack.org/418530 | 03:29 |
| mnaser | yeah i think the problem is the state has to be maintained in two places :\ | 03:29 |
| johnsom | No, actually, the problem is in neutron-lbaas. The other drivers have the same issue | 03:29 |
| mnaser | dang | 03:30 |
| johnsom | It's a long story, but it all starts with neutron-lbaas using auto-commit with the database and not actual transactions for the "test->lock" code paths | 03:30 |
| johnsom | Most of which is spread over a bunch of classes/methods | 03:31 |
| mnaser | darn | 03:31 |
| johnsom | I mean, this is part of why we started Octavia | 03:31 |
| mnaser | ok so good news #1 https://github.com/terraform-providers/terraform-provider-openstack/pull/149 | 03:31 |
| johnsom | Yep, as I mentioned, the support is in-flight now | 03:32 |
| mnaser | https://github.com/terraform-providers/terraform-provider-openstack/blob/master/CHANGELOG.md#100-november-08-2017 | 03:33 |
| mnaser | even better its released | 03:33 |
| johnsom | I also stress tested the Octavia API to show it doesn't have these issues.... | 03:33 |
| mnaser | yeah doesnt surprise me | 03:33 |
| mnaser | the whole back and forth will always cause issues | 03:33 |
| johnsom | I started this: https://review.openstack.org/#/c/516468 but then realized this is a hack that works around the issues for create, but there will still be other issues without a re-write | 03:35 |
| mnaser | i guess the only consumer left is | 03:35 |
| mnaser | the dashboard for octavia | 03:35 |
| johnsom | No, octavia-dashboard is Octavia native API | 03:36 |
| mnaser | okay well that covers most bases i guess | 03:36 |
| johnsom | https://github.com/openstack/octavia-dashboard | 03:36 |
| johnsom | Yeah, we hope to start the deprecation clock on neutron-lbaas in Queens. Just need to get the driver spec/framework merged | 03:37 |
| mnaser | johnsom: what sort of oddities can i expect running v1 and v2 side by side | 03:38 |
| johnsom | Octavia v1 and v2? | 03:39 |
| mnaser | yeah | 03:39 |
| mnaser | obviously v1 would be firewalled | 03:39 |
| johnsom | Works fine, just note v1 has little security. Otherwise, no issues. | 03:39 |
| mnaser | johnsom: i was thinking of keeping v1 only exposed to localhost | 03:40 |
| mnaser | and v2 would be public-ly exposed | 03:40 |
| johnsom | Yeah, works fine. We still run our gates with both enabled and tests for both | 03:43 |
| openstackgerrit | Hengqing Hu proposed openstack/octavia-dashboard master: Fix Angular tabs https://review.openstack.org/522094 | 03:55 |
| *** slaweq has joined #openstack-lbaas | 03:59 | |
| *** slaweq has quit IRC | 04:04 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Fix a gate breakage for tempest https://review.openstack.org/522082 | 04:19 |
| openstackgerrit | Merged openstack/octavia-dashboard master: Able to edit connection limit of listener https://review.openstack.org/520060 | 04:20 |
| mnaser | johnsom: exposed v2 api, the lb's which couldnt be deleted are status ERROR (i guess it auto switches to ERROR state because of operation timeout) | 04:36 |
| mnaser | provisioning_status in V2 api is PENDING_DELETE though | 04:36 |
| mnaser | oddly enough i dont see anything in the logs regarding the id of that specific load balancer in regards to a delete operation | 04:36 |
| johnsom | Hmmm, then something else went wrong. Can you look in the controller worker log and see if there is an error in there? | 04:37 |
| johnsom | What version of Octavia do you have? | 04:37 |
| mnaser | openstack-octavia-api-1.0.0-0.20170817030618.2c8cc57.el7.centos.noarch | 04:37 |
| mnaser | this is 1.0.0 though, just not the tagged release | 04:37 |
| johnsom | Ok , so pik | 04:37 |
| johnsom | pike | 04:37 |
| mnaser | yep | 04:38 |
| mnaser | hmm | 04:39 |
| mnaser | NotFound: Security group rule 63705156-cb62-4d57-97a2-a4677442f5ef does not exist | 04:39 |
| mnaser | i wonder if that is what made the delete fail | 04:39 |
| mnaser | and now its just in some weird condition | 04:40 |
| johnsom | It could be. I just posted a patch to make the network driver handle strange situations better. A missing security group might be in that group. | 04:40 |
| mnaser | i can imagine that as something which would put things in that state | 04:41 |
| johnsom | https://review.openstack.org/#/c/517455/ | 04:41 |
| johnsom | That driver was not very tolerant to things disappearing underneath it. | 04:41 |
| mnaser | that can probably explain the issue here hm | 04:41 |
| johnsom | And that error triggered a revert? | 04:42 |
| mnaser | johnsom: im not sure what happened but i guess it raised an exception and its stuck in pending delete? | 04:43 |
| mnaser | http://paste.openstack.org/show/627026/ | 04:43 |
| mnaser | it doesnt seem like it handles the exception? | 04:44 |
| mnaser | johnsom: i think i found it | 04:47 |
| mnaser | https://github.com/openstack/octavia/blob/master/octavia/network/drivers/neutron/allowed_address_pairs.py#L159 | 04:47 |
| mnaser | this has no exception handling | 04:47 |
| mnaser | johnsom: handle the exception raised of NotFound and "pass" because we asssume that it has been deleted | 04:48 |
| johnsom | Yep. Bummer, I didn't catch that one in my patch. It's still a problem. Do you want to make a patch for it or should I? | 04:49 |
| mnaser | johnsom: ill do one right now | 04:49 |
| mnaser | and a bug so we can hopefully backport | 04:49 |
| johnsom | Yeah, we can backport that | 04:50 |
| mnaser | fyi https://storyboard.openstack.org/#!/story/2001300 | 04:52 |
| johnsom | With the capstone tasks on all the flows now, you can't end in PENDING_* anymore. It will always either complete or go into ERROR state | 04:54 |
| mnaser | yeah, its progress :D | 04:54 |
| mnaser | glad i can find this quick and hopefully we can get it merged and backported quickly as a very major and vocal user is surfacing so i want to get it all cleared up | 04:55 |
| johnsom | +1 | 04:55 |
| mnaser | johnsom: i dont think this is something that can be checked by a unit test eh? :\ | 04:55 |
| johnsom | Many of us are on vacation this week, but hopefully we can get it in quick | 04:56 |
| johnsom | (technically I am on vacation myself) | 04:56 |
| mnaser | ha soorry :P | 04:56 |
| johnsom | Well, yes, it can be covered by a unit test | 04:56 |
| mnaser | mock the delete to raise a NotFound? | 04:57 |
| johnsom | Yep | 04:57 |
| johnsom | There should be some other examples there for NotFound | 04:57 |
| mnaser | cool! ill get the test in first | 04:57 |
| johnsom | We catch it in other parts of that code | 04:57 |
| mnaser | and then make sure it fails then redo it | 04:57 |
| johnsom | https://github.com/openstack/octavia/commit/c05a8cfb88bd1219a086e8826f7783b64f1e8537 | 04:59 |
| johnsom | mnaser ^^^ That is probably how you got there too | 05:01 |
| mnaser | yep | 05:05 |
| mnaser | got a test up now | 05:05 |
| mnaser | running the unit tests locally with it failing hopefully and ill do the fix | 05:06 |
| *** slaweq has joined #openstack-lbaas | 05:09 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Fix a gate breakage for tempest https://review.openstack.org/522082 | 05:09 |
| *** slaweq has quit IRC | 05:13 | |
| openstackgerrit | Mohammed Naser proposed openstack/octavia master: Handle race condition deleting security group rule https://review.openstack.org/522114 | 05:16 |
| mnaser | johnsom: ^ doneso :) | 05:16 |
| *** yamamoto_ has joined #openstack-lbaas | 05:18 | |
| *** bzhao has quit IRC | 05:18 | |
| *** bbzhao has quit IRC | 05:18 | |
| johnsom | Yeah, looks pretty good. I will watch for the gate, then +2 it | 05:19 |
| *** yamamoto has quit IRC | 05:21 | |
| *** oanson_ has joined #openstack-lbaas | 05:52 | |
| *** oanson has quit IRC | 05:52 | |
| *** oanson_ is now known as oanson | 05:52 | |
| openstackgerrit | Mohammed Naser proposed openstack/octavia master: Handle race condition deleting security group rule https://review.openstack.org/522114 | 05:53 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: ACTIVE-ACTIVE: Initial distributor driver https://review.openstack.org/519509 | 05:53 |
| *** slaweq has joined #openstack-lbaas | 05:54 | |
| *** slaweq has quit IRC | 05:58 | |
| openstackgerrit | Guoqiang Ding proposed openstack/neutron-lbaas master: Modify a detail in parse_stat https://review.openstack.org/522124 | 06:05 |
| openstackgerrit | Guoqiang Ding proposed openstack/neutron-lbaas master: Fix deprecated translation https://review.openstack.org/522127 | 06:19 |
| *** rtjure has quit IRC | 06:29 | |
| *** fnaval has quit IRC | 06:30 | |
| *** rtjure has joined #openstack-lbaas | 06:32 | |
| *** openstackgerrit has quit IRC | 06:33 | |
| *** fnaval has joined #openstack-lbaas | 07:01 | |
| *** fnaval has quit IRC | 07:06 | |
| *** openstackgerrit has joined #openstack-lbaas | 07:06 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron-lbaas master: Imported Translations from Zanata https://review.openstack.org/521435 | 07:06 |
| *** Alex_Staf has joined #openstack-lbaas | 07:08 | |
| *** rcernin has quit IRC | 07:17 | |
| *** ipsecguy_ has joined #openstack-lbaas | 07:22 | |
| *** ipsecguy has quit IRC | 07:22 | |
| *** yamamoto_ has quit IRC | 07:27 | |
| *** Alex_Staf has quit IRC | 07:32 | |
| *** eN_Guruprasad_Rn has joined #openstack-lbaas | 07:44 | |
| *** armax has quit IRC | 07:44 | |
| *** armax has joined #openstack-lbaas | 07:45 | |
| *** armax has quit IRC | 07:45 | |
| *** armax has joined #openstack-lbaas | 07:46 | |
| *** armax has quit IRC | 07:46 | |
| *** armax has joined #openstack-lbaas | 07:46 | |
| *** armax has quit IRC | 07:47 | |
| *** armax has joined #openstack-lbaas | 07:47 | |
| *** threestrands has quit IRC | 07:47 | |
| *** armax has quit IRC | 07:47 | |
| *** armax has joined #openstack-lbaas | 07:48 | |
| *** armax has quit IRC | 07:48 | |
| *** pcaruana has joined #openstack-lbaas | 07:49 | |
| *** armax has joined #openstack-lbaas | 07:49 | |
| *** armax has quit IRC | 07:49 | |
| *** armax has joined #openstack-lbaas | 07:50 | |
| *** armax has quit IRC | 07:50 | |
| *** tesseract has joined #openstack-lbaas | 08:14 | |
| *** slaweq has joined #openstack-lbaas | 08:16 | |
| *** slaweq has quit IRC | 08:17 | |
| *** slaweq has joined #openstack-lbaas | 08:17 | |
| *** yamamoto has joined #openstack-lbaas | 08:27 | |
| *** Alex_Staf has joined #openstack-lbaas | 08:28 | |
| *** yamamoto has quit IRC | 08:35 | |
| *** krypto has joined #openstack-lbaas | 08:41 | |
| *** krypto has quit IRC | 08:41 | |
| *** krypto has joined #openstack-lbaas | 08:41 | |
| openstackgerrit | Chenghui Yu proposed openstack/octavia master: Enable some off-by-default checks https://review.openstack.org/522178 | 08:54 |
| *** slaweq has quit IRC | 08:55 | |
| *** eN_Guruprasad_Rn has quit IRC | 09:07 | |
| sapd_ | hi every one! I'm installing octavia , I create octavia service end created endpoints for this service. But when I use command "openstack loadbalancer list" , the output is "Could not find requested endpoint in Service Catalog." | 09:09 |
| *** yamamoto has joined #openstack-lbaas | 09:11 | |
| *** slaweq has joined #openstack-lbaas | 09:17 | |
| *** eN_Guruprasad_Rn has joined #openstack-lbaas | 09:19 | |
| *** yamamoto has quit IRC | 09:19 | |
| *** slaweq has quit IRC | 09:23 | |
| *** yamamoto has joined #openstack-lbaas | 09:23 | |
| *** bar_ has joined #openstack-lbaas | 09:26 | |
| *** AlexeyAbashkin has joined #openstack-lbaas | 09:38 | |
| *** apuimedo has quit IRC | 09:54 | |
| *** krypto has quit IRC | 10:03 | |
| *** bar_ has quit IRC | 10:03 | |
| *** krypto has joined #openstack-lbaas | 10:04 | |
| *** krypto has quit IRC | 10:04 | |
| *** krypto has joined #openstack-lbaas | 10:04 | |
| *** salmankhan has joined #openstack-lbaas | 10:08 | |
| *** sri_ has quit IRC | 10:13 | |
| *** bar_ has joined #openstack-lbaas | 10:14 | |
| *** krypto has quit IRC | 10:15 | |
| *** krypto has joined #openstack-lbaas | 10:16 | |
| *** annp has quit IRC | 10:24 | |
| *** krypto has quit IRC | 10:38 | |
| *** krypto has joined #openstack-lbaas | 10:40 | |
| *** salmankhan has quit IRC | 10:46 | |
| *** salmankhan has joined #openstack-lbaas | 10:48 | |
| *** krypto has quit IRC | 10:49 | |
| *** krypto has joined #openstack-lbaas | 10:49 | |
| *** krypto has quit IRC | 10:49 | |
| *** krypto has joined #openstack-lbaas | 10:49 | |
| *** fnaval has joined #openstack-lbaas | 11:03 | |
| *** fnaval has quit IRC | 11:08 | |
| *** AlexeyAbashkin has quit IRC | 11:31 | |
| *** AlexeyAbashkin has joined #openstack-lbaas | 11:34 | |
| *** sanfern has joined #openstack-lbaas | 11:46 | |
| *** bar_ has quit IRC | 11:52 | |
| *** bar_ has joined #openstack-lbaas | 11:54 | |
| *** ianychoi has quit IRC | 12:00 | |
| *** ianychoi has joined #openstack-lbaas | 12:00 | |
| *** yamamoto has quit IRC | 12:20 | |
| openstackgerrit | Chenghui Yu proposed openstack/octavia master: Enable some off-by-default checks https://review.openstack.org/522178 | 12:24 |
| *** yamamoto has joined #openstack-lbaas | 12:24 | |
| *** AlexeyAbashkin has quit IRC | 12:29 | |
| *** sticker has joined #openstack-lbaas | 12:32 | |
| *** bar_ has quit IRC | 12:43 | |
| *** AlexeyAbashkin has joined #openstack-lbaas | 12:44 | |
| *** bar_ has joined #openstack-lbaas | 12:46 | |
| *** spectr has joined #openstack-lbaas | 12:53 | |
| openstackgerrit | Nir Magnezi proposed openstack/octavia master: Update devstack plugin and examples https://review.openstack.org/503638 | 12:57 |
| *** spectr has quit IRC | 13:04 | |
| *** fnaval has joined #openstack-lbaas | 13:04 | |
| *** fnaval has quit IRC | 13:08 | |
| *** bar_ has quit IRC | 13:10 | |
| *** kobis has joined #openstack-lbaas | 13:17 | |
| *** yamamoto has quit IRC | 13:21 | |
| *** yamamoto has joined #openstack-lbaas | 13:22 | |
| *** eN_Guruprasad_Rn has quit IRC | 13:29 | |
| *** AlexeyAbashkin has quit IRC | 13:43 | |
| *** links has quit IRC | 13:44 | |
| *** AlexeyAbashkin has joined #openstack-lbaas | 13:50 | |
| *** leitan has joined #openstack-lbaas | 13:52 | |
| *** bar_ has joined #openstack-lbaas | 14:06 | |
| *** salmankhan has quit IRC | 14:16 | |
| bcafarel | sorry my comment in https://review.openstack.org/#/c/518992/ turned into a wall of text… but feedback welcome on the idea | 14:20 |
| bcafarel | maybe I'll rewrite it as a ML post | 14:21 |
| *** salmankhan has joined #openstack-lbaas | 14:24 | |
| *** spectr has joined #openstack-lbaas | 14:27 | |
| openstackgerrit | Nir Magnezi proposed openstack/octavia master: Update devstack plugin and examples https://review.openstack.org/503638 | 14:34 |
| *** fnaval has joined #openstack-lbaas | 14:36 | |
| *** rcernin has joined #openstack-lbaas | 14:51 | |
| *** sticker_ has joined #openstack-lbaas | 14:53 | |
| *** sticker__ has joined #openstack-lbaas | 14:54 | |
| *** krypto has quit IRC | 14:55 | |
| *** krypto has joined #openstack-lbaas | 14:55 | |
| *** sticker has quit IRC | 14:57 | |
| *** sticker has joined #openstack-lbaas | 14:57 | |
| *** sticker_ has quit IRC | 14:58 | |
| *** armax has joined #openstack-lbaas | 14:58 | |
| *** sticker_ has joined #openstack-lbaas | 14:59 | |
| *** sticker__ has quit IRC | 15:00 | |
| *** sticker__ has joined #openstack-lbaas | 15:00 | |
| *** krypto has quit IRC | 15:01 | |
| *** krypto has joined #openstack-lbaas | 15:02 | |
| *** sticker has quit IRC | 15:02 | |
| *** Alex_Staf has quit IRC | 15:03 | |
| *** spectr has quit IRC | 15:03 | |
| *** sticker_ has quit IRC | 15:04 | |
| *** sticker__ has quit IRC | 15:05 | |
| *** sticker has joined #openstack-lbaas | 15:16 | |
| *** rcernin has quit IRC | 15:18 | |
| *** salmankhan has quit IRC | 15:22 | |
| *** salmankhan has joined #openstack-lbaas | 15:23 | |
| dmellado | bcafarel: what did you do now? :P | 15:31 |
| bcafarel | dmellado: just me rambling on how to build amphora images from distro packages, with fun things like selinux enabled :) | 15:33 |
| dmellado | bcafarel: oh, centhoras? | 15:33 |
| bcafarel | hey that's a cool name | 15:35 |
| dmellado | heh | 15:36 |
| cgoncalves | +1 | 15:36 |
| dmellado | btw, I'm spending some cycles on a related effort | 15:37 |
| dmellado | amphernetes | 15:38 |
| dmellado | so while being cramped up this week I'll catch up you folks on your next week's meetings :P | 15:38 |
| cgoncalves | dmellado: 2 awesome alias names in a row?! go, dmellado! | 15:39 |
| rm_work | lol | 15:48 |
| rm_work | those are pretty good :P | 15:48 |
| *** AlexeyAbashkin has quit IRC | 15:49 | |
| dmellado | \o/ | 15:55 |
| xgerman_ | o/ | 16:12 |
| xgerman_ | so Michael will be out the rest of the week and it’s probably all me (and nmagnezi today until | 16:13 |
| xgerman_ | I am off for the rest of the week | 16:13 |
| xgerman_ | best to ping me before EOB today if you need reviews by a core (and then lobby Nir for the +A) | 16:14 |
| nmagnezi | o/ | 16:16 |
| nmagnezi | xgerman_, no weekly meeting today? | 16:17 |
| xgerman_ | he asked me to run it ;-) | 16:17 |
| xgerman_ | but should be light | 16:17 |
| mnaser | darn | 16:18 |
| mnaser | i was gonna ping but im pretty sure my job is about to timeout in devstack on 522114 :( | 16:18 |
| *** slaweq has joined #openstack-lbaas | 16:21 | |
| *** sticker has quit IRC | 16:27 | |
| xgerman_ | mmh | 16:28 |
| *** kobis has quit IRC | 16:33 | |
| mnaser | yup it just timed out | 16:33 |
| mnaser | will recheck asap | 16:33 |
| xgerman_ | ok, Michael made his latest patch depends_on https://review.openstack.org/#/c/522077/ | 16:40 |
| xgerman_ | but with that merged it should all work? | 16:41 |
| *** slaweq has quit IRC | 16:45 | |
| *** tesseract has quit IRC | 16:57 | |
| johnsom | sapd_ You need to use “openstack endpoint create” to register the endpoints in keystone | 17:00 |
| *** eN_Guruprasad_Rn has joined #openstack-lbaas | 17:01 | |
| *** krypto has quit IRC | 17:02 | |
| *** salmankhan has quit IRC | 17:08 | |
| *** aojea has joined #openstack-lbaas | 17:10 | |
| *** salmankhan has joined #openstack-lbaas | 17:10 | |
| *** salmankhan has quit IRC | 17:14 | |
| *** pcaruana has quit IRC | 17:18 | |
| *** salmankhan has joined #openstack-lbaas | 17:25 | |
| *** Alex_Staf has joined #openstack-lbaas | 17:30 | |
| *** sanfern has quit IRC | 17:30 | |
| *** AlexeyAbashkin has joined #openstack-lbaas | 17:33 | |
| *** AlexeyAbashkin has quit IRC | 17:38 | |
| *** krypto has joined #openstack-lbaas | 17:47 | |
| *** krypto has quit IRC | 17:47 | |
| *** krypto has joined #openstack-lbaas | 17:47 | |
| *** yamamoto_ has joined #openstack-lbaas | 17:51 | |
| *** yamamoto has quit IRC | 17:55 | |
| *** yamamoto has joined #openstack-lbaas | 17:57 | |
| *** aojea has quit IRC | 17:58 | |
| *** aojea has joined #openstack-lbaas | 17:58 | |
| *** aojea has quit IRC | 17:58 | |
| *** aojea has joined #openstack-lbaas | 17:58 | |
| *** yamamoto_ has quit IRC | 17:59 | |
| *** sshank has joined #openstack-lbaas | 18:03 | |
| *** krypto has quit IRC | 18:04 | |
| *** sshank has quit IRC | 18:11 | |
| *** aojea has quit IRC | 18:17 | |
| *** sshank has joined #openstack-lbaas | 18:17 | |
| *** jniesz has joined #openstack-lbaas | 18:31 | |
| openstackgerrit | Carlos Goncalves proposed openstack/python-octaviaclient master: Add Quota client API and OSC support https://review.openstack.org/518767 | 18:36 |
| *** sanfern has joined #openstack-lbaas | 18:43 | |
| cgoncalves | johnsom: hi | 18:57 |
| cgoncalves | johnsom: re quota cli: what do you think of e.g. 'loadbalancer quota unset --pool'? | 18:58 |
| cgoncalves | obviously we can't use 'loadbalancer quota set --pool None' | 18:59 |
| mugsie | we did "reset" for designate - https://github.com/openstack/python-designateclient/blob/master/setup.cfg#L148 | 19:02 |
| xgerman_ | +1 | 19:06 |
| cgoncalves | mugsie: hmm that's not a bad idea. I just thought that 'unset' was the best pratice/most used one | 19:18 |
| cgoncalves | thanks! | 19:18 |
| mugsie | cgoncalves: yeah, I am not sure that the osc cli can actually reset / unset a quota at all | 19:21 |
| * mugsie just started looking at code | 19:21 | |
| bar_ | cgoncalves, you could make '--pool none' work by taking advantage of add_argument's type keyword. | 19:21 |
| *** longstaff has joined #openstack-lbaas | 19:21 | |
| bar_ | note that when I commented about that feature in your patch, I didn't realize at the time that remove_quota() essentially performs set_quota_to_defaults() | 19:24 |
| cgoncalves | bar_: not sure I follow you on the type keyword | 19:25 |
| cgoncalves | bar_: yeah, remote_quota() sets all quotas to default | 19:25 |
| mugsie | set --foo none would drive me nuts | 19:26 |
| bar_ | so why bother with selective unsetting? see what I mean? | 19:26 |
| cgoncalves | bar_: the none option doesn't sound very osc-compliant | 19:26 |
| bar_ | What I'm trying to say is that though it's technically possible to achieve '--pool none', you shouldn't. The reasoning is that selective unsetting has little use IMO | 19:28 |
| cgoncalves | bar_: if a user wants to revert a specific quota limit, he needs to reset_quota + set_quota again for all quotas he wants to set | 19:28 |
| bar_ | correct. | 19:28 |
| cgoncalves | but it's not that bad, I believe, and if someone requests quota-specific resetting option we could add in the future | 19:29 |
| bar_ | I agree. | 19:30 |
| *** slaweq has joined #openstack-lbaas | 19:30 | |
| openstackgerrit | Carlos Goncalves proposed openstack/python-octaviaclient master: Add Quota client API and OSC support https://review.openstack.org/518767 | 19:32 |
| *** eN_Guruprasad_Rn has quit IRC | 19:32 | |
| bar_ | I think resetting a single value to default has limited use-case. I'm more concerned with the naming of the resetting call: remove_quota() sounds misleading to me. | 19:32 |
| *** eN_Guruprasad_Rn has joined #openstack-lbaas | 19:32 | |
| *** salmankhan has quit IRC | 19:33 | |
| cgoncalves | bar_: agreed, though that's the name octavia api has | 19:33 |
| cgoncalves | https://developer.openstack.org/api-ref/load-balancer/v2/index.html#remove-a-quota | 19:33 |
| *** eN_Guruprasad_Rn has quit IRC | 19:35 | |
| openstackgerrit | Doug Hellmann proposed openstack/octavia master: do not add octavia directory to import path https://review.openstack.org/522371 | 19:42 |
| bar_ | cgoncalves, yeah. I'm aware of it. This functionality could benefit from better naming, even if the actual call remains the same. e.g. ResetQuota | 19:44 |
| *** aojea has joined #openstack-lbaas | 19:52 | |
| *** sanfern has quit IRC | 20:01 | |
| xgerman_ | #startmeeting Octavia | 20:01 |
| *** sanfern has joined #openstack-lbaas | 20:01 | |
| openstack | Meeting started Wed Nov 22 20:01:15 2017 UTC and is due to finish in 60 minutes. The chair is xgerman_. Information about MeetBot at http://wiki.debian.org/MeetBot. | 20:01 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 20:01 |
| *** openstack changes topic to " (Meeting topic: Octavia)" | 20:01 | |
| openstack | The meeting name has been set to 'octavia' | 20:01 |
| xgerman_ | #topic Announcements | 20:01 |
| *** openstack changes topic to "Announcements (Meeting topic: Octavia)" | 20:01 | |
| *** sanfern has quit IRC | 20:01 | |
| xgerman_ | so johnsom won’t attend today so it’s all me ;-) | 20:01 |
| *** sanfern has joined #openstack-lbaas | 20:02 | |
| nmagnezi | and me :D | 20:02 |
| *** sanfern has quit IRC | 20:02 | |
| longstaff | hi | 20:02 |
| xgerman_ | :-) | 20:02 |
| jniesz | hi | 20:02 |
| *** sanfern has joined #openstack-lbaas | 20:02 | |
| xgerman_ | #chair nmagnezi | 20:03 |
| openstack | Current chairs: nmagnezi xgerman_ | 20:03 |
| *** sanfern has quit IRC | 20:03 | |
| nmagnezi | xgerman_, ^^ i hope that chair is not Ikea made | 20:03 |
| *** sanfern has joined #openstack-lbaas | 20:03 | |
| nmagnezi | :) | 20:03 |
| xgerman_ | ok, I don’t really have any announcements other that most of us will be out until Monday | 20:03 |
| *** hogepodge has joined #openstack-lbaas | 20:03 | |
| *** sanfern has quit IRC | 20:04 | |
| xgerman_ | #topic Brief progress reports / bugs needing review | 20:04 |
| *** openstack changes topic to "Brief progress reports / bugs needing review (Meeting topic: Octavia)" | 20:04 | |
| *** slaweq has quit IRC | 20:04 | |
| *** sanfern has joined #openstack-lbaas | 20:04 | |
| *** sanfern has quit IRC | 20:04 | |
| xgerman_ | johnsom made some good progress on #link https://review.openstack.org/#/c/519509/ - please review | 20:05 |
| xgerman_ | we also need reviews for #link https://review.openstack.org/#/c/509957/ | 20:06 |
| nmagnezi | #link https://review.openstack.org/#/c/519509/ | 20:06 |
| xgerman_ | #link https://review.openstack.org/#/c/458308/ | 20:06 |
| jniesz | for the provider spec, auth token approach sounded ok to me | 20:06 |
| xgerman_ | QoS is pretty close but I also discovered that we don’t enable the qos extension in our devstack | 20:07 |
| *** salmankhan has joined #openstack-lbaas | 20:07 | |
| longstaff | ok thanks jniesz | 20:07 |
| bar_ | I've created a story about QoS progress: https://storyboard.openstack.org/#!/story/2001310 | 20:07 |
| nmagnezi | xgerman_, yeah, I also saw such comment from bar_ . any reason not to include that addition in the same patch? | 20:07 |
| bar_ | I made devstack deployment work earlier today | 20:07 |
| xgerman_ | we can add to that patch or make that patch dependant on the devstack update | 20:08 |
| nmagnezi | ack | 20:09 |
| xgerman_ | then we probably also need to tempest test… | 20:09 |
| xgerman_ | any updates on our test efforts? | 20:09 |
| nmagnezi | Alex_Staf, ^ ? | 20:10 |
| nmagnezi | we have a scenarios gate for the tempest plugin repo now, don't we? | 20:10 |
| xgerman_ | yeah, I think so, too ;-) | 20:11 |
| xgerman_ | #topic Provider driver spec | 20:11 |
| *** openstack changes topic to "Provider driver spec (Meeting topic: Octavia)" | 20:11 | |
| *** salmankhan has quit IRC | 20:11 | |
| Alex_Staf | hi guys | 20:12 |
| xgerman_ | hi | 20:12 |
| xgerman_ | I will make another topic for you after that ;-) | 20:12 |
| Alex_Staf | no news from my side . I bugged u enough on the plugin | 20:12 |
| xgerman_ | ok | 20:12 |
| Alex_Staf | I do have some request though I submitted | 20:12 |
| Alex_Staf | regarding HA documentation etc | 20:12 |
| xgerman_ | ok, it’s definitely on our radar | 20:13 |
| Alex_Staf | cool | 20:13 |
| Alex_Staf | another thing | 20:13 |
| *** sshank has quit IRC | 20:13 | |
| Alex_Staf | I think we need some api call tool that request for the octavia processes status - like neutron agent-list but for octavia | 20:14 |
| xgerman_ | yes, we talked about that in Denver at the PTG | 20:15 |
| jniesz | yea, where the driver would be required to give the status | 20:15 |
| Alex_Staf | I think it will help in the automated tests as well - to getr status after failover , for example | 20:15 |
| jniesz | and maybe an error field to list an error | 20:15 |
| Alex_Staf | xgerman_, really? I missed that I guess =\ | 20:15 |
| xgerman_ | it was on one of the provider driver days | 20:15 |
| xgerman_ | we should make sure the spec includes that call | 20:15 |
| *** longstaf_ has joined #openstack-lbaas | 20:16 | |
| jniesz | that is something we should require drivers to implement | 20:16 |
| nmagnezi | xgerman_, not sure, but do o-hm and o-hk even use RPC? is not, I wonder how can we achieve an "agent-list" for octavia services | 20:16 |
| nmagnezi | btw I think we should be discussing the provider spec at the moment.. :P | 20:17 |
| xgerman_ | we are doing it - agent list is something all providers need | 20:17 |
| xgerman_ | but we call it more “component” list | 20:17 |
| nmagnezi | ack | 20:18 |
| xgerman_ | nmagnezi we can start having the octavia components sending heartbeats | 20:18 |
| nmagnezi | xgerman_, yup. | 20:18 |
| xgerman_ | or use one of the new etcd integrations… | 20:19 |
| *** longstaff has quit IRC | 20:19 | |
| Alex_Staf | xgerman_, another question . I discussed with nir the octavia operation and we know that api sends the request to the workers ( ha setup 3 controllers- 3 api,worker,HM,houseKeeper) , and it send the task to the messaging que and one of the workers pick it up and execute right ? | 20:19 |
| nmagnezi | do we have a story for that? | 20:19 |
| nmagnezi | if not, maybe Alex_Staf can submit one | 20:19 |
| xgerman_ | yes | 20:20 |
| Alex_Staf | nmagnezi, octavia components sending heartbeats ? | 20:20 |
| jniesz | etcd seems to make sense for status / health | 20:20 |
| nmagnezi | Alex_Staf, about the octavia components list | 20:20 |
| Alex_Staf | xgerman_, how do the other 2 workers are not executing the request again ? | 20:20 |
| xgerman_ | jniesz +1 | 20:20 |
| xgerman_ | we pop the request from the queue | 20:20 |
| xgerman_ | and if the worker with the request dies you are SoL | 20:21 |
| Alex_Staf | when it was received it removed fro mthe que ? | 20:21 |
| xgerman_ | yes | 20:21 |
| Alex_Staf | cool | 20:21 |
| Alex_Staf | what is SoL ? | 20:21 |
| Alex_Staf | nmagnezi, lets discuss that tomorrow - the story | 20:21 |
| xgerman_ | the idea is that one day we use task board and taskflow will manage persisting tasks | 20:21 |
| xgerman_ | out of luck | 20:21 |
| nmagnezi | O_o | 20:22 |
| xgerman_ | ok, after we went on a tangent is there anything longstaf_ needs to move ahead with the spec | 20:23 |
| xgerman_ | ? | 20:23 |
| jniesz | to limit provider drivers from only updating the status of their objects (amphora, loadbalancer, listener...) with auth token | 20:23 |
| jniesz | from provider, do we need to add something like "owner" field to the tables | 20:23 |
| xgerman_ | we have a provider field on LB | 20:23 |
| xgerman_ | that’s what i thought we would be using | 20:23 |
| jniesz | ok, I don't remember seeing that field added in the spec | 20:24 |
| jniesz | for lb | 20:24 |
| jniesz | i guess it is implied through flavor | 20:25 |
| xgerman_ | it has been there forever | 20:25 |
| kong | hi, guys, does that make sense to support http communication between worker and amphora? | 20:25 |
| jniesz | yes, nm | 20:25 |
| longstaf_ | for auth token: Octavia would generate a token and pass to provider driver when loading driver, and drivers include token on calls to update Octavia | 20:25 |
| longstaf_ | does this make sense? | 20:26 |
| *** bar_ has quit IRC | 20:26 | |
| nmagnezi | kong, kindly wait for the open part :) | 20:26 |
| *** fnaval has quit IRC | 20:26 | |
| kong | ooh, sorry, i didn't know i am in a 'meeting room'... | 20:26 |
| xgerman_ | longstaf_ I see two authentications | 20:27 |
| xgerman_ | 1) Is the token to update status/stats | 20:27 |
| xgerman_ | 2) is user/pwd to read from Octavia API if needed | 20:27 |
| xgerman_ | (1) since we are using a library we can leave that for implementation details | 20:28 |
| xgerman_ | there might be multiple “drivers” with different auth methods | 20:28 |
| xgerman_ | I am not sure what volume those updates are and if we get a ton every second we likely need to do soemthing different then when they are more sporadic | 20:29 |
| jniesz | I think the use-case is for the driver to update status/stats of only objects that it owns | 20:29 |
| jniesz | and the library having protection, so one driver doesn't update status of objects from another | 20:30 |
| xgerman_ | indeed | 20:30 |
| jniesz | shouldn't that be a single auth method, since it would be part of the library | 20:30 |
| xgerman_ | I like the library to only do the updates… for anyhting else a driver should use the Octavia API | 20:31 |
| jniesz | yea, agreed | 20:31 |
| jniesz | strictly for status / stats update | 20:31 |
| jniesz | anything else should leverage existing apiv | 20:32 |
| xgerman_ | +1 | 20:32 |
| longstaf_ | agreed | 20:32 |
| *** slaweq has joined #openstack-lbaas | 20:32 | |
| xgerman_ | anyhow I like to keep auth/transport driver specific so we can swap that out if we can’t handle the volume | 20:33 |
| xgerman_ | or somebody has a weird network topology | 20:33 |
| xgerman_ | ok, moving on | 20:35 |
| xgerman_ | #topic Open Discussion | 20:35 |
| *** openstack changes topic to "Open Discussion (Meeting topic: Octavia)" | 20:35 | |
| nmagnezi | kong, ^ | 20:37 |
| kong | hi | 20:37 |
| kong | just wanna ask does that make sense to support http communication between worker and amphora? we don't want to use https in our ci environment, it will increase our complexity | 20:37 |
| nmagnezi | to the best of my knowledge it is will not work out of the box | 20:37 |
| kong | yeah | 20:38 |
| nmagnezi | i looked at it once | 20:38 |
| xgerman_ | ok, so we generate the certs ourselves and out the amphora-id in them | 20:38 |
| xgerman_ | then we check if the amphora-id in the server cert matches the one we are supposed to talk to | 20:38 |
| nmagnezi | kong, can't you have the same workflow we have in devstack? | 20:38 |
| xgerman_ | which means if we go to http that would all need to be refactored | 20:38 |
| nmagnezi | It's CI... it might fit. | 20:39 |
| xgerman_ | +1 | 20:39 |
| nmagnezi | xgerman_, +1 | 20:39 |
| nmagnezi | xgerman_, that's exactly what I saw when I looked at it in the context of tripleO support | 20:39 |
| kong | i want to know if adding an option makes sense | 20:40 |
| kong | to the upstream | 20:40 |
| kong | we don't want to maintain private code :-) | 20:40 |
| nmagnezi | can you describe the usecase? | 20:40 |
| xgerman_ | yeah, I can’t see anyone wanting less security ;-) | 20:40 |
| nmagnezi | what would anyone who use Octavia in production would prefer http over https? | 20:41 |
| nmagnezi | what/ why | 20:41 |
| jniesz | shouldn't CI match what you would run in prod? | 20:41 |
| xgerman_ | +1 | 20:41 |
| nmagnezi | +1 | 20:41 |
| kong | ok, i know. doesn't make sense to add this kind of thing just for ci | 20:41 |
| kong | yeah, in our prod, we use https | 20:42 |
| nmagnezi | sorry for that guys but I need to drop | 20:42 |
| xgerman_ | yeah, since the SSL stuff has a lot of custom things you likely need to test that | 20:42 |
| nmagnezi | o/ | 20:42 |
| kong | just because the specialty of our ci | 20:42 |
| kong | i mean, special case | 20:43 |
| xgerman_ | well, I think we don’t see a need for upstream at the moment but we won’t stop you | 20:43 |
| kong | we will try to figure out another way for that, thanks | 20:43 |
| xgerman_ | ok | 20:43 |
| xgerman_ | anything else to discuss? | 20:44 |
| xgerman_ | #endmeeting | 20:45 |
| *** openstack changes topic to "Welcome to LBaaS / Octavia - Queens development is now open." | 20:45 | |
| openstack | Meeting ended Wed Nov 22 20:45:04 2017 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 20:45 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/octavia/2017/octavia.2017-11-22-20.01.html | 20:45 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/octavia/2017/octavia.2017-11-22-20.01.txt | 20:45 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/octavia/2017/octavia.2017-11-22-20.01.log.html | 20:45 |
| xgerman_ | Happy Thanksgiving… | 20:45 |
| longstaf_ | Happy Thansgiving | 20:45 |
| Alex_Staf | Happy Thanksgiving guys | 20:45 |
| jniesz | happy thanksgiving | 20:48 |
| openstackgerrit | Doug Hellmann proposed openstack/octavia master: Change how path munging works for sphinx docs https://review.openstack.org/521211 | 20:51 |
| openstackgerrit | Doug Hellmann proposed openstack/octavia master: ignore api-ref/build directory https://review.openstack.org/522385 | 20:51 |
| *** Alex_Staf has quit IRC | 20:53 | |
| *** longstaf_ has quit IRC | 20:54 | |
| *** sshank has joined #openstack-lbaas | 21:28 | |
| *** bar_ has joined #openstack-lbaas | 21:34 | |
| *** slaweq has quit IRC | 21:41 | |
| *** threestrands has joined #openstack-lbaas | 21:45 | |
| *** threestrands has quit IRC | 21:45 | |
| *** threestrands has joined #openstack-lbaas | 21:45 | |
| *** fnaval has joined #openstack-lbaas | 21:45 | |
| *** fnaval has quit IRC | 21:49 | |
| *** AlexeyAbashkin has joined #openstack-lbaas | 22:09 | |
| *** AlexeyAbashkin has quit IRC | 22:13 | |
| *** rcernin has joined #openstack-lbaas | 22:20 | |
| *** sshank has quit IRC | 22:42 | |
| *** aojea has quit IRC | 22:44 | |
| *** sshank has joined #openstack-lbaas | 22:44 | |
| *** aojea has joined #openstack-lbaas | 22:44 | |
| *** aojea has quit IRC | 22:48 | |
| *** aojea has joined #openstack-lbaas | 22:53 | |
| openstackgerrit | Bar RH proposed openstack/octavia master: Rename 'remove' to 'reset' in Quota API https://review.openstack.org/522402 | 22:54 |
| *** leitan has quit IRC | 22:55 | |
| *** leitan has joined #openstack-lbaas | 22:56 | |
| openstackgerrit | Bar RH proposed openstack/octavia master: Rename 'remove' to 'reset' in Quota API https://review.openstack.org/522402 | 22:59 |
| *** aojea has quit IRC | 22:59 | |
| *** AlexeyAbashkin has joined #openstack-lbaas | 23:08 | |
| *** jniesz has quit IRC | 23:10 | |
| *** AlexeyAbashkin has quit IRC | 23:12 | |
| *** bar_ has quit IRC | 23:18 | |
| *** ipsecguy_ has quit IRC | 23:46 | |
| *** leitan has quit IRC | 23:54 | |
| *** ipsecguy has joined #openstack-lbaas | 23:55 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!