| openstackgerrit | Santosh Sharma proposed openstack/neutron-lbaas: Changes to wait for loadbalancer status after any CRUD operation https://review.openstack.org/168046 | 00:14 |
|---|---|---|
| openstackgerrit | Santosh Sharma proposed openstack/neutron-lbaas: Changes to wait for loadbalancer status after any CRUD operation https://review.openstack.org/168046 | 00:18 |
| Santosh_NS | https://review.openstack.org/#/c/168046/ for review Above changes are required to run tempest api tests with async client operations. | 00:23 |
| *** xgerman has quit IRC | 00:39 | |
| *** mwang2_ has joined #openstack-lbaas | 00:39 | |
| *** mwang2 has quit IRC | 00:41 | |
| *** mwang2_ has quit IRC | 00:46 | |
| *** BrianShang_ has joined #openstack-lbaas | 00:54 | |
| *** jamiem has quit IRC | 00:56 | |
| *** BrianShang has quit IRC | 00:56 | |
| *** redrobot has quit IRC | 00:59 | |
| *** woodster_ has quit IRC | 01:00 | |
| openstackgerrit | Merged openstack/neutron-lbaas: Updating .coveragerc to exclude the tests https://review.openstack.org/169050 | 01:00 |
| *** mordred has quit IRC | 01:01 | |
| *** bharath has quit IRC | 01:03 | |
| *** redrobot has joined #openstack-lbaas | 01:06 | |
| *** mordred has joined #openstack-lbaas | 01:06 | |
| *** redrobot is now known as Guest48074 | 01:06 | |
| *** SumitNaiksatam has joined #openstack-lbaas | 01:10 | |
| openstackgerrit | Madhusudhan Kandadai proposed stackforge/octavia: Implements the haproxy amphora agent api server https://review.openstack.org/160034 | 01:13 |
| *** woodster_ has joined #openstack-lbaas | 01:14 | |
| *** jamiem has joined #openstack-lbaas | 01:16 | |
| *** bharath has joined #openstack-lbaas | 01:22 | |
| *** madhu_ak has quit IRC | 01:23 | |
| *** bharath has quit IRC | 01:31 | |
| *** bharath has joined #openstack-lbaas | 01:33 | |
| *** bharath has quit IRC | 02:05 | |
| *** bharath has joined #openstack-lbaas | 02:06 | |
| *** jamiem has quit IRC | 02:10 | |
| *** jamiem has joined #openstack-lbaas | 02:11 | |
| *** bharath has quit IRC | 02:25 | |
| openstackgerrit | Phillip Toohill proposed openstack/neutron-lbaas: WIP: Neutron LBaaS: TLS Barbican Scenario Test https://review.openstack.org/164828 | 02:30 |
| openstackgerrit | Phillip Toohill proposed openstack/neutron-lbaas: Neutron LBaaS: Load Balancer Basic Scenario Test https://review.openstack.org/158904 | 02:33 |
| openstackgerrit | Phillip Toohill proposed openstack/neutron-lbaas: WIP: Neutron LBaaS: TLS Barbican Scenario Test https://review.openstack.org/164828 | 02:34 |
| *** fnaval has quit IRC | 03:15 | |
| *** fnaval has joined #openstack-lbaas | 03:30 | |
| *** sbfox has joined #openstack-lbaas | 03:31 | |
| *** chlong has joined #openstack-lbaas | 03:48 | |
| *** chlong has quit IRC | 03:49 | |
| *** jamiem has quit IRC | 03:53 | |
| *** woodster_ has quit IRC | 04:00 | |
| *** sbfox has quit IRC | 04:32 | |
| *** dougwig has quit IRC | 04:46 | |
| *** ctracey has quit IRC | 04:46 | |
| *** dougwig has joined #openstack-lbaas | 04:48 | |
| *** ctracey has joined #openstack-lbaas | 04:49 | |
| *** kobis has joined #openstack-lbaas | 05:02 | |
| *** kobis has quit IRC | 05:13 | |
| *** xgerman has joined #openstack-lbaas | 05:19 | |
| *** kobis has joined #openstack-lbaas | 05:23 | |
| *** bharath has joined #openstack-lbaas | 05:26 | |
| *** kobis has quit IRC | 05:29 | |
| *** xgerman_ has joined #openstack-lbaas | 05:30 | |
| *** xgerman has quit IRC | 05:30 | |
| *** bharath has quit IRC | 05:31 | |
| *** xgerman__ has joined #openstack-lbaas | 05:32 | |
| *** xgerman_ has quit IRC | 05:35 | |
| *** xgerman__ has quit IRC | 05:38 | |
| *** BrianShang_ has quit IRC | 06:06 | |
| *** woodster_ has joined #openstack-lbaas | 06:13 | |
| *** mlavalle has quit IRC | 06:19 | |
| *** ajmiller__ has joined #openstack-lbaas | 06:31 | |
| *** ajmiller_ has quit IRC | 06:34 | |
| *** mwang2_ has joined #openstack-lbaas | 06:42 | |
| *** fnaval has quit IRC | 06:42 | |
| *** mwang2_ has quit IRC | 06:46 | |
| *** sbalukoff has quit IRC | 07:17 | |
| *** amotoki has joined #openstack-lbaas | 07:23 | |
| *** niusmallnan has joined #openstack-lbaas | 07:40 | |
| niusmallnan | hi, who knows, what is the common way to make the LBaaS agent highly available? | 07:42 |
| *** kobis has joined #openstack-lbaas | 07:43 | |
| *** fnaval has joined #openstack-lbaas | 07:49 | |
| *** kobis has quit IRC | 08:05 | |
| *** woodster_ has quit IRC | 08:20 | |
| openstackgerrit | Santosh Sharma proposed openstack/neutron-lbaas: Changes to wait for loadbalancer status after any CRUD operation https://review.openstack.org/168046 | 08:28 |
| Santosh_NS | https://review.openstack.org/#/c/168046/ for review Above changes are required to run tempest api tests with async client operations. | 08:31 |
| *** SumitNaiksatam has left #openstack-lbaas | 08:57 | |
| *** niusmallnan has left #openstack-lbaas | 09:15 | |
| openstackgerrit | Santosh Sharma proposed openstack/neutron-lbaas: Changes to wait for loadbalancer status after any CRUD operation (Changes required to support async client operations) https://review.openstack.org/168046 | 09:33 |
| Santosh_NS | ping | 11:47 |
| *** amotoki has quit IRC | 13:33 | |
| *** openstackgerrit has quit IRC | 13:36 | |
| *** openstackgerrit has joined #openstack-lbaas | 13:36 | |
| *** apuimedo has joined #openstack-lbaas | 13:46 | |
| *** localloop127 has joined #openstack-lbaas | 14:02 | |
| *** apuimedo has quit IRC | 14:05 | |
| *** apuimedo has joined #openstack-lbaas | 14:14 | |
| *** TrevorV|Home has joined #openstack-lbaas | 14:20 | |
| *** woodster_ has joined #openstack-lbaas | 14:29 | |
| *** openstackgerrit has quit IRC | 14:39 | |
| *** openstackgerrit has joined #openstack-lbaas | 14:39 | |
| *** sbalukoff has joined #openstack-lbaas | 15:06 | |
| *** ajmiller__ has quit IRC | 15:06 | |
| *** ajmiller has joined #openstack-lbaas | 15:07 | |
| *** ajmiller has quit IRC | 15:12 | |
| *** ajmiller has joined #openstack-lbaas | 15:17 | |
| *** amotoki has joined #openstack-lbaas | 15:35 | |
| *** mwang2_ has joined #openstack-lbaas | 15:53 | |
| *** fnaval has quit IRC | 15:57 | |
| *** bharath has joined #openstack-lbaas | 16:01 | |
| *** fnaval has joined #openstack-lbaas | 16:13 | |
| *** SumitNaiksatam has joined #openstack-lbaas | 16:19 | |
| *** bharath has quit IRC | 16:32 | |
| *** bharath__ has joined #openstack-lbaas | 16:32 | |
| *** xgerman has joined #openstack-lbaas | 16:48 | |
| *** localloop127 has quit IRC | 16:51 | |
| TrevorV|Home | xgerman, you around? | 16:57 |
| xgerman | yes | 16:57 |
| TrevorV|Home | I manually performed the SFTP file transfer via paramiko for the SSH driver, and the file ends up with 664 permissions. | 16:57 |
| TrevorV|Home | is 600 what we want on the amphora, or 644, or what? | 16:58 |
| xgerman | we want 600 for cetrificate pem files | 16:58 |
| xgerman | since they contain private keys | 16:59 |
| xgerman | haproxy file doesn't matter | 16:59 |
| TrevorV|Home | Alright, but for consistency I'll do 600 for all. | 17:09 |
| xgerman | sounds good ;-) | 17:12 |
| TrevorV|Home | xgerman, ptoohill has jinja set up to accept custom log value locations type of thing. | 17:16 |
| TrevorV|Home | log file*** | 17:16 |
| xgerman | perfext | 17:16 |
| ptoohill | well, its not there yet | 17:17 |
| TrevorV|Home | He's suggesting a separate review to make it actually set up that way though, since its outside the scope of the ssh_driver itself. | 17:17 |
| ptoohill | but it will be | 17:17 |
| TrevorV|Home | (should let him speak for himself :P ) | 17:17 |
| ptoohill | :) | 17:17 |
| ptoohill | ill make update to take all those in as params and just default to whats in the template if None is provided | 17:17 |
| ptoohill | the template is a config file, in fact the template can be a custom template placed anywhere | 17:18 |
| xgerman | you know our views differ here... but sounds ok | 17:18 |
| ptoohill | if the operator wants to do something slightly different with the templates they should be able to | 17:19 |
| ptoohill | other wise this could have been all stringed up with no need for a template | 17:19 |
| ptoohill | kinda like v1 was | 17:19 |
| xgerman | well, sno template is just a mess... but this will work for me... | 17:20 |
| ptoohill | sno? | 17:21 |
| ptoohill | er no*? | 17:21 |
| ptoohill | :P | 17:21 |
| ptoohill | fair enough | 17:21 |
| ptoohill | itll be configurable | 17:21 |
| *** SumitNaiksatam has quit IRC | 17:37 | |
| openstackgerrit | min wang proposed stackforge/octavia: health manager service https://review.openstack.org/160061 | 17:42 |
| *** bharath__ has quit IRC | 17:51 | |
| *** bharath has joined #openstack-lbaas | 18:01 | |
| *** bharath has quit IRC | 18:02 | |
| *** bharath has joined #openstack-lbaas | 18:02 | |
| *** SumitNaiksatam has joined #openstack-lbaas | 18:03 | |
| *** SumitNaiksatam has left #openstack-lbaas | 18:04 | |
| *** SumitNaiksatam has joined #openstack-lbaas | 18:05 | |
| *** madhu_ak has joined #openstack-lbaas | 18:12 | |
| openstackgerrit | Michael Johnson proposed stackforge/octavia: Updating the disk image builder code https://review.openstack.org/170578 | 18:13 |
| openstackgerrit | Trevor Vardeman proposed stackforge/octavia: Amphora SSH Driver https://review.openstack.org/160964 | 18:17 |
| *** raj_ has joined #openstack-lbaas | 18:25 | |
| raj_ | hi looking for stable version of devstack havin lbass | 18:25 |
| raj_ | while dealing with kilo,getting lot of errors | 18:25 |
| johnsom | raj_ what version of kilo were you trying? | 18:35 |
| *** sbfox has joined #openstack-lbaas | 18:37 | |
| raj_ | hey johnsom , are you asking for lbaas version ? | 18:38 |
| johnsom | kilo version | 18:39 |
| johnsom | Are you using one of the milestone releases, release candidate, etc. | 18:40 |
| johnsom | I would recommend using a release candidate version of Kilo | 18:42 |
| raj_ | i am trying to use github repo | 18:47 |
| raj_ | I am not sure which versio i am getting by that | 18:47 |
| raj_ | I am pulling it from master branch | 18:50 |
| johnsom | This page may help: https://chapter60.wordpress.com/2015/02/20/installing-openstack-lbaas-version-2-on-kilo-using-devstack/ | 18:51 |
| *** kobis has joined #openstack-lbaas | 18:54 | |
| *** sbfox has quit IRC | 19:03 | |
| *** sbfox has joined #openstack-lbaas | 19:04 | |
| *** madhu_ak has quit IRC | 19:08 | |
| *** sbfox has quit IRC | 19:08 | |
| *** raj_ has quit IRC | 19:11 | |
| *** barclaac has joined #openstack-lbaas | 19:13 | |
| *** barclaac|2 has quit IRC | 19:16 | |
| *** kobis1 has joined #openstack-lbaas | 19:19 | |
| *** mwang2_ has quit IRC | 19:19 | |
| *** kobis has quit IRC | 19:20 | |
| *** madhu_ak has joined #openstack-lbaas | 19:29 | |
| *** apuimedo has quit IRC | 19:29 | |
| *** TrevorV|Home has quit IRC | 19:31 | |
| *** mspreitz has joined #openstack-lbaas | 19:36 | |
| mspreitz | Does anybody here know whether Neutron's LBaaS can do IPv6 on the front side while doing IPv4 on the back? | 19:37 |
| *** barclaac has quit IRC | 19:38 | |
| *** madhu_ak_ has joined #openstack-lbaas | 19:39 | |
| *** madhu_ak_ has quit IRC | 19:41 | |
| *** madhu_ak has quit IRC | 19:42 | |
| *** amotoki has quit IRC | 19:47 | |
| *** mwang2 has joined #openstack-lbaas | 19:47 | |
| *** kobis1 has quit IRC | 20:03 | |
| *** sbfox has joined #openstack-lbaas | 20:04 | |
| *** kobis has joined #openstack-lbaas | 20:04 | |
| openstackgerrit | min wang proposed stackforge/octavia: Update specs in amphora-driver-interface https://review.openstack.org/170599 | 20:05 |
| openstackgerrit | min wang proposed stackforge/octavia: Update specs in amphora-driver-interface https://review.openstack.org/170599 | 20:06 |
| *** sbfox has quit IRC | 20:09 | |
| *** kobis has quit IRC | 20:09 | |
| *** kobis has joined #openstack-lbaas | 20:20 | |
| mspreitz | Does anybody here know whether Neutron's LBaaS can do IPv6 on the front side while doing IPv4 on the back? | 20:33 |
| rm_work | it should work, I think -- blogan> | 20:34 |
| rm_work | ? | 20:34 |
| blogan | mspreitz: i have not actually tested that out, but I would think it would | 20:39 |
| *** kobis has quit IRC | 20:40 | |
| rm_work | so really that's a question for the deployment backend, I think -- for example, if it is the HAProxy backend, then it is whether or not HAProxy supports this | 20:42 |
| rm_work | or if it's A10 as the backend, does A10's LB support this | 20:42 |
| blogan | haproxy does, but since the vip port is created in neutron_lbaas, that would be a factor as well | 20:42 |
| rm_work | since Neutron-LBaaS really just handles the deployment/configuration of the LB solution | 20:43 |
| blogan | but yeah after that, depends on what the driver is capable of | 20:43 |
| blogan | i dont see how any of the software load balancers wouldn't support that though | 20:45 |
| mspreitz | And none of this requires both IPv4 and IPv6 subnets on the same Neutron network, right? The LB's front side can be on a different network than the backends, with no Neutron router between them, right? | 20:58 |
| mspreitz | blogan: rm_work: ^^ | 20:58 |
| rm_work | depends on the plugin again, i assume | 20:59 |
| rm_work | but, probably | 20:59 |
| mspreitz | rm_work: let's say the HAProxy plugin for LBaaS, to make it concrete | 20:59 |
| rm_work | for HAProxy that would work, since we'd just plug the member networks as well as the frontend network | 20:59 |
| mspreitz | thanks | 21:00 |
| dougwig | you could put both ends on non-neutron networks with some drivers, if you were insane. :) | 21:03 |
| *** sbfox has joined #openstack-lbaas | 21:06 | |
| *** ajmiller has quit IRC | 21:07 | |
| dougwig | do any of our tests actually exercise the v6 code path? | 21:07 |
| mspreitz | dougwig: good question! | 21:09 |
| blogan | mspreitz: i actually do not believe the haproxy one right now supports frontend and backend on separate networks, so ipv6 frontend and ipv4 backend would not work (unless there's some way to get ipv4 and ipv6 addresses on the same subnet) | 21:10 |
| blogan | dougwig: not that i know of | 21:10 |
| dougwig | which of us should file a bug to get that addressed? | 21:10 |
| blogan | reason being is we wanted to get parity with v1, and v1 did nto support this either | 21:10 |
| dougwig | the tests, i mean. | 21:10 |
| *** sbfox has quit IRC | 21:10 | |
| blogan | dougwig; he who smelt it dealt it | 21:11 |
| blogan | need to put that multi-network load balancing implementation in the vancouver talks | 21:12 |
| blogan | bc the question will come up as to who is responsible for setting up the connectivity with the members, the plugin, driver, or user | 21:12 |
| dougwig | err, huh? | 21:13 |
| blogan | so members have the subnet_id on them now right? | 21:14 |
| dougwig | isn't the fix for that to switch to octavia as the ref? or several of the vendors can handle that scenario just fine as well. | 21:14 |
| blogan | okay if octavia becomes the ref then that would fix it yes | 21:14 |
| blogan | but octavia will be creating the amphora ports on the member networks | 21:14 |
| blogan | how does your driver connect up the frontend and backend if they're on two different networks? | 21:15 |
| blogan | wait thats why you needed that query in yoru driver right? | 21:15 |
| rm_you | blogan: wait, why does haproxy care? | 21:15 |
| dougwig | it either leaves it as an exercise for the grader, or plumbs itself directly into the vxlan/vlan and goes direct. | 21:15 |
| rm_you | if the routes exist at the system level, why does it matter? | 21:15 |
| blogan | the haproxy driver in neutron-lbaas is running in a network namespace | 21:16 |
| dougwig | only matters if you want to keep the traffic off your neutron node. | 21:16 |
| rm_you | oh | 21:16 |
| rm_you | right, haproxy-namespace | 21:16 |
| rm_you | i was thinking haproxy octavia | 21:16 |
| blogan | so it either needs a interface on the member networks, or its gateway does | 21:16 |
| *** ajmiller has joined #openstack-lbaas | 21:17 | |
| blogan | or its gateway's gateway, but you get the point | 21:17 |
| dougwig | unless you want to telepathy into the network, it does need a path to the other device, yes. | 21:17 |
| *** apuimedo has joined #openstack-lbaas | 21:17 | |
| blogan | dougwig: what do you mean by grader? | 21:17 |
| dougwig | we have a mode where we set ourselves up as a VTEP endpoint with OVS, and are just linked directly into the tenant mesh. | 21:18 |
| dougwig | blogan: that's where we just trust the implicit ip routing of the network. | 21:18 |
| dougwig | those last two statements are separate, not related. | 21:18 |
| blogan | well either way, to answer mspreitz question, the haproxy ns driver currently does not support it | 21:20 |
| blogan | octavia should though | 21:20 |
| rm_you | blogan: the real answer is "it's not up to neutron-lbaas really" | 21:21 |
| mspreitz | blogan: thanks | 21:21 |
| blogan | rm_you: it is if the ref impl doesn't support it | 21:22 |
| *** sbalukoff has quit IRC | 21:22 | |
| blogan | rm_you: i mean its neutron-lbaas's problem | 21:22 |
| rm_you | eh | 21:22 |
| rm_you | i think of that as a plugin still | 21:22 |
| rm_you | just happens to be the default one | 21:22 |
| blogan | and the one we maintain | 21:23 |
| blogan | especially with dougwig's shim driver revolution | 21:23 |
| rm_you | but that hopefully zero deployments use | 21:23 |
| dougwig | well, it's our problem in that our ref impl is kinda lame. | 21:24 |
| dougwig | the fix for that is to speed octavia into place, IMO. | 21:24 |
| blogan | its a problem we're working on to fix though! | 21:24 |
| dougwig | yes, not arguing that | 21:24 |
| blogan | the network driver i've been working has been coming along well, after i thought it was done but then realized i missed a ton of shit with security groups | 21:25 |
| openstackgerrit | min wang proposed stackforge/octavia: Add health check amphora-driver-interface https://review.openstack.org/170599 | 21:26 |
| dougwig | blogan: that part just got towed away temporarily. | 21:30 |
| blogan | i hope you get caught in my undertow | 21:31 |
| openstackgerrit | min wang proposed stackforge/octavia: Add health check amphora-driver-interface https://review.openstack.org/170599 | 21:37 |
| *** madhu_ak has joined #openstack-lbaas | 21:49 | |
| madhu_ak | dougwig: the lbaasv2 api CI check is failing for me : https://review.openstack.org/#/c/165627/ :( | 21:49 |
| dougwig | madhu_ak: yep, that's a problem in my driver. it won't hold up for change. | 21:50 |
| madhu_ak | hmm I see.. but please let me know if I need to make any changes to make it work | 21:51 |
| dougwig | madhu_ak: ok, i will. | 21:55 |
| madhu_ak | thanks! | 21:55 |
| mspreitz | blogan: rm_you: Let me try the 4-4 version of the question. Can I use LBaaS with HAProxy plugin where the LB listens on one network and the backends are on a different network? | 22:13 |
| rm_you | apparently not with the HAProxy namespace plugin | 22:15 |
| *** mspreitz has quit IRC | 22:17 | |
| dougwig | unless we broke it, that driver used to let you put in any ip address, not just a nova vm reference. so as long as it's routable IPs, you could add the backend on a different network that way. | 22:17 |
| *** sbfox has joined #openstack-lbaas | 22:21 | |
| *** sbfox has quit IRC | 22:25 | |
| rm_you | dougwig: but blogan was saying the namespace driver needed them to be subnets on the same network for it to workj | 22:27 |
| rm_you | for some reason, i don't really know | 22:27 |
| *** xgerman has quit IRC | 22:43 | |
| *** amotoki has joined #openstack-lbaas | 22:48 | |
| *** amotoki has quit IRC | 22:52 | |
| *** mlavalle has joined #openstack-lbaas | 23:01 | |
| openstackgerrit | Al Miller proposed stackforge/octavia: Add devstack plugin for octavia https://review.openstack.org/167796 | 23:38 |
| *** mspreitz has joined #openstack-lbaas | 23:42 | |
| *** ajmiller has quit IRC | 23:42 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!