*** bzhao_ has quit IRC | 00:09 | |
*** armax has joined #openstack-lbaas | 00:22 | |
openstackgerrit | Merged openstack/octavia master: Support QoS policy apply on vip port https://review.openstack.org/441912 | 00:43 |
---|---|---|
*** reedip has quit IRC | 00:59 | |
*** fnaval has joined #openstack-lbaas | 01:09 | |
*** amotoki has quit IRC | 01:16 | |
*** amotoki has joined #openstack-lbaas | 01:22 | |
openstackgerrit | Merged openstack/octavia master: blueprint LBaaS Alternative Monitoring IP/Port https://review.openstack.org/453001 | 01:59 |
openstackgerrit | Merged openstack/python-octaviaclient master: Initialize plugin for OSC https://review.openstack.org/446223 | 02:13 |
openstackgerrit | XieYingYun proposed openstack/octavia master: Optimize the link address https://review.openstack.org/454427 | 02:16 |
*** amotoki has quit IRC | 03:00 | |
*** reedip has joined #openstack-lbaas | 03:09 | |
*** amotoki has joined #openstack-lbaas | 03:19 | |
*** armax has quit IRC | 03:20 | |
*** ianychoi has quit IRC | 03:36 | |
*** amotoki has quit IRC | 03:37 | |
*** ianychoi has joined #openstack-lbaas | 03:42 | |
*** links has joined #openstack-lbaas | 03:47 | |
openstackgerrit | Ankur proposed openstack/octavia master: Octavia Feature Classification https://review.openstack.org/451177 | 03:55 |
*** amotoki has joined #openstack-lbaas | 04:00 | |
*** amotoki has quit IRC | 04:01 | |
*** amotoki has joined #openstack-lbaas | 04:01 | |
*** voelzmo has joined #openstack-lbaas | 04:02 | |
*** voelzmo has quit IRC | 04:10 | |
*** reedip has quit IRC | 04:11 | |
*** fnaval has quit IRC | 04:18 | |
*** voelzmo has joined #openstack-lbaas | 04:20 | |
*** reedip has joined #openstack-lbaas | 04:21 | |
openstackgerrit | Ankur proposed openstack/octavia master: Introduce TESTING.rst for quick developer reference https://review.openstack.org/403938 | 04:27 |
*** yamamoto has quit IRC | 04:30 | |
*** yamamoto has joined #openstack-lbaas | 04:30 | |
*** voelzmo has quit IRC | 04:43 | |
*** reedip has quit IRC | 04:44 | |
*** rcernin has joined #openstack-lbaas | 04:48 | |
*** rcernin has quit IRC | 04:55 | |
*** rcernin has joined #openstack-lbaas | 04:55 | |
*** rcernin is now known as rcernin|wfh | 04:55 | |
*** yamamoto_ has joined #openstack-lbaas | 05:07 | |
*** yamamoto has quit IRC | 05:11 | |
*** charcol has joined #openstack-lbaas | 05:15 | |
*** yamamoto has joined #openstack-lbaas | 05:20 | |
*** yamamoto_ has quit IRC | 05:24 | |
*** foutatoro has quit IRC | 05:38 | |
*** csomerville has quit IRC | 05:55 | |
*** yamamoto_ has joined #openstack-lbaas | 06:02 | |
*** cody-somerville has joined #openstack-lbaas | 06:03 | |
*** yamamoto has quit IRC | 06:06 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron-lbaas-dashboard master: Updated from global requirements https://review.openstack.org/454505 | 06:10 |
openstackgerrit | OpenStack Proposal Bot proposed openstack/octavia master: Updated from global requirements https://review.openstack.org/454507 | 06:12 |
openstackgerrit | Ankur proposed openstack/python-octaviaclient master: [WIP] Add rest of load balancer commands https://review.openstack.org/454516 | 06:14 |
*** charcol has quit IRC | 06:25 | |
*** yamamoto_ has quit IRC | 06:30 | |
*** pcaruana has joined #openstack-lbaas | 06:34 | |
*** yamamoto has joined #openstack-lbaas | 06:38 | |
*** voelzmo has joined #openstack-lbaas | 06:42 | |
*** cody-somerville has quit IRC | 07:06 | |
*** cody-somerville has joined #openstack-lbaas | 07:07 | |
*** tesseract has joined #openstack-lbaas | 07:07 | |
*** aojea_ has joined #openstack-lbaas | 07:16 | |
*** yamamoto has quit IRC | 07:31 | |
*** yamamoto has joined #openstack-lbaas | 07:32 | |
*** gcheresh has joined #openstack-lbaas | 07:42 | |
*** yamamoto has quit IRC | 08:09 | |
*** gcheresh has quit IRC | 08:37 | |
*** gcheresh has joined #openstack-lbaas | 08:45 | |
*** yamamoto has joined #openstack-lbaas | 08:46 | |
*** voelzmo has quit IRC | 08:58 | |
*** voelzmo has joined #openstack-lbaas | 09:05 | |
*** voelzmo has quit IRC | 09:06 | |
*** yamamoto has quit IRC | 09:13 | |
*** yamamoto has joined #openstack-lbaas | 09:15 | |
*** gongysh has quit IRC | 09:28 | |
*** reedip has joined #openstack-lbaas | 09:29 | |
*** gongysh has joined #openstack-lbaas | 09:30 | |
*** sputnik13 has quit IRC | 09:40 | |
*** sputnik13 has joined #openstack-lbaas | 09:41 | |
*** amotoki has quit IRC | 10:10 | |
*** links has quit IRC | 10:16 | |
*** tesseract has quit IRC | 10:30 | |
*** links has joined #openstack-lbaas | 10:32 | |
*** tesseract has joined #openstack-lbaas | 10:44 | |
*** gcheresh has quit IRC | 11:03 | |
*** links has quit IRC | 11:04 | |
*** links has joined #openstack-lbaas | 11:05 | |
*** voelzmo has joined #openstack-lbaas | 11:05 | |
*** blogan__ has joined #openstack-lbaas | 11:07 | |
*** blogan_ has quit IRC | 11:07 | |
*** openstackgerrit has quit IRC | 11:18 | |
*** pcaruana has quit IRC | 11:22 | |
reedip | rm_work : hi | 11:33 |
xgerman | hi | 12:15 |
reedip | rm_work : As I would be travelling back to India today, I didnt get too much time to work ( shopping for good stuff, you see ) , so the HM patch is open :) | 12:22 |
xgerman | no worries… I think we can tackle that next week ;-) | 12:26 |
xgerman | also R naming vote opened | 12:27 |
reedip | Yeah, ... I lov Raspberry Pi :) | 12:28 |
*** gongysh has quit IRC | 12:29 | |
xgerman | I had a feeling that would win | 12:30 |
reedip | It should, but then Openstack Rock also sounds good :D | 12:52 |
*** gcheresh has joined #openstack-lbaas | 12:57 | |
*** yamamoto has quit IRC | 13:03 | |
*** catintheroof has joined #openstack-lbaas | 13:03 | |
*** gcheresh has quit IRC | 13:06 | |
*** yamamoto has joined #openstack-lbaas | 13:07 | |
*** chlong has joined #openstack-lbaas | 13:33 | |
*** gongysh has joined #openstack-lbaas | 13:50 | |
*** amotoki has joined #openstack-lbaas | 14:01 | |
*** ipsecguy has joined #openstack-lbaas | 14:05 | |
*** ipsecguy_ has quit IRC | 14:05 | |
*** amotoki has quit IRC | 14:19 | |
*** links has quit IRC | 14:25 | |
*** armax has joined #openstack-lbaas | 14:41 | |
*** blogan__ is now known as blogan | 14:49 | |
*** fnaval has joined #openstack-lbaas | 14:54 | |
*** voelzmo has quit IRC | 15:02 | |
*** fnaval has quit IRC | 15:03 | |
*** gongysh has quit IRC | 15:04 | |
*** rcernin|wfh has quit IRC | 15:04 | |
*** fnaval has joined #openstack-lbaas | 15:05 | |
*** catintheroof has quit IRC | 15:06 | |
*** catintheroof has joined #openstack-lbaas | 15:06 | |
*** catintheroof has quit IRC | 15:07 | |
*** catintheroof has joined #openstack-lbaas | 15:15 | |
*** catintheroof has quit IRC | 15:16 | |
*** catintheroof has joined #openstack-lbaas | 15:26 | |
*** catintheroof has quit IRC | 15:27 | |
*** amotoki has joined #openstack-lbaas | 15:30 | |
*** gcheresh has joined #openstack-lbaas | 15:37 | |
*** aojea_ has quit IRC | 15:42 | |
*** openstackgerrit has joined #openstack-lbaas | 15:51 | |
openstackgerrit | Michael Johnson proposed openstack/octavia master: Explicitly list neutron api extensions for tests https://review.openstack.org/454769 | 15:51 |
*** amotoki has quit IRC | 16:15 | |
*** voelzmo has joined #openstack-lbaas | 16:18 | |
openstackgerrit | Merged openstack/octavia master: Updated from global requirements https://review.openstack.org/454507 | 16:27 |
*** KeithMnemonic has joined #openstack-lbaas | 16:35 | |
*** KeithMnemonic1 has joined #openstack-lbaas | 16:35 | |
*** bbzhao has quit IRC | 16:36 | |
*** bbzhao has joined #openstack-lbaas | 16:37 | |
*** tesseract has quit IRC | 16:42 | |
openstackgerrit | Merged openstack/neutron-lbaas-dashboard master: Updated from global requirements https://review.openstack.org/454505 | 16:50 |
*** catintheroof has joined #openstack-lbaas | 16:51 | |
*** armax has quit IRC | 16:57 | |
*** catintheroof has quit IRC | 16:57 | |
*** bbzhao has quit IRC | 17:22 | |
*** bbzhao has joined #openstack-lbaas | 17:22 | |
*** chlong_ has joined #openstack-lbaas | 17:46 | |
*** armax has joined #openstack-lbaas | 17:47 | |
*** kobis has joined #openstack-lbaas | 17:58 | |
openstackgerrit | German Eichberger proposed openstack/octavia master: ACTIVE-ACTIVE Topology: Initial Distributor Driver Mixin https://review.openstack.org/313006 | 18:04 |
xgerman | ok, rewrote that first patch | 18:04 |
*** kobis has quit IRC | 18:08 | |
*** kobis has joined #openstack-lbaas | 18:08 | |
*** kobis has quit IRC | 18:09 | |
johnsom | Nice. Ready for a review pass? | 18:18 |
xgerman | sure, jenkins hates it but that should be an easy fix | 18:19 |
johnsom | Ok, maybe I will give it a look this afternoon | 18:19 |
xgerman | cool | 18:20 |
openstackgerrit | German Eichberger proposed openstack/octavia master: ACTIVE-ACTIVE Topology: Initial Distributor Driver Mixin https://review.openstack.org/313006 | 18:40 |
*** kobis has joined #openstack-lbaas | 18:47 | |
xgerman | arrgh, forgot to fix the data model | 18:52 |
rm_work | johnsom: i have a new guy that wants to join octavia upstream contribution!! | 18:53 |
rm_work | in a couple of hours I'll be walking him through the process to get set up to push reviews | 18:53 |
rm_work | need to find him a good intro bug | 18:54 |
johnsom | Awesome! Welcome | 18:54 |
rm_work | anything that comes to mind before I dive into the bug tracker? | 18:54 |
rm_work | I know we had quite a few around the v2 api that should be simple | 18:54 |
rm_work | dunno if all were claimed yet | 18:54 |
johnsom | I am happy to do project overview and code layout training if you don't want to do it | 18:54 |
johnsom | Yeah, there are a number of open bugs worth looking at | 18:55 |
johnsom | Use the lbaas-merge tag filter | 18:55 |
reedip | hey johnsom , rm_work , do we have a developer doc guide which tells about the state transition in Octavia? | 18:56 |
*** diltram_ has joined #openstack-lbaas | 18:56 | |
reedip | I mean what states occur during create/update/delete? | 18:57 |
*** diltram has quit IRC | 18:57 | |
johnsom | reedip Take a look at this: https://developer.openstack.org/api-ref/load-balancer/v2/index.html#status-codes | 18:58 |
reedip | also, a completely separate query, do we have a plan to Purge Loadbalancers of a tenant ? ( ankur-gupta-f4 , this might be needed in SDK ) | 18:58 |
reedip | johnsom ok thanks, will do | 18:58 |
reedip | johnsom, yeah , I got what I wanted, thanks | 18:59 |
johnsom | reedip Let me know if that meets your needs | 18:59 |
johnsom | reedip So given a tenant delete all of their resources? | 18:59 |
reedip | johnsom :yes | 18:59 |
johnsom | reedip We have cascade delete, but I don't think we have an RFE for purge tenant | 19:00 |
reedip | johnsom : I guess we need one, and as SDK and OSC is being created right now, this would a good time to integrate it ( or keep it handy) | 19:01 |
rm_work | that sounds like an admin thing right? | 19:01 |
reedip | xgerman: something similar in FWaaS as well, I think ^^ | 19:01 |
rm_work | admin API is on the short list for summit topics, I assume | 19:01 |
johnsom | reedip Go for it! Grin | 19:01 |
reedip | rm_work , yeah , it would be admin dependent | 19:01 |
johnsom | We are doing an integrated API and using RBAC to protect "admin" actions. | 19:02 |
johnsom | The framework is there, we just haven't finished it yet | 19:02 |
johnsom | for the RBAC stuff. | 19:03 |
rm_work | right i mean | 19:03 |
rm_work | we need to talk about / plan | 19:03 |
reedip | johnsom : but the purge action is a bit different, like we already have purge for Ports, networks, compute instances, storages | 19:03 |
rm_work | what we want to see for admin commands | 19:03 |
rm_work | and how we structure it | 19:03 |
johnsom | Yeah | 19:03 |
reedip | rm_work. johnsom: but load balancers, firewalls are not directly connected to these resources, so we need to keep these on track as well | 19:03 |
reedip | I mean they are, but they are not tracked during purge | 19:04 |
rm_work | right but | 19:06 |
rm_work | we just need to talk about how we want to go about doing this stuff | 19:06 |
rm_work | and make a spec for the "admin" stuff | 19:06 |
rm_work | even if it isn't a separate endpoint | 19:06 |
rm_work | so we can do it once, right, and not have to deprecate random weird one-off things people wrote, or maintain a bunch of stuff with really weird structure forever | 19:07 |
reedip | rm_work : you mean about the purge? | 19:07 |
rm_work | yes | 19:07 |
rm_work | we DO need that | 19:07 |
rm_work | I think we can agree it's something we'll want to see | 19:07 |
rm_work | but we need to have a discussion about admin commands as a whole | 19:07 |
reedip | rm_work : purge command would simply call the delete. But the call of the delete action would be in the OSC and octaviaclient | 19:08 |
rm_work | right | 19:08 |
rm_work | ah | 19:08 |
reedip | so I think the focus would be how to deal with it there | 19:08 |
rm_work | you just mean in client | 19:08 |
rm_work | not as an admin api action? | 19:08 |
xgerman | yeah, I am co fused about the need for purge, too | 19:08 |
rm_work | that's just a delete-loop | 19:08 |
rm_work | we don't need anything special for that | 19:08 |
reedip | rm_work : do you think admin api also needs a purge action ? | 19:08 |
reedip | wont that be a bit complicated? | 19:08 |
xgerman | I thought admin was more stuff like migrate load balancer vm, rotate images, chamge topology | 19:09 |
rm_work | for LBID in $(openstack loadbalancer list -f json | jq .id); do openstack loadbalancer delete $LBID; done | 19:09 |
rm_work | xgerman: right | 19:09 |
rm_work | a tenant suspend/purge might make sense though | 19:09 |
rm_work | as a single API call for an admin to make | 19:09 |
xgerman | yep, we also don’t want to put oo much orchestration into our API | 19:09 |
ankur-gupta-f4 | also note i don't think we have cascade delete in v2 yet | 19:09 |
ankur-gupta-f4 | so there may be orphaned resources | 19:09 |
ankur-gupta-f4 | i believe | 19:09 |
rm_work | right, maybe | 19:10 |
xgerman | I wrote cascade delete so it must be somewhere | 19:10 |
rm_work | though as I discovered, ALL delete was cascade delete in v1, by accident | 19:10 |
rm_work | sooooo | 19:10 |
*** aojea has joined #openstack-lbaas | 19:10 | |
xgerman | yes, that was what I wanted ;-) | 19:10 |
xgerman | no accident | 19:10 |
rm_work | xgerman: yeah, in neutron-lbaas, dunno fi we have implemented that functionality in octavia's v2 implementation yet | 19:10 |
rm_work | lolol | 19:10 |
rm_work | xgerman sabotaging the delete command :P | 19:10 |
reedip | hehe | 19:10 |
xgerman | no, it had a bug before and didn’t delete everyhting | 19:11 |
openstackgerrit | German Eichberger proposed openstack/octavia master: ACTIVE-ACTIVE Topology: Initial Distributor Driver Mixin https://review.openstack.org/313006 | 19:11 |
xgerman | let’s hope this works | 19:11 |
johnsom | I'm with rm_work, this needs a spec. I don't want any "graph" BS going forward | 19:13 |
reedip | now I am confused | 19:14 |
rm_work | xgerman: we HAD a cascade delete command. it was technically broken, and wouldn't run | 19:14 |
rm_work | the NORMAL delete command was cascading, but since it wasn't intended to cascade, it was doing so improperly T_T | 19:14 |
rm_work | I submitted fixes for both issues like... only three weeks ago | 19:15 |
xgerman | well, I wrote that stuff like two years ago | 19:15 |
rm_work | heh yeah | 19:15 |
xgerman | and we entertained the idea that delete was broken and needed to cascade | 19:15 |
xgerman | because horizon would call delete on LB and expect cascade | 19:16 |
johnsom | reedip For the purge, let's write up an RFE or spec calling out the need and proposed approach. | 19:16 |
johnsom | xgerman but we have a path /cascade or something like that to trigger the cascade. I would have just done a param myself, but.... | 19:16 |
xgerman | yeah, that is v2 | 19:17 |
xgerman | I think we were chatting v1 ;-) | 19:17 |
rm_work | no, that is also v1 | 19:18 |
xgerman | at that point we can do whatever — I found the cascading delete battle in lbaaasv2 neutron silly from the get-go | 19:18 |
rm_work | https://github.com/openstack/octavia/blob/master/octavia/api/v1/controllers/load_balancer.py#L289 | 19:18 |
reedip | johnsom : the purge was for OSC , but if you think we need a purge from Admin side in the API as well, I wil | 19:18 |
reedip | :) | 19:19 |
rm_work | reedip: i think it doesn't make sense in the client | 19:19 |
*** openstackgerrit has quit IRC | 19:19 | |
rm_work | it's doing an orchestration layer in the client for something that is a single line of BASH I can type up in 30 seconds | 19:19 |
reedip | rm_work : yes, exactly right...but not with LB but also ALL tenant resources | 19:19 |
rm_work | and automated systems should be able to do their own logic for this based on ... whatever | 19:20 |
reedip | so u execute a purge commands and the whole MATRIX code just crashes :D | 19:20 |
rm_work | if the LB is gone, nothing else is provisioned | 19:20 |
xgerman | nah, most people don’t have that many LB and most clouds don’t have that many tenants | 19:20 |
rm_work | it's all just DB objects | 19:20 |
reedip | ok | 19:20 |
rm_work | so nothing but LBs matters functionally | 19:20 |
reedip | lets break it into 2 sections then. I guess you guys agree an API side purge command is required | 19:20 |
xgerman | but agree with johnsom we should write a spec/RfE and see | 19:20 |
rm_work | yep | 19:20 |
rm_work | we need to discuss and spec | 19:21 |
rm_work | maybe it turns out we need an API command | 19:21 |
rm_work | maybe we don't | 19:21 |
reedip | from the OSC side, I suggest keep a purge command as well | 19:21 |
rm_work | I am receptive to the idea, as it sounds like a common admin task | 19:21 |
reedip | for the API side/RFE I will write the spec | 19:21 |
reedip | for the OSC side, I will discuss with ankur-gupta-f4 and dtroyer | 19:21 |
johnsom | reedip Thank you | 19:21 |
xgerman | +1 | 19:21 |
rm_work | "oh shit we need to shut down this tenant NOW because they are violating EULA" -> POST /admin/tenant_destroy | 19:22 |
rm_work | or whatever | 19:22 |
rm_work | seems like a realistic need | 19:22 |
reedip | ok , great .. cya guys soon, time for my flight :D | 19:22 |
xgerman | rm_work you have tenants? | 19:22 |
rm_work | yeah | 19:26 |
rm_work | we're on Liberty :P | 19:26 |
johnsom | Sigh | 19:27 |
rm_work | except Octavia!!! | 19:27 |
rm_work | We'll be running master, lagging about a month probably | 19:27 |
johnsom | That is one of the big bonus from doing the merge.... | 19:28 |
rm_work | yep | 19:28 |
rm_work | we're practically release-independent in reality | 19:28 |
rm_work | because all we do is call APIs with stuff that's been stable for years | 19:28 |
rm_work | want to run Octavia-master on a cloud that is running Kilo? no worries | 19:31 |
rm_work | should "just work" | 19:31 |
johnsom | That is the plan... | 19:31 |
rm_work | keystone/nova/neutron APIs have been stable forever for the stuff we touch | 19:31 |
johnsom | Well, kilo. I know we did some work on neutron in kilo for our stuff | 19:31 |
rm_work | hmm | 19:32 |
rm_work | yeah maybe | 19:32 |
*** chlong_ has quit IRC | 19:37 | |
*** chlong has quit IRC | 19:37 | |
*** aojea has quit IRC | 19:39 | |
*** aojea has joined #openstack-lbaas | 19:40 | |
*** chlong has joined #openstack-lbaas | 19:42 | |
*** aojea has quit IRC | 19:45 | |
*** kobis has quit IRC | 19:49 | |
*** reedip has quit IRC | 20:02 | |
rm_work | johnsom: actually I am going to have him look at fixing the threading and sleep() model in our daemons | 20:11 |
rm_work | healthmanager and housekeeping | 20:11 |
rm_work | so they won't suck as much | 20:11 |
rm_work | since I think we have a simple fix for that | 20:11 |
rm_work | and removing our dependency on eventlet entirely | 20:12 |
*** openstackgerrit has joined #openstack-lbaas | 20:13 | |
openstackgerrit | Adam Harwell proposed openstack/octavia master: [WIP] Removing dependency on eventlet https://review.openstack.org/454873 | 20:13 |
rm_work | that's part 1 ^^ | 20:13 |
*** voelzmo has quit IRC | 20:31 | |
*** darkfires has joined #openstack-lbaas | 20:39 | |
darkfires | anyone know why I might be getting WARNING stevedore.named [req-267acb74-c8e2-4692-afd5-a0bf61dc8ba3 - - - - -] Could not load neutron_lbaas.drivers.haproxy.namespace_driver.HaproxyNSDriver ? when i did an strace it didn't even try to open anything in neutron_lbaas/drivers/* | 20:40 |
*** aojea has joined #openstack-lbaas | 20:45 | |
*** gcheresh has quit IRC | 20:57 | |
*** voelzmo has joined #openstack-lbaas | 21:01 | |
*** voelzmo has quit IRC | 21:05 | |
*** chlong has quit IRC | 21:08 | |
johnsom | Hi | 21:11 |
johnsom | Are you trying to use the old namespace driver, Octavia, or other? | 21:11 |
*** catintheroof has joined #openstack-lbaas | 21:12 | |
johnsom | rm_work yeah, the housekeeping is annoying and a pretty easy fix. Healthmanager I think the delay shutting down is mostly the processing finishing up. | 21:14 |
*** blogan_ has joined #openstack-lbaas | 21:14 | |
darkfires | yeah the old namespace driver on newton | 21:15 |
johnsom | Hmm, ok. Check two things: | 21:15 |
*** catintheroof has quit IRC | 21:16 | |
*** blogan has quit IRC | 21:16 | |
*** blogan_ is now known as blogan | 21:16 | |
johnsom | In neutron_lbaas.conf, under the "[service_providers]" section, service_provider should be: LOADBALANCERV2:Haproxy:neutron_lbaas.drivers.haproxy.plugin_driver.HaproxyOnHostPluginDriver:default | 21:19 |
johnsom | in neutron.conf, "[DEFAULT]", service_plugins should have "lbaasv2" in the list | 21:20 |
darkfires | both those are already set just like that | 21:20 |
johnsom | That said, you should consider the octavia driver. It brings HA, etc. that the namespace driver doesn't have | 21:20 |
johnsom | And neutron-lbaas is installed? | 21:21 |
*** armax has quit IRC | 21:21 | |
darkfires | it is | 21:21 |
darkfires | openstack-neutron-lbaas-9.2.0-1.el7.noarch python-neutron-lbaas-9.2.0-1.el7.noarch | 21:21 |
johnsom | Hmmm, yeah, not very familiar with that packaging. It comes up in a "pip list"? | 21:23 |
johnsom | Is it running under py2.7 and you are running neutron under py3.5? | 21:23 |
darkfires | no it's the centos 7.3 packaging so no pip on it, have everything else working without issue so im surprised thats not | 21:24 |
darkfires | py2.7.5 no 3.5 installed | 21:24 |
johnsom | Yeah, I'm not sure why it's not finding it either. | 21:25 |
johnsom | I was just thinking pip would check the python eggs and show if it is showing up there. | 21:25 |
darkfires | i could install pip | 21:26 |
darkfires | # pip list | grep lbaas neutron-lbaas (9.2.0) | 21:27 |
johnsom | Hmm, well there goes that idea | 21:27 |
darkfires | i edited the stevedore src code to try and narrow it down ... and purposely made a config error on the interface driver for comparison | 21:29 |
darkfires | http://dpaste.com/110V4TN | 21:29 |
darkfires | you can see the interface driver it actually loads the possibilities | 21:30 |
darkfires | but the first one doesn't load anything at all | 21:30 |
darkfires | http://dpaste.com/25K0X7F | 21:32 |
darkfires | files are definitely there tho | 21:32 |
darkfires | but yeah since that driver is smoking crack i guess ill try the octavia one | 21:33 |
xgerman | let us know how that goes | 21:34 |
darkfires | thanks for your help btw.. much appreciated :) | 21:34 |
johnsom | That is what most of us use. Same lbaasv2 in the service_plugin list, service_provider = LOADBALANCERV2:Octavia:neutron_lbaas.drivers.octavia.driver.OctaviaDriver:default | 21:34 |
johnsom | The fact it didn't find the bridge interface driver either is interesting | 21:35 |
darkfires | no i purposely did that | 21:35 |
darkfires | for a point of comparison in debuging | 21:35 |
darkfires | the bridge thing | 21:35 |
darkfires | wanted to see what it would do with something that already worked (loading the list of available interfaces) | 21:36 |
johnsom | Ok. | 21:36 |
johnsom | Cores can you take a quick look at: https://review.openstack.org/#/c/446082/ | 21:38 |
johnsom | It's the base cookiecutter for the tempest plugin. It would be nice to have that landed | 21:39 |
johnsom | Hey, the DIB patch merged... Maybe our py3 gates are happy now | 21:43 |
*** aojea has quit IRC | 21:44 | |
openstackgerrit | Ankur proposed openstack/octavia master: Introduce TESTING.rst for quick developer reference https://review.openstack.org/403938 | 21:48 |
*** KeithMnemonic has quit IRC | 22:15 | |
*** KeithMnemonic1 has quit IRC | 22:15 | |
darkfires | using octavia is wayy more involved than just switching the driver, wants barbican and other stuff.. going to have to debug the other driver the hard way i guess | 22:20 |
johnsom | Actually both of those drivers use barbican | 22:21 |
johnsom | But, yes, octavia is more invovled | 22:21 |
darkfires | could that be why i'm getting could not load then, because of barbican? | 22:22 |
johnsom | I don't think so. stevedore loading the basic driver should be way earlier than loading barbican | 22:23 |
johnsom | Barbican is optional in both as well | 22:25 |
rm_work | yeah, I run without barbican configured at all | 22:35 |
darkfires | 2017-04-07 18:47:10.329 10341 WARNING neutron.services.provider_configuration [-] The configured driver neutron_lbaas.agent_scheduler.ChanceScheduler has been moved, automatically using neutron_lbaas.agent_scheduler.ChanceScheduler instead. Please update your config files, as this automatic fixup will be removed in a future release. | 22:52 |
darkfires | lol | 22:52 |
johnsom | hmm, probably related to the agent refactor going on in neutron. | 22:54 |
darkfires | DEBUG neutron.services.provider_configuration [-] Service providers = ['LOADBALANCERV2:Haproxy:neutron_lbaas.drivers.haproxy.plugin_driver.HaproxyOnHostPluginDriver:default'] parse_service_provider_opt /usr/lib/python2.7/site-packages/neutron/services/provider_configuration.py:170 WARNING stevedore.named [-] Could not load neutron_lbaas.drivers.haproxy.plugin_driver.HaproxyOnHostPluginDriver | 23:02 |
darkfires | as much as i hate the thought i think i might just end up sticking haproxy inside a vm for the customer instead of messing around with this thing | 23:06 |
*** csomerville has joined #openstack-lbaas | 23:13 | |
*** cody-somerville has quit IRC | 23:16 | |
*** JudeC has joined #openstack-lbaas | 23:23 | |
rm_work | JudeC: o/ | 23:23 |
rm_work | johnsom: JudeC is going to be starting to contribute soon hopefully :) | 23:23 |
JudeC | :D | 23:24 |
johnsom | Hi JudeC Welcome! | 23:34 |
johnsom | Please feel free to reach out to me if you have any questions | 23:35 |
*** armax has joined #openstack-lbaas | 23:38 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!