*** vivek-ebay has quit IRC | 00:03 | |
*** VijayB has quit IRC | 00:03 | |
*** mlavalle has quit IRC | 00:04 | |
openstackgerrit | Brandon Logan proposed openstack/neutron-lbaas: Fixed bug where session persistence was removed https://review.openstack.org/155169 | 00:10 |
---|---|---|
*** vivek-ebay has joined #openstack-lbaas | 00:10 | |
*** chlong has quit IRC | 00:15 | |
*** VijayB has joined #openstack-lbaas | 00:16 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron-lbaas: Updated from global requirements https://review.openstack.org/160570 | 00:16 |
vivek-ebay | vivek | 00:22 |
ajmiller | blogan dougwig aish I've come across a small problem with the devstack work. I hardcoded the service_provider in neutron_lbaas.conf to point to the synchronous_namespace_driver. I've got a 2-line fix that will choose a better default, and allow it to be overridden in localrc. | 00:23 |
ajmiller | Just a heads-up in case anyone else runs into this. | 00:24 |
Aish | should i make any changes in the doc? | 00:24 |
Aish | ajmiller: ^^ | 00:24 |
ajmiller | Aish, yes, but hold off until I make sure this works. | 00:25 |
Aish | okies.. :) | 00:25 |
*** VijayB has quit IRC | 00:38 | |
*** VijayB has joined #openstack-lbaas | 00:44 | |
*** ajmiller has quit IRC | 00:49 | |
openstackgerrit | Franklin Naval proposed openstack/neutron-lbaas: Neutron LBaaS: Creating Pool API Tests https://review.openstack.org/153001 | 00:58 |
*** vivek-ebay has quit IRC | 01:04 | |
*** jamiem has quit IRC | 01:05 | |
openstackgerrit | Franklin Naval proposed openstack/neutron-lbaas: Neutron LBaaS: Initial Load Balancer Tests https://review.openstack.org/153016 | 01:06 |
openstackgerrit | Franklin Naval proposed openstack/neutron-lbaas: Neutron LBaaS: Creating Pool API Tests https://review.openstack.org/153001 | 01:07 |
*** sbfox1 has quit IRC | 01:14 | |
*** VijayB has quit IRC | 01:22 | |
*** Aish has quit IRC | 01:25 | |
*** sbalukoff has quit IRC | 01:29 | |
*** xgerman has quit IRC | 01:29 | |
johnsom | blogan Octavia question, why is the API calling back to the controller for a load balancer create? I thought that was an API/DB only action as I don't allocate an Amp until a listener create. | 01:32 |
johnsom | When we were at the whiteboard in Seattle we noted that LB create was an API only action. | 01:33 |
*** vivek-ebay has joined #openstack-lbaas | 02:01 | |
blogan | johnsom: ping | 02:29 |
*** amotoki has joined #openstack-lbaas | 02:32 | |
*** Varun_Lodaya has quit IRC | 02:34 | |
*** VijayB has joined #openstack-lbaas | 02:34 | |
*** crc32 has joined #openstack-lbaas | 02:49 | |
*** ajmiller has joined #openstack-lbaas | 02:51 | |
*** Varun_Lodaya has joined #openstack-lbaas | 03:02 | |
*** vivek-eb_ has joined #openstack-lbaas | 03:04 | |
*** vivek-ebay has quit IRC | 03:05 | |
*** madhu has quit IRC | 03:08 | |
*** VijayB has quit IRC | 03:16 | |
openstackgerrit | Phillip Toohill proposed openstack/neutron-lbaas: Return all data for list pools https://review.openstack.org/160530 | 03:24 |
*** blogan_ has joined #openstack-lbaas | 03:26 | |
*** Varun_Lodaya has quit IRC | 03:33 | |
*** blogan_ has quit IRC | 03:34 | |
*** Varun_Lodaya has joined #openstack-lbaas | 03:37 | |
openstackgerrit | Al Miller proposed openstack/neutron-lbaas: Allow LBaaS service provider to be configurable in devstack localrc https://review.openstack.org/160625 | 03:40 |
*** madhu has joined #openstack-lbaas | 03:46 | |
*** amotoki has quit IRC | 04:02 | |
*** crc32 has quit IRC | 04:22 | |
*** sbfox has joined #openstack-lbaas | 04:23 | |
*** sbfox has quit IRC | 04:23 | |
*** madhu has quit IRC | 04:42 | |
*** fnaval has quit IRC | 04:48 | |
*** ajmiller has quit IRC | 04:51 | |
*** ajmiller has joined #openstack-lbaas | 04:52 | |
*** madhu has joined #openstack-lbaas | 04:53 | |
*** sbalukoff has joined #openstack-lbaas | 05:03 | |
*** Aish has joined #openstack-lbaas | 05:07 | |
*** Aish has quit IRC | 05:09 | |
*** rm_you has joined #openstack-lbaas | 05:19 | |
*** rm_you has joined #openstack-lbaas | 05:19 | |
*** fnaval has joined #openstack-lbaas | 05:30 | |
*** blogan_ has joined #openstack-lbaas | 05:34 | |
*** madhu_ has joined #openstack-lbaas | 05:35 | |
*** madhu has quit IRC | 05:39 | |
*** blogan_ has quit IRC | 05:50 | |
*** blogan_ has joined #openstack-lbaas | 05:50 | |
*** vivek-eb_ has quit IRC | 05:51 | |
openstackgerrit | Al Miller proposed openstack/neutron-lbaas: Fix argument to invocation of lbaas agent https://review.openstack.org/160644 | 06:01 |
*** madhu__ has joined #openstack-lbaas | 06:02 | |
*** madhu_ has quit IRC | 06:05 | |
*** sbfox has joined #openstack-lbaas | 06:06 | |
*** rm_you| has joined #openstack-lbaas | 06:17 | |
*** rm_you has quit IRC | 06:19 | |
*** redrobot has quit IRC | 06:21 | |
*** redrobot has joined #openstack-lbaas | 06:25 | |
*** redrobot is now known as Guest32544 | 06:25 | |
*** madhu__ has quit IRC | 06:30 | |
*** clev has quit IRC | 06:39 | |
*** juliancash has quit IRC | 06:49 | |
*** apuimedo has joined #openstack-lbaas | 07:08 | |
openstackgerrit | Phillip Toohill proposed stackforge/octavia: Jinja Haproxy templates https://review.openstack.org/157537 | 07:15 |
openstackgerrit | Phillip Toohill proposed stackforge/octavia: Jinja Haproxy templates https://review.openstack.org/157537 | 07:18 |
*** blogan_ has quit IRC | 07:20 | |
*** woodster_ has quit IRC | 07:20 | |
*** blogan_ has joined #openstack-lbaas | 07:20 | |
*** openstackgerrit has quit IRC | 07:22 | |
*** openstackgerrit has joined #openstack-lbaas | 07:22 | |
*** blogan_ has quit IRC | 07:23 | |
*** blogan_ has joined #openstack-lbaas | 07:24 | |
*** fnaval has quit IRC | 07:26 | |
*** darsh29_ has joined #openstack-lbaas | 07:37 | |
*** jkoelker has quit IRC | 07:38 | |
openstackgerrit | Brandon Logan proposed openstack/neutron-lbaas: Fixed synchronous driver's import redirect https://review.openstack.org/160677 | 07:46 |
*** jkoelker has joined #openstack-lbaas | 07:48 | |
*** Varun_Lodaya has quit IRC | 07:57 | |
*** Varun_Lodaya has joined #openstack-lbaas | 07:58 | |
*** Varun_Lodaya has quit IRC | 07:58 | |
*** Varun_Lodaya has joined #openstack-lbaas | 08:02 | |
*** blogan_ has quit IRC | 08:05 | |
*** Varun_Lodaya has quit IRC | 08:06 | |
openstackgerrit | Merged openstack/neutron-lbaas: Finish Agent Scheduler and LB Agent Driver Piece https://review.openstack.org/156131 | 08:11 |
openstackgerrit | Merged openstack/neutron-lbaas: Listener, Pool, Member, and HealthMonitor Managers https://review.openstack.org/156132 | 08:11 |
openstackgerrit | Phillip Toohill proposed stackforge/octavia: Jinja Haproxy templates https://review.openstack.org/157537 | 08:20 |
*** jkoelker has quit IRC | 08:27 | |
*** darsh29_ has quit IRC | 08:37 | |
*** evgenyf has joined #openstack-lbaas | 08:37 | |
*** BrianShang has joined #openstack-lbaas | 08:44 | |
*** vivek-ebay has joined #openstack-lbaas | 08:48 | |
*** kobis has joined #openstack-lbaas | 08:50 | |
*** vivek-ebay has quit IRC | 08:53 | |
*** sbfox has quit IRC | 09:00 | |
openstackgerrit | Evgeny Fedoruk proposed openstack/neutron-lbaas: TLS capability extension implementation for lbaas v2 https://review.openstack.org/145085 | 09:15 |
*** apuimedo has quit IRC | 09:22 | |
*** apuimedo has joined #openstack-lbaas | 09:24 | |
*** kobis1 has joined #openstack-lbaas | 10:13 | |
*** kobis has quit IRC | 10:13 | |
*** amotoki has joined #openstack-lbaas | 10:34 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron-lbaas: Updated from global requirements https://review.openstack.org/160570 | 10:57 |
*** kobis1 has quit IRC | 12:00 | |
*** kobis has joined #openstack-lbaas | 12:01 | |
*** evgenyf has quit IRC | 12:06 | |
*** evgenyf has joined #openstack-lbaas | 12:11 | |
*** bharath has quit IRC | 13:01 | |
*** jschwarz has joined #openstack-lbaas | 13:25 | |
*** ajmiller has quit IRC | 13:31 | |
*** banix has joined #openstack-lbaas | 13:32 | |
*** banix has quit IRC | 13:34 | |
*** woodster_ has joined #openstack-lbaas | 13:37 | |
*** banix has joined #openstack-lbaas | 13:39 | |
*** jamiem has joined #openstack-lbaas | 14:00 | |
*** chlong has joined #openstack-lbaas | 14:27 | |
*** banix has quit IRC | 14:36 | |
*** banix has joined #openstack-lbaas | 14:37 | |
*** apuimedo has quit IRC | 14:52 | |
*** fnaval has joined #openstack-lbaas | 14:55 | |
*** banix has quit IRC | 14:55 | |
*** clev has joined #openstack-lbaas | 14:57 | |
openstackgerrit | Al Miller proposed openstack/neutron-lbaas: Fix argument to invocation of lbaas agent https://review.openstack.org/160644 | 15:09 |
openstackgerrit | Al Miller proposed openstack/neutron-lbaas: Allow LBaaS service provider to be configurable in devstack localrc https://review.openstack.org/160625 | 15:17 |
johnsom | blogan: pong | 15:45 |
*** fnaval has quit IRC | 15:52 | |
*** sballe__ has joined #openstack-lbaas | 15:53 | |
*** xgerman has joined #openstack-lbaas | 15:53 | |
*** Aish has joined #openstack-lbaas | 15:54 | |
*** madhu has joined #openstack-lbaas | 15:54 | |
*** ajmiller has joined #openstack-lbaas | 15:55 | |
*** banix has joined #openstack-lbaas | 16:05 | |
*** fnaval has joined #openstack-lbaas | 16:06 | |
*** amotoki has quit IRC | 16:16 | |
*** jkoelker has joined #openstack-lbaas | 16:17 | |
*** Varun_Lodaya has joined #openstack-lbaas | 16:20 | |
*** enikanorov has quit IRC | 16:28 | |
*** enikanorov has joined #openstack-lbaas | 16:28 | |
*** Santosh_NS_ has joined #openstack-lbaas | 16:34 | |
*** kobis has left #openstack-lbaas | 16:34 | |
*** blogan_ has joined #openstack-lbaas | 16:35 | |
*** madhu_ has joined #openstack-lbaas | 16:46 | |
*** madhu has quit IRC | 16:48 | |
openstackgerrit | Al Miller proposed openstack/neutron-lbaas: Allow LBaaS service provider to be configurable in devstack https://review.openstack.org/160625 | 16:48 |
enikanorov | blogan: ping | 16:49 |
*** Guest32544 is now known as redrobot | 16:51 | |
blogan_ | enikanorov: hi | 16:51 |
enikanorov | blogan: can you point me to a breif description of the project status? | 16:52 |
enikanorov | have the project switched to api v2, does drivers support that, etc | 16:53 |
blogan_ | ah, i can just tell you | 16:53 |
enikanorov | yep, that also will do :) | 16:53 |
blogan_ | v2 api has been merged, the agent v2 is in the process of merging, tls termination should be merging after that, and there are drivers for v2 that are WIPs right now | 16:54 |
blogan_ | not sure if they will make it for Kilo | 16:54 |
*** madhu has joined #openstack-lbaas | 16:54 | |
enikanorov | so 'agent' means haproxy impl for api v2? | 16:57 |
*** madhu_ has quit IRC | 16:57 | |
dougwig | blogan_: it's function setup_develop { | 16:58 |
*** evgenyf has quit IRC | 16:58 | |
blogan_ | oh same thing | 16:58 |
blogan_ | i mean | 16:58 |
blogan_ | not | 16:58 |
blogan_ | but is | 16:58 |
blogan_ | enikanorov: yes | 16:58 |
dougwig | enikanorov: check the neutron minutes yesterday for a status. | 16:59 |
blogan_ | enikanorov: had to just do a v2 bc a simple shim was not worth the edge cases, hassle, work | 16:59 |
*** Aish has quit IRC | 16:59 | |
*** madhu has quit IRC | 16:59 | |
blogan_ | dougwig: is that function run for every openstack repo? | 16:59 |
dougwig | look in devstack/plugin.sh ! | 17:00 |
dougwig | :) | 17:00 |
blogan_ | oh der | 17:00 |
blogan_ | im stacking from a clean vm, hopefully this reproduces it | 17:01 |
*** kobis has joined #openstack-lbaas | 17:04 | |
*** Varun_Lodaya has quit IRC | 17:06 | |
*** VijayB has joined #openstack-lbaas | 17:08 | |
enikanorov | blogan: thanks! | 17:11 |
*** chlong has quit IRC | 17:17 | |
*** barclaac|2 has joined #openstack-lbaas | 17:19 | |
*** sbfox has joined #openstack-lbaas | 17:22 | |
*** Aish has joined #openstack-lbaas | 17:26 | |
*** madhu has joined #openstack-lbaas | 17:27 | |
*** VijayB has quit IRC | 17:27 | |
*** Varun_Lodaya has joined #openstack-lbaas | 17:36 | |
openstackgerrit | Al Miller proposed openstack/neutron-lbaas: Allow LBaaS service provider to be configurable in devstack https://review.openstack.org/160625 | 17:36 |
*** vivek-ebay has joined #openstack-lbaas | 17:37 | |
*** kobis has quit IRC | 17:38 | |
openstackgerrit | Al Miller proposed openstack/neutron-lbaas: Allow LBaaS service provider to be configurable in devstack https://review.openstack.org/160625 | 17:41 |
ajmiller | ^^ These changes fix the path to the plugin (was too sleepy last night when I did that), and also rebased the patch. | 17:44 |
blogan_ | ajmiller: do believe the removal of lbaas from devstack caused neutron grenade failures | 17:45 |
blogan_ | we we're reverting it | 17:45 |
ajmiller | Ohh | 17:46 |
blogan_ | grenade tests juno upgrading to the current fix | 17:47 |
*** sbfox has quit IRC | 17:49 | |
blogan_ | juno devstack had lbaas in neutron, obvioulsy we are not anymore and the devstack changes removed the setup_develop of neutron-lbaas out of devstack and into our plugin, but grenade's upgrade script doesn't call plugin installs | 17:50 |
blogan_ | ill bbl | 17:51 |
*** blogan_ has quit IRC | 17:51 | |
ajmiller | OK | 17:51 |
dougwig | ajmiller: grenade imports devstack and calls install_neutron, so i think it needs to be taught about our external plugin | 17:54 |
ajmiller | Yes, OK, in hindsight that makes sense. | 17:55 |
ajmiller | Is that something that needs to be done in grenade, or as a neutron devstack hook similar to what we did for q-lbaas flag? | 17:56 |
*** VijayB has joined #openstack-lbaas | 17:58 | |
*** sbfox has joined #openstack-lbaas | 18:01 | |
*** VijayB has quit IRC | 18:03 | |
*** jorgem has joined #openstack-lbaas | 18:14 | |
*** crc32 has joined #openstack-lbaas | 18:20 | |
*** VijayB_ has joined #openstack-lbaas | 18:22 | |
*** jorgem1 has joined #openstack-lbaas | 18:24 | |
*** jorgem has quit IRC | 18:27 | |
openstackgerrit | Al Miller proposed openstack/neutron-lbaas: Fix argument to invocation of lbaas agent https://review.openstack.org/160644 | 18:41 |
blogan | ajmiller: ive been thinking about that too, and i think it can be done both ways, but not sure which one is the right way | 18:45 |
ajmiller | OK. What makes the best sense for quickly getting the tests unstuck? Might there be a short-term and longer-term solution? | 18:46 |
*** jorgem1 has quit IRC | 18:47 | |
*** jorgem has joined #openstack-lbaas | 18:49 | |
*** SumitNaiksatam has joined #openstack-lbaas | 18:50 | |
blogan | revert, which is already up | 18:51 |
*** xgerman has quit IRC | 18:53 | |
*** xgerman has joined #openstack-lbaas | 18:55 | |
*** jschwarz has quit IRC | 19:02 | |
*** jorgem has quit IRC | 19:11 | |
*** ajmiller has quit IRC | 19:23 | |
*** ajmiller has joined #openstack-lbaas | 19:24 | |
*** jorgem has joined #openstack-lbaas | 19:33 | |
*** jorgem has quit IRC | 19:33 | |
*** jorgem has joined #openstack-lbaas | 19:34 | |
*** jorgem has quit IRC | 19:35 | |
*** sbfox has quit IRC | 19:40 | |
*** mlavalle has joined #openstack-lbaas | 19:41 | |
*** sbfox has joined #openstack-lbaas | 19:43 | |
openstackgerrit | Trevor Vardeman proposed stackforge/octavia: Amphora SSH Driver https://review.openstack.org/160964 | 20:16 |
blogan | ajmiller: let me know when you have a chance to discuss getting the agent up and running | 20:16 |
ajmiller | blogan, OK, getting some lunch now, I'll ping again in a few minutes. | 20:18 |
ajmiller | Will be quick lunch | 20:18 |
blogan | okay | 20:19 |
*** Aish has quit IRC | 20:21 | |
openstackgerrit | Trevor Vardeman proposed stackforge/octavia: Moving to novaclient v2 https://review.openstack.org/160969 | 20:25 |
*** sbalukoff has quit IRC | 20:29 | |
*** sbalukoff has joined #openstack-lbaas | 20:31 | |
TrevorV | blogan dougwig if you guys could look this through after gate, we could possibly get this in quick-like | 20:31 |
TrevorV | https://review.openstack.org/160969 | 20:31 |
ajmiller | blogan I'm back | 20:31 |
blogan | okay what issues are you seeing | 20:32 |
ajmiller | blogan I will set things up fresh so I've got a clean slate to work with and ping you back | 20:37 |
blogan | okay sounds good | 20:37 |
TrevorV | xgerman please don't hold back if you'd like to review it as well, its just a quick fix | 20:39 |
xgerman | I asked johnsom to look at it | 20:41 |
xgerman | but it looks fine | 20:41 |
johnsom | TrevorV: Can't we just check for Client and not have to worry with the versions? | 20:46 |
TrevorV | johnsom I don't follow | 20:47 |
*** Aish has joined #openstack-lbaas | 20:47 | |
TrevorV | johnsom you mean just have "novaclient.client.Client" and hope for the best? | 20:48 |
johnsom | Could we just assert that it is an instance of the parent Client and not need to check against v1_1 or v2? | 20:48 |
johnsom | I guess what I don't know is if we require v2 as the code change just changes this from v1_1 to v2. | 20:49 |
TrevorV | Well what was v1_1 is now v2. As in there is no code in the v1_1 directory any longer. | 20:50 |
TrevorV | So I suppose we don't explicitly need v2, but then we might not have explicitly needed v1_1, I just followed what was there previously | 20:50 |
johnsom | Well, it was only a week or two ago it was changed to v1_1. | 20:51 |
johnsom | I see there is a v3 api too. | 20:51 |
TrevorV | Then I have NO idea what I was looking at. | 20:52 |
TrevorV | I also can't say I understand how to get a "versionless" client... | 20:52 |
johnsom | https://review.openstack.org/#/c/157567/ | 20:52 |
johnsom | http://docs.openstack.org/developer/python-novaclient/api.html | 20:53 |
TrevorV | In that documentation you specify a version and build a client. | 20:54 |
johnsom | I think novaclient.v2.client.Client inherits from the parent Client | 20:54 |
TrevorV | I just changed the test to assert "novaclient.client.Client" and it threw a typeError | 20:55 |
openstackgerrit | Ihar Hrachyshka proposed openstack/neutron-lbaas: oslo.log: clean up remaining usages of incubator module https://review.openstack.org/160976 | 20:55 |
TrevorV | novaclient.client.Client isn't a class, type, or tuple of classes and types. | 20:55 |
johnsom | Ok, that answers my question.... | 20:55 |
TrevorV | Oh... ha ha okay. | 20:55 |
TrevorV | Sorry I didn't quite understand... I was lost in most of that :S | 20:55 |
johnsom | Sorry, I probably wasn't clear either | 20:56 |
ajmiller | blogan OK, ready to go. | 20:56 |
blogan | okay, you just did an install of devstack? | 20:57 |
blogan | or stack of devstack | 20:57 |
ajmiller | I am using a devstack set up with commit 3c46916bae of the "Haproxy Namespace Driver" patch tree, plus my recent devstack patches | 20:58 |
blogan | aho kay | 20:58 |
ajmiller | Then ran stack.sh | 20:58 |
ajmiller | The agent is running, but failing with: | 20:59 |
blogan | so neutron-lbaas is at the Haproxy Namespace Driver commit | 20:59 |
blogan | which agent? | 20:59 |
ajmiller | 015-03-03 12:59:00.251 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager File "/opt/stack/neutron-lbaas/neutron_lbaas/services/loadbalancer/drivers/common/agent_driver_base.py", line 70, in get_ready_devices | 20:59 |
ajmiller | 2015-03-03 12:59:00.251 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager agents = self.plugin.get_lbaas_agents(context, | 20:59 |
ajmiller | 2015-03-03 12:59:00.251 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager | 20:59 |
ajmiller | 2015-03-03 12:59:00.251 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager AttributeError: 'LoadBalancerPluginv2' object has no attribute 'get_lbaas_agents' | 20:59 |
ajmiller | [service_providers] | 20:59 |
ajmiller | service_provider = LOADBALANCERV2:Haproxy:neutron_lbaas.services.loadbalancer.drivers.haproxy.plugin_driver.HaproxyOnHostPluginDriver:default | 20:59 |
blogan | should be | 21:00 |
blogan | service_provider = LOADBALANCERV2:Haproxy:neutron_lbaasdrivers.haproxy.plugin_driver.HaproxyOnHostPluginDriver:default | 21:00 |
blogan | neutron_lbaas.drivers | 21:00 |
ajmiller | OK. | 21:00 |
ptoohill | missing a '.' there? | 21:01 |
blogan | service_provider = LOADBALANCERV2:Haproxy:neutron_lbaas.drivers.haproxy.plugin_driver.HaproxyOnHostPluginDriver:default | 21:01 |
blogan | actually that commit should have a line commented out for it | 21:02 |
blogan | but depends on how devstack copies those | 21:02 |
ajmiller | Yes, that is my mistake. | 21:02 |
*** VijayB_ has quit IRC | 21:03 | |
ajmiller | OK, agent is now running. | 21:03 |
ajmiller | So now it ran for a minute, then produced: | 21:04 |
ajmiller | 2015-03-03 13:04:21.842 ERROR neutron_lbaas.services.loadbalancer.agent.agent_manager [-] Unable to retrieve ready devices | 21:04 |
ajmiller | 2015-03-03 13:04:21.842 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager Traceback (most recent call last): | 21:04 |
ajmiller | 2015-03-03 13:04:21.842 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager File "/opt/stack/neutron-lbaas/neutron_lbaas/services/loadbalancer/agent/agent_manager.py", line 152, in sync_state | 21:04 |
ajmiller | 2015-03-03 13:04:21.842 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager ready_instances = set(self.plugin_rpc.get_ready_devices()) | 21:04 |
ajmiller | 2015-03-03 13:04:21.842 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager File "/opt/stack/neutron-lbaas/neutron_lbaas/services/loadbalancer/agent/agent_api.py", line 36, in get_ready_devices | 21:04 |
ajmiller | 2015-03-03 13:04:21.842 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager return cctxt.call(self.context, 'get_ready_devices', host=self.host) | 21:05 |
ajmiller | 2015-03-03 13:04:21.842 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager File "/usr/local/lib/python2.7/dist-packages/oslo_messaging/rpc/client.py", line 156, in call | 21:05 |
ajmiller | 2015-03-03 13:04:21.842 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager retry=self.retry) | 21:05 |
ajmiller | 2015-03-03 13:04:21.842 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager File "/usr/local/lib/python2.7/dist-packages/oslo_messaging/transport.py", line 90, in _send | 21:05 |
ajmiller | 2015-03-03 13:04:21.842 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager timeout=timeout, retry=retry) | 21:05 |
ajmiller | 2015-03-03 13:04:21.842 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager File "/usr/local/lib/python2.7/dist-packages/oslo_messaging/_drivers/amqpdriver.py", line 417, in send | 21:05 |
ajmiller | 2015-03-03 13:04:21.842 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager retry=retry) | 21:05 |
ajmiller | 2015-03-03 13:04:21.842 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager File "/usr/local/lib/python2.7/dist-packages/oslo_messaging/_drivers/amqpdriver.py", line 406, in _send | 21:05 |
ajmiller | 2015-03-03 13:04:21.842 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager result = self._waiter.wait(msg_id, timeout) | 21:05 |
ajmiller | 2015-03-03 13:04:21.842 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager File "/usr/local/lib/python2.7/dist-packages/oslo_messaging/_drivers/amqpdriver.py", line 296, in wait | 21:05 |
ajmiller | 2015-03-03 13:04:21.842 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager reply, ending = self._poll_connection(msg_id, timer) | 21:05 |
ptoohill | could you slap that in gist? | 21:05 |
ptoohill | or paste | 21:05 |
ajmiller | 2015-03-03 13:04:21.842 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager File "/usr/local/lib/python2.7/dist-packages/oslo_messaging/_drivers/amqpdriver.py", line 243, in _poll_connection | 21:05 |
ajmiller | 2015-03-03 13:04:21.842 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager self._raise_timeout_exception(msg_id) | 21:05 |
ajmiller | 2015-03-03 13:04:21.842 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager File "/usr/local/lib/python2.7/dist-packages/oslo_messaging/_drivers/amqpdriver.py", line 212, in _raise_timeout_exception | 21:05 |
ajmiller | 2015-03-03 13:04:21.842 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager _('Timed out waiting for a reply to message ID %s.') % msg_id) | 21:05 |
ajmiller | 2015-03-03 13:04:21.842 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager MessagingTimeout: Timed out waiting for a reply to message ID ed961fd922f44d6daa8c796cf32991c6. | 21:05 |
ajmiller | 2015-03-03 13:04:21.842 TRACE neutron_lbaas.services.loadbalancer.agent.agent_manager | 21:05 |
ajmiller | This is what I remember seeing last night, and I think what Aish saw this morning too. | 21:05 |
ajmiller | Yeah, that's probably better, just a sec. | 21:05 |
ajmiller | https://gist.github.com/allenmiller/2e5f2ea14fe7930ac452 | 21:07 |
ptoohill | this was on a fresh stack? | 21:08 |
ajmiller | yes | 21:08 |
ajmiller | the only thing that wasn't fresh was I made the edit to the service provider above, killed/restarted q-svc and lbaas | 21:11 |
*** apuimedo has joined #openstack-lbaas | 21:12 | |
*** BrianShang_ has joined #openstack-lbaas | 21:13 | |
TrevorV | HOLY OUTPUT ajmiller | 21:14 |
blogan | looks like its got issues connecting to rabbit | 21:14 |
ptoohill | thats what i was thinking, but trying to reproduce | 21:14 |
blogan | ah ajmiller | 21:15 |
blogan | you're still running lbaas v1 agent | 21:15 |
ajmiller | ?!?!?! | 21:15 |
blogan | you need to run neutron-lbaasv2-agent | 21:15 |
ajmiller | Doh! | 21:15 |
blogan | had to keep teh v1 agent around so we dont screw people's existing deploys | 21:15 |
*** BrianShang has quit IRC | 21:15 | |
*** VijayB has joined #openstack-lbaas | 21:15 | |
ajmiller | Rerunning with v2 agent executable now... | 21:17 |
ajmiller | That *might* make a difference ;) | 21:18 |
blogan | just might | 21:18 |
ajmiller | ubuntu@decima-devstack:~/devstack$ neutron lbaas-loadbalancer-create --name lb1 private-subnet | 21:21 |
ajmiller | Created a new loadbalancer: | 21:21 |
ajmiller | +---------------------+--------------------------------------+ | 21:21 |
ajmiller | | Field | Value | | 21:21 |
ajmiller | +---------------------+--------------------------------------+ | 21:21 |
ajmiller | | admin_state_up | True | | 21:21 |
ajmiller | | description | | | 21:21 |
ajmiller | | id | e83fabc7-44ce-4090-96a2-4c961c6d8c38 | | 21:21 |
ajmiller | | listeners | | | 21:21 |
ajmiller | | name | lb1 | | 21:21 |
ajmiller | | operating_status | OFFLINE | | 21:21 |
ajmiller | | provider | haproxy | | 21:21 |
ptoohill | :D | 21:22 |
ajmiller | | provisioning_status | PENDING_CREATE | | 21:22 |
ajmiller | | tenant_id | 93505fe563ed48bb98945931fc53e1fe | | 21:22 |
ajmiller | | vip_address | 10.0.0.3 | | 21:22 |
ajmiller | | vip_port_id | f5501a1b-6577-40f4-ab66-9a0deaad6484 | | 21:22 |
ajmiller | | vip_subnet_id | e5958632-714b-44d4-89ab-9928abf330c2 | | 21:22 |
blogan | did it go ACTIVE/ONLINE? | 21:22 |
ajmiller | +---------------------+--------------------------------------+ | 21:22 |
ajmiller | ubuntu@decima-devstack:~/devstack$ neutron lbaas-loadbalancer-show lb1 | 21:22 |
ajmiller | +---------------------+--------------------------------------+ | 21:22 |
ajmiller | | Field | Value | | 21:22 |
ajmiller | +---------------------+--------------------------------------+ | 21:22 |
ajmiller | | admin_state_up | True | | 21:22 |
ajmiller | | description | | | 21:22 |
ajmiller | | id | e83fabc7-44ce-4090-96a2-4c961c6d8c38 | | 21:22 |
ajmiller | | listeners | | | 21:22 |
ajmiller | | name | lb1 | | 21:22 |
ajmiller | | operating_status | ONLINE | | 21:22 |
ajmiller | | provider | haproxy | | 21:22 |
ajmiller | | provisioning_status | ACTIVE | | 21:22 |
ajmiller | | tenant_id | 93505fe563ed48bb98945931fc53e1fe | | 21:22 |
ajmiller | | vip_address | 10.0.0.3 | | 21:22 |
ajmiller | | vip_port_id | f5501a1b-6577-40f4-ab66-9a0deaad6484 | | 21:22 |
ajmiller | | vip_subnet_id | e5958632-714b-44d4-89ab-9928abf330c2 | | 21:22 |
ajmiller | +---------------------+--------------------------------------+ | 21:22 |
ajmiller | Cool beans! | 21:22 |
ptoohill | yay | 21:23 |
blogan | of course you really dont have a loadbalancer or haproxy process spun up | 21:23 |
blogan | but it did at least go to the agent, and the agent set that statuses | 21:23 |
ajmiller | Right, gotta get some instances and listener running | 21:23 |
blogan | you'll have to add a listener to get the real haproxy process spun up | 21:24 |
ptoohill | and a few magical beans | 21:24 |
TrevorV | beans beans the magical fruit. The more you eat, the more you | 21:28 |
blogan | loot? | 21:29 |
TrevorV | scoot. | 21:29 |
ajmiller | ubuntu@decima-devstack:~/devstack$ curl 10.0.0.3 | 21:30 |
ajmiller | Welcome to 10.0.0.4 | 21:30 |
ajmiller | ubuntu@decima-devstack:~/devstack$ curl 10.0.0.3 | 21:30 |
ajmiller | Welcome to 10.0.0.6 | 21:30 |
ajmiller | ubuntu@decima-devstack:~/devstack$ curl 10.0.0.3 | 21:30 |
ajmiller | Welcome to 10.0.0.4 | 21:30 |
ajmiller | ubuntu@decima-devstack:~/devstack$ curl 10.0.0.3 | 21:30 |
ajmiller | Welcome to 10.0.0.6 | 21:30 |
ajmiller | ubuntu@decima-devstack:~/devstack$ curl 10.0.0.3 | 21:30 |
ajmiller | Welcome to 10.0.0.4 | 21:30 |
ajmiller | ubuntu@decima-devstack:~/devstack$ curl 10.0.0.3 | 21:30 |
ajmiller | Welcome to 10.0.0.6 | 21:30 |
ajmiller | So it works. | 21:30 |
blogan | yay i'm not crazy | 21:31 |
blogan | well i'm nto crazy this one time | 21:32 |
ajmiller | Well, not *that* crazy anyway | 21:32 |
*** juliancash has joined #openstack-lbaas | 21:33 | |
ajmiller | Any recent thoughts on the devstack problem? | 21:33 |
blogan | well luckily devstack still builds without that pack, even with the plugin in tree | 21:33 |
blogan | so that gives us time to figure it out | 21:34 |
blogan | i think it'll have to be a conversation with the devstack folks and grenade folks on the best place to put it | 21:34 |
blogan | my first inclincation is that grenade should be updated to allow plugins | 21:34 |
blogan | liket hat | 21:34 |
blogan | but i could see why someone would want to just put the plugin installation in the install_neutron devstack method | 21:35 |
madhu | fnaval: Just checking we are gathering now? | 21:35 |
*** xgerman has quit IRC | 21:36 | |
ajmiller | OK, good. Seems like a design oversight, and since this plugin architecture is the way forward it seems like grenade should get upgraded.. Meanwhile, I will keep make this v2 agent fix to my patch stream so things should work once we move forward. | 21:36 |
juliancash | \o/ | 21:36 |
*** bharath has joined #openstack-lbaas | 21:36 | |
fnaval | hi madhu - my update: worked on polishing up some api smoke tests: test_load_balancers, test_listeners, test_pools. currently working on polishing up test_load_balancer_basic.py scenario and test_members.py o/ | 21:36 |
blogan | ajmiller: yeah add it as another reveiw dependent on that review your on | 21:37 |
*** vivek-ebay has quit IRC | 21:37 | |
openstackgerrit | Franklin Naval proposed openstack/neutron-lbaas: Neutron LBaaS: Load Balancer Basic Scenario Test https://review.openstack.org/158904 | 21:39 |
openstackgerrit | Franklin Naval proposed openstack/neutron-lbaas: Neutron LBaaS: Load Balancer Basic Scenario Test https://review.openstack.org/158904 | 21:39 |
madhu | fnaval: sure.. I think we may want to sync up with you for the right setup..my update for now is, we are not able to run the tests..erro saying: neutron support is required.. | 21:39 |
fnaval | madhu: i think your set up got messed up somehow | 21:39 |
fnaval | make sure you checkout this PR on your neutron_lbaas opt/stack directory: git fetch https://review.openstack.org/openstack/neutron-lbaas refs/changes/89/158189/10 && git checkout FETCH_HEAD | 21:40 |
fnaval | then make the appropriate changes in the conf files | 21:41 |
fnaval | then restart q-srv | 21:41 |
fnaval | then run the agent | 21:41 |
madhu | yes, I do agree. thinking of creating the new vm and if you could let me know the things that you set up would be helpful. In that way, we finish the testing stuffs, write reviews | 21:41 |
fnaval | its outlined at the end of the test strategy | 21:41 |
madhu | I did follow the docs fnaval, maybe we need to sync up offline? | 21:42 |
openstackgerrit | Phillip Toohill proposed openstack/neutron-lbaas: Haproxy Namespace Driver https://review.openstack.org/158189 | 21:43 |
fnaval | perhaps. or is there someone physically there with you that would be able to get the environment sett up at your office? | 21:43 |
fnaval | someone who has it already working | 21:43 |
fnaval | they can take snapshot of that image and then share with you | 21:44 |
*** bharath_ has joined #openstack-lbaas | 21:44 | |
madhu | yes, I have the devstack (kilo) with LBaaS V2 with the right agent, able to create an instance with its listeners.. its working cool for me | 21:44 |
*** bharath has quit IRC | 21:45 | |
fnaval | hmmm | 21:45 |
madhu | but when I try to run the tests, it is not helping though | 21:45 |
fnaval | did you checkout ptoohill's newest change? | 21:45 |
madhu | yes, I did pull this one: git fetch https://review.openstack.org/openstack/neutron-lbaas refs/changes/89/158189/10 && git checkout FETCH_HEAD | 21:46 |
ptoohill | did you cherry-pick the tests ontop of that commit | 21:46 |
ptoohill | so you can run the agent and the current tests together? | 21:46 |
ptoohill | or vice versa would work also | 21:47 |
ptoohill | i think | 21:47 |
madhu | after getting the latest ptoohills change, I ran the migration, setup.py install, restarted the q-svc screen and finally ran the agent | 21:47 |
madhu | then I got this patch: git fetch https://review.openstack.org/openstack/neutron-lbaas refs/changes/16/153016/17 && git checkout FETCH_HEAD on top it | 21:48 |
ptoohill | think you need to cherry pick that for tests purposes | 21:48 |
madhu | did change the netron_lbaas conf file | 21:48 |
fnaval | yeah, you'll probably need to cherry-pick that | 21:48 |
fnaval | git fetch https://review.openstack.org/openstack/neutron-lbaas refs/changes/16/153016/17 && git cherry-pick FETCH_HEAD | 21:49 |
fnaval | on top of ptoohill's change | 21:49 |
openstackgerrit | Al Miller proposed openstack/neutron-lbaas: Fix invocation of lbaas agent https://review.openstack.org/160644 | 21:49 |
madhu | okay let me try that | 21:50 |
*** xgerman has joined #openstack-lbaas | 21:51 | |
openstackgerrit | Trevor Vardeman proposed stackforge/octavia: Amphora SSH Driver https://review.openstack.org/160964 | 21:55 |
openstackgerrit | Al Miller proposed openstack/neutron-lbaas: Allow LBaaS service provider to be configurable in devstack https://review.openstack.org/160625 | 22:18 |
*** bharath_ has quit IRC | 22:20 | |
openstackgerrit | Joshua Harlow proposed stackforge/octavia: Remove the 2.6 classifier https://review.openstack.org/161008 | 22:28 |
johnsom | ^^^ I am adding that to the agenda tomorrow. TaskFlow wants to stop supporting 2.6 and noticed we have it in our setup.cfg | 22:30 |
blogan | im all for killing 2.6 | 22:31 |
johnsom | +1 | 22:31 |
blogan | did yosemite upgrade to 2.7? | 22:31 |
*** jamiem has quit IRC | 22:31 | |
johnsom | I have no idea | 22:33 |
dougwig | mavericks was 2.7 | 22:33 |
blogan | shows how out of date i am with os x | 22:34 |
dougwig | only reason to support 2.6 is for icehouse. | 22:34 |
dougwig | rhel6. | 22:34 |
dougwig | or juno. | 22:34 |
*** vivek-ebay has joined #openstack-lbaas | 22:38 | |
openstackgerrit | Merged stackforge/octavia: Moving to novaclient v2 https://review.openstack.org/160969 | 22:38 |
blogan | i thought neutron killed 2.6 | 22:38 |
blogan | so why wouldn't we follow suit | 22:38 |
*** banix has quit IRC | 22:39 | |
dougwig | kilo killed 2.6, because rhel6 was dropped. | 22:41 |
dougwig | i think we should kill it. i'm surprised it's not already. | 22:41 |
xgerman | +1 | 22:42 |
blogan | finally those jobs started on that revert review | 22:42 |
*** vivek-ebay has quit IRC | 22:42 | |
openstackgerrit | Franklin Naval proposed openstack/neutron-lbaas: Neutron LBaaS: Adding test cases for Members. https://review.openstack.org/154237 | 22:42 |
dougwig | yell at xgerman. it's apparently an issue with hpcloud that stalled the gate. | 22:42 |
dougwig | :) | 22:42 |
blogan | ohhhh | 22:42 |
blogan | wouldn't have happened with raxcloud | 22:43 |
xgerman | I can't believe that | 22:43 |
xgerman | our cloud has been good today :-) | 22:43 |
blogan | sounds like it rained | 22:45 |
blogan | ... | 22:45 |
blogan | get it | 22:45 |
blogan | :( | 22:45 |
dougwig | sigh. | 22:45 |
dougwig | towgan tried to make a funny. | 22:45 |
dougwig | my boy band joke was better. and that is a low low bar. | 22:45 |
blogan | except yours was in front of a lot of people | 22:47 |
blogan | so your failure was much more crashy and burny | 22:47 |
dougwig | heh, it was meant to fail. the joke was the follow-up, which *did* work. | 22:47 |
blogan | what was the follow up? | 22:47 |
blogan | next talk we do, we should do it informercial style. you can do the voiceover like the informercial "have you ever tried load balancing with v1" and then i could be up there banging on a server with a screw driver and it bursts into flames | 22:49 |
xgerman | lol | 22:50 |
xgerman | did we even get accepted | 22:50 |
blogan | im not sure if they've announced which ones did yet | 22:50 |
openstackgerrit | Franklin Naval proposed openstack/neutron-lbaas: Neutron LBaaS: Adding test cases for Members. https://review.openstack.org/154237 | 22:50 |
xgerman | well, if we know unofficially i would be happyt know, too | 22:50 |
blogan | well i officially and unofficially do not nkow | 22:51 |
*** openstackgerrit has quit IRC | 22:51 | |
*** openstackgerrit has joined #openstack-lbaas | 22:52 | |
openstackgerrit | Phillip Toohill proposed openstack/neutron-lbaas: Added agent callbacks https://review.openstack.org/156133 | 22:53 |
openstackgerrit | Franklin Naval proposed openstack/neutron-lbaas: Neutron LBaaS: Creating Pool API Tests https://review.openstack.org/153001 | 22:55 |
openstackgerrit | Phillip Toohill proposed openstack/neutron-lbaas: Agent, device driver interface, and callback API https://review.openstack.org/152666 | 22:56 |
openstackgerrit | Phillip Toohill proposed openstack/neutron-lbaas: Completion of Agent Manager https://review.openstack.org/158188 | 22:56 |
*** vivek-ebay has joined #openstack-lbaas | 22:59 | |
openstackgerrit | Phillip Toohill proposed openstack/neutron-lbaas: Completion of Agent Manager https://review.openstack.org/158188 | 23:00 |
openstackgerrit | Phillip Toohill proposed openstack/neutron-lbaas: Haproxy Namespace Driver https://review.openstack.org/158189 | 23:00 |
*** VijayB has quit IRC | 23:06 | |
*** VijayB_ has joined #openstack-lbaas | 23:08 | |
*** vivek-ebay has quit IRC | 23:09 | |
*** vivek-ebay has joined #openstack-lbaas | 23:09 | |
xgerman | fnaval I am trying to understand some of the tests - you guys say import tempest but the tempest-lib docs at http://docs.openstack.org/developer/tempest-lib/usage.html#rest-client suggest to write import tempest_lib | 23:26 |
xgerman | now I am confused why we put something into the requirements but use something else | 23:27 |
Varun_Lodaya | Hey Guys, I was trying out some features, had a question, is there any plan to add a backup member to the pool? | 23:27 |
Varun_Lodaya | Today there is no way to move existing connections to a different server is a member dies | 23:28 |
dougwig | Varun_Lodaya: no current plans, but anything is on the table for consideration. please add your ideas here: https://etherpad.openstack.org/p/lbaas-vancouver-planning | 23:28 |
Varun_Lodaya | ok cool | 23:28 |
blogan | Varun_Lodaya: none discussed, but we have that feature in our current CLB offering at Rax, so getting it in would be fine with me, the details of it would need to be discussed | 23:29 |
Varun_Lodaya | thanks @dougwig | 23:29 |
fnaval | xgerman: i think i followed the pattern that is established in the tempest tests | 23:29 |
fnaval | or the ones in the neutron repo | 23:29 |
fnaval | just to be consistent with their import scheme | 23:30 |
fnaval | as we started the testing in the hackathon, i dont think i readily followed the tempest documentation; so there may be some tech debt there. | 23:31 |
*** chlong has joined #openstack-lbaas | 23:31 | |
fnaval | we just wanted to build something that worked - basically hacked existing code that was already working | 23:31 |
xgerman | mmh, I think neutron is in the process of moving their tests in tree -- so they might not have made the switch to tempest_lib | 23:31 |
fnaval | Hopefully someone later will find the correct way to do it | 23:32 |
xgerman | yep, so I think as it is written the neutron-lbaas tests need to be copied into the tempest directory on devstack | 23:32 |
fnaval | well, if it's 1.0 tests, they probably wont work since they're using totally different client | 23:34 |
fnaval | the api has definitely changed | 23:34 |
xgerman | as it looks to me if you import tempest you need to be in tree with the tempest project | 23:35 |
xgerman | /opt/stack/tempest/tempest/api/network | 23:35 |
fnaval | unless you have it already installed as a pip package | 23:36 |
fnaval | you can check if tempest is installed by doing "pip freeze" | 23:36 |
xgerman | so you install tempest instead of tempest-lib? | 23:37 |
fnaval | on the box that i'm running devstack on, i see both | 23:37 |
fnaval | oh wait | 23:37 |
fnaval | sorry | 23:37 |
fnaval | i see tempest-lib | 23:37 |
fnaval | tempest-lib==0.1.0 | 23:37 |
xgerman | ah | 23:37 |
fnaval | and i see this: -e git://git.openstack.org/openstack/tempest.git@17f81d4cc83569438fe11b1ee6ee2afe74c0b501#egg=tempest-master | 23:38 |
xgerman | tempest-lib==0.3.0 | 23:38 |
xgerman | ok, fnaval the git thing gets you around having to use import tempest_lib | 23:38 |
fnaval | xgerman: here's what my pip freeze looks like on my devstack set up: https://gist.github.com/fnaval/120231a1c5a3bd09afee | 23:39 |
fnaval | there's a number of git -e 's in there | 23:39 |
fnaval | bbiab | 23:40 |
xgerman | understood - i am just pondering if we should switch to tempest-lib or keep bringing int he main tempest | 23:40 |
xgerman | probably a question for wiser man like nlogan and dougwig - though I lean to swicthign to tempest-lib | 23:40 |
xgerman | blogan | 23:41 |
*** VijayB_ has joined #openstack-lbaas | 23:42 | |
*** VijayB_ has quit IRC | 23:43 | |
dougwig | xgerman, fnaval - sounds like a question for #openstack-qa to me. | 23:46 |
xgerman | well, it seems we set our project up for tempest-lib by putting it into test-requirements... but qa is a good place to make sure that was the right choice | 23:48 |
xgerman | asking now | 23:50 |
openstackgerrit | min wang proposed stackforge/octavia: Create database models for healthmanager https://review.openstack.org/149802 | 23:50 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!