| *** SumitNaiksatam has quit IRC | 00:01 | |
| rm_work | k got it i think | 00:03 |
|---|---|---|
| rm_work | now the annoying random places that i missed :P | 00:03 |
| rm_work | the REST driver was a little closer to accurate | 00:05 |
| rm_work | submitting a draft | 00:05 |
| rm_work | once all tox finishes | 00:05 |
| *** jorgem has quit IRC | 00:09 | |
| openstackgerrit | Banashankar k proposed openstack/neutron-lbaas: Octavia driver unit tests Added unit tests for the octavia driver. https://review.openstack.org/210644 | 00:11 |
| rm_work | also there's some issues in py34 allowed_address_pairs driver that are being masked by exception handling | 00:14 |
| rm_work | possibly just with the tests | 00:14 |
| openstackgerrit | Adam Harwell proposed openstack/octavia: Correct usage and configuration of CertManager/Generator with Stevedore https://review.openstack.org/212858 | 00:15 |
| rm_work | xgerman: ^^ | 00:15 |
| *** minwang2 has quit IRC | 00:17 | |
| *** madhu_ak has quit IRC | 00:20 | |
| johnsom | rm_work Is the test supposed to be for the ssh driver while changes are to rest driver? | 00:25 |
| rm_work | changes are to both | 00:25 |
| rm_work | but the rest driver was way closer to accurate already | 00:25 |
| johnsom | Oh, ok | 00:25 |
| rm_work | so tests didn't need to change for it | 00:25 |
| rm_work | the old classloader method was replaced with Stevedore | 00:26 |
| rm_work | but never got fully removed | 00:26 |
| *** haigang has joined #openstack-lbaas | 00:26 | |
| rm_work | and ssh driver was just never using it at all | 00:26 |
| *** mlavalle has quit IRC | 00:29 | |
| openstackgerrit | Adam Harwell proposed openstack/octavia: test_plug_vip technically not testing properly due to mock issue in py34 https://review.openstack.org/212868 | 00:46 |
| rm_work | and this is an interesting one ^^ | 00:46 |
| rm_work | randomly caught it while running other tox tests | 00:46 |
| rm_work | name slightly misleading | 00:46 |
| rm_work | wasn't sure how to word it | 00:46 |
| rm_work | damnit and the commit title is too long anyway | 00:48 |
| openstackgerrit | Adam Harwell proposed openstack/octavia: test_plug_vip not testing properly in py34 https://review.openstack.org/212868 | 00:50 |
| rm_work | fixed | 00:50 |
| rm_work | bbl | 00:50 |
| *** haigang has quit IRC | 00:50 | |
| *** diogogmt has joined #openstack-lbaas | 00:52 | |
| *** iwi1 has quit IRC | 00:52 | |
| *** bharath has quit IRC | 00:53 | |
| *** vivek-ebay has quit IRC | 01:05 | |
| rm_you | though, technically i'm ALWAYS here | 01:27 |
| rm_you | so... there's that | 01:27 |
| *** bana_k has quit IRC | 01:34 | |
| *** haigang has joined #openstack-lbaas | 01:39 | |
| *** fnaval has quit IRC | 01:41 | |
| *** minwang2 has joined #openstack-lbaas | 01:44 | |
| *** fnaval has joined #openstack-lbaas | 01:52 | |
| *** crc32 has quit IRC | 01:53 | |
| *** haigang has quit IRC | 01:56 | |
| *** haigang has joined #openstack-lbaas | 01:59 | |
| *** mikeymeitbual has quit IRC | 02:08 | |
| *** KunalGandhi has quit IRC | 02:46 | |
| *** enikanorov2 has quit IRC | 02:55 | |
| *** KunalGandhi has joined #openstack-lbaas | 03:10 | |
| *** crc32 has joined #openstack-lbaas | 03:28 | |
| crc32 | you there blogan? | 03:34 |
| *** mixos has joined #openstack-lbaas | 03:35 | |
| *** mixos has left #openstack-lbaas | 03:36 | |
| *** crc32 has quit IRC | 03:36 | |
| *** crc32 has joined #openstack-lbaas | 03:40 | |
| *** KunalGandhi has quit IRC | 03:43 | |
| *** KunalGandhi has joined #openstack-lbaas | 03:43 | |
| *** crc32 has quit IRC | 03:44 | |
| *** KunalGan_ has joined #openstack-lbaas | 03:52 | |
| *** bharath has joined #openstack-lbaas | 03:54 | |
| *** KunalGandhi has quit IRC | 03:56 | |
| *** bharath has quit IRC | 03:58 | |
| *** diogogmt has quit IRC | 04:00 | |
| *** KunalGan_ has quit IRC | 04:09 | |
| *** enikanorov2 has joined #openstack-lbaas | 04:10 | |
| *** vivek-ebay has joined #openstack-lbaas | 04:19 | |
| *** vivek-ebay has quit IRC | 04:21 | |
| *** vivek-ebay has joined #openstack-lbaas | 04:27 | |
| *** KunalGandhi has joined #openstack-lbaas | 04:29 | |
| *** KunalGandhi has quit IRC | 05:00 | |
| *** numan has joined #openstack-lbaas | 05:01 | |
| *** vivek-ebay has quit IRC | 05:14 | |
| *** SumitNaiksatam has joined #openstack-lbaas | 05:21 | |
| *** vivek-ebay has joined #openstack-lbaas | 05:23 | |
| *** chlong has quit IRC | 05:41 | |
| *** minwang2 has quit IRC | 05:43 | |
| *** vivek-ebay has quit IRC | 05:43 | |
| *** vivek-ebay has joined #openstack-lbaas | 05:44 | |
| *** chlong has joined #openstack-lbaas | 05:44 | |
| *** haigang has quit IRC | 05:45 | |
| *** haigang has joined #openstack-lbaas | 05:46 | |
| *** haigang has quit IRC | 05:47 | |
| *** haigang has joined #openstack-lbaas | 05:47 | |
| *** vivek-ebay has quit IRC | 05:48 | |
| *** minwang2 has joined #openstack-lbaas | 05:48 | |
| *** minwang2 has quit IRC | 06:05 | |
| *** ganeshna has joined #openstack-lbaas | 06:44 | |
| rm_work | blogan: don't suppose... | 06:53 |
| rm_work | actually, anyone else around? | 06:53 |
| *** ganeshna_ has joined #openstack-lbaas | 06:58 | |
| *** ganeshna has quit IRC | 07:00 | |
| *** fnaval has quit IRC | 07:04 | |
| *** numan has quit IRC | 07:17 | |
| *** haigang has quit IRC | 07:21 | |
| *** chlong has quit IRC | 07:36 | |
| *** chlong has joined #openstack-lbaas | 07:39 | |
| *** bharath__ has joined #openstack-lbaas | 07:40 | |
| *** eezhova has quit IRC | 07:52 | |
| *** ganeshna_ has quit IRC | 07:54 | |
| *** ganeshna has joined #openstack-lbaas | 07:55 | |
| *** ganeshna has quit IRC | 08:16 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron-lbaas: Updated from global requirements https://review.openstack.org/212261 | 08:18 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/octavia: Updated from global requirements https://review.openstack.org/212265 | 08:20 |
| *** eezhova has joined #openstack-lbaas | 08:27 | |
| *** numan has joined #openstack-lbaas | 08:33 | |
| *** ganeshna has joined #openstack-lbaas | 08:45 | |
| *** fnaval has joined #openstack-lbaas | 09:05 | |
| *** fnaval has quit IRC | 09:09 | |
| *** ganeshna has quit IRC | 09:11 | |
| *** ganeshna has joined #openstack-lbaas | 09:12 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron-lbaas: Updated from global requirements https://review.openstack.org/212261 | 09:17 |
| *** apuimedo has joined #openstack-lbaas | 09:17 | |
| rm_work | err, so, uhh, I was bored for a bit waiting on some stuff to finish, so I gave this a shot... | 09:45 |
| openstackgerrit | Adam Harwell proposed openstack/octavia: Single Call Create (WIP) https://review.openstack.org/213066 | 09:45 |
| rm_work | going to start a devstack build using that patch and see what happens, rofl | 09:46 |
| rm_work | prolly just go to sleep while that happens... feel free to poke holes in that implementation before I wake up | 09:47 |
| rm_work | hmm, I wonder if anyone else had actually been working on that -- didn't see any patchsets though, so whatev :P | 09:51 |
| rm_work | s/patchsets/CRs/ | 09:51 |
| rm_work | only took about an hour and a half, so no big deal either way | 09:52 |
| *** haigang has joined #openstack-lbaas | 09:56 | |
| *** numan has quit IRC | 10:00 | |
| *** haigang has quit IRC | 10:03 | |
| *** numan has joined #openstack-lbaas | 10:16 | |
| *** numan has quit IRC | 10:20 | |
| openstackgerrit | Adam Harwell proposed openstack/octavia: Single Call Create (WIP) https://review.openstack.org/213066 | 10:27 |
| *** numan has joined #openstack-lbaas | 10:32 | |
| *** numan has quit IRC | 10:40 | |
| rm_work | oh, cool, it works | 10:41 |
| *** bharath__ has quit IRC | 10:41 | |
| rm_work | hurrah :P | 10:41 |
| * rm_work goes to sleep | 10:42 | |
| *** ganeshna has quit IRC | 10:50 | |
| *** ganeshna has joined #openstack-lbaas | 10:51 | |
| *** ganeshna has quit IRC | 10:53 | |
| *** numan has joined #openstack-lbaas | 10:53 | |
| *** ganeshna has joined #openstack-lbaas | 10:53 | |
| *** haigang has joined #openstack-lbaas | 10:55 | |
| *** ganeshna_ has joined #openstack-lbaas | 11:14 | |
| *** ganeshna has quit IRC | 11:14 | |
| *** ganeshna has joined #openstack-lbaas | 11:20 | |
| *** ganeshna_ has quit IRC | 11:22 | |
| *** numan has quit IRC | 11:35 | |
| *** bharath_ has joined #openstack-lbaas | 11:42 | |
| *** bharath_ has quit IRC | 11:47 | |
| *** ganeshna_ has joined #openstack-lbaas | 11:57 | |
| *** ganeshna has quit IRC | 11:59 | |
| *** haigang has quit IRC | 12:05 | |
| *** ganeshna_ has quit IRC | 12:21 | |
| *** diogogmt has joined #openstack-lbaas | 13:05 | |
| *** eezhova has quit IRC | 13:08 | |
| *** eezhova has joined #openstack-lbaas | 13:10 | |
| *** KunalGandhi has joined #openstack-lbaas | 13:12 | |
| *** KunalGan_ has joined #openstack-lbaas | 13:14 | |
| *** iwi1 has joined #openstack-lbaas | 13:17 | |
| *** KunalGandhi has quit IRC | 13:17 | |
| *** diogogmt has quit IRC | 13:19 | |
| *** vivek-ebay has joined #openstack-lbaas | 13:28 | |
| *** bharath has joined #openstack-lbaas | 13:31 | |
| *** bharath has quit IRC | 13:36 | |
| *** fnaval has joined #openstack-lbaas | 13:45 | |
| *** ganeshna_ has joined #openstack-lbaas | 14:12 | |
| *** ganeshn__ has joined #openstack-lbaas | 14:14 | |
| *** ganeshna_ has quit IRC | 14:14 | |
| *** chlong has quit IRC | 14:19 | |
| *** vivek-eb_ has joined #openstack-lbaas | 14:20 | |
| *** vivek-ebay has quit IRC | 14:22 | |
| *** numan has joined #openstack-lbaas | 14:23 | |
| *** fnaval has quit IRC | 14:30 | |
| *** bharath has joined #openstack-lbaas | 14:32 | |
| *** bharath has quit IRC | 14:37 | |
| *** minwang2 has joined #openstack-lbaas | 14:38 | |
| *** alejandrito has joined #openstack-lbaas | 14:44 | |
| *** fnaval has joined #openstack-lbaas | 14:46 | |
| johnsom | You are a wild man.... | 14:46 |
| *** bharath has joined #openstack-lbaas | 14:47 | |
| *** numan has quit IRC | 14:49 | |
| *** fnaval has quit IRC | 14:50 | |
| openstackgerrit | Merged openstack/octavia: Updated from global requirements https://review.openstack.org/212265 | 14:52 |
| *** fnaval has joined #openstack-lbaas | 14:53 | |
| *** diogogmt has joined #openstack-lbaas | 15:01 | |
| *** TrevorV has joined #openstack-lbaas | 15:05 | |
| *** mlavalle has joined #openstack-lbaas | 15:10 | |
| *** iwi1 has quit IRC | 15:12 | |
| *** KunalGan_ has quit IRC | 15:20 | |
| *** madhu_ak has joined #openstack-lbaas | 15:22 | |
| *** ganeshn__ has quit IRC | 15:25 | |
| *** openstackgerrit has quit IRC | 15:31 | |
| *** ganeshna has joined #openstack-lbaas | 15:31 | |
| *** openstackgerrit has joined #openstack-lbaas | 15:32 | |
| *** ganeshna_ has joined #openstack-lbaas | 15:39 | |
| *** ganeshn__ has joined #openstack-lbaas | 15:41 | |
| *** ganeshna has quit IRC | 15:42 | |
| *** ganeshna_ has quit IRC | 15:43 | |
| *** iwi has joined #openstack-lbaas | 15:50 | |
| *** woodster_ has joined #openstack-lbaas | 16:04 | |
| *** KunalGandhi has joined #openstack-lbaas | 16:16 | |
| *** enikanorov2 has quit IRC | 16:20 | |
| *** enikanorov2 has joined #openstack-lbaas | 16:26 | |
| *** SumitNaiksatam has quit IRC | 16:28 | |
| *** SumitNaiksatam has joined #openstack-lbaas | 16:30 | |
| *** madhu_ak has quit IRC | 16:31 | |
| *** madhu_ak has joined #openstack-lbaas | 16:31 | |
| *** minwang2 has quit IRC | 16:37 | |
| *** apuimedo has quit IRC | 16:39 | |
| *** bharath has quit IRC | 16:46 | |
| *** ganeshn__ has quit IRC | 16:54 | |
| kfox1111 | what does lbaasv1 mode https do? since it doesn't support termination? | 16:54 |
| *** madhu_ak_ has joined #openstack-lbaas | 16:55 | |
| *** madhu_ak has quit IRC | 16:57 | |
| *** madhu_ak has joined #openstack-lbaas | 17:04 | |
| *** madhu_ak_ has quit IRC | 17:05 | |
| *** KunalGandhi has quit IRC | 17:07 | |
| johnsom | It creates a TCP configuration for the flows. | 17:08 |
| *** madhu_ak_ has joined #openstack-lbaas | 17:09 | |
| *** madhu_ak has quit IRC | 17:10 | |
| *** crc32 has joined #openstack-lbaas | 17:11 | |
| *** crc32 has quit IRC | 17:13 | |
| *** minwang2 has joined #openstack-lbaas | 17:13 | |
| *** iwi has quit IRC | 17:19 | |
| *** crc32 has joined #openstack-lbaas | 17:27 | |
| *** madhu_ak has joined #openstack-lbaas | 17:32 | |
| *** ajmiller_ has joined #openstack-lbaas | 17:33 | |
| *** madhu_ak_ has quit IRC | 17:35 | |
| *** madhu_ak has quit IRC | 17:35 | |
| *** madhu_ak has joined #openstack-lbaas | 17:36 | |
| *** ajmiller has quit IRC | 17:36 | |
| *** madhu_ak_ has joined #openstack-lbaas | 17:38 | |
| *** madhu_ak has quit IRC | 17:41 | |
| openstackgerrit | Banashankar k proposed openstack/neutron-lbaas: Octavia driver unit tests Added unit tests for the octavia driver. https://review.openstack.org/210644 | 17:46 |
| openstackgerrit | Banashankar k proposed openstack/neutron-lbaas: Adding TLS args to octavia driver https://review.openstack.org/209682 | 17:46 |
| openstackgerrit | Banashankar k proposed openstack/neutron-lbaas: WIP - Octavia driver https://review.openstack.org/174114 | 17:46 |
| *** bana_k has joined #openstack-lbaas | 17:46 | |
| *** bharath has joined #openstack-lbaas | 17:47 | |
| *** bharath has quit IRC | 17:52 | |
| *** bharath has joined #openstack-lbaas | 17:59 | |
| *** madhu_ak_ has quit IRC | 18:01 | |
| TrevorV | johnsom, you around today my man? | 18:05 |
| johnsom | Yep. After my extended standup today I'm going to do a test run with Octavia devstack and update the etherpad. | 18:06 |
| johnsom | Assuming that is what you want to ask about... grin | 18:06 |
| johnsom | So, maybe an hour or so | 18:09 |
| *** woodster_ has quit IRC | 18:10 | |
| TrevorV | Sounds good | 18:12 |
| TrevorV | The failover shouldn't be complete yet | 18:12 |
| TrevorV | I haven't put in the listener/loadbalancer updates | 18:12 |
| TrevorV | Hoping to get more of an idea of what needs done, ya know? | 18:12 |
| johnsom | Yep. That is why I'm going to do an end-to-end run with debug on and do some log parsing to make sure I see all of the tasks being run in normal flows. That should give me something to compare with our list and make sure I don't miss something.... | 18:14 |
| TrevorV | Awesome johnsom that'll be a HUGE help. | 18:14 |
| *** SumitNaiksatam has quit IRC | 18:15 | |
| rm_work | I did update my devstack script to be *current*, so it actually works now | 18:17 |
| rm_work | https://gist.github.com/rm-you/d7fbe613d525f12dc447 | 18:17 |
| rm_work | though might need to tweak which octavia ref it pulls down | 18:18 |
| *** vivek-eb_ has quit IRC | 18:21 | |
| rm_work | johnsom: did you see my random CR from last night for single-call LB? | 18:21 |
| johnsom | Yeah, I commented that you are a wild man.... | 18:21 |
| johnsom | Haven't had a chance to review though | 18:22 |
| rm_work | it ... "works" | 18:23 |
| rm_work | though i didn't test every object cause i am not sure where some of that stuff goes | 18:23 |
| rm_work | like health_monitors or TLS | 18:23 |
| rm_work | those might not be in yet but shouldn't be hard to add if they aren't | 18:23 |
| johnsom | rm_work FYI, there is a octavia/devstack directory that has sample localrc and the plugin.sh script that runs with devstack. We probably should get the barbican stuff in there. | 18:27 |
| johnsom | Also, I have switched to IMAGE_URLS+=",http://download.cirros-cloud.net/0.3.3/cirros-0.3.3-x86_64-disk.img" for cirros | 18:27 |
| johnsom | It's a new location and a much better image | 18:27 |
| rm_work | ok | 18:31 |
| rm_work | updated that | 18:40 |
| TrevorV | rm_work, http://www.dxracer.com/us/en-us/product/1/pc_gaming_chair/king_series/oh-kf00-nw-zero/ | 18:45 |
| crc32 | 2015-08-14 18:47:25.313 | 'pyroute2' is not in global-requirements.txt <-- is this what killed my ./stack.sh run? | 18:56 |
| johnsom | Yes, you need to update your dev stack environment, specifically /opt/stack/requirements | 18:57 |
| johnsom | https://review.openstack.org/#/c/168764/ <-- Merged 8/9/15 | 18:59 |
| crc32 | git pulled | 18:59 |
| TrevorV | johnsom, any luck? Still in progress my man? | 19:06 |
| johnsom | Have the log, parsing now | 19:07 |
| TrevorV | Cool. | 19:07 |
| johnsom | First one will be no spares path | 19:07 |
| TrevorV | Alright | 19:08 |
| *** vivek-ebay has joined #openstack-lbaas | 19:08 | |
| *** crc32 has quit IRC | 19:25 | |
| *** minwang2 has quit IRC | 19:28 | |
| *** ajmiller_ has quit IRC | 19:28 | |
| *** minwang2 has joined #openstack-lbaas | 19:28 | |
| *** ajmiller has joined #openstack-lbaas | 19:29 | |
| *** enikanorov2 has quit IRC | 19:35 | |
| *** enikanorov2 has joined #openstack-lbaas | 19:49 | |
| *** crc32 has joined #openstack-lbaas | 19:50 | |
| TrevorV | johnsom, so that's the logs of a startup for a completely functioning load balancer? | 19:51 |
| johnsom | Yes, through member creation | 19:52 |
| johnsom | I'm pulled into another meeting, but after this ends I'm going to pull that into the failover list as necessary | 19:53 |
| TrevorV | Alright johnsom thanks. I'm not sure how much longer I'll be around. Some family stuffs going on, but hopefully I'll still be around for a couple more hours | 19:54 |
| johnsom | Ok, have a good weekend. I kind of also want to do a diagram off of this, just to have handy | 19:55 |
| xgerman | johnsom there is a way to turn task flow stuff into a diagram FYI | 19:56 |
| xgerman | Have a great weekend TrevorV | 19:56 |
| johnsom | Yep | 19:57 |
| TrevorV | That would be awesome! Could use it in future docs as well to describe work-flow, y eah? | 19:57 |
| TrevorV | yeah*** | 19:57 |
| rm_you | blogan pointed out that single-call should prolly happen mostly post-queue and my impl was pre-queue, so i'm going to start over :P | 20:00 |
| rm_you | though that CR is still ... interesting in an academic sense, i think | 20:00 |
| xgerman | let’s not get too hung up with things we don’t necessarily need for Liberty | 20:01 |
| xgerman | I am not even sure if Neutron will get the “get me a network” done | 20:01 |
| rm_you | yeah, i suppose so -- is there something else I could help with | 20:01 |
| blogan | xgerman: neutron-lbaas willg et the single create callf or sure, just prob not liberty unles rm_you does it | 20:01 |
| rm_you | i might look at the single-call in my off-time tho <_< which is a little weird since WFH my time all blurs together anyway >_< | 20:01 |
| xgerman | blogan agreed | 20:01 |
| xgerman | I am single middle trying to get us to become ref implementation in Liberty | 20:02 |
| xgerman | :-) | 20:02 |
| xgerman | single mindedly | 20:02 |
| rm_you | yeah, what would help with that | 20:02 |
| blogan | rm_you: docs | 20:02 |
| rm_you | lol | 20:02 |
| xgerman | :-) | 20:02 |
| * rm_you backs away slowly | 20:02 | |
| * rm_you backs away more quickly | 20:03 | |
| xgerman | we probably need to do more tests to find holes/bugs | 20:03 |
| * rm_you sprints away | 20:03 | |
| * TrevorV notes rm_you sprinting backwards o_0 | 20:03 | |
| xgerman | it worries me that dougwig’s driver hasn’t merged | 20:03 |
| xgerman | is there something we can do to expedite that | 20:04 |
| xgerman | ? | 20:04 |
| johnsom | Yeah, I am a bit freaked out over the driver | 20:04 |
| dougwig | xgerman: i think we need to get it merged in some form to let folks iterate. | 20:04 |
| dougwig | but are dependent commits and Depends-On not able to fill the gap? | 20:04 |
| *** alejandrito has quit IRC | 20:04 | |
| rm_you | bana_k was working on it AFAIK? | 20:04 |
| rm_you | i saw more patchsets this morning | 20:05 |
| xgerman | yep | 20:05 |
| xgerman | and I like that stuff merged ASAP before we get into the crunch and the gate fails and... | 20:05 |
| rm_you | lol yeah | 20:05 |
| johnsom | bana_k did some unit tests and is working on updating for the TLS stuff | 20:05 |
| rm_you | well, i can't help with that, it's neutron-lbaas | 20:05 |
| rm_you | (can't help with getting it merged, don't have +2) | 20:06 |
| *** vivek-ebay has quit IRC | 20:06 | |
| rm_you | i could help WORK ON IT, but i think bana_k has it covered | 20:06 |
| xgerman | maybe review and heckle pothole and blogan :-) | 20:06 |
| johnsom | rm_work A pass over this would help: https://review.openstack.org/#/c/201882/ | 20:06 |
| johnsom | It's next on my list of things to look at and work on | 20:06 |
| rm_you | ah right | 20:06 |
| rm_you | i have stayed away from that since it still seemed to be in heavy development | 20:07 |
| xgerman | yep, that’s another thing we need to square away quickly | 20:07 |
| rm_you | is it stable? | 20:07 |
| xgerman | only tests will tell | 20:07 |
| blogan | dougwig, xgerman: we can get it merged, but if Liberty releases and it doesn't do proper status updating that would be disappoitning to everyone | 20:07 |
| xgerman | yeah, since rm_work was looking for work — is there something he can do to make that all awesome | 20:08 |
| xgerman | ? | 20:08 |
| johnsom | rm_work I don't know yet. It's been on my back burner a bit. I have signed up to add some config file/data passing to it. | 20:08 |
| blogan | TrevorV: you going to be abelt o work on that piece? or anyone else? | 20:08 |
| xgerman | I am stuck in internal projects so my time budget is unpredictable | 20:08 |
| blogan | rm_you can if he wants to, not sure if he'll start backing away all the way to mexico though | 20:08 |
| blogan | xgerman: same, but i may have to force myself to do this piece | 20:09 |
| xgerman | that’s not too far from your neck of the woods :-) | 20:09 |
| rm_you | xgerman: that has been me, but i'm in a holding pattern right now for internal stuff so i have more time | 20:09 |
| rm_you | but might have to disappear again at a moment's notice <_< | 20:09 |
| blogan | you've already disappeared | 20:09 |
| rm_you | <_< | 20:09 |
| blogan | you're just a ghost in the machine now | 20:09 |
| xgerman | lol | 20:10 |
| rm_you | T_T | 20:10 |
| rm_you | speaking of, afk for the next 45m | 20:10 |
| xgerman | ok, so we need stats in that driver | 20:10 |
| xgerman | mmh | 20:10 |
| rm_you | let me know if you decide on some coding task you need me on | 20:11 |
| rm_you | stats as in... bandwidth/connections/etc? | 20:11 |
| blogan | xgerman: stats in the octavia driver? | 20:11 |
| rm_you | or stats as in "status of member nodes" | 20:11 |
| rm_you | or something else | 20:11 |
| xgerman | ajmiller was trying to make that more fancy but right now it’s just bytes in/out per lb | 20:11 |
| blogan | so both of those are going to be itneresting to accomplish | 20:11 |
| rm_you | i think that's all that is necessary | 20:11 |
| rm_you | strictly speaking | 20:11 |
| rm_you | at least for initial launch | 20:12 |
| blogan | octavia doesn't even have a get stats call yet | 20:12 |
| rm_you | it's just exposing data from haproxy | 20:12 |
| ajmiller | blogan rm_you xgerman Yes, I have finally gotten back to the listener stats patch. | 20:12 |
| xgerman | mmh, I though cc32’s latest UDPm incarnations ends that data | 20:12 |
| rm_you | dunno, i'll read through it later today | 20:13 |
| xgerman | so all we need to do is tally it up somewhere in the DB | 20:13 |
| blogan | yes just exposing the data but thats a bit more involved, whetehr we just ahve an api go directly to the haproxy instances or if we those stored in the db fromt he heartbeats, those arent in place | 20:13 |
| rm_you | do need to afk for 30-45 tho right now | 20:13 |
| ajmiller | I still have fantasies of getting it into Liberty.... | 20:13 |
| *** woodster_ has joined #openstack-lbaas | 20:13 | |
| blogan | xgerman: those stats aren't being inserted as far as i know | 20:13 |
| xgerman | yep, that code is missing | 20:13 |
| xgerman | and also the code to get it out | 20:14 |
| blogan | not even sure we have a table for that | 20:14 |
| blogan | yep | 20:14 |
| blogan | and then member status is going to be a huge deal for neutron-lbaas to get from octavia | 20:14 |
| xgerman | but member status is in the DB | 20:14 |
| blogan | bc neutron-lbaas simply polling octavia's api for all the members is not scalable | 20:14 |
| blogan | xgerman: neutron-lbaas needs to have its db updated with what octavia has | 20:14 |
| xgerman | let’s not overcomplicate for now. If we add some bulk call to get all memebers | 20:15 |
| blogan | well just fyi, no other drivers (except the namespace driver) update the member statuses | 20:16 |
| blogan | in neutron-lbaas | 20:16 |
| blogan | as far as i can remember | 20:16 |
| blogan | but unfortnately, the namespace driver does | 20:16 |
| *** madhu_ak has joined #openstack-lbaas | 20:16 | |
| xgerman | :-( | 20:16 |
| blogan | dougwig: any thoughts on this? | 20:17 |
| blogan | dougwig: specifically, should updating the member's operating status be a requirement to be the ref impl | 20:17 |
| TrevorV | blogan, sorry, were you asking if I was going to do the piece that updates the status up the tree throuh neutron lbaas? | 20:19 |
| TrevorV | I can def do that | 20:19 |
| xgerman | mmh, as bogan said member status won’t scale | 20:20 |
| xgerman | blogan | 20:20 |
| blogan | TrevorV: i mean the piece that makes the octavia commands async | 20:20 |
| blogan | xgerman: separate issue, octavia driver currently just gets a call to create an lb and passes that command to octavia, once octavia responds, the driver sets the provisioning status to ACTIVE, even though its PENDING_CREATE in octavia | 20:21 |
| xgerman | mmh | 20:21 |
| blogan | so a thread needs to be spawned that polls that particular lb until its active or ERROR (or timesout) | 20:21 |
| blogan | i don't tink that will have a scaling issue since its just for one single lb (or listener, pool, etc) and its a short lived thread | 20:22 |
| xgerman | well, can we use our queue to send back updates? | 20:22 |
| blogan | the octavia queue? | 20:22 |
| xgerman | yep | 20:22 |
| xgerman | of course new queue but we could add a bakchannel | 20:23 |
| blogan | we could do something like that, but that requires code in both projects, whereas just the polling requires code in neutron-lbaas | 20:23 |
| xgerman | we also need to move potentially member status and stats | 20:23 |
| blogan | but we will have to do something like that for a scalable solution to get the member statuses | 20:24 |
| xgerman | yep, my thinking as well | 20:24 |
| blogan | xgerman: yeah but i stil think polling for this piece (not the member status) is the better wya to go for now | 20:25 |
| xgerman | yeah, but if we have to create a queue anyway… just saying :-) | 20:25 |
| blogan | true | 20:25 |
| blogan | depends on if we need the memebr status for liberty or not then | 20:26 |
| johnsom | I think it is a must. How can users troubleshoot? | 20:26 |
| blogan | its a must for the product, i'm not convinced its a must for Liberty considering what is already in neutron-lbaas | 20:27 |
| xgerman | +1 | 20:27 |
| xgerman | also one week is short | 20:27 |
| xgerman | actually two weeks | 20:27 |
| blogan | i think we're all spread a little thin too unfortunately | 20:27 |
| xgerman | :-( | 20:28 |
| blogan | if only openstack just paid me | 20:28 |
| blogan | TrevorV: are you already working on this or should i start on it? | 20:29 |
| TrevorV | I don't know that I understand what you said blogan. Isn't the status update the same thing your discussing at this point? | 20:29 |
| TrevorV | Just HOW that happens? | 20:29 |
| xgerman | well, there is the thing where you ask Octavia to fire up a listener, lb , etc and then you fork a thread to make sure it actually happened | 20:30 |
| blogan | TrevorV: 2 issues, all members that octavia has may have their operating status's changed at any time in their lifecycle, so neutron-lbaas needs to be updated on all of those status changes | 20:30 |
| blogan | TrevorV: the issue I'm asking you to complete is the whole driver just setting the provisioning status to ACTIVE when octavia hasn't reported it as ACTIVE | 20:30 |
| TrevorV | Okay, so you're talking about neutron lbaas member health monitoring, but asking me if I'm going to propogate the appropriate status according to Octavia, and then updating accordingly with its actual status | 20:31 |
| TrevorV | ? | 20:31 |
| blogan | TrevorV: just for a single entity, so a user creates an lb through neutron lbaas, neutron lbaas calls octavia to create the lb, neutron-lbaas then returns to the user that the lb is in PENDING_CREATE. meanwhile the octavia drvier spawned a thread that is polling octavia waiting for it to become ACTIVE, and once it does it sets the status in the DB | 20:35 |
| blogan | in the neturon-lbaas DB | 20:35 |
| TrevorV | Yeah, that's what I was talking about the stuff you suggest I work on. Yeah I can do that. | 20:36 |
| TrevorV | Have some more work in the failover stuff, but I can do both | 20:36 |
| TrevorV | I won't be able to really get to it today, I'm on babysitting duty tonight, but I can maybe on Sunday/Monday, yeah? | 20:36 |
| xgerman | sure - sounds good. I am camping this weekend so won’t be off much help | 20:37 |
| *** vivek-ebay has joined #openstack-lbaas | 20:37 | |
| xgerman | will see if I can write some stats code later today so we at least get that into the Octavia DB | 20:38 |
| blogan | TrevorV: i might get ocd on this and attempt to do it, just to see how hard it is, i'll let you know, unless you really want to do it | 20:38 |
| blogan | TrevorV: but it needs to be done ASAP, and you plan on doing it in another review correct? | 20:39 |
| TrevorV | in another review than... what? | 20:39 |
| TrevorV | In a separate review than my delete fix review? | 20:39 |
| *** vivek-eb_ has joined #openstack-lbaas | 20:40 | |
| blogan | yes | 20:40 |
| blogan | if so then we can get it all merged | 20:40 |
| TrevorV | I can do it in the same review, just chalk it up to "driver optimizations" or something. | 20:41 |
| *** vivek-ebay has quit IRC | 20:41 | |
| TrevorV | What do you think? | 20:41 |
| TrevorV | If YOU want to get at it, then by all means | 20:41 |
| TrevorV | You'll probably get to it before I would | 20:42 |
| blogan | TrevorV: i'd rather get the deletes working in as well and not have to wait on this to get done | 20:42 |
| TrevorV | Right, but deletes are still "pending" until the octavia driver goes in | 20:42 |
| blogan | yes but is it pending bc the code is not done? or bc its waiting on the octavia driver? | 20:42 |
| TrevorV | delete fixes are waiting for octavia driver | 20:45 |
| blogan | but the code is good right? | 20:45 |
| TrevorV | I'm saying if that's pending for something specific you might have the appropriate time. | 20:45 |
| TrevorV | Yeah | 20:45 |
| blogan | well i'm talking about just getting the octavia driver merged right now | 20:45 |
| blogan | adn then the deletes | 20:45 |
| blogan | ah but it needs more tests | 20:46 |
| TrevorV | The deletes needs more tests? | 20:46 |
| blogan | no driver | 20:46 |
| blogan | dougwig's | 20:46 |
| blogan | im gonig to work on that right now unless he has | 20:47 |
| TrevorV | Oh yeah, bana_k was doing that I thought | 20:47 |
| blogan | dougwig ? | 20:47 |
| blogan | ohhh | 20:47 |
| blogan | bana_k: around? | 20:47 |
| TrevorV | Check dependent reviews, blogan should be there already | 20:47 |
| blogan | just tls and delete | 20:47 |
| blogan | oh nvm its after tls | 20:48 |
| TrevorV | Gotcha. | 20:48 |
| TrevorV | Yeah, idk | 20:48 |
| TrevorV | hah | 20:48 |
| blogan | oh good, thanks bana_k | 20:48 |
| blogan | not sure this should depend on the tls review | 20:48 |
| blogan | ok time to do a bit of git magic | 20:49 |
| TrevorV | It should if he plans to have unit tests for the tls review as well, right blogan ? | 20:49 |
| blogan | yeah but there arent any | 20:49 |
| TrevorV | Oh. | 20:49 |
| TrevorV | Well. | 20:49 |
| TrevorV | My b | 20:49 |
| TrevorV | Alright, I have to step away | 20:49 |
| blogan | so really this should depend on the octavia driver reveiw, and phil's tls review should have the unit test additions | 20:49 |
| TrevorV | johnsom, I should have some changes up this weekend or Monday, earliest. Sorry I can't get to it ATM :( | 20:49 |
| TrevorV | +1 blogan | 20:50 |
| TrevorV | Have a good weekend all! | 20:50 |
| *** TrevorV has quit IRC | 20:50 | |
| johnsom | TrevorV Sounds good. We are all busy these days. I'm starting on my work for crc32 now. | 20:50 |
| bana_k | blogan | 20:57 |
| bana_k | yes | 20:57 |
| bana_k | sorry was away | 20:57 |
| bana_k | I have unit tests for it. | 20:58 |
| blogan | bana_k: np, thanks for doing that | 20:58 |
| bana_k | I followed the other drivers unit tests cases | 20:58 |
| blogan | bana_k: but i dont think you need to be dependent on the tls reveiw | 20:58 |
| bana_k | and covered most of it. | 20:58 |
| blogan | bana_k: just the Octavia driver | 20:58 |
| bana_k | Please someone take a look at it and let me know if thats all we want or needs some change. | 20:59 |
| blogan | bana_k: quick look over i did, it looks good but ill take a more in depth look here in a sec once i get the root driver review fixed (a pep8 issue) | 20:59 |
| bana_k | Yes I am dependent on the tls changes, as the rest call args changes | 21:00 |
| bana_k | thanks blogan | 21:00 |
| bana_k | kiran took a loot at it and suggested couple of cosmetic changes | 21:01 |
| bana_k | so fixed them | 21:01 |
| rm_work | blogan: honestly i think all of those could be one review, the TLS Additions is an addon to a driver that isn't even merged yet | 21:01 |
| rm_work | why not just include it | 21:01 |
| rm_work | and typically tests are included in the original review, not a sub-review | 21:01 |
| rm_work | so like | 21:01 |
| blogan | rm_work: true | 21:02 |
| rm_work | you could just pull down that chain and --reset down to the first commit, amend it, and submit it all as the original single CR | 21:02 |
| blogan | and trevor's is just 3 line change | 21:02 |
| rm_work | yeah | 21:02 |
| blogan | so 2 reviews can be rolled in | 21:02 |
| blogan | the tests probably could too | 21:02 |
| rm_work | yes | 21:02 |
| blogan | just would inflte the lines but meh it makes sense | 21:03 |
| blogan | bana_k: you ahve any changes in flight? | 21:03 |
| rm_work | also, i am more a fan of a queue backchannel than polling for status updates, but i always choose the "better" solution over "quick" | 21:03 |
| rm_work | even when sometimes we need quick | 21:03 |
| rm_work | but that's still my vote | 21:03 |
| bana_k | no all are committed | 21:03 |
| blogan | rm_work: polling for member status or just the single entity upon a create, update, delete? | 21:03 |
| blogan | bana_k: okay i think i'm going to combien it all into one reveiw and give us all Co-Authorship | 21:04 |
| rm_work | any polling on the neutron-lbaas side | 21:04 |
| bana_k | ok sounds good :) | 21:04 |
| rm_work | really any status update in octavia should trigger adding an event to a queue that could be read by neutron-lbaas | 21:04 |
| rm_work | essentially a queue that is a status change event stream | 21:04 |
| rm_work | and we'd have a consumer on the neutron-lbaas side watching that stream and making updates in the neutron db | 21:05 |
| blogan | rm_work: it is the right way but thats going to be a bit more involved, unless you want to work on that piece and you're sure you get it done soonish | 21:05 |
| rm_work | to reflect the changes | 21:05 |
| rm_work | hmm | 21:05 |
| rm_work | I could look into it | 21:05 |
| blogan | but the fact it requires code in both octavia and neutron-lbaas gives me pause for it being done before liberty | 21:05 |
| rm_work | just need to figure out where to hook in on the octavia side | 21:05 |
| rm_work | we have one week left? O_o | 21:05 |
| blogan | 2 | 21:05 |
| rm_work | yeah might be tough but i think i could do it | 21:05 |
| blogan | well trevor and/or i will do the polling piece as a backup plan | 21:05 |
| rm_work | the curious part is actually not whether i can DO it, but whether we can get it merged <_< | 21:05 |
| blogan | oh yes it can be done, but can it be done well enough in time to be merged | 21:06 |
| rm_work | i'll take that as my task then | 21:06 |
| rm_work | queue for status updates | 21:06 |
| rm_work | k | 21:06 |
| blogan | oslo_messaging hooray! | 21:06 |
| rm_work | i am still going to think of it as an "event stream" | 21:06 |
| rm_work | anywho, brb again | 21:06 |
| blogan | well it'll probably be used for stats as well | 21:06 |
| rm_work | ugh k | 21:07 |
| rm_work | noted | 21:07 |
| rm_work | will plan for that eventuality | 21:07 |
| blogan | great | 21:07 |
| *** jerrygb has joined #openstack-lbaas | 21:24 | |
| bana_k | I am kinda free now. Let me know if there is any work for me | 21:25 |
| *** vivek-eb_ has quit IRC | 21:26 | |
| *** vivek-ebay has joined #openstack-lbaas | 21:26 | |
| *** jerrygb has quit IRC | 21:30 | |
| *** vivek-eb_ has joined #openstack-lbaas | 21:41 | |
| *** vivek-ebay has quit IRC | 21:44 | |
| johnsom | bana_k If you have some time, testing out Octavia on devstack and filing bugs could be handy. I've been working on that some this week too. | 21:53 |
| johnsom | Or there is always docs... grin | 21:53 |
| openstackgerrit | Brandon Logan proposed openstack/neutron-lbaas: Octavia driver https://review.openstack.org/174114 | 22:19 |
| blogan | dougwig, ajmiller, pothole, xgerman, johnsom, rm_work, bana_k: ^^ | 22:22 |
| ajmiller | ack | 22:23 |
| bana_k | I can take devstack testing :P | 22:26 |
| johnsom | ack | 22:27 |
| bana_k | can i use https://gist.github.com/rm-you/d7fbe613d525f12dc447 to launch a devstack ? | 22:28 |
| bana_k | for testing | 22:28 |
| bana_k | blogan : ack | 22:29 |
| johnsom | You can | 22:29 |
| bana_k | ok cool. | 22:29 |
| xgerman | looking | 22:30 |
| johnsom | I'm going to post one with the neutron_lbaas.conf commented line in it | 22:33 |
| bana_k | and on the other hand, whats need to be documented ? | 22:34 |
| openstackgerrit | Michael Johnson proposed openstack/neutron-lbaas: Octavia driver https://review.openstack.org/174114 | 22:36 |
| johnsom | Ok, now looks good to me | 22:37 |
| xgerman | bana_k there is some debrief thing I forgot how to find we need to add some better LBaaS chapter | 22:37 |
| xgerman | johnsom they look identical? or is Friday afternoon catching up with me | 22:39 |
| johnsom | Friday is catching up to you, it has been a long week. I modified neutron_lbaas.conf to have the octavia service string in it commented out | 22:39 |
| xgerman | yep, it is | 22:40 |
| johnsom | Crumb, and looking at it, I put it in the wrong section. | 22:40 |
| johnsom | Just a sec | 22:40 |
| bana_k | oh ok. I can try to take a look at it. | 22:41 |
| openstackgerrit | Michael Johnson proposed openstack/neutron-lbaas: Octavia driver https://review.openstack.org/174114 | 22:41 |
| *** vivek-eb_ has quit IRC | 22:42 | |
| *** vivek-ebay has joined #openstack-lbaas | 22:42 | |
| *** ajmiller_ has joined #openstack-lbaas | 22:45 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia: health manager service https://review.openstack.org/160061 | 22:47 |
| *** ajmiller has quit IRC | 22:49 | |
| openstackgerrit | Pengtao Huang proposed openstack/neutron-lbaas: change if to elif https://review.openstack.org/212560 | 22:49 |
| *** Purandar has joined #openstack-lbaas | 23:04 | |
| openstackgerrit | Al Miller proposed openstack/neutron-lbaas: Filter get_pool_members to return members from the desired pool https://review.openstack.org/210968 | 23:05 |
| Purandar | hello | 23:09 |
| *** purandarkakde has joined #openstack-lbaas | 23:16 | |
| *** Purandar has quit IRC | 23:18 | |
| openstackgerrit | Sherif Abdelwahab proposed openstack/octavia: Amphora Flows and Service Drivers for Active Standby https://review.openstack.org/206252 | 23:19 |
| xgerman | Purandar hi | 23:23 |
| rm_you | bana_k: if you are doing testing for octavia, replace the refss/changes/##/######/# with the one from the patchset you want to test (on the enable_plugin octavia line) | 23:24 |
| rm_you | the one in there now was my not-accurate single-call CR | 23:24 |
| rm_you | which you probably don't care about :P | 23:25 |
| rm_you | (line 16) | 23:25 |
| bana_k | oh ok :). will change that accordingly | 23:25 |
| bana_k | thanks | 23:25 |
| bana_k | so any preference on which patch I should be testing. why to overlap the testing | 23:26 |
| rm_you | you can remove the ref/* | 23:27 |
| rm_you | and just have the line be: | 23:27 |
| rm_you | enable_plugin octavia https://review.openstack.org/openstack/octavia | 23:27 |
| xgerman | rm_you I am a fan of the queue as well so I am happy to do some part of it | 23:27 |
| rm_you | then it'll just test on octavia master | 23:27 |
| rm_you | xgerman: alright, i'm going to start taking a look, but pointers at where to hook in to the existing event flow would be good | 23:28 |
| bana_k | yes thats what we want? | 23:28 |
| xgerman | well, if things are done right we need to hook into the HealthMixin and I need to write a stats mixin | 23:28 |
| rm_you | basically just need to find where to shove a "enqueue_status_event(object_type, object_id, new_status)" call | 23:28 |
| rm_you | and that should be *it* on the octavia side | 23:29 |
| rm_you | pretty simple | 23:29 |
| rm_you | then the neutron-lbaas side needs to have a consumer that watches the queue and does db updates | 23:29 |
| xgerman | yeah, a QueueHealthMixin and QueueStatsMixin | 23:29 |
| xgerman | yep | 23:29 |
| rm_you | I have "mixed feelings" about using a ton of mixins | 23:29 |
| rm_you | but I guess that's alright | 23:29 |
| xgerman | well, that way we can replace them with something else in prod | 23:30 |
| xgerman | (e.g. we will send stats straight into some ceilometer queue IHMO) | 23:30 |
| rm_you | true | 23:30 |
| rm_you | http://www.artima.com/weblogs/viewpost.jsp?thread=246341 | 23:30 |
| rm_you | https://www.artima.com/weblogs/viewpost.jsp?thread=246483 | 23:31 |
| rm_you | part 1 and 2 | 23:31 |
| xgerman | well, we can always make a health and stats driver | 23:32 |
| rm_you | lol | 23:32 |
| rm_you | such driver | 23:32 |
| rm_you | many plugin | 23:32 |
| rm_you | wow | 23:32 |
| rm_you | i honestly don't have a problem either way -- the mixin option is less configurable though I think? | 23:33 |
| rm_you | basically i'd like the code to throw a "status change event" and then have some other code (configurable) decide what to do with it | 23:33 |
| rm_you | so maybe plugin is more apt | 23:33 |
| rm_you | plugin/driver | 23:33 |
| xgerman | so here is the format of the message | 23:34 |
| xgerman | https://review.openstack.org/#/c/201882/29/octavia/controller/healthmanager/heartbeat_udp.py | 23:34 |
| xgerman | I will just use that and put it on the queue for now | 23:34 |
| rm_you | am I looking at the json thing in dorecv? | 23:34 |
| rm_you | ah, are you going to do it? | 23:34 |
| xgerman | crc32 sends that to the controlelr with UDP I turn around and put it on the queue | 23:35 |
| xgerman | yep, I can do that | 23:35 |
| rm_you | xgerman: well again, i'd rather have it call out to some interface | 23:35 |
| rm_you | which then lets us configure what happens to it | 23:35 |
| xgerman | so a mixin is not an interface? | 23:36 |
| rm_you | either sent to queue, or logged, or whatever | 23:36 |
| rm_you | well | 23:36 |
| rm_you | unless I'm mistaken, to switch which option is used, you'd have to edit the code to use the different mixin | 23:36 |
| rm_you | right? | 23:36 |
| rm_you | so like a StatusChangeLogMixin vs. StatusChangeEventQueueMixin | 23:36 |
| xgerman | not necessarily | 23:36 |
| rm_you | you'd have to actually switch which one the code inherits | 23:36 |
| crc32 | johnsom said he would write code to inject an ampphora conf file instead of the status_sender.json. xgerman wants to stop using status_sender.json | 23:36 |
| johnsom | Yep | 23:37 |
| rm_you | err, what? | 23:37 |
| rm_you | i mean | 23:37 |
| crc32 | sorry I heard json thing in dorescv and that you meant /etc/amphora/status_sender.json | 23:38 |
| rm_you | i don't know exactly what you're referring to, but i really don't care what the message looks like -- i just wanted to "handle" (log/queue/whatever) specific events, like status changes | 23:38 |
| xgerman | ok, cool' | 23:38 |
| rm_you | somewhere in here, octavia gets a notification that a status has changed on a member/vip/listener/whatever | 23:38 |
| rm_you | i just want to put a hook there | 23:39 |
| xgerman | yeah, I can handle the hooking | 23:39 |
| rm_you | k | 23:39 |
| rm_you | then let me know when you have a CR? | 23:39 |
| rm_you | were you going to do that TODAY? | 23:39 |
| rm_you | if not, then maybe just let me do it because i will probably be working on it this weekend while you are off camping | 23:40 |
| rm_you | since i kinda want to take Monday off :P | 23:40 |
| xgerman | well, if you write the receiving part I can fill in the sender Monday and we can test it Tuesday ;-) | 23:41 |
| xgerman | I might also be able to do it tonight... | 23:41 |
| rm_you | I'll prolly just do both sides if you don't start today | 23:41 |
| rm_you | lol | 23:41 |
| rm_you | i tend to just get focused and hammer stuff out | 23:41 |
| rm_you | it's fine | 23:41 |
| rm_you | plus i have a vision now :P | 23:42 |
| xgerman | here is how mixing can be loaded dynamically: https://review.openstack.org/#/c/206252/11/octavia/amphorae/drivers/driver_base.py | 23:43 |
| rm_you | loool | 23:43 |
| xgerman | and I hope our visions overlap — aka we either use some driver or mixin to write to the queue | 23:43 |
| rm_you | so we use a mixture of dynamic mixins and stevedore drivers? | 23:43 |
| rm_you | that's ... >_> | 23:43 |
| rm_you | just got finished removing the other class-loading mechanism from the cert stuff | 23:44 |
| rm_you | i don't know if using BOTH is a great idea, but that's fine | 23:44 |
| rm_you | I'll look at both ways, but it'll be really similar either way | 23:44 |
| rm_you | mixins might be "easier"? maybe | 23:44 |
| xgerman | not sure — I speed that stuff out in http://octavia.io/review/master/specs/version0.5/amphora-driver-interface.html | 23:45 |
| xgerman | speced | 23:45 |
| rm_you | alright | 23:45 |
| rm_you | sounds good then | 23:45 |
| xgerman | cool | 23:46 |
| rm_you | not as familiar with this method but i'll figure it out | 23:46 |
| rm_you | do you have a base abstract Mixin and then extend it for the actual Mixins? | 23:46 |
| xgerman | yep | 23:46 |
| rm_you | k | 23:46 |
| rm_you | yeah ok i see that from the VRRP example | 23:47 |
| rm_you | I'll give that a shot this weekend | 23:47 |
| xgerman | If I get to it tonight I will send you an e-mail or something | 23:48 |
| rm_you | i'll be here | 23:49 |
| rm_you | this is my gaming machine :P | 23:49 |
| rm_you | I am always online either here or rm_work because i never am away from a computer for more than about 15 minutes <_< | 23:49 |
| xgerman | get a life! | 23:50 |
| rm_you | lol | 23:50 |
| *** ajmiller_ has quit IRC | 23:54 | |
| *** mixos has joined #openstack-lbaas | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!