| *** fnaval has quit IRC | 00:04 | |
| *** diltram has quit IRC | 00:09 | |
| *** diltram_ has joined #openstack-lbaas | 00:09 | |
| openstackgerrit | Merged openstack/neutron-lbaas: Fix a typo https://review.openstack.org/399311 | 00:35 |
|---|---|---|
| *** yamamoto_ has joined #openstack-lbaas | 00:56 | |
| *** kbyrne has quit IRC | 00:57 | |
| *** armax has quit IRC | 00:59 | |
| *** yuanying has quit IRC | 01:00 | |
| *** kbyrne has joined #openstack-lbaas | 01:01 | |
| *** yuanying has joined #openstack-lbaas | 01:03 | |
| *** bana_k has quit IRC | 01:08 | |
| *** bana_k has joined #openstack-lbaas | 01:11 | |
| *** agarner has quit IRC | 01:13 | |
| *** robcresswell has quit IRC | 01:13 | |
| *** blogan_ has quit IRC | 01:13 | |
| *** HenryG has quit IRC | 01:13 | |
| *** basilAB has quit IRC | 01:13 | |
| *** greghaynes has quit IRC | 01:13 | |
| *** bedis has quit IRC | 01:13 | |
| *** dasanind has quit IRC | 01:13 | |
| *** Kdecherf has quit IRC | 01:13 | |
| *** mugsie_ has quit IRC | 01:13 | |
| *** sbalukoff has quit IRC | 01:13 | |
| *** blogan has joined #openstack-lbaas | 01:14 | |
| *** bedis_ has joined #openstack-lbaas | 01:14 | |
| *** aleph1 has joined #openstack-lbaas | 01:14 | |
| *** mugsie_ has joined #openstack-lbaas | 01:14 | |
| rm_work | johnsom: ugh later than i thought | 01:14 |
| *** sbalukoff has joined #openstack-lbaas | 01:14 | |
| rm_work | johnsom: if you need to run i don't blame you, otherwise i'm here | 01:14 |
| *** mugsie_ has quit IRC | 01:14 | |
| *** mugsie_ has joined #openstack-lbaas | 01:14 | |
| *** greghaynes has joined #openstack-lbaas | 01:14 | |
| *** Kdecherf has joined #openstack-lbaas | 01:14 | |
| *** HenryG has joined #openstack-lbaas | 01:15 | |
| *** robcresswell has joined #openstack-lbaas | 01:19 | |
| johnsom | Yeah, getting close to date night time. Tomorrow? | 01:20 |
| rm_work | kk | 01:20 |
| johnsom | Ugh, net split... I can see irccloud queuing my posts | 01:20 |
| *** diltram_ has quit IRC | 01:23 | |
| *** diltram has joined #openstack-lbaas | 01:24 | |
| *** dasanind has joined #openstack-lbaas | 01:27 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia: Add quota support to Octavia https://review.openstack.org/360794 | 01:30 |
| johnsom | Darn it, didn't get all the tests done I had hoped. Really good coverage on the quota checks now. I will have to pick it up in the morning. | 01:32 |
| *** ducttape_ has joined #openstack-lbaas | 01:32 | |
| *** bana_k has quit IRC | 01:36 | |
| *** kevo has quit IRC | 01:37 | |
| openstackgerrit | JingLiu proposed openstack/neutron-lbaas: Set access_policy for messaging's dispatcher https://review.openstack.org/415805 | 02:08 |
| *** ducttape_ has quit IRC | 02:08 | |
| *** ipsecguy has quit IRC | 02:09 | |
| *** madhu_ak has joined #openstack-lbaas | 02:15 | |
| *** catintheroof has joined #openstack-lbaas | 02:22 | |
| *** gongysh has joined #openstack-lbaas | 02:25 | |
| *** ipsecguy has joined #openstack-lbaas | 02:25 | |
| *** ducttape_ has joined #openstack-lbaas | 02:28 | |
| *** yamamoto_ has quit IRC | 02:41 | |
| *** ducttape_ has quit IRC | 02:48 | |
| *** catintheroof has quit IRC | 02:50 | |
| *** yuanying has quit IRC | 03:01 | |
| *** ducttape_ has joined #openstack-lbaas | 03:02 | |
| *** yuanying has joined #openstack-lbaas | 03:06 | |
| *** ducttape_ has quit IRC | 03:37 | |
| *** yamamoto_ has joined #openstack-lbaas | 03:45 | |
| *** amotoki has joined #openstack-lbaas | 03:50 | |
| *** yuanying has quit IRC | 03:56 | |
| *** yuanying has joined #openstack-lbaas | 03:56 | |
| *** yuanying has quit IRC | 04:01 | |
| openstackgerrit | Joe Mills proposed openstack/neutron-lbaas: scenario test: Open up port for second listener https://review.openstack.org/417277 | 04:02 |
| *** saju_m has joined #openstack-lbaas | 04:03 | |
| *** gongysh has quit IRC | 04:09 | |
| *** cody-somerville has joined #openstack-lbaas | 04:13 | |
| *** csomerville has quit IRC | 04:15 | |
| *** madhu_ak has quit IRC | 04:28 | |
| *** links has joined #openstack-lbaas | 04:40 | |
| *** ducttape_ has joined #openstack-lbaas | 04:45 | |
| *** gongysh has joined #openstack-lbaas | 04:52 | |
| *** yuanying has joined #openstack-lbaas | 04:57 | |
| *** ducttape_ has quit IRC | 04:59 | |
| *** saju_m has quit IRC | 05:45 | |
| *** bana_k has joined #openstack-lbaas | 06:15 | |
| *** gongysh has quit IRC | 06:15 | |
| *** reedip_ has joined #openstack-lbaas | 06:15 | |
| *** amotoki has quit IRC | 06:25 | |
| *** Alex_Stef has joined #openstack-lbaas | 06:33 | |
| *** gongysh has joined #openstack-lbaas | 06:42 | |
| *** bana_k has quit IRC | 06:52 | |
| *** rcernin has quit IRC | 06:55 | |
| *** ducttape_ has joined #openstack-lbaas | 07:00 | |
| *** gcheresh_ has joined #openstack-lbaas | 07:01 | |
| *** reedip_ has quit IRC | 07:01 | |
| *** ducttape_ has quit IRC | 07:05 | |
| *** gcheresh_ has quit IRC | 07:07 | |
| *** tesseract has joined #openstack-lbaas | 07:09 | |
| *** pcaruana has joined #openstack-lbaas | 07:10 | |
| *** rcernin has joined #openstack-lbaas | 07:13 | |
| *** kobis has joined #openstack-lbaas | 07:18 | |
| *** nmagnezi_ has joined #openstack-lbaas | 07:34 | |
| *** kobis has quit IRC | 07:47 | |
| *** yuanying has quit IRC | 07:51 | |
| *** Alex_Stef has quit IRC | 08:06 | |
| *** kevo has joined #openstack-lbaas | 08:14 | |
| *** mjblack has quit IRC | 08:24 | |
| *** robcresswell has quit IRC | 08:31 | |
| *** greghaynes has quit IRC | 08:31 | |
| *** harlowja has quit IRC | 08:31 | |
| *** matt-borland has quit IRC | 08:31 | |
| *** links has quit IRC | 08:31 | |
| *** ipsecguy has quit IRC | 08:31 | |
| *** diltram has quit IRC | 08:31 | |
| *** bedis_ has quit IRC | 08:31 | |
| *** mhayden has quit IRC | 08:31 | |
| *** openstackgerrit has quit IRC | 08:31 | |
| *** Kdecherf has quit IRC | 08:31 | |
| *** mugsie_ has quit IRC | 08:31 | |
| *** adam_g has quit IRC | 08:31 | |
| *** cody-somerville has quit IRC | 08:31 | |
| *** kbyrne has quit IRC | 08:31 | |
| *** reedip has quit IRC | 08:31 | |
| *** kong has quit IRC | 08:31 | |
| *** Kiall has quit IRC | 08:31 | |
| *** dasanind has quit IRC | 08:31 | |
| *** sbalukoff has quit IRC | 08:31 | |
| *** rm_work has quit IRC | 08:31 | |
| *** BlackDex has quit IRC | 08:31 | |
| *** kevo has quit IRC | 08:31 | |
| *** rcernin has quit IRC | 08:31 | |
| *** yamamoto_ has quit IRC | 08:31 | |
| *** HenryG has quit IRC | 08:31 | |
| *** blogan has quit IRC | 08:31 | |
| *** jschwarz has quit IRC | 08:31 | |
| *** nmagnezi_ has quit IRC | 08:31 | |
| *** pcaruana has quit IRC | 08:31 | |
| *** tesseract has quit IRC | 08:31 | |
| *** aleph1 has quit IRC | 08:31 | |
| *** kevinbenton has quit IRC | 08:31 | |
| *** ajo has quit IRC | 08:31 | |
| *** jidar has quit IRC | 08:31 | |
| *** ankur-gupta-f has quit IRC | 08:31 | |
| *** pck has quit IRC | 08:31 | |
| *** mestery has quit IRC | 08:31 | |
| *** sindhu has quit IRC | 08:34 | |
| *** Kiall has joined #openstack-lbaas | 08:37 | |
| *** kevo has joined #openstack-lbaas | 08:37 | |
| *** mjblack has joined #openstack-lbaas | 08:37 | |
| *** cody-somerville has joined #openstack-lbaas | 08:37 | |
| *** nmagnezi_ has joined #openstack-lbaas | 08:37 | |
| *** rcernin has joined #openstack-lbaas | 08:37 | |
| *** pcaruana has joined #openstack-lbaas | 08:37 | |
| *** tesseract has joined #openstack-lbaas | 08:37 | |
| *** links has joined #openstack-lbaas | 08:37 | |
| *** ipsecguy has joined #openstack-lbaas | 08:37 | |
| *** dasanind has joined #openstack-lbaas | 08:37 | |
| *** diltram has joined #openstack-lbaas | 08:37 | |
| *** robcresswell has joined #openstack-lbaas | 08:37 | |
| *** HenryG has joined #openstack-lbaas | 08:37 | |
| *** Kdecherf has joined #openstack-lbaas | 08:37 | |
| *** greghaynes has joined #openstack-lbaas | 08:37 | |
| *** sbalukoff has joined #openstack-lbaas | 08:37 | |
| *** mugsie_ has joined #openstack-lbaas | 08:37 | |
| *** aleph1 has joined #openstack-lbaas | 08:37 | |
| *** bedis_ has joined #openstack-lbaas | 08:37 | |
| *** blogan has joined #openstack-lbaas | 08:37 | |
| *** harlowja has joined #openstack-lbaas | 08:37 | |
| *** matt-borland has joined #openstack-lbaas | 08:37 | |
| *** ankur-gupta-f has joined #openstack-lbaas | 08:37 | |
| *** mhayden has joined #openstack-lbaas | 08:37 | |
| *** openstackgerrit has joined #openstack-lbaas | 08:37 | |
| *** kevinbenton has joined #openstack-lbaas | 08:37 | |
| *** adam_g has joined #openstack-lbaas | 08:37 | |
| *** jschwarz has joined #openstack-lbaas | 08:37 | |
| *** pck has joined #openstack-lbaas | 08:37 | |
| *** reedip has joined #openstack-lbaas | 08:37 | |
| *** BlackDex has joined #openstack-lbaas | 08:37 | |
| *** kong has joined #openstack-lbaas | 08:37 | |
| *** rm_work has joined #openstack-lbaas | 08:37 | |
| *** ajo has joined #openstack-lbaas | 08:37 | |
| *** mestery has joined #openstack-lbaas | 08:37 | |
| *** jidar has joined #openstack-lbaas | 08:37 | |
| *** Kiall is now known as Guest87200 | 08:39 | |
| *** kbyrne has joined #openstack-lbaas | 08:40 | |
| *** ctracey has quit IRC | 08:40 | |
| *** jsheeren has joined #openstack-lbaas | 08:42 | |
| *** yamamoto has joined #openstack-lbaas | 08:43 | |
| *** sindhu has joined #openstack-lbaas | 08:51 | |
| *** Alex_Stef has joined #openstack-lbaas | 08:58 | |
| *** ducttape_ has joined #openstack-lbaas | 09:01 | |
| *** ctracey has joined #openstack-lbaas | 09:02 | |
| *** ducttape_ has quit IRC | 09:06 | |
| *** armax has joined #openstack-lbaas | 09:06 | |
| *** gcheresh_ has joined #openstack-lbaas | 09:08 | |
| *** nmagnezi_ has quit IRC | 09:20 | |
| *** yuanying has joined #openstack-lbaas | 09:24 | |
| *** gcheresh_ has quit IRC | 09:42 | |
| *** Alex_Stef has quit IRC | 09:56 | |
| *** pck has quit IRC | 09:56 | |
| *** pck has joined #openstack-lbaas | 09:56 | |
| *** pck_ has joined #openstack-lbaas | 10:06 | |
| *** pck has quit IRC | 10:06 | |
| *** amotoki has joined #openstack-lbaas | 10:15 | |
| *** pck has joined #openstack-lbaas | 10:16 | |
| *** pck_ has quit IRC | 10:16 | |
| *** kobis has joined #openstack-lbaas | 10:50 | |
| *** kobis has quit IRC | 10:54 | |
| *** amotoki has quit IRC | 10:57 | |
| *** ducttape_ has joined #openstack-lbaas | 11:02 | |
| *** ducttape_ has quit IRC | 11:06 | |
| *** yamamoto has quit IRC | 11:13 | |
| *** mugsie_ has quit IRC | 11:18 | |
| *** kevo has quit IRC | 11:19 | |
| *** amotoki has joined #openstack-lbaas | 11:20 | |
| *** mugsie_ has joined #openstack-lbaas | 11:25 | |
| *** gongysh has quit IRC | 11:29 | |
| *** gongysh has joined #openstack-lbaas | 11:30 | |
| *** gongysh has quit IRC | 11:36 | |
| *** amotoki has quit IRC | 11:50 | |
| *** amotoki has joined #openstack-lbaas | 11:52 | |
| *** kobis has joined #openstack-lbaas | 11:57 | |
| *** kobis has quit IRC | 12:06 | |
| *** amotoki has quit IRC | 12:06 | |
| *** kobis has joined #openstack-lbaas | 12:09 | |
| *** amotoki has joined #openstack-lbaas | 12:13 | |
| *** Alex_Stef has joined #openstack-lbaas | 12:18 | |
| *** kobis has quit IRC | 12:18 | |
| *** amotoki has quit IRC | 12:28 | |
| *** catintheroof has joined #openstack-lbaas | 12:32 | |
| *** catintheroof has quit IRC | 12:32 | |
| *** catintheroof has joined #openstack-lbaas | 12:32 | |
| *** ducttape_ has joined #openstack-lbaas | 12:33 | |
| *** amotoki has joined #openstack-lbaas | 12:35 | |
| *** catintheroof has quit IRC | 12:36 | |
| *** catintheroof has joined #openstack-lbaas | 12:37 | |
| *** ducttape_ has quit IRC | 12:38 | |
| *** amotoki has quit IRC | 13:03 | |
| *** amotoki has joined #openstack-lbaas | 13:05 | |
| *** amotoki has quit IRC | 13:12 | |
| *** amotoki has joined #openstack-lbaas | 13:15 | |
| *** Alex_Stef has quit IRC | 13:15 | |
| *** yamamoto has joined #openstack-lbaas | 13:15 | |
| *** amotoki has quit IRC | 13:16 | |
| *** anilvenkata has joined #openstack-lbaas | 13:27 | |
| *** ducttape_ has joined #openstack-lbaas | 13:47 | |
| *** jsheeren has quit IRC | 13:49 | |
| *** amotoki has joined #openstack-lbaas | 13:54 | |
| *** beardedeagle has joined #openstack-lbaas | 13:57 | |
| *** yamamoto has quit IRC | 13:57 | |
| *** anilvenkata has quit IRC | 13:59 | |
| *** anilvenkata has joined #openstack-lbaas | 14:00 | |
| *** chlong has joined #openstack-lbaas | 14:02 | |
| *** anilvenkata has quit IRC | 14:02 | |
| *** anilvenkata has joined #openstack-lbaas | 14:02 | |
| *** amotoki has quit IRC | 14:10 | |
| *** ducttape_ has quit IRC | 14:31 | |
| *** links has quit IRC | 14:37 | |
| xgerman | o/ | 14:39 |
| *** Dave has quit IRC | 14:41 | |
| *** gcheresh_ has joined #openstack-lbaas | 14:50 | |
| *** ducttape_ has joined #openstack-lbaas | 14:55 | |
| *** gcheresh_ has quit IRC | 14:57 | |
| *** ducttape_ has quit IRC | 15:01 | |
| *** ducttape_ has joined #openstack-lbaas | 15:01 | |
| *** ducttape_ has quit IRC | 15:06 | |
| *** Alex_Stef has joined #openstack-lbaas | 15:08 | |
| *** ducttape_ has joined #openstack-lbaas | 15:28 | |
| *** ankur-gupta-f1 has joined #openstack-lbaas | 15:41 | |
| *** Dave has joined #openstack-lbaas | 15:49 | |
| *** rcernin has quit IRC | 15:53 | |
| *** TrevorV has joined #openstack-lbaas | 15:57 | |
| *** _ducttape_ has joined #openstack-lbaas | 15:58 | |
| *** ducttape_ has quit IRC | 16:02 | |
| *** Dave has quit IRC | 16:05 | |
| *** Alex_Stef has quit IRC | 16:08 | |
| johnsom | o/ | 16:10 |
| *** _ducttape_ has quit IRC | 16:10 | |
| ankur-gupta-f1 | johnsom: morning | 16:11 |
| *** ducttape_ has joined #openstack-lbaas | 16:11 | |
| *** cody-somerville has quit IRC | 16:11 | |
| *** cody-somerville has joined #openstack-lbaas | 16:11 | |
| *** ducttape_ has quit IRC | 16:17 | |
| *** ducttape_ has joined #openstack-lbaas | 16:17 | |
| *** Alex_Stef has joined #openstack-lbaas | 16:31 | |
| *** ducttape_ has quit IRC | 16:40 | |
| *** ducttape_ has joined #openstack-lbaas | 16:40 | |
| *** Alex_Stef has quit IRC | 16:41 | |
| *** mixos has joined #openstack-lbaas | 16:42 | |
| *** mixos has quit IRC | 16:46 | |
| *** tesseract has quit IRC | 16:47 | |
| *** bana_k has joined #openstack-lbaas | 16:50 | |
| *** fnaval has joined #openstack-lbaas | 16:57 | |
| *** amotoki has joined #openstack-lbaas | 16:59 | |
| diltram | ankur-gupta-f1: back again in the castle? | 17:02 |
| ankur-gupta-f1 | diltram: not back till Monday. Dont worry. We will have a 2 hr pow wow to figure everything out and start crushing this API stuff | 17:03 |
| diltram | I'm just thinking ankur-gupta-f1 about rebasing this all patches on top of my pagination/sorting patch | 17:04 |
| diltram | based on this we can start reviewing all patches and they will work after merging this pagination | 17:04 |
| johnsom | Sounds like a decent plan | 17:04 |
| ankur-gupta-f1 | We can deal with it next week, don't want to add additional dependencies. currently they are still all dependent on the yet to be merged base classes | 17:04 |
| ankur-gupta-f1 | specifically they are all dependent (or should be) on the Test Base class for V2 API patch. | 17:05 |
| diltram | I know but even they should all depend on previous because right now we will loose all reviews because there will be huge merge conflict | 17:05 |
| diltram | but if I will make this test base class dependent of pagination | 17:06 |
| diltram | I will have the whole beautifull chain of patches | 17:06 |
| diltram | which will be mergable | 17:06 |
| ankur-gupta-f1 | okay that makes sense. | 17:08 |
| ankur-gupta-f1 | johnsom: could use your eyes on https://review.openstack.org/#/c/405599/ and https://review.openstack.org/#/c/405621/ | 17:09 |
| johnsom | ok | 17:10 |
| ankur-gupta-f1 | thanks | 17:10 |
| openstackgerrit | Lubosz Kosnik (diltram) proposed openstack/octavia: Implement sorting and pagination for octavia https://review.openstack.org/382147 | 17:15 |
| *** kobis has joined #openstack-lbaas | 17:29 | |
| diltram | Octavia will be the first project with working py3x dsvm test :P | 17:32 |
| *** bana_k has quit IRC | 17:32 | |
| diltram | dims implemented that stuff in keystone but in not working way :p | 17:32 |
| johnsom | Ha | 17:36 |
| johnsom | Assuming we can get the gates merged... | 17:38 |
| openstackgerrit | Michael Johnson proposed openstack/octavia: Updated from global requirements https://review.openstack.org/417229 | 17:44 |
| openstackgerrit | Michael Johnson proposed openstack/octavia: Remove an erroneous MarkHealthMonitorActiveInDB task https://review.openstack.org/409403 | 17:44 |
| *** kevo has joined #openstack-lbaas | 17:45 | |
| *** woodster_ has joined #openstack-lbaas | 17:47 | |
| *** jerrygb has joined #openstack-lbaas | 17:50 | |
| *** gcheresh_ has joined #openstack-lbaas | 17:54 | |
| diltram | it's gonna work :) | 17:58 |
| diltram | jlvillal - which is OSIC member - works on enabling it everywhere | 17:58 |
| diltram | we just need to align names :) | 17:59 |
| diltram | he by mistake in commit msg specified different name that used in code | 17:59 |
| diltram | ok reboot, my display is not working L/ | 18:00 |
| diltram | :/ | 18:00 |
| *** pcaruana has quit IRC | 18:01 | |
| *** bana_k has joined #openstack-lbaas | 18:22 | |
| *** gcheresh_ has quit IRC | 18:23 | |
| *** _beardedeagle has joined #openstack-lbaas | 18:31 | |
| *** _beardedeagle has quit IRC | 18:35 | |
| *** beardedeagle has quit IRC | 18:35 | |
| *** Alex_Stef has joined #openstack-lbaas | 18:54 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia: Add quota support to Octavia https://review.openstack.org/360794 | 18:56 |
| *** jerrygb_ has joined #openstack-lbaas | 18:57 | |
| *** amotoki has quit IRC | 18:58 | |
| *** jerrygb has quit IRC | 19:00 | |
| *** ducttape_ has quit IRC | 19:00 | |
| *** ducttape_ has joined #openstack-lbaas | 19:00 | |
| *** gcheresh_ has joined #openstack-lbaas | 19:17 | |
| *** raginbajin has quit IRC | 19:21 | |
| *** ducttape_ has quit IRC | 19:23 | |
| *** ducttape_ has joined #openstack-lbaas | 19:24 | |
| *** ducttape_ has quit IRC | 19:24 | |
| *** ducttape_ has joined #openstack-lbaas | 19:24 | |
| *** anilvenkata has quit IRC | 19:37 | |
| *** jerrygb_ has quit IRC | 19:43 | |
| *** TrevorV has quit IRC | 19:46 | |
| *** TrevorV has joined #openstack-lbaas | 19:48 | |
| *** Alex_Stef has quit IRC | 19:56 | |
| *** amotoki has joined #openstack-lbaas | 19:59 | |
| *** jerrygb has joined #openstack-lbaas | 20:02 | |
| *** jerrygb has quit IRC | 20:03 | |
| *** jerrygb has joined #openstack-lbaas | 20:03 | |
| *** beardedeagle has joined #openstack-lbaas | 20:11 | |
| *** jerrygb_ has joined #openstack-lbaas | 20:13 | |
| *** jerrygb has quit IRC | 20:15 | |
| *** jerrygb has joined #openstack-lbaas | 20:17 | |
| *** jerrygb_ has quit IRC | 20:19 | |
| *** TrevorV has quit IRC | 20:20 | |
| *** csomerville has joined #openstack-lbaas | 20:30 | |
| *** TrevorV has joined #openstack-lbaas | 20:32 | |
| *** cody-somerville has quit IRC | 20:33 | |
| *** Dave has joined #openstack-lbaas | 21:00 | |
| *** gongysh has joined #openstack-lbaas | 21:15 | |
| *** catintheroof has quit IRC | 21:18 | |
| *** gcheresh_ has quit IRC | 21:27 | |
| rm_work | johnsom: can you check my logic on https://review.openstack.org/#/c/416519/3/octavia/network/drivers/neutron/allowed_address_pairs.py@95 ? | 21:28 |
| johnsom | Sure, looking | 21:28 |
| openstackgerrit | Merged openstack/octavia: Updated from global requirements https://review.openstack.org/417229 | 21:30 |
| openstackgerrit | Merged openstack/octavia: Introduce API Decorators https://review.openstack.org/405621 | 21:30 |
| johnsom | rm_work You are correct, there is no real point to that iteration over the subnets. We should just put the subnet_id in the "fixed_ips" field (which is a strange overload IMHO) and call it a day. | 21:32 |
| rm_work | kk | 21:32 |
| rm_work | I'm never sure with the ports stuff because I'm not sure i fully understand what happens under the hood there | 21:33 |
| rm_work | oh, as for what we were discussing earlier (your concerns about using FLIPs), the way they're implemented here the swing time is in milliseconds | 21:34 |
| rm_work | so it shouldn't be a problem to use FLIPs instead of the AAP ports | 21:35 |
| johnsom | Interesting | 21:35 |
| rm_work | it's not GARP based, it's static routes | 21:35 |
| johnsom | Upstream it's like 30+ seconds | 21:35 |
| rm_work | yeah | 21:35 |
| rm_work | not here :) | 21:36 |
| johnsom | Hmm | 21:36 |
| rm_work | also looking at options for configuration of amps besides rest_api | 21:36 |
| rm_work | there's some stuff that would be kinda "out of band" that could work | 21:36 |
| rm_work | getting more details still | 21:37 |
| rm_work | that'd simplify the image a ton, no need for a mgmt interface, no need for netns, no need for agent | 21:37 |
| rm_work | my main concerns are security, consistency of updates, and speed of updates | 21:38 |
| johnsom | You are creeping my out with "no agent, no netns" | 21:39 |
| rm_work | heh | 21:39 |
| rm_work | well, i'm not yet convinced myself, but the concept of getting rid of that stuff is alluring | 21:39 |
| johnsom | Ha | 21:39 |
| *** armax has quit IRC | 21:39 | |
| johnsom | Added my comment and -1'd that patch | 21:40 |
| rm_work | people here want to poll a config management system like Consul.io so I'm in the research phase | 21:40 |
| *** armax has joined #openstack-lbaas | 21:40 | |
| johnsom | Ah, yeah, we considered that at a "place" I used to work | 21:40 |
| *** gcheresh_ has joined #openstack-lbaas | 21:41 | |
| johnsom | That exact package too | 21:41 |
| xgerman | but then the designaye people couneled against it | 21:41 |
| rm_work | lol | 21:41 |
| rm_work | hmm | 21:41 |
| rm_work | then maybe I can get a jumpstart | 21:41 |
| johnsom | Well, it brought up other issues | 21:41 |
| rm_work | I'm concerned about the security model primarily | 21:42 |
| rm_work | supposedly the amps could register as nodes in it as well for health monitoring? | 21:42 |
| xgerman | that sounds more like etcd | 21:42 |
| rm_work | which is interesting? but makes me concerned about scale | 21:42 |
| johnsom | Bingo | 21:43 |
| rm_work | heh yeah was looking at that too | 21:43 |
| xgerman | to be fair those things are fairly scalable and k8 is using etcd for it;s work | 21:43 |
| xgerman | so I would assume it works | 21:43 |
| rm_work | so you've already basically gone down both of these routes and decided they weren't feasible | 21:43 |
| xgerman | we have been counseled against consul from the DNS people | 21:44 |
| xgerman | I think etcd was never considered | 21:44 |
| xgerman | but those things behave beastly in netsplits and failures | 21:44 |
| xgerman | security might be ok - I think they have RBAC | 21:45 |
| johnsom | Yeah, it was a few years ago. I looked at etcd too. SOP issues, netsplits, preference to decentralized solutions, secure communication paths, trusted to untrusted, etc. All thoughts that went into it | 21:45 |
| rm_work | the other thing I had always kinda considered was using a configuration "drive" | 21:45 |
| rm_work | https://blueprints.launchpad.net/cinder/+spec/multi-attach-volume | 21:45 |
| rm_work | supposedly this has been implemented for a while? | 21:45 |
| xgerman | we have a config drive in Octavia to distrinute certs and configuration | 21:46 |
| rm_work | so you could have the controller attach and write configs to tiny cinder volumes | 21:46 |
| rm_work | which are mounted read-only on the amps | 21:46 |
| rm_work | (on multiple amps!) | 21:46 |
| johnsom | Yeah, we do already have config drive there, but it is limited in it's scope of use. | 21:46 |
| rm_work | sorry i didn't mean real config-drive | 21:46 |
| rm_work | that's why i tried to put it in quotes and such | 21:46 |
| johnsom | Currently it is one drive to one amp | 21:46 |
| xgerman | mmh, so you like to change config while the amp is running but not use the REST interface because yo like them all to discover it themselves | 21:47 |
| rm_work | I mean, if it's running with a shared volume | 21:47 |
| rm_work | it can literally be running with the haproxy configs on that volume | 21:47 |
| rm_work | and soft-restart on i-notify changes | 21:47 |
| johnsom | What locking fun you will have..... Going to run OCFS2? hahahahaha | 21:47 |
| rm_work | completely out-of-band config | 21:47 |
| rm_work | well, if it's read-only on amps | 21:47 |
| rm_work | it shouldn't have serious issues | 21:48 |
| xgerman | well, I like the etcd type things better for that - then you can have some sort of feedback who actually applied there change | 21:48 |
| johnsom | Anyhow, ummm, if you feel the need.... | 21:48 |
| xgerman | yeah, mostly I can see people putting things on it so they can update without rolling amps | 21:49 |
| rm_work | hmm | 21:49 |
| xgerman | well, I guess yiu like that as a replacement for our REST/Healthmonitoring setup | 21:49 |
| rm_work | well, i'm interested in any of the negatives you found about Consul, so i can argue those points and see if there's any good solution | 21:49 |
| rm_work | really it's a networking thing | 21:49 |
| rm_work | i'd like to not have to plug two interfaces | 21:50 |
| rm_work | it would simplify things insanely lots | 21:50 |
| xgerman | consul would be running on mgmt net or would you run it on the Interwebs? | 21:50 |
| rm_work | and if we didn't need an agent, that'd simplify things quite a lot again | 21:50 |
| rm_work | simple is good | 21:50 |
| rm_work | erg | 21:50 |
| rm_work | soooooo | 21:50 |
| rm_work | there isn't really ... a difference here | 21:50 |
| xgerman | we had ssh maybe you want to return to that :-) | 21:50 |
| rm_work | which is why i'm kinda disliking it right now | 21:51 |
| johnsom | I really don't see how you won't have an agent with native haproxy | 21:51 |
| xgerman | yep, you will need somethign to push health | 21:51 |
| rm_work | err, well | 21:51 |
| rm_work | i guess it'll need just the health part | 21:52 |
| rm_work | but not the rest-api part | 21:52 |
| rm_work | no binding on a port | 21:52 |
| rm_work | and the health part is tiny | 21:52 |
| johnsom | We did push model to make security a priority, but you could always do a pull agent | 21:52 |
| xgerman | well, in the case of etcd they usually use TCP so health would be more heavy | 21:52 |
| rm_work | yeah, it's essentially still an agent, just pulling from Consul | 21:52 |
| rm_work | I guess what mean is, it's not a huge block of agent code in our codebase | 21:53 |
| johnsom | You might also talk with arimth, he is facing a bunch of issues with an "agentless" trove | 21:53 |
| rm_work | and it doesn't bind to a port | 21:53 |
| xgerman | well, I have seen Kubernetes using etcd so i think it might work | 21:53 |
| rm_work | do you remember any of the specific negatives of Consul? | 21:53 |
| johnsom | I gave you a list a few lines back. What I remember off my head | 21:54 |
| rm_work | I mean, security-model-wise, to be truly secure it'd still need all of the same multi-net/netns stuff | 21:54 |
| xgerman | besides doing unspeakable things to DNS… I assume it comes with the (un)reliabelity of DNS and it’s updates | 21:54 |
| rm_work | ah "SOP issues, netsplits, preference to decentralized solutions, secure communication paths, trusted to untrusted, etc" | 21:55 |
| rm_work | got it | 21:55 |
| rm_work | yeah, i'm more interested in true out-of-band | 21:55 |
| xgerman | yeah, netsplits and failures are a bitch with those systems. etcd goes in read-only BTW | 21:55 |
| johnsom | Well, we have a security model were actions always flow from a more trusted environment to the less trusted. | 21:55 |
| rm_work | like what I mentioned with using multi-attach volumes | 21:55 |
| johnsom | arimth is fighting with one trove vm sending fake messages up on behalf of another trove vm. He's getting into a bunch of strange signing things. | 21:56 |
| xgerman | trove? they are still a thing? | 21:57 |
| johnsom | Our model is so much cleaner | 21:57 |
| xgerman | yeah, we are pretty fail safe | 21:58 |
| xgerman | as long as you can keep the mysql chugging everyhting can fail and you come back | 21:58 |
| rm_work | still not sure i see issues with a config volume approach :P | 21:59 |
| johnsom | Yeah, that is "better", but you still have locking stuff to deal with. | 22:00 |
| rm_work | why? | 22:00 |
| rm_work | i mean | 22:00 |
| johnsom | Well, and does your cinder scale to that | 22:01 |
| xgerman | say you have two controllers | 22:01 |
| rm_work | right but only one controller can act on a LB config at once | 22:01 |
| rm_work | right? | 22:01 |
| rm_work | we ALREADY lock there | 22:01 |
| johnsom | plus you have controllers writing and amps reading/writing | 22:01 |
| *** catintheroof has joined #openstack-lbaas | 22:01 | |
| rm_work | wait, amps write? | 22:01 |
| xgerman | health messages? | 22:01 |
| rm_work | oh | 22:01 |
| johnsom | +1 | 22:01 |
| rm_work | i'm still imagining that being via UDP | 22:02 |
| rm_work | i guess that means for security we still need two interfaces... hmm | 22:02 |
| xgerman | ok, so instead of the rest calls you would push a haproxy config to the drive and some “software” will see the change and do soemthign meaningful | 22:02 |
| rm_work | though in my case it doesn't matter | 22:02 |
| *** TrevorV has quit IRC | 22:02 | |
| rm_work | yeah, i mean, can set up the image to auto-soft-restart haproxy on i-notify events | 22:03 |
| rm_work | assuming those work on cinder volumes | 22:03 |
| johnsom | I come back to, is this necessary. In clouds virtual networks are basically free and readily available. | 22:03 |
| xgerman | aha, so you would stop having one haproxy process for listener | 22:03 |
| rm_work | lulz | 22:03 |
| rm_work | johnsom: unfortunately there are no custom networks here | 22:04 |
| rm_work | so i'm in an interesting position :P | 22:04 |
| rm_work | there is exactly one subnet you can bind to | 22:04 |
| rm_work | period | 22:04 |
| rm_work | end of cloudstory :P | 22:04 |
| johnsom | So create your own tunnel | 22:04 |
| rm_work | lol | 22:04 |
| * xgerman is glad they hired rm_work instead of me | 22:04 | |
| johnsom | Hahaha, you can IPSec your mgmt lan | 22:04 |
| rm_work | that's just... lol | 22:05 |
| rm_work | i mean | 22:05 |
| xgerman | ssh tunnel? | 22:05 |
| rm_work | actually that'd save me from having to plug two interfaces | 22:05 |
| johnsom | Solves your problem | 22:05 |
| rm_work | it's ... funny, but i'm not immediately seeing how it's a bad idea | 22:05 |
| rm_work | lol | 22:05 |
| johnsom | It doesn't use up ports | 22:05 |
| rm_work | hmmmmmmmmmmmmmmmmmm | 22:06 |
| johnsom | If you guys do it, I get my DNS fees waived..... | 22:06 |
| johnsom | Grin | 22:06 |
| rm_work | lol | 22:06 |
| rm_work | can't use netns with that though right? because if you moved the original interface into a netns and tried to leave the tun interface outside it ... WOULD that work? or would it break | 22:07 |
| xgerman | you cna map whatever you want to nentns | 22:07 |
| xgerman | but tun would be mgmt so outside ns? | 22:07 |
| johnsom | You can make it work, there are ways to "share" | 22:08 |
| johnsom | It gets a bit mind twisty, but can be done | 22:08 |
| *** TrevorV has joined #openstack-lbaas | 22:08 | |
| xgerman | I have seen people running their vpn agent in a. container so… | 22:09 |
| johnsom | Yeah, exactly | 22:09 |
| rm_work | hmm | 22:10 |
| rm_work | i think that might actually be ... the way to go | 22:10 |
| rm_work | though billing for bandwidth gets a bit tricky :P | 22:10 |
| rm_work | i guess you can estimate standard traffic caused by HM noise and just map-remove it from everything | 22:11 |
| rm_work | but it's imprecise | 22:11 |
| xgerman | can’t you use iptables in the ns? | 22:11 |
| johnsom | You could measure it with iptables rules | 22:11 |
| xgerman | +1 | 22:11 |
| rm_work | unless you're billing from haproxy stats i guess :P | 22:11 |
| johnsom | Yeah, what he said | 22:11 |
| rm_work | magic | 22:12 |
| *** jerrygb has quit IRC | 22:13 | |
| *** gongysh_ has joined #openstack-lbaas | 22:15 | |
| *** gongysh has quit IRC | 22:17 | |
| *** matt-borland has quit IRC | 22:28 | |
| *** gcheresh_ has quit IRC | 22:37 | |
| *** armax has quit IRC | 22:46 | |
| *** ducttape_ has quit IRC | 23:00 | |
| *** _beardedeagle has joined #openstack-lbaas | 23:01 | |
| diltram | rm_work: I didn't saw your vote for this code https://review.openstack.org/#/c/399117/ | 23:03 |
| diltram | so I'm pinging you :) | 23:03 |
| rm_work | ah | 23:03 |
| diltram | it's not complete work, require additional patches but they will be added later | 23:04 |
| diltram | after merging API :) | 23:04 |
| rm_work | I am weak on policy stuff | 23:04 |
| diltram | but it's not really policy | 23:04 |
| diltram | it's just some code which enables policy | 23:04 |
| diltram | there are just maybe 3 or 4 rules | 23:05 |
| diltram | because it's a ground for moving policy rules from nlbaas into octavia | 23:05 |
| *** beardedeagle has quit IRC | 23:05 | |
| rm_work | BTW What are the copyright rules now? I can never keep track... are you supposed to put some company copyright at the top or is it all implied now? >_< | 23:06 |
| rm_work | also BTW, was this copy/pasted from somewhere? | 23:06 |
| johnsom | Same, yeah, if you create a file it should have a copyright | 23:06 |
| diltram | yes, this code is moved from nova | 23:07 |
| rm_work | keep the original copyright from that code then, since you're just doing mods? | 23:08 |
| rm_work | like, names? | 23:08 |
| rm_work | if i can see this is exactly what nova used, it's prolly an easy +2 | 23:08 |
| johnsom | Yeah, it should keep the original. You can add a line for yourself if you modify | 23:08 |
| openstackgerrit | Michael Johnson proposed openstack/octavia: Add quota support to Octavia https://review.openstack.org/360794 | 23:09 |
| *** kevo has quit IRC | 23:12 | |
| diltram | it's taken as is from nova | 23:15 |
| rm_work | alright | 23:15 |
| rm_work | then I'm sold :P | 23:15 |
| diltram | if there is no copyrights there was no copyrights | 23:15 |
| diltram | :P | 23:15 |
| diltram | my code is just octavia/common/context and some changes in octavia/common/policy - because nova is using global var I'm using a class | 23:16 |
| *** gongysh_ has quit IRC | 23:17 | |
| *** _beardedeagle has quit IRC | 23:21 | |
| rm_work | ok so | 23:21 |
| rm_work | we already have the PKI in place for IPSEC via cert... | 23:22 |
| rm_work | we re-use the existing ca / client certs that are already on the amps | 23:22 |
| rm_work | the boot interface *is* the vip, we create the IPSEC tunnel and move the vip into the netns | 23:23 |
| rm_work | bam | 23:23 |
| diltram | rm_work: what you're trying to build? | 23:23 |
| diltram | :P | 23:23 |
| rm_work | it really doesn't seem like a horrible idea at first glance, i'm still trying to figure out why it's not | 23:23 |
| *** fnaval has quit IRC | 23:23 | |
| rm_work | trying to figure out a security model for management net without the ability to create neutron networks | 23:24 |
| johnsom | Yeah, I don't think it is a bad solution actually. | 23:25 |
| johnsom | It doesn't solve your act/stndby issue but is a start | 23:25 |
| rm_work | i'm playing around with strongswan | 23:25 |
| *** fnaval has joined #openstack-lbaas | 23:25 | |
| rm_work | i've never actually done ipsec before but it doesn't look TOO bad, just really horrible docs/guides | 23:25 |
| *** TrevorV has quit IRC | 23:26 | |
| johnsom | I can help you with it. I ran ~6,000 ipsec tunnels to retail stores all over the world. | 23:27 |
| rm_work | lol | 23:27 |
| johnsom | Though, now, with ikev2 your job is a lot easier | 23:27 |
| rm_work | yeah it seems that way | 23:27 |
| rm_work | the old way seems ... ugh | 23:27 |
| johnsom | Also, compare strongswan and openswan. One is kind of dying, but I always get them confused. | 23:27 |
| rm_work | but yeah, if we wanted this to be an option, i'm willing to put in some effort on it | 23:27 |
| rm_work | supposedly strongswan is the living one i think? | 23:28 |
| rm_work | but will doublecheck | 23:28 |
| rm_work | i wonder if we run into issues with having strongswan as a binary dep? though maybe it doesn't matter if it's not the default/reference | 23:29 |
| johnsom | Yeah, it looks like strongswan is what ubuntu bundles, so probably the best bet | 23:29 |
| johnsom | It's in our image, so just add an element if there isn't one already. vpnaas (now dead) uses one of them, so there might be an element already | 23:30 |
| rm_work | heh | 23:30 |
| johnsom | pretty easy to add an element either way | 23:31 |
| rm_work | yep | 23:33 |
| *** fnaval has quit IRC | 23:34 | |
| rm_work | just trying to figure out how to configure the client side in a way that's less server-server network bridge, and more client-server | 23:35 |
| rm_work | all the guides seem to assume you want to bridge the networks of two routers <_< | 23:35 |
| *** ducttape_ has joined #openstack-lbaas | 23:36 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia: Add support for policy.json https://review.openstack.org/399117 | 23:37 |
| rm_work | did that need a rebase? | 23:40 |
| johnsom | Yeah, I couldn't check the test coverage | 23:40 |
| rm_work | ah | 23:42 |
| openstackgerrit | Merged openstack/octavia: Introduct Test Base classes for V2 API https://review.openstack.org/405599 | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!