| *** ducttape_ has joined #openstack-lbaas | 00:05 | |
| *** ducttape_ has quit IRC | 00:05 | |
| *** ducttape_ has joined #openstack-lbaas | 00:05 | |
| *** harlowja has joined #openstack-lbaas | 00:06 | |
| diltram | sbalukoff: I'm gere | 00:08 |
|---|---|---|
| sbalukoff | diltram: Sorry to bug you! johnsom already got it, eh. | 00:09 |
| diltram | ok, np :) | 00:09 |
| diltram | sbalukoff, johnsom: heard about that? https://wiki.fd.io/view/VPP/What_is_VPP%3F | 00:19 |
| johnsom | Yes | 00:20 |
| johnsom | There is an openstack project proposed to use it | 00:22 |
| *** bdeschenes has joined #openstack-lbaas | 00:22 | |
| johnsom | http://osdir.com/ml/openstack-dev/2016-10/msg00258.html | 00:22 |
| *** bana_k has quit IRC | 00:23 | |
| sbalukoff | diltram: That's cool, eh! | 00:24 |
| diltram | nice, didn't know that there is something using it already | 00:25 |
| *** ducttape_ has quit IRC | 00:26 | |
| johnsom | Well, it's proposed... | 00:28 |
| openstackgerrit | Merged openstack/neutron-lbaas: Fix a gate issue with a mock and deprecated call https://review.openstack.org/386270 | 00:34 |
| openstackgerrit | Michael Johnson proposed openstack/neutron-lbaas: Create a consolidated API test https://review.openstack.org/386240 | 00:34 |
| *** bana_k has joined #openstack-lbaas | 00:52 | |
| *** yuanying has quit IRC | 01:03 | |
| *** gongysh_ has quit IRC | 01:06 | |
| *** yuanying has joined #openstack-lbaas | 01:07 | |
| *** fnaval has quit IRC | 01:09 | |
| *** ducttape_ has joined #openstack-lbaas | 01:13 | |
| *** fnaval has joined #openstack-lbaas | 01:19 | |
| *** yuanying_ has joined #openstack-lbaas | 01:23 | |
| *** bdeschenes has quit IRC | 01:25 | |
| *** bana_k has quit IRC | 01:26 | |
| *** yuanying has quit IRC | 01:26 | |
| *** derekjhyang has joined #openstack-lbaas | 01:49 | |
| *** gongysh_ has joined #openstack-lbaas | 02:09 | |
| *** cody-somerville has joined #openstack-lbaas | 02:24 | |
| *** cody-somerville has quit IRC | 02:24 | |
| *** cody-somerville has joined #openstack-lbaas | 02:24 | |
| *** gongysh_ has quit IRC | 02:37 | |
| *** ducttape_ has quit IRC | 02:43 | |
| *** yuanying_ has quit IRC | 02:51 | |
| *** yuanying has joined #openstack-lbaas | 02:52 | |
| *** armax has quit IRC | 03:09 | |
| *** tovin07 has joined #openstack-lbaas | 03:10 | |
| *** sputnik13 has joined #openstack-lbaas | 03:12 | |
| *** yuanying has quit IRC | 03:15 | |
| *** yuanying has joined #openstack-lbaas | 03:15 | |
| *** yuanying_ has joined #openstack-lbaas | 03:17 | |
| *** yuanying has quit IRC | 03:18 | |
| *** yuanying_ has quit IRC | 03:22 | |
| *** ducttape_ has joined #openstack-lbaas | 03:23 | |
| *** yamamoto has quit IRC | 03:30 | |
| *** ducttape_ has quit IRC | 03:39 | |
| *** yuanying has joined #openstack-lbaas | 03:52 | |
| *** links has joined #openstack-lbaas | 03:59 | |
| *** sputnik13 has quit IRC | 04:15 | |
| *** anilvenkata has joined #openstack-lbaas | 04:28 | |
| *** gongysh_ has joined #openstack-lbaas | 04:30 | |
| openstackgerrit | Stephen Balukoff proposed openstack/octavia: Load system sysctl in amphora network namespace https://review.openstack.org/386328 | 04:36 |
| *** ducttape_ has joined #openstack-lbaas | 04:39 | |
| *** yamamoto has joined #openstack-lbaas | 04:44 | |
| *** ducttape_ has quit IRC | 04:45 | |
| *** bana_k has joined #openstack-lbaas | 05:01 | |
| *** sputnik13 has joined #openstack-lbaas | 05:06 | |
| *** bana_k has quit IRC | 05:29 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/octavia: Updated from global requirements https://review.openstack.org/386370 | 05:41 |
| *** ducttape_ has joined #openstack-lbaas | 05:41 | |
| *** ducttape_ has quit IRC | 05:45 | |
| *** sputnik13 has quit IRC | 05:48 | |
| *** amotoki has joined #openstack-lbaas | 05:51 | |
| *** woodster_ has quit IRC | 05:55 | |
| *** numans has joined #openstack-lbaas | 06:12 | |
| *** pcaruana has joined #openstack-lbaas | 06:23 | |
| *** jsheeren has joined #openstack-lbaas | 06:40 | |
| *** ducttape_ has joined #openstack-lbaas | 06:42 | |
| *** ducttape_ has quit IRC | 06:46 | |
| *** jsheeren has quit IRC | 06:50 | |
| *** jsheeren has joined #openstack-lbaas | 06:51 | |
| *** tesseract- has joined #openstack-lbaas | 07:05 | |
| *** rcernin has joined #openstack-lbaas | 07:20 | |
| openstackgerrit | Stephen Balukoff proposed openstack/octavia: Update haproxy logging configuration https://review.openstack.org/386427 | 07:21 |
| *** amoralej|off is now known as amoralej | 07:27 | |
| *** ducttape_ has joined #openstack-lbaas | 07:42 | |
| *** ihrachys has joined #openstack-lbaas | 07:43 | |
| *** ihrachys has quit IRC | 07:45 | |
| *** ducttape_ has quit IRC | 07:47 | |
| *** ihrachys has joined #openstack-lbaas | 07:48 | |
| *** sputnik13 has joined #openstack-lbaas | 07:52 | |
| *** gtrxcb has joined #openstack-lbaas | 07:55 | |
| *** eezhova has joined #openstack-lbaas | 08:00 | |
| *** gtrxcb has quit IRC | 08:10 | |
| openstackgerrit | Miguel Angel Ajo proposed openstack/octavia: Enable support for multinode in the gate hook https://review.openstack.org/378677 | 08:15 |
| openstackgerrit | Miguel Angel Ajo proposed openstack/octavia: WIP/DONOTMERGE: multinode ignore server_ca (checking) https://review.openstack.org/386450 | 08:17 |
| *** ipsecguy has quit IRC | 08:28 | |
| *** eezhova has quit IRC | 08:31 | |
| *** ducttape_ has joined #openstack-lbaas | 08:43 | |
| *** ducttape_ has quit IRC | 08:48 | |
| *** ipsecguy has joined #openstack-lbaas | 08:50 | |
| *** eezhova has joined #openstack-lbaas | 09:01 | |
| openstackgerrit | Merged openstack/octavia: Updated from global requirements https://review.openstack.org/386370 | 09:09 |
| *** ihrachys has quit IRC | 09:10 | |
| *** ihrachys has joined #openstack-lbaas | 09:13 | |
| *** ihrachys has quit IRC | 09:16 | |
| openstackgerrit | Miguel Angel Ajo proposed openstack/octavia: WIP/DONOTMERGE: multinode ignore server_ca (checking) https://review.openstack.org/386450 | 09:17 |
| *** Kiall has joined #openstack-lbaas | 09:37 | |
| *** amotoki has quit IRC | 09:41 | |
| *** ihrachys has joined #openstack-lbaas | 09:43 | |
| *** ducttape_ has joined #openstack-lbaas | 09:44 | |
| *** ihrachys has quit IRC | 09:44 | |
| *** openstack has joined #openstack-lbaas | 09:48 | |
| *** ducttape_ has quit IRC | 09:48 | |
| *** amotoki has joined #openstack-lbaas | 09:55 | |
| *** tovin07 has quit IRC | 10:00 | |
| *** amotoki has quit IRC | 10:04 | |
| *** openstack has joined #openstack-lbaas | 10:17 | |
| *** sputnik13 has quit IRC | 10:20 | |
| *** jsheeren has quit IRC | 10:33 | |
| *** ducttape_ has joined #openstack-lbaas | 10:45 | |
| *** amotoki has joined #openstack-lbaas | 10:49 | |
| *** ducttape_ has quit IRC | 10:50 | |
| *** gongysh_ has quit IRC | 10:58 | |
| *** kobis has joined #openstack-lbaas | 11:05 | |
| openstackgerrit | Miguel Angel Ajo proposed openstack/octavia: Enable support for multinode in the gate hook https://review.openstack.org/378677 | 11:16 |
| openstackgerrit | Miguel Angel Ajo proposed openstack/octavia: Increase the verbosity of haproxy.rest_api_driver failures https://review.openstack.org/386544 | 11:16 |
| *** kobis has quit IRC | 11:20 | |
| *** kobis has joined #openstack-lbaas | 11:23 | |
| *** amoralej is now known as amoralej|lunch | 11:28 | |
| *** yamamoto has quit IRC | 11:29 | |
| *** ducttape_ has joined #openstack-lbaas | 11:46 | |
| *** strigazi is now known as strigazi_AFK | 11:48 | |
| *** ducttape_ has quit IRC | 11:51 | |
| *** kobis has quit IRC | 12:00 | |
| *** links has quit IRC | 12:02 | |
| *** links has joined #openstack-lbaas | 12:14 | |
| *** amotoki has quit IRC | 12:15 | |
| *** yamamoto has joined #openstack-lbaas | 12:20 | |
| *** ducttape_ has joined #openstack-lbaas | 12:20 | |
| *** yamamoto has quit IRC | 12:24 | |
| *** yamamoto has joined #openstack-lbaas | 12:26 | |
| *** anilvenkata is now known as anilvenkata_afk | 12:32 | |
| *** yamamoto has quit IRC | 12:36 | |
| *** yamamoto has joined #openstack-lbaas | 12:37 | |
| *** openstack has joined #openstack-lbaas | 12:47 | |
| *** ducttape_ has quit IRC | 12:49 | |
| *** ducttape_ has joined #openstack-lbaas | 13:04 | |
| *** reedip has quit IRC | 13:06 | |
| *** bdeschenes has joined #openstack-lbaas | 13:10 | |
| *** eezhova has quit IRC | 13:12 | |
| *** eezhova has joined #openstack-lbaas | 13:14 | |
| *** eezhova has quit IRC | 13:15 | |
| *** links has quit IRC | 13:16 | |
| *** ducttape_ has quit IRC | 13:17 | |
| *** eezhova has joined #openstack-lbaas | 13:18 | |
| *** eezhova has quit IRC | 13:21 | |
| *** portdirect has quit IRC | 13:24 | |
| *** portdirect has joined #openstack-lbaas | 13:24 | |
| *** eezhova has joined #openstack-lbaas | 13:26 | |
| *** eezhova has quit IRC | 13:26 | |
| *** yamamoto has quit IRC | 13:30 | |
| *** gongysh has joined #openstack-lbaas | 13:37 | |
| *** ducttape_ has joined #openstack-lbaas | 13:39 | |
| *** yamamoto has joined #openstack-lbaas | 13:42 | |
| *** eezhova has joined #openstack-lbaas | 13:42 | |
| *** yamamoto has quit IRC | 13:44 | |
| *** yamamoto has joined #openstack-lbaas | 13:44 | |
| openstackgerrit | Miguel Angel Ajo proposed openstack/octavia: Fixes pregenerated certificates validity to 50y https://review.openstack.org/386623 | 13:46 |
| openstackgerrit | Miguel Angel Ajo proposed openstack/octavia: Enable support for multinode in the gate hook https://review.openstack.org/378677 | 13:47 |
| openstackgerrit | Miguel Angel Ajo proposed openstack/octavia: Fixes pregenerated certificates validity to 50y https://review.openstack.org/386623 | 13:47 |
| ajo | sorry for the excessive noise ;) | 13:47 |
| *** amoralej|lunch is now known as amoralej | 13:50 | |
| *** fnaval has quit IRC | 13:57 | |
| *** fnaval has joined #openstack-lbaas | 14:08 | |
| *** ducttape_ has quit IRC | 14:17 | |
| ajo | 2016-10-14 14:22:01.475446 | + ./safe-devstack-vm-gate-wrap.sh:main:L616: sudo mv /home/jenkins/workspace/gate-octavia-v1-dsvm-scenario-multinode-nv/devstack-gate-cleanup-host.txt /opt/stack/logs/ | 14:24 |
| ajo | 2016-10-14 14:22:01.481642 | + ./safe-devstack-vm-gate-wrap.sh:main:L618: exit 0 | 14:24 |
| ajo | 2016-10-14 14:22:01.482755 | [Zuul] Task exit code: 0 | 14:24 |
| ajo | 2016-10-14 14:22:06.090446 | [Zuul] Job complete, result: SUCCESS | 14:24 |
| ajo | oh yeah folks! :D | 14:24 |
| ajo | logan-, blogan, dougwig, sbalukoff , ptoohill ^ | 14:26 |
| ajo | :) | 14:26 |
| ajo | multinode scenario working ':D | 14:26 |
| ajo | in 27m, amazing :) | 14:27 |
| ajo | https://review.openstack.org/#/c/378677/ | 14:27 |
| *** ducttape_ has joined #openstack-lbaas | 14:32 | |
| *** ducttape_ has quit IRC | 14:33 | |
| *** ducttape_ has joined #openstack-lbaas | 14:34 | |
| openstackgerrit | Miguel Angel Ajo proposed openstack/octavia: Enable support for multinode in the gate hook https://review.openstack.org/378677 | 14:39 |
| *** amotoki has joined #openstack-lbaas | 14:41 | |
| *** bdeschenes has quit IRC | 14:48 | |
| *** amotoki has quit IRC | 14:50 | |
| *** portdirect has quit IRC | 14:57 | |
| *** portdirect has joined #openstack-lbaas | 14:58 | |
| *** portdirect has quit IRC | 14:58 | |
| *** portdirect_ has joined #openstack-lbaas | 14:58 | |
| *** portdirect_ is now known as portdirect | 14:59 | |
| *** rcernin has quit IRC | 15:02 | |
| *** strigazi_AFK is now known as strigazi | 15:04 | |
| *** pcaruana has quit IRC | 15:15 | |
| *** armax has joined #openstack-lbaas | 15:20 | |
| diltram | ajo: great job man :) | 15:31 |
| *** bana_k has joined #openstack-lbaas | 15:40 | |
| blogan | ajo: awesome! | 15:40 |
| diltram | best copyrights ever :P # Copyright 2012 Your Mom | 15:46 |
| ajo | diltram, blogan :) | 15:46 |
| ajo | lol | 15:46 |
| ajo | thanks diltram blogan I sweated a bit to get it up , but now it works :D I will propose it to non-voting check once it's merged : | 15:51 |
| ajo | so we can gather stats and compare to the singlenode | 15:51 |
| blogan | diltram: where is that copyright? | 15:51 |
| diltram | https://review.openstack.org/#/c/382147/3 | 15:51 |
| blogan | of course it'd be carlos | 15:52 |
| diltram | yeah :P | 15:52 |
| ajo | lol | 15:52 |
| *** bana_k has quit IRC | 15:52 | |
| blogan | ajo: yeah the multinode will be quite informative, for better or worse ;) | 15:53 |
| diltram | +2 | 15:53 |
| johnsom | blogan oye.... | 15:54 |
| *** tesseract- has quit IRC | 15:54 | |
| blogan | johnsom: the second i saw what diltram wrote, i knew who it'd be | 15:55 |
| diltram | :P | 15:55 |
| ajo | thanks diltram, please note I have a depends-on https://review.openstack.org/#/c/381965/ (devstack fix for bug), other projects, like ironic, have included that in their gate_hook | 15:56 |
| ajo | so until that merges, I will include it in ours too, so we don't have to wait | 15:56 |
| xgerman | blogan what happened to octavia.io? RAX needs to save? | 15:56 |
| ajo | diltram, long history short, some of the multinode-capable clouds use the 10.1.x.x range for their nodes, and the neutron subnet pool in 10.0.0.0/8 by default breaks it | 15:56 |
| blogan | xgerman: ptoohill was maintainer of that, i only consumed it | 15:57 |
| xgerman | ok, the waybackmachine is our only hope | 15:57 |
| diltram | ajo: heard that, I saw e-mail on mailing list about that | 15:57 |
| ajo | diltram, please also note those two depended patches in octavia: https://review.openstack.org/#/c/386544/1 & https://review.openstack.org/#/c/386623/2 | 15:57 |
| ajo | well, actually first, is not a real dependency, I will remove as a base | 15:58 |
| johnsom | xgerman I see the domain still works: sonar.octavia.io | 15:58 |
| ajo | the 2nd one (regenerated pregenerated certs...) is | 15:58 |
| ajo | apparently, I cleated a client.pem for 10 years, and the CA was just for 1 month ':D | 15:58 |
| xgerman | well, I was hoping for our specs and docs... | 15:58 |
| ajo | cleated->created | 15:58 |
| ajo | ':D | 15:58 |
| johnsom | xgerman This works well now: http://docs.openstack.org/developer/octavia/ or the docs job | 15:58 |
| xgerman | oh, we move don | 15:59 |
| diltram | yeah, we can even kill that old octavia.io because it leads to missunderstandings - someone is updating that website? | 16:00 |
| johnsom | It auto updates | 16:00 |
| diltram | ok | 16:00 |
| diltram | Review could not be loaded. - I can see this | 16:00 |
| johnsom | Just don't kill the domain! We need it for our coverage gate | 16:01 |
| blogan | diltram: octavia.io reads all communication and understands all...by suggesting killing it you've activated it to take over the world! | 16:01 |
| diltram | hahahha :p | 16:01 |
| diltram | https://review.openstack.org/#/c/386328/ - this change is based on test results? | 16:03 |
| *** gongysh has quit IRC | 16:04 | |
| johnsom | It is a problem. I looked into it when they first raised it. | 16:05 |
| openstackgerrit | Miguel Angel Ajo proposed openstack/octavia: Enable support for multinode in the gate hook https://review.openstack.org/378677 | 16:05 |
| openstackgerrit | Miguel Angel Ajo proposed openstack/octavia: Fixes pregenerated certificates validity to 50y https://review.openstack.org/386623 | 16:05 |
| johnsom | The created namespaces start blank | 16:05 |
| ajo | ok, those two are now the only patches missing for multinode ^ other deps removed :) | 16:06 |
| *** bana_k has joined #openstack-lbaas | 16:07 | |
| *** yamamoto has quit IRC | 16:09 | |
| *** yamamoto has joined #openstack-lbaas | 16:11 | |
| rm_work | TIL: python only flyweights -5 through 256 | 16:24 |
| rm_work | so, don't rely on "is" for integer comparisons | 16:24 |
| *** ihrachys has quit IRC | 16:28 | |
| *** numans has quit IRC | 16:38 | |
| *** bana_k has quit IRC | 16:41 | |
| *** amoralej is now known as amoralej|off | 17:00 | |
| *** cody-somerville has quit IRC | 17:06 | |
| *** yamamoto has quit IRC | 17:06 | |
| *** yamamoto has joined #openstack-lbaas | 17:07 | |
| *** ihrachys has joined #openstack-lbaas | 17:12 | |
| *** yamamoto has quit IRC | 17:12 | |
| *** yamamoto has joined #openstack-lbaas | 17:16 | |
| *** links has joined #openstack-lbaas | 17:18 | |
| *** ihrachys has quit IRC | 17:19 | |
| *** eezhova has quit IRC | 17:24 | |
| *** bana_k has joined #openstack-lbaas | 17:26 | |
| *** cody-somerville has joined #openstack-lbaas | 17:29 | |
| *** cody-somerville has quit IRC | 17:29 | |
| *** cody-somerville has joined #openstack-lbaas | 17:29 | |
| johnsom | dougwig ping | 17:30 |
| dougwig | johnsom: ack | 17:31 |
| johnsom | Can you look at Andreas comment here: https://review.openstack.org/#/c/386245/ Am I missing something when it comes to stable branches? I'm leaving the old targets | 17:32 |
| njohnston | johnsom: I think he is saying that https://review.openstack.org/386240 sets up the consolidated API test for master. But since there is no version constraint on the new test, it will run against master + stable/newton + stable/mitaka etc. So you may need to use version constraints so that versions without https://review.openstack.org/386240 have working tests. | 17:37 |
| johnsom | Hmm, ok, I will look into that | 17:38 |
| johnsom | Thanks! | 17:38 |
| njohnston | johnsom: By way of example, I had to do something like that with the fwaas tempest tests, so I defined a 'legacy' job for old branches, and my new 'v1' and 'v2' test sets would run against everything else. https://github.com/openstack-infra/project-config/blob/master/zuul/layout.yaml#L2214-L2221 | 17:39 |
| johnsom | njohnston Thanks! | 17:39 |
| njohnston | johnsom: Any time! | 17:40 |
| *** pcaruana has joined #openstack-lbaas | 17:48 | |
| *** anilvenkata_afk has quit IRC | 17:55 | |
| *** portdirect has quit IRC | 18:02 | |
| *** cody-somerville has quit IRC | 18:02 | |
| *** cody-somerville has joined #openstack-lbaas | 18:03 | |
| *** cody-somerville has quit IRC | 18:03 | |
| *** cody-somerville has joined #openstack-lbaas | 18:03 | |
| *** portdirect has joined #openstack-lbaas | 18:03 | |
| *** eezhova has joined #openstack-lbaas | 18:15 | |
| dougwig | johnsom: do the consolidated jobs just timeout on non-nested-virt clouds? | 18:27 |
| johnsom | I am consolidating the API test jobs which use the noop drivers | 18:27 |
| johnsom | The scenario is the that spins up amps. Currently our scenario test times out and fails | 18:28 |
| *** ducttape_ has quit IRC | 18:31 | |
| openstackgerrit | Adam Harwell proposed openstack/octavia: Implement sorting and pagination for octavia https://review.openstack.org/382147 | 18:33 |
| johnsom | dougwig This would be an example of that: http://logs.openstack.org/40/386240/2/check/gate-neutron-lbaasv2-dsvm-scenario/b134378/ | 18:34 |
| johnsom | Ugh, in other scenario test issues, I'm not sure what to do about this: http://logs.openstack.org/40/386240/2/check/gate-neutron-lbaasv2-dsvm-scenario/cc6f667/logs/screen-o-cw.txt.gz#_2016-10-14_17_40_30_919 | 18:36 |
| johnsom | I added a "give it one more try" but that seems to have not worked. | 18:36 |
| johnsom | Not happy with flask/werkz | 18:36 |
| johnsom | I can't keep retrying on 404's as that is a valid response in some cases. | 18:37 |
| *** ihrachys has joined #openstack-lbaas | 18:37 | |
| *** ihrachys has quit IRC | 18:37 | |
| *** ihrachys has joined #openstack-lbaas | 18:38 | |
| *** ihrachys has quit IRC | 18:39 | |
| openstackgerrit | Adam Harwell proposed openstack/octavia: Add quota support to Octavia https://review.openstack.org/360794 | 18:40 |
| rm_work | yeah we should convert is to use gunicorn | 18:40 |
| rm_work | it isn't hard T_T | 18:40 |
| rm_work | *convert it | 18:40 |
| rm_work | that would solve multiple problems | 18:40 |
| *** links has quit IRC | 18:41 | |
| *** derekjhyang_ has joined #openstack-lbaas | 18:43 | |
| *** johnsom_ has joined #openstack-lbaas | 18:44 | |
| *** vmtyler_ has joined #openstack-lbaas | 18:45 | |
| *** Aju has joined #openstack-lbaas | 18:49 | |
| *** derekjhyang has quit IRC | 18:51 | |
| *** vmtyler has quit IRC | 18:51 | |
| *** johnsom has quit IRC | 18:51 | |
| *** afranc has quit IRC | 18:51 | |
| *** tries_ has quit IRC | 18:51 | |
| *** derekjhyang_ is now known as derekjhyang | 18:52 | |
| *** ducttape_ has joined #openstack-lbaas | 18:53 | |
| *** johnsom_ is now known as johnsom | 18:53 | |
| *** vmtyler_ is now known as vmtyler | 18:53 | |
| *** tries_ has joined #openstack-lbaas | 18:57 | |
| *** eezhova has quit IRC | 19:04 | |
| *** cody-somerville has quit IRC | 19:08 | |
| *** cody-somerville has joined #openstack-lbaas | 19:08 | |
| *** cody-somerville has quit IRC | 19:08 | |
| *** cody-somerville has joined #openstack-lbaas | 19:08 | |
| *** dlundquist has quit IRC | 19:09 | |
| *** dlundquist has joined #openstack-lbaas | 19:11 | |
| rm_work | johnsom: looking to see how tedious it will actually be to shove gunicorn in there | 19:11 |
| rm_work | it will kill the guru stuff, i think... not sure what else it'll interfere with | 19:11 |
| rm_work | the interesting part may actually be reading the config values | 19:11 |
| rm_work | because it'll have to be done outside of python | 19:11 |
| rm_work | or ... hmm | 19:12 |
| *** cody-somerville has quit IRC | 19:18 | |
| *** cody-somerville has joined #openstack-lbaas | 19:25 | |
| *** cody-somerville has quit IRC | 19:25 | |
| *** cody-somerville has joined #openstack-lbaas | 19:25 | |
| rm_work | yeah ok I think this is actually pretty easy | 19:27 |
| rm_work | almost done maybe? | 19:27 |
| *** cody-somerville has quit IRC | 19:32 | |
| rm_work | took longer to get devstack up and an amphora created to test this, than it did to write it, i think | 19:33 |
| rm_work | but, i should wait until I know it is working :P | 19:33 |
| *** cody-somerville has joined #openstack-lbaas | 19:34 | |
| *** cody-somerville has quit IRC | 19:34 | |
| *** cody-somerville has joined #openstack-lbaas | 19:34 | |
| rm_work | seems to work? | 19:51 |
| openstackgerrit | Adam Harwell proposed openstack/octavia: Run amphora agent with gunicorn https://review.openstack.org/386758 | 19:51 |
| rm_work | johnsom: ^^ | 19:51 |
| johnsom | Hmm cool! I at lunch, but will take a look later | 19:53 |
| rm_work | OK | 19:53 |
| rm_work | It doesn't do auto-reloading (though I could enable it?) but I think we explicitly reload the agent on changes, don't we? | 19:53 |
| rm_work | it's easy enough to add the reload option | 19:54 |
| rm_work | 'reload': True, | 19:54 |
| rm_work | johnsom: anywho... leave comments, probably it isn't perfect, but should be easy enough to make a few tweaks. i'm heading to bed | 19:55 |
| rm_work | basically i am hoping it should pass the scenarios | 19:57 |
| *** Aju has quit IRC | 20:05 | |
| rm_work | ... i lied about going to bed, i'm watching the test runs | 20:06 |
| *** eezhova has joined #openstack-lbaas | 20:17 | |
| *** eezhova has quit IRC | 20:17 | |
| *** pcaruana has quit IRC | 20:18 | |
| rm_work | looking good | 20:20 |
| rm_work | if there's regressions, it should be on incidentals, like cert replacement reloading and such | 20:21 |
| rm_work | the server itself seems to work fine | 20:21 |
| rm_work | how were you able to reproduce issues with werkzeug? | 20:21 |
| rm_work | if you have a CR that causes issues to surface, can you rebase it on mine and see what happens? | 20:22 |
| *** strigazi is now known as strigazi_AFK | 20:22 | |
| *** _ducttape_ has joined #openstack-lbaas | 20:22 | |
| diltram | listener, loadbalancer and minimal passed | 20:22 |
| rm_work | yeah pool is about to pass, watching it | 20:23 |
| *** ducttape_ has quit IRC | 20:23 | |
| rm_work | scenario is still setting up devstack >_> | 20:23 |
| rm_work | i think the ovh host is shitty | 20:23 |
| rm_work | wait, i think those all use noop >_> | 20:24 |
| rm_work | erg | 20:24 |
| rm_work | prolly need to wait for scenario | 20:25 |
| rm_work | lol scenario finally starting to actually run tests | 20:26 |
| rm_work | took 35 minutes for devstack to install on that node T_T | 20:27 |
| diltram | hahahaha | 20:27 |
| rm_work | i'm not sure if this will even be definitive, might time out waiting for the amphora to come up, but because the host is bad, not because the change didn't work >_< | 20:29 |
| rm_work | do me a favor and recheck it if it fails? I'm going to go to bed, can't wait another 30m for this to actually spin an amp >_< | 20:30 |
| diltram | yeah, I will be here to take a look :P | 20:31 |
| rm_work | ah, just as i was about to lock my screen, first test passed | 20:32 |
| rm_work | so obviously it started the agent properly | 20:33 |
| rm_work | is johnsom back from lunch yet? :P | 20:34 |
| johnsom | Yep, just catching up | 20:34 |
| rm_work | the way I implemented gunicorn is the least disruptive to the current workflow | 20:35 |
| diltram | looks working :P | 20:35 |
| rm_work | we could go as deep as switching out internals of the config and upstart init to make it just run gunicorn directly | 20:36 |
| rm_work | but that's a lot more work | 20:36 |
| johnsom | I know Brandon proposed gunicorn before | 20:36 |
| rm_work | yeah | 20:36 |
| rm_work | i did too | 20:36 |
| rm_work | i just never had time to do it | 20:36 |
| rm_work | and it looks like we waited the appropriate amount of time, because 19.0 added the ability to do what i'm doing | 20:36 |
| johnsom | Yeah, I think we have fought with issues on the old platform long enough | 20:38 |
| diltram | second is ok | 20:38 |
| diltram | so it's working | 20:38 |
| rm_work | yep | 20:40 |
| rm_work | just responded to dougwig on the CR | 20:40 |
| rm_work | basically, re: global-reqs, IDGAF | 20:40 |
| rm_work | does anyone else care? :/ | 20:42 |
| openstackgerrit | Stephen Balukoff proposed openstack/octavia: Load system sysctl in amphora network namespace https://review.openstack.org/386328 | 20:43 |
| diltram | rm_work: you're right in my opinion | 20:44 |
| diltram | so if go is not approved as programing language we can't use this new http go smth? | 20:44 |
| openstackgerrit | Stephen Balukoff proposed openstack/octavia: Update haproxy logging configuration https://review.openstack.org/386427 | 20:50 |
| diltram | passed | 20:50 |
| dougwig | diltram: that merged. water under the bridge. | 20:55 |
| diltram | I don't understad what do you mean? | 20:55 |
| dougwig | rm_work: replied replied. | 20:56 |
| dougwig | diltram: the test http server is merged; it's all good. | 20:56 |
| rm_work | dougwig: besides it being in g-r.txt or not, I still don't want to add gunicorn to our requirements.txt because it isn't needed for octavia to run, it is ONLY for the amphorae | 20:56 |
| dougwig | rm_work: right, see my reply | 20:56 |
| rm_work | i was about to say "i could maybe be convinced to add an amphorae-requirements.txt" | 20:56 |
| diltram | dougwig: +1 | 20:57 |
| dougwig | i will assume the "maybe" in that sentence is due to too much sake. | 20:57 |
| rm_work | the issue with this is that now the packages that ARE in global-reqs aren't going to be tracked | 20:59 |
| rm_work | re: versions and such | 20:59 |
| rm_work | (I mean, the bot won't update them anymore) | 21:00 |
| sbalukoff | Hmmm.... | 21:00 |
| rm_work | so, see my new patchset, but i don't like it a whole lot | 21:00 |
| rm_work | err, momentarily | 21:00 |
| diltram | but even adding anything into dib elements will not be tracked by bot | 21:01 |
| sbalukoff | What's the license on gunicorn? | 21:01 |
| rm_work | yeah | 21:01 |
| johnsom | Ok, just getting back into the conversation after a phone call. So, gunicorn is not openstack approved? | 21:01 |
| rm_work | so, no difference? | 21:01 |
| rm_work | johnsom: it's not in g-r but #infra seems to think there wouldn't be a fight to get it there | 21:01 |
| dougwig | but we don't want it in requirements.txt if it's not required for the server to run, i'd think | 21:02 |
| rm_work | right | 21:02 |
| rm_work | so even if it were in g-r i wouldn't want to add it to requirements | 21:02 |
| johnsom | Where did we put pyroute2? | 21:02 |
| rm_work | requirements.txt | 21:03 |
| rm_work | which sucks | 21:03 |
| rm_work | but i moved it | 21:03 |
| rm_work | urg connection issues | 21:03 |
| rm_work | ok should post shortly | 21:03 |
| openstackgerrit | Adam Harwell proposed openstack/octavia: Run amphora agent with gunicorn https://review.openstack.org/386758 | 21:03 |
| johnsom | Yeah, so, it should be there. We use that when we build the image I think | 21:03 |
| rm_work | ^^ see that | 21:03 |
| *** cody-somerville has quit IRC | 21:03 | |
| *** cody-somerville has joined #openstack-lbaas | 21:04 | |
| *** cody-somerville has quit IRC | 21:04 | |
| *** cody-somerville has joined #openstack-lbaas | 21:04 | |
| rm_work | but yeah, now none of those things are tracked by the openstack bot | 21:04 |
| johnsom | Yeah, so why move it out? | 21:04 |
| rm_work | I've never been a fan of installing that stuff in requirements.txt | 21:05 |
| rm_work | but if we're going to have an amphora-requirements.txt then that's where it should live | 21:05 |
| johnsom | Fair, but that argues for moving the agent into it's own project | 21:05 |
| dougwig | i'd bet we could modify the bot to check for additional requirement files, if that's the only holdup | 21:05 |
| rm_work | honestly i liked how patchset one went | 21:05 |
| rm_work | dougwig: that'd be interesting, worth investigating | 21:05 |
| dougwig | or a separate repo, as johnsom suggests. | 21:05 |
| johnsom | dougwig +1 | 21:06 |
| dougwig | rm_work: your first patchset was a disgusting hack | 21:06 |
| rm_work | yeah, and that's been long in the works | 21:06 |
| rm_work | i actually tried splitting it once but it was going to be a lot more work than i had time for | 21:06 |
| johnsom | Hahaha, we are going to end up with octavia-lib | 21:06 |
| *** matt-borland has quit IRC | 21:09 | |
| sbalukoff | Heh! Probably. | 21:10 |
| sbalukoff | And whatever pattern we follow for the amphora image is going to follow for the distributor image. :P | 21:11 |
| diltram | but even splitting amphora agent into new repo will be good | 21:11 |
| rm_work | yes but not in scope for this >_< | 21:11 |
| sbalukoff | True. | 21:11 |
| sbalukoff | diltram: That's where we end up getting octavia-lib. | 21:11 |
| diltram | Anyone knows what part of code is required by base octavia and amphora agent? | 21:12 |
| diltram | as a shared code? | 21:12 |
| *** cody-somerville has quit IRC | 21:13 | |
| sbalukoff | Not off the top of my head, but I would guess many things from octavia/common. | 21:13 |
| johnsom | Not off their head, probably not. | 21:13 |
| johnsom | octavia/amphorae/backends down would be a start | 21:13 |
| rm_work | i actually do | 21:13 |
| rm_work | most of it | 21:13 |
| rm_work | it is a lot | 21:13 |
| johnsom | But, I think we need to step back from that edge. | 21:13 |
| rm_work | i have an abandoned CR for it... | 21:13 |
| rm_work | the rabbit-hole was deeeep | 21:14 |
| johnsom | I'm not sure now is the right time to try to do that split. | 21:14 |
| rm_work | IMO it is not | 21:14 |
| johnsom | I still don't see the advantage of the amphora-requirements if you are still installing them all in the amp. I commented as such on the patch. | 21:16 |
| rm_work | ? | 21:17 |
| rm_work | amphora-requirements.txt is for things that are ONLY required in the amphora | 21:17 |
| rm_work | we absolutely need cryptography, but it's also needed by the main app | 21:17 |
| johnsom | So we are saving a few packages from the controllers? | 21:18 |
| rm_work | yes :/ | 21:18 |
| rm_work | maybe this is not worth the optimization? | 21:18 |
| rm_work | since the amp is really where optimizing matters, and we aren't | 21:18 |
| johnsom | It seems like saving space in the amp is more important | 21:18 |
| rm_work | well, as we're talking, i'm TRYING to propose gunicorn to g-r | 21:19 |
| rm_work | but having environment issues | 21:19 |
| dougwig | with the ssh driver, none. :) | 21:21 |
| rm_work | ^^ :P | 21:22 |
| rm_work | all this nonsense goes away | 21:22 |
| rm_work | ah, maybe now IS a good time to resurrect that | 21:22 |
| dougwig | since amps are meant to be black boxes, and we've intermingled this much already, that is arguing really well that we should separate those repos. IMO. | 21:22 |
| * johnsom glares at you people | 21:22 | |
| rm_work | and do it the way i wanted to do it to begin with.. | 21:22 |
| * dougwig apologizes for the ssh troll, even if it is superior. | 21:22 | |
| rm_work | aww, thought you were serious :( still really want the ssh driver back | 21:23 |
| rm_work | though, we still need a process of some sort to do heartbeats | 21:23 |
| diltram | in my opinion if Adam has time and don't want to work on something more urgent he has a green light to make that split | 21:23 |
| dougwig | i want it back too, but the team has decided, so let it lie. | 21:23 |
| rm_work | :/ | 21:25 |
| johnsom | diltram my concern is it would be a distraction for reviewing, splitting in-flight patches (act/act), ... I of course can't/will not stop him, but it has some downside to do it now. | 21:25 |
| diltram | true | 21:25 |
| johnsom | Plus, they are arguing for a new amp driver over SSH, which didn't really work the first time (yes, my opinion) | 21:26 |
| diltram | so like I said - if he don't want to help with merge :P | 21:26 |
| rm_work | FYI, I in no way want to deal with that split right now :P | 21:26 |
| rm_work | "new" lol | 21:26 |
| rm_work | I guess technically | 21:26 |
| diltram | but what is value added of ssh driver? | 21:27 |
| dougwig | i mean, this entire argument is because we have two projects in one. separate repos, or separate requirements files, or ... we end up with hacks in the bowels of scripts or just LCD. pick your poison. | 21:27 |
| johnsom | Hey, I said "my opinion"... grin | 21:27 |
| rm_work | ugh g-r is not working here | 21:27 |
| rm_work | because i don't have *system-level* python3 | 21:27 |
| dougwig | diltram: the "image" becomes native ubuntu or cirros, and no code to update on the amp, ever. | 21:27 |
| rm_work | dougwig: so, technically not true unfortunately, unless we deploy and run a heartbeat/stats-sender via sftp | 21:28 |
| diltram | ok - understand | 21:28 |
| * johnsom thinks dougwig started his weekend early... | 21:28 | |
| rm_work | can someone else do this proposal? T_T | 21:28 |
| rm_work | global-requirements proposals require this constraints hackery and i can't get their stuff to run right on my box | 21:29 |
| rm_work | it REALLY wants system-level python2 and python3 and i don't have both (I use py3 via pyenv) | 21:29 |
| dougwig | rm_work: i can. what's the g-r line, including license? | 21:31 |
| *** portdirect_ has joined #openstack-lbaas | 21:31 | |
| *** portdirect has quit IRC | 21:31 | |
| *** portdirect_ is now known as portdirect | 21:31 | |
| rm_work | gunicorn>=19.0 # MIT | 21:32 |
| dougwig | rm_work: https://review.openstack.org/386790 | 21:37 |
| dougwig | why are the two of us, of all people, doing work to maintain the crappy rest agent? | 21:38 |
| dougwig | sorry, that slipped out | 21:38 |
| openstackgerrit | Adam Harwell proposed openstack/octavia: Run amphora agent with gunicorn https://review.openstack.org/386758 | 21:38 |
| * johnsom glares again | 21:38 | |
| rm_work | dougwig: should I depends-on: I211dc771aa95147c0f1d9e6ac1a65a7e164b33c2 | 21:38 |
| rm_work | ? | 21:38 |
| johnsom | Yes | 21:39 |
| openstackgerrit | Adam Harwell proposed openstack/octavia: Run amphora agent with gunicorn https://review.openstack.org/386758 | 21:40 |
| rm_work | done | 21:40 |
| rm_work | happy now? you monsters | 21:40 |
| johnsom | Ha | 21:40 |
| johnsom | Do you want to expand on the text of Doug's patch? It's, um, basic | 21:41 |
| rm_work | yeah uhhh | 21:41 |
| rm_work | was thinking about it | 21:41 |
| dougwig | wait and see what happens. more words is just more things to argue about. | 21:41 |
| johnsom | Ha, true | 21:42 |
| johnsom | I hope this fixes all the strangeness we have had with ports getting lost, requests not getting responded to, or partial responses | 21:43 |
| rm_work | whelp, already responded | 21:43 |
| dougwig | rm_work: but if you change it, do add a co-author. forgot that. | 21:43 |
| johnsom | If it does I will buy rm_work a tapas | 21:43 |
| rm_work | heh | 21:43 |
| rm_work | yeah, uhh... so, how's it going, people with corp cards? :P | 21:44 |
| rm_work | who wants to get some dinner? ^_^ | 21:44 |
| rm_work | i tried to keep my response pretty short but to the point | 21:45 |
| rm_work | what channel do g-r changes go to? | 21:45 |
| dougwig | infra? | 21:47 |
| rm_work | I thought so but i don't see it there | 21:47 |
| rm_work | ah well, now that it's 7am i should head to bed >_> | 21:47 |
| sbalukoff | Sleep well, man. | 21:50 |
| openstackgerrit | Adam Harwell proposed openstack/octavia: Add quota support to Octavia https://review.openstack.org/360794 | 21:54 |
| rm_work | ^^ missed updating the migration downrevision >_> | 21:54 |
| rm_work | oh hey diltram it looks like it is ACTUALLY testing quotas now? :P | 22:00 |
| rm_work | see failures http://sonar.octavia.io/archive/octavia/1c992690482cac6ae4ac18d555f2a23e13358e68/2394/console_log.txt.gz | 22:00 |
| rm_work | as you were saying, the test was using the wrong default | 22:00 |
| rm_work | i think that was tyou | 22:00 |
| openstackgerrit | Adam Harwell proposed openstack/octavia: Add quota support to Octavia https://review.openstack.org/360794 | 22:03 |
| diltram | here | 22:03 |
| diltram | https://review.openstack.org/#/c/360794/33/etc/octavia.conf | 22:03 |
| diltram | default quota is 20 | 22:03 |
| rm_work | ah for just the one | 22:03 |
| rm_work | ok need to update again | 22:03 |
| openstackgerrit | Adam Harwell proposed openstack/octavia: Add quota support to Octavia https://review.openstack.org/360794 | 22:04 |
| rm_work | there we go | 22:04 |
| rm_work | corrected | 22:05 |
| rm_work | and NOW sleep maybe T_T | 22:05 |
| johnsom | Ugh, this is going to be a fun weekend. The power is already blipping on and off | 22:10 |
| dougwig | johnsom: got your generator and plenty of beer? | 22:38 |
| johnsom | Batteries, natural gas, and yes, beer | 22:39 |
| johnsom | wine, bourbon, etc. | 22:39 |
| dougwig | then it's just like camping. with more water and looting. | 22:39 |
| johnsom | Just the basics | 22:39 |
| *** _ducttape_ has quit IRC | 23:05 | |
| openstackgerrit | Merged openstack/octavia: Load system sysctl in amphora network namespace https://review.openstack.org/386328 | 23:10 |
| openstackgerrit | Dustin Lundquist proposed openstack/octavia: Scenario: test with concurrent clients https://review.openstack.org/385640 | 23:20 |
| *** afranc has joined #openstack-lbaas | 23:28 | |
| *** gongysh has joined #openstack-lbaas | 23:37 | |
| *** ducttape_ has joined #openstack-lbaas | 23:38 | |
| openstackgerrit | Dustin Lundquist proposed openstack/octavia: Scenario: test with concurrent clients https://review.openstack.org/385640 | 23:41 |
| *** bana_k has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!