| *** yamamoto has joined #openstack-lbaas | 00:03 | |
| *** yamamoto has quit IRC | 00:09 | |
| *** hongbin has joined #openstack-lbaas | 00:40 | |
| *** longkb has joined #openstack-lbaas | 00:40 | |
| openstackgerrit | ZhaoBo proposed openstack/neutron-lbaas master: Fix the None value of Q_PLUGIN_CONF_FILE if enable neutron-* https://review.openstack.org/575687 | 00:55 |
|---|---|---|
| *** yamamoto has joined #openstack-lbaas | 01:05 | |
| *** yamamoto has quit IRC | 01:10 | |
| *** openstack has joined #openstack-lbaas | 01:20 | |
| *** ChanServ sets mode: +o openstack | 01:20 | |
| *** phuoc has joined #openstack-lbaas | 01:23 | |
| *** ramishra has joined #openstack-lbaas | 01:54 | |
| *** yamamoto has joined #openstack-lbaas | 02:07 | |
| *** yamamoto has quit IRC | 02:12 | |
| *** annp has joined #openstack-lbaas | 02:31 | |
| *** yamamoto has joined #openstack-lbaas | 03:09 | |
| *** yamamoto has quit IRC | 03:14 | |
| openstackgerrit | Merged openstack/octavia master: Fix health manager to be spare amphora aware https://review.openstack.org/576668 | 03:16 |
| *** gans has joined #openstack-lbaas | 03:27 | |
| openstackgerrit | Merged openstack/octavia master: Change tox.ini coverage to always generate html https://review.openstack.org/577321 | 03:27 |
| *** yamamoto has joined #openstack-lbaas | 04:10 | |
| *** yamamoto has quit IRC | 04:15 | |
| *** hongbin has quit IRC | 04:22 | |
| *** nmanos has joined #openstack-lbaas | 04:46 | |
| *** nmanos has quit IRC | 05:08 | |
| *** nmanos has joined #openstack-lbaas | 05:09 | |
| *** yamamoto has joined #openstack-lbaas | 05:12 | |
| *** yamamoto has quit IRC | 05:18 | |
| *** yamamoto has joined #openstack-lbaas | 05:30 | |
| *** nmanos has quit IRC | 05:43 | |
| *** nmanos has joined #openstack-lbaas | 05:44 | |
| openstackgerrit | Adit Sarfaty proposed openstack/octavia master: Providers: propose changes to create pool API https://review.openstack.org/574241 | 06:22 |
| *** yboaron has quit IRC | 06:36 | |
| *** ispp has joined #openstack-lbaas | 06:45 | |
| *** ispp has quit IRC | 06:47 | |
| *** gans has quit IRC | 06:59 | |
| *** kobis has joined #openstack-lbaas | 06:59 | |
| *** ispp has joined #openstack-lbaas | 07:05 | |
| *** pcaruana has joined #openstack-lbaas | 07:07 | |
| *** tesseract has joined #openstack-lbaas | 07:20 | |
| *** luksky has joined #openstack-lbaas | 07:46 | |
| *** velizarx has joined #openstack-lbaas | 07:50 | |
| *** longkb1 has joined #openstack-lbaas | 07:57 | |
| *** yboaron has joined #openstack-lbaas | 07:58 | |
| *** longkb has quit IRC | 07:59 | |
| *** longkb has joined #openstack-lbaas | 08:21 | |
| *** longkb1 has quit IRC | 08:23 | |
| *** sapd has quit IRC | 08:59 | |
| openstackgerrit | ZhaoBo proposed openstack/octavia master: UDP jinja template https://review.openstack.org/525420 | 09:15 |
| openstackgerrit | ZhaoBo proposed openstack/octavia master: UDP for [2] https://review.openstack.org/529651 | 09:15 |
| *** sapd has joined #openstack-lbaas | 09:23 | |
| *** salmankhan has joined #openstack-lbaas | 09:31 | |
| openstackgerrit | ZhaoBo proposed openstack/octavia master: UDP for [2] https://review.openstack.org/529651 | 09:39 |
| *** kobis has quit IRC | 09:53 | |
| *** peereb has joined #openstack-lbaas | 09:59 | |
| *** peereb has quit IRC | 10:01 | |
| *** peereb has joined #openstack-lbaas | 10:01 | |
| *** peereb has quit IRC | 10:02 | |
| *** peereb has joined #openstack-lbaas | 10:03 | |
| *** ispp has quit IRC | 10:03 | |
| *** peereb has quit IRC | 10:04 | |
| *** peereb has joined #openstack-lbaas | 10:04 | |
| *** peereb has quit IRC | 10:05 | |
| *** kobis has joined #openstack-lbaas | 10:19 | |
| *** atoth has joined #openstack-lbaas | 11:15 | |
| *** ispp has joined #openstack-lbaas | 11:25 | |
| *** ispp has quit IRC | 11:34 | |
| *** ispp has joined #openstack-lbaas | 11:40 | |
| *** annp has quit IRC | 11:44 | |
| *** ispp has quit IRC | 11:46 | |
| *** ispp has joined #openstack-lbaas | 11:46 | |
| *** longkb has quit IRC | 11:49 | |
| *** amuller has joined #openstack-lbaas | 11:54 | |
| *** phuoc has quit IRC | 11:56 | |
| *** phuoc has joined #openstack-lbaas | 11:57 | |
| *** ispp has quit IRC | 12:05 | |
| *** velizarx has quit IRC | 12:15 | |
| *** velizarx has joined #openstack-lbaas | 12:43 | |
| *** ramishra has quit IRC | 13:04 | |
| *** ramishra has joined #openstack-lbaas | 13:06 | |
| openstackgerrit | ZhaoBo proposed openstack/octavia master: UDP for [2] https://review.openstack.org/529651 | 13:07 |
| openstackgerrit | ZhaoBo proposed openstack/octavia master: UDP for [3][5][6] https://review.openstack.org/539391 | 13:07 |
| *** KeithMnemonic has joined #openstack-lbaas | 13:08 | |
| *** luksky has quit IRC | 13:19 | |
| *** yboaron_ has joined #openstack-lbaas | 13:29 | |
| *** ispp has joined #openstack-lbaas | 13:30 | |
| *** yboaron has quit IRC | 13:31 | |
| *** kobis has quit IRC | 13:48 | |
| *** yboaron_ has quit IRC | 13:59 | |
| *** kobis has joined #openstack-lbaas | 14:08 | |
| *** blake has joined #openstack-lbaas | 14:11 | |
| *** kobis has quit IRC | 14:11 | |
| *** kobis has joined #openstack-lbaas | 14:12 | |
| *** yboaron_ has joined #openstack-lbaas | 14:13 | |
| *** kobis has quit IRC | 14:16 | |
| *** kobis has joined #openstack-lbaas | 14:24 | |
| *** yamamoto has quit IRC | 14:26 | |
| openstackgerrit | ZhaoBo proposed openstack/octavia master: UDP for [3][5][6] https://review.openstack.org/539391 | 14:27 |
| dulek | Hey, we've started seeing this lately: http://logs.openstack.org/60/577760/1/check/kuryr-kubernetes-tempest-daemon-containerized-octavia/a4a1048/controller/logs/screen-o-api.txt.gz#_Jun_25_12_01_23_977628 | 14:33 |
| dulek | Had you changed something around amphora IP allocation lately? | 14:33 |
| dulek | I've tried looking up who's reserving that IP before Octavia, but first reservation in Neutron log happens when Octavia gets create LB request… | 14:34 |
| *** phuoc has quit IRC | 14:40 | |
| *** phuoc has joined #openstack-lbaas | 14:45 | |
| *** fnaval has joined #openstack-lbaas | 14:46 | |
| *** kobis has quit IRC | 14:47 | |
| *** phuoc has quit IRC | 14:49 | |
| *** phuoc has joined #openstack-lbaas | 14:54 | |
| *** phuoc has quit IRC | 14:58 | |
| *** sapd1 has joined #openstack-lbaas | 14:59 | |
| sapd1 | nmagnezi: Could you help me write unit test? | 15:06 |
| johnsom | dulek Hi, we actually don't do anything to specifically allocate certain IPs. Unless the user specifies one, all of the IP allocation is done by neutron for us. Either via DHCP or neutron passing us an IP. | 15:13 |
| *** pcaruana has quit IRC | 15:13 | |
| *** yamamoto has joined #openstack-lbaas | 15:16 | |
| *** kobis has joined #openstack-lbaas | 15:17 | |
| *** kobis has quit IRC | 15:18 | |
| *** velizarx has quit IRC | 15:20 | |
| *** phuoc has joined #openstack-lbaas | 15:21 | |
| johnsom | dulek It looks like that error is from the VIP, so I wonder if that test is passing in a fixed IP. | 15:21 |
| *** phuoc has quit IRC | 15:25 | |
| *** blake has quit IRC | 15:35 | |
| openstackgerrit | sapd proposed openstack/octavia master: Exclude amphora instances already DELETED when failover loadbalancer https://review.openstack.org/577122 | 15:38 |
| *** phuoc has joined #openstack-lbaas | 15:40 | |
| openstackgerrit | sapd proposed openstack/octavia master: Exclude amphora instances already DELETED when failover loadbalancer https://review.openstack.org/577122 | 15:41 |
| *** phuoc has quit IRC | 15:45 | |
| *** ramishra has quit IRC | 15:47 | |
| *** kobis has joined #openstack-lbaas | 15:56 | |
| *** phuoc has joined #openstack-lbaas | 15:59 | |
| *** ispp has quit IRC | 16:01 | |
| *** phuoc_ has joined #openstack-lbaas | 16:05 | |
| *** phuoc has quit IRC | 16:06 | |
| *** yboaron_ has quit IRC | 16:06 | |
| openstackgerrit | Carlos Goncalves proposed openstack/octavia master: Add upgrade guide https://review.openstack.org/577295 | 16:08 |
| *** phuoc_ has quit IRC | 16:15 | |
| *** tesseract has quit IRC | 16:18 | |
| *** AlexeyAbashkin has joined #openstack-lbaas | 16:28 | |
| openstackgerrit | Carlos Goncalves proposed openstack/octavia master: Add upgrade guide https://review.openstack.org/577295 | 16:33 |
| openstackgerrit | Merged openstack/neutron-lbaas master: Fix the None value of Q_PLUGIN_CONF_FILE if enable neutron-* https://review.openstack.org/575687 | 16:42 |
| *** velizarx has joined #openstack-lbaas | 16:43 | |
| *** kobis has quit IRC | 16:44 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Fix configured haproxy restarts https://review.openstack.org/577238 | 16:50 |
| *** Alexey_Abashkin has joined #openstack-lbaas | 17:04 | |
| *** AlexeyAbashkin has quit IRC | 17:07 | |
| *** AlexeyAbashkin has joined #openstack-lbaas | 17:07 | |
| *** velizarx has quit IRC | 17:07 | |
| *** Alexey_Abashkin has quit IRC | 17:09 | |
| openstackgerrit | sapd proposed openstack/octavia master: Exclude amphora instances already DELETED when failover loadbalancer https://review.openstack.org/577122 | 17:12 |
| *** salmankhan has quit IRC | 17:33 | |
| *** AlexeyAbashkin has quit IRC | 17:49 | |
| *** harlowja has joined #openstack-lbaas | 17:55 | |
| harlowja | whatsss up | 17:55 |
| johnsom | harlowja Hey, welcome back from vacation Grin | 17:58 |
| harlowja | :-P | 17:58 |
| sapd1 | :) | 17:58 |
| johnsom | So, I am 90% sure I have asked this before, so get the ruler out to slap my wrist | 17:58 |
| johnsom | I have a solution, but wondered if I could do this a better way. | 17:58 |
| harlowja | sup | 17:58 |
| johnsom | I have an unordered flow, that re-merges into a main flow. | 17:59 |
| johnsom | I want to make it "ok" for some of the unordered sub-flows to fail, and not revert the parent main flow. | 17:59 |
| johnsom | Is there a way to do that? I played with retry, but it seemed like it always reverted the parent | 17:59 |
| johnsom | End game, I want it to try to update the configs on all of the amphora in a LB, and passively fail (mark them in ERROR) should they not be reachable. The use case is when more than one of the amps in an LB fail at the same time. | 18:00 |
| johnsom | I worked around this in my tasks, but wondered if there is a better way to do "best effort" sub flows | 18:01 |
| harlowja | hmmm | 18:13 |
| * harlowja thinking | 18:14 | |
| * harlowja paging back in taskflow | 18:14 | |
| johnsom | Take your time, it's mostly intellectual at this point as I worked around it. | 18:14 |
| harlowja | i think i'd need a kind of 'CONTINUE/SKIP' enum for retrying | 18:16 |
| harlowja | i can add, one does not seem to exist | 18:18 |
| *** Swami has joined #openstack-lbaas | 18:31 | |
| *** JudeC_ has joined #openstack-lbaas | 18:40 | |
| *** Swami_ has joined #openstack-lbaas | 18:48 | |
| *** sapd1 has quit IRC | 18:55 | |
| *** atoth has quit IRC | 18:58 | |
| *** amuller has quit IRC | 19:27 | |
| *** pcaruana has joined #openstack-lbaas | 19:27 | |
| *** yamamoto has quit IRC | 19:44 | |
| *** blake has joined #openstack-lbaas | 19:45 | |
| *** yamamoto has joined #openstack-lbaas | 19:46 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Clarify IPv6 support for the lb-mgmt-net https://review.openstack.org/577913 | 19:48 |
| *** yamamoto has quit IRC | 19:49 | |
| *** yamamoto has joined #openstack-lbaas | 19:49 | |
| *** yamamoto has quit IRC | 19:49 | |
| *** JudeC__ has joined #openstack-lbaas | 20:02 | |
| *** JudeC_ has quit IRC | 20:02 | |
| *** JudeC_ has joined #openstack-lbaas | 20:05 | |
| *** JudeC__ has quit IRC | 20:09 | |
| *** KeithMnemonic has quit IRC | 20:40 | |
| *** pcaruana has quit IRC | 20:43 | |
| *** yamamoto has joined #openstack-lbaas | 20:50 | |
| *** yamamoto has quit IRC | 20:58 | |
| *** jlaffaye_ has joined #openstack-lbaas | 21:01 | |
| *** jlaffaye_ has quit IRC | 21:03 | |
| *** jlaffaye_ has joined #openstack-lbaas | 21:03 | |
| *** harlowja has quit IRC | 21:04 | |
| *** jlaffaye has quit IRC | 21:06 | |
| *** fnaval has quit IRC | 21:37 | |
| *** rcernin has joined #openstack-lbaas | 21:45 | |
| *** yamamoto has joined #openstack-lbaas | 21:54 | |
| *** yamamoto has quit IRC | 21:59 | |
| lxkong | good morning, all. I came here to ask question again :-) | 22:12 |
| lxkong | does octavia supports feature similar to this https://aws.amazon.com/blogs/aws/elb-connection-draining-remove-instances-from-service-with-care/? | 22:12 |
| *** harlowja has joined #openstack-lbaas | 22:14 | |
| openstackgerrit | Carlos Goncalves proposed openstack/octavia master: Add upgrade guide https://review.openstack.org/577295 | 22:22 |
| johnsom | lxkong Yes, set the member weight to zero | 22:23 |
| lxkong | johnsom: awesome, i will test now | 22:23 |
| *** blake has quit IRC | 22:50 | |
| *** yamamoto has joined #openstack-lbaas | 22:56 | |
| *** yamamoto has quit IRC | 23:02 | |
| lxkong | johnsom: i'm a little confused. When updating the weight, i can see `/usr/sbin/service haproxy-{listener_id} reload`, will that lead to interruption to the current connections? | 23:25 |
| johnsom | No | 23:25 |
| johnsom | lxkong reload just tells HAProxy to reload it's configuration. It will spawn a second temporary process and do a graceful hand off | 23:26 |
| johnsom | lxkong I will say, on versions prior to Ubuntu bionic, there is a *slight* chance a new connection could get blocked, but it's pretty small. | 23:27 |
| johnsom | lxkong A restart would interrupt traffic | 23:27 |
| lxkong | johnsom: i just tested that setting the weight to 0 on one member causes new requirests all go to another one. But I really need to think about how to verify the connections on the first one still alive | 23:28 |
| johnsom | lxkong you could setup a "simple" web server with netcat that streams /dev/zero or /dev/urandom | 23:30 |
| lxkong | johnsom: so, now it depends on the end user to manually set weight to 0 before manually removing the member from the pool, right? | 23:32 |
| johnsom | { echo -e "HTTP/1.0 200 OK\r\n\r\n"; cat /dev/zero } | sudo nc -l -p 80 -w 1 | 23:32 |
| johnsom | lxkong Correct | 23:32 |
| lxkong | johnsom: i am using `while true; do echo -e "HTTP/1.1 200 OK\r\n\r\n${OUTPUT}\r" | sudo nc -l -p 80; done` | 23:32 |
| johnsom | Sure, there you go | 23:33 |
| johnsom | Well, wait, that just does multiple connections, not a large transfer that would test it's not interrupted | 23:33 |
| lxkong | yeah, need modification | 23:34 |
| lxkong | johnsom: does that make sense we support some param when removing member? so the member weight could be set to 0 and stay for some time before actually removed from the pool? | 23:35 |
| lxkong | like other clouds do | 23:35 |
| johnsom | lxkong That seems pretty proprietary, I don't think any (or many) of the hardware load balancers do that | 23:36 |
| johnsom | I mean it can stay weight zero forever, or use heat to set it to zero and then remove the member. | 23:37 |
| lxkong | johnsom: ok, i know, it's up to the orchestration tooling the user is using | 23:38 |
| johnsom | lxkong Even that AWS blog post says deregistering the instance is an additional step. | 23:38 |
| lxkong | what's the default value of weight if not specified? | 23:38 |
| johnsom | 1 | 23:39 |
| lxkong | ok, i will do some more tests, thank for all the answers | 23:39 |
| johnsom | Sure, NP | 23:40 |
| *** yamamoto has joined #openstack-lbaas | 23:57 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!