| openstackgerrit | Merged openstack/octavia master: Deprecate the Spares Pool feature for removal in X https://review.opendev.org/741686 | 00:44 |
|---|---|---|
| rm_work | johnsom: seeing some weirdness possibly again in failover, looking into it, just a heads up in case you were also aware of any issues | 01:21 |
| rm_work | ran into `octavia.network.base.PortNotFound: port not found` during a failover of an LB that had a user's bring-your-own-port | 01:22 |
| rm_work | still trying to track down what caused the failover to begin with, it happened while i was out, and everything is gone now, so sifting through logs | 01:23 |
| johnsom | No, I don’t know of any open issues with failover | 01:23 |
| rm_work | kk | 01:23 |
| rm_work | failover failed, ended up in ERROR with 4 amp records <_< | 01:23 |
| rm_work | manual failover fixed it somehow | 01:23 |
| johnsom | That seems like the old code behavior | 01:24 |
| johnsom | Is like the worker updated and the hm not? | 01:25 |
| rm_work | don't think that's possible given how many iterations it's been, but yeah i will look | 01:25 |
| rm_work | if we are still running OOOOLD HM code somehow, I will boggle | 01:26 |
| rm_work | hmm looks like cert refresh failed due to a timeout O_o | 01:47 |
| rm_work | and then... it tried to replace the amp maybe? | 01:49 |
| rm_work | or... some other failover was attempted actually much later | 01:49 |
| rm_work | and that's when the port not found error happened -- trying to figure out which port | 01:49 |
| rm_work | `vrrp_port = self.get_port(amp.vrrp_port_id)` | 01:51 |
| rm_work | so, the vrrp port didn't exist... weird, i thought that was a solved issue | 01:51 |
| rm_work | 1) we might want to retry on a connection timeout on a cert rotation? | 01:57 |
| openstackgerrit | Merged openstack/octavia master: Refactoring amphora stats driver interface https://review.opendev.org/737111 | 02:10 |
| rm_work | `1 retries with interval 5 seconds have failed for retry-octavia-delete-amphora-flow-07a4407d-35e7-49e5-a18f-183c965a9527. Giving up.` then it does a "revert-delete-port" but obviously that's bogus, you can't revert that :D | 02:11 |
| rm_work | then the port doesn't exist for the failover | 02:11 |
| rm_work | which... i thought you made it create all new ports, but I remember there was one you had to reuse -- hmm | 02:12 |
| johnsom | It should recreate everything, but if we don’t own the port I think I don’t mess with it | 02:13 |
| rm_work | yeah this is the vrrp port anyway | 02:14 |
| rm_work | trying to figure out why a failover 30m later fails because it can't find the vrrp port that was deleted earlier | 02:15 |
| johnsom | That so seems like old code. The base ports are all rebuilt now | 02:15 |
| johnsom | Because nova is dumb and will not release them | 02:16 |
| rm_work | yeah.... doesn't look like old code involved though, which is why i'm confused. | 02:19 |
| rm_work | hmmm ok | 02:25 |
| rm_work | https://github.com/openstack/octavia/blob/master/octavia/controller/worker/v1/flows/amphora_flows.py#L650-L651 | 02:29 |
| rm_work | https://github.com/openstack/octavia/blob/master/octavia/controller/worker/v1/flows/amphora_flows.py#L320-L323 | 02:29 |
| *** rcernin has quit IRC | 02:29 | |
| rm_work | https://github.com/openstack/octavia/blob/master/octavia/network/drivers/neutron/allowed_address_pairs.py#L688 | 02:30 |
| rm_work | johnsom: ^^ and there is where no matter what happens, during a failover it will try to get the vrrp port, fail, and revert | 02:31 |
| rm_work | if the vrrp port is gone | 02:31 |
| rm_work | if i'm reading this right | 02:31 |
| johnsom | Hmmm | 02:31 |
| johnsom | I will have to look tomorrow, having dinner now | 02:32 |
| johnsom | Seems like it should already have rebuilt that port by then, but that is off my head | 02:32 |
| *** armax has quit IRC | 03:10 | |
| *** armax has joined #openstack-lbaas | 03:10 | |
| *** zzzeek has quit IRC | 03:18 | |
| *** zzzeek has joined #openstack-lbaas | 03:19 | |
| *** sapd1_x has joined #openstack-lbaas | 03:27 | |
| *** rcernin has joined #openstack-lbaas | 03:29 | |
| *** rcernin has quit IRC | 03:29 | |
| *** rcernin has joined #openstack-lbaas | 03:29 | |
| *** zzzeek has quit IRC | 03:55 | |
| *** zzzeek has joined #openstack-lbaas | 03:58 | |
| *** sapd1_x has quit IRC | 04:03 | |
| *** sapd1_x has joined #openstack-lbaas | 04:04 | |
| *** njohnston has quit IRC | 04:22 | |
| *** sapd1_x has quit IRC | 04:35 | |
| *** zzzeek has quit IRC | 04:45 | |
| *** zzzeek has joined #openstack-lbaas | 04:48 | |
| *** zzzeek has quit IRC | 05:17 | |
| *** zzzeek has joined #openstack-lbaas | 05:18 | |
| *** gcheresh_ has joined #openstack-lbaas | 05:26 | |
| *** sapd1 has quit IRC | 05:48 | |
| *** sapd1 has joined #openstack-lbaas | 05:48 | |
| *** vishalmanchanda has joined #openstack-lbaas | 06:14 | |
| *** ataraday_ has joined #openstack-lbaas | 06:34 | |
| openstackgerrit | Ann Taraday proposed openstack/octavia stable/ussuri: Add missing reload method in amphora noop driver https://review.opendev.org/739064 | 06:42 |
| openstackgerrit | Ann Taraday proposed openstack/octavia stable/ussuri: Fix missing params in amphora base and noop driver https://review.opendev.org/739065 | 06:42 |
| openstackgerrit | Ann Taraday proposed openstack/octavia stable/ussuri: Fix missing dependency in amphora create flow https://review.opendev.org/742401 | 06:43 |
| *** ramishra has quit IRC | 06:51 | |
| *** ramishra has joined #openstack-lbaas | 07:24 | |
| *** sapd1_x has joined #openstack-lbaas | 07:52 | |
| *** rcernin has quit IRC | 08:20 | |
| gthiemonge | ataraday_: Hi, could you review https://review.opendev.org/#/c/714004/ please? | 08:21 |
| *** ianychoi__ has quit IRC | 08:40 | |
| kklimonda | Hi, I've deployed octavia on ussuri with ovn ml2 driver with amphora driver - it seems to be working fine, however I constantly see the following in ovn-controller logs: | 09:08 |
| kklimonda | ```pinctrl|INFO|Claiming virtual lport a5a3f456-622a-46fa-94e7-9805b153e640 for this chassis with the virtual parent b8f77b5e-6886-4506-ae55-7f76b374a07c``` | 09:08 |
| kklimonda | it seems to be repeating every 5 seconds for every vip port | 09:14 |
| *** sapd1_x has quit IRC | 09:15 | |
| *** lxkong has joined #openstack-lbaas | 09:28 | |
| ataraday_ | gthiemonge, Hi! Done | 09:29 |
| *** lxkong has quit IRC | 09:34 | |
| openstackgerrit | Ann Taraday proposed openstack/octavia stable/train: Fix memory consumption issues with default connection_limit https://review.opendev.org/747662 | 09:36 |
| *** lxkong has joined #openstack-lbaas | 09:39 | |
| ataraday_ | gthiemonge, Question about https://review.opendev.org/#/c/738381/ - do we still need some changes for amphora backend? | 09:46 |
| kklimonda | btw, could octavia/ovn combo be used to implement active-active LB, where ovn takes a role of a "director" and balances traffic to available amphorae? | 09:55 |
| gthiemonge | ataraday_: yes, amphora support will be included in a follow-up commit (in the next cycle), but some other providers might start implementing sctp now | 10:21 |
| openstackgerrit | Merged openstack/octavia stable/ussuri: Add missing reload method in amphora noop driver https://review.opendev.org/739064 | 10:43 |
| *** rcernin has joined #openstack-lbaas | 10:52 | |
| openstackgerrit | Merged openstack/octavia stable/ussuri: Fix missing params in amphora base and noop driver https://review.opendev.org/739065 | 11:07 |
| openstackgerrit | Merged openstack/octavia stable/ussuri: Fix missing dependency in amphora create flow https://review.opendev.org/742401 | 11:07 |
| *** sapd1_x has joined #openstack-lbaas | 11:17 | |
| ataraday_ | gthiemonge, OK, thanks! | 11:17 |
| *** rcernin has quit IRC | 11:28 | |
| openstackgerrit | Merged openstack/octavia-tempest-plugin master: Adjust scenario tests for NotImplemented skip https://review.opendev.org/714004 | 11:39 |
| *** rcernin has joined #openstack-lbaas | 11:45 | |
| *** sapd1_x has quit IRC | 11:56 | |
| openstackgerrit | Ann Taraday proposed openstack/octavia stable/ussuri: Update amphora v2 for the failover refactor https://review.opendev.org/750632 | 12:06 |
| *** jamesdenton has quit IRC | 12:08 | |
| *** njohnston has joined #openstack-lbaas | 12:08 | |
| *** jamesdenton has joined #openstack-lbaas | 12:09 | |
| openstackgerrit | Ann Taraday proposed openstack/octavia master: Bump taskflow version https://review.opendev.org/750638 | 12:23 |
| *** rcernin has quit IRC | 12:41 | |
| ataraday_ | I will miss today meeting, I want to raise for consideration this change https://review.opendev.org/#/c/750638/ - if we want to bump taskflow version in V. | 13:02 |
| openstackgerrit | Gregory Thiemonge proposed openstack/octavia stable/ussuri: Fix memory consumption issues with default connection_limit https://review.opendev.org/747651 | 13:05 |
| *** ataraday_ has quit IRC | 13:07 | |
| *** rcernin has joined #openstack-lbaas | 13:14 | |
| *** TrevorV has joined #openstack-lbaas | 13:24 | |
| *** sapd1_x has joined #openstack-lbaas | 13:27 | |
| *** rcernin has quit IRC | 13:41 | |
| johnsom | kklimonda OVN errors should be discussed in the #openstack-neutron channel. We can't help you much with errors in OVN. | 13:53 |
| kklimonda | fair enough | 13:56 |
| johnsom | kklimonda Also on the active-active topic. Maybe. Right now it doesn't support all of the load balancing capabilities that would be needed by a distributor. | 13:57 |
| kklimonda | johnsom: is there something I could read about what's lacking on ovn side? | 14:03 |
| johnsom | kklimonda https://docs.openstack.org/octavia/latest/user/feature-classification/index.html | 14:03 |
| rm_work | johnsom: yeah i am pretty sure this is an issue | 14:19 |
| rm_work | what is the best way to step through this with you? | 14:19 |
| rm_work | Hangout? | 14:19 |
| johnsom | rm_work Sure, I'm in a meeting for another 30 or so | 14:20 |
| rm_work | kk np me too | 14:20 |
| *** armax has joined #openstack-lbaas | 14:21 | |
| johnsom | rm_work I am free now, so whenever you are ready | 14:35 |
| rm_work | kk one minute | 14:35 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Handle network deltas when delete member https://review.opendev.org/734002 | 14:40 |
| rm_work | johnsom: alright | 14:46 |
| rm_work | johnsom: https://meet.google.com/btb-hvbw-epu | 14:47 |
| *** ramishra has quit IRC | 15:30 | |
| *** zzzeek has quit IRC | 15:31 | |
| *** zzzeek has joined #openstack-lbaas | 15:32 | |
| *** zzzeek has quit IRC | 15:37 | |
| *** zzzeek has joined #openstack-lbaas | 15:38 | |
| *** zzzeek has quit IRC | 15:42 | |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Failover stop threshold https://review.opendev.org/656811 | 15:44 |
| *** zzzeek has joined #openstack-lbaas | 15:45 | |
| *** ramishra has joined #openstack-lbaas | 15:46 | |
| *** zzzeek has quit IRC | 15:50 | |
| *** zzzeek has joined #openstack-lbaas | 15:51 | |
| *** zzzeek has quit IRC | 15:56 | |
| *** zzzeek has joined #openstack-lbaas | 15:59 | |
| johnsom | #startmeeting Octavia | 16:00 |
| openstack | Meeting started Wed Sep 9 16:00:49 2020 UTC and is due to finish in 60 minutes. The chair is johnsom. Information about MeetBot at http://wiki.debian.org/MeetBot. | 16:00 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 16:00 |
| *** openstack changes topic to " (Meeting topic: Octavia)" | 16:00 | |
| openstack | The meeting name has been set to 'octavia' | 16:00 |
| gthiemonge | o/ | 16:00 |
| johnsom | Hi everyone | 16:01 |
| johnsom | Happy feature freeze week. | 16:01 |
| rm_work | o/ | 16:01 |
| johnsom | #topic Announcements | 16:01 |
| *** openstack changes topic to "Announcements (Meeting topic: Octavia)" | 16:01 | |
| johnsom | Final release of python-octaviaclient is this week! | 16:01 |
| johnsom | Feature freeze for octavia is this week! | 16:02 |
| rm_work | semi-distracted, got a second meeting overlapping right now >_< | 16:02 |
| johnsom | #link https://releases.openstack.org/victoria/schedule.html | 16:02 |
| johnsom | Kendall did a nice write up of the intern work on Octavia: | 16:03 |
| johnsom | #link https://superuser.openstack.org/articles/contributor-qa-a-college-students-experience-contributing-to-the-openstack-ussuri-release/ | 16:03 |
| johnsom | If you haven't seen it | 16:03 |
| johnsom | The summit schedule just went live as well: | 16:03 |
| johnsom | #link https://www.openstack.org/summit/2020/summit-schedule | 16:03 |
| johnsom | I have not looked to see if we are on it. lol | 16:04 |
| *** zzzeek has quit IRC | 16:04 | |
| johnsom | Any other announcements this week? | 16:04 |
| *** zzzeek has joined #openstack-lbaas | 16:05 | |
| johnsom | #topic Brief progress reports / bugs needing review | 16:05 |
| *** openstack changes topic to "Brief progress reports / bugs needing review (Meeting topic: Octavia)" | 16:05 | |
| rm_work | "The amphorae of application delivery & security with Octavia" by A10 :P | 16:05 |
| rm_work | lol | 16:05 |
| johnsom | My focus has pretty much been purely on reviews and rebasing patches. | 16:05 |
| gthiemonge | reviews and internal stuff here | 16:06 |
| gthiemonge | BTW ataraday asked if we could review https://review.opendev.org/#/c/750638/ | 16:06 |
| johnsom | gthiemonge Ah, thanks for highlighting that. It's not on the priority review list and I would have missed it. | 16:07 |
| johnsom | I have added it | 16:08 |
| gthiemonge | thanks johnsom | 16:08 |
| johnsom | Ok, on to the main event: | 16:08 |
| johnsom | #topic Priority bug reviews for Victoria | 16:08 |
| *** openstack changes topic to "Priority bug reviews for Victoria (Meeting topic: Octavia)" | 16:08 | |
| johnsom | #link https://etherpad.opendev.org/p/octavia-priority-reviews | 16:08 |
| johnsom | We need to land anything from the top two (three, but lib is done) sections by tomorrow-ish. | 16:09 |
| johnsom | If there is anything there that you are ok not making it into Victoria, please move it down. | 16:09 |
| johnsom | Otherwise, please help prioritize getting those reviewed and merged. | 16:10 |
| johnsom | I think we are doing pretty good really. We have got a lot of work done over the last week or two. Thank you! | 16:10 |
| *** sapd1_x has quit IRC | 16:10 | |
| johnsom | Any other discussion/questions/comments on the priority list? | 16:11 |
| johnsom | #topic Victoria cycle highlights | 16:12 |
| *** openstack changes topic to "Victoria cycle highlights (Meeting topic: Octavia)" | 16:12 | |
| johnsom | It's that time again where we submit highlights for the release. | 16:13 |
| johnsom | I have put up a proposed patch: | 16:13 |
| johnsom | #link https://review.opendev.org/750528 | 16:13 |
| johnsom | Please review that and give me comments by Friday when I will remove the WIP flag and submit them. | 16:13 |
| johnsom | Just a reminder, the highlights are up to five items to highlight for the release for press reasons. It's not a replacement for the release notes which are much more detailed. | 16:15 |
| johnsom | #topic Open Discussion | 16:15 |
| *** openstack changes topic to "Open Discussion (Meeting topic: Octavia)" | 16:15 | |
| johnsom | Any other topics today? | 16:15 |
| johnsom | gthiemonge just pointed out to me that the grenade job is timing out. That is probably due to the new scenario tests. I will fix that here after the meeting. | 16:16 |
| johnsom | After this week we are in feature freeze. No new features should be merged. | 16:18 |
| johnsom | We can continue to work on bug fixes, docs, and tests however. | 16:18 |
| johnsom | If there are no more topics I will give you all some time back to do reviews! | 16:19 |
| gthiemonge | thanks johnsom | 16:19 |
| johnsom | o/ | 16:20 |
| johnsom | #endmeeting | 16:20 |
| *** openstack changes topic to "Discussions for OpenStack Octavia | Priority bug review list: https://etherpad.openstack.org/p/octavia-priority-reviews" | 16:20 | |
| openstack | Meeting ended Wed Sep 9 16:20:21 2020 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 16:20 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/octavia/2020/octavia.2020-09-09-16.00.html | 16:20 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/octavia/2020/octavia.2020-09-09-16.00.txt | 16:20 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/octavia/2020/octavia.2020-09-09-16.00.log.html | 16:20 |
| *** ramishra has quit IRC | 16:43 | |
| *** dosaboy has quit IRC | 16:51 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Update grenade tempest waiter check interval https://review.opendev.org/750736 | 16:52 |
| johnsom | let's give that a try | 16:52 |
| johnsom | Ah, that isn't going to be enough. I need to narrow the job list down again. Before we ran one smoke check, now it's running a bunch. | 16:55 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Update grenade job to run one smoke test https://review.opendev.org/750736 | 16:57 |
| *** spatel has joined #openstack-lbaas | 17:02 | |
| spatel | I have question for octavia, I am testing in lab and we are running VLAN base provider so no more NATing (My cisco router is gateway for all vms). when i create LB in that case amphora attaching it as floating IP (instead of directly adding external IP to amphora VM) | 17:05 |
| spatel | How does my external IP get attached to direct on amphora and not do NAT. | 17:05 |
| johnsom | Octavia doesn't use floating IPs | 17:05 |
| spatel | johnsom: good to see you again :) | 17:06 |
| johnsom | o/ | 17:06 |
| spatel | I just spun up lb1 and ssh into amphora and found only single IP address on VM which is lbaas-mgmt | 17:07 |
| johnsom | So, when you create your load balancer, you specify any neutron network as your VIP network. They can be a provider network. | 17:07 |
| johnsom | Yeah, all tenant traffic is isolated in a network namespace. | 17:07 |
| johnsom | "sudo ip netns exec amphora-haproxy ip a" | 17:07 |
| spatel | on controller node? | 17:07 |
| johnsom | The tenant traffic interfaces are all in that network namespace inside the amphora | 17:08 |
| johnsom | Did you find the interface you expected? | 17:10 |
| spatel | johnsom: ah! i can see it now | 17:11 |
| johnsom | Note that you will not see the VIP address with the "ip a" command, it will be a base port IP | 17:11 |
| spatel | earlier i was using "ip a" :) | 17:11 |
| spatel | johnsom: you nailed it | 17:11 |
| johnsom | Yeah, that will only show the control plane interface on the lb-mgmt-net | 17:11 |
| johnsom | We isolate the tenant traffic for security | 17:12 |
| spatel | johnsom: look at this output - http://paste.openstack.org/show/797663/ | 17:13 |
| johnsom | Yep, that looks pretty good to me | 17:14 |
| spatel | so i have only single interface ens3 and all traffic flowing through ? or it has two distinct interface ens3 and eth1 ? | 17:14 |
| johnsom | Currently, for tenant traffic you have one interface. When you add a member server, that has a different subnet than the VIP, it will add another interface. | 17:15 |
| spatel | hmm | 17:15 |
| johnsom | If the pool member is on the same network as the VIP, you will only have one interface. | 17:16 |
| spatel | i have pool member on same subnet | 17:16 |
| spatel | what are these two address ? 192.168.1.14 and 192.168.1.23 ? | 17:17 |
| johnsom | Yeah, that creates a "one-armed" load balancer. | 17:17 |
| spatel | don't you think it should have single IP (if everything located on single subnet?) | 17:18 |
| johnsom | 14 is the base port address. The amphora synchronization traffic and VRRP use that interface. The 23 is the VIP interface. The 14 address is permanent to this amphora. The 23 address will move between amphora in failure cases. | 17:18 |
| spatel | I have noticed my lb1 created two amphora, look like its Active-standby | 17:19 |
| johnsom | Yes. that topology allows failover between the amphora in around a second. Less if you do manual tuning. | 17:20 |
| johnsom | If you were to nova delete this amphora you are on, your traffic through the load balancer will continue on the standby. | 17:20 |
| johnsom | And then Octavia will rebuild the missing amphora | 17:21 |
| spatel | Lets say if i configure for SINGLE amphora (not HA) in that case if my compute node die with some reason, does octavia create new amphora on other compute and heal my services itself? | 17:21 |
| johnsom | Yes, it does. But it will take longer to resume traffic flow. Typically 30 seconds to a minute. | 17:21 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Add a release note about HAProxy 2.0 https://review.opendev.org/750531 | 17:27 |
| spatel | hmm can we adjust that time to lower? | 17:27 |
| johnsom | For SINGLE instances most of the time is waiting for a new VM to boot in nova. | 17:27 |
| spatel | it would be great if octvia integrate with senlin so it will auto-scale ELB (like AWS) | 17:27 |
| johnsom | For ACTIVE/STANDBY you can configure how aggressive it is with the octavia.conf. | 17:27 |
| johnsom | Auto scale is on the list of to-do features. It won't need senlin, it will handle it on it's own. | 17:28 |
| johnsom | Work started on this recently with the new metrics work. | 17:28 |
| johnsom | What level of connections and bandwidth are you expecting to need? | 17:29 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Add a release note about HAProxy 2.0 https://review.opendev.org/750531 | 17:30 |
| johnsom | Ugh, the focal jobs switched. now the lower constraints job is failing with a missing package dependency | 17:42 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Fix the lower-const job by bumping cffi to 1.14.0 https://review.opendev.org/750767 | 17:55 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Fix the lower-constraints job for focal https://review.opendev.org/750767 | 18:19 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Fix the lower-constraints job for focal https://review.opendev.org/750767 | 18:25 |
| *** spatel has quit IRC | 18:28 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Fix the lower-constraints job for focal https://review.opendev.org/750767 | 18:31 |
| johnsom | blah, what a chain of required changes. Sorry for the noise | 18:31 |
| *** zzzeek has quit IRC | 18:33 | |
| *** zzzeek has joined #openstack-lbaas | 18:34 | |
| *** zzzeek has quit IRC | 18:41 | |
| *** zzzeek has joined #openstack-lbaas | 18:42 | |
| *** zzzeek has quit IRC | 18:50 | |
| *** zzzeek has joined #openstack-lbaas | 18:50 | |
| *** spatel has joined #openstack-lbaas | 18:51 | |
| spatel | johnsom: thanks! sorry i was away for sometime so didn't see your last messages | 18:51 |
| johnsom | No worries | 18:51 |
| spatel | "Auto scale is on the list of to-do features" do you know how they are going to do active-active? | 18:52 |
| spatel | they need one more LB in front of octavia to scale horizontally | 18:53 |
| johnsom | Well, auto scale also include more RAM and vCPUs as well. | 18:53 |
| johnsom | There are three active/active proposals. Just no developers to work on them at the moment. | 18:53 |
| spatel | johnsom: how does AWS doing that, i believe they are using CNAME style RR method. | 18:54 |
| *** zzzeek has quit IRC | 18:55 | |
| johnsom | Yeah, that will break session persistence. AWS has three different "load balancers" that handle it differently | 18:55 |
| *** zzzeek has joined #openstack-lbaas | 18:56 | |
| spatel | hmm | 18:56 |
| *** zzzeek has quit IRC | 19:03 | |
| *** zzzeek has joined #openstack-lbaas | 19:05 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Fix the lower-constraints job for focal https://review.opendev.org/750767 | 19:17 |
| johnsom | Oh boy py3.8 opened a can of worms. I guess I shouldn't have lost track of that test patch I was running. | 19:18 |
| johnsom | async is now a reserved word, lol | 19:18 |
| *** zzzeek has quit IRC | 19:21 | |
| *** zzzeek has joined #openstack-lbaas | 19:23 | |
| johnsom | pyroute2 released today with the fix to our regression | 19:29 |
| johnsom | Well, the regression in pyroute2 | 19:29 |
| rm_work | i thought that happened in like 3.6 | 19:32 |
| johnsom | Well we had versions in lower-constraints of sqlalchemy and pyroute2 that used it.... | 19:32 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Fix the lower-constraints job for focal https://review.opendev.org/750767 | 19:40 |
| johnsom | I guess it was in oslo.db, not sqlalchemy | 19:40 |
| *** vishalmanchanda has quit IRC | 19:45 | |
| *** openstackgerrit has quit IRC | 20:17 | |
| *** openstackgerrit has joined #openstack-lbaas | 20:28 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Fix the lower-constraints job for focal https://review.opendev.org/750767 | 20:28 |
| *** spatel has quit IRC | 20:34 | |
| rm_work | i meant i thought async became reserved in python in like 3.6 | 20:39 |
| *** spatel has joined #openstack-lbaas | 20:46 | |
| *** spatel has quit IRC | 20:52 | |
| *** TMM has quit IRC | 21:04 | |
| *** TMM has joined #openstack-lbaas | 21:04 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Update grenade job to run one smoke test https://review.opendev.org/750736 | 21:22 |
| *** gcheresh_ has quit IRC | 21:31 | |
| rm_work | k this looks good, I am assuming the commands work because i don't know how to actually verify other than yeah the syntax seems valid :D | 22:12 |
| johnsom | lol | 22:20 |
| johnsom | When you get a chance, could you take a quick pass at these gate fix patches: https://review.opendev.org/#/c/750767/ and https://review.opendev.org/#/c/750736/ | 22:24 |
| johnsom | The lower-constraint one was mostly done but zuul got restarted, so it started over. | 22:24 |
| *** TrevorV has quit IRC | 22:42 | |
| *** rcernin has joined #openstack-lbaas | 22:54 | |
| *** rcernin has quit IRC | 22:56 | |
| *** rcernin has joined #openstack-lbaas | 22:56 | |
| *** zzzeek has quit IRC | 23:17 | |
| *** zzzeek has joined #openstack-lbaas | 23:19 | |
| *** dosaboy has joined #openstack-lbaas | 23:28 | |
| *** zzzeek has quit IRC | 23:38 | |
| *** zzzeek has joined #openstack-lbaas | 23:40 | |
| *** zzzeek has quit IRC | 23:59 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!