| *** ysandeep|out is now known as ysandeep | 06:15 | |
| opendevreview | conna proposed openstack/octavia master: Fix update/delete listener ca/crl error https://review.opendev.org/c/openstack/octavia/+/845093 | 07:45 |
|---|---|---|
| *** ysandeep is now known as ysandeep|lunch | 07:58 | |
| *** ysandeep|lunch is now known as ysandeep | 08:44 | |
| johnsom | tweining I only see one open patch on the stable branches and the parent for that chain is marked WIP. | 14:31 |
| *** ysandeep is now known as ysandeep|out | 14:31 | |
| johnsom | https://review.opendev.org/c/openstack/octavia/+/839854 | 14:31 |
| tweining | right. I found 2 more but they are all W-1. | 14:32 |
| tweining | sorry, I didn't check it before | 14:32 |
| johnsom | tweining I will fix the pep8 here: https://review.opendev.org/c/openstack/octavia/+/839848 | 14:34 |
| tweining | I saw that failure this morning, but I wasn't sure what's the problem here in .pylintrc. | 14:36 |
| johnsom | I tend to just fix the issues in the code rather than commenting out checks unless it is really obvious that it's a rule different than OpenStack has used. Otherwise they usually come back to haunt us later. | 14:37 |
| johnsom | Ha, yeah, ok, new pylint as there are more/new things coming out after I fixed that one. | 14:38 |
| johnsom | Though these are just warnings and the fix for the patch seems to pass ok. So, I will create another patch for the pylint warnings. | 14:41 |
| tweining | cool thanks | 14:42 |
| opendevreview | Michael Johnson proposed openstack/octavia master: Fix new pylint issues https://review.opendev.org/c/openstack/octavia/+/845146 | 14:47 |
| opendevreview | Michael Johnson proposed openstack/octavia stable/yoga: Fix new pylint issues https://review.opendev.org/c/openstack/octavia/+/845042 | 14:48 |
| opendevreview | Michael Johnson proposed openstack/octavia stable/yoga: Fix compile_amphora_details when using UDP listeners https://review.opendev.org/c/openstack/octavia/+/839848 | 14:48 |
| opendevreview | Michael Johnson proposed openstack/octavia stable/xena: Fix new pylint issues https://review.opendev.org/c/openstack/octavia/+/845043 | 14:48 |
| opendevreview | Michael Johnson proposed openstack/octavia stable/wallaby: Fix new pylint issues https://review.opendev.org/c/openstack/octavia/+/845044 | 14:49 |
| johnsom | I wondered about those brackets, but didn't try it as it would clear the votes | 14:56 |
| johnsom | But I did that on the other patch, nevermind. It is still before my work hours and I need my coffee | 14:58 |
| tweining | it's only a cosmetic issue. I find the less brackets the better. :) | 14:58 |
| johnsom | Yeah, this isn’t lisp | 14:59 |
| tweining | this morning I had to decrypt some awk code that I found in the devstack plugin.sh. I never used awk before and I though I won't ever need to learn it. but then I found an easier way to do the same using grep and cut. | 15:04 |
| johnsom | awk and sed are useful tools, but other tools have certainly evolved over time. | 15:10 |
| tweining | johnsom: https://paste.centos.org/view/20027194 this is how I set up my IPv6 network in TripleO standalone. maybe you spot some mistake? | 15:10 |
| tweining | I use sed from time to time | 15:11 |
| johnsom | Hmm, that looks basically right. Let me look at a few things | 15:16 |
| johnsom | The fact that your neighbor discovery failed is the indicator in my book | 15:17 |
| johnsom | Maybe the neutron router for the subnet needs an "external" interface? | 15:19 |
| johnsom | Mine gets setup with public IPv6 (firewalled of course) so the router1 devstack sets up has and interface with a routable address. | 15:25 |
| johnsom | I'm just guessing the lb-mgmt-router is missing an external interface | 15:26 |
| opendevreview | Michael Johnson proposed openstack/octavia master: Fix new pylint issues https://review.opendev.org/c/openstack/octavia/+/845146 | 15:27 |
| opendevreview | Michael Johnson proposed openstack/octavia stable/yoga: Fix new pylint issues https://review.opendev.org/c/openstack/octavia/+/845042 | 15:27 |
| opendevreview | Michael Johnson proposed openstack/octavia stable/xena: Fix new pylint issues https://review.opendev.org/c/openstack/octavia/+/845043 | 15:27 |
| opendevreview | Michael Johnson proposed openstack/octavia stable/wallaby: Fix new pylint issues https://review.opendev.org/c/openstack/octavia/+/845044 | 15:27 |
| tweining | thanks, I will do some research in that area. I just also found out about "ipv6_ra_mode: Determines who sends Router Advertisements." in the docs, which sounds related. | 15:28 |
| johnsom | No, that isn't the issue, I was thinking of tempest hitting the VIPs. This is a local interface, It's up, there is a route there. It's something else. | 15:40 |
| tweining | apart from this issue I wonder how these subnet settings would look like in a real edge deployment. I guess there will need to be multiple subnets? | 15:42 |
| johnsom | Yes, each AZ has it's own lb-mgmt-net setting in the definition | 15:42 |
| johnsom | It could be the same in a stretched L2 design, but in most cases it's unique subnets. | 15:43 |
| johnsom | Hmm, these could make fun LB hosts: https://www.theregister.com/2022/06/08/intel_security_reference_design/ | 15:44 |
| johnsom | They even have QAT | 15:44 |
| tweining | I always thought Atoms would be Intel's low end CPUs | 15:48 |
| johnsom | Atoms are higher end parts with defects in some areas that get disabled. Grin | 15:49 |
| tweining | anyway, 100Gbps ethernet doesn't sound like low end at all. | 15:50 |
| johnsom | Yeah, with QAT you can get your encryption up to those rates too without eating a bunch of cores. | 15:51 |
| opendevreview | Spencer Harmon proposed openstack/octavia master: Add event notifications for load balancers. https://review.opendev.org/c/openstack/octavia/+/831051 | 16:19 |
| spencerharmon | I seem to be failing pep8 tox tests for files unrelated to my changes. I can't reproduce the failure locally. If anyone has the cycles, could you help point me to what files are causing the failure? | 18:58 |
| spencerharmon | #link https://zuul.opendev.org/t/openstack/build/2ecc5c26cd774248ba4638be66866097 | 18:58 |
| tweining | spencerharmon: johnsom already proposed a fix for those errors https://review.opendev.org/c/openstack/octavia/+/845146 | 19:01 |
| tweining | but Zuul is not convinced yet ;) | 19:01 |
| spencerharmon | Oh no. What to do when the gate won't let you fix the gate? Ahhhh! | 19:02 |
| spencerharmon | Thanks for the context. That'll save me from pulling out any more hair :) | 19:02 |
| tweining | I'd suggest to wait a bit till a fix is merged and then try a recheck | 19:03 |
| spencerharmon | Excellent. Will do. Thanks so much! | 19:03 |
| tweining | you're welcome :) | 19:04 |
| opendevreview | Dr. Jens Harbott proposed openstack/octavia master: Fix pyroute2.netns mock in some tests https://review.opendev.org/c/openstack/octavia/+/845171 | 20:21 |
| opendevreview | Michael Johnson proposed openstack/octavia stable/wallaby: Fix new pylint issues https://review.opendev.org/c/openstack/octavia/+/845044 | 20:37 |
| opendevreview | Michael Johnson proposed openstack/octavia stable/xena: Fix new pylint issues https://review.opendev.org/c/openstack/octavia/+/845043 | 20:37 |
| opendevreview | Michael Johnson proposed openstack/octavia master: Fix pyroute2.netns mock in some tests https://review.opendev.org/c/openstack/octavia/+/845171 | 20:57 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!