Loadbalancer occasionally timesout

[AdamJSoftware]

Environmental Info:
RKE2 Version: v2.7.6

Node(s) CPU architecture, OS, and Version:
Linux apps2-pool1-070c2e68-qbsw9 5.15.0-89-generic #99-Ubuntu SMP Mon Oct 30 20:42:41 UTC 2023 x86_64 x86_64 x86_64 GNU/Linux * 3

Cluster Configuration:
3 servers

Describe the bug:
This only happens on one of my clusters. My others don't have this issue. When I create a loadbalancer, about 1/3 of my requests timeout, and the other ones are quite slow. This cluster also frequently displays fsnotify: too many open files when trying to read logs which leads me to believe that it is an issue with too many open files but that can also be a red hearing. I also remember having issues creating a horizontal pod scaler but I don't know if that's related to this. I am routing all the LB traffic to traefik BTW.

[brandond]

RKE2 doesn't ship with a load-balancer controller or Traefik ingress controller, so these would both be components that you deployed yourself? I'm not sure how we can help with things that we don't provide.

RKE2 Version: v2.7.6

This looks like a Rancher version, not rke2?

[AdamJSoftware]

My bad. I have some further development anyways. I was at RKE2 version v1.26.4 but then I upgraded to version v1.26.10+rke2r2 and I noticed that the timeouts significantly diminished, there are still some requests that occasionally timeouts but significantly less.

[brandond]

Closing, please feel free to reopen if there is anything RKE2-specific we can address.