Aaron Johnson

It looks like `nginx_ingress_controller_ssl_expire_time_seconds` is missing when curling metrics from the `ingress-nginx-controller-metrics` service but only on some calls whereas other calls have the metrics ``` $ curl http://172.16.6.62:10254/metrics | grep...

yeah this is really annoying and took me a long time to figure out it was Tilix doing this. This should be a must fix IMO

It would be nice if we could use the standard convention label names and for taints in eksctl (below example doesn't work but should) ```yaml labels: node-role.kubernetes.io/gitlab-worker: true taints: node-role.kubernetes.io/gitlab-worker:...

I hit #2296 which says discussion is taking place here

right, but why is `node-role.kubernetes.io` restricted? Isn't that the standard convention for how kubectl determines node roles?

ah I think I understand now. So if I want to make use of standard convention node labels I'll need to apply like `kubectl label nodes ip-172-27-121-8.ec2.internal node-role.kubernetes.io/worker=true` on my...

@aszlig done, look good? (I pushed the non-squashed/rebased branch to `rescuefixes-old` in my fork if you want to compare with the old branch)

I decided to give Immich a go on kubernetes with the helm chart and I gotta say this is really well done so great work 😃 (Facial recognition was the...

We currently work around this limitation with a script that temporarily removes the MutatingWebhookConfiguration and ValidatingWebhookConfiguration objects, creates TargetGroupBindings that the secondary aws-load-balancer-controllers use, then recreate the MutatingWebhookConfiguration and ValidatingWebhookConfiguration...

Closing, if this is still not addressed in your case please comment @israelshenkar