MongoDB exposed via Internal NLB causing sporadic connection timeouts

We identified what the problem was and tested the solution and its now working. Incase anyone comes across this. Make sure you set the annotation:

service.beta.kubernetes.io/aws-load-balancer-target-group-attributes: preserve_client_ip.enabled=false

However as our mongodb is already deployed, we had to manually look for the LB Target Group and Disable this.

My only concern left is if we are okay to leave this as is, with the manual change to the target group, is it going to revert?

Is it possible to add the above annotation to our existing config and reapply?

Resources:

https://kubernetes-sigs.github.io/aws-load-balancer-controller/v2.6/guide/service/annotations/