Switching from Cluster IP to Load balancer hangs in pending forever

I’ve deployed the postgres db, operator using helm charts. I was able to connect to the DB using PG client pod.

I need to switch from clusterIP to load balancer so that I can connect to the DB from external clients.

When I tried to add expose type to LoadBalancer, and run helm upgrade, pgbouncer service hangs in pending mode as shown below

Any thoughts would be much appreciated.

NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE SELECTOR
ppg-db-pgbouncer LoadBalancer 172.20.133.11 5432:32060/TCP 2d2h postgres-operator.crunchydata.com/cluster=ppg-db,postgres-operator.crunchydata.com/role=pgbouncer

Hello @vamsi ,

out of the output you shared it is not clear what the issue is.

Can you run kubectl describe svc ppg-db-pgbouncer and show the output?

I don’t see EXTERNAL-IP in your output, so my guess would be that there is some issue with getting it within your k8s env.

  1. Which k8s flavor are you using? (AWS, GKE, k3s, etc).
  2. Which version of the operator it is? I assume latest 2.4, but maybe I’m wrong.