The UI is not able to pick the correct domain name for the pods which is supposed to be uhn7klr2c.local
It's taking Kubernetes' default domain name which is cluster.local
This is why it's not able to reach the pods, and hence it's showing status unreachable.
I want to understand how and from where UI reads the pod names/domain names etc.
Also, what's the significance of KUBE_DOMAIN ? Is it being honored properly in this case?
The override for the custom domain is properly being written in the Postgres table (availability_zones).
Pods' names are showing correctly in logs and kubectl outputs.
Pods and containers are communicating well.
The issue is only with the UI. This ticket is open for a month now and it's a blocker for the customer (Rakuten).
Rakuten uses IPv6 everywhere in their infra.
Rakuten uses robin.io provided k8s and storage for the cluster.
ZD ticket - 1514
The UI is not able to pick the correct domain name for the pods which is supposed to be
uhn7klr2c.local
It's taking Kubernetes' default domain name which iscluster.local
This is why it's not able to reach the pods, and hence it's showing statusunreachable
.I want to understand how and from where UI reads the pod names/domain names etc. Also, what's the significance of KUBE_DOMAIN ? Is it being honored properly in this case?
The override for the custom domain is properly being written in the Postgres table (availability_zones). Pods' names are showing correctly in logs and kubectl outputs. Pods and containers are communicating well.
The issue is only with the UI. This ticket is open for a month now and it's a blocker for the customer (Rakuten).
Rakuten uses IPv6 everywhere in their infra. Rakuten uses robin.io provided k8s and storage for the cluster.
https://yugabyte.zendesk.com/agent/tickets/1514