Restore AKS Backup does not work as expected

Alexander Redmann 20 Reputation points


we are experimenting with the backup function of the AKS cluster, which seems to be working. We are performing a daily backup of the entire cluster, meaning all namespaces. When we select a specific restore point for just one namespace, we have to delete the namespace first because Azure cannot overwrite existing resources. When we perform the restore, some pods remain in a pending status because a secret is missing. Are these not being backed up? How can we resolve this issue?
It is an grafana, prometheus, loki stack with an operator and the pod kube-prometheus-stack-grafana- has the error events:
Normal Scheduled 4m34s default-scheduler Successfully assigned monitoring/kube-prometheus-stack-grafana-7cf5785ff8-b7gl9 to aks-default-25289621-vmss000002

Warning Failed 4m8s (x4 over 4m34s) kubelet Error: secret "kube-prometheus-stack-grafana" not found

Normal Pulled 4m8s (x4 over 4m34s) kubelet Container image "" already present on machine


Azure Kubernetes Service (AKS)
Azure Kubernetes Service (AKS)
An Azure service that provides serverless Kubernetes, an integrated continuous integration and continuous delivery experience, and enterprise-grade security and governance.
1,987 questions
0 comments No comments
{count} votes

Accepted answer
  1. akinbade abiola 8,625 Reputation points

    Hello Alexander Redmann,

    Thanks for your question.

    If you also want to back up cluster-scoped resources, secrets, and persistent volumes, select the items under other options. Please remember to check this option.


    Screenshot that shows various backup configurations.


    You can mark it 'Accept Answer' if this helped.

    1 person found this answer helpful.

0 additional answers

Sort by: Most helpful