Backup and Restore

  • Backup Candidates

    • Resource configuration

    • ETCD cluster

    • Persistent Volumes

Resource Configuration

  • We can easily get the configs declared declaratively in yaml format.

  • If you create it imperatively & do not document it then it would be hard to redeploy if something goes wrong.

  • Hence we query the kube-apiserver and save all resources configs for all objects created on server

  • kubectl get all -A -o yaml > back.yaml

  • We use tools like velero (ARK) that covers every resources excluded by above commands

PreviousCluster UpgradesNextETCD Backup

Last updated