Raju Singh:
Can anyone help me solving below scenario on etcd backup/restore :
Cert/key files have been given in base node along with the etcd backup file from last successful backup (/opt/snapshot.db) . Master node doesn’t have etcdctl installed and it runs “etcd” as a service NOT pod . How do we restore the last backup successfully in this case . Any leads/solution welcome @Tej_Singh_Rana, @Mohamed Ayman, @Vijin Palazhi…
Mohamed Ayman:
the etcdctl is a client tool to communicate with the etcd which exposed and running on the master. so u can add
--endpoints
with the private ip of the master node:2379 from the base node. you can check this useful guide https://github.com/mmumshad/kubernetes-cka-practice-test-solution-etcd-backup-and-restore
Raju Singh:
Thanks Ayman for reverting.
What about " --data-dir=/var/lib/etcd-from-bkp" which we pass while restoring , we need to modify the volume for(/var/lib/etcd) this mount point in etcd manifests file .
I was wondering in case of etcd as service, will this mount/directory exist in master ??