The process of backing up an etcd cluster is straightforward, but there are multiple ways that you can approach this task:
- Create a backup or snapshot of the storage disk for etcd. This is especially valid in cloud scenarios where you can easily manage backups outside your Kubernetes cluster.
- Perform a manual snapshot of etcd using the etcdctl command. You need to manage the backup files yourself: upload them to external storage, and apply a retention policy.
- Use Velero (formerly Heptio Ark (https://velero.io/)), which can perform snapshots, manage them in external storage, and restore them if needed. Additionally, it can be used to perform backups of persistent volumes using Restic integration (https://velero.io/docs/master/restic/).
- Use etcd-operator (https://github.com/coreos/etcd-operator) to provision etcd clusters on top of Kubernetes. You can...