29_backup_k8s_etcd.yml - Usage Guide

Modifies State k3s kubeadm

Takes a snapshot of the etcd database (the heart of Kubernetes) and stores it safely.

Playbook: playbooks/29_backup_k8s_etcd.yml

What it does

  • Detects whether the cluster uses k3s or kubeadm (etcdctl).
  • Takes the snapshot and saves it under /var/backups/etcd with a timestamp.
  • Protects the backup directory with 0700 and snapshot files with 0600.
  • After a successful backup, removes snapshots older than the retention period.
  • Runs only on the first server in the master nodes to avoid duplicate work.

Parameters (optional)

Variable Default Description
etcd_backup_retention_days 30 After a successful backup, deletes snapshots older than this many days. Must be a positive integer. 
ansible-playbook -i inventories/musteri_a/hosts.ini playbooks/29_backup_k8s_etcd.yml \
  --extra-vars 'etcd_backup_retention_days=14'

An etcd snapshot also contains Kubernetes Secret data. Restrict access to the backup directory to authorized users and copy the backup to a separate, encrypted location.

Sample output 

################################################################################
# HOST: master1
################################################################################
k3s detected. Running k3s etcd-snapshot...
Retention: snapshots older than 30 days were removed.
SUCCESS: k3s etcd snapshot saved to /var/backups/etcd