Velero + Kopia Data Mover
Snapshot any CSI storage class (EBS, Ceph RBD, Longhorn, OpenEBS ZFS) and stream encrypted deduplicated blocks to eu-north-1 or eu-central-1.
Decouple your cluster state from single-cloud lock-in. Continuously replicate etcd snapshots, CSI persistent volumes, and Helm/ArgoCD manifests into immutable European S3 vaults.
Snapshot any CSI storage class (EBS, Ceph RBD, Longhorn, OpenEBS ZFS) and stream encrypted deduplicated blocks to eu-north-1 or eu-central-1.
Configure Nordic Data Sync as your native off-cluster Longhorn backupstore so bare-metal node failures recover persistent volumes on a standby cluster in minutes.
Back Harbor, GitLab Container Registry, and Loki/Thanos observability chunks with flat-rate €9/TB S3 storage and zero cross-AZ pull charges.
apiVersion: batch/v1
kind: CronJob
metadata:
name: etcd-s3-immutable-snapshot
namespace: kube-system
spec:
schedule: "*/30 * * * *"
jobTemplate:
spec:
template:
spec:
containers:
- name: snapshot-uploader
image: amazon/aws-cli:2.18.0
command:
- /bin/sh
- -c
- |
aws --endpoint-url https://de.nordicdatasync.com \
s3 cp /snapshots/etcd-latest.db \
s3://k8s-fra1-etcd/cluster-01/$(date -u +%Y%m%d-%H%M).db \
--object-lock-mode COMPLIANCE \
--object-lock-retain-until-date $(date -u -d "+14 days" +%Y-%m-%dT%H:%M:%SZ)
restartPolicy: OnFailure
Replace expensive hyperscaler buckets for Prometheus long-term metrics and Loki log chunks:
type: S3
config:
bucket: observability-metrics-arn1
endpoint: se.nordicdatasync.com
region: eu-north-1
insecure: false
signature_version2: false
put_user_metadata:
X-Nordic-Tier: nvme-coalesced
http_config:
idle_conn_timeout: 90s
response_header_timeout: 2m