Migrating a Kubernetes Cluster to AnubizHost
Managed Kubernetes (EKS, GKE, AKS, DOKS) is convenient but expensive at sustained load and locks you into the cloud provider's network and IAM. This guide covers moving a Kubernetes workload to a self-managed k3s or k0s cluster on AnubizHost offshore dedicated servers.
Need this done for your project?
We implement, you ship. Async, documented, done in days.
Choose Self-Hosted Kubernetes Distribution
For most teams, k3s (Rancher) or k0s (Mirantis) is the right choice on AnubizHost - they're lightweight, single-binary, and run well on VPS or dedicated. Full upstream Kubernetes (kubeadm) is fine for teams that want vanilla.
Sizing: 3-node control plane for HA, 3+ workers depending on workload. AnubizHost dedicated servers with NVMe and EPYC CPU run k3s clusters with hundreds of pods comfortably.
Provision Cluster Infrastructure
Order 3 AnubizHost dedicated servers in Romania or Iceland. Install k3s server on each control plane node: curl -sfL https://get.k3s.io | INSTALL_K3S_EXEC="server --cluster-init" sh -. Join workers: curl -sfL https://get.k3s.io | K3S_URL=https://server:6443 K3S_TOKEN=xxx sh -.
Configure metallb for LoadBalancer services (cloud LB is not available on dedicated). Install longhorn or rook-ceph for persistent volumes.
Export Manifests from Source Cluster
kubectl get all,cm,secret,pvc,ingress \
--all-namespaces -o yaml > cluster-export.yamlClean the export: remove resourceVersion, uid, cluster-specific annotations. For cloud-specific resources (LoadBalancer with cloud annotations, StorageClass referencing cloud provisioners), translate to self-hosted equivalents.
Apply to new cluster: kubectl apply -f cluster-export-cleaned.yaml.
Migrate Stateful Workloads with Velero
Velero handles PV migration cleanly. Install Velero on the source cluster with backup target pointing at AnubizHost MinIO. Take a backup including PVs: velero backup create migration --include-namespaces my-ns. Install Velero on target pointing at same MinIO bucket. Restore: velero restore create --from-backup migration.
For active databases, take a logical backup (pg_dump, etc.) and restore after Velero brings up the pod skeleton.
DNS Cutover and Decommission
Update Ingress controller external IP / DNS to point at the AnubizHost cluster's metallb VIP. Verify application paths work. Keep source cluster running for 7 days as rollback. Related reading: offshore dedicated, Docker Compose migration, switching from AWS.
Related Services
Why Anubiz Host
Ready to get started?
Skip the research. Tell us what you need, and we'll scope it, implement it, and hand it back — fully documented and production-ready.