ClusterWarden: DataBase clusters the easy way
ClusterWarden provisions fully replicated Percona XtraDB and PostgreSQL clusters on your servers with a few clicks, then keeps them running for you — rolling maintenance with zero downtime, automatic failover and self-healing, and scheduled, verified, encrypted backups with one-click restore. Enterprise database resilience, without the enterprise headcount.
YOUR PERSONAL MYSQL & POSTGRES DBA
Your personal DBA, on duty 24/7.
ClusterWarden does what a seasoned database administrator does — builds the cluster, applies patches in a safe order, takes and verifies the backups, spots a struggling node before it becomes an outage, fails over when a primary dies, and knows when to leave things alone and ask you first. The difference is that it never sleeps, never forgets a step, and never lets a routine task slip because it was busy — so you get senior-DBA judgement on every cluster, every hour of every day, without hiring one.
Easy Cluster setup
Give ClusterWarden a few servers, choose Percona XtraDB Cluster or PostgreSQL (including pgEdge active-active), and it builds a fully replicated, load-balanced, health-checked cluster end to end. What used to take a DBA a full day is done before your coffee is cold.
Zero-downtime maintenance
Rolling restarts, OS patching and reboots happen one member at a time, with each node verified back in the cluster before the next is touched. Add or retire members, apply security updates and tune MySQL settings without ever taking the service offline.
Failover and self-healing
When a primary fails, ClusterWarden promotes a standby and re-homes the rest of the cluster. Unhealthy members are diagnosed and repaired automatically, escalating to a full rebuild only when a healthy donor can re-seed them.
Database backups
Hot physical backups and logical dumps of whole servers or selected databases run on the schedules you set, are verified before they're trusted, and are encrypted and offloaded to S3-compatible, SFTP or FTP storage with independent local and offsite retention. Restore any of them to any cluster in one click.
Always-on monitoring
Every node's health, replication state, disk, memory, CPU and internet exposure is checked around the clock, with alerts that only fire on confirmed trouble — and a public status page so you can watch ClusterWarden itself.
Built for teams and providers
Multi-tenant workspaces with per-workspace quotas, two-factor confirmation for destructive actions, a full audit trail and a provisioning API let you run one cluster or hundreds — for your own company or for your customers.