Now supported

Three cluster engines, one personal DBA

ClusterWarden now builds and runs your Percona XtraDB, pgEdge PostgreSQL and Valkey clusters.

Plus single-node MariaDB & MySQL Patched, backed up and monitored, 24/7

Configure your Cluster
Percona XtraDB ClusterSynchronous multi-primary MySQL
PostgreSQL · pgEdgeDistributed multi-master Postgres
Valkey ClusterSharded in-memory key-value

ClusterWarden provisions fully replicated Percona XtraDB, PostgreSQL and Valkey clusters on your servers in a few clicks, then keeps them running for you with zero-downtime maintenance, automatic failover and verified, encrypted backups.

Enterprise database resilience, without the enterprise headcount.

Everything a senior DBA does, on every cluster

ClusterWarden handles the work a seasoned database administrator would, and knows when to leave things alone and ask you first.

  • Builds the clusterFully replicated and production-ready in a few clicks.
  • Patches in a safe orderRolling maintenance, one node at a time, with zero downtime.
  • Takes and verifies backupsScheduled, encrypted, and restorable in one click.
  • Spots trouble earlyCatches a struggling node before it becomes an outage.
  • Fails over automaticallySelf-heals when a primary dies, so your database stays up.
  • Asks before risky changesKnows when to leave things alone and check with you first.
Never sleeps

Watching every cluster, every hour of every day.

Never forgets a step

Every runbook followed the same way, every time.

Never lets a task slip

Routine work gets done, even on your busiest days.

Running single-node MariaDB or MySQL? ClusterWarden manages those too, turning it into your private DBA as a service.

Configure your Cluster

How ClusterWarden works

From bare servers to a self-healing cluster

Easy cluster setup

Give ClusterWarden a few servers and choose an engine. 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.

Percona XtraDB ClusterPostgreSQLpgEdge active-active

Zero-downtime maintenance

Restarts, OS patching and reboots roll through one member at a time, and each node is 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.

Rolling restartsOS patchingScale in & out

Failover & 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.

Auto-promotionAuto-repairSafe rebuilds

Database backups

Hot physical backups and logical dumps of whole servers or selected databases run on your schedule, are verified before they're trusted, then encrypted and offloaded with independent local and offsite retention. Restore any of them to any cluster in one click.

Physical & logicalEncryptedS3 · SFTP · FTP

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, plus a public status page so you can watch ClusterWarden itself.

24/7 checksConfirmed alerts onlyPublic status page

Built for teams & providers

Multi-tenant workspaces with per-workspace quotas, two-factor confirmation for destructive actions, a full audit trail and a provisioning API. Run one cluster or hundreds, for your own company or for your customers.

Multi-tenant2FA safeguardsAudit trailAPI