Raff backs up your VMs on a schedule you set: incremental backups that copy only changed blocks, stored on a separate pool, restorable in one click. Every account gets a free backup storage pool of 50 GB plus 10 to 15 percent of each VM's disk, shared with snapshots.
Backups and snapshots share an account-wide free pool: 50 GB plus 10 percent of each standard VM's disk (15 percent on premium). Only storage above the pool is billed, at the live rate below.
Set the schedule once. From then on the platform takes the backups, prunes the old ones, and keeps them off your VM's disk.
Daily or weekly, at the hour you choose, with multiple schedules per VM if you want belt and suspenders.
The first backup is full; every one after copies only changed blocks. Fast to run, small to store.
Pick any recovery point and restore the VM in place. Powered off, overwritten, back in minutes.
Backups live on a dedicated pool, not on your VM’s disk. A dead disk cannot take its backups with it.
Point-in-time snapshots before risky changes, for VMs and block storage volumes alike.
Keep 2 to 100 recovery points per VM. Old backups age out automatically; no cron scripts.
Managed Postgres and MySQL add continuous point-in-time recovery on top, restorable to any minute.
Backups are taken from the running VM on the platform side. No agent to install, no load window to plan.
Backups, schedules and snapshots are first-class in the REST API, the raff CLI and Terraform.
Use a snapshot before an upgrade you might want to undo. Use backups for everything you can't afford to lose. One honest rule: a VM with backups enabled can't also take snapshots, so pick backups for anything that matters.
| Snapshot | Backup | |
|---|---|---|
| What is it for | A checkpoint before a risky change | Scheduled protection against loss |
| How it runs | Manual, on demand | Automatic, daily or weekly |
| What it stores | Disk state at one moment | Full base + incremental changes |
| Where it lives | With the VM | Separate backup pool |
| Retention | Until you delete it | Keep 2–100 points, auto-pruned |
| Billing | Shared free pool, then per GB | Shared free pool, then per GB |
Restores are in-place: the VM is powered off and overwritten with the recovery point you picked. Full walkthrough in the backup docs, and database-level recovery lives with Managed Databases.
Backups you have to remember are backups you don't have. Set it once and the platform remembers for you.
Open the Backups tab on any VM, or enable backups right in the create flow.
Daily or weekly, at your chosen time, plus how many recovery points to keep (2 to 100).
Backups run, chains grow by changed blocks only, old points age out. Restore is one click away.
Backups are one layer of keeping your stack safe. Everything below plays on the same team.
You set a schedule, daily or weekly, and Raff backs up the VM automatically. The first backup is full; every one after copies only changed blocks. Backups are stored on a separate pool and restore in one click.
Every account gets a free backup-storage pool: 50 GB plus 10 percent of each standard VM's disk (15 percent on premium), shared across backups and snapshots. Only storage above the pool is billed per GB-month.
You choose a retention of 2 to 100 recovery points per VM. When a new backup completes, the oldest point beyond your retention is pruned automatically.
Pick a recovery point on the VM's Backups tab and restore. The VM powers off and is overwritten in place with that point, then boots again. Restores stay on the same VM.
Yes. A VM can have multiple schedules, for example daily at 02:00 and weekly on Sunday, all sharing the same retention setting.
A snapshot is a manual checkpoint before a risky change; a backup is scheduled protection stored on a separate pool. One rule to know: a VM with backups enabled cannot also take snapshots, so use backups for anything important.
On a dedicated backup storage pool, separate from the storage your VM's disk runs on. A problem with the VM or its disk does not touch its backups.
Scheduled, incremental, stored separately, and free inside your storage pool. 3,000+ builders run on Raff in our us-east region, backed by a 99.9% uptime SLA in writing.
Guides to help you plan, build, and scale on Raff.
Series
Server HardeningCover cloud server baseline security, patching, vulnerability reduction, post-deployment hardening, and secure decommissioning decisions.
Series
Startup Infrastructure, Cost & ReadinessHelp startup and SaaS teams move from MVP infrastructure to production readiness, cost control, reliability, diligence, and additional platform complexity only when justified.
Series
Database Reliability & OperationsHelp small teams operate production databases with monitoring, replication, failover, backups, recovery testing, incident readiness, and clear operational ownership.