Your backups outlive us.

You are not locked to us surviving

This is the most important fact on the page, so it goes first. Your vault is a plain Restic repository. There is no proprietary format, no custom client, and no ServerCrate-specific anything between you and your data. Restic is open source, widely deployed, and reads its own repositories anywhere they live.

That means your backups do not depend on ServerCrate as a company continuing to exist. If we vanished tomorrow, your snapshots are still standard Restic, and you can point Restic at them from any machine, or copy them wholesale to another host, with the same tool you already use. The lock-in risk that makes single-operator providers scary simply isn't present in this architecture.

Redundancy, stated honestly

Every vault sits on a ZFS pool with cryptographic checksums on every block. Silent corruption is detected on read and healed from redundancy rather than quietly served back to you. On top of that, vault data is mirrored to a second pool on separate disks, so a single drive failure does not put your snapshots at risk.

What we will not claim: we are not a multi-region cloud. Storage runs on owned hardware in a single Los Angeles location with redundant power and a business-grade uplink. That is the right tradeoff for homelab and small-team backup, and the wrong one if you need geographic redundancy. If geo-redundancy matters to you, treat ServerCrate as one leg of a 3-2-1 strategy, not the whole thing. The exit commands below make running a second copy elsewhere trivial.

Restores are verified, not assumed

An untested backup is a hope, not a backup. So we don't assume recoverability, we measure it. An independent machine outside our primary infrastructure restores from a live vault over the real REST endpoint and verifies the result byte for byte against a cryptographic manifest. It runs on a different host than the storage node on purpose, because recovery onto a clean machine is the thing that actually matters.

The exit, in five minutes

You can leave whenever you want, and you do not need us to help you do it. Everything below is stock Restic against your vault's REST endpoint, using the connection string from your portal. Step one proves you have a complete, verified copy. Step two pulls every snapshot to any other Restic-compatible target. Step three is closing your account.

1. Confirm you have everything, verified.

# your vault, your password, both from your portal
export RESTIC_REPOSITORY="rest:https://USER:TOKEN@YOURVAULT.vault.servercrate.net/"
export RESTIC_PASSWORD="your-repository-password"

restic snapshots                 # list everything stored
restic check --read-data         # verify every byte, not just the index

2. Copy every snapshot to a new home. rsync.net, BorgBase, your own rest-server, a USB disk — anything Restic can write to.

# point at the destination and initialize it
export RESTIC_REPOSITORY="sftp:you@your-new-host:/backups/vault"
restic init

# pull the full history out of ServerCrate, preserving snapshots
restic copy \
  --from-repo "rest:https://USER:TOKEN@YOURVAULT.vault.servercrate.net/" \
  --from-password-file sc.pwd

3. Close the account. Cancel from the portal, or email support@servercrate.net. No retention games, no exit fees, no "call to cancel."

Who actually runs this

ServerCrate is built and operated independently, in the open, by a small team that answers its own email. We think that is a feature: there is no support tier between you and the person who can fix your problem, and the security model is designed so that the size of the operator never has to be something you trust. The architecture, not the headcount, is what protects your data.

One point of clarity, because it comes up: an older, unrelated game-server hosting business once used a similar name. This ServerCrate — encrypted Restic backup on ZFS — is a separate operation with its own infrastructure, founded in late 2025. If you find old threads about a different "ServerCrate," they are not about this product.