Off-site backup hosting: capacity, encryption and restore time
Plan off-site backup storage using recovery objectives, independent keys, deletion permissions and a full restore test. Compare prepared Vault capacities.
Content revised · English
- Architecture to evaluate
- Recovery point / recovery time / keys
- Prepared starting points
- 3 plans
- Responsibilities
- Application, rights and recovery
A fit shortlist, not installed software or live inventory.
Keep these in view.
- Recovery point / recovery time / keys.
- Make deletion resistance an actual control.
- Test a complete restore.
Recovery point / recovery time / keys
Choose how much recent work you can lose (RPO) and how long recovery may take (RTO). Restic, Borg and Kopia can provide encrypted incremental backup workflows; rclone is a transport tool, and a self-hosted S3-compatible service is another system to maintain.
Make deletion resistance an actual control
A snapshot, a versioned repository or an encrypted archive is not automatically immutable. Retention enforcement, object-lock configuration, separate administration and deletion permissions determine the boundary. Keep recovery keys outside both the source host and the public web root.
Test a complete restore
A theoretical 1 TB transfer at 1 Gbps needs about 2 hours 13 minutes before overhead; decompression, database restore and validation add time. Test an application restore under realistic bandwidth, then record the result and remaining dependencies. A successful backup job is only one stage.
Record the recovery test
Document the source data size, backup date, restore destination, effective transfer rate, decryption steps, application validation and elapsed recovery time. Retain failures as well as the successful result. Confirm that someone can retrieve the keys and instructions when the original server and its credentials are unavailable.
| Mechanism | Useful for | Independent protection requires |
|---|---|---|
| Snapshot | Fast rollback of disk state | Separate backup and application consistency |
| Encrypted archive | Confidentiality of backup contents | Separate keys and restore testing |
| Versioned repository | Earlier file states | Deletion controls and retention |
| Immutable design | Resistance to deletion for a period | Enforced lock, tested permissions and separate admin |
Capacity to consider.
Match these starting points to the representative application. Software installation and live inventory are not connected; the exact specification and period total come from the catalogue.
Vault 2
Store backups, archives or authorised media on a 2 TB HDD configuration with two virtual CPUs and 4 GB of memory. The 10 TB monthly transfer allowance and 1 Gbps port suit scheduled file movement, with rTorrent and Jellyfin as the listed software stack choices.
- Compute
- 2 vCPU
- Memory
- 4 GB DDR4
- Storage
- 2 TB HDD
- Transfer
- 10 TB / month
billed monthly
Vault 8
Grow an archive to 8 TB of HDD storage with four virtual CPUs, 8 GB of memory and unmetered transfer on a 1 Gbps port. A seedbox software stack and an S3-compatible object-storage option support file workflows, with access controls and retention configured for your use case.
- Compute
- 4 vCPU
- Memory
- 8 GB DDR4
- Storage
- 8 TB HDD
- Transfer
- Unmetered
billed monthly
Vault 16
Combine 16 TB of HDD storage, four virtual CPUs and 16 GB of memory for larger backup sets and file collections. The configuration adds cold-storage snapshots and LUKS encryption to unmetered transfer on a 1 Gbps port, while recovery and key management remain part of your storage design.
- Compute
- 4 vCPU
- Memory
- 16 GB DDR4
- Storage
- 16 TB HDD
- Transfer
- Unmetered
billed monthly
Before this workload goes live.
- Define maximum acceptable data loss and total application recovery time.
- Estimate retained backup size, growth and effective transfer capacity.
- Separate the recovery key and repository administration from the source host.
- Verify deletion and retention permissions with the intended backup tool.
- Restore the application to a separate destination and record the elapsed result.
Questions, answered.
Is S3 object storage included?
A storage server and a managed object endpoint are different products. A compatible self-hosted service can be designed, but no S3 endpoint is created.
Are backups encrypted automatically?
No automatic backup workflow is connected. Configure encryption in the selected tool and keep its recovery keys outside the source and repository administration paths.
Does Borg or restic make the repository immutable?
Not by itself. Enforced retention and deletion permissions, including separate administration or an appropriate lock design, determine deletion resistance.
How fast can I recover?
Measure the whole restore and application validation. A theoretical 1 TB at 1 Gbps already takes about 2 hours 13 minutes before overhead, restore processing and testing.
What is kept in my account workspace?
Your configuration, selected plan, image, location and billing term are kept with the order and payment request references. Quotes, payment-state reads and support conversations remain linked to your account. Read fulfilment details separately from payment status.