For IT management
A backup strategy that works out in an emergency.
AegisGrid puts proven rules such as 3-2-1-1-0 into practice with a single tool: several destinations in one run, immutable copies, regular checks, automatic check restores, clear retention and a signed log that shows what happened.
The 3-2-1-1-0 rule with AegisGrid
The German Federal Office for Information Security (BSI) recommends the 3-2-1 rule: three copies, two media, one off site. The 3-2-1-1-0 rule adds an immutable copy and zero errors in the check. With AegisGrid, everything is created in a single run – and on the backup server every step is documented in the evidence log.
| Rule | Meaning | Implementation with AegisGrid | Evidence |
|---|---|---|---|
| 3 copies | The original plus two backups | Up to eight destinations in one run; the source is read only once. | Every incoming copy in the log of the backup server |
| 2 media | Different storage | Local storage, USB disk, backup server in the network, cloud storage or export to tape. | Destinations and result per run, exports in the log |
| 1 off site | One copy at another location | Encrypted transfer over TLS to a remote backup server or to the AegisGrid cloud storage; alternatively mirroring from the backup server on site. | Arrival and mirroring with computer, address and amount of data |
| 1 immutable | Offline or locked against deletion | WORM storage with a retention lock; export of single backups as an archive. | Every deletion and every change of the retention |
| 0 errors | Verified recoverability | Check of all stored objects, comparison with the original, repair from a second copy and automatic check restore that starts the VMs in a sandbox. | Checks, damage and check restores with their result, monthly as a signed report |
What makes the difference in operation
No backup chains
Every run is a complete version that can be restored on its own. There is no dependence on earlier full or incremental backups and no merging – a damaged intermediate version does not endanger the following ones.
Little storage, little bandwidth
Thanks to deduplication, only what the destination is still missing is transferred and stored – also between computers with similar data. Free areas of NTFS and ext file systems are skipped.
Scheduling and dependencies
The job scheduler starts backups on a schedule or after other jobs have finished, also in parallel, and sends results by email. Running backups can be paused or limited in bandwidth.
Retention by rules
Every backup carries its own retention period in days, weeks, months or years – different per destination, seven days locally and twelve months in the cloud storage, for example. A scheduled maintenance job cleans up expired versions; locked versions remain until their lock ends.
Robust against interruptions
If a transfer breaks off, the next backup continues there instead of starting from scratch. If an optional destination fails, the others keep running.
Secure updates
Updates and components are signed with Ed25519; only what the signature confirms is installed. Old versions are not installed again.
One solution for the whole environment
Instead of separate tools per platform, AegisGrid backs up physical servers, virtual machines and Microsoft 365 with the same method and into the same stores. That simplifies documentation, training and emergency planning.
For connecting your own tools, the job scheduler provides a documented HTTPS API.
- Windows Server with VSS shadow copies
- Linux servers with LVM snapshots
- Hyper-V, application-consistent through recovery checkpoints
- Proxmox VE and Xen: whole VMs with all disks
- Microsoft 365: Exchange Online, OneDrive, SharePoint
Questions from IT management
Can AegisGrid back up locally and off site at the same time?
Yes. One run can serve up to eight destinations, a local store and a remote backup server or the cloud storage, for example. The source is read only once, and every destination stores only what it is still missing.
How do we prove that backups can be restored?
AegisGrid checks all stored objects for integrity and, on Linux, can compare a backup with the snapshot it was read from. Results go to the people responsible by email. The backup server also regularly restores backups as a test, checks the file systems and starts virtual machines in an isolated sandbox; the check report comes with screenshots. Everything together is in the signed evidence report. With Managed Backup we set this up for you and evaluate it.
Can AegisGrid be integrated into existing monitoring?
Yes. Results arrive by email, and the job scheduler offers an HTTPS API through which jobs, progress and results can be queried.
Test AegisGrid on your own systems
In a pilot project we back up selected servers and VMs and restore them as a test – with measured times.