Manual

AegisGrid Manual

Everything about setting up and running AegisGrid: from the first test to the backup server and troubleshooting. The examples work as printed, the output comes from real runs.

Program version: 2.0.0-alpha (version 2, deduplicating store) · As of: 6 November 2026

All chapters

1 Introduction
- What AegisGrid can do
- The programs
- Basic terms
- Overview of the flows
2 Installation
- Requirements
- Installing with bkp_setup
- First test
3 Quick start: back up and restore locally
- Creating test data
- First backup
- More backups
- What is in the store
- Restoring
- Important options of the agent
- Choosing the ID hash (--hash)
4 Setting up the backup server (bkp_target)
- Step 1: create a certificate and key
- Step 2: create an API key
- Step 3: start the server
- Step 4: first backup over the network
- Several destinations: sinks
- Firewall and network
- Crash-safe network backup
5 Consistent backups: snapshots and hooks
- LVM snapshots
- Hooks: freezing the guest
- Windows: VSS shadow copy
6 Stores and runs
- Layout of a store
- Backup names per store
- Generations
- Encryption
- S3 object storage
- Viewing stores and runs: bkp_dump
- Managing stores safely
7 Restore
- From a local store
- Straight from the server
- At a certain point in time (point in time)
- From the sinks of a profile
- From an archive
- Damaged data: partial restore
- Virtual machines
- Restoring a VM in place
- Emergency: host set up anew
- Checking the restore
8 Rescue medium: restoring a computer without a system
- Making the medium
- Starting from the medium
- The steps in the program
- Network: WLAN and a fixed address
- Other hardware
- A computer as a new VM
- Limits of the rescue medium
9 Backing up files and folders
- A file profile
- Exclusions
- Restoring files
- A run as a folder
- Files from disk and VM backups
- Ransomware warning
- Checks on file runs
10 Maintenance
- Overview: store-info
- Retention and prune
- Changing the retention afterwards: set-retention
- Deleting a run: delete
- Scrub: checking for bit rot
- Repair: replacing damage from a second copy
- Mirror: mirroring a store
- Export and import: a run as an archive
- WORM: retention locks
- Rebuilding the index
- Certificates and keys
- Scheduling maintenance
- The target scheduler
- Consumption of the jobs
11 Check restore: restoring backups as a test
- The steps
- Counting and reading files
- Checking for viruses
- Where the check data is
- Setting up a chain
- The VM sandbox
- Can the server do it? The system check
- Dependent VMs: first the DC, then Exchange
- Reports and screenshots
12 Evidence for NIS2 and BSI
- What is recorded
- How the log is protected
- Checking, exporting, reporting
13 Monitoring
- Live view: bkp_monitor
- Scheduler jobs at a glance: bkp_sched list
- A job in detail: bkp_sched show
- For scripts: --dump
- Controlling running jobs: pause, background mode, bandwidth
- Progress window on the Windows desktop
- Monitoring the success of backups
- Windows event log
14 Management console (Windows)
- Installation
- Releasing a scheduler for other consoles
- Preparing the scheduler
- Adding a scheduler
- Help in the dialogs
- The main window
- Creating and changing jobs
- Schedule editor
- Destinations and profiles
- Starting, cancelling, logs
- Switching jobs off and on
- Holding running backups
- Progress and control of running jobs
- Restoring files
- Restoring a drive, disk or VM
- Scheduler and email settings
- Evidence
- When the connection does not work
15 Target console (Windows)
- Installation
- Preparing the backup server
- Adding a server
- The main window
- Data: stores and runs
- Browsing files
- Actions and scheduler
- Accesses, releases, settings
- Email settings
- History
- When the connection does not work
16 Web interface
- The first visit: browser warning and fingerprint
- Signing in
- Web users
- Own certificate
- Activation
- When it does not work
17 Scenarios step by step
- Scenario A: backing up a KVM guest over the network (daily, a new generation on Sundays)
- Scenario B: backing up a Windows server with VSS
- Scenario C: complete restore after a disk failure
- Scenario D: restoring to a certain point in time
- Scenario E: tidying up and mirroring storage
- Scenario F: renewing the server certificate and exchanging the API key
18 Backing up Microsoft 365 (bkp_m365)
- App registration in Entra ID
- Setting up a job
- Storage, retention and checking
- Exporting
19 Restoring Microsoft 365
- Restore app in Entra ID
- Writing back
- What comes back
- Result and exit codes
- In the administration console
20 Backing up Hyper-V VMs (bkp_hyperV)
- Requirements
- Backing up
- Settings per VM and job in the scheduler
- Restoring
- vTPM and guardian
21 Contact and branding
- Where it can be seen
- How it gets onto the machines
- Moving to the new names
22 Site server (bkp_sitesrv)
- What the machines show of it
- A site server of your own
- Isolated network
- Site server for partners
- Monitoring and administration API
- When something is wrong
23 Automatic updates (operators)
- Commands
- Setting up
- What happens during a run
- Which update for which system
- Missing libraries
- Rollback
- Activating the installation (bkp_license)
- Trial version and license
24 Security
- Transport: TLS and pinning
- Authentication: API keys and ACL
- Updates: Ed25519 signature
- Data at rest
- Checklist
25 Storage needs and chunk size
- How the chunk size works
- Measuring instead of guessing
- Changing the chunk size
26 Troubleshooting
- Backup problems
- Restore problems
- Update problems
27 Switching for FULL/DIFF/INC users
- Comparison
- What is different when restoring
- Retention and freeing space
- Damage assessment
- Space display
- Schedule before and after
- Checklist for the update at the customer
A Option reference
- bkp_agent
- bkp_restore
- bkp_sched
- bkp_target
- bkp_maintenance
- bkp_monitor, bkp_dump, bkp_restore_legacy
- bkp_updater and bkp_license
B Store format and network protocol
- Objects and packs
- Runs and manifests
- Network protocol
C Known limitations
D Glossary