Backups You Can Actually Restore
Backup engineering, restore testing and disaster-recovery planning.
Synchresis Solutions designs, tests and monitors backup and disaster-recovery arrangements for servers and virtual machines. We work with 3-2-1 offsite backups, ZFS snapshots and replication, Restic, Borg, and bare-metal and VM restores.
A successful backup job is only part of the evidence. The useful question is whether you can recover the required data and applications into a working environment when you need them.
Plan the recovery as well as the backup
3-2-1 and offsite copies
Review where the copies live, which failures they protect against and who can access them. Check retention and encrypted storage against the data you need to recover.
Snapshots and replication
Use ZFS snapshots and replication with a clear view of what they retain. Review how deletions, corruption or an unavailable host affect the recovery options.
Restore testing
Test recovering files, guests or a complete system. Check the recovered application and its dependencies, not only whether the restore command finishes.
Recovery objectives
Plan recovery point and recovery time objectives around the actual business impact. Identify the systems, credentials, network paths and people needed for the runbook.
Recovery across platforms
We work across Linux, FreeBSD, Windows Server and virtualization environments. A recovery plan needs to account for disk formats, guest drivers, firmware, DNS and the application’s own state. Our virtualization support and server engineering services address those dependencies.
If an incident is already underway, tell us which backup copies remain and what has happened since the last known good state. Recovery depends on what is available; we establish that before proposing changes.
Questions before you get started
Can you check whether our existing backups work?
Yes. We can review the configuration and work with you on a restore test that checks the recovered service.
Can replication replace every other backup?
The right design depends on the failures you need to survive. We review retention, isolation and recovery requirements rather than assuming one replicated copy covers every scenario.
Can you guarantee recovery before seeing the backups?
No. We first establish the available copies, their state and the system dependencies. Recovery objectives and the proposed work are agreed for your environment.
Review your backup and recovery plan
Tell us what runs, what is backed up and which outage you need to survive. For an active incident, call.