Operations

Unraid OS update maintenance

Before the update

1. Back up the Unraid boot device configuration.
2. Check and apply plugin updates.
3. Stop Docker and, if applicable, VM services.
4. Stop the array cleanly and confirm it is unmounted.
5. Review the target release and confirm the rollback path.

Apply a stable release

1. Open the OS update interface and select the stable branch.
2. Download and apply the target release.
3. Wait for the completion message before restarting.
4. Reboot and monitor the boot.

Post-update validation

Host

• Management interface reachable.
• Array starts cleanly.
• Filesystems mount without errors.
• Hardware drivers load.

Service

• Docker engine starts.
• Dependent containers start in order.
• Reverse-proxied names resolve.
• Metrics and health probes recover.

Hardware consideration

After a platform-level change, verify boot priority and hardware-specific driver modules before restoring the full application workload.

Change record

Record for each maintenance window: previous version, target version, backup location, start time, outcome, observed issues, and whether rollback was required.

Backup and recovery

Asset	Protection	Recovery use
AdGuard Home	Baseline config stored on Unraid	Rebuild DNS service quickly
Caddy	Config backup stored on Unraid	Restore reverse-proxy routing
Unraid boot device	Configuration archive before updates	Restore host settings and boot config
Proxmox workloads	Proxmox Backup Server under evaluation	Test host and workload recovery

Maturity

Configuration backups and documented rollback are in place. Proxmox Backup Server is being tested and is not represented as a completed production backup platform.

Rollback paths

Management interface available

1. Open the OS update page and select the automatically preserved previous release.
2. Restore and reboot.
3. Perform post-rollback validation.

Host not bootable

1. Power down safely.
2. Mount the boot device on another system.
3. Restore the previous release files or a known-good backup.
4. Reinstall the device and validate boot and array state.

Recovery priorities

1. Host access and storage integrity.
2. DNS and private network access.
3. Reverse proxying and certificates.
4. Monitoring, before lower-priority applications.
5. Application containers, in dependency order.

Operational checklists

After a service change

• [ ] Configuration validates.
• [ ] Service reloads or restarts cleanly.
• [ ] FQDN resolves from a LAN and a mesh client.
• [ ] Certificate is valid for the requested name.
• [ ] Upstream responds through Caddy.
• [ ] Prometheus target is healthy.
• [ ] Grafana shows current data.
• [ ] Health script completes without false recovery.

During an incident

See the incident-response checklist.