Migrating Proxmox Workloads to New RAID Storage With Verified Recovery Paths

A virtualization host had workloads spread across legacy SAS storage that needed consolidation onto a new three-drive RAID5 virtual disk. A network-critical firewall workload meant that losing the wrong storage at the wrong time could also remove access to the preferred recovery path.

This was a contained storage migration and workload recovery exercise. A separate degraded OS RAID1 was outside the project and was not repaired as part of this work.

The work had to preserve an independent recovery route while legacy disks were rebuilt. Long-running RAID initialization also needed to be treated as its own monitoring gate rather than a reason to interrupt healthy controller activity.

• Verified the preferred remote Proxmox Backup Server recovery path before destructive work.\n- Verified all 12 pieces of an independent portable backup archive with SHA-256 checksums.\n- Verified archive structure and readability, then performed a controlled file extraction and confirmed the extracted file was readable and non-empty.\n- Confirmed current backup verification before proceeding.\n- Moved the network-critical firewall workload away from the disks being rebuilt.\n- Created and validated the new three-drive RAID5 target.\n- Restored or migrated four target virtual machines to the replacement storage.

Created the replacement RAID target and completed the in-scope workload restore or migration after recovery readiness was proven.

The migration kept recovery, workload placement, controller state, and post-change connectivity in view instead of treating the work as a disk-copy task.

Rollback material was retained until the replacement storage and workloads were validated. Legacy workload storage targets were disabled only after that validation.

The replacement RAID5 finished healthy and Optimal with all three members Online. Four target virtual machines were placed on the replacement storage, and post-migration network and connectivity validation passed.

Evidence included verified backup material, archive readability and extraction checks, current backup verification, healthy replacement RAID state, completed workload placement, and post-change connectivity checks.

This demonstrates Proxmox storage migration, RAID and controller operations, backup verification, restore planning, dependency analysis, rollback planning, workload restoration, and post-change verification. The maintenance lesson was clear: customer-visible interruption, reserved maintenance time, contingency time, and long-running background completion are not automatically the same thing.