When RAID 5 Remembers Something You Don’t
You reboot the server… and suddenly the controller insists the array is foreign.
No drives failed. Nothing was touched. No maintenance window.
Just a warning that feels both urgent and completely unclear.
This message doesn’t mean “import now.”
It means the controller found a mismatch between what the disks say and what its cache/NVRAM believes—and it’s protecting your data until the conflict is resolved.
Handled wrong, this is where data loss starts.
What You See
- “Foreign Config Detected” on boot or during POST
- All drives show Online or Optimal
- Array refuses to mount
- Controller prompts: Import / Clear / Ignore
- Logs show config mismatch, stale timestamps, or unexpected metadata
Why It Happens
1. Metadata Epoch Drift
If the controller’s cache/NVRAM wasn’t fully committed before shutdown, on-disk metadata is now “older” than the controller expects.
Ref: TN-R5-001 §4, §6
2. Mid-write power loss
A single incomplete parity-write cycle can break alignment, leading to a foreign warning instead of a rebuild.
Ref: TN-R5-001 §6
3. Silent RAID 5 penalty (read-modify-write)
If a survivor hit a latent sector error, the controller may distrust its own parity and refuse to assemble the array.
Ref: TN-R5-001 §3
4. Controller safety lockout
If identity signatures (slot, WWN, timestamps) don’t match, controllers stop to avoid corrupting the stripe map.
Ref: TN-R5-001 §4, §8
5. Accidental drive shuffle
Even brief reseating or moving one drive to a new slot can trigger mismatched IDs.
Ref: TN-R5-001 §4
What NOT To Do
- Do NOT import the foreign config immediately
- Do NOT clear foreign (erases layout references)
- Do NOT force the array online
- Do NOT rebuild until parity is validated
- Do NOT re-seat or reorder drives trying to “fix it”
Any of these can lock the array into the wrong layout—or overwrite the last good metadata.
What You CAN Do
- Freeze the configuration exactly as it is
- Export controller config
- Capture NVRAM/cache state
- Clone all member disks sector-by-sector
- Verify slot → serial → WWN identity
- Validate metadata timestamps and signature coherence
- Reconstruct parity externally before admitting drives back into the controller
This preserves the only evidence needed to restore the real RAID 5 layout.
Diagnostic Overview
- Array Type: RAID 5 — Single Parity Set
- Controller State: Foreign Config Detected / Members Online
- Likely Cause: Metadata Epoch Mismatch or Signature Drift
- Do NOT: Import or Clear Foreign Before Imaging
- Recommended Action: Clone Members, Export Controller Config, Validate Metadata Against TN-R5-001
RAID Triage Center – RAID 5 Triage – RAID 5 Technical Notes