RAID 6 failure response
RAID 6 normally tolerates two unavailable members. When a third disk drops out or becomes unreadable, the controller can no longer reconstruct every stripe using normal redundancy.
Understand the symptom
RAID 6 distributes data and two parity values across members. Three missing blocks in the same stripe exceed that protection. Recovery can still benefit from disks the controller marked failed, because they may contain readable areas. Partially rebuilt targets, stale members and disks with different bad-sector maps can contribute complementary data. Reconstruction must establish order, offsets, stripe size, parity rotation and member generation on working images.
What you may see
The controller no longer has enough complete members for ordinary dual-parity reconstruction.
The rebuild workload may expose weak sectors on another survivor while writing a target.
The controller may permit a risky assembly without proving the selected members are coherent.
Unreadable or stale stripes can affect files differently across the volume.
Protect the source
These operations can overwrite parity, metadata and partially recoverable member state.
A failed member may still contain critical readable stripes unavailable elsewhere.
Choosing members solely by health status can combine incompatible points in time.
Do this now
Disable automatic rebuilds, scrubs, initialization and application writes.
Map bays, serial numbers, removal times, replacements and rebuild targets.
Capture RAID level, stripe settings, event logs, alerts and rebuild progress.
Preserve disks marked failed, stale, spare or removed and avoid individual write tests.
NEPFOSYS process
Each member is imaged independently to preserve the best readable areas from all generations.
Geometry, stale state, partial rebuild boundaries and unreadable stripes are evaluated on copies.
Shares, databases, virtual disks and files are validated against business priorities.
A clear next step
Start a case with the exact symptoms, incident timeline, prior attempts, encryption details, and the files that matter most. NEPFOSYS will route the case to the appropriate DFW intake or nationwide mail-in path.
Start with a case reviewQuestions, answered
Normal RAID 6 redundancy cannot reconstruct three missing blocks in the same stripe. Recovery may still be possible when failed members retain complementary readable data.
Not without preserving all members. A forced assembly can use stale or unreadable data and may write new state that complicates independent recovery.
It may contain valid reconstructed stripes up to a boundary that can supplement unreadable areas on other disks.
DFW local · Nationwide mail-in