Two drives are failed or missing
RAID 6 can tolerate two unavailable members only when the remaining set is readable and represents one consistent array state.
Review this condition →RAID 6 data recovery
RAID 6 data recovery after multiple drive failures.
A RAID 6 may be offline even when several members remain readable. ADR evaluates each drive independently, protects unstable sources, and reconstructs the array away from the failed controller before recovered data is written elsewhere.
The array does not need to be forced online, rebuilt, or mounted before ADR can begin evaluating the member set.
Failure historyIdentify when each member changed
Member coverageCheck readability independently
Two parity setsValidate geometry and consistency
Protected resultRecover to separate storage
Start with the symptom
What changed before the RAID 6 went offline?
RAID 6 protects against two member failures during normal operation. Recovery becomes a different problem when failures occurred at different times, a rebuild changed one member, or additional drives are readable only in part.
A third member is unstable
Stop the rebuild and protect the available sources before overlapping read errors reduce the remaining recovery options.
Review this condition →The rebuild failed or stalled
Keep the removed member and partial rebuild target. Do not restart the process merely to see whether it advances.
Review this condition →Another member failed during the rebuild
Stop before additional reads or writes compound an already inconsistent member set.
Review this condition →The virtual disk is missing
Healthy drive labels do not prove the controller has assembled the correct member set or geometry.
Review this condition →The array changed after intervention
Preserve every original and replacement member and record each rebuild, import, force-online, or repair action.
Review this condition →The configuration is lost
Do not recreate the RAID 6 from memory. Order, offsets, stripe geometry, and parity behavior can be tested virtually.
Review this condition →The recovery path
Protect the members before deciding which array state to trust.
ADR works from the evidence on the individual drives. Candidate layouts are tested without asking the controller to rewrite the only available member set.
- 1
Document the set
Record the enclosure, controller, RAID level, bay order, serial numbers, failures, replacements, and exact sequence of events.
- 2
Protect unstable sources
Image members that are slow, inconsistent, unreadable in regions, or rejected by the controller.
- 3
Test the layout
Establish member generations, order, offsets, stripe geometry, and dual-parity behavior outside the controller.
- 4
Recover priority data
Validate the reconstructed filesystem and recover the files, databases, or virtual machines needed first to separate storage.
We can get started now
Tell ADR what the RAID 6 reports today.
Start with the controller or NAS model, member count and capacities, current status of every drive, rebuild or replacement history, and the systems or data the business needs first.