⚠ Do now
Free tool · Message decoder
Type a word from the screen, the alert email or the boot message, or pick your platform. Each message shows what it means, how urgent it is, the one thing not to do, and the vendor document it comes from.
All 35 messages
What it means: The controller found disks carrying RAID metadata it doesn't recognise as current, usually after disks were moved or temporarily went missing.
Do not import or clear it until the members are imaged. With a tested backup, Dell's order is import first, from the drive that went foreign last.
Sources: Dell: PERC 11 User's Guide: Foreign configuration found error message; Dell: PowerEdge: Importing or Clearing Foreign Configurations on PERC controllers (000139205)
What it means: The controller is holding unwritten (dirty) cache for a virtual disk that went offline or missing, pending import or discard.
Do not discard the preserved cache before importing any foreign configuration. Dell warns you might lose data.
Sources: Dell: PERC 9 User's Guide: Offline or missing virtual drives with preserved cache error message; Dell: PERC 9 User's Guide: Managing preserved cache
What it means: Configured disks were removed or are no longer reachable, often because of SAS cabling.
Do not press a key to continue before checking cables. Proceeding marks these virtual disks OFFLINE and inaccessible.
Source: Dell: PERC 9 User's Guide: Missing disks in virtual disk error message
What it means: Whole virtual disks can't be found, because disks were removed or have become inaccessible.
Do not continue past the prompt. Proceeding removes those virtual disks from the configuration, so power off and check the cables.
Source: Dell: PERC 9 User's Guide: Missing virtual disks error message
What it means: A redundant virtual disk has one or more failed or inaccessible member disks but is still running.
Do not pull a second disk. Back up or image first, then replace only the failed member.
Source: Dell: PERC 9 User's Guide: Degraded state of virtual disks
What it means: A drive's SMART predictive-failure flag is set and the drive may need replacing soon.
Do not replace the drive unless all other drives in the array are online.
Source: HPE: HPE ProLiant Gen9 Troubleshooting Guide, Volume II: Error Messages (Part 795673-007, Edition 7)
What it means: A rebuild or surface scan found bad sectors it couldn't recover, because fault tolerance was already reduced or the same spot is bad on another drive. Reads of those blocks will return errors.
Do not assume the rebuild left the data intact. HPE recommends back up and restore.
Source: HPE: HPE ProLiant Gen9 Troubleshooting Guide, Volume II: Error Messages (Part 795673-007, Edition 7)
What it means: Logical drives that previously failed are attached, often after an incorrect power-up or power-down order.
Do not press F2. It accepts data loss and re-enables the logical drives.
Source: HPE: HPE ProLiant Gen9 Troubleshooting Guide, Volume II: Error Messages (Part 795673-007, Edition 7)
What it means: A logical drive has failed because a drive or cable is failed or not installed correctly.
Do not swap drives before checking that cables are secure and drives are fully seated.
Source: HPE: HPE ProLiant Gen9 Troubleshooting Guide, Volume II: Error Messages (Part 795673-007, Edition 7)
What it means: The controller sees no valid array configuration. Connections may be missing or wrong, or drives were moved.
Do not create a new array. HPE says power down immediately and return the drives to their original locations.
Sources: HPE: HPE ProLiant Gen9 Troubleshooting Guide, Volume II: Error Messages (Part 795673-007, Edition 7); HPE: HPE Smart Array Controllers – Drive Procedures (a00020470en_us)
What it means: A failed or replacement drive has not been rebuilt yet, and the controller asks whether to start the rebuild.
Do not press F1 yet: F1 starts the rebuild. With no tested backup, power down at the prompt and have the drives imaged.
Source: HPE: HPE ProLiant Gen9 Troubleshooting Guide, Volume II: Error Messages (Part 795673-007, Edition 7)
What it means: A drive is failed, wrong, unsupported or disconnected, so the logical drive is running without full fault tolerance.
Do not remove another drive. HPE warns all data might be lost if another drive fails.
Source: HPE: HPE ProLiant Gen9 Troubleshooting Guide, Volume II: Error Messages (Part 795673-007, Edition 7)
What it means: Replacement drives went into the wrong bays, or are the wrong type or size.
Do not press F2. It resets the configuration and all data is lost. Put the drives back in the correct bays.
Source: HPE: HPE ProLiant Gen9 Troubleshooting Guide, Volume II: Error Messages (Part 795673-007, Edition 7)
What it means: Drives aren't responding because of a failed drive, a loose or failed cable, an unpowered external enclosure or an incorrect power sequence.
Do not press F2 to fail the non-responding drives. Power down and check cables and enclosures first.
Source: HPE: HPE ProLiant Gen9 Troubleshooting Guide, Volume II: Error Messages (Part 795673-007, Edition 7)
What it means: Power was lost while data sat in the write-back cache. The data was preserved and is being written out, and HPE says nothing was lost.
Do not keep cutting power abruptly. Use orderly shutdowns.
Source: HPE: HPE ProLiant Gen9 Troubleshooting Guide, Volume II: Error Messages (Part 795673-007, Edition 7)
What it means: Backup power for the cache was unavailable (battery depleted or disconnected, or cache backup or restore failed), so cached writes were lost.
Do not assume files written just before the outage are intact. Check them before relying on them.
Source: HPE: HPE ProLiant Gen9 Troubleshooting Guide, Volume II: Error Messages (Part 795673-007, Edition 7)
What it means: The Smart Storage battery that protects the controller cache is degraded, failed or discharged.
Do not just reboot. HPE's first step is to power down and remove all power from the server.
Source: HPE: HPE ProLiant Gen9 Troubleshooting Guide, Volume II: Error Messages (Part 795673-007, Edition 7)
What it means: A physical or logical drive in the array has failed.
Do not rebuild or recreate anything until the failed physical and logical drives are identified.
Source: HPE: HPE ProLiant Gen9 Troubleshooting Guide, Volume II: Error Messages (Part 795673-007, Edition 7)
What it means: A virtual drive with double redundancy (such as RAID 6) has lost a member but still has some protection left.
Do not leave it unattended. One more failure removes the remaining protection.
Sources: Broadcom: Glossary, 12Gb/s MegaRAID Tri-Mode Software; Broadcom: StorCLI2: Virtual Drive Information Display
What it means: Member failures have made the virtual drive's data inaccessible.
Do not delete and recreate the array. Dell notes that path loses all data on it.
Sources: Broadcom: Glossary, 12Gb/s MegaRAID Tri-Mode Software; Broadcom: StorCLI2: Virtual Drive Information Display; Dell: PowerEdge: How to fix Double Faults and Punctures in RAID Arrays (000139251)
What it means: Controller firmware found an unrecoverable error on the drive and dropped it from use.
Do not force the drive back to good and into the array before the cause is diagnosed.
Source: Broadcom: Glossary, 12Gb/s MegaRAID Tri-Mode Software
What it means: RST shows the RAID volume as failed. On RST 19.5, 20.0 and 20.1 a RAID 1/5/10 volume can be wrongly shown as Failed instead of Degraded after a SATA drive is unplugged during active I/O, which blocks access.
Do not hot-unplug any other SATA member drive. Check the RST version against Intel's fixed releases.
Source: Intel: [Critical][RST 19.5] RAID 1/5/10 Array Failure Causes Data Corruption (000099952)
What it means: One or more member drives have failed or stopped working. The system may still run but is at risk of data loss.
Do not keep running degraded without a backup and a replacement drive.
Source: Intel: Rebuild Degraded Intel RSTe RAID Array Using Auto Rebuild Feature (000024286)
What it means: Drives failed or went missing and hold outdated copies of the data, so resilience is reduced but data is still readable.
Do not reset (wipe) any pool drive to 'fix' it. Reconnect or replace drives, then repair.
Source: Microsoft: Storage Spaces and Storage Spaces Direct health and operational states
What it means: Drives are failed or missing, but those drives hold up-to-date copies, so resilience is reduced.
Do not replace the missing drives before trying to reconnect them, because they hold current data.
Source: Microsoft: Storage Spaces and Storage Spaces Direct health and operational states
What it means: The virtual disk has lost data because too many drives failed.
Do not delete and recreate the virtual disk unless a backup exists. That is Microsoft's restore-from-backup path.
Source: Microsoft: Storage Spaces and Storage Spaces Direct health and operational states
What it means: Too many drives used by the virtual disk failed, are missing or have stale data, so it is detached.
Do not replace failed drives before every missing drive and server is reconnected.
Source: Microsoft: Storage Spaces and Storage Spaces Direct health and operational states
What it means: The pool lost quorum (most drives failed or offline), and Windows locked its configuration read-only.
Do not force the pool back to read-write before the missing drives are reconnected.
Source: Microsoft: Storage Spaces and Storage Spaces Direct health and operational states
What it means: The drive has become separated from the pool.
Do not run Reset-PhysicalDisk while the drive may hold the only good copy. Reset erases the drive.
Source: Microsoft: Storage Spaces and Storage Spaces Direct health and operational states
What it means: Storage Spaces found old pool metadata on the drive. It is usually a temporary state.
Do not jump to Reset-PhysicalDisk. It wipes the drive, so try Repair-VirtualDisk first.
Source: Microsoft: Storage Spaces and Storage Spaces Direct health and operational states
What it means: The drive carries metadata from a different pool.
Do not reset the drive if it came from another pool you still need. Reset wipes it.
Source: Microsoft: Storage Spaces and Storage Spaces Direct health and operational states
What it means: The drive is missing. On Storage Spaces Direct a server may be down.
Do not replace or reset the drive before checking for a down server or a disconnected cable.
Source: Microsoft: Storage Spaces and Storage Spaces Direct health and operational states
What it means: The drive reports that it is close to failing.
Do not put off the replacement.
Source: Microsoft: Storage Spaces and Storage Spaces Direct health and operational states
What it means: The array is defined but not running, and every I/O to it errors.
Do not run mdadm --assemble --force or --create without images of the members. mdadm warns a forced array may contain data corruption.
Sources: Linux kernel documentation (kernel.org): RAID arrays (md) – Linux kernel admin guide; mdadm project, via man7.org: mdadm(8) manual page
What it means: The member was kicked out of active use because of a detected fault or unacknowledged bad blocks.
Do not remove another member while the array is degraded.
Source: Linux kernel documentation (kernel.org): RAID arrays (md) – Linux kernel admin guide
No listed message matches . Try one or two distinctive words from the screen, or choose All. If it is not listed, photograph the screen and call 1300 806 557.
Messages are quoted exactly as each vendor publishes them. Where the vendor shows a placeholder such as X, # or (x), your screen shows a bay, disk, volume or device name instead. The meaning and the Do not line are a plain-English summary of the linked vendor document, not a diagnosis of your system. If your screen shows something that is not listed, photograph it and call 1300 806 557.
Linux, ZFS or SMART output? Use the evidence reader → Rebuild or stop? checker →
Questions
Dell PERC, HPE Smart Array, Broadcom MegaRAID, Intel RST, Windows Storage Spaces and Linux mdadm: 35 messages in all.
Each message is the vendor’s own wording, and each one links to the vendor document it comes from.
Photograph the screen and call us before anyone clicks anything.
Message on the screen
Photograph the screen and talk it through with an engineer before anything is written to the drives.