Just an hour ago I rebooted my unraid 5.12a server because I couldn't write to my shares. When the reboot finished I noticed a red-balled drive (disk7 sdf).

This drive is mounted in a eSATA tower and there was a power-outage. This tower wasn't on a UPS, while the server where it is connected to is.

 

What is the best thing to do now? Can I just stop the array-unassign the drive-start the array-stop the array-reassign the drive-start the array?

 

Do I risk that the drive is getting cleared and formatted in this sitation?

 

Is it better to unassign the Parity disk too?

 

Syslog is attached.

syslog-2012-12-02.zip

Can someone pls help?

 

The best approach is to

Stop array.

Unassign drive.

Start array with it unassigned.

Stop array once more.

Reassign drive.

Start array and let the drive get reconstructed.

 

Do NOT remove the parity disk. It is critical to the reconstruction.

 

Remember,the drive turned RED when a write to it failed, so it must be reconstructed.

 

Joe L.

Thanks Joe, I knew I missed something. Will the rebuild automatically start after reassigning the drive?

 

When I now play a file that is on this drive, will it in fact be "reconstructed" from the parity disk?

Data rebuild in progress now, going at a low speed of approx. 35 MB/s. One of the drawbacks of an eSATA tower. Time to build all drives in one case. A lot less clutter and less prone to errors. And faster off course.

Once it has rebuilt, I would recommend you to do a parity check (with NO corrections written to the parity drive).  This will check that the data just written to the restored drive can all be read correctly.