mdadm problem with replacing disk

Hi all,

In a mdadm RAID-5 setup I have to replace a disk whose health is deteriorating (although not yet kicked out my mdadm).
So I added a new disk with

mdadm /dev/md0 --add /dev/sdf
mdadm --detail

shows the new disk correctly as a spare.

After issuing

mdamd /dev/md0 --replace /dev/sda

mdadm starts rebuilding, but after a while stops. That is, the process is still there, the disk led on the system lights continuously without flickering, but nothing seems to happen anymore, and

mdadm --detail /dev/md0

hangs forever. Also the system won’t shutdown properly (I have to do a hard reset).

Nothing can be seen in the journal or in the dmesg output.

To be sure I did a new try with the RAID-system unmounted, but the result is the same.

The data on the disk can be read: I made a full backup before issuing the replacement.

Any idea what might be happening here?


Anyone any ideas? Are there logs available I might have overseen?