Re: Remaining SATA (and other) issues

just another remark :

I originally tested on a 'gconcat' device : as said, one of the
improvements of Alexander's/sos_at_ work is I now, after each
crash, can 'fsck -t ufs -p /dev/concat/data' (sometimes minor
human intervention needed) and start over again.

I notice that after my last test just using plain /dev/ad6s1, after
the crash (and reboot) /dev/ad6s1 has gone (and (or, because(?)))
'fdisk partition table' in invalid.

IIRC gconcat stores metadata at the end of the disk, and fdisk
(and 'bsdlabel as well?') is at the beginning. 

Might this indicate some very low numbered sectors are written
erronously?

Best, Arno