new kernels and disk crash!

[Riccardo Mottola <riccardo.mottola%libero.it@localhost>]

Hi,

I had to reset and/or power-cycle my SS20 a lot this past fortnight, due 
to Unstable 10.0.

It worked quite well for days using single-processor. I put it under 
load and the ambient temperature wasn't too hot either.

After retrying MP again I get a freeze, reboot and fsck dies with bad 
sectors:

Starting file system checks:
[ 474.0821440] sd1(esp0:0:3:0):  Check Condition on CDB: 0x28 00 00 4d 
9d 60 00 00 20 00
[ 474.1337740]     SENSE KEY:  Media Error
[ 474.1458575]    INFO FIELD:  5086569
[ 474.1575165]  COMMAND INFO:  -1 (0xffffffff)
[ 474.1675130]      ASC/ASCQ:  Read Retries Exhausted
[ 474.1748000]          SKSV:  Actual Retry Count: 172
[ 474.1831180] sd1a: error reading fsbn 5086560 of 5086560-5086591 (sd1 
bn 5086560; cn 3418 tn 3 sn 18)
CANNOT READ: BLK 5086560

is this just bad luck and coincidence? Or can a bad kernel have ruined 
the disk and/or be the disk okay and kernel being stupid? This fsck 
under single-processor, so stable.

I fear I need to find a new disk, not easy a "not noisy" server disk.

sd1... which disk is it on the SS20.. scsi 3. Do you remember if the top 
or bottom one?

Riccardo