This message is from the T13 list server.

On Wed, 29 Jan 2003 15:19:50 -0700, Edward A. Gardner wrote:
>This message is from the T13 list server.
>Actually it's a trivial change for ATAPI.  All it requires is that the 
>protocol spec (ATAPI spec) require that the host (SCSI initiator) issue a 
>REQUEST SENSE immediately after receiving a CHECK CONDITION status.

Huh? I don't get it... We need to require a host to do something it
should do anyway? OK, maybe it should be required just to flush out
the standard and make it more complete.

It seems to me there is no excuse to not issue a REQUEST SENSE
following any check condition (aka error). You need the REQUEST SENSE
data to log to the system error log file(s). 

OH NO! WHAT AM I THINKING? SORRY... I forgot we are talking about toy
computers here. Toy computers don't need error logs - just reboot
them when you see the blue screen. Yea, right, that's the way it
works!

No wonder IBM still sells large numbers of their mainframes - systems
that really don't break - systems that recover from hardware errors
and software problems without being rebooted - systems that  really
will run 24x365 without being rebooted.

Hale



*** Hale Landis *** www.ata-atapi.com ***



Reply via email to