Well, this request was a dud: no responses. But, what the heck, let
me try again. What is the procedure to get the kernel to report some
or all of the request sense bytes for SCSI errors?
For example, this appears to be the default error reporting...at
least in my case:
> > SCSI disk error : host0 channel0 id1 lun0 return code = 2800002
> > Info fld = 0xb19573, current sd08:10: sense key hardware Error
> > SCSI Disk I/O error : dev 08:10, sector 11638130
That is not terribly useful.
Dan
Dan Jones wrote (3/14):
>
> I wish to understand the settings and different reporting levels of
> the "scsi log token" where token <= error. I think that I have that
> right. In particular, I am interested in improved error reporting
> i.e. I really want to log both hard and soft errors with full request
> sense data without a blizzard of trivia. Doable?
>
> One of my software buddies took pity on me and found:
>
> /*
> * Usage: echo "scsi log token #N" > /proc/scsi/scsi
> * where token is one of [error,scan,mlqueue,mlcomplete,llqueue,
> * llcomplete,hlqueue,hlcomplete]
> */
>
> This gives me license to play around, but without the ability
> to generate different kinds of errors, it is all guesswork to
> me.
>
> I already know that I don't want the top level as it reports
> timers ad nauseam.
>
--
Dan Jones, Storage Engineer VA Linux Systems
V:(408)542-5737 F:(408)745-9130 1382 Bordeaux Drive
[EMAIL PROTECTED] Sunnyvale, CA 94089
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to [EMAIL PROTECTED]
