Re: ktrace-ing a command that locks up the machine

[Mouse <mouse%Rodents-Montreal.ORG@localhost>]

>>> Suggestion: put the ktrace file on a filesystem mounted -o sync.
>> That (with ktrace -s) gave me an empty file.
> If you've got a spare disk (or space partition) try
>  	ktrace -f /dev/rwdxx ...
> Maybe that will help.

Maybe...but does ktrace always write sector-aligned sector-sized
chunks?  If not, /dev/rwd* probably won't even accept the writes.

But I'd be surprised if this works any better than a filesystem mounted
-o sync, even if the writes are suitably aligned and sized.  Might be
worth trying - under some circumstances even things with a low
estimated chance of working are - anyway....

At this point, if I were the person dealing with this, I'd start
scattering debugging in the relevant kernel code paths.

/~\ The ASCII				  Mouse
\ / Ribbon Campaign
 X  Against HTML		mouse%rodents-montreal.org@localhost
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B