dtrace ustack kernel panic

[Andriy Gapon]

on 02/08/2011 20:46 maestro something said the following:
> Hi,
> 
> just finished installing FreeBSD-9BETA1 and recompiling the kernel with dtrace.
> 
> This is even worse, I have the same behavior as mentioned here:
> http://freebsd.1045724.n5.nabble.com/bin-158431-dtrace-crash-in-dt-proc-lookup-when-attaching-to-PID-assert-dpr-NULL-tt4535367.html#none

Kind of a mentoring note: it would be much shorter and much more useful to paste
"PR 158431" or, even better, an http URL to the said PR in FreeBSD PR DB web
interface.  I know, I know, it's really Google to blame, right? :-)

> i.e., dtrace regardless of whether with or without any probes just quits with
> the following error message
> 
> Assertion failed: (dpr != NULL), file
> /usr/src/cddl/lib/libdtrace/../../../cddl/contrib/opensolaris/lib/libdtrace/common/dt_proc.c,
> line 751.
> 
> that's inside dt_proc_lookup()
> 
> I think I have to give up on ustack(), too bad

Or maybe you still have a chance to help us here, maybe it will even help you in
the long term.
It would be good if you pasted a little bit more of the output that you got.
Please note that in that PR the (meaningful) output starts with:
dtrace: no probes specified
I.e. it starts with what dtrace considers to be a fatal condition.

I tried to run dtruss (as you did) and I got this pre-amble before the assertion:
[some dtrace script body]
: probe description proc:::exit does not match any probes

I guess that in my case I got it because my userland was not compiled with CTF
support.  Not sure about yours...

Of course, it's still rather bad that dtrace crashes when it prematurely exits.
But maybe it doesn't crash in the correct environment...  I don't know.

-- 
Andriy Gapon