Re: fatal signal and .rfl locks

[Eric Siegerman]

On Thu, May 31, 2001 at 01:01:51PM -0400, Donald Sharp wrote:
> On Thu, May 31, 2001 at 08:30:43AM -0700, Dennis Jones wrote:
> > When I do an update, I get the message,
> > 
> > "Terminated with fatal signal 11"
> > 
> > It seems to be happening in the same place everytime, so I went to the
> > directory on the CVS server where it stops, did an "ls" and found some
> > #cvs.rfl.... files there.  If I try to list those files (using ls -la) to
> > get their time stamp, I get a segmentation fault.
> 
> You get a seg fault from ls?  
> 
> That's bizarre.  Are you setting your LD_LIBRARY_PATH to something strange?
> 
> If not you could be having some sort of weird disk fault/errors..

Or a hacked system.  Say they replaced /bin/ls with a custom
version that wouldn't show the files they planted -- a fairly
common trick -- but their bogus ls program is broken.  Or they
replaced libc.so.X with a broken version, for similar reasons --
that'd explain why both cvs and ls have similar problems.

Or some other problem.  Either way, I'd be very worried until
I'd figured out what was going on.

Or maybe something just went bad in the running kernel, and the
reboot's all that was needed.  But I'd want to put a fair amount
of effort into convincing myself that's all it was.

--

|  | /\
|-_|/  >   Eric Siegerman, Toronto, Ont.        address@hidden
|  |  /
With sufficient thrust, pigs fly just fine. However, this is not
necessarily a good idea.
        - RFC 1925 (quoting an unnamed source)