per-connection sshd doesn't always pass on SIGQUIT

[Bob Proulx]

Damien Miller wrote:
> Philippe Cerfon wrote:
> > Are you going to merge the patch of yours?
>
> Does it solve your problem?

Was there an actual problem?  As I understood it there was no problem
but just the question of why SIGQUIT was different.

I think SIGQUIT should not be passed on because SIGQUIT is different
in that the purpose of it is for debugging to dump a core image and it
does not make sense to pass it along.  Adding passing it along would
actually create a bug in that then if used it would cause multiple
processes to dump a core image, all named "core" on traditional
platforms, and the last one dumped would be the one there rather than
the one targeted with the SIGQUIT.

Bob