While I realize that I may not be able to expect to run at
19200 without errors, it is annoying that both ppp and slip seem to
lock up after a certain number of errors ... only at 19200.

	I have run for four days connected at 9600 --- I managed to
get down that monstr sup during that time.  I hope that I don't have
to do that anytime soon.

	Anyways... after a small amount of intensive transfering to
the netbsd machine at 19200 using either ppp or slip, I find that I
can still see the router trying to send something (receive light
flashing no the modem), but my netbsd machine fails to send after
that.  In fact, kill -HUP does not have the expected effect on the
pppd process.

	When I hangup the modem (read power cycle), pppd quits as
expected.  I can then restart a 9600 connection which will work
forever.

	Is this some behaviour that is inherent in the kernel?

	I have also been getting mysterious sig 11's on my compile
processes when I try to compile the new source... by I'm working on
tracking that down myself.

Dave.

-- 
----------------------------------------------------------------------------
|David Gilbert, PCI, Richmond Hill, Ontario.  | Two things can only be     |
|Mail:      dgilbert@pci.on.ca                |  equal if and only if they |
|http://www.pci.on.ca/~dgilbert               |   are precisely opposite.  |
---------------------------------------------------------GLO----------------