Mike,
I'm using the diald-0.16 (presumably it has the .5 additions in it) which was supplied
with my SuSE 6.1 distribution, but I too have the
routing errors mentioned below but they relate to the sl0 device. route -nv and
ifconfig -a show much more than is supposed to be there.
When diald tried to close down an idle link it always emits these errors and usually
dies on the spot, often leaving pppd running with
no hangup.
I don't understand any of this. I'm using the same diald configuration that worked
well for me on SuSE 5.3 (2.0.36 kernel). I suspect
SuSE's setup utility has screwed up the routing. Can you give me any idea of what the
routing table should look like with diald down,
with diald up and with diald and ppp up? What default routes should be set up by the
system _before_ diald is started? The system has an
eth0 device as well which leads to some Windows boxes though I haven't set up IP
Masq/firewalling yet.
Apologies for echoing this mail to your own mailbox but my last few posts to the list
have failed to appear.
Mike Jagdis wrote:
> On Sat, 20 Mar 1999, Michael Meissner wrote:
>
> > Or alternatively, download http://diald.unix.ch/diald-0.98.3.tar.gz and install
> > that (I don't remember if there are warnings). Note, I found that using the
> > slip proxy in the 0.98.3 release to be buggy (diald dies),
>
> Hmmm... Now that shouldn't happen :-).
>
> > but the ethertap
> > proxy support seems to work fine, providing you build ethertap directly into
> > the kernel (I had problems when I built it as a module), and create the
> > appropriate /dev/tap{0,1,2,3,4,5,6,7,8} nodes.
>
> You shouldn't need to create /dev/tap? nodes. Diald uses the
> new socket based interface rather than the old device node
> interface (which is only there for backward compatibility).
>
> > For both the 0.16.5 and the 0.98.3 releases, 2.[21].x kernels, and all of the
> > releases of nettools that I recall, I get syslog entries of the form:
> >
> > Mar 19 14:44:34 tinman diald[2749]: Nonzero exit status (7) on command
>'/sbin/route add 207.244.95.150 metric 0 dev tap0'
> > Mar 19 14:44:35 tinman diald[2749]: Nonzero exit status (7) on command
>'/sbin/route add default metric 0 netmask 0.0.0.0 dev tap0'
> >
> > These seem harmless, but probably indicates some tuning is needed for 2.[12].x
> > kernels.
>
> Yeah, they're complaints about routes that already exist or have
> already been deleted. The 2.2 kernels do this automatically. When
> I get round to it I'll stop reporting the errors. At least I will
> if I find where the exit status' are documented...
>
> Mike
--
[EMAIL PROTECTED] Ralph Clark, Virgo Solutions Ltd (UK)
__ _
/ / (_)__ __ ____ __ * Powerful * Flexible * Compatible * Reliable *
/ /__/ / _ \/ // /\ \/ / *Well Supported * Thousands of New Users Every Day*
/____/_/_//_/\_,_/ /_/\_\ The Cost Effective Choice - Linux Means Business!
-
To unsubscribe from this list: send the line "unsubscribe linux-diald" in
the body of a message to [EMAIL PROTECTED]
