On Tue, Dec 21, 1999 at 07:07:07AM +0000, Martin Veasey wrote:
> Please forgive me if this has been asked 10,000 times before - I can't
> find an answer using on-line resources. I'm having difficulty establishing
> an on-demand modem dialout to my ISP.
>
> pppd works fine when handled manually, dialing, negotiating and
> setting up a defaultroute for my small network.
>
> Using diald is proving to be more of a problem. It appears to load OK
> from boot.local, but appears not to work ... specifically, it doesn't
> pick up it's role as the default router to the outside - pinging
> external addresses whether names or IP numbers just cause lock-up and
> time-out.
This seems very weird considering that the sl0 interface is set up as
defaultroute. Is diald receiving the requests? Is it attempting to
connect? Perhaps the connect script is failing. I wish I could tell
exactly what was going wrong from this side of things.
If diald is not attempting to dial, have you made sure that the daemon
is actually running? Have you attempted to bring up the connection
using dctrl?
Those are some things you can try to attempt to get a better idea of
what's happening.
> Starting the diald daemon manually leads to a number of interesting
> effects:
>
> From the log:
>
> Dec 19 22:45:57 rusty diald[430]: SIGTERM. Termination request
> received.
> Dec 19 22:45:57 rusty diald[430]: Diald is dieing with code 0
> Dec 19 22:46:45 rusty modprobe: modprobe: Can't locate module tap0
> ... 14 other module error messages (from this out-of-the-box SuSE 6.3
> installation and make/install modules doesn't fix it) ...
> Dec 19 22:46:45 rusty modprobe: modprobe: Can't locate module tap15
> Dec 19 22:46:45 rusty diald[848]: start sl0: SIOCSIFMETRIC: Operation
> not supported
The interesting question is why is diald receiving SIGTERM? The other
information is fairly standard stuff. The "Diald is dieing with code 0"
is clearly a result of receiving SIGTERM. The tap0-tap15 errors are
not important, and the SIOCSIFMETRIC error is not abnormal either.
>
> From route -n
> Kernel IP routing table
> Destination Gateway Genmask Flags Metric Ref Use Iface
> "ISP gateway" 0.0.0.0 255.255.255.255 UH 0 0 0 sl0
> "ISP gateway" 0.0.0.0 255.255.255.255 UH 0 0 0 sl0
> 192.168.0.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0
> 127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo
> 0.0.0.0 0.0.0.0 0.0.0.0 U 0 0 0 sl0
>
> looks wrong. Surely there shouldn't be two SLIP entries at the top and
> the defaultroute option hasn't engaged properly at the bottom.
Nope. Actually, your routing table looks correct.
> I use this command line to set up diald
> diald /dev/modem mode ppp connect '/usr/sbin/chat -f
> /etc/ppp/demon-on.ppp'\lock modem crtscts defaultroute local "static
> IP address" remote "ISP gateway"
> I haven't included the script because that works fine under pppd.
> Has anyone got any pointers as to what I should be trying at this
> point as I'm sure I'm missing something obvious.
Hmmm. Do you have a diald options file? Do you use some kind of filter
file that will determine when diald should actually bring up the link?
Do you include any options in your pppd options file that shouldn't
be there (described, I believe in the diald manpage)?
Again, I can't really diagnose the situation from here, so these possibilities
might not help, but I wish you good luck!
Tim
-
To unsubscribe from this list: send the line "unsubscribe linux-diald" in
the body of a message to [EMAIL PROTECTED]
