Pal Dahle wrote:
> I have trouble with diald! I've tried to collect all relevant
> info in this
> e-mail. I hope someone of you have time to help me. Please tell me if
> there's anything more I should supply.
[snip]
> Redhat 6.1
> Linux kernel 2.2.12-20
> pppd version 2.3.10-3
You've kept up with the official RedHat updates. You are behind on released
versions though.
> diald version 0.99.1-2 installed using Redhat Package Manager
Consider upgrading to 0.99.4 if you have any other problems (most likely
problem, if diald stops dialing out if ISP hangs up on you).
> My network address is 192.168.1.1
> I have IP-masquerading active
>
> /etc/diald.conf
> ---------------------------------------
> debug 31
> mode ppp
> pppd-options user <my isp user-name>
pppd-options noauth user <my isp user-name>
> connect "/usr/sbin/chat -v -f /etc/ppp/chatf"
> device /dev/modem
> speed 115200
> modem
> lock
> crtscts
> local 192.168.0.1
> remote 192.168.0.2
> dynamic
> defaultroute
> include /usr/lib/diald/standard.filter
> ---------------------------------------
[snip]
> Mar 15 20:57:46 grieg diald[673]: keepup parsing error. Got
> token 'tcp.ssl'.
> Not a known tcp service port.
/usr/lib/diald/standard.filter assumes port 443 is named 'ssl'
/etc/services provided by RedHat names it 'https'
Either change the tcp.ssl lines in standard.filter to read tcp.https (my
preference) OR add an additional line to /etc/services using the name ssl
(ssl 443/tcp).
Note: RedHat likes alias names to be listed after the port definition in
/etc/services (see 'man services' for info). Diald ignores anything to the
right of the port definition so aliases must be listed on separate lines,
this matches format of the IANA port number listings (which use the name
https for port 443).
[snip]
> Mar 15 20:57:47 grieg modprobe: can't locate module tap0
Your kernel supports ethertap devices so diald tries to use them (in
preference to slip).
(Assuming the rpm was created by someone who knew what they were doing)
read the explanation in /usr/doc/diald-0.99.1/README.ethertap
In your case ethertap was compiled as a module.
add to /etc/conf.modules
--
alias tap0 ethertap
options tap0 -o tap0 unit=0
alias tap1 ethertap
options tap1 -o tap1 unit=1
# repeat as above for 2-14
alias tap15 ethertap
options tap15 -o tap15 unit=15
--
Actually the first one will probably be all you'll ever need. There are a
max of 16 tap devices tap0-tap15
[snip]
> Mar 15 20:58:02 grieg diald[674]: Trigger: udp 192.168.0.1/1024
> 202.12.27.33/53
>
> # what is happening here? something triggers diald! What is it?
Domain name lookup from 192.168.0.1.
[snip]
> Mar 15 20:58:27 grieg modprobe: can't locate module char-major-108
Bogus error message generated by pppd 2.3.10, it's corrected in 2.3.11.
[char-major-108 exists in kernels 2.3+]
> Mar 15 20:58:27 grieg pppd[695]: The remote system is required to
> authenticate itself but I
You need to use the 'noauth' pppd option IF there is an existing default
route when pppd starts.
When you dial in normally there is not a default route (so no problem).
When diald manages the call there is a default route (to the diald proxy
device) so you'll see this error unless you use the 'noauth' option. I
added it to your pppd-options line in diald.conf above.
Effectively what is happening is that you are asking the ISP server to log
onto your machine. 'noauth' says it doesn't have to.
Hope this helps,
Lourdes
-
To unsubscribe from this list: send the line "unsubscribe linux-diald" in
the body of a message to [EMAIL PROTECTED]
