I've been using diald-0.16 for some years with no problem. Last
weekend I upgraded to RH 6.0, so I upgraded to diald-0.99.1 while I
was at it. I'm now not able to get demand dialing to work -- moments
after establishing a connection, it announces it is taking down an
idle link. If I force the connection up, it stays up. In what
follows, I have one major and two minor questions:
(major) how can I get my link to come up and stay up?
(minor) what is the purpose of an Ethertap device, and why should we
now be using them? I didn't find the description in
/usr/doc/kernel-doc-2.2.5/networking/ethertap.txt to be much help.
(minor) what is the meaning of the errors relating to SIOCSIFMETRIC?
I haven't been able to find that symbol in the kernel include files,
nor in the source for diald.
Some things I've tried (relative to the diald.conf and logs following) --
I took the defaultroute out of my diald.conf (which got rid of the
duplicate path to my ISP's gateway), and created an addroute shell
script (which was able to create a gateway through the remote IP). I
tried to create a gateway for the proxy device through 0.0.0.0 (as
recommended by the diald man page), but got errors. None of these
ideas fixed the problem.
The rest of this message is information that might be helpful in
finding the problem.
Here's the configuration part of my diald.conf:
device /dev/modem
mode ppp
local 204.134.124.67
remote 204.134.124.254
defaultroute
dynamic
crtscts
#netmask 255.255.255.0
connect /usr/lib/diald/connect
fifo /var/run/diald.ctl
accounting-log /var/log/diald.log
redial-timeout 5
If I'm not running diald, my routes look like this:
[root@noodlenoggin /root]# route
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
204.134.124.223 * 255.255.255.255 UH 0 0 0 eth0
204.134.124.68 * 255.255.255.255 UH 0 0 0 eth0
204.134.124.69 * 255.255.255.255 UH 0 0 0 eth0
204.134.124.67 * 255.255.255.255 UH 0 0 0 eth0
127.0.0.0 * 255.0.0.0 U 0 0 0 lo
(notes: 204.134.124.67 is my gateway machine; 68, 69, and 223 are
machines on my home network).
Once I start diald, the routing looks like this:
[root@noodlenoggin /root]# route
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
204.134.124.223 * 255.255.255.255 UH 0 0 0 eth0
204.134.124.254 * 255.255.255.255 UH 0 0 0 tap0
204.134.124.254 * 255.255.255.255 UH 0 0 0 tap0
204.134.124.68 * 255.255.255.255 UH 0 0 0 eth0
204.134.124.69 * 255.255.255.255 UH 0 0 0 eth0
204.134.124.67 * 255.255.255.255 UH 0 0 0 eth0
127.0.0.0 * 255.0.0.0 U 0 0 0 lo
default * 0.0.0.0 U 0 0 0 tap0
(that's right, 254 shows up twice)
When I try to dial out, the following is logged to /var/log/messages:
Jun 22 22:07:33 noodlenoggin diald[4992]: start tap0: SIOCSIFMETRIC: Operation not
supported
Jun 22 22:08:38 noodlenoggin diald[4992]: Calling site 204.134.124.254
Jun 22 22:08:39 noodlenoggin connect: Initializing Modem
Jun 22 22:08:40 noodlenoggin connect: Dialing system
Jun 22 22:09:10 noodlenoggin connect: Connected
Jun 22 22:09:10 noodlenoggin connect: Logging in
Jun 22 22:09:11 noodlenoggin connect: Starting Comm Protocol
Jun 22 22:09:12 noodlenoggin connect: Protocol started
Jun 22 22:09:12 noodlenoggin diald[4992]: Connected to site 204.134.124.254
Jun 22 22:09:12 noodlenoggin diald[4992]: Running pppd (pid = 5036).
Jun 22 22:09:13 noodlenoggin kernel: CSLIP: code copyright 1989 Regents of the
University of California
Jun 22 22:09:13 noodlenoggin kernel: PPP: version 2.3.3 (demand dialling)
Jun 22 22:09:13 noodlenoggin kernel: PPP line discipline registered.
Jun 22 22:09:13 noodlenoggin kernel: registered device ppp0
Jun 22 22:09:13 noodlenoggin pppd[5036]: pppd 2.3.7 started by root, uid 0
Jun 22 22:09:13 noodlenoggin pppd[5036]: Using interface ppp0
Jun 22 22:09:13 noodlenoggin pppd[5036]: Connect: ppp0 <--> /dev/ttyS3
Jun 22 22:09:16 noodlenoggin kernel: PPP BSD Compression module registered
Jun 22 22:09:17 noodlenoggin kernel: PPP Deflate Compression module registered
Jun 22 22:09:19 noodlenoggin pppd[5036]: local IP address 204.134.124.67
Jun 22 22:09:19 noodlenoggin pppd[5036]: remote IP address 204.134.124.250
Jun 22 22:09:20 noodlenoggin diald[4992]: New addresses: local 204.134.124.67, remote
204.134.124.250, broadcast 0.0.0.0
Jun 22 22:09:20 noodlenoggin diald[4992]: start ppp0: SIOCSIFMETRIC: Operation not
supported
Jun 22 22:09:25 noodlenoggin diald[4992]: Closing down idle link.
Jun 22 22:09:25 noodlenoggin diald[4992]: start tap0: SIOCSIFMETRIC: Operation not
supported
Jun 22 22:09:25 noodlenoggin pppd[5036]: Terminating on signal 2.
Jun 22 22:09:25 noodlenoggin pppd[5036]: Connection terminated.
Jun 22 22:09:25 noodlenoggin pppd[5036]: Connect time 0.2 minutes.
Jun 22 22:09:25 noodlenoggin pppd[5036]: Sent 202 bytes, received 339 bytes.
Jun 22 22:09:25 noodlenoggin pppd[5036]: Exit.
Jun 22 22:09:25 noodlenoggin diald[4992]: start tap0: SIOCSIFMETRIC: Operation not
supported
Jun 22 22:09:25 noodlenoggin diald[4992]: start tap0: SIOCADDRT: File exists
Jun 22 22:09:26 noodlenoggin diald[4992]: start tap0: SIOCADDRT: File exists
Jun 22 22:09:26 noodlenoggin diald[4992]: Disconnected. Call duration 14 seconds.
Jun 22 22:09:26 noodlenoggin diald[4992]: IP transmitted 0 bytes and received 60 bytes.
Jun 22 22:09:27 noodlenoggin diald[4992]: Delaying 5 seconds before clear to dial.
For a few seconds, my routing table looks like this:
[root@noodlenoggin /root]# route
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
204.134.124.250 * 255.255.255.255 UH 0 0 0 ppp0
204.134.124.250 * 255.255.255.255 UH 0 0 0 ppp0
204.134.124.223 * 255.255.255.255 UH 0 0 0 eth0
204.134.124.68 * 255.255.255.255 UH 0 0 0 eth0
204.134.124.69 * 255.255.255.255 UH 0 0 0 eth0
204.134.124.67 * 255.255.255.255 UH 0 0 0 eth0
127.0.0.0 * 255.0.0.0 U 0 0 0 lo
default * 0.0.0.0 U 0 0 0 ppp0
--
Joseph J. Pfeiffer, Jr., Ph.D. Phone -- (505) 646-1605
Department of Computer Science FAX -- (505) 646-1002
New Mexico State University http://www.cs.nmsu.edu/~pfeiffer
-
To unsubscribe from this list: send the line "unsubscribe linux-diald" in
the body of a message to [EMAIL PROTECTED]
