More diald problems.
Just to continue with the problems I'm having with Diald - I'd appreciate
any suggestions/comments about what I've provided below. TIA.
Note: I have "debug 31" enabled for diald.
When I try to ping a remote site, diald "cranks up" the connection to my
ISP; however, the ping never received a response. Monitoring
/var/log/messages shows diald accepting the ping packets. After about 30
seconds, diald drops the connection.
After troubleshooting this, I figured out why the ping wouldn't work once
the connection came up. There was no route to the default gateway (my
ISP's interface). So, I added "route add default gw $5" to /etc/ppp/ip-up
and now the ping works fine. However, diald still drops the connection
after 30 seconds even though it's still accepting packets coming from the
ping command. So, after a default timeout, diald again re-establishes the
connection to my ISP and ping starts receiving responses for another 30
seconds until diald drops the connection again. This will continue in an
endless cycle.
I've left /etc/diald/diald.conf alone so it's configured exactly as it came
in the .deb distribution.
In the diald man page, under "addroute", something was mentioned about how
newer kernels will have problems if routes aren't configured properly for
the sl0 and ppp0 interfaces before and after the connections are
established. Could this be my problem? I've gone over my routing tables
and things look okay. I've included them below FYI.
Before diald initiates the connection to my ISP, "route -n" provides:
Destination Gateway Mask Flags Metric Interface
127.0.0.3 0.0.0.0 255.255.255.255 UH 2000 sl0
x.y.z.128 0.0.0.0 255.255.255.224 U 0 eth0
127.0.0.0 0.0.0.0 255.0.0.0 U 0 l0
0.0.0.0 127.0.0.3 0.0.0.0 UG 2000 sl0
>From the docs, everything appears to be configured properly. Note, a dest
of 0.0.0.0 equals "default". Also, note that x.y.z.128 defines my network
segment which is a subnet of a class C address. Thus the .128 net address
and the .224 subnet mask.
After diald initiates the connection to the ISP, "route -n" provides:
Destination Gateway Mask Flags Metric Interface
127.0.0.3 0.0.0.0 255.255.255.255 UH 2000 sl0
a.b.c.d 0.0.0.0 255.255.255.255 UH 2000 ppp0
x.y.z.128 0.0.0.0 255.255.255.224 U 0 eth0
127.0.0.0 0.0.0.0 255.0.0.0 U 0 l0
0.0.0.0 127.0.0.3 0.0.0.0 UG 2000 sl0
Note, this is the "broken" routing table as there is no default gateway to
my ISP's interface. So, I've had to manually establish one using
"/etc/ppp/ip-up" (as described above). Once I've done this, routing works
fine and the table is as follows:
Destination Gateway Mask Flags Metric Interface
127.0.0.3 0.0.0.0 255.255.255.255 UH 2000 sl0
a.b.c.d 0.0.0.0 255.255.255.255 UH 2000 ppp0
x.y.z.128 0.0.0.0 255.255.255.224 U 0 eth0
127.0.0.0 0.0.0.0 255.0.0.0 U 0 l0
0.0.0.0 a.b.c.d 0.0.0.0 UG 0 ppp0
0.0.0.0 127.0.0.3 0.0.0.0 UG 2000 sl0
So, finally, my questions are:
1. Why does diald drop my connection 30 seconds after establishing it,
even though there are packets being sent across that would normally keep it
alive. (Note - I can do anything across the connection once established,
but no matter what (ftp, telnet, etc.) diald still kills the connection.)
2. From the docs, I thought diald would establish the default route to the
gateway automatically. Am I wrong in my assumption? If so, is what I did
to "fix" the problem the "right" way to go about it?
TIA for your help,
Kevin Traas
--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
debian-user-REQUEST@lists.debian.org . Trouble? e-mail to Bruce@Pixar.com
Reply to: