'traceroute' doesn't seem to work !!!!!

[Simon R Wootton]

We are trying to run the traceroute command on a rp7410 running hpux 11i. Even when trying to trace the route to the connected router (which works fine) we just get the :

1 * * *
2 * * * etc

Is something not enabled or switched off within unix ??

All help rewarded !!

Cheers
Simon

[Geoff Wild]

Is the default router set correctly?

Rgds...Geoff

[Simon R Wootton]

Default 'gateway' is set correctly, is the default 'router' a different setting, and if so, where can I find it ??

Thanks
Simon

[Uday_S_Ankolekar]

1.Can you ping to the router?
2. Is the default gateway on rp7410 set? (You can check with netstat -rn command)
3.Does traceroute to self ( rp7410) works?
4. Is ipaddress, subnet are configured correct in the server?

Check all these and let forum know

-USA..

[Jim Butler]

try ping -o to the router

i.e., ping -o router

netstat -nr should show your routes.

If that doesn't work, check your interface

ifconfig lan0 (where you get the lan number from running "lanscan")

Make sure your interface is up, and pingable from the subnet.

route add default routernameoripadd 1

[Geoff Wild]

OK - what is the ip of your gateway, your server, and subnet mask.


Rgds...Geoff

[Elmar P. Kolkman]

Is the network interface which routes the default route the first network interface mentioned when doing a netstat -i ?

There is a bug in traceroute (which has troubled thousands, well hundreds, well many anyway) resulting in it only working on the first network card... and this is not by definition lan0, to make it more complicated.

[Simon R Wootton]

Our connections are fine, the router talks to the rp7410 and we can access the rp7410 from other sites no problem. It's just the traceroute command - I think the bug in traceroute could well be to blame - the first lan card it finds is not the one talking to the router ..... Add us to the list of confused 1000's !

[Jeff Schussele]

Hi Simon,

Ak Elmar alluded to with a certain patch level, HP introduced the need to use -i lanX switch to force the specific I/F to be used. Else it will use the first I/F output in a netstat -in command.

HTH,
Jeff

[Ron Kinner]

Many network admins have a touch of paranoia and have turned off/blocked the ICMP messages which traceroute relies on. If everything else is working then don't worry about it. Just make a note so that if you have routing troubles in the future you will know that traceroute is not going to help solve the problem.

You should still be able to traceroute to other hosts (including your own) on the same network as long as they are not listening on port 33434 (check with netstat -an |grep 33434 and if you find a process listening on the port then change to a different one with traceroute -p 33435 hostnameOrIP for example.)
Best to try with both hostnames and IP addresses to verify that it is not a DNS problem.

Ron