default route don't work after couple of minutes

[Nadeem Rahman]

I have added a default route:
route add default 10.2.30.1 1
it works good for a minute or two after that it give me message destination unreachable.

If I delete and add the route again, it will work good just for a minute or two.

Please see the attachment for more information.

Can anybody give me some advice.

Thanks in advance

[Rita C Workman]

My first question would be is that 10.#.#.# a Class A or is it (and this is my guess) a private IP, that could be affected by DHCP assigns.

You want to have standard and static IP's assigned to hubs/bridges/routers, with something other than a 10.#.#.# number.

/rcw

[Denis Gass]

is 10.2.30.70 your host address?
and in this this case, why do you route this adress
onto lan0 instead of l0?
you have
10.2.30.70/255.255.255.255
10.2.30.70 UH 0 432880 lan0 4136

you should have
10.2.30.70/255.255.255.255
127.0.0.1 UH 0 432880 l0 4136

test and traceroute your router
--
Denis

[Nadeem Rahman]

Thanks for the quick reply, yes this is a private IP Addr and it is assigned by a DHCP server.

10.2.30.1 is a static gateway address.

Thanks

[Berlene Herren · Accepted solution]

Why is your route mask 0.0.0.0?

Is this 11.X? If so, you may have a dead gateway, if this is connecting to a firewall that has ICMP turned off.

Run this:

#ndd -get /dev/ip ip_ire_status | grep -e IRE_GATEWAY -e flag

and see if the output for the default route says "IRE_GATEWAY DEAD". If so, you've got the dead gateway problem. This typically means that the default gateway is a firewall that refuses to answer ICMP echo requests.

Eliminate the problem by using ndd to turn off dead gateway detection. The following command checks the current dead gateway detection setting:

#ndd -get /dev/ip ip_ire_gw_probe

If this shows a 1, you have detection on, with 0 representing detection turned off.

3. Turn off the dead gateway probe by using

#ndd -set /dev/ip ip_ire_gw_probe 0

4. Reset the existing dead gateway by deleting the default route, then adding it back again. Use the following two commands:

route delete default 1.2.3.4
route add default 1.2.3.4 1

where "1.2.3.4" represents your gateway's IP address, and the final 1 in the add command represents that you have one hop to get to your gateway.

To keep this off after a reboot, edit
/etc/rc.config.d/nddconf and add the following lines:

TRANSPORT_NAME[0]=ip
NDD_NAME[0]=ip_ire_gw_probe
NDD_VALUE[0]=0

Berlene

[Denis Gass]

Hey, let the wizards be with you!
Could it be the dhcp server broadcasting another default
route?

[Volker Borowski]

This could be a lot, depending on the network structure.
Just let the default route "know" a "cheaper" route to your target, you will receive an icmp-redirect at first, and may be the other router is blocking you.
Is this machine running RIP or anything like this (gated or so).

I would proceed this way:
install route;ping target (and watch your clock)
if it works at first, and is gone after x times 60 sec, something is RIPing (Default timeout is one minute).
When it does not work any more, stop ping and traceroute immediately, to see if it is still using the gateway.
ping the gateway
I'll bet somebody is telling your machine to go somewhere else.
Do the netstat -r when it stops working, to see if you get M- or D- flags.

But I do not know if all this helps
Good hunting
Volker

[Nadeem Rahman]

Thanks to everybody my problem is resolved. Special thanks to Berlene Herren, it was a dead gateway problem.