Rick wrote :
>
> Office lan: 10.64.216.0/26
> Office Host: 10.64.216.55/26
> Gateway - office side (eth1): 10.64.216.62/26
> - lab side (eth3): 10.64.216.65/26
> Lab Cisco router - 10.64.216.66
> Lab subnets: 10.64.216.64/26
> 10.64.216.128/26
> 10.64.216.192/26
> 10.64.217.0/24
> 10.64.218.0/24
>
> The routing table on the gateway looks like this:
(I rewrote the routes so they are more readable and removed the
irrelevant loopback and link-local routes)
10.64.216.0/26 dev eth1
10.64.216.64/26 dev eth3
10.64.218.0/24 gw 10.64.216.66 dev eth3
10.64.217.0/24 gw 10.64.216.66 dev eth3
default gw 10.64.216.1 dev eth1
What is 10.64.216.1 ?
> From the office host we can reach any host on the 10.64.217.0/24 and
> 10.64.218.0/24 subnets via the lab router implying the gateway is
> routing to 10.64.216.66.
>
> Our problem is that we can't work out how to route traffic directly to
> 10.64.216.64/26 subnet from the office host.We can't contact either
> the lab Cisco router interface or any other hosts in the subnet.
What happens exactly ? Any "destination unreachable" error messages ?
Have you traced the packets on each expected traversed interface ?
What do the routing tables look like on the office host 10.64.216.55,
on the lab hosts in the 10.64.216.64/26 subnet and on the lab Cisco router ?
> On the surface I would have thought we didn't need a specific route as
> the network is directly attached
Do you mean a route to 10.64.216.64/26 ? Yes you need one, either on the
office host itself or on its default gateway. If its default gateway is the
lab-office gateway, the route is already present. Else, if its default
gateway is another router or it has no default gateway, you need to add the
route on the host itself or on its default gateway, saying that the gateway
for this destination is 10.64.216.62.
|