Misconfigured Routing Tables

The most common cause of connectivity problems across a network is that the
routing tables have not been properly defined. In this scenario, your datagrams are
going out to the remote destination, and datagrams are being sent back to your
system but are taking a bad route on the way to your network. This problem
occurs when the advertised routes for your network point to the wrong router (or
do not point to any router).
This is a very common problem with new or recently changed networks. It is not
at all unusual for somebody to forget to define the route back to your new network.
Just because the datagrams are going out does not mean that return datagrams
are coming back in on the same route.
The only way to successfully diagnose this problem is to use the traceroute program
from both ends of a connection, seeing where in the network path the problem
occurs. If you stop getting responses after the second or third hop on
outbound tests, then it is highly likely that the router at that juncture has an incorrect
routing entry for your network, or doesn’t have any entry at all.