Odd routing problem on work network

[Confused]

Hey guys, this has me stumped, so maybe you can shed some light

We have one IP address (10.1.1.35) which, even when set on different computers, is unable to reach 10.1.3.10, our Alpha server for our Hospital Administration database. The two subnets are at two physically different locations, and are connected by a Cisco 2650 at either location, with a direct kilostream serial link between the two.

So it's:

10.1.1.35 - Switch - (Ethernet 10.1.1.1) Cisco 2650 (Serial 192.168.20.253) - WAN Link - (Serial 192.168.20.254) Cisco 2650 (Ethernet 10.1.3.1) - Switch - 10.1.3.10

From 10.1.1.35, I can connect to any other IP address on the 10.1.3.0 network, other than 10.1.3.10.

From any other address on the 10.1.1.0 network, I can connect to 10.1.3.10 just fine.

All physical links are fine, as if one of them was not working, then it would be causing chaos, as no one could connect to the administration system and the hospital would grind to a halt, and my phone would be ringing off the hook!

In doing a Tracert from any machine which has the IP address 10.1.1.35 to 10.1.3.10 gets the following:

1 1 ms 1 ms 1 ms 10.1.1.1
2 xx ms xx ms xx ms 192.168.20.254
3 * * * Request timed out.
4 * * * Request timed out.

etc til the tracert finishes.

From my machine (10.1.1.61) a tracert to 10.1.3.10 gets:

1 1 ms 1 ms 1 ms 10.1.1.1
2 xx ms xx ms xx ms 192.168.20.254
3 xx ms xx ms xx ms 10.1.3.10

(times are xx because of the current traffic on the WAN link, it changes)


My suspicion is that there is something set in the Router on the 10.1.3.0 network that is sending traffic from 10.1.1.35 off to some IP address that doesn't exist, rather than the right one, or some other weird thing to do with the router.

I have now set 3 computers up with the address 10.1.1.35, and all of them do the same, so it's not a HOSTS or computer error on this end.


Any thoughts/help/advice is much appreciated


Garry

 

[Confused]

We've almost definitely decided to try restarting the remote router tonight (when no one is using the system) if no one here has any ideas, so suggest anything other than "restart the router"


Garry

 

[subflava]

Your problem is so specific that it makes one think it's an ACL-type of problem. Basically, any combination of computers and IP addresses can reach the other network except that particular combination 10.1.1.35<--->10.1.3.10? That would rule out any subnet masking errors or route errors (I'm assuming you don't have any routes more specific than /24 on both routers).

If it's not an ACL on one of the routers or the Alpha server, then that really is an "odd" problem. Sorry, but not much help here.

 

[nightowl]

I was thinking the same thing about an ACL being in place that is not known about. Other than that I am not sure what might be going on. It looks like it might be on the router that is connecting to the 10.1.3.0 network since it hits that router, possibly for outbound traffic on the ethernet interface.