I am trying to setup routing under windows 2003, but I cant get it working. See the routing table below.

[font="Fixedsys"]
Active Routes:
Network Destination        Netmask          Gateway       Interface  Metric
         0.0.0.0          0.0.0.0      192.168.3.1      192.168.3.5      1
        10.1.0.0      255.255.0.0      10.64.1.129      10.64.6.105      1
       10.64.0.0    255.255.248.0      10.64.6.105      10.64.6.105     20
     10.64.6.105  255.255.255.255        127.0.0.1        127.0.0.1     20
  10.255.255.255  255.255.255.255      10.64.6.105      10.64.6.105     20
       127.0.0.0        255.0.0.0        127.0.0.1        127.0.0.1      1
     192.168.3.0    255.255.255.0      192.168.3.5      192.168.3.5     20
     192.168.3.5  255.255.255.255        127.0.0.1        127.0.0.1     20
   192.168.3.255  255.255.255.255      192.168.3.5      192.168.3.5     20
       224.0.0.0        240.0.0.0      10.64.6.105      10.64.6.105     20
       224.0.0.0        240.0.0.0      192.168.3.5      192.168.3.5     20
 255.255.255.255  255.255.255.255      10.64.6.105      10.64.6.105      1
 255.255.255.255  255.255.255.255      192.168.3.5      192.168.3.5      1
Default Gateway:       192.168.3.1
===========================================================================[/font]

The IP address of the box is 10.64.6.105, The IP address of the router that routes to the 10.1.0.0 network is 10.64.1.129, I used the command

Route add 10.1.0.0 mask 255.255.0.0 10.64.6.129

I am trying to route to the 10.1.0.0 network, and as you can see it kind of looks ok but if I tracert to 10.1.1.1 the first hop just goes * * * Request timed out.

The server can ping the router IP, I can tracert to the router and it works but it wont route for some reason. Does that table look right, or is there another way.

I now have this

Active Routes:
Network Destination        Netmask          Gateway       Interface  Metric
         0.0.0.0          0.0.0.0      192.168.3.1      192.168.3.5     20
        10.1.0.0      255.255.0.0      10.64.1.129      10.64.6.105      1
       10.64.0.0    255.255.248.0      10.64.6.105      10.64.6.105     20
     10.64.6.105  255.255.255.255        127.0.0.1        127.0.0.1     20
  10.255.255.255  255.255.255.255      10.64.6.105      10.64.6.105     20
       127.0.0.0        255.0.0.0        127.0.0.1        127.0.0.1      1
     192.168.3.0    255.255.255.0      192.168.3.5      192.168.3.5     20
     192.168.3.5  255.255.255.255        127.0.0.1        127.0.0.1     20
   192.168.3.255  255.255.255.255      192.168.3.5      192.168.3.5     20
       224.0.0.0        240.0.0.0      10.64.6.105      10.64.6.105     20
       224.0.0.0        240.0.0.0      192.168.3.5      192.168.3.5     20
 255.255.255.255  255.255.255.255      10.64.6.105      10.64.6.105      1
 255.255.255.255  255.255.255.255      192.168.3.5      192.168.3.5      1
Default Gateway:       192.168.3.1
===========================================================================
Persistent Routes:
 Network Address          Netmask  Gateway Address  Metric
        10.1.0.0      255.255.0.0      10.64.1.129       1

C:\Documents and Settings\Administrator>ping 10.64.1.129

Pinging 10.64.1.129 with 32 bytes of data:

Reply from 10.64.1.129: bytes=32 time=5ms TTL=64

Ping statistics for 10.64.1.129:
   Packets: Sent = 1, Received = 1, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
   Minimum = 5ms, Maximum = 5ms, Average = 5ms

As you can see the server can ping that gateway.

Can I just sanity check the IPs?

Box: 10.64.6.105

Router: 10.64.1.129

To get to the 10.1.0.0 network you added a route for that network pointing to 10.64.6.129 for the GW.

Is 6.129 the interface IP that's facing the network that the server is on, and 1.129 the interface facing the other side?

6.105 is the IP address of the server on the same network as the router,

1.129 is the router/gateway IP

the 10.1 network is where I want to connect.

our network subnet is 255.255.248.0 hence the .6.xx and .1.xx being on the same network.

If you do a tracert this is the result.

C:\Documents and Settings\Administrator>tracert 10.1.1.1

Tracing route to 10.1.1.1 over a maximum of 30 hops

 1     *        *        *     Request timed out.

Now, if I try this from a sun solaris box I get this (names removed)

xxxxxx root :: is >traceroute 10.1.1.1
traceroute to 10.1.1.1 (10.1.1.1), 30 hops max, 40 byte packets
1  nnnnnnn (10.64.1.129)  5.728 ms  3.091 ms  3.463 ms
2 ^C

The route on the unix box is the same.

I have also tried it from XP and it doesnt connect either, so it looks like a windows thing.

Got ya. So you're using the 2k3 box to route from 10.64.x.x to 10.1.x.x? Are you pinging/tracerting from another machine then?

Same machine. the windows box is either mis routing, or not routing between the 10.64 network and the 10.1 network.

I am rusty when it comes to windows and routing, especially when it doesnt want to play ball.

If it's the same machine then it should be able to ping the 10.1 network anyway as it must have an interface on that network? And you've added a route pointing to 10.64.1.129, so you must be using another router.

Which device is responsible for the routing between 10.64.x.x and 10.1.x.x? It sounds like you are trying to use the W2k3 box unless 10.64.1.129 has an interface on both networks then it won't route for you.

its all getting confusing,

The 10.1 network is external and a third party suppliers network.

The 10.64 network is our internal lan address.

The router on 10.64.1.129 routes between both these networks, and it works fine with Sun Solaris and is sitting on the 10.64 network.

The server I am wanting to access the 10.1 network is here, and connected to the 10.64 network and has one other NIC on a 192.168.3 network for other purposes. This server can route between the 192.168.3 network and 10.64 network using its own lan cards because its acting as a router between these. What it doesnt seem able to do is to route to the 10.1 network over the 10.64 nic.

OK, makes sense now. Looks correct in that case.. can you be sure there's no firewall intervention anywhere?

Cant see any, I have tried the same route addition on a 2000 server, and an XP workstation and it does not work on these either.

Does it work from anything else other than the Solaris box? Next logical step in my mind is to test on a Windows machine using the Solaris box's IP address however I imagine that's not particularly practical.

iirc, you have to enable routing on the NICs in the w2k3 machine, has that been done? Otherwise maybe the windows firewall is not properly disabled and is dropping packets for you?

Do a 'pathping' - it's like a traceroute but shows you which interface it's using. The default gateway/interface on the 192 address suggests you've got multiple NICs?

Edit - sorry, my bad, I missed your post describing what the 192 address NIC does.

*shrug*

Edited 23 January, 200917 yr by Kaiserb_uk

Right, your problem is that you have a /8 ( 255.0.0.0) subnet mask on the 10.64.x.y network. This means that the server will always look locally for the 10.1.0.0/16 network and never to the router, so no matter what your routing table tells you, the /8 mask always takes precedence.

Is the subnet mask on the 10.64 network correct? If the Solaris machine works, I'd say that it has a different subnet mask on it, more than likely a /16.

Hope this helps Manny

The subnet on the 10.64 network is 255.255.248.0

The subnet on the 10.1 network should be 255.255.0.0,

the 10.1 network is external and the subnet was passed to me by the third party.

the 10.64 subnet is correct, onsite we have various equipment with a range of 10.64.1.1 to 10.64.7.254 (on our local lan).

maybe the 10.1 network should be 255.255.255.0

As long as the subnet of the 10.1 network is 255.192.0.0 or less, it should be fine as it's not interfering with your local subnet.

Should the route be just one hop? Could there be a second hop router that's not configured properly?

Do you have access to the box you want to reach? If so, can you install wireshark or other? I'm wondering if your ping / traffic is getting to that box, but the routing on that box with the multipls NICs is not correct so the return path never makes it back to your box.

Would it be possible to have a simple graphical drawing of the boxes / routers / NICs? It might help get clarity

Quick diagram in MS paint,

The vpn in the diagram works fine as does RRAS, I am using the VPN to access the server at the moment and other servers on our lan.

The subnet on the 10.64 network is 255.255.248.0

The subnet on the 10.1 network should be 255.255.0.0,

the 10.1 network is external and the subnet was passed to me by the third party.

the 10.64 subnet is correct, onsite we have various equipment with a range of 10.64.1.1 to 10.64.7.254 (on our local lan).

maybe the 10.1 network should be 255.255.255.0

oops! Missed that entry on the routing table.

Right, I wonder if the PC is using the 192.168.3.5 interface as a source? Can you add a route to the .129 router for the 192.168.3.5 host and point it at the 10.64.1.129 IP ? this would be a quick way of seeing if this is the case.