why a subinterface on cisco router

This is standard practice when creating frame-relay connections.
Yes, it is technically possible to use just the primary interface, but for ease of scalability, we generally start the first PVC on a sub-interface. If you ever need to add a 2nd PVC, just add another sub-if.
General practice is to use sub-if for all point-point PVC's and primary interface on point-multipoint frame-relay connections (rare any more).
"Because the wizard prompted me" is not a good answer and I would be wary of hiring this installer again.

The route statements and the serial configuration look suspicous to me.
You only have 2 endpoints on a point-point link. Your's and the other end. Your route statement show 3 static routes to 3 different endpoints at the other end of the PVC - strange indeed.
We also do not use /24 subnets on point-to-point FR PVC's. Typically it is a /30 subnet since there can only be two nodes.
No default route configured?  Does this configuration even work at all?

Good catch on the static routes. I suspect the head office has multiple PVCs and the tech created a static route to the LAN at each of the other sites. But a far better solution is a single default route, and let the head office router figure it out.

hello irmoore,

this configuration works the rosignol suboffice and head office are able

to comminucate okay

>> Your route statement show 3 static routes to 3 different endpoints at the other end of the PVC - strange indeed.

could elaborate on th statement

the static routes was included to allow packs to the other destinations.

the are four other sub office in our frame-relay pool

and head office is configure to talk to each .

the rosignol suboffice can quickly be allow to connect to the other suboffice

by asking the local telephone company to build the resp PVC to the desirable

suboffice.

hope the above can be follow if not i will map out the inter suboffice communication.

thanks

hello Irmoore,

Correction to earlier post

>> the are four other sub office in our frame-relay pool

It should read the are three suboffice and head office in our frame-relay

pool a TOTAL of four locations.




The communication pool consists of head office and three other suboffice.

The head office router has static routes for all suboffice.

For each location the local telephone office build a VC as follows:-

head office to new amsterdam

head office to linden

head off to rosignol

Our database and other server are all house at head office, hence

all traffic is directed to head office to retrieve resources.

We have have included static routes in all the suboffice routers mapped to

the other suboffice .

Example the rosignol sub office ( refer the running-config printout )

there are static mapping

 for new amsterdam                208.130.2.0 255.255.255.0 208.130.4.2
              inden                         208.130.3.0 255.255.255.0 208.130.4.3
              head office                 208.130.1.0 255.255.255.0 208.130.4.1

new amsterdam sub office static  mappings will be
 
                                        linden          208.130.3.0  255.255.255.0 208.130.4.3
                                        head office  208.130.1.0  255.255.255.0 208.130.4.1
                                        rosignol       208.130.5.0  255.255.255.0 208.130.4.4

and the linden suboffice  follows the same pattern.

In the current setup if we place another database  server in suboffice (eg new amsterdam)

and clients  from linden suboffice wants to access resources the traffic is routed

thru head office router  to the server at new amsterdam.

If for some reason head office router is out of commission the clients will

not be able to access info from the new amsterdam office,

because there are no V.C from new amsterdam to any two suboffice

The V.C exist between HEAD OFFICE AND NEW AMSTERDAM.

All that is now need is for the local telephone company build a V.C

from linden to new amsterdam and since all the suboffice router has knowledge

each subnet traffic can now be routed thru linden router.

The above may sound confusing and may be look awkward to configure

but we did this simulation before and it work okay.

Any suggestion on a more simple configuration?

perhaps by appling subinterface configuration might prove less confusing.

Regards

Jomo