Why does my router crash when IPZ routing is done with rather large, fragmented updates?
When IPX DEMAND RIP is used with updates from the remote device that are large and fragmented, it causes the packets to be queued before they can be re-assembled and processed, which fills up the receiving queue. If you have such a situation, it is best to use IPX static routes, or use the default IPX gateway and IPX SAP Helper features as opposed to DEMAND RIP updates.