Re: Hop problem


John Lekashman (lekash@orville.nas.nasa.gov)
Thu, 21 Jan 88 09:13:37 PST


        From tcp-ip-RELAY@SRI-NIC.ARPA Thu Jan 21 00:40:29 1988
        Received: Thu, 21 Jan 88 00:40:26 PST by orville.nas.nasa.gov (5.54/1.2)
        Received: Thu, 21 Jan 88 00:41:41 PST by ames-nas.arpa (5.51/1.2)
        Message-Id: <8801210841.AA12538@ames-nas.arpa>
        Received: from CUNYVM.CUNY.EDU by SRI-NIC.ARPA with TCP; Wed 20 Jan 88 14:17:55-PST
        Received: from UNMB.BITNET by CUNYVM.CUNY.EDU ; Wed, 20 Jan 88 17:17:52 EST
        Date: Wed, 20 Jan 88 15:16 MDT
        From: <STGEORGE%UNMB.BITNET@CUNYVM.CUNY.EDU>
        Subject: Hop problem
        To: tcp-ip@sri-nic.arpa
        X-Original-To: tcp-ip@sri-nic.arpa
        Status: RO

        Can anyone tell me the status of thw work being done to get around the
        limitation of TCP/IP to a hop of 15 nodes? Thanks.

Said work is complete, in that the current limit is 255, assuming all
intervening gateways only decrement the time to live by one.
This has always been the state. Perhaps the particular IP
you are running initializes the time to live at 15, assuming
this to be an upper limit. There is also the existence of
an implementation of a routing protocol called RIP, which
assumes a distance of 16 is infinite. Perhaps this is
causing you some problem. All you need do is correct all
machines to initialize the ttl field to be more than 15,
and you will win. I doubt seriously if anyone is trying
actively to repair this as a default case.

                                        john



This archive was generated by hypermail 2.0b3 on Thu Mar 09 2000 - 14:40:41 GMT