ietf
[Top] [All Lists]

Re: pinging problem

2001-06-10 00:00:03
From: Vijayendra Sharma
To: ietf(_at_)ietf(_dot_)org
Sent: 09-06-2001 13:30
Subject: pinging problem

Hello sir,
   Can you help me with this problem,
 Will machines A and B able to ping eachother,


      A                                B
1)192.168.0.1/16-----Hub------192.168.0.2/24

2)192.168.0.1/16-----Switch------192.168.0.2/24

3)192.168.0.1/16-----Router------192.168.0.2/24

Doesn't it bother anybody else that in cases 1 and 2, the netmask widths
are inconsistent?  That's probably very bad network design, unless you
*really* understand what you're doing.  Yes, we've done things like
that ourselves (for instance, we played *all* sorts of games when we
were subnetting our first /16 into /22's back in '89 - amazing how much
stuff then got upset at a /22 subnet mask on a 'class B' address block ;)

The addresses 192.168.0.1 and 192.168.0.2 should have the same netmask
width - even if you're using a /32 for case (3) as per the recent RFC
regarding the numbering of point-to-point lenghts.

The problem is that if there is a *third* host (for instance 192.168.1.1),
host A will say "this is on the local subnet along with host B, because
it's inside my /16", but host B will say "Host A is inside my /24,
but the host C is off the local subnet because it's outside the /24".
This *will* break things.

                                Valdis Kletnieks
                                Operating Systems Analyst
                                Virginia Tech



<Prev in Thread] Current Thread [Next in Thread>