Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!husc6!cmcl2!rutgers!ucla-cs!zen!ucbvax!TWG.ARPA!jerry
From: jerry@TWG.ARPA ("Jerry Scott")
Newsgroups: comp.protocols.tcp-ip
Subject: RE: SUN 3.4 problems
Message-ID: <8708290110.AA13873@ucbvax.Berkeley.EDU>
Date: Fri, 28-Aug-87 20:51:00 EDT
Article-I.D.: ucbvax.8708290110.AA13873
Posted: Fri Aug 28 20:51:00 1987
Date-Received: Sun, 30-Aug-87 06:20:03 EDT
Sender: daemon@ucbvax.BERKELEY.EDU
Reply-To: "Jerry Scott" <jerry@twg.arpa>
Organization: The ARPA Internet
Lines: 45

I think your problem has to do with ICMP Mask Requests from your Sun 3.4.
In release 3.3 and 3.4, diskless suns starting asking the net for
network masks.  This is all well and good, but when you have hosts
on your net giving you the wrong answer...

There is a bug in 4.3BSD and (sorry to say) Wollongong's VAX/VMS release
3.0 networks that causes them to respond to ICMP Mask Requests with
the wrong answer (actually the answer is right but is not byte swapped
into network order before sent out onto the cable).

Anyway the fix for 4.3BSD is to modify the file netinet/ip_icmp.c
in routine icmp_input as follows:

CHANGE:
	case ICMP_MASKREQ:
		if (icmplen < ICMP_MASKLEN || (ia = ifptoia(ifp)) == 0)
			break;
		icp->icmp_type = ICMP_MASKREPLY;
		icp->icmp_mask = ia->ia_netmask;

TO:
	case ICMP_MASKREQ:
		if (icmplen < ICMP_MASKLEN || (ia = ifptoia(ifp)) == 0)
			break;
		icp->icmp_type = ICMP_MASKREPLY;
		icp->icmp_mask = htonl(ia->ia_netmask);

To fix your Wollongong host, use anonymous ftp from another Wollongong
VMS machine, set ftp transfer type to BINARY and transfer structure to
RECORD, and get file ip_icmp.o.

By the way, there is another work-around if you are familiar with
dealing with disk partitions for client hosts on your sun servers.
Get into the root partitions for the diskless hosts and modify
the /etc/rc.boot files to issue the "ifconfig" command properly,
for example:
	ifconfig ie0 192.5.36.4 ...  netmask 255.255.255.0

Hope this helps, I think this was discussed earlier but your message
shows the symptom from the sun's point of view.  I hope Sun support
is taking note of this.

Jerry Scott

------