Xref: utzoo comp.protocols.tcp-ip:13286 alt.sys.sun:1616 Newsgroups: comp.protocols.tcp-ip,alt.sys.sun,jpl.sun-nets Path: utzoo!utgpu!freedom!elroy.jpl.nasa.gov!news From: cyamamot@kilroy.jpl.nasa.gov (Cliff Yamamoto) Subject: Our Sun isn't "RIPing" Message-ID: <1990Oct7.000448.26450@elroy.jpl.nasa.gov> Originator: cyamamot@kilroy Keywords: RIP, router, passive Sender: news@elroy.jpl.nasa.gov (Usenet) Nntp-Posting-Host: kilroy.jpl.nasa.gov Organization: Jet Propulsion Laboratory, Pasadena, CA Date: Sun, 7 Oct 90 00:04:48 GMT Greetings network gurus! We have a Sun Sparc+ running as a file server and router for our building's LAN. We also have a Banyan server running as passive router on the same physical ethernet. On this same ethernet, we have various PCs. Some are running PC-NFS while others are running PC/TCP on top of Vines. A picture is probably in order: PC . . . PC . . . PC . . . etc Banyan | | | +--Server--+ | | | | | x.x.63.1 +-----------+-----------+---------------+----------+-----------+ x.x.63.10 x.x.91.20 x.x.91.21 x.x.63.107 x.x.91.1 | Sun Sparc+ where: x.x.63.x is Tcp/Ip & NFS | x.x.91.x is Banyan/Vines x.x.1.137 x.x.1.xxx | x.x.1.xx is Labwide backbone ----------------------------+ | | x.x.1.165 Sun other A hosts & subnets The situation: - We have /etc/gateways on the Sun Sparc+ configured as required. - All the PCs on the 63.x and 91.x subnets can ping each other fine. - All the PCs on the 63.x can ping the x.x.1.xxx hosts and vice-versa. - The Sun Sparc+ can ping all hosts on all subnets. The problem: - NONE of the 91.x hosts can ping the x.x.1.xxx hosts and vice-versa. The probable cause? - The Sun Sparc+ is not sending out RIP info to the hosts on x.x.1.xxx about the 91.x subnet. The test: - We added an explicit /etc/gateway routing entry at Sun A. It basically tells Sun A that all x.x.91.0 traffic should route to x.x.1.165 with a metric of 1 and passive. THIS WORKED!! Sun A was able to ping all PCs on the 91.x subnet and vice-versa. The question: So does anyone know what causes this? We don't have access to a sniffer so it makes it difficult to see if the Sparc+ is sending out info. Is there something we missed? Any help would be appreciated. Thanks in advance! Cliff Yamamoto