jet@karazm.math.uh.edu (J. Eric Townsend) (10/08/90)
Well, not random, but unpredictable. It's only happened a couple of times. Platform: Sun 4/60C, SunOS 4.0.3c 2 CDC Wren drives 1 Sun cart tape unit 1 GigaTrend DAT unit 2 Telebit T1000 modems Here's the message as it appeared in /var/adm/messages: Sep 6 15:08:45 karazm vmunix: Setting subnet mask to 0xfffff000 The correct mask for us is "0xffff0000". I checked le0 with ifconfig, and sure enough, it had been changed. So I put it back. :-) Any ideas on what causes this and how I can stop it? J. Eric Townsend -- University of Houston Dept. of Mathematics (713) 749-2120 Internet: jet@uh.edu Bitnet: jet@UHOU Skate UNIX(r)
jhd@maths.bath.ac.uk (10/27/90)
In article <1990Oct8.002204.13482@rice.edu> jet@karazm.math.uh.edu (J. Eric Townsend) writes: >Sep 6 15:08:45 karazm vmunix: Setting subnet mask to 0xfffff000 > >The correct mask for us is "0xffff0000". I checked le0 with ifconfig, and >sure enough, it had been changed. So I put it back. :-) Any ideas on what >causes this and how I can stop it? What causes it is that some other machine on your network broadcasts an ICMP address mask reply message, and your ever-trusting Sun just believes it. The solution is to find out WHO is doing this broadcast. Needless to say, this is easier said than done. Getting hold of tcpdump and running it with an appropriate filter (for broadcasts) would do it, but it's a tedious process. The alternative solution is hunt around your physical network for some machine that has this incorrect netmask. Good luck! James Davenport jhd@maths.bath.ac.uk