>>>>> On Sat, 23 Feb 2002 08:46:16 +1100, >>>>> Edwin Groothuis <[EMAIL PROTECTED]> said:
>> > And the interface configuration: >> > gif0: flags=8051<UP,POINTOPOINT,RUNNING,MULTICAST> mtu 1280 >> > tunnel inet 203.173.130.126 --> 206.123.31.114 >> > inet6 fe80::250:8bff:feb9:2d24%gif0 prefixlen 64 scopeid 0x9 >> > inet6 3ffe:b80:2:460::2 --> 3ffe:b80:2:460::1 prefixlen 128 >> >> Hmm, and what command did you type to cause this problem? If possible, >> please give me the network topology as well. > The problem was when "route add -inet6 default -interface gif0" was > used instead of the "route ... <ipaddress>". For the rest the > configuration and the commands were the same. > This is a layout of the 'network': (snip) > The symptons were that if I setup a TCP-session, the original message > is at: > >http://www.freebsd.org/cgi/getmsg.cgi?fetch=279914+284924+/usr/local/www/db/text/2002/freebsd-hackers/20020217.freebsd-hackers Okay, I understand the configuration. I did a similar test on a 4.5-RELEASE box, but could not reproduce the problem. The main difference in my test case is I did not use ppp, so I guess ppp is somehow related to this problem... Now I'd like to know: - did you see the same problem in an environment without ppp? - what is the result of netstat -rnal when the kernel is sending neighbor solicitations? - I guess some process (or kernel) modifies the route to the destination. Could you run 'route -n monitor' while the session to see the modification? Thanks, JINMEI, Tatuya Communication Platform Lab. Corporate R&D Center, Toshiba Corp. [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-net" in the body of the message
