Many thanks to the following people:
miker@il.us.swissbank.com (Mike Raffety)
sitongia@ozzel.hao.ucar.edu (Leonard Sitongia)
eckhard@ts.go.dlr.de (Eckhard Rueggeberg)
mel@maths.aberdeen.ac.uk (Andrew Mellanby)
Problem was :
>
> using traceroute I found a HOP throw "localhost" (see below).
> What can be the reason for these "bad" HOP?
> All maschines are SUN-Sparc Server with SunOS 4.1.3.
>
> Example:
>
> > phoenix{af}1/2: traceroute gambler
> > traceroute to gambler (149.204.150.1), 30 hops max, 40 byte packets
> > 1 mythos2 (149.204.28.3) 2 ms 2 ms 2 ms
> > 2 localhost (127.0.0.1) 25 ms 20 ms 20 ms
> > 3 gambler (149.204.150.1) 24 ms 24 ms 23 ms
> > phoenix{af}1/3:
>
>
> I thing the Problem is on the mashine mythos2 - but I can't find it.
>
>
Solution is/could be :
Between the 1. and 2. HOP there is a point-to-point-connection. We are using
sun's software INR-6.0 without seperate TCP/IP-addresses for the ptp.
This is possible with INR-6.0.
At an other ptp-connection we are using own TCP/IP-addresses.
There I can see one of the ptp-TCP/IP-addresses if I use traceroute
to a host on the other side. Every other (OS, architecture, netmask,
route-daemon, etc.) is the same
So, only different is:
- one ptp with own TCP/IP-addresses ---> "correct" traceroute
- one ptp without own TCP/IP-addresses ---> "bad" traceroute
Finaly, I think the problem results from the INR-6.0 software.
Kind Regards,
Andreas Fenner
System Administrator
'''
(o o)
/-----oOO--(_)--OOo----------------------------------------------------------\
| |
| Dipl.Ing. Tel.: +49 - 711 / 869 - 6093 |
| Andreas Fenner Fax.: +49 - 711 / 869 - 6049 |
| Alcatel-SEL |
| PS/ETD-4 |
| Motorstrasse 55 e-mail : af@bk.sel.de |
| D-70499 Stuttgart uucp : ...!neckar!af |
| Germany suninfo : ...!suninfo!neckar!af |
| DECNet : 61.137::af |
\----------------------------------------------------------------------------/
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:08:31 CDT