From: Kenn Humborg (kenn at domain bluetree.ie)
Date: Wed 09 Feb 2000 - 13:15:28 GMT
> I'm having a strange problem which I hope someone can finger. I've just
> upgraded a box used as an ISDN gateway to SuSE 6.3 and rsh / ssh
> connections from it to internal / external boxes break down with a message
> rlogin: read: Connection reset by peer.
> The ssh breakdowns have the same message, minus the rlogin. It seems that
> they happen at the same time as rsh breakdowns. However, that has only
> happened twice and each time I flipped from the VC with the dead ssh to
> that with the rsh, the rsh was dead also. I suspect that the same thing
> causes both errors, but what the **** is it ?
Make absolutely sure that you don't have another machine
with the same IP address. When this happens, all hell
I've seen this: A and B have same IP address. I'm on
machine C. I telnet to A. My machine sends out an
ARP request, A answers (by luck) and the TCP connection
is opened. After a while, the ARP entry expires and
is re-requested. B answers this time. The next TCP
packet from C goes to B. B says "Hey! I don't have an
open connection from C with that source-port/dest-port
pair" and sends back a connection reset.
This archive was generated by hypermail 2.1.6 : Thu 06 Feb 2003 - 13:05:23 GMT