RE: Re: connection refused|
"Norberto Altalef" <nalt,AT,ub,DOT,edu,DOT,ar>|
Mon, 11 Feb 2002 19:58:16 +0100|
>ciped-cb at the remote site has disappeared, leaving the UDP socket
>unattended (i.e. nonexistent) so that the periodic key exchange
>("kxchg") attempt between local and remote ciped-cb is failing.
>> After that I lose the link with this site.
>Contact with that site has already been lost by the time "kxchg -
>refused" errors are occurring -- the cipcbN interface on the remote
>end cannot continue to exist in the absence of ciped-cb.
>Check /var/log/messages at the remote end to see if ciped-cb explained
>why it was departing.
Thanks for your help Karl.
Really only one of the five remotes sites is lossing the connection.
Al the sites are identical (hardware and software).
The ciped-cb in the remote continues running when the error msg appears in
the central site.
No error messages in the remote site.
I verified during the weekend (no activity in the remote site) and I had no
The first errors began on monday morning. Traffic related ?
I suspect a link problem, but I'm not sure, since in the same times, that I
lose the connection, the link becomes very very slow (ping 3 sec versus
normal 50 msec)
But, the link runs OK if I'm not using the cipe VPN.
May the link have some restrictions to UDP packets ?
Let me point that this site has a different link that the others.
It has a point to point digital line (64 K) and the others microwave links
of higher speeds.
I will thanks any advice to try to solve this problem.
>Set "maxerr -1" in the options file to prevent ciped-cb from exiting
>merely because of temporary loss of connectivity to the far end.
I will try this option, but I think that there is some particular issue with
this remote site.
Many thanks in advance