From: VE6TN on Shaw IDC (yajx.bhlmah@relay.tunkki.fi)
Date: Fri Dec 13 2002 - 18:20:04 EET
This makes sense.
My ve6tn-7 cluster has been experiencing much the same internode
communications failures. What I did is implement a script that monitors the
logfile and when it sees a string of @@@@@@@ symbols it executes the clx -f
command to recover. This script is working but isn't really the solution to
the problem.
I have noticed that if I reduce the number of nodes connected, then CLX is
more stable.
Barry VE6TN
----- Original Message -----
From: "Peter" <ecs.dpbjistj@dmzglobal.net>
To: <apkmhqm.fdfsggjvsg@teilar.gr>
Sent: Friday, December 13, 2002 6:10 AM
Subject: inter-node communication
> Hi
>
> I found the problem and have the solution.
> My link partner has about 570 nodes in this configuration. (a dx-spider
> cluster) But CLX seems to have trouble to handle more as 500 nodes. When
I
> accept only the EU nodes the total goes down to 300 Clx can handle this
> amount fine
>
> Ben, have you time to increase the value somewhere in the software?
>
> Peter PB0AIU
> Sysop PI4CC
>
> --
> Internet DX-cluster telnet://dxc.pi4cc.net
> DX-cluster info http://www.pi4cc.nl/dxcluster
>
>
This archive was generated by hypermail 2b30 : Fri Dec 13 2002 - 18:25:07 EET