Hi!
We have two AlphaServer 1000A running DU 4.0D (all patches) and TC-ASE 1.5.
Both servers have two 10/100 mbit ethernet-interfaces. We use tu0 with 10
mbit,half-duplex as the heart-beat interface, tu1 with 100 mbit,full duplex
as the lan-interface. Both interfaces are configured in the /etc/rc.config
as described in the last mail from Dave Campbell, i.e.
NETDEV_0="tu0"
NETDEV_1="tu1"
If I disconnect the lan-cable of the active node no rearrangment
of the services occur. The former
active node generates error-messages about a network-failure the
other node generates error-messages about NFS problems (the active nodes
servers files from a raid via NFS). But both systems are not able to solve
the problem within max. 15 minutes. I have to reconnect the lan-cable
to solve the problem.
Shutting down the active node or switching off the power makes no
problems. The second node takes all jobs within a few seconds.
What is wrong here??
Thanks in advance!
Wolf
--
Dr. Wolf Dieter Brandt MPI fuer Stroemungsforschung
Tel. : +49 (0)551 5176 435 Bunsenstrasse 10
Fax : +49 (0)551 5176 409 D 37073 Goettingen
EMail: wolf_at_chaos.gwdg.de WWW: http://www.chaos.gwdg.de/
Received on Thu Jun 24 1999 - 08:10:36 NZST