[Veritas-bu] time-out problem after end writing

Hi All

I have a strange time-out error on some Windows 2000 clients, with multiple
NICs, one is a decicated backup LAN

The backup is running fine connecting, writing to tape, after the end
writing message the backup times out and generates a status 41.

I have tried the usually network trouble-shouting, ping, traceroute,
putting the backup server in the host-file on the clients

>From the bpinetd on the client:

01-11-02 15:02:30: [1012]: INF - tcb privileges enabled
01-11-02 15:02:30: [1012]: INF - Timer not created
01-11-02 15:02:30: [1012]: INF - CreateNamedPipe (NetBackup Process
Information)
01-11-02 15:02:30: [1012]: INF - ConnectNamedPipe (NetBackup Process
Information): (WIN32 536: Waiting for a process to open the other end of
the pipe. )
01-11-02 15:02:31: [1012]: INF - CreateNamedPipe (NetBackup Client Request)
01-11-02 15:02:31: [1012]: INF - ConnectNamedPipe (NetBackup Client
Request): (WIN32 536: Waiting for a process to open the other end of the
pipe. )
01-11-02 15:02:31: [1012]: TCP - success: 172.16.35.12 is a local address
01-11-02 15:02:31: [1012]: INF - Starting listen processes
01-11-02 15:02:31: [980]: INF - accepted socket: -1
01-11-02 15:02:31: [980]: ERR - unable to start BPJAVA-msvc process (WIN32
2: The system cannot find the file specified. )
01-11-02 15:02:53: [980]: INF - accepted socket: 552
01-11-02 15:02:54: [980]: INF - BPCD process started
01-11-02 15:03:38: [980]: INF - accepted socket: 568
01-11-02 15:03:38: [980]: INF - BPCD process started

I have the suspicion that the clients is trying reach the backup server
through another NIC than the one on backup LAN

Does anybody know how to get netbackup to show which ip-adress/machine is
expected to be the other end.

Thanks in advance

Michael
CSC