This section contains information on common troubleshooting issues.
If the Primary Domain Controller (PDC) is a Windows 2000 server, Windows workstations might not be able to access the CIFS server.
To fix this using iManager:
In a Web browser, enter the following in the address (URL) field:
At the login prompt, enter the server administrator username and password.
In the left frame, click> .
Type the NetWare server name where CIFS is running, or browse and select one.
Click, then click the tab.
Ensure that the Mode is set to Third Party Domain and that the Group Name is set correctly.
Specify the name and IP address for the Primary Domain Controller, then click.
Enter CIFSSTOP and then CIFSSTRT at the server console.
If the PDC and CIFS servers are on different subnets, you must specify an IP address for the PDC on the CIFS Config property page for the Server object.
Trying to change a password from a Windows 9.x or NT4 client workstation might fail if the default network username and password in the client's cache are not a valid combination on the CIFS server where the password change was attempted.
To avoid this problem, validate the username and password combination in the client's cache on the CIFS server before attempting the password change. You can validate by attempting to access the file system on the CIFS server by browsing to Network Neighborhood.
You might occasionally find that your CIFS server is not visible using Network Neighborhood. This can happen if you have a domain that contains only Windows 95/98 and no NT/2000/XP servers or clients. To correct this problem, enter the server name or IP address in Find Computer.
Native File Access for Windows is supported only on NSS volumes. Traditional volumes are not available to Windows users.
If you have CIFS configured to run in ACTIVE/ACTIVE mode and you stop and then restart CIFS on a server, you must bring the virtual server resource offline and then online again to cause the CIFS ADD command to be executed from the load script. This is required for the cluster to be aware that CIFS has been restarted.
If the CIFS virtual server (cluster-enabled volume) goes into a comatose state when starting, it could indicate that the CIFS ADD command fails to execute in the virtual server load script. This problem is common if servers running CIFS do not have trustee rights to the CIFS virtual server.
Ensure servers running CIFS have trustee rights to the CIFS Virtual Server object.