Check that nmbd is running:

ps -eaf | grep nmbd

result:

root       3422      1  0 10:30 ?        00:00:00 /usr/sbin/nmbd -D
root       5178   4327  0 11:14 pts/0    00:00:00 grep --color=auto nmbd

If nmbd does not appear in the list reported by ps, it should be started as root using the normal means (/usr/local/samba/bin/nmbd -D).

Samba servers have a special NetBIOS name, __SAMBA__, to which they will always respond.

By querying the server for this name, we can verify that nmbd is working correctly. The -U flag is used to specify the address to which the query should be sent.

nmblookup -U 127.0.0.1 __SAMBA__

result:

querying __SAMBA__ on 127.0.0.1
192.168.0.2 __SAMBA__<00>
192.168.1.2 __SAMBA__<00>

If nmbd is not running, the following error message will be shown:

name_query failed to find name __SAMBA__

This can also be caused by not including the loopback interface name for the interfaces parameter in smb.conf and setting bind interfaces only = yes

nmblookup -U 127.0.0.1 ShareWiz

result:

querying ShareWiz on 127.0.0.1
192.168.0.2 ShareWiz<00>
192.168.1.2 ShareWiz<00>

If an domain was not found by nmbd the result would be:

querying xyz on 127.0.0.1
name_query failed to find name xyz

A possible reason for a failure is that the server was unable to register its NetBIOS name.

If this is the case, you can locate the host that currently owns the name by sending a name query request to the broadcast address of the local subnet.

nmblookup -B 192.168.1.255 ShareWiz

result:

querying ShareWiz on 192.168.1.255
192.168.1.3 ShareWiz<00>

In this example, the name ShareWiz has been registered by a host at address 192.168.1.3 and not our Samba server.

To correct this problem is to rename the rogue machine.

If it failed the result would be:

querying ShareWiz on 192.168.1.255
name_query failed to find name ShareWiz