Hi,
xinetd users have reported that the nmbd daemon does not run from
xinetd.
The error that shows up is this:
Opening sockets 137
bind succeeded on port 138
open_sockets: Broadcast sockets opened.
Netmask for eth0 = 255.255.255.128
Derived broadcast address 130.151.17.255
Added interface ip=130.151.17.230 bcast=130.151.17.255
nmask=255.255.255.128
bind failed on port 137 socket_addr=130.151.17.230 (Address already in use)
make_subnet: Failed to open nmb socket on interface 130.151.17.230 for
port 137. Error was Address already in use
ERROR: Failed when creating subnet lists. Exiting.
One user reported that adding "flags=REUSE" to the netbios-ns entry in
/etc/xinetd.conf "solves" the problem:
service netbios-ns
{
socket_type = dgram
protocol = udp
wait = yes
user = root
server = /usr/sbin/nmbd
server_args = -a
flags = REUSE
}
As you see, the entry for this service in xinetd.conf is pretty much
the same as in inetd.conf. However, the behavior is different with
inetd than with xinetd: when run from inetd, when nmbd receives a
SIGTERM signal it finishes and inetd does not fork another nmbd until
it sees traffic on UDP port 137. The problem with xinetd is that it
forks another nmbd right after the running nmbd finishes, without
waiting for traffic on UDP port 137.
We (the Debian maintainers of the xinetd and samba packages) are
trying to figure out whose problem is this (nmbd or xinetd) but so far
we have had no success.
Any clues?
Thanks in advance.
E.-
--
Eloy A. Paris
Information Technology Department
Rockwell Automation Venezuela
Telephone: +58-2-9432311 Fax: +58-2-9431645