we have samba [mostly] working in a VPN application connecting two Mandrake subnets using SSH. if all machines define the same workgroup name, and we define a single domain master browser, (with the other subnet defined as a preferred, local, master browser,) and a single WINS server, after a little while, all shares are visible to all WIN9x machines. this samba stuff is fantastic! 2 questions: 1. we want to define multiple workgroups. should there be as many domain master browsers? 2. is there any way to 'force' an election and all that name exchange/sync stuff? after restarting NMBD and SMBD to effect changes it's like watching a pot boil. as usual, many thanks for your fine work. kim
I am running samba-2.2.2 on RedHat 7.1 as a DMB and WINS in
one subnet and samba-2.2.1a-4 on RedHat 7.2 as a LMB on
another subnet.
an abbreviated DMB's smb.conf: where aaa.bbb.ccc.ddd is the IP
address for PHYS2, the LMB and www.xxx.yyy.zzz is the IP address
for PHYS1, the DMB:
[global]
workgroup = PHYSICS
netbios name = PHYS1
. . .
socket options = TCP_NODELAY SO_RCVBUF=8192 SO_SNDBUF=8192
remote browse sync = aaa.bbb.ccc.ddd (PHYS2)
remote announce = aaa.bbb.ccc.ddd (PHYS2)
local master = yes
os level = 64
domain master = yes
preferred master = yes
domain logons = yes
name resolve order = wins bcast lmhosts hosts
wins support = yes
wins proxy = yes
dns proxy = yes
an abbreviated LMB's smb.conf:
[global]
workgroup = PHYSICS
netbios name = PHYS2
...
security = user
socket options = TCP_NODELAY SO_RCVBUF=8192 SO_SNDBUF=8192
remote browse sync = www.xxx.yyy.zzz (PHYS1)
remote announce = www.xxx.yyy.zzz
local master = yes
os level = 64
domain master = no
preferred master = yes
wins support = no
wins server = www.xxx.yyy.zzz (PHYS1)
dns proxy = yes
So, browsing across the two subnets works for the workgroup
Physics. However, I would like to use these two servers to
inable browsing for several workgroups. i.e. if the Physics
workgroup is the Entire Neighborhood, I want to enable
browsing across subnets for all the workgroups in Physics,
not just one workgroup.
Has anyone acomplished this? Any suggestions?
thanks
--
Charles C. Windsor | ph: 404-651-2279
Georgia State University | fax: 404-651-1427
Physics & Astronomy | email: cwindsor@gsu.edu