[Samba] smbd crashing machine

[Ken Schneider]

I have the following setup:

Compaq Proliant 6000 w/ 4 cpu's, 512 MB ram
SuSE PRO 8.0
samba 2.2.5 w/winbind
using ACL's (acl-2.0.11-2)

Using samba to provide access to a share using MS NT server for
authentication via winbind. All works well until the following occurs:

I am getting the following errors showing in my messages file ( and also
using dmesg):

 <1>Unable to handle kernel NULL pointer dereference at virtual address
0000002d
 printing eip:
c014c3ac
*pde = 00000000
Oops: 0000
CPU:    2
EIP:    0010:[<c014c3ac>]    Not tainted
EFLAGS: 00010246
eax: c6e3a080   ebx: 00000000   ecx: 00000000   edx: dd48e000
esi: 00000000   edi: 00000000   ebp: 00008001   esp: dd48ff34
ds: 0018   es: 0018   ss: 0018
Process smbd (pid: 30096, stackpage=dd48f000)
Stack: c6e3a080 ffffffff 0821c440 dd48ff8c dce838a8 c0146170 c6e3a080
00008001
       00008000 ca6ce000 0821c440 bfffed8c c180edf8 00000000 00000004
ca374160
       c013a0af ca6ce000 00008001 00000000 dd48ff8c 0000001d ca374160
c180ce20
Call Trace: [<c0146170>] [<c013a0af>] [<c013a41e>] [<c01070f3>]

Code: f6 43 2d 10 74 59 f7 c5 00 08 00 00 0f 85 d7 00 00 00 8b 82

At some point the machine crashes with an entry as above and the only
way to recover is to do a hard shutdown (ouch!)

If any additional info is needed please let me know.

-- 
Ken Schneider
Senior UNIX Administrator
Network Administrator
Radiation Therapy Services, Inc.
kschneider-AT-rtsx.com