Distribution: WhiteBox Enterprise Linux 3.0 (RedHat Enterprise clone)
Hardware Environment: IBM x330, dual Pentium III, 2 GB memory, qlogic qla2300
fibre channel connected disks
Software Environment: NFS-exported filesystems are XFS.
Problem Description:

My NFS server has crashed 3 times the last three weeks. I only managed to catch
the latest oops:

Unable to handle kernel paging request at virtual address 9c010000
 printing eip:
c013c9a5
*pde = 00000000
Oops: 0002 [#1]
SMP
CPU:    1
EIP:    0060:[<c013c9a5>]    Not tainted
EFLAGS: 00010012   (2.6.6)
EIP is at free_block+0x65/0xf0
eax: ed2f7000   ebx: d8ff9000   ecx: d8ff9f58   edx: 9c010000
esi: f7ffd5a0   edi: 00000004   ebp: f7ffd5b8   esp: c0529f78
ds: 007b   es: 007b   ss: 0068
Process swapper (pid: 0, threadinfo=c0529000 task=f7f890e0)
Stack: f7ffd5c8 00000018 f7fcb410 00000018 f7fcb400 f7fcb410 f7ffd5a0 c013d1b0
       f7ffd5a0 f7ffd620 c2022580 c0529fd0 c013d371 c0529000 c20236a0 00000001
       00000286 c013d2b0 c20236a0 c2022580 c0529fd0 c0125184 c0529fd0 c0529fd0
Call Trace:
 [<c013d1b0>] drain_array+0x70/0xb0
 [<c013d371>] reap_timer_fnc+0xc1/0x1a0
 [<c013d2b0>] reap_timer_fnc+0x0/0x1a0
 [<c0125184>] run_timer_softirq+0xc4/0x160
 [<c0121195>] __do_softirq+0xb5/0xc0
 [<c01098bc>] do_softirq+0x4c/0x60
 =======================
 [<c0113f5c>] smp_apic_timer_interrupt+0xcc/0x130
 [<c0104880>] default_idle+0x0/0x40
 [<c0107306>] apic_timer_interrupt+0x1a/0x20
 [<c0104880>] default_idle+0x0/0x40
 [<c01048ad>] default_idle+0x2d/0x40
 [<c0104946>] cpu_idle+0x46/0x50
 [<c011d917>] __call_console_drivers+0x57/0x60
 [<c011da20>] call_console_drivers+0x90/0x120
                                                                                
                                                                                
Code: 89 02 8b 43 0c c7 03 00 01 10 00 31 d2 c7 43 04 00 02 20 00
 <0>Kernel panic: Fatal exception in interrupt
In interrupt handler - not syncing


Steps to reproduce: