Jeremy,
After weeks of bliss, I had a mishap with autofs-4.0.0pre6-1. The kernel
running at the time was 2.2.15pre15 (SMP) + Tronds NFSv3 patch + D.
Higgens server and NLM patch (quite stable actually). The hardware is 2x
P3 500Mhz. w/ U2W Scsi - 128MB of memory.
I was having trouble with my server (permission hosed up), and tried to
shutdown autofs on the client:
Mar 20 18:24:57 pii automount[30491]: >> mount: server1:/server1 failed, reason given
by server: Permission denied
Mar 20 18:24:57 pii automount[30491]: mount(nfs): nfs: mount failure server1:/server1
on /net/server1
Mar 20 18:26:14 pii automount[23312]: attempting to mount entry /net/cy
Mar 20 18:26:32 pii gpm[6694]: No data
Mar 20 18:26:35 pii gpm[6694]: No data
Mar 20 18:28:57 pii automount[23301]: shutting down, path = /misc
Mar 20 18:28:57 pii automount[30528]: expired /net/cy
Mar 20 18:28:57 pii automount[30529]: expired /net/home
Mar 20 18:28:57 pii automount[23312]: shutting down, path = /net
Whereupon this occurred:
ksymoops 2.3.3 on i686 2.2.15pre15. Options used
-V (default)
-k /proc/ksyms (default)
-l /proc/modules (default)
-o /lib/modules/2.2.15pre15/ (default)
-m /lib/modules/2.2.15pre15/System.map (specified)
Mar 20 18:28:57 pii kernel: kmem_free: Bad obj addr (objp=c3ec1760, name=dentry_cache)
Mar 20 18:28:57 pii kernel: Unable to handle kernel NULL pointer dereference at
virtual address 00000000
Mar 20 18:28:57 pii kernel: current->tss.cr3 = 0ed86000, %cr3 = 0ed86000
Mar 20 18:28:57 pii kernel: *pde = 00000000
Mar 20 18:28:57 pii kernel: Oops: 0002
Mar 20 18:28:57 pii kernel: CPU: 1
Mar 20 18:28:57 pii kernel: EIP: 0010:[kmem_cache_free+368/404]
Mar 20 18:28:57 pii kernel: EIP: 0010:[<c01247c8>]
Using defaults from ksymoops -t elf32-i386 -a i386
Mar 20 18:28:57 pii kernel: EFLAGS: 00010286
Mar 20 18:28:57 pii kernel: eax: 0000003e ebx: c3ec1760 ecx: cbeac000 edx:
cbeac000
Mar 20 18:28:57 pii kernel: esi: cffef620 edi: 00000286 ebp: 00000000 esp:
cbeadf20
Mar 20 18:28:57 pii kernel: ds: 0018 es: 0018 ss: 0018
Mar 20 18:28:57 pii kernel: Process automount (pid: 23312, process nr: 40,
stackpage=cbead000)
Mar 20 18:28:57 pii kernel: Stack: cf27d600 cc052134 c3ec17dc c3ec1660 c0134c2f
cffef620 c3ec1760 c3ec1760
Mar 20 18:28:57 pii kernel: c3ec1660 c0134f40 c3ec1760 00000000 00000000
cc3e2a20 c0160c88 cf27d600
Mar 20 18:28:57 pii kernel: cc3e2a44 cc3e2a20 00000000 00000001 c0160b2b
cc3e2a20 cefb8d80 ffffffe7
Mar 20 18:28:57 pii kernel: Call Trace: [dput+223/328] [shrink_dcache_sb+284/296]
[autofs_catatonic_mode+120/128] [autofs_root_ioctl+255/404] [sys_ioctl+437/484]
[system_call+52/56]
Mar 20 18:28:57 pii kernel: Call Trace: [<c0134c2f>] [<c0134f40>] [<c0160c88>]
[<c0160b2b>] [<c0132451>] [<c010a1d4>]
Mar 20 18:28:57 pii kernel: Code: c7 05 00 00 00 00 00 00 00 00 eb 0f 56 53 68 9e 26
1f c0 e8
>>EIP; c01247c8 <kmem_cache_free+170/194> <=====
Trace; c0134c2f <dput+df/148>
Trace; c0134f40 <shrink_dcache_sb+11c/128>
Trace; c0160c88 <autofs_catatonic_mode+78/80>
Trace; c0160b2b <autofs_root_ioctl+ff/194>
Trace; c0132451 <sys_ioctl+1b5/1e4>
Trace; c010a1d4 <system_call+34/38>
Code; c01247c8 <kmem_cache_free+170/194>
00000000 <_EIP>:
Code; c01247c8 <kmem_cache_free+170/194> <=====
0: c7 05 00 00 00 00 00 movl $0x0,0x0 <=====
Code; c01247cf <kmem_cache_free+177/194>
7: 00 00 00
Code; c01247d2 <kmem_cache_free+17a/194>
a: eb 0f jmp 1b <_EIP+0x1b> c01247e3
<kmem_cache_free+18b/194>
Code; c01247d4 <kmem_cache_free+17c/194>
c: 56 pushl %esi
Code; c01247d5 <kmem_cache_free+17d/194>
d: 53 pushl %ebx
Code; c01247d6 <kmem_cache_free+17e/194>
e: 68 9e 26 1f c0 pushl $0xc01f269e
Code; c01247db <kmem_cache_free+183/194>
13: e8 00 00 00 00 call 18 <_EIP+0x18> c01247e0
<kmem_cache_free+188/194>
-------------------------------
It left something hosed up in the kernel, and I had to reboot. I hope
this is useful.
Steve
