Re: Understanding cores...
Re: Understanding cores...
- Subject: Re: Understanding cores...
- From: "Brian Bechtel" <email@hidden>
- Date: Mon, 8 Jan 2007 09:24:47 -0800
On 1/6/07, Michael Tuexen <email@hidden> wrote:
I'm looking at a core of a MacPro. It crashes a lot and it use an
NKE. So I would
like to understand why the system is crashing, if it is related to
the NKE and
finally how to fix it...
[...]
(gdb) paniclog
panic(cpu 2 caller 0x001A3135): Unresolved kernel trap (CPU 2, Type
14=page fault), registers:
CR0: 0x8001003b, CR2: 0xffffffe0, CR3: 0x011b0000, CR4: 0x000006e0
EAX: 0x00000000, EBX: 0x352c4038, ECX: 0x352c4000, EDX: 0x352dd008
CR2: 0xffffffe0, EBP: 0x00000000, ESI: 0x352f4010, EDI: 0x004ae85c
EFL: 0x00010046, EIP: 0x00135f3b, CS: 0x00000008, DS: 0x03820010
Backtrace, Format - Frame : Return Address (4 potential args on stack)
0x252a3d68 : 0x128d1f (0x3c9540 0x252a3d8c 0x131df4 0x0)
0x252a3da8 : 0x1a3135 (0x3cf1f4 0x2 0xe 0x3cea24)
0x252a3eb8 : 0x19a8d4 (0x252a3ed0 0x352f4000 0x252a3f28 0x352f4000)
Backtrace terminated-invalid frame pointer 0x0
You have a page fault. The offending address is in CR2: 0xffffffe0
The IP at the time of the fault was in EIP: 0x00135f3b
The "showcurrentstacks" demonstrates that the other 3 cores were idle.
There isn't enough information here to diagnose this further.
The systems runs MacOS X 10.4.8 Server. So could it be that the
system locked up and
the Watchdog forced it to reboot. Would we then see a trace like the
one above?
No, if the watchdog forced it to reboot, you wouldn't be able to
attach and get the "showcurrentstacks". A reboot is not a panic,
although a panic can cause a reboot.
You need to examine the code around that IP and see why it's faulting.
On a Mac Pro I have access to, that's in the code "idle_thread" which
tells me... not much.
task vm_map ipc_space #acts pid proc command
0x03808da0 0x013e3f3c 0x037d2ef0 54 0 0x004d2200 kernel_task
activation thread pri state wait_queue wait_event
0x03822e68 0x03822e68 0 IR
reserved_stack=0x24ff8000
kernel_stack=0x252a0000
stacktop=0x252a3d68
0x252a3d68 0x128d1f <panic+382>
0x252a3da8 0x1a3135 <kernel_trap+1538>
0x252a3eb8 0x19a8d4 <trap_from_kernel+19>
stackbottom=0x252a3eb8
try "switchtoact 0x03822e68" and examine the state of the thread.
Sorry, you'll just have to debug further. There isn't enough
information for anyone to help you more than that.
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Darwin-kernel mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden