Subject: port-i386/29899: PR: kernel panic with NetBSD 3.0_BETA during /etc/daily
To: None <port-i386-maintainer@netbsd.org, gnats-admin@netbsd.org,>
From: None <zza@serpens.de>
List: netbsd-bugs
Date: 04/06/2005 07:00:01
	Note: There was a bad value `' for the field `Priority'.
	It was set to the default value of `medium'.

	Note: There was a bad value `' for the field `Severity'.
	It was set to the default value of `serious'.

>Number:         29899
>Category:       port-i386
>Synopsis:       kernel panic with NetBSD 3.0_BETA during /etc/daily
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    port-i386-maintainer
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Wed Apr 06 07:00:00 +0000 2005
>Originator:     Bernhard Moellemann
>Release:        NetBSD 3.0_BETA
>Organization:
Home
	
>Environment:
	
	
System: NetBSD arcanic 3.0_BETA NetBSD 3.0_BETA (ARCANIC) #0: Fri Apr 1 11:04:53 CEST 2005 root@arcanic:/mnt/root/usr/src/sys/arch/i386/compile/ARCANIC i386
Architecture: i386
Machine: i386
>Description:
Panic during the daily find. The machine crashed two times before while
heavy disk activity but X was aktive, so I blindly types sync at this
occasions and it booted.

In the evening I switchted to console and got the following at the
next morning:

uvm_fault(0x34ffe00, 0x196d000, 0, 2) -> 0xe kernel: page fault trap, code=0
stopped in pid 15536.1 (find) at   netbsd: uvm_pageremove+0x5a:   movl %eax,0x14(%edx)

I requested a backtrace: (or was it traceback?)

>bt
uvm_pageremove(c09bf080, c09bf080, caaa39ec, c03dceeb, c077e680) at uvm_pageremove+0x5a
uvm_pagefree(c09bf080, 123d8, 0, c09af040, 123e0000) at uvm_pagefree+0x149
uvm_km_pgremove(c077e6800, 123d0000, 123e0000, c077e6800, 40500) at uvm_km_pgremove+0x7a
uvm_unmap_remove(c077e700, d23d0000, d23e0000, caaa3aa4) at uvm_unmap_remove+0x27e
uvm_unmap1(...) at uvm_unmap1+0xa0
uvm_km_free(...) at uvm_km_free+0x2d
bufpool_pagefree(...) at bufpool_pagefree+0x19
pool_allocator_free(...) at pool_allocator_free+0x16
pr_rmpage(...) at pr_rmpage+0x84
pool_do_put(...) at pool_do_put+0xcf
allocbuf(...) at allocbuf+0x1d3
getbkl(...) at getbkl+0x153
bio_doread(...) at bio_doread+0x3a
ffs_read(...) at ffs_read+0x1c0
VOP_READ(...) at VOP_READ+0x34
ufs_readdir(...)
VOP_READDIR(...)
vn_readdir(...)
sys_getdents(...)
syscall_plain() at syscall_plain+0x7e
-----syscall (number 272)

I wrote this down to paper, so there may by typos and I got lazy near the
end, thus the ... instead of boring numbers.

The machine did run memtest86 over night and reported no errors.

You find the kernel options, /etc/release and dmesg.boot at
http://circinus.de/panic30/

>How-To-Repeat:
Perhaps it happens again in the next days ...
>Fix:

>Unformatted: