NetBSD Problem Report #38905

From jukka+moray@salmi.ch  Mon Jun  9 07:32:40 2008
Return-Path: <jukka+moray@salmi.ch>
Received: from mail.netbsd.org (mail.netbsd.org [204.152.190.11])
	by narn.NetBSD.org (Postfix) with ESMTP id 069E263B8BC
	for <gnats-bugs@gnats.NetBSD.org>; Mon,  9 Jun 2008 07:32:40 +0000 (UTC)
Message-Id: <20080609073237.1E9E31A901@moray.salmi.ch>
Date: Mon,  9 Jun 2008 09:32:37 +0200 (CEST)
From: j+nbsd@2008.salmi.ch
To: gnats-bugs@gnats.NetBSD.org
Subject: Soekris net4501 running NetBSD 4.0_STABLE crashes
X-Send-Pr-Version: 3.95

>Number:         38905
>Category:       kern
>Synopsis:       Soekris net4501 running NetBSD 4.0_STABLE crashes
>Confidential:   no
>Severity:       serious
>Priority:       high
>Responsible:    kern-bug-people
>State:          closed
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Mon Jun 09 07:35:00 +0000 2008
>Closed-Date:    Fri Dec 28 12:15:24 +0000 2018
>Last-Modified:  Fri Dec 28 12:15:24 +0000 2018
>Originator:     Jukka Salmi
>Release:        NetBSD 4.0_STABLE
>Organization:
>Environment:
System: NetBSD octopus.salmi.ch 4.0_STABLE NetBSD 4.0_STABLE (OCTOPUS) #0: Fri Jan  4 16:25:35 UTC 2008  root@moray.salmi.ch:/b/build/nbsd/4/i386/sys/arch/i386/compile/OCTOPUS i386
Architecture: i386
Machine: i386
>Description:
After running fine for several months a Soekris net4501 system running
a slightly modified NET4501 kernel (kernel config file see below)
suddenly dropped into ddb twice during the last week. A backtrace from
the second crash:

uvm_fault(0xc66c82a0, 0, 1) -> 0xe
kernel: supervisor trap page fault, code=0
Stopped in pid 909.1 (ksh) at   netbsd:fr_delstate+0x9: movl    0x20(%eax),%edx
fr_delstate(c7002ba8,c0210780,0,0,c011beba) at netbsd:fr_delstate+0x9
fr_timeoutstate(d7848ffa,c700299c,c70029c0,c018a257,0) at netbsd:fr_timeoutstate+0x43
fr_slowtimer(0,c70029bc,c70029c0,c018a251,40500) at netbsd:fr_slowtimer+0x15
softclock(0,c66c82a0,c70029e0,c01cf19a,0) at netbsd:softclock+0x1b8
softintr_dispatch(0,c01f0010,c66c0030,c66d0010,c7000010) at netbsd:softintr_dispatch+0x6e
Xsoftclock() at netbsd:Xsoftclock+0x26
--- interrupt ---
memset(2cf1000,c7002a9c,c0160fdc,c7002a88,0) at netbsd:memset+0x2d
uvm_pagealloc_strat(0,0,0,c6faec9c,2) at netbsd:uvm_pagealloc_strat+0x1d0
uvmfault_promote(c7002bec,c7002be4,c6fbfee4,c6fbfee4,0) at netbsd:uvmfault_promote+0xc5
uvm_fault_internal(c66c82a0,8083000,2,0,c020e08c) at netbsd:uvm_fault_internal+0xb02
trap() at netbsd:trap+0x611
--- trap (number 6) ---

The system acts as a filtering bridge. Its kernel config file:

--------------------------------------------------------------------------------
# OCTOPUS

# for use with
# $NetBSD: NET4501,v 1.54 2006/11/28 16:24:55 christos Exp $

include "arch/i386/conf/NET4501"

no options INSECURE
no options LKM
no options USERCONF

options DDB_FROMCONSOLE=0
options DDB_TEE_MSGBUF=1

no options COMPAT_NOMID
no options COMPAT_09
no options COMPAT_10
no options COMPAT_11
no options COMPAT_12
no options COMPAT_13
no options COMPAT_14
no options COMPAT_15
no options COMPAT_16
no options COMPAT_20
no options COMPAT_30
no options COMPAT_43
no options COMPAT_386BSD_MBRPART

no file-system NFS

no options SOFTDEP
no options GATEWAY
no options INET6
no options MROUTING
no options PIM
#options NMBCLUSTERS=4096

options IPFILTER_LOG

no options NFS_BOOT_DHCP,NFS_BOOT_BOOTPARAM

options MEMORY_DISK_HOOKS
options MEMORY_DISK_IS_ROOT
options MEMORY_DISK_SERVER=0
options MEMORY_DISK_ROOT_SIZE=20480
options MEMORY_RBFLAGS=0

options DUMP_ON_PANIC=0

no config netbsd
config netbsd root on ? type ? dumps on none

no cbb* at pci?
no cardslot* at cbb?
no cardbus* at cardslot?
no pcmcia* at cardslot?

no ath* at cardbus?
no atw* at cardbus?
no wi* at pci?
no wi* at pcmcia?

no pseudo-device gif
no pseudo-device faith
no pseudo-device stf
no pseudo-device vlan

options BRIDGE_IPF

pseudo-device md 1
pseudo-device ksyms
--------------------------------------------------------------------------------

>How-To-Repeat:
unknown

>Fix:
dito

>Release-Note:

>Audit-Trail:
From: "Brian A. Seklecki" <lavalamp@spiritual-machines.org>
To: j+nbsd@2008.salmi.ch, gnats-bugs@gnats.NetBSD.org
Cc: 
Subject: re: kern/38905 Soekris net4501 running NetBSD 4.0_STABLE crashes
Date: Sat, 11 Oct 2008 17:00:00 -0400 (EDT)

 Jukka:

 Does the problem persist?  Did you end up replacing the unit?

 I know its a pain in the ass, but if nothing in software changes (or 
 possibly, the behavior of the network traffic and software), perhaps its 
 time to run a memtext86+ test?

 I know its a pain in the ass, but there's an mini-IDE header on the 4501.

 [Just going through old Soekris-related issues]

 ~BAS

From: Jukka Salmi <j+nbsd@2008.salmi.ch>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: kern/38905 Soekris net4501 running NetBSD 4.0_STABLE crashes
Date: Sun, 12 Oct 2008 01:15:54 +0200

 Brian A. Seklecki --> netbsd-bugs (2008-10-11 22:20:04 +0000):
 > The following reply was made to PR kern/38905; it has been noted by GNATS.
 > 
 > From: "Brian A. Seklecki" <lavalamp@spiritual-machines.org>
 > To: j+nbsd@2008.salmi.ch, gnats-bugs@gnats.NetBSD.org
 > Cc: 
 > Subject: re: kern/38905 Soekris net4501 running NetBSD 4.0_STABLE crashes
 > Date: Sat, 11 Oct 2008 17:00:00 -0400 (EDT)
 > 
 >  Jukka:
 >  
 >  Does the problem persist?  Did you end up replacing the unit?
 >  
 >  I know its a pain in the ass, but if nothing in software changes (or 
 >  possibly, the behavior of the network traffic and software), perhaps its 
 >  time to run a memtext86+ test?

 I haven't seen this problem since, it occurred only once. Not that
 this would make the problem smaller...


 Regards, Jukka

 -- 
 bashian roulette:
 $ ((RANDOM%6)) || rm -rf ~

State-Changed-From-To: open->closed
State-Changed-By: maya@NetBSD.org
State-Changed-When: Fri, 28 Dec 2018 12:15:24 +0000
State-Changed-Why:
This bug predates some significant changes in ipfilter that make the backtrace hard to follow. While I don't know if the bug is solved, it's hard to make an investigation based on a backtrace when the functions on it no longer exist.
If you are still having problems, please report a new bug or reply.


>Unformatted:

Home	PR Database Search