NetBSD Problem Report #41013

From www@NetBSD.org  Sat Mar 14 10:11:58 2009
Return-Path: <www@NetBSD.org>
Received: from mail.netbsd.org (mail.netbsd.org [204.152.190.11])
	by www.NetBSD.org (Postfix) with ESMTP id EA85363BAB8
	for <gnats-bugs@gnats.netbsd.org>; Sat, 14 Mar 2009 10:11:58 +0000 (UTC)
Message-Id: <20090314101158.86A0463BA0A@www.NetBSD.org>
Date: Sat, 14 Mar 2009 10:11:58 +0000 (UTC)
From: 6bone@6bone.informatik.uni-leipzig.de
Reply-To: 6bone@6bone.informatik.uni-leipzig.de
To: gnats-bugs@NetBSD.org
Subject: netbsd5-rc2 crash
X-Send-Pr-Version: www-1.0

>Number:         41013
>Category:       kern
>Synopsis:       netbsd5-rc2 crash - kernel diagnostic assertion "vp->v_usecount == 1" - quota
>Confidential:   no
>Severity:       serious
>Priority:       high
>Responsible:    kern-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Sat Mar 14 10:15:00 +0000 2009
>Last-Modified:  Wed Mar 14 13:00:14 +0000 2012
>Originator:     Uwe Toenjes
>Release:        netbsd-5-rc2
>Organization:
University of Leipzig
>Environment:
netbsd-5-rc2 amd64
>Description:
at high disk load the following error occurs nearly every day

panic: WARNING: SPL NOT LOWERED ON TRAP EXIT 6 0                                
kernel diagnostic assertion "vp->v_usecount == 1" failed: file "/usr/src/sys/ke0

WARNING: SPL NOT LOWERED ON SYSCALL 1 -11008 EXIT ffe00000 7                    
fatal breakpoint trapWARNING: SPL NOT LOWERED ON SYSCALL 1 -10352 EXIT ffe000007
 in supervisor mode                                                             
WARNING: SPL NOT LOWERED ON SYSCALL 1 16 EXIT ffffda6c 7                        
trap type 1 code 0 rip ffffffff80535105 cs 8 rflags 246 cr2  7f7ffd6ed040 cpl 00
SWtARoNppINeGd:  inS PLp idNO 1T 94L5O2W.ER1 E(D fOiNn dS) YaStCA L  L  1  7   7
0x5:  leave                                                                    +
db{3}> 
trace
breakpoint() at netbsd:breakpoint+0x5
panic() at netbsd:panic+0x24d
__kernassert() at netbsd:__kernassert+0x2d
getnewvnode() at netbsd:getnewvnode+0x2a0
ffs_vget() at netbsd:ffs_vget+0x49
ufs_lookup() at netbsd:ufs_lookup+0x7c3
VOP_LOOKUP() at netbsd:VOP_LOOKUP+0x80
lookup() at netbsd:lookup+0x34b
namei() at netbsd:namei+0x1a4
do_sys_stat() at netbsd:do_sys_stat+0x44
sys___lstat30() at netbsd:sys___lstat30+0x2a
syscall() at netbsd:syscall+0xb6
db{3}> show registers
ds          0x6b0
es          0x10
fs          0x660
gs          0x87
rdi         0
rsi         0x3f8
rbp         0xffff800125a106a0
rbx         0xffff800125a106b0
rdx         0x8
rcx         0
rax         0x1
r8          0xffffffff80a0b78a  copyright+0x8adaa
r9          0x1
r10         0xffff800125a105a0
r11         0x8
r12         0x104
r13         0xffffffff80a19220  copyright+0x98840
r14         0xffff800125a10818
r15         0x1
rip         0xffffffff80535105  breakpoint+0x5
cs          0x8
rflags      0x246
rsp         0xffff800125a106a0
ss          0x10
netbsd:breakpoint+0x5:  leave

>How-To-Repeat:

>Fix:

>Release-Note:

>Audit-Trail:
From: 6bone@6bone.informatik.uni-leipzig.de
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: kern/41013: netbsd5-rc2 crash
Date: Sun, 29 Mar 2009 15:10:20 +0200 (CEST)

 the problem also occurs with netbsd-rc3.

From: David Holland <dholland-bugs@netbsd.org>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: kern/41013: netbsd5-rc2 crash
Date: Sun, 29 Mar 2009 17:08:55 +0000

 On Sat, Mar 14, 2009 at 10:15:00AM +0000, 6bone@6bone.informatik.uni-leipzig.de wrote:
  > at high disk load the following error occurs nearly every day
  > 
  > panic: WARNING: SPL NOT LOWERED ON TRAP EXIT 6 0
  > kernel diagnostic assertion "vp->v_usecount == 1" failed: file "/usr/src/sys/ke0
  >  [...]
  > ffs_vget() at netbsd:ffs_vget+0x49
  > ufs_lookup() at netbsd:ufs_lookup+0x7c3
  > VOP_LOOKUP() at netbsd:VOP_LOOKUP+0x80

 So, I deduce at least one of the fses involved is ffs -- but is it
 vanilla ffs, softdep, or wapbl? And are you also pounding on any other
 filesystem types, e.g. tmpfs?

 -- 
 David A. Holland
 dholland@netbsd.org

From: 6bone@6bone.informatik.uni-leipzig.de
To: gnats-bugs@NetBSD.org
Cc: kern-bug-people@netbsd.org, gnats-admin@netbsd.org, netbsd-bugs@netbsd.org
Subject: Re: kern/41013: netbsd5-rc2 crash
Date: Mon, 30 Mar 2009 09:16:07 +0200 (CEST)

 On Sun, 29 Mar 2009, David Holland wrote:

 > So, I deduce at least one of the fses involved is ffs -- but is it
 > vanilla ffs, softdep, or wapbl? And are you also pounding on any other
 > filesystem types, e.g. tmpfs?

 I am using only vanilla FFSv2. The root partition is located at an 
 internal raid (scsi, controller perc 5i), some other partitions up to a 
 size of 3 TB are located on external storages (scsi, controller perc 4e). 
 Quota is enabled at some partitions.

 The crash results in an automatic reboot. So I get only a trace if I set 
 ddb.onpanic=1. I will attach a new trace from last night:


 Regards
 Uwe


 uvm_fault(0xffffffff80ccc7e0, 0x0, 1) -> e
 panic: fatal page faultkernel diagnostic assertion "vp->v_usecount == 1" 
 failede

 trap type 6 code 0 rip ffffffff803ce8b3 cs 8 rflags 10246 cr2  70 cpl 0 
 rsp fff0
 kfaetrnale lb: rpeaakgpeo finatul ttr atpra ip,n  csuodpeer=0vi
 sStopped in pid 0.61 (system) at netbsd:qsync+0x103:     movq 
 0x70(%rdx,%rax8
 ),%r14
 db{0}> trace
 qsync() at netbsd:qsync+0x103
 ffs_sync() at netbsd:ffs_sync+0x29f
 VFS_SYNC() at netbsd:VFS_SYNC+0x33
 sync_fsync() at netbsd:sync_fsync+0x85
 VOP_FSYNC() at netbsd:VOP_FSYNC+0x71
 sched_sync() at netbsd:sched_sync+0x15d
 db{0}> show registers
 ds          0x3b0
 es          0xf759
 fs          0x7ac0
 gs          0x12
 rdi         0xffff800072e303b0
 rsi         0
 rbp         0xffff800072837b20
 rbx         0xffff800072e303b0
 rdx         0
 rcx         0
 rax         0
 r8          0
 r9          0
 r10         0x1
 r11         0
 r12         0xffff80006d2b9450
 r13         0xffff80009d2a0018
 r14         0
 r15         0xffff80007300af78
 rip         0xffffffff803ce8b3  qsync+0x103
 cs          0x8
 rflags      0x10246
 rsp         0xffff800072837af0
 ss          0x10
 netbsd:qsync+0x103:     movq    0x70(%rdx,%rax,8),%r14
 db{0}> continue
 or mode
 WARNING: SPL NOT LOWERED ON SYSCALL 1 -20704 EXIT ffe00000 7
 trap type 1 code 0 rip ffffffff80535625 cs 8 rflags 246 cr2  7f7ffd68e020 
 cpl 00
 Stopped in pid 28842.1 (find) at        netbsd:breakpoint+0x5:  leave
 db{1}> continue
 uvm_fault(0xffffffff80ccc7e0, 0x0, 1) -> e

 dump to dev 4,1 not possible
 atal page faultrebooting...


From: Andrew Doran <ad@netbsd.org>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: kern/41013: netbsd5-rc2 crash
Date: Fri, 3 Apr 2009 07:26:32 +0000

 Quota related?

From: Andrew Doran <ad@netbsd.org>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: kern/41013: netbsd5-rc2 crash
Date: Fri, 3 Apr 2009 07:27:00 +0000

 I looked at the getnewvnode() path last week and didn't see how this
 can happen..

From: 6bone@6bone.informatik.uni-leipzig.de
To: gnats-bugs@NetBSD.org
Cc: kern-bug-people@netbsd.org, gnats-admin@netbsd.org, netbsd-bugs@netbsd.org
Subject: Re: kern/41013: netbsd5-rc2 crash
Date: Sat, 4 Apr 2009 11:42:21 +0200 (CEST)

 On Fri, 3 Apr 2009, Andrew Doran wrote:

 > Quota related?

 For testing I have all devices mounted without userquota or groupquota. 
 But it will need some days without a crash before I can say if it helps.


 Uwe

From: 6bone@6bone.informatik.uni-leipzig.de
To: gnats-bugs@NetBSD.org
Cc: kern-bug-people@netbsd.org, gnats-admin@netbsd.org, netbsd-bugs@netbsd.org
Subject: Re: kern/41013: netbsd5-rc2 crash
Date: Sat, 11 Apr 2009 22:24:15 +0200 (CEST)

 On Sat, 4 Apr 2009, 6bone@6bone.informatik.uni-leipzig.de wrote:

 > For testing I have all devices mounted without userquota or groupquota. But 
 > it will need some days without a crash before I can say if it helps.

 With disabled quota the crash doesn't occur.


 Uwe

From: 6bone@6bone.informatik.uni-leipzig.de
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: kern/41013: netbsd5-rc2 crash
Date: Sun, 3 May 2009 09:41:51 +0200 (CEST)

 the error also occurs with netbsd5-stable. with enabled quota the kernel 
 crashes almost every day.

 is it possible to incrase the priority of the problem report to high?


 I attached the kernel trace from last night:

 uvm_fault(0xffffffff80ccd7e0, 0x0, 1) -> e
 panic: fatal page faultkernel diagnostic assertion "vp->v_usecount == 1" 
 failede

 trap type 6 code 0 rip ffffffff803ce9e3 cs 8 rflags 10246 cr2  70 cpl 0 
 rsp fff0
 kfeartnaell :b rpeaagkepo ifnatu lttr tapra pi,n  csoudpeer=v0
 sStopped in pid 0.61 (system) at netbsd:qsync+0x103:     movq 
 0x70(%rdx,%rax8
 ),%r14
 db{3}> trace
 qsync() at netbsd:qsync+0x103
 ffs_sync() at netbsd:ffs_sync+0x2eb
 VFS_SYNC() at netbsd:VFS_SYNC+0x33
 sync_fsync() at netbsd:sync_fsync+0x85
 VOP_FSYNC() at netbsd:VOP_FSYNC+0x71
 sched_sync() at netbsd:sched_sync+0x15d
 db{3}> show registers
 ds          0xa268
 es          0xffe9
 fs          0xcac0
 gs          0x12
 rdi         0xffff8000c385a268
 rsi         0
 rbp         0xffff800071f5cb20
 rbx         0xffff8000c385a268
 rdx         0
 rcx         0
 rax         0
 r8          0
 r9          0
 r10         0x1
 r11         0
 r12         0xffff800072c5f000
 r13         0xffff8000a586ac28
 r14         0
 r15         0xffff800072d86d48
 rip         0xffffffff803ce9e3  qsync+0x103
 cs          0x8
 rflags      0x10246
 rsp         0xffff800071f5caf0
 ss          0
 netbsd:qsync+0x103:     movq    0x70(%rdx,%rax,8),%r14

>Unformatted:

NetBSD Home
NetBSD PR Database Search

(Contact us) $NetBSD: query-full-pr,v 1.39 2013/11/01 18:47:49 spz Exp $
$NetBSD: gnats_config.sh,v 1.8 2006/05/07 09:23:38 tsutsui Exp $
Copyright © 1994-2007 The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.