NetBSD Problem Report #8931

Received: (qmail 22338 invoked from network); 1 Dec 1999 12:23:42 -0000
Message-Id: <199912011223.OAA07953@roboti.cs.hut.fi>
Date: Wed, 1 Dec 1999 14:23:37 +0200 (EET)
From: Antti Kantee <pooka@iki.fi>
Reply-To: pooka@iki.fi
To: gnats-bugs@gnats.netbsd.org
Subject: softdep filesystem freeze
X-Send-Pr-Version: 3.95

>Number:         8931
>Category:       kern
>Synopsis:       softdep filesystem freeze
>Confidential:   no
>Severity:       critical
>Priority:       high
>Responsible:    kern-bug-people
>State:          closed
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Wed Dec 01 04:24:00 +0000 1999
>Closed-Date:    Sun Oct 29 21:47:50 +0000 2000
>Last-Modified:  Sun Oct 29 21:47:50 +0000 2000
>Originator:     
>Release:        -current ~Nov 25th
>Organization:
>Environment:
System: NetBSD/alpha 1.4P


>Description:

Well, I can only guess if this is actually a softdep problem, but I can't
think of any other explanation for it.

Today I was compiling ghostscript and started untarring a large tar file
containing a lot of small files. Suddenly everything froze. I could ping the
box just great, but nothing else worked. Jogging up to it revealed that
it hadn't panic()'d, and I could type random text at the login-prompt, but
I never got asked for the password.

After that I had a little fun with ddb, here's something hopefully useful:

idle()
mi_switch()
tsleep()
getnewbuf()
getblk()
bread()
ffs_blkatoff()
ufs_lookup()
lookup()
namei()
vn_open()
sys_open()
syscall()
XentSys()


>How-To-Repeat:

Dunno, banging some filesystem real hard for a while might trigger it. I
haven't managed to repeat it myself, but I will keep trying..

>Fix:

no idea, sorry


>Release-Note:
>Audit-Trail:

From: Greg Oster <oster@cs.usask.ca>
To: pooka@iki.fi
Cc: gnats-bugs@gnats.netbsd.org
Subject: Re: kern/8931: softdep filesystem freeze 
Date: Wed, 01 Dec 1999 08:45:55 -0600

 Antti Kantee writes:
 > 
 > >Number:         8931
 > >Category:       kern
 > >Synopsis:       softdep filesystem freeze
 > >Confidential:   no
 > >Severity:       critical
 > >Priority:       high
 > >Responsible:    kern-bug-people
 > >State:          open
 > >Class:          sw-bug
 > >Submitter-Id:   net
 > >Arrival-Date:   Wed Dec  1 04:24:00 1999
 > >Last-Modified:
 > >Originator:     
 > >Organization:
 > >Release:        -current ~Nov 25th
 > >Environment:
 > System: NetBSD/alpha 1.4P
 > 
 > 
 > >Description:
 > 
 > Well, I can only guess if this is actually a softdep problem, but I can't
 > think of any other explanation for it.

 Can you check to see if you have this version of src/sys/kern/vfs_bio.c:
 /*      $NetBSD: vfs_bio.c,v 1.61 1999/11/26 17:18:15 fvdl Exp $        */

 Also: was this on a normal filesystem, or did you have this on a RAID set?

 > >How-To-Repeat:
 > 
 > Dunno, banging some filesystem real hard for a while might trigger it. I
 > haven't managed to repeat it myself, but I will keep trying..

 The above version of vfs_bio.c should fix this particular problem... 
 (It looks like you may have an older version of that file..)

 Later...

 Greg Oster



From: Antti Kantee <pooka@iki.fi>
To: Greg Oster <oster@cs.usask.ca>
Cc: gnats-bugs@gnats.netbsd.org
Subject: Re: kern/8931: softdep filesystem freeze
Date: Wed, 1 Dec 1999 17:17:53 +0200

 On Wed Dec 01 1999 at 08:45:55 -0600, Greg Oster wrote:

 > > Well, I can only guess if this is actually a softdep problem, but I can't
 > > think of any other explanation for it.
 > 
 > Can you check to see if you have this version of src/sys/kern/vfs_bio.c:
 > /*      $NetBSD: vfs_bio.c,v 1.61 1999/11/26 17:18:15 fvdl Exp $        */

 Uh, I really should start tracking my kernels with more care. But from
 looking at the date, it seems that the specific kernel was compiled on
 Nov 26th and since the version of vfs_bio described above would have
 shown up on the supscan of 27th, I guess it's safe to say: I had an
 earlier version.

 > Also: was this on a normal filesystem, or did you have this on a RAID set?

 Good old normal ffs.

 > > Dunno, banging some filesystem real hard for a while might trigger it. I
 > > haven't managed to repeat it myself, but I will keep trying..
 > 
 > The above version of vfs_bio.c should fix this particular problem... 
 > (It looks like you may have an older version of that file..)

 Great if it really does fix the problem (and sorry for the noise). I'd
 like to listen to things a bit though before this pr is closed.

 -- 
 Antti Kantee <pooka@iki.fi>

From: Greg Oster <oster@cs.usask.ca>
To: Antti Kantee <pooka@iki.fi>
Cc: gnats-bugs@gnats.netbsd.org
Subject: Re: kern/8931: softdep filesystem freeze 
Date: Wed, 01 Dec 1999 09:27:07 -0600

 Antti Kantee writes:
 > On Wed Dec 01 1999 at 08:45:55 -0600, Greg Oster wrote:
 > 
 > > > Well, I can only guess if this is actually a softdep problem, but I can't
 > > > think of any other explanation for it.
 > > 
 > > Can you check to see if you have this version of src/sys/kern/vfs_bio.c:
 > > /*      $NetBSD: vfs_bio.c,v 1.61 1999/11/26 17:18:15 fvdl Exp $        */
 > 
 > Uh, I really should start tracking my kernels with more care. But from
 > looking at the date, it seems that the specific kernel was compiled on
 > Nov 26th and since the version of vfs_bio described above would have
 > shown up on the supscan of 27th, I guess it's safe to say: I had an
 > earlier version.

 Ok.

 > > Also: was this on a normal filesystem, or did you have this on a RAID set?
 > 
 > Good old normal ffs.

 Ok.  This is good to know. :)  I *tried* to reproduce what you saw on a 
 non-RAID partition, but was unable to...

 > > > Dunno, banging some filesystem real hard for a while might trigger it. I
 > > > haven't managed to repeat it myself, but I will keep trying..
 > > 
 > > The above version of vfs_bio.c should fix this particular problem... 
 > > (It looks like you may have an older version of that file..)
 > 
 > Great if it really does fix the problem (and sorry for the noise). 

 It certainly fixes a symptom of what may be an underlying problem :-/

 > I'd like to listen to things a bit though before this pr is closed.

 For sure.. I'm leaving this for Frank to close :)

 Thanks for the additional info.

 Later...

 Greg Oster



From: Manuel Bouyer <bouyer@antioche.lip6.fr>
To: Antti Kantee <pooka@iki.fi>
Cc: gnats-bugs@gnats.netbsd.org
Subject: Re: kern/8931: softdep filesystem freeze
Date: Wed, 1 Dec 1999 17:44:12 +0100

 On Wed, Dec 01, 1999 at 02:23:37PM +0200, Antti Kantee wrote:
 > >How-To-Repeat:
 > 
 > Dunno, banging some filesystem real hard for a while might trigger it. I
 > haven't managed to repeat it myself, but I will keep trying..

 If you can, get a core dump when it's in this state: go to debugger
 (ctrl-alt-del) and then 'call cpu_reboot(0x104)'

 --
 Manuel Bouyer, LIP6, Universite Paris VI.           Manuel.Bouyer@lip6.fr
 --
State-Changed-From-To: open->closed 
State-Changed-By: pooka 
State-Changed-When: Sun Oct 29 23:46:41 EET 2000 
State-Changed-Why:  
I haven't had the problem in ages now. 
Guess the revision mentioned fixed it. 
>Unformatted:

NetBSD Home
NetBSD PR Database Search

(Contact us) $NetBSD: query-full-pr,v 1.51 2026/08/10 02:28:17 riastradh Exp $
$NetBSD: gnats_config.sh,v 1.10 2026/05/13 22:00:09 riastradh Exp $
Copyright © 1994-2026 The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.