Journalled File System general discussion

J. Peters | 1 May 2006 17:40

Picon

Re: [Q] FC5 and JFS crash

3 more crashes this weekend:

[root <at> XXXhp360-XX log]# last -f /var/log/wtmp.1
reboot   system boot  2.6.16-1.2080_FC Sun Apr 30 09:55          (23:15)
XXXXXXX  pts/2        :1.0             Sun Apr 30 09:33 - crash  (00:21)
XXXXXXX  pts/1        :1.0             Sun Apr 30 09:33 - crash  (00:21)
XXXXXXX  pts/0        XXX.XX.XX.XXX    Sun Apr 30 09:32 - crash  (00:22)

I am leaning towards hardware issues at this point vs. software. This system was
setup with FC5 out of the box.

> ----- Original Message -----
> From: "Dave Kleikamp" <shaggy <at> austin.ibm.com>
> To: "J. Peters" <japeters <at> mail.com>
> Subject: Re: [Jfs-discussion] [Q] FC5 and JFS crash
> Date: Sat, 29 Apr 2006 09:02:57 -0500
> 
> 
> On Fri, 2006-04-28 at 19:10 -0700, J. Peters wrote:
> > Has anyone kicked the tires on FC5 and JFS?
> >
> > I am getting a crash using this configuration:
> >
> > rpm -q kernel-smp
> > kernel-smp-2.6.15-1.2054_FC5
> > rpm -q jfsutils
> > jfsutils-1.1.10-4

No Oops! in dmesg or /var/log/messages.

(Continue reading)

J. Peters | 5 May 2006 16:15

Picon

Re: [Q] FC5 and JFS crash

Hardware passed diagnostics.

The reason why there is no Kdump "vmcore" is that CONFIG_PROC_VMCORE has to
be set to configure the pseudo filesystem "/proc/vmcore" in the kernel.  This
is not set by default in FC5 as given by "cat /proc/filesystems":

cat /proc/filesystems
nodev   sysfs
nodev   rootfs
nodev   bdev
nodev   proc
nodev   cpuset
nodev   binfmt_misc
nodev   debugfs
nodev   securityfs
nodev   sockfs
nodev   usbfs
nodev   pipefs
nodev   futexfs
nodev   tmpfs
nodev   inotifyfs
nodev   eventpollfs
nodev   devpts
        ext2
nodev   ramfs
nodev   hugetlbfs
        iso9660
nodev   mqueue
        ext3
        jfs

(Continue reading)

Ingo | 21 May 2006 19:37

Picon

JFS added uid, gid and umask options, added officially?

Hi Dave,

in February you provided me with a patch for SUSE 10.0 to enable
the use of JFS-partititions created by OS/2, adding uid, gid and umask.
It really works fine for me, but now SUSE-update proposes to also
update the kernel (which I refused for the time).

My question is therefore whether your patch has been incorporated
in the official distribution (and thus updates), so I can safely update
the kernel and the associated modules?

Best regards,
Ingo

P.S.: any news regarding the 'directory size problem'?

-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642

Dave Kleikamp | 22 May 2006 14:49

Picon

Re: JFS added uid, gid and umask options, added officially?

On Sun, 2006-05-21 at 19:37 +0200, Ingo wrote:
> Hi Dave,
> 
> in February you provided me with a patch for SUSE 10.0 to enable
> the use of JFS-partititions created by OS/2, adding uid, gid and umask.
> It really works fine for me, but now SUSE-update proposes to also
> update the kernel (which I refused for the time).
> 
> My question is therefore whether your patch has been incorporated
> in the official distribution (and thus updates), so I can safely update
> the kernel and the associated modules?

No, I'm pretty sure SuSE hasn't incorporated that patch.  SuSE's kernel
is based on 2.6.16, and the uid,gid,umask patch was first included in
the 2.6.17-rc* mainline kernel.  I would assume the same patch would
apply to the newer SuSE kernel.

> Best regards,
> Ingo
> 
> P.S.: any news regarding the 'directory size problem'?

Sorry, I've put this off.  I'll try to look at it again soon.  It helps
to keep bugging me.  
--

-- 
David Kleikamp
IBM Linux Technology Center

-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?

(Continue reading)

Paul Drynoff | 30 May 2006 10:39

Picon

jfs: What may be wrong?

I have 2.6.16-gentoo-r7, compiled without any debug support,
but recently I found in my log such strange messages, 
what may be wrong?

May 30 12:20:05 localhost [<c01b3026>] __get_metapage+0x3c/0x3a1
May 30 12:20:05 localhost [<e18680a1>] _nv002668rm+0x1d/0x2c [nvidia]
May 30 12:20:05 localhost [<e18680a1>] _nv002668rm+0x1d/0x2c [nvidia]
May 30 12:20:05 localhost [<c01ae86c>] dtSearch+0x15b/0x7b3
May 30 12:20:05 localhost [<c0156005>] __d_lookup+0x85/0xbc
May 30 12:20:05 localhost [<c01ac5dd>] get_UCSname+0x3d/0xe9
May 30 12:20:05 localhost [<c019f5fa>] jfs_lookup+0xa2/0x15e
May 30 12:20:05 localhost [<c01521de>] poll_freewait+0x35/0x3e
May 30 12:20:05 localhost [<c0159359>] mntput_no_expire+0x14/0x59
May 30 12:20:05 localhost [<c014fdac>] link_path_walk+0xb0/0xbb
May 30 12:20:05 localhost [<c014efd7>] __lookup_hash+0xad/0xc9
May 30 12:20:05 localhost [<c015037c>] do_unlinkat+0x5a/0x122
May 30 12:20:05 localhost [<c01199c5>] sys_gettimeofday+0x23/0x51
May 30 12:20:05 localhost [<c01029e3>] sysenter_past_esp+0x54/0x75
May 30 12:20:05 localhost [<c01b3026>] __get_metapage+0x3c/0x3a1
May 30 12:20:05 localhost [<e18680a1>] _nv002668rm+0x1d/0x2c [nvidia]
May 30 12:20:05 localhost [<e18680a1>] _nv002668rm+0x1d/0x2c [nvidia]
May 30 12:20:05 localhost [<c01ae86c>] dtSearch+0x15b/0x7b3
May 30 12:20:05 localhost [<c0156005>] __d_lookup+0x85/0xbc
May 30 12:20:05 localhost [<c01ac5dd>] get_UCSname+0x3d/0xe9
May 30 12:20:05 localhost [<c019f5fa>] jfs_lookup+0xa2/0x15e
May 30 12:20:05 localhost [<c01521de>] poll_freewait+0x35/0x3e
May 30 12:20:05 localhost [<c0159359>] mntput_no_expire+0x14/0x59
May 30 12:20:05 localhost [<c014fdac>] link_path_walk+0xb0/0xbb
May 30 12:20:05 localhost [<c011263b>] do_page_fault+0x168/0x4ae
May 30 12:20:05 localhost [<c014efd7>] __lookup_hash+0xad/0xc9

(Continue reading)

Dave Kleikamp | 30 May 2006 17:13

Picon

Re: jfs: What may be wrong?

On Tue, 2006-05-30 at 12:39 +0400, Paul Drynoff wrote:
> I have 2.6.16-gentoo-r7, compiled without any debug support,
> but recently I found in my log such strange messages, 
> what may be wrong?

This could be caused by a corrupted directory on disk.  This patch
should do a better job of detecting what the problem is.  You should be
able to fix the problem by running "fsck -f" against the partition.

I don't know what the cause of the corruption may be.  I'd be interested
if it is reproducible after fixing it with fsck (assuming fsck does fix
it).

> May 30 12:20:05 localhost [<c01b3026>] __get_metapage+0x3c/0x3a1
> May 30 12:20:05 localhost [<e18680a1>] _nv002668rm+0x1d/0x2c [nvidia]
> May 30 12:20:05 localhost [<e18680a1>] _nv002668rm+0x1d/0x2c [nvidia]
> May 30 12:20:05 localhost [<c01ae86c>] dtSearch+0x15b/0x7b3
> May 30 12:20:05 localhost [<c0156005>] __d_lookup+0x85/0xbc
> May 30 12:20:05 localhost [<c01ac5dd>] get_UCSname+0x3d/0xe9
> May 30 12:20:05 localhost [<c019f5fa>] jfs_lookup+0xa2/0x15e
> May 30 12:20:05 localhost [<c01521de>] poll_freewait+0x35/0x3e
> May 30 12:20:05 localhost [<c0159359>] mntput_no_expire+0x14/0x59
> May 30 12:20:05 localhost [<c014fdac>] link_path_walk+0xb0/0xbb
> May 30 12:20:05 localhost [<c014efd7>] __lookup_hash+0xad/0xc9
> May 30 12:20:05 localhost [<c015037c>] do_unlinkat+0x5a/0x122
> May 30 12:20:05 localhost [<c01199c5>] sys_gettimeofday+0x23/0x51
> May 30 12:20:05 localhost [<c01029e3>] sysenter_past_esp+0x54/0x75

diff -Nurp linux-2.6.16/fs/jfs/jfs_dtree.c linux/fs/jfs/jfs_dtree.c
--- linux-2.6.16/fs/jfs/jfs_dtree.c	2006-03-19 23:53:29.000000000 -0600

(Continue reading)

Nico -telmich- Schottelius | 31 May 2006 09:38

fsck.jfs segfaults

Hello dear jfs-developers!

Have a look at: 

http://creme.schottelius.org/~nico/linux/debug/fs/fsck.jfs-segmentation-fault-2006-05-31

Afair I had this problem some years ago without dm-crypt, but I cannot
remember the situation.

Is it possible, if the disk (hde) is broken, it could cause fsck.jfs
to segfault, because it gets 'non valid' information from the harddisk?

Please CC-me and tell me how to debug / what could be the fault.

Nico

--

-- 
:x

Dave Kleikamp | 31 May 2006 17:03

Picon

Re: fsck.jfs segfaults

On Wed, 2006-05-31 at 09:38 +0200, Nico -telmich- Schottelius wrote:
> Hello dear jfs-developers!
> 
> Have a look at: 
> 
> http://creme.schottelius.org/~nico/linux/debug/fs/fsck.jfs-segmentation-fault-2006-05-31
> 
> Afair I had this problem some years ago without dm-crypt, but I cannot
> remember the situation.
> 
> Is it possible, if the disk (hde) is broken, it could cause fsck.jfs
> to segfault, because it gets 'non valid' information from the harddisk?

It's possible, but that would be a bug.

> Please CC-me and tell me how to debug / what could be the fault.

It would be useful to run jfs_fsck under gdb:

gdb /sbin/jfs_fsck
> run /dev/mapper/nirvana
<jfs_fsck should segfault>
> where

> 
> Nico
> 
--

-- 
David Kleikamp
IBM Linux Technology Center

(Continue reading)

Group

gmane.comp.file-systems.jfs.general.

Search Archive

Page

1

Articles in period: 8

May 2006

Language

Change language

Options

Current view: Threads only / Showing only 20 lines / Not hiding cited text.
Change to All messages, whole messages, or hide cited text.

Post a message
NNTP Newsgroup
Classic Gmane web interface
XML

XML

RSS Feed
List Information

About Gmane

Recent entries

[PATCH] jfs: Functions jfs_freeze() and jfs_unfreeze() always return 0 (16 May 2013)
jfs resize (21 May 2013)
xattr performance (17 May 2013)
[GIT PULL] jfs for 3.10 (3 May 2013)
Negative value for inode??? (24 Jan 2013)
[PATCH] jfs: avoid undefined behavior from left-shifting by 32 bits (7 Jan 2013)
[PATCH 12/21] jfs: drop vmtruncate (15 Dec 2012)
[PATCH 12/21] jfs: drop vmtruncate (3 Nov 2012)
[GIT PULL] jfs bug fix for 3.7 (23 Oct 2012)
[PATCH 12/21] jfs: drop vmtruncate (20 Oct 2012)
[PATCH 00/21 v3] drop vmtruncate (20 Oct 2012)
[PATCH] jfs: Fix FITRIM argument handling (16 Oct 2012)
finding inodes via block (12 Oct 2012)
[PATCH 12/22] jfs: drop vmtruncate (6 Oct 2012)
[GIT PULL] JFS updates for 3.7 (3 Oct 2012)
[PATCH] fs/jfs: Patch with various fixes for the JFS code. (21 Sep 2012)
fs/jfs: TRIM support for JFS Filesystem (18 Sep 2012)
Oh noes! fsck stuck :-/ (9 Sep 2012)
[PATCH] JFS: use list_move instead of list_del/list_add (6 Sep 2012)
[PATCH 12/21] jfs: drop vmtruncate (31 Aug 2012)
[PATCH 2/8] drivers/block/xen-blkback:Evalue bio->bi_rw after calling (30 Jul 2012)

Archives

9	May 2013
2	March 2013
2	February 2013
10	January 2013
2	December 2012
7	November 2012
16	October 2012
17	September 2012
7	August 2012
36	July 2012
17	June 2012
10	May 2012
7	March 2012
1	January 2012
23	December 2011
14	November 2011
3	October 2011
3	September 2011
5	August 2011
5	July 2011
13	June 2011
25	May 2011
21	April 2011
9	March 2011
4	February 2011
7	December 2010
20	November 2010
13	October 2010
6	September 2010
2	August 2010
12	July 2010
17	June 2010
6	May 2010
10	April 2010
8	March 2010
22	February 2010
29	January 2010
42	December 2009
10	November 2009
5	October 2009
18	September 2009
6	August 2009
43	July 2009
24	June 2009
10	May 2009
18	April 2009
10	March 2009
1	January 2009
1	December 2008
1	November 2008
2	October 2008
37	September 2008
26	August 2008
32	July 2008
26	June 2008
36	May 2008
42	April 2008
38	March 2008
21	February 2008
25	January 2008
20	December 2007
35	November 2007
58	October 2007
31	September 2007
49	August 2007
27	July 2007
9	June 2007
11	May 2007
2	April 2007
29	March 2007
10	February 2007
32	January 2007
24	December 2006
18	November 2006
16	October 2006
14	September 2006
17	August 2006
20	July 2006
21	June 2006
8	May 2006
21	April 2006
8	March 2006
38	February 2006
22	January 2006
11	December 2005
25	November 2005
49	October 2005
50	September 2005
36	August 2005
51	July 2005
50	June 2005
15	May 2005
36	April 2005
54	March 2005
25	February 2005
62	January 2005
53	December 2004
25	November 2004
14	October 2004
52	September 2004
5	August 2004
63	July 2004
30	June 2004
33	May 2004
43	April 2004
75	March 2004
9	February 2004
31	January 2004
15	December 2003
7	November 2003
14	October 2003
64	September 2003
8	August 2003
10	July 2003
19	June 2003
23	May 2003
14	April 2003
17	March 2003
20	February 2003
41	January 2003
6	December 2002
30	November 2002
57	October 2002
40	September 2002
29	August 2002
49	July 2002
6	June 2002

Design

Zawodny | Right Menu | Left Menu

Your Own Design

Paste the URL of your CSS below. Download CSS template