xfs-masters
[Top] [All Lists]

[xfs-masters] [Bug 794] XFS File System Corrupt in the OLD ARM (ARM920T)

To: xfs-master@xxxxxxxxxxx
Subject: [xfs-masters] [Bug 794] XFS File System Corrupt in the OLD ARM (ARM920T) Platform
From: bugzilla-daemon@xxxxxxxxxxx
Date: Mon, 6 Oct 2008 11:17:00 -0700
Reply-to: xfs-masters@xxxxxxxxxxx
Sender: xfs-masters-bounce@xxxxxxxxxxx
http://oss.sgi.com/bugzilla/show_bug.cgi?id=794





------- Additional Comments From david.kim@xxxxxxxxxxxxxxxxx  2008-10-06 11:16 
CST -------
(In reply to comment #1)
> (In reply to comment #0)
> > Hi,
> > I have problem using XFS file system with OLD ARM (ARM920T) platform.
> > Current configuration is followings.
> > CPU: SAMSUNG S3C2800 (ARM920T core)
> > Kernel version : 2.4.33 (I patched recent PATCH regarding fix dir2 
shortform 
> > structures on ARM old ABI)
> Any other patches applied?

  No, I just applied it. Do I need to patch any more?


> > XFSprogs : version 2.9 4 (I can't use 2.10.1 because of version mismatch)
> Which version mismatch is that?

I got the following kernel error message like "XFS:bad version" when I made a 
xfs file system with 2.10.1 xfsprogs-utility.

[root@dvr]$ ./mkfs.xfs.2.10.1 -V
mkfs.xfs version 2.10.1
[root@dvr]$ 
[root@dvr]$ ./mkfs.xfs.2.10.1 -f /dev/hda1
meta-data=/dev/hda1              isize=256    agcount=4, agsize=30524000 blks
         =                       sectsz=512   attr=2
data     =                       bsize=4096   blocks=122096000, imaxpct=25
         =                       sunit=0      swidth=0 blks
naming   =version 2              bsize=4096   ascii-ci=0
log      =internal log           bsize=4096   blocks=32768, version=2
         =                       sectsz=512   sunit=0 blks, lazy-count=0
realtime =none                   extsz=4096   blocks=0, rtextents=0
[root@dvr]$ 
[root@dvr]$ mount -t xfs /dev/hda1 /pdvr01
mount: wrong fs type, bad option, bad superblock on /dev/hda1,
       missing codepage or other error
       In some cases useful info is found in syslog - try
       dmesg | tail  or so

=== I GOT FOLLOWING DMESG =======================

XFS: bad version
XFS: SB validate failed


And, sometimes I got following error even I used 2.9.4 xfsprogs-utility.

[root@dvr]$ ./mkfs.xfs.2.9.4 -V
mkfs.xfs version 2.9.4
[root@dvr]$ ./mkfs.xfs.2.9.4 -f /dev/hda1 
meta-data=/dev/hda1              isize=256    agcount=16, agsize=7631000 blks
         =                       sectsz=512   attr=0
data     =                       bsize=4096   blocks=122096000, imaxpct=25
         =                       sunit=0      swidth=0 blks, unwritten=1
naming   =version 2              bsize=4096  
log      =internal log           bsize=4096   blocks=32768, version=1
         =                       sectsz=512   sunit=0 blks, lazy-count=0
realtime =none                   extsz=4096   blocks=0, rtextents=0
[root@dvr]$ 
[root@dvr]$ mount -t xfs /dev/hda1 /pdvr01
mount: wrong fs type, bad option, bad superblock on /dev/hda1,
       missing codepage or other error
       In some cases useful info is found in syslog - try
       dmesg | tail  or so

=== I GOT FOLLOWING DMESG =======================

XFS mounting filesystem ide0(3,1)
Filesystem "ide0(3,1)": corrupt inode 128 (bad size 576460752303423488 for 
local inode).  Unmount and run xfs_repair.
0x0: 49 4e 41 ed 01 01 00 02 00 00 00 00 00 00 00 00 
Filesystem "ide0(3,1)": XFS internal error xfs_iformat(5) at line 531 of file 
xfs_inode.c.  Caller 0xc00c40cc
This architecture does not implement dump_stack()
XFS: failed to read root inode



> > Actually, I'm developing DVR(Digital Video Recorder) with above 
configuration.
> > Usually DVR application is trying to make directory whenever needs record 
> > video data to HDD, however, I got sometimes "Unknown error 990" and after 
that 
> > I can't access any data from the HDD through the XFS file system.
> > In case of DVR, customer couldn't check or repair broken file system at 
all 
> > anytime. 
> > My question is what's the problem of error 990 message?
> see what dmesg tells, you when you get this, and attach it please.

Currently, I can't reproduce it. However, It might be "File system has to be 
shutdown...". anyway, I'll attach when I get the error again.

By the way, I also got "No more space on device" error message, even HDD space 
is remained over 450G. It's weired.




> > And, is there any other problem in the XFS file system with ARM old 
platform?
> You should run the xfsqa suite to give it a good workout...

How do I get this one?
There is no xfsqa in the xfsprogrs-utility package.


> > Somebody told that can't use XFS file system with old ARM in the linux 
kernel 
> > mailing list. I just wondering this is true or not.
> on some hardware there is also a "vmap cache aliasing problem" but that 
usually
> shows up in log write problems (thanks hch for reminding me of this 
yesterday) :)
> Can you build your xfs.ko with -g in the CFLAGS and attach it?  I'll look at 
the
> structure alignment.

file size is over 4M, even gziped. I can't attach it. Is there any other way? 
I have a ftp server here. I can open our server.

Thank you for your help!!

Regards,
David.



-- 
Configure bugmail: http://oss.sgi.com/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.


<Prev in Thread] Current Thread [Next in Thread>