xfs
[Top] [All Lists]

Re: mkfs on large h/w RAID fails

Subject: Re: mkfs on large h/w RAID fails
From: Steve Lord <lord@xxxxxxx>
Date: 28 Jan 2002 13:46:32 -0600
Sender: owner-linux-xfs@xxxxxxxxxxx
On Mon, 2002-01-28 at 13:39, Dave Sill wrote:
> [Please reply only to the mailing list.]
> 
> Steve Lord <lord@xxxxxxx> wrote:
> > On Mon, 2002-01-28 at 12:00, Dave Sill wrote:
> > > Austin Gonyou <austin@xxxxxxxxxxxxxxx> writes:
> > > 
> > > > can you try 2.4.14 or 2.4.17 and also upgrade to the new tools?
> > > 
> > > Do I need the XFS kernel to run mkfs.xfs or can I just install the new
> > > tools?
> > 
> > Any old kernel should do.
> 
> Thought so. Unfortunately, I'm still seeing the same problem after
> installing xfsprogs-1.3.17:
> 
> [root@malachite rpm]# mkfs.xfs -f /dev/sdb1
> meta-data=/dev/sdb1              isize=256    agcount=224, agsize=1048576 blks
> data     =                       bsize=4096   blocks=234870292, imaxpct=25
>          =                       sunit=0      swidth=0 blks, unwritten=0
> naming   =version 2              bsize=4096  
> log      =internal log           bsize=4096   blocks=28670
> realtime =none                   extsz=65536  blocks=0, rtextents=0
> mkfs.xfs: read failed: Input/output error
> mkfs.xfs: data size check failed
> mkfs.xfs: mount initialization failed
> [root@malachite rpm]# 

OK, something is lying about the size of your device, it is coming in
at 234870292 4K blocks. The data size check failed means it attempted to
read the last block of the device and the read got an error. 

You could try

mkfs -t xfs -f -d size=234870290b /dev/sdb1

or some other number smaller than the actual device size and see if
that works.

An strace -v output of the failure would also be useful

Steve


> 
> -Dave
-- 

Steve Lord                                      voice: +1-651-683-3511
Principal Engineer, Filesystem Software         email: lord@xxxxxxx


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