on boot, access beyond end of device

[Phil Meyer]

Mike -- EMAIL IGNORED wrote:
> My minimal FC7 box has 125M ram, 20G hard drive, 400 MHZ. I have been 
> successfully using it for a web server. Recently it has been generating
> a huge number of:
>
>    attempt to access beyond end of device
>
> Below are some messages produced on a reboot showing the problems.  I
> note that df shows no sda2.  After suggestions on a previous thread,
> I ran memory and disk tests, and found no problems.
>
> Thanks for your assistance.
>
> Mike.
>
> --
>
> Feb 25 16:39:54 mbrc20 kernel:  sda: sda1 sda2
> Feb 25 16:39:54 mbrc20 kernel:  sda: p2 exceeds device capacity
> Feb 25 16:39:54 mbrc20 kernel: sd 0:0:0:0: [sda] Attached SCSI disk
> Feb 25 16:39:54 mbrc20 kernel: scsi 0:0:1:0: Direct-Access     IOMEGA   
> ZIP 100          14.A PQ: 0 ANSI: 5
> Feb 25 16:39:54 mbrc20 kernel: sd 0:0:1:0: [sdb] Attached SCSI removable 
> disk
> Feb 25 16:39:54 mbrc20 kernel: scsi 1:0:0:0: CD-ROM            PLEXTOR  
> CD-R   PX-W2410A 1.02 PQ: 0 ANSI: 5
> Feb 25 16:39:54 mbrc20 kernel: scsi 1:0:1:0: CD-ROM            NEC      
> CD-ROM DRIVE:28D 3.04 PQ: 0 ANSI: 5
> Feb 25 16:39:54 mbrc20 kernel: device-mapper: ioctl: 4.11.0-ioctl 
> (2006-10-12) initialised: dm-devel at redhat.com
> Feb 25 16:39:54 mbrc20 kernel: attempt to access beyond end of device
> Feb 25 16:39:54 mbrc20 kernel: sda: rw=0, want=40017743, limit=39062500
> Feb 25 16:39:54 mbrc20 kernel: attempt to access beyond end of device
> Feb 25 16:39:54 mbrc20 kernel: sda: rw=0, want=40017743, limit=39062500
> Feb 25 16:39:54 mbrc20 kernel: attempt to access beyond end of device
> Feb 25 16:39:54 mbrc20 kernel: sda: rw=0, want=39923925, limit=39062500
> Feb 25 16:39:54 mbrc20 kernel: Buffer I/O error on device dm-1, logical 
> block 131056
> Feb 25 16:39:54 mbrc20 kernel: attempt to access beyond end of device
> Feb 25 16:39:54 mbrc20 kernel: sda: rw=0, want=39923925, limit=39062500
> Feb 25 16:39:54 mbrc20 kernel: Buffer I/O error on device dm-1, logical 
> block 131056
>
>   

compare the numbers between:

# cat /proc/partitions

and the first column of:

# df

What might have happened long long ago and far far away, is that fdisk 
incorrectly calculated the partition size.  It has been known to do that.

Allot of the brand new Nokia N810s are shipping with this problem, due 
to the manufacturer copying the disk image of a slightly different model 
disk to the current one.

If the partition sizes between those two commands do not match, then you 
are at a trouble spot.  You should not add any more data to that drive.  
The symptoms described match exactly the symptoms of filling up a drive 
with invalid partition sizes.  It can work for years, as long as you 
don't go above a certain percentage used, depending upon the size of the 
error.

I would recommend, if they do not match, that you backup, wipe, 
repartition, mkfs, boot from rescue and check again, and if ok restore.