Attempt to access beyond end of device
 

This is what I have:
RedHat AS 5.1 64 bit
Qlogic HBAs (failover mode)
EVA storage
LVM
VG INFORMATION:
--- Volume group ---
VG Name myvg
System ID
Format lvm2
Metadata Areas 24
Metadata Sequence No 8
VG Access read/write
VG Status resizable
MAX LV 0
Cur LV 3
Open LV 3
Max PV 0
Cur PV 24
Act PV 24
VG Size 383.91 GB
PE Size 4.00 MB
Total PE 98280
Alloc PE / Size 97280 / 380.00 GB
Free PE / Size 1000 / 3.91 GB
VG UUID XwE1BI-BNn9-LfPl-9XKW-M0y2-t7bb-5mMHtN

LV INFORMATION:
--- Logical volume ---
LV Name /dev/myvg/mylv
VG Name ora01wvg
LV UUID cxHbI5-CgqR-AMbp-GgQv-QqKd-Mfdw-RXsAvP
LV Write Access read/write
LV Status available
# open 1
LV Size 185.00 GB
Current LE 47360
Segments 13
Allocation inherit
Read ahead sectors 0
Block device 253:37

Filesystem = ext3


I got the following message in my messages file. I have a couple of questions surrounding it.
* Is it possible to prevent RedHat from mounting in read-only when it encounters a filesystem problem? I would rather it just unmount the filesystem than to mount read-only.

* Is this a storage/LVM/OS issue?


Mar 22 06:54:16 server kernel: attempt to access beyond end of device
Mar 22 06:54:16 server kernel: dm-37: rw=0, want=367001608, limit=367001600
Mar 22 06:54:16 server kernel: EXT3-fs error (device dm-37): read_block_bitmap: Cannot read block bitmap - block_group = 1400, block_bitmap = 45875200
Mar 22 06:54:16 server kernel: Aborting journal on device dm-37.
Mar 22 06:54:16 server kernel: ext3_abort called.
Mar 22 06:54:16 server kernel: EXT3-fs error (device dm-37): ext3_journal_start_sb: Detected aborted journal
Mar 22 06:54:16 server kernel: Remounting filesystem read-only
Mar 22 06:54:16 server kernel: EXT3-fs error (device dm-37) in ext3_prepare_write: IO failure

As to a file system being mounted read-only, you can try to "re-mount" it read-write in this fashion:

mount -o remount,rw /your-directory-name-here


As to the potential LVM related issues, you should make sure that you have logical volumes,
formatted to the same size as you've told the LVM they are ( in the sense of resize2fs ) , mounted
in the right place and such. For example on my system, in what you've shown us, if the logical
volume you're showing us in what you've posted, was meant to be a part of the volume you've
shown us, the VG name field for the logical volume would be the same as the VG name field
in the volume group information. Are we missing some output here?