kernel: hda: dma_intr: status=0x51 { DriveReady SeekComplete Error } kernel: hda: dma_intr: error=0x84 { DriveStatusError BadCRC }
Okay, there are a couple issues here... First off, are you running the fsck on the partition or the logical volume? You can't run fsck on an LVM physical volume because it doesn't have a standard file system, it has lvm meta data on it. If you run fsck, it has to be on the logical volume (/dev/VolGroupName/LogVolWhatever).
Now... You seem discouraged with LVM... The above errors are pointing to a drive failing. Are you doing any sort of raid or similar? lvm2 in linux is used to manipulate partitions on the fly and allows for all sorts of spanning and moving and expanding and stuff like that. It isn't like IBM's evms where you can mirror volumes and fun stuff like that. So if you want to protect your data you still need some sort of raid. Going back to just plain ext3 wouldn't help you in this case because it's a hardware failure that's causing your troubles. It's similar to running raid0 and losing 1 drive.
j