What you want to do cannot be done in one step because the various storage layers have no way to communicate that their size has changed to the other layers. For increasing the size, you need to work from the disk up to the filesystem skipping the steps that don't apply to you. (To decrease, you would go in the reverse order). I included the command to check the size at each step to make sure it worked.
Code:
rescan the disk(s) to update the size:
echo 1 > /sys/block/sdX/device/rescan
cat /sys/block/sdX/size
dm-multipath
multipathd -k"resize map mpathXX"
cat /sys/block/dmX/size
md RAID - I have not found a good way to do this. Here's a slow way:
for each drive, fail it then add back and resync
after all are complete grow the array to max size
cat /proc/mdstat
Physical volume
pvresize
pvs
Logical volume
lvresize
lvs
filesystem
now you get to run gfs2_grow
df
