I have performed 5 offline defragmentations over the past 3 years. We
are running Exchange 2000 Standard Edition. All defrags have
previously been run, successfully, over the network (GigE) to a hard
disk with ample storage for the copy. The current server doesn't have
the HD space to perform the operation anyway but using the network.
Most recently, I have attached a low end high capacity hard drive via
1394. The first time I ran the defrag (6mos ago), this approach was
successful. Last weekend, after successfully defragging the priv1.etb
file (using eseutil), the priv1.stm took FOREVER, over 12 hours to be
exact. This was painful, but I did not want to interupt the process,
so I waited, and waited, and waited. Hours later the process
completed, but a) the CPU was maxed necessitating a reboot b)the
information store started but showed only a total reduction of 1GB
(from 12 to 11) for the entire defrag process. Exchange System Manager
shows me that the total mailbox size is about 6GB. The public store
size is very small. Obviously the defrag didn't do its job. I am going
to try a different approach this evening.

Has anyone seen this behavior before? Any recommendations?