A volume of size 2tb with 7 luns (250 GB each) inside that are allocated to a exchange server. The problem is I see the volume is 98% full from the filer side, but when I look in the SME it shows "each lun uses only 50 GB (out of 250Gb)" which means there should be 1.4tb still available but it I see only 47gb available.
Is there any thing needs to be done to see the updated used space on the filer end.
this is expected behavior. If you fill a lun with 1gb, the lun on the netapp has 1gb of written blocks. If you now delete the 1gb of blocks on the lun, Windows will just delete the references in the master file table, it will not free each and every one of these 1gb of blocks. Thus the netapp machine doesnt know about the free blocks either and the lun stays at 1gb utilization. You will need to run snapdrive free space reclaimation to free these blocks. Be aware that this is a cpu and i/o intensive task.
Considering you are using thick provisioned LUNs (space reservation was selected during the creation of them), you will always see the volume usage from the filer side corresponding to the sum of all LUN sizes (in this case ~1.750GB).
If you use THIN provisioned LUNs, your volume usage will reflect only the space used in each LUN (in this case ~350GB).
Now speaking about space reclamation, consider the following: you have a 250GB LUN, you write 100GB on it, and then, deletes 50GB.
For the OS, you are using only 50GB, for filer still 100GB. Now if you run space reclamation from the SD on this LUN, the filer will reflect those present 50GB.