I've got a IBM/Netfinity 6000R server that is finally getting full - and since it has a single 100BaseT network connection, the backup times are long, and about to get worse. I have a large library I thought it might be possible to stop services for that library only, back up that SQL database and the IDMDS folders for it on the data drive, then restart those services, and stop the rest, backup the other libraries' folders, restart those. It seemed like a good idea. But when we tried it (using Backup Exec) the backup process failed because it couldn't take a snapshot with some of the libraries running and some not running.
Is there any way to segment the backups like this, to reduce off-line time for individual libraries by backing them up separately ?
I know I need a new server with a bigger pipe, and more volumes (OS on C, SQL on D, Data on E presently), a new server is 'in the pipeline', but management wants to increase the quantity of data on my CS 5.4 system by 50% without waiting for the new server, and the weekly full backup is taking 19 hours already, and my estimate of 31 hours with the new data is unacceptable.