Selecting A Cleaning Strategy; Setting Up Host-Managed Drive Cleaning - Oracle StorageTek SL150 User Manual

cleaning media, as described in
page 3-10.

Selecting a Cleaning Strategy

LTO drives sometimes need cleaning because the clearance between LTO magnetic
tape media and the read/write heads is very small. When the tape is streaming past
the recording heads at maximum speed, small clearances maximize data transfer. But
at other times, when the tape is positioning and tensioning to accommodate slow I/O
sources, tape can intermittently touch the recording heads in the drive. The tape wears
heavily under these conditions, and small deposits of recording media accumulate on
the recording surfaces of the drive.
LTO drives are designed to remove normal levels of contamination automatically.
Small internal brushes sweep the debris away before it can build up and become a
problem. So many LTO drives never require supplementary cleaning with cleaning
cartridges. However, slow or intermittent data sources may not be able to provide
enough data to keep a drive streaming. The drive may overrun and reposition often
enough to cause unusually heavy wear to tapes and abnormally hard, heavy deposits
on recording heads. In extreme cases, the internal brushes can no longer remove the
buildup and read/write errors begin to increase. When the errors exceed the error
correction thresholds set by the drive, the drive returns a tape alert and requests
cleaning.
The SL150 user interface notifies you when LTO drives request cleaning, and most host
backup and storage management applications recognize the requests as well. You can
thus handle the required cleanings in either of two ways.
You can automatically manage cleaning operations using host applications like Oracle
Secure Backup, Symantec NetBackup, or IBM Tivoli Storage Manager. You configure
the application for Automatic, Tape Alert, or reactive cleaning, depending on the
application. The application then cleans drives when the drives issue the
corresponding request. See
details.
If host application-managed cleaning is impractical, you can manage cleaning by
monitoring the library's user interface for cleaning messages and responding
accordingly. When a drive requires cleaning, the Library Health indicator at the top of
the interface is set to Degraded, the Health Table displays the drives that need to be
cleaned, and the Health property of the affected drive requests cleaning. You then
make a note of the affected drive(s) and carry out the cleaning as described in
Library User Interface to Clean the Degraded Drive"
automatic under this option, you must monitor drives regularly and respond quickly
when needed.
Note, however, that routine cleaning in the absence of an alert is emphatically NOT
recommended! Cleaning cartridges are abrasive, and overuse can damage LTO drives.
You should not schedule regular cleanings using a host application, nor should you
perform cleanings after some specified number of mounts. Clean only when the drive
tells you that cleaning is necessary.

Setting Up Host-Managed Drive Cleaning

In most cases, you want the host application that manages the library or the backup
schedule to control cleaning operations. You want to set up the host application to
automatically initiate cleaning when a drive requests it. This approach minimizes
disruption to data operations, minimizes operator workload, and insures that
cleanings are done when necessary and only when necessary.
"Providing the Required Cleaning Cartridges"
"Setting Up Host-Managed Drive Cleaning"
Configuring Drive Cleaning Operations
on page 3-7for
on page 7-7. Since cleaning is not
System Configuration 3-7
on
"Use the