2 Monitoring File System Operations

Correctly configured Oracle HSM file systems need little routine administrative intervention. You do, however, need to monitor each system for abnormalities. In general, you monitor two things: availability and utilization.

Availability is conceptually straightforward and easy to monitor. When key components such as a host system, network interface, file system, or storage subsystem become unavailable, core functionality is abruptly lost or degraded, and alerts are displayed in the administrative interfaces and logs.

Utilization issues are more subtle and require more judgment on your part. When usage is excessive, lack of resources, such as storage media, can stop the archiving process as effectively as a component failure, even though the system is functioning normally. Utilization levels that presage imminent trouble in one organization may be perfectly consistent with years of trouble-free operation in another. So recognizing trends and rates is crucial when you are monitoring utilization. A resource that is 80% used is fine if utilization grows 1% per year but a crisis if growth exceeds 1% every week.

Oracle HSM provides three monitoring interfaces:

• Oracle HSM Manager

• samu

• Log and Trace Files

Each has its strengths, depending on your working style and habits.

Oracle HSM Manager

Oracle HSM Manager is a browser-based graphical user interface that lets administrators monitor and control all aspects of file-system operations. The Oracle HSM Manager browser interface pages are divided into three sections:

• Banner

• Navigation tree

• Content pane

The banner displays the name of the application and status information, such as the last time data was updated in the browser interface, the name and role of the user that is currently logged in, the name of the management station that is hosting the Oracle HSM Manager software, and the current number and type of unacknowledged faults.

The navigation tree on the left side of the interface contains the server menu and a hierarchical listing of the available displays. Clicking a link in the navigation tree causes the corresponding display to appear in the content pane.

The Monitoring node of the navigation tree is your principle monitoring resource. You can list and sort all faults detected on monitored file systems and equipment. You can configure automatic email alerts. You can list all jobs that are currently running. The Monitoring node also contains a link to the monitoring Dashboard, a popup window that alerts you to problems and provides quick links to summary information on the following areas of concern:

• Daemons

• File Systems

• Archive Media Utilization

• Tape Libraries

• Library Drives

• Volume Pending Load Request

• Unusable Volumes

• Archiving Copy Queue

• Staging Queue

The Metrics and Reports node of the navigation tree provides a comprehensive range of status and utilization reports as well as a link to System Details. The System Details page lets you quickly review the Oracle HSM configuration and provides quick, convenient access to the log and trace files.

A comprehensive help system fully documents the use of the Oracle HSM Manager.

samu

The samu operator utility is a text-based, menu-driven configuration and management interface that you launch from the command line. It is a convenient, lightweight way to monitor Oracle HSM devices, file system activity, and error messages.

The samu utility is in some respects similar to the UNIX vi editor. You select displays, set display options, navigate within and between displays, enter commands, refresh displays, and quit the utility using similar control key sequences. The last line of each display window displays error messages. Displays refresh automatically unless an error occurs, in which case the display halts until the operator takes further action. When desired, you can take snapshots of display windows for later reference.

The h command opens help screens that list all of the keyboard short cuts, commands. and parameters. You can also consult the samu man page and the Oracle Hierarchical Storage Manager and StorageTek QFS samuCommand Reference in the Oracle HSM Customer Documentation Library (http://docs.oracle.com/en/storage/#sw) for additional information.

The following is a typical samu monitoring display:

Archiver status                  samu      5.4     12:24:10 Mar 19 2014
 
sam-archiverd: Waiting for resources
 
sam-arfind:  samma1 mounted at /samma1
Files waiting to start 0    schedule 70,524    archiving 0
Monitoring file system activity.
 
sam-arfind:  DISKVOL1 mounted at /diskvols/DISKVOL1
Files waiting to start 0    schedule 0    archiving 0
Monitoring file system activity. 

                           samu on samqfshost1

Log and Trace Files

The Oracle HSM software performs comprehensive logging and, when configured, tracing. So you may wish to monitor the following files, particularly when problems arise:

• /var/adm/messages

• /var/adm/sam-log

• /var/opt/SUNWsamfs/trace/ (holds trace files for the daemons and processes)

• /var/opt/SUNWsamfs/devlog/ (holds logs for the devices configured in the /etc/opt/SUNWsamfs/mcf file)

• /var/opt/SUNWsamfs/archiver.log

• /var/opt/SUNWsamfs/stager.log

• /var/opt/SUNWsamfs/recycler.log

• additional archiving logs specific to file systems (if configured).

For information on configuring logging and tracing, see the Oracle Hierarchical Storage Manager and StorageTek QFS Installation and Configuration Guide in the Oracle HSM Customer Documentation Library (http://docs.oracle.com/en/storage/#sw).