User`s guide

Monitoring computer health
Monitoring health
Intel Server Manager monitors important computer functions and resources so it can alert you to
problems as soon as it becomes aware of them. Depending on the computer's hardware,
resources and functions, Intel Server Manager can monitor any of the following:
• Chassis intrusion—Monitors when the system's chassis is opened.
• Drive failure prediction—Monitors a
S.M.A.R.T. drive for potential disk failure.
• Drive space—Monitors how much drive space remains on each logical drive.
Thresholds
are configurable for each logical drive.
• ECC error detection—Monitors the detection of ECC (error-correcting code) memory.
• Fan speeds—Monitors the speed for cooling fans installed in the computer.
• Performance—Monitors performance data of counters you specify (for components such
as drives, memory, network traffic, and so forth).
• Power supplies—Monitors the status of each power supply on the system.
• Services—Monitors the system services you have selected.
• Temperatures—Monitors the temperature of vital components in your system.
• Virtual memory—Monitors how much virtual memory remains available to the computer.
Thresholds are configurable.
• Voltages—Monitors the power voltages on the computer's power supply lines.
When a problem occurs in one of the areas listed above, the computer's health status changes
from Normal to either Warning or Critical, depending on the event and its severity. (The status
icon for the computer will include a Warning
or Critical icon.) You can observe a
computer's health change using one of the following Intel Server Manager tools:
• Summary—The computer summary page displays a description of the problem.
• System—Under the System link in the left pane, click the item generating the health
status change (such as Drives or Memory). The page for the item includes a description
of the problem and steps you can take to resolve the problem.
• Alerts—ISM includes five different
alert actions you can select to notify you of health
changes.
You can select the health contributors that determine the computer's health status. By default all
health contributors are selected. If you deselect an item it no longer contributes to the overall
health, but you will still receive alerts for that item. For example, if you decide that Performance
counters do not need to be included in overall health warnings for a server, you can deselect
Performance. You will, however, receive any alerts you have selected for Performance counters.
For an out-of-band IPMI server, a limited number of health contributors can be monitored through
the server's BMC. These are indicated with "(IPMI)" in the list of health contributors. These are
not configurable, so they are monitored when the server is out of band whether or not you have
deselected them on the list of health contributors.
When you receive health status notifications for a blade server or chassis management module
(MM), you can view the details in the management module's web interface provided by the
45