HP D7171A - NetServer - LPr Manual page 7

ZKLWHýSDSHU
+ 3 ý 1 H W 6 H U Y H U ý 0 D Q D J H P H Q W
Each voltage has an upper and lower threshold for which an alert can be issued. If any
voltage is out of range, an error is logged in the Event log. For an extreme voltage
deviation from range, the system issues shutdown commands to prevent hardware damage
and protect customer data from unstable circuitry that can result from fluctuating voltage
levels. (Thresholds can be set so that the system will shut down upon over-temperature
conditions as well.)
In addition, on certain HP NetServers, the power supplies will generate a pre-failure
condition. Again, this pre-failure condition is monitored by the system and alerts are
generated and logged in the Event log, recording the specific power supply that has
encountered the pre-failure condition, which will be replaced under the system warranty.
Processors
Servers with multiple CPU processors are commonplace. Typically these servers are
designed for high-performance applications that require more CPU power than is currently
available from a single processor. CPU faults are usually unrecoverable errors, meaning
that neither the operating system nor the hardware can manage the fault conditions. If a
server encounters a CPU fault, the operating system halts and users lose their network
connections. When this happens, the CPU needs to be taken offline for reconfiguration,
after which the servers can be restarted (sometimes automatically) and users can
reconnect to the server.
Some server manufacturers have chosen to monitor certain aspects of processor functions,
such as the L2 cache error corrections. HP has studied this failure mode and has
determined that its frequency of occurrence is so low that monitoring it for a Pre-Failure
Warranty is not appropriate. HP has instead implemented fault-resilient boot, so that a
failed processor is removed from the configuration during POST.
How to Take Advantage of the HP Pre-Failure Warranty
This section details how system administrators are warned of potential Pre-Failure
Warranty events. Because HP TopTools is the standard monitoring mechanism, the
procedures are described in terms of what happens with HP TopTools.
HP TopTools identifies pre-failure conditions by displaying a yellow status indicator to show
that a component is degraded. Along with the yellow color code, HP TopTools also displays
messages recommending replacement of the component in a pre-failure condition.
Upon receiving a HP TopTools alert that a component has degraded, the system
administrator follows the on-screen instructions or contacts an authorized HP service
provider to determine procedures and eligibility for replacement of the affected component
under warranty.
7