System information

Chapter 10. Using the Advanced Systems Management Interface 295
Figure 10-16 shows the options you can choose relating to a system dump. You can choose
to change one or more of the system dump options, or you can choose to change one or
more of the options and initiate a system dump.
Figure 10-16 Initiating a system dump
Dump policy
Select the policy to determine when system dump data is collected. The As needed policy
instructs the service processor (SP) to collect system dump data only when the SP
determines it is necessary — typically only when a specific failure has not been identified.
The Always policy instructs the SP to collect the dump data after any hardware or firmware
error resulting in a system crash. The default policy is As needed.
Hardware content
Select the policy to determine how much hardware data is collected for a system dump. The
Automatic policy instructs the service processor (SP) to collect the hardware data that it
determines is necessary, depending on the particular failure. Automatic is the default policy.
Note that collection of hardware data can be time consuming. In some cases, the user may
wish to override the default policy. The Minimum policy instructs the SP to collect the
minimum amount of hardware data. This selection allows the user to minimize the hardware
data portion of the system dump.
Note that if this option is selected, the debug data collected for some errors may be
insufficient. The Medium policy instructs the SP to collect a moderate amount of hardware
data. This option strikes a balance between dump collection time and amount of data
available for problem debug.
The Maximum policy instructs the SP to collect the maximum amount of hardware data. Note
that if this selection is chosen, the collection of hardware data can be quite time consuming,
especially for systems with a large number of processors.
Service processor dump
Use this procedure only under the direction of your service provider. You can initiate a service
processor dump in order to capture and preserve error data after a service processor
application failure, external reset, or user request for a service processor dump. This
information can be used to resolve a hardware or server firmware problem. The existing
service processor dump is considered valid if neither the server firmware not the HMC has
collected the previous failure data. Figure 10-17 shows that you can select the option to
enable or disable a system processor dump, or initiate a service processor dump.