Replacement of Failed Processors

LPMC and HPMC

If an active processor fails with a Low Priority Machine Check (LPMC), in an iCOD partition, its processor capacity is replaced instantly by an inactive processor, if any are available in the partition. The failing processor is marked for deconfiguration during the next system reboot.

If an active processor fails with a High Priority Machine Check (HPMC), then upon reboot, the failed processor is deconfigured and its processor capacity is instantly replaced by an inactive processor, if any are available in the partition.

	NOTE: In both of the above scenarios the failed processor should be replaced in a timely manner using your normal hardware support process.

Failed Monarch Processors

Monarch processors (see Title not available for a definition) that are failing with a LPMC are not instantly replaced. When a monarch processor experiences a LPMC, the LPMC monitor marks the processor for deconfiguration; however, the LPMC monitor cannot deactivate the processor, unless the system is rebooted. Deactivation of a monarch processor is not possible because it is the controlling processor of the operating system (CPU 0). Therefore, an iCOD processor cannot replace a (failing) monarch processor.

If your iCOD system has only one active processor, it is considered a monarch processor, and it cannot be replaced online. A reboot of the system is required to replace the failing monarch processor.

If there are multiple active processors in your iCOD system, one of them is designated as the monarch processor, and the other (non-monarch) processors can be replaced online. If the monarch processor fails, it cannot be replaced without a reboot.