Managing CPUs when importing the RAID

The number of CPUs specified for a RAID is stored in the RAID configuration file. However, when importing a RAID to a system or migrating it to another node in an HA cluster, it is important to note that the number of available CPUs may differ.

Tip:

When allocating all available CPUs on a system to a RAID by enumerating them (e. g., 0-3) instead of using the -all flag, please note that only the specified CPUs will be used for this RAID on the system to which it is imported.

If the specified number of CPUs is invalid for the destination node system, the following error will occur:

"Error: The specified CPU numbers (<CPU_numbers>) are greater than the allowed CPU number: <number_of_CPUs>. Ignoring 'cpu allowed' parameter."

In this scenario, after migration, the RAID will operate on all available CPUs in the destination node system and will obtain an additional parameter - cpu_allowed_configured - which indicates the number of CPUs specified in this RAID's the configuration file.

Figure 1. 'cpu_allowed_configured' parameter