Disk call home

The IBM Storage Scale RAID pdisk is an abstraction of a physical disk. A pdisk corresponds to exactly one physical disk, and belongs to exactly one de-clustered array within exactly one recovery group.

Overall flow for call home

The management server, either Power9 or Utility Node VM, is the central point for call home in the entire cluster. The management server is responsible of collecting call home information from all the nodes that it manages and sending it to IBM. The data that is sent belongs to one of these categories:
  • Daily or weekly heartbeat data: General cluster information that is used for proactive alerts. It can also be used as data points when resolving support tickets.
  • Support tickets: An internal issue is detected by mmhealth (typically a hardware issue) that requires IBM intervention (and possible part replacement).
In the following systems, the I/O nodes collect and store call home data locally, and the management server retrieves local data later.
  • IBM Storage Scale System 3000
  • IBM Storage Scale System 3200
  • IBM Storage Scale System 3500
  • IBM Storage Scale System 5000
Figure 1. Call home flow
Unified call home from GUI
Proactive alerts: Internal IBM tools, which are improved continuously, generate these alerts. Therefore, these alert are not the same although IBM Storage Scale System on the cluster is not changed.
Internal issues: These issues are detected by the IBM Storage Scale System code on the cluster – their detection mechanism is specific to the cluster code level.

Start of changeFrom IBM Storage Scale System 6.2.1.0 onward, IBM Electronic Service Agent (ESA) is removed (the RPM is no longer available); the unified call home is used to create service requests. Although ESA is still installed on Power9 management server for backward compatibility, ESA is deactivated after the node is updated.End of change

The IBM Storage Scale RAID pdisk is an abstraction of a physical disk. A pdisk maps to a physical disk and belongs a declustered array in a recovery group.

The disk hospital is a key feature of the IBM Storage Scale RAID that asynchronously diagnoses errors and faults in the storage subsystem.

A pdisk includes the following attributes:
  • State
    ok
    The pdisk is healthy and operational.
    diagnosing
    The IBM Storage Scale RAID disk hospital is performing a diagnosis when an error occurred.
    missing
    The IBM Storage Scale RAID is cannot communicate with a disk. If a missing disk is reconnected and operational, the disk state displayed as ok.

    For a complete list of pdisk states and further information on pdisk configuration and administration, see IBM Storage Scale RAID Administration.

    Any pdisk with the dead, missing, failing, or slow state is a non-functioning pdisk.

  • Unique worldwide name (WWN)
  • Field replaceable unit (FRU) code
  • Physical location code
Start of changeThe disk hospital applies the replace flag to replace disks immediately, regardless of the replacement threshold of their declustered array. A call home notification is generated when the total number of non-functioning pdisks exceeds the replacement threshold, but only for disks that receive the replace flag.End of change
Note: Start of changeMissing disks must be examined and replaced if no other problem is found.End of change
Figure 2. IBM Storage Scale System call home block
IBM Storage Scale System Call Home Block Diagram

The callback script communicates with mmhealth and captures a call home ticket information in the call home staging area of the node. The call home node (the management server, MS) uploads this data and submit a support ticket to IBM through mmcallhome.