An administrator is updating disk firmware using LCM and receives the following message: “there have been 10 or more cluster services restarts within 15 minutes in the Controller VM” The message appears as each drive is upgraded. What is the reason this message is being generated?

During a Life Cycle Manager (LCM) disk firmware update, each disk is taken offline from the cluster's storage pool to be safely updated. This process is a logical "ejection" of the drive from the perspective of the cluster services. The removal and subsequent re-addition of a disk causes cluster services, particularly Stargate which manages the disk I/O, to restart in order to recognize the change in storage topology. Since this happens for each drive being updated, it can lead to multiple service restarts in a short period, triggering the observed alert.

A. A node has been rebalanced. Cluster rebalancing is a background data migration task managed by Curator and is not the direct cause of service restarts during a firmware update.
B. A drive has been reformatted. Reformatting is a destructive data operation and is not part of a standard, non-disruptive firmware upgrade procedure performed by LCM.
C. A node has been rebooted. A node or CVM reboot is a per-node operation. The message appearing "as each drive is upgraded" indicates a more granular, per-drive event is the trigger.

1. Life Cycle Manager (LCM) Guide v2.6, Chapter: "Performing an Inventory and Updating Software or Firmware", Section: "Firmware Updates". The guide describes the rolling, one-component-at-a-time nature of updates.

2. Nutanix Support Portal, KB-3517, "Troubleshooting Stargate crashes and latency issues". This article explains that Stargate restarts can be triggered by disk removal or failure events as the service reconfigures its I/O domain.