The vCenter Alarm "Cannot connect to storage" is unexpectedly disabled following the use of the "Destroy Datastore" workflow within the Pure Storage vSphere Plugin.
search cancel

The vCenter Alarm "Cannot connect to storage" is unexpectedly disabled following the use of the "Destroy Datastore" workflow within the Pure Storage vSphere Plugin.

book

Article ID: 437229

calendar_today

Updated On:

Products

VMware vCenter Server

Issue/Introduction

  • The vCenter root-level alarm "Cannot connect to storage" is found to be disabled without manual intervention.
  • Configured Email actions for the alarm are removed.
  • The issue specifically occurs after performing a Right-click > Pure Storage > Destroy Datastore operation using the Pure Storage vSphere Plugin.
  • Aria Operations for Logs (formerly Log Insight) attributes the reconfiguration to the user who was logged in at the time.
  • Native vCenter datastore deletion workflows do not trigger this behavior.

Environment

Vcenter 8.x

 

Cause

The root cause is a scope violation bug within the Pure Storage vSphere Plugin.

During the automated "Destroy Datastore" workflow, the plugin programmatically reaches up to the vCenter root level to modify or disable the alarm, likely as an attempt to suppress expected alerts while a datastore is being decommissioned. However, the plugin's script fails to revert the alarm to its original enabled state and restore email actions once the process completes.

The reason logs incorrectly blame the user is that third-party plugins utilize the active vCenter session token of the logged-in administrator to execute API calls. Therefore, vCenter records the plugin's automated API script as a manual ReconfigureAlarm event initiated by that user.

Resolution

To ensure critical storage connectivity alarms remain active and email alerts are sent during actual failures, follow the recommendations below:

Immediate Workaround

Immediately cease using the Pure Storage plugin to destroy datastores. Instead, utilize the native vCenter workflow for all datastore decommissioning:

  1. Unmount the datastore from all hosts.
  2. Detach the storage device.
  3. Destroy/Remove the volume on the Pure Storage Array.
  4. Delete the datastore in vCenter.
Software Update

Check the Pure Storage Compatibility Matrix and upgrade the Pure vSphere Plugin to the latest available version to determine if this bug has been patched in a newer release.

Vendor Escalation

If the plugin upgrade does not resolve the issue, open a support case directly with the Pure Storage Support Team to report the orchestration flaw within their vSphere plugin code.

Powered by Wolken Software