NSX Upgrade Precheck Warning Regarding Pending Host Reboots and DRS

Products

• VMware NSX
• VMware SDDC Manager / VCF Installer

Issue/Introduction

During the NSX upgrade process, the upgrade precheck workflow pauses and displays the following warning message: "Check if there are hosts managed by this NSX that require a reboot and whether DRS is enabled on their clusters".

Environment

• VMware NSX 
• VMware Cloud Foundations (VCF)

 

Cause

The NSX Upgrade Coordinator incorrectly flags a pending host reboot requirement due to a stale state or synchronization issue within the NSX Manager cluster, causing the precheck to fail even when the underlying ESXi hosts do not actually require a reboot.

Resolution

1. Connect to the ESXi hosts managed by the NSX instance via SSH in root mode.

2. Execute the following command to verify the actual host reboot status:
vim-cmd hostsvc/hostsummary | grep -i reboot

   If rebootRequired is set to true, then the host reboot is required.  See Error while checking if host reboot is pending from a previous action.

   If the command returns no output (indicating no actual reboots are pending on the hosts), perform a rolling reboot of the NSX Manager cluster. To perform a rolling reboot, reboot the NSX Manager nodes one by one. Ensure each manager node fully initializes and cluster synchronization is healthy before rebooting the next node.

3. Once the rolling reboot of all NSX Managers is complete, return to the NSX Upgrade Coordinator and re-run the upgrade precheck bundle. The stale warnings should be cleared.
   If host reboots and/or a rolling reboot of the NSX managers does not clear the warning, please open a case with Broadcom Support. See Creating and managing Broadcom cases
   
   Please provide the NSX log bundle when opening the case with Broadcom support - see Collect Support Bundles

Additional Information

Run the NSX Upgrade Pre-check Bundle

NSX Upgrade Guide: Configuring and Upgrading Hosts

NSX Upgrade Guide: Upgrade Hosts Manually