Upgrading NSX on Hosts Fails with "Hosts Upgrade has failed. To retry the upgrade, exit the host from NSX Maintenance Mode"

Products

• VMware NSX

Issue/Introduction

When attempting to upgrade NSX on a host, the upgrade fails with the following error in the NSX Manager UI:

 

Hosts Upgrade has failed. To retry the upgrade, exit the host from NSX Maintenance Mode.

The detailed error message shows:

 

Please reboot the host immediately to discard the unfinished update. Unexpected error while upgrading upgrade unit: Install of offline bundle failed on host <UUID> with error: [LiveInstallationError] VMware_bootbank_nsx-esx-datapath_4.2.3.1.0-8.0.24954728: Error in running [/etc/init.d/nsx-datapath start upgrade]: Return code: 1 Output: start upgrade begin Exception: Traceback (most recent call last): File "/etc/init.d/nsx-datapath", line 2066, in <module> fromNsxVersion, _ = nsxesxutils.getNSXFromVersionUpgrade() File "/usr/lib/vmware/nsx-esx-datapath/lib64/python/nsxesxutils.py", line 428, in getNSXFromVersionUpgrade vib_list = localcli('software vib list') File "/usr/lib/vmware/nsx-esx-datapath/lib64/python/nsxesxutils.py", line 276, in localcli raise Exception('Failed to execute localcli cmd: software vib list, rc: 1, out: err: Killed') Exception: Failed to execute localcli cmd: software vib list, rc: 1, out: err: Killed Errors: Got no data from process. Command "/usr/lib/vmware/esxcli-software vib.list" exited with error code: 1 It is not safe to continue. Please reboot the host immediately to discard the unfinished update.

The error indicates that the nsx-datapath VIB installation failed due to issues executing the localcli command during the upgrade.

Environment

VMware NSX 4.2.3.1

Cause

The failure occurs when the upgrade process cannot execute the localcli command to validate or list installed VIBs during the upgrade of the nsx-esx-datapath module.

Contributing factors include:

• A temporary connectivity or communication issue between the host and NSX Manager during upgrade bundle application.

• Incomplete or corrupted execution of the offline bundle install process, causing the localcli command to fail.

• Resource/process constraints on the host that interrupt command execution, resulting in “Got no data from process” and premature termination.

Resolution

Workaround:

• Reboot the affected host to clear out the unfinished update state.

  • This ensures incomplete or corrupted bundle installation is discarded.

• After the host is back online, navigate to NSX Manager → System → Upgrade.

• Retry the upgrade for the affected host by selecting the host and choosing Retry in the Upgrade section.

• Confirm that the host upgrades successfully with a status of SUCCESS.

 

 

Additional Information

 

• This error is typically transient and related to temporary command or bundle execution failures during the host upgrade process.

• Always ensure hosts have sufficient resources and stable management plane connectivity before retrying the upgrade.

• Review /var/log/esxupdate.log and /var/log/vobd.log on the host for additional error details if the retry fails.

• If failures persist across retries, verify bundle accessibility, confirm host health, and open a support request with VMware/Broadcom for further analysis.