NSX Logical Switch State has failed alert in Aria Operations
Article ID: 438883
Updated On:
Products
Issue/Introduction
Aria Operations is reporting an alert from NSX Manager indicating that the Logical Switch State has failed. In the NSX Manager UI, the logical segments appear in a Success state under the Policy view, but show a Failed Config state under the Manager view.
From the desired_state_manager.json file in the NSX Manager support bundle, several logical switches are reported as being in a failed state and not fully realized . The snippet below shows how these entries appear in the file:
"/nsxapi/api/v1/logical-switches/status": { "fully_realized_switches": <total number of switches realised>, "last_update_timestamp": <last update timestamp in unix format>, "total_switches": <total number of switches>},
To identify the switches reporting a failed state, the following command was used:
grep -B1 ' "state": "failed"' desired_state_manager.json
Example output:
grep -B1 ' "state": "failed"' desired_state_manager.json "logical_switch_id": "########-####-####-####-##########2c", "state": "failed"
No LogicalSwitchStateAutoRectifier events were found in the NSX Manager syslog under /var/log/. The following command confirms that no such events were logged, indicating the background recovery task is not running:
grep -ar "LogicalSwitchStateAutoRectifier: registering out-of-sync-hosts for LogicalSwitches" syslo* | wc -l0
Environment
VMware NSX
VMware Aria Operations 8.x
Cause
This issue can occur when there are connectivity problems with vCenter or when the EAM/VPXD services have crashed. If the condition persists for an extended period, the LogicalSwitchStateAutoRectifier function may stop enqueueing failed logical switches in NSX Manager, leaving the LogicalSwitchHostSyncer with an empty coordinator queue. As a result, the logical switch status remains in a FAILED state.
Resolution
Once connectivity to vCenter or the EAM/VPXD service is restored, reboot the NSX Manager to trigger a logical switch resync. Alternatively, you can update the logical segment description in the NSX Manager UI to force a resync for that specific segment.
Additional Information
Additional KBs that may help resolve the underlying issue in NSX:
Feedback
