Transport Node creation fails with error: "Reason: LogicalSwitchBulkMsg waiting for reply from MPA node"

Products

• VMware NSX
• VMware vCenter Server

Issue/Introduction

• Creating Transport Node fails in NSX-T UI with the error: "error (Reason: LogicalSwitchBulkMsg waiting for reply from MPA node)"
• Shows other errors about communication with the Manager.
• NSX-T Transport Node failed to deploy, but NSX thinks it's deployed.
• Alarm can be seen for communication.management_channel_to_transport_node_down.
  
• There are no vibs installed on the Transport Node.
• The host can ping the VIP and all NSX managers.

Environment

VMware NSX-T 3.x

VMware NSX

‎VMware vCenter Server 7.x

Cause

This issue is caused by a temporary communication failure to the transport node during the initial deployment process, simulating the errors above.

Resolution

1. Put the host in Maintenance mode on the vCenter.
2. Remove the host from the cluster.
3. In NSX, you should see the host move from the cluster to "other nodes".
4. Go into each error, and click "Resolve".
5. Verify the status of the node. (Should see communication is working).
6. Remove NSX, click Force Delete NSX.
7. NSX should show un-configured.
8. Back in vCenter, take host out of maintenance mode.
9. Put it back in the cluster.
10. NSX should automatically deploy to the node.
11. Wait about 5 minutes after deploy completes, and refresh the status, Node should come up with all stats as green.

 

Additional Information

Once the host is up and the status is verified as successful, workloads can be migrated or provisioned back to the host.