For edge nodes deployed using NSX Manager, if there are changes to VM external to NSX Manager such that the VM MoRef at vSphere changes, the deployment config at NSX Manager may have discrepancies as compared to the current edge VM config at the compute manager. For example, the VM is removed from vSphere inventory and consequently added back.
Edge VM lifecycle management operations like Update and reporting current state activity are impacted.
Edge routing continues to function.
From var/log/proton/localhost_access_log.txt
2020-05-11T10:36:38.865Z - "POST /nsxapi/api/v1/transport-nodes/af9a4d90-aa9b-48d3-a520-b739ad7a4f97?action=refresh_node_configuration&resource_type=EdgeNode HTTP/1.1" 400 224 4448 4448
Relevant log location
/var/log/proton/nsxapi.log
Sample log entry:
2020-05-11T10:36:38.863Z INFO http-nio-127.0.0.1-7440-exec-16 AuditingServiceImpl - FABRIC [nsx@6876 audit="true" comp="nsx-manager" level="INFO" reqId="f0633a4d-aeea-41c5-8a99-df071b85db4a" subcomp="manager" username="admin"] UserName="admin", ModuleName="TransportNodeLcm", Operation="RefreshTransportNode", Operation status="failure", New value=["af9a4d90-aa9b-48d3-a520-b739ad7a4f97"]
2020-05-11T10:36:38.864Z INFO http-nio-127.0.0.1-7440-exec-16 NsxBaseRestController - - [nsx@6876 comp="nsx-manager" level="INFO" subcomp="manager"] Error in API /nsxapi/api/v1/transport-nodes/af9a4d90-aa9b-48d3-a520-b739ad7a4f97?action=refresh_node_configuration&resource_type=EdgeNode caused by exception com.vmware.nsx.management.edge.common.exceptions.EdgeException: {"moduleName":"FABRIC","errorCode":16047,"errorMessage":"[Fabric] Refresh edge null placement configuration failed. Check network connectivity of the edge node."}
Steps to reproduce this issue:
(a)
1. Remove edge VM from VC inventory.
2. Register edge VM to the VC.
(b)
1. Remove edge VM from VC inventory.
2. Remove VC from NSX.
3. Register edge VM to the VC.