After rolling reboot of NSX Managers, ESXi host status shows as Degraded in NSX UI

Products

• VMware NSX

Issue/Introduction

• Following a Rolling reboot of NSX Manager cluster,  ESXi host goes in Degraded state and the tunnels are reported as Down:
  
  
  
  
• Inconsistent BFD (Bidirectional Forwarding Detection) session states observed for the host.

Environment

VMware NSX

Cause

• In vmkernel.log of host, the below log entries can be seen which points to Duplicate IP:
  
  2026-02-19T23:13:39.943Z In(182) vmkernel: cpu8:2103583)vdl2: VDL2BFDCheckAndUpdateSessionMac:1347: [nsx@6876 comp="nsx-esx" subcomp="vdl2-22224315"]BFD local vtep segment is same as remote vtep segment###.##.###.#
2026-02-19T23:13:39.943Z In(182) vmkernel: cpu8:2103583)BFD_CreateSession:440:[nsx@6876 comp="nsx-esx" subcomp="bfd"]session: 0xf5e299a3, local: ###.##.###.#, remote: ###.##.###.#, type: overlay
  
  
• The "Degraded" status results from the NSX management plane failing to resynchronize session states quickly after a shift in TEP IP assignments, especially if a duplicate IP segment issue occurred during the transition. Stale entries in the ARP tables or BFD session states prevent tunnels from re-establishing automatically despite functional physical connectivity.

Resolution

To force a resynchronization of the management plane and clear stale BFD session states, place the impacted ESXi host into Maintenance Mode via vCenter Server.

Entering Maintenance Mode triggers a configuration update in NSX that often restores tunnel status automatically.