Multiple ESXi hosts report a Down status within the NSX Manager UI. This condition often results in the NSX-NCP pod entering a CrashLoopBackOff state because it cannot reach the API server.
NSX_VDL2 agent on the ESXi host is reported as down.net-dvs -l on the host shows: com.vmware.common.opaqueDvs.status.component.vdl2 = down/var/log/vobd.log indicate TEP vmkernel interfaces (e.g., vmk10) are unable to obtain a DHCP lease. Unable to obtain a DHCP lease on interface vmk10 vmkernel.log show: No valid v4 tep ip for a long time of vmnic[vmk10] on VDS
The issue occurs when the Transport Node Profile applied to the cluster is configured to use DHCP for IPv4 Assignment, but the environment does not provide a DHCP server for the TEP network, or the deployment requires an IP Pool. This prevents the TEP interface from receiving an IP address, breaking control plane connectivity.
Update the Transport Node Profile to use an IP Pool for TEP address assignment:
vmk10 receives a valid IP address.For general port connectivity verification, ensure ports 1234, 1235, and 443 are open between ESXi hosts and the NSX Manager cluster.
nc -zv <NSX-Manager-IP><port>