Error "Controller service to Transport node's connection is down for too long" and ESXi hosts disconnect after powering on a newly deployed NSX Manager node.
Article ID: 447333
Updated On:
Products
Issue/Introduction
When you power on and connect the network adapter of a newly deployed NSX Manager node, you may observe that ESXi hosts transition to a disconnected state in the NSX Manager UI.
Additionally, the system displays the following alarm: Controller service to Transport node's connection is down for too long.
During troubleshooting, you may notice that pinging the newly deployed NSX Manager from the ESXi hosts or other Layer 3 devices is unsuccessful, while pinging from Layer 2 devices (such as other NSX Manager VMs) succeeds
Environment
VMware NSX
VMware ESXi
Cause
This issue occurs if the default gateway is configured incorrectly on the newly deployed NSX Manager node. An incorrect gateway prevents necessary Layer 3 routing between the NSX Manager and the ESXi hosts.
You can verify if this is the cause by checking the routing table on the impacted NSX Manager using the route -n command.
Resolution
To resolve this issue, you must redeploy the node with the correct network configuration:
Detach and delete the incorrectly configured NSX Manager node.Detach the Failed NSX Manager Node from the NSX Manager Cluster
Redeploy a new NSX Manager node.
During the initial deployment phase, ensure that you specify the correct gateway IP address.
Once the new node is deployed with the proper routing configuration, the impacted ESXi hosts will successfully reconnect.
Additional Information
Configure a static IP address and netmask. Make sure an appropriate network route is also configured. You can use the optional gateway argument, or set a route using the set route command.
You cannot configure a static IP address for NSX Manager or a controller because their static IP address is set at installation and cannot be changed.
Feedback
