Cluster expansion (add host) fails if the cluster transport nodes are not configured with overlay transport zone
search cancel

Cluster expansion (add host) fails if the cluster transport nodes are not configured with overlay transport zone

book

Article ID: 318244

calendar_today

Updated On:

Products

VMware Cloud Foundation

Issue/Introduction

  • KB contains the workaround for fixing the above error, while expanding the cluster helps them in unblocking.
  • Error from /var/log/vmware/vcf/domainmanager/domainmanager.log in SDDC manager:

    [YYYY-MM-DDTHH:MM:SS] ERROR [vcf_dm,af02dd018ab871af,e853] [c.v.v.h.s.v.HostSpecValidator,dm-exec-19] Failed to fetch IP address pool details from NSX-T cluster <nsxt-cluster-vip-fqdn>
    java.lang.IllegalStateException: Unable to find any hostswitch of type VDS in the Transport node with Id: <transport-node-id>

Environment

VMware Cloud Foundation 4.5

Cause

As per VCF design, the transport nodes in the cluster (backed by NSX) should be associated with an overlay transport zone, while vlan transport zone is optional. This issue occurs if the customer has removed the overlay transport zone associated with transport nodes.

Resolution

Workaround:

Follow the steps below:


1. Find the Transport Node Profile for the cluster. It usually follows the naming pattern:
<vcenter-vm-name>-<cluster-name>



2. Verify that at least 1 standard host switch is associated with the overlay transport zone. If it exists, skip to step 3, else, create an overlay transport zone using the steps below.

If there is no overlay transport zone associated with the host switch, add one transport zone

2.1 Create a transport zone with name overlay-tz-<nsxt-vip-fqdn> and select Traffic type overlay.

2.2 Click on the manage option and add a tag with the name "vcf" and scope "vcf-orchestration" to the transport zone created above.

2.3 Update a standard switch in the cluster Transport Node Profile by adding this overlay transport zone.

3. Verify that the transport zone identified in step 2 contains the VCF tags.



4. Apply the transport node profile on the cluster.



5. Wait for the cluster reconfiguration.

6. Once the cluster reconfiguration is done, retry the failed workflow from the SDDC manager. Since the workflow is still in validation and phase, and no configuration has been written on the host, a new add host workflow can also be started.