NSX-T proxy crashes upon enabling Publish FQDN

Products

• VMware NSX

Issue/Introduction

• On the NSX-T manager, you have publish FQDN enabled, whereby transport node will use the FQDN to communicate with manager node and not the IP address.
• New configuration changes cannot be updated on the host temporarily.
• Checking the ESXi /var/run/vobd.log logs , you can see similar messages: 

2022-02-28T14:24:50.008Z: [UserWorldCorrelator] 11655566254937us: [vob.uw.core.dumped.encrypted] /usr/lib/vmware/nsx-proxy/bin/nsx-proxy(107224770) /var/core/nsx-proxy-zdump.001 vmware:key/fqid/<VMWARE-NULL>/BG%2dNKP%2d01/ASEAAgEAxZWQRQrmST2%2fFf7NYmwnnQEIAAwAEAAgAAQAQUVTLTI1NgCA34adTZLG1pRiZP6hPfN3rExdTYFVs7N3LgPBdJ5wDAux1kz%2fBdXWBmkSdMWQ9j0OorLLgrSJdqSYF06ghgEA

• Checking the ESXi /var/run/nsx-syslog log, you see the following entries:

nsx-proxy[52815696]:NSX 52815696 - [nsx@6876 comp="nsx-esx"subcomp="nsx-proxy" tid="52815702" level="INFO"] updateIPFqdnState called with list size
...
nsx-proxy[52815696]: NSX 52815696 - [nsx@6876 comp="nsx-esx" subcomp="nsx-proxy" tid="52815696" level="INFO"] NestdbClient::GetCcpStatusMessageCB: Start with size: -1144411644

Environment

VMware NSX-T Data Center 3.x
VMware NSX-T Data Center

Cause

The GetCcpStatusMessageCB has a reference to ccpList as an argument. This is because of passing locally created ccpList to updateIPFqdnState and the same list is being used as a reference argument to bind GetCcpStatusMessageCB.

Resolution

The issue is resolved in NSX-T 3.2.1.0 available at Broadcom Downloads.

Workaround:
If this issue is impacting, you can revert the publish FQDN settings.
Please review the following guide on details to revert publish FQDN NSX-T installation guide.