ESXi Netcpa process may crash in Environment with L2 Bridges configured

Products

• VMware NSX

Issue/Introduction

Symptoms:

• On the Fabric status of the dashboard, Host Communication Channel Status may show as down.  
• Control plane agent to controller may show as down on Host Preparation Tab.

• You may see the following log entries on the vmkernel.log of the affected ESXi host:

2018-11-29T08:52:01.624Z cpu6:11936471)WARNING: vdrb: VdrCpVdrBridgeNetworkFdbUpdate:3792: CP:[I:0x2714] Bridge Network FDB Update: Failed to post message status: Limit exceeded
 

• You may see the following on the syslog of the affected ESXi host:

2019-06-01T23:24:39.055Z esxi_host_fqdn.com netcpa: error netcpa[26D2221700] [Originator@6876 sub=Default] Connection error with controller ##.##.##.##:0 on source port 0, triggering reconnect
2019-06-01T23:24:43.726Z esxi_host_fqdn.com netcpa: error netcpa[26D2221700] [Originator@6876 sub=Default] Connection error with controller ##.##.##.##:0 on source port 0, triggering reconnect
2019-06-01T23:24:44.902Z esxi_host_fqdn.com netcpa: error netcpa[26D2221700] [Originator@6876 sub=Default] Connection error with controller ##.##.##.##:0 on source port 0, triggering reconnect

Note: The preceding log excerpts are only examples. Date, time, and environmental variables may vary depending on your environment.

Cause

The issue is caused by the size of the messages being sent to the NSX Controllers from the ESXi host for bridging updates. This may cause the netcpa connection to flap which in turn will cause some log spew, this log spew will cause the netcpa crash.

Resolution

This is resolved in NSX for vSphere version 6.4.5.

Workaround:
Currently there is no workaround.

Additional Information

Impact/Risks:
Netcpa process on the ESXi host will crash.