HCX - Communication failure over NE HA extended network post infra outage

Products

• VMware HCX

Issue/Introduction

• After a Network Extension (NE) High Availability (HA) failover due to an infrastructure outage, VMs are not reachable between the sites on one of the extended networks.
• The HCX tunnels between the sites remain up and operational.
• Post-NE HA failover, one of the NE Appliance’s vNICs, associated with the affected extended network, is not passing traffic.
• Other extended networks on the same NE Appliance are working fine.
• The following log messages are observed in /var/log/messages log of the affected NE Appliance after it becomes the active node (the vNIC number may vary):
  

   cgw 1316 - - [Err-Tasker] : cmd (/usr/sbin/ip link set dev vNic_1 up) done, error: Timeout
   cgw 1316 - - [Err-configer] : cmd(execCmd<ip link set dev vNic_1 up>) failed: /usr/sbin/ip link set dev vNic_1 up: Timeout

 

Environment

HCX 4.9.x and earlier

Resolution

This issue is resolved in VMware HCX version 4.10, available at Broadcom downloads.

If you are having difficulty finding and downloading software, please review the Download Broadcom products and software KB.

Workaround:  

Re-extend the affected network.

1. Remove affected network extension
2. Extend affected network