Error: FlinkContainer service is not running in dual stack environment - Aria Operations for Networks

Products

• VCF Operations for Networks

Issue/Introduction

Upgrade or service health checks fail in dual-stack environments. The FlinkContainer service shows as unhealthy or not running, often accompanied by Nginx start failures.

• Following error is observed on UI:
  
• check-service-health.sh reports Problem: FlinkContainer javaservice is not running.
• Nginx service fails to start with the following error:
  

  nginx: [emerg] socket() [::]:80 failed (97: Address family not supported by protocol)

• Restapilayer logs (/var/log/arkin/restapilayer/) show:
  

  rpcConfigCacheConfig { host: "localhost" }

• yarn node -list shows Number-of-Running-Containers as zero.

Environment

• VCF Operations for Networks (formerly vRealize Network Insight)
• Dual-stack (IPv4/IPv6) configuration

Cause

In dual-stack environments where the hostname is changed from the default, the Flink service may attempt to bind to localhost or localhost6 incorrectly. If nodename is not consistently configured as localhost6 in the health check scripts and configuration, the container count returns zero.

Resolution

Fixed in release 6.14.0 and higher. See Download Broadcom Products and Software to download this release.

If you believe you have encountered this issue, please open a support request with  VCF Operations for Networks and refer to this KB article for a workaround .

For more information, see Creating and managing Broadcom support cases.

Additional Information

To speak with a customer representative or a Support Engineer see Contact Support. Scroll to the bottom of the page and click on your respective region.