Host core dump files unrelated to NSX services trigger "Application Crashed" alarms in NSX UI upon upgrading to version 4.0.0.1

Products

• VMware NSX

Issue/Introduction

• NSX is constantly reporting Application Crashed Alarms for many Hosts
• Log entries similar to the below are seen in NSX Manager var/log/syslog:

  NSX 3597 MONITORING [nsx@6876 alarmId="########-####-####-####-########62a5" alarmState="OPEN" comp="nsx-manager" entId="########-####-####-####-########93e4" errorCode="MP701099" eventFeatureName="infrastructure_service" eventSev="CRITICAL" eventState="On" eventType="application_crashed" level="FATAL" nodeId="########-####-####-####-########93e4" subcomp="monitoring"] Application on NSX node esxi01.example.com has crashed. The number of core files found is 3. Collect the Support Bundle including core dump files and contact VMware Support team.

• Core dumps are found present in /var/core on the ESXi host unrelated to NSX-T:

  /var/core/xxxx-xxxx-drivers-plugin-util-zdump.001

Environment

VMware NSX

Cause

The issue is caused due to the NSX Manager matching the core dumps without validating that they are utilized in NSX.

Resolution

This issue is resolved in VMware NSX 4.0.1.1, available at Broadcom downloads.

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

Workaround:

• Remove the core dumps from /var/core on the ESXi hosts using the below command:

  root> rm -rf /var/core/*

 

Additional Information