The vpxd service crashes with out of memory errors due to incorrect content in vpxd.cfg

Products

• VMware vCenter Server

Issue/Introduction

• vpxd crashes from time to time;
  
  
• In /var/log/vmware/vpxd/vpxd-xxxx.log, there is repeated "Current value xxxxxxx exceeds soft limit 1048576".
  
  
• When vpxd crashed, the backtrace is like below:
  
  yyyy-mm-ddThh:mm:ss.xxxZ warning vpxd[xxxxxxx] [Originator@xxxx sub=Memory checker] Current value xxxxxxx exceeds soft limit 1048576. Shutting down process.
yyyy-mm-ddThh:mm:ss.xxxZ panic vpxd[xxxxxxx] [Originator@xxxx sub=Default]
-->
--> Panic: Memory exceeds hard limit. Panic
--> Backtrace:
--> [backtrace begin] product: VMware VirtualCenter, version: 8.0.2, build: build-23319993, tag: vpxd, cpu: x86_64, os: linux, buildType: release
--> backtrace[00] libvmacore.so[0x0053D3AB]
--> backtrace[01] libvmacore.so[0x0043197E]: Vmacore::System::Stacktrace::CaptureFullWork(unsigned int)
--> backtrace[02] libvmacore.so[0x00444252]: Vmacore::System::SystemFactory::CreateBacktrace(Vmacore::Ref<Vmacore::System::Backtrace>&)
--> backtrace[03] libvmacore.so[0x00516A74]
--> backtrace[04] libvmacore.so[0x00516B8C]: Vmacore::PanicExit(char const*)
--> backtrace[05] libvmacore.so[0x004316A0]: Vmacore::System::ResourceChecker::DoCheck()
--> backtrace[06] libvmacore.so[0x003840E1]
--> backtrace[07] libvmacore.so[0x0037DBE6]
--> backtrace[08] libvmacore.so[0x003834F9]
--> backtrace[09] libvmacore.so[0x0051D093]
--> backtrace[10] libpthread.so.0[0x00008EAE]
--> backtrace[11] libc.so.6[0x000FDE2F]
--> backtrace[12] (no module)
--> [backtrace end]


• When checking the vCenter Server with the vdt tool, there is error like the one below:
  
          VC Machine ID Check
            [FAIL]    Machine ID Check
                        Failed to process the vpxd.cfg file!  Investigate the invalid XML
                        Documentation:     https://knowledge.broadcom.com/external/article?legacyId=82751

Environment

vCenter Server 7.x

vCenter Server 8.x

Cause

There is incorrect content in the /etc/vmware-vpx/vpxd.cfg file. when this happens, vpxd is supposed to have a hard limit of ## GB, but due to the corrupt vpxd.cfg file, vpxd never used the hard limit from the file. Instead, it used the default of 1 GB. It would not take long for the vpxd service to reach this limit and the memory checker to detect that the hard limit had been reached. Thus, the vpxd service crashes.

Resolution

Follow article below to correct content in /etc/vmware-vpx/vpxd.cfg

vCenter is down after wrong configuration for vCLS and reboot