vCenter Service (vpxd) Fails to Start with error "Excessive depth in document: 256"

Products

• VMware vCenter Server

Issue/Introduction

• The vSphere Client fails to load; attempting to access the vCenter Server via FQDN returns the below error:

  No Healthy Upstream

• vCenter Server service (vpxd) is confirmed down when checking status using below command:

  service-control --status --all

• Manual attempts to start the vpxd service result in an immediate service crash or panic.
• The /var/log/vmware/vpxd/vpxd.log file contains the following critical errors:
  
  

  ERROR vpxd [vc@4413 sub="Default" opID="HB-host-8##8@17###2-6####9f2"] An unrecoverable problem has occurred, stopping the VMware VirtualCenter service. Error: Error[VdbODBCError] (-1) "ODBC error: (2200N) - ERROR: invalid XML content
  DETAIL: Line 1: Excessive depth in document: 256 use XML_PARSE_HUGE option
  9</uuid><contentId>94f44ce14##########9e6f8fede0</contentId><parent><fileName
  
  Error while executing the query" is returned when executing SQL statement "UPDATE VPX_VM_VIRTUAL_DEVICE SET DEVICE_TEXT =? WHERE ID =? AND DEVICE_KEY =? "
  CRITICAL vpxd [sub="Default" opID="8##8@17###2-6####9f2"]
  
  -- >Panic: Unrecoverable VmRootError. Panic!
  -> Backtrace:
  [backtrace begin] product: VMware VirtualCenter, version: 9.1.0, build: build-25573614, tag: vpxd, cpu: x86_64, os: linux, buildType: release
  backtrace[00] libvmacore.so[0x0047DDB4]
  backtrace[01] libvmacore.so[0x0036D2B/
  x0036D2BA]: Vmacore :: System: :Stacktrace: :CaptureFullWork(unsigned int)
  backtrace[02] libvmacore.so[0x0037E727
  backtrace[03] Libvmacore.so[@x004564A8]
  ovmacore. soLex664s64
  backtrace[04] libvmacore.
  Libvmacore. so[0x0045663C]: Vmacore: :PanicExit(char const*)
  backtrace[05] vpxd[0x02E33530]
  backtrace[06] vpxd[@x02E13833]
  backtrace[07] vpxd[0x0
  backtrace[08] vpxd[@x02E04F69]
  backtrace[09] vpxd[0x02E12DA4]
  backtrace [10] libvmacore. so[0x002C460C]
  backtrace [11] Libvmacor
  Libvmacore.so[0x002Cc1Cc]
   backtrace [13] libc. so GM
  
  "HB-host-8##8@17###2-6####9f2""] [VdbStatement] value = 6#3#
  ERROR vpxd 5956 [sub="Default" opID="HB-host-8##8@17###2-6####9f2"] [VdbStatement] [2]datatype: 1, size: 4, arraySize: 0
  ERROR vpxd 5956 [sub="Default" opID="HB-host-8###01
  ID="8##8@17###2-6####9f2"] [VdbStatement] value = ##01
  ERROR vpxd 5956 [vc@4413 sub="Default" opID="HB-host-8##8@17###2-6####9f2"] [Vdb: : IsRecoverableErrorCode] Unable to recover from 2200N:1
  2026-08-24T16:29:11.857Z ERROR vpxd 5956 [vc@4413 sub="Default" opID="HB-host-8##8@17###2-6####9f2""] [VdbStatement] SQLError was thrown: "ODBC error: (2200N) - ERROR: invalid XML content
  DETAIL: Line 1: Excessive depth in document: 256 use XML_PARSE_HUGE option
  
  
  

 

 

 

 

Environment

VMware vCenter Server 9.x

Cause

A virtual machine in the inventory has a snapshot chain depth that exceeds the internal XML parser's nesting limit (256 levels). This usually occurs when a VM accumulates an excessive number of snapshots (typically 250+), causing the database query updating the VM's device inventory (VPX_VM_VIRTUAL_DEVICE) to fail due to XML complexity.

Resolution

1. Take SSH to the vCenter using root credentials.
2. Identify the Problematic VM by accessing the vCenter database via SSH:

    /opt/vmware/vpostgres/current/bin/psql -d VCDB -U postgres 

3. Query the VM ID from the logs (e.g., ID 6#3#):

   SELECT name FROM vpx_vm WHERE id=6#3#;

4. Verify if the VM has snapshot chain.
5. Log in to the specific ESXi host where the VM resides. Since the vCenter service is currently in a panic state, you must perform the consolidation directly on the ESXi host where the VM resides.
6. After consolidating the snapshots on the host, attempt to restart the vpxd service:

    service-control --start vmware-vpxd