ESXi host reboots unexpectedly with Event 44 "Host had been powered off" in hostd logs
search cancel

ESXi host reboots unexpectedly with Event 44 "Host had been powered off" in hostd logs

book

Article ID: 432839

calendar_today

Updated On:

Products

VMware vCenter Server

Issue/Introduction

An ESXi host within a VMware Cloud Foundation (VCF) or vSphere cluster reboots suddenly without a scheduled task or Purple Screen of Death (PSOD). Symptoms:

  • Host disconnects from vCenter Server.

  • Virtual Machines (VMs) are restarted by vSphere HA on other nodes.

  • The following error is observed in /var/log/hostd.log: Event 44 : Host had been powered off. The poweroff was not the result of a kernel error, deliberate reboot, or shut down. This could indicate a hardware issue.

Environment

VMware vSphere ESXi 8.x

Cause

The ESXi kernel detected a boot-up sequence following an ungraceful termination of the previous session. Because the log explicitly excludes "kernel error" (PSOD) and "deliberate reboot," the cause is external to the ESXi software layer. This is typically triggered by:

  • Physical power loss (PSU failure, PDU outage).

  • Hardware-initiated reset (Motherboard/Chipset watchdog timer).

  • Thermal shutdown (CPU/Chassis overheating).

Resolution

 

  1. Review vCenter Events: Confirm no automated workflows (e.g., Update Manager/Lifecycle Manager) were active at the time of the crash.

  2. Analyze Out-of-Band (OOB) Logs: * Log into the Hardware Management Console (iDRAC, iLO, BMC).

    • Review the System Event Log (SEL) and Lifecycle Log for entries matching the timestamp in hostd.log.

    • Look for keywords: Power Input Lost, Power Supply Fault, Critical Temperature, or Fatal Bus Error.

  3. Check Physical Layer: * Verify both Power Supply Units (PSUs) are functional.

    • Ensure the host is connected to redundant power sources (PDU A and PDU B).

  4. Verify Firmware Baseline: Ensure the server BIOS and Power Management Controller firmware match the recommended recipe for your specific ESXi/VCF version.

  5. Diagnostic Commands: Execute the following via ESXi Shell to check for hardware-reported errors:

    • localcli hardware ipmi sel list

    • esxcli hardware ipmi sdr list

 

Additional Information

If no hardware events are found in the OOB management logs, investigate the upstream power infrastructure. If the issue recurs on the same physical blade/server, engage the hardware vendor for a component stress test (Motherboard/Memory).

Powered by Wolken Software