Edit

Windows Server 2022 on VMware won't restart because of unrecoverable I/O error

Summary

When you try to restart a Windows Server 2022 virtual machine (VM) that runs on VMware, the VM doesn't restart. Instead, the VM displays an "unrecoverable input/output (I/O) operation" error. This error occurs if the operating system can't flush a registry hive to disk. This condition can affect system stability or data integrity.

This article helps you identify the component that's causing the failure and resolve the issue.

Symptoms

You restart a Windows Server 2022 VM that runs in a VMware environment. During the restart process, you receive an error message that indicates that an I/O operation failed during a registry hive flush. You might also see Event ID 6 logged in the System event log. The event details resemble the following example:

Source: Kernel-General
Event ID: 6
Description: The operating system failed to flush data to the transaction log.

The issue doesn't recur consistently.

Cause

This issue occurs if Windows Server 2022 doesn't flush a registry hive to disk. This behavior can affect system stability or data integrity. Several factors can cause this behavior, including:

  • Resource constraints, such as memory exhaustion
  • Interference from file system filter drivers
  • Temporary system performance degradation during the restart process

Because the error is isolated to the registry flush process and doesn't recur consistently, hardware failure isn't likely to be a factor.

Resolution

To resolve this issue, you have to identify the component that triggered the failure, identify the underlying issue in that component, and then fix that issue.

Step 1: Update your configuration

  1. Verify that the VM has the latest Windows Server 2022 updates and the latest VMware drivers installed.
  2. Verify that you're using the latest version of VMware Tools and any related drivers.
  3. Verify that all third-party filter drivers, such as antivirus or backup software drivers, are up to date.
  4. Apply any missing updates.
  5. To determine whether the updates fixed the issue, restart the VM.

Step 2: Collect data while you reproduce the issue

  1. Set up a monitoring tool such as Performance Monitor (PerfMon). Monitor the following counters on the VM:
    • CPU usage
    • Memory usage
    • Disk usage
  2. Restart the VM.
  3. After the error occurs, save the performance data.

Step 3: Analyze the data and update or adjust your configuration

  1. Review the performance data. In particular, look for spikes in resource usage.
  2. In Event Viewer, review the System and Application logs. Look for warnings or errors that occur before the registry hive flush operation fails.
  3. Use the performance and event data to identify which components might be causing issues. In particular, watch for log entries that occur at the same time as spikes in resource usage. This information can help pinpoint which component, such as a specific driver or process, is at the root of the I/O failure.
  4. Resolve any underlying issues that you discover during your analysis.
  5. To determine whether the updates fixed the issue, restart the VM.

Data collection

If you contact Microsoft Support, you can attach this data to your support request.

  • Exact error message text and screenshots of the relevant events in Event Viewer.

  • Complete System and Application event data that covers the period when the issue occurred. You can export this information from Event Viewer.

  • The performance data that you collected during Step 2.

  • Windows Server 2022 version and build details.

    Note: To see the exact version and build number, run winver at a command prompt on the affected computer.

  • VMware configuration details, including virtual hardware settings and VMware Tools version

References

  • Last updated on