vSAN EPD Service fails to start with inaccessible OSDATA and Local Datastore on ESXi 8.0U3
search cancel

vSAN EPD Service fails to start with inaccessible OSDATA and Local Datastore on ESXi 8.0U3

book

Article ID: 442173

calendar_today

Updated On:

Products

VMware vSAN

Issue/Introduction

  • The vSAN Entry Persistence Daemon (EPD) service status is reported as Abnormal or fails to start.
  • Some Firmware upgrading was done without reboot.
  • The OSDATA partition and local datastores are inaccessible or could not be opened in the vSphere Client.
  • System logs (such as vmkernel.log, vpxa, or vsanmonitoring) are missing or have stopped updating.
  • Manual attempts to restart the EPD service (/etc/init.d/epd restart) fail with errors indicating missing persistent storage:
    • INIT: No persistent storage found to backup the DB into.
    • INIT: Failed to clear epd memory reservation.
  • From the DCUI console log, I/O Error can be found on the boot disk. 

Environment

VMware vSphere ESXi 8.0

Cause

The EPD service requires access to the /scratch partition (located on OSDATA for ESXi 8.x) to initialize its database (epd-storeV2.db). If the local boot disk or the storage controller hosting it fails or becomes disconnected (e.g., following an incomplete firmware upgrade or hardware failure), the ESXi host cannot mount the persistent storage required for EPD and system logging.

Resolution

  1. Enter Maintenance Mode: Place the affected ESXi host into Maintenance Mode. For vSAN-enabled hosts, select the Ensure Accessibility data evacuation mode to maintain virtual machine availability while the host is offline.

  2. Perform a System Reboot: Reboot the host to re-initialize the hardware and finalize any pending firmware updates. During the boot process, observe the POST (Power-On Self-Test) or hardware management console (e.g., HPE iLO, Dell iDRAC) to verify if the boot device is correctly recognized by the storage controller.

  3. Validate Boot Disk Integrity: If the host fails to boot or cannot locate the operating system, verify the physical and logical status of the boot drive within the RAID controller settings. If the disk is reported as failed, "Unconfigured," or missing, engage your hardware vendor for further diagnostic testing and potential drive replacement.

Powered by Wolken Software