Backups fail with cycle redundancy check (Error 23)

Topic

When you try to run a backup for a Windows system protected by a Datto backup agent, you receive the error message:

"Volume_READ_ERROR_CRC - Error Occurred reading the protected volume: Data error (cycle redundancy check)"

Environment

  • Datto Windows Agent
  • Datto Endpoint Backup for PCs
  • Datto Endpoint Backup with Disaster Recovery
  • Datto Backup for Microsoft Azure
  • Datto Endpoint Backup

Description

Error: 23 is a Windows error code indicating that the specified device is encountering a Data Error (cyclic redundancy check or CRC). Common causes are:

  • Disk errors
  • Filesystem errors
  • Degraded RAID arrays

Identifying the Drive

The logs can be collected from the local machine to verify the error and identify the specific volume (drive letter) that is failing.

The log locations for the agent can be found here based on agent used:

For SIRIS and ALTO devices the logs can also be viewed within the device.

  1. Navigate to the device GUI.via the Datto Partner Portal or locally if you have access.
  2. Select the Protect tab.
  3. Click the Backup Logs link for the agent.
  4. Select Show Agent Logs.

The following is an example of the log that will appear in the agent logs from the error.

EXAMPLE  
2024-01-07T15:56:56.835Z 2024-01-07T07:56:56.835-0800 [TID 0016292] [INFO:] Beginning copying volume 0d223928-0000-0000-0000-100000000000 (E:\)
2024-01-07T15:56:56.999Z 2024-01-07T07:56:56.999-0800 [TID 0016292] [INFO:] Beginning execution of the Configure Hole Punching phase.
2024-01-07T15:56:57.000Z 2024-01-07T07:56:57.000-0800 [TID 0016292] [INFO:] Finished execution of the Configure Hole Punching phase.
2024-01-07T15:56:57.442Z 2024-01-07T07:56:57.442-0800 [TID 0016292] [INFO:] Creating resume file block range provider for 0d223928-0000-0000-0000-100000000000 starting at block 787008 (3.00195 GiB)
2024-01-07T15:56:57.768Z 2024-01-07T07:56:57.768-0800 [TID 0016292] [INFO:] Beginning execution of the Copy Chunk Ranges and Punch Holes phase.
2024-01-07T15:56:58.142Z 2024-01-07T07:56:58.142-0800 [TID 0016292] [ERROR:] Failed to read length 4194304 on block device \\?\Volume{0d223928-0000-0000-0000-100000000000}\: Data error (cyclic redundancy check). | ERROR(Win32) : 23 (0x00000017)
2024-01-07T15:56:58.144Z 2024-01-07T07:56:58.144-0800 [TID 0016292] [ERROR:] Failed to read the chunk at offset 3223584768
2024-01-07T15:56:58.324Z 2024-01-07T07:56:58.324-0800 [TID 0016292] [INFO:] Beginning execution of the Backup Pass Finalization phase.
2024-01-07T15:56:58.469Z 2024-01-07T07:56:58.469-0800 [TID 0016292] [INFO:] Finished execution of the Backup Pass Finalization phase.
2024-01-07T15:56:58.636Z 2024-01-07T07:56:58.636-0800 [TID 0016292] [ERROR:] Backup error occurred during backup run: Error occurred reading the protected volume: Data error (cyclic redundancy check) (VOLUME_READ_ERROR_CRC)

The drive letter will be listed at the end of the "Begin copying volume" line that occurs just prior to the error. In this example, the failing drive is the E volume. If you are unable to identify the drive letter by searching the agent logs, contact Datto technical support for assistance

Common causes and solutions:

  • Check for disk errors by investigating the server's event logs at the time of the backup failure. Look for events containing the Volume GUID and/or the volume snapshot GUID for clues about disk errors.

    • If any are found, run chkdsk /r followed by a reboot of the system, in order to try and repair them.
    • If a chkdsk /r fails to resolve the errors, replacement of faulty hardware, may be required to repair the protected system.

  • A specific common cause are bad blocks on the volume - when this occurs you will see errors generated in the System event log, like this:

    Log Name: System
    Source: Disk
    Date: 5/14/2018 8:55:10 AM
    Event ID: 7
    Description:
    The device, \Device\Harddisk0\DR0, has a bad block.

  • If a chkdsk /r fails to resolve the bad block errors in the event log, you may need to do a chkdsk /B in order re-evaluate the bad clusters on the volume.

  • If the volume is using a RAID array, check the array's health, ensure that it is not in a degraded state, and resolve any errors the RAID controller reports.

  • Verify disk health of the volume using utilities like SMART.

Additional Resources