13.5. Checking the consistency of a virtual disk

Procedure

1. Use the qemu-img check command on the image you want to test. For example:

   # qemu-img check <test-name.qcow2>
   
   No errors were found on the image.
   327434/327680 = 99.92% allocated, 0.00% fragmented, 0.00% compressed clusters
   Image end offset: 21478375424

   If the check finds problems on the disk image, the output of the command looks similar to the following:

   167 errors were found on the image.
   Data may be corrupted, or further writes to the image may corrupt it.
   
   453368 leaked clusters were found on the image.
   This means waste of disk space, but no harm to data.
   
   259 internal errors have occurred during the check.
   Image end offset: 21478375424

2. To attempt repairing the detected issues, use the qemu-img check command with the -r all option. Note, however, that this might fix only some of the problems.

   警告

   Repairing the disk image can cause data corruption or other issues. Back up the disk image before attempting the repair.

   # qemu-img check -r all <test-name.qcow2>
   
   [...]
   122 errors were found on the image.
   Data may be corrupted, or further writes to the image may corrupt it.
   
   250 internal errors have occurred during the check.
   Image end offset: 27071414272

   This output indicates the number of problems found on the disk image after the repair.

3. If further disk image repairs are required, you can use various libguestfs tools in the guestfish shell.