Resolve system disk errors on a Windows instance-阿里云帮助中心

This topic describes how to resolve system disk errors on a Windows instance.

Background

System disk errors on a Windows instance can prevent patch and driver installations, and may even stop the instance from starting.

Symptoms

You can run the chkdsk or sfc /VERIFYONLY command in Command Prompt to check the disk status.

If the disk has errors, the command output includes error messages. For example, if system files are corrupted, the output is similar to the following:

C:\Windows\system32>chkdsk
The type of the file system is NTFS.
WARNING! F parameter not specified.
Running CHKDSK in read-only mode.
Stage 1: Examining basic file system structure...
Attribute record (80, "") from file record segment 1C078
is corrupt.
Attribute record (80, "") from file record segment 1C079
is corrupt.
Attribute record (80, "") from file record segment 1C07B
is corrupt.
Attribute record (80, "") from file record segment 3C762
is corrupt.
289280 file records processed.
File verification completed.
  6224 large file records processed.
Errors found. CHKDSK cannot continue in read-only mode.
C:\Windows\system32>

Solution

Follow these steps to manually repair the system disk.

Connect to the ECS instance.

For more information, see Choose a connection method.
Open Command Prompt as an administrator.
Run the following command to scan and verify the integrity of all protected system files.
```
sfc /scannow
```
If the command output reports no integrity violations, the system file integrity is verified.
Run the following command to check the system disk for errors.
```
chkdsk c:
```
If the command output reports no problems, the system disk is healthy.
Restart the instance in the ECS console.

For more information, see Restart an instance.

If the manual repair fails, you can roll back the system disk using a snapshot. For more information, see Roll back a cloud disk by using a snapshot.