VMware vDR error -2241 問題排除

VMware Data Recovery (VDR) 是vSphere V4.0 Enterprise版本以上附贈的備份工具，可以供100以下的Guest時備份使用。
但既然是附贈的，常常也會出包，最可怕的莫過於所有的備份還原點統統失蹤！也就是說先前做的備份統統無法正常存取，如下圖所示：

這時候先不要急著重做備份，因為VDR的備份是靠Store上的資料檔作索引。可能只是單純索引檔壞了，VDR無法找到適當的還原點資料。確認此問題的方法就是到VDR的log去看看是否有出現 Trouble reading from destination volume, error -2241 ( Destination index invalid/damaged)這樣的Log，若有的話，就可以依照這篇文章所摘錄的兩篇VMWare KB進行修復動作。

原文出自此篇官方KB
1. 手動強制進行Integrity Check，修復索引檔損壞部份

Running a manual integrity check on a VMware Data Recovery deduplication store

Purpose

This article provides instructions on running a manual integrity check on a VMware Data Recovery (VDR) deduplication store.

Resolution

To run a manual integrity check:

 

Note: This process takes longer for larger datastores or shares.

1. Connect to the VMware Data Recovery appliance in your vCenter client.
2. Click the Configuration tab and select destinations.
3. Select a deduplication store.
4. Select Integrity check in the top-right corner.
   
   Note: Alternatively, right-click on the deduplication store, and select Integrity check.
   
5. Select Yes to initiate a manual integrity check, when prompted.

You can monitor the progress of your integrity check under the Reports tab in running tasks.

Additional Information

Notes:

• A manual integrity check is required to remove damaged restore points on the deduplication store. Mark them for deletion prior to running your integrity check.
• During a manual integrity check, backup, and restore jobs are not run.

For more information, see the VMware Data Recovery Administration Guide.

原文出自此篇官方KB
2. 強迫重新建立索引檔(適用於方法1.失敗時)

VMware Data Recovery Integrity Check fails with the error: Trouble reading from destination volume, error -2241

Symptoms

• VMware Data Recovery Integrity Check fails.
• Backups do not complete.
• When running Integrity Check, you see the errors:
  
  9/6/2009 3:36:40 PM: Executing Integrity Check
  9/6/2009 3:36:40 PM: To Backup Set /...
  9/6/2009 3:36:40 PM: Trouble reading from destination volume, error -2241 ( Destination index invalid/damaged)
  9/6/2009 3:36:40 PM: Backup Set "/" will be locked until the restore point with errors are deleted and integrity check succeeds.
  9/6/2009 3:36:40 PM: Task incomplete

Resolution

This issue occurs if:

• There are bad restore points detected. Clear the bad restore points (mark them for deletion) and perform another integrity check.
• Lock (.lck) files are present on the destination datastore.

Note: For more information on performing an integrity check, see Running a manual integrity check on a VMware Data Recovery deduplication store (1029296).

To remove the lock files:

1. Log in to the VMware Data Recovery appliance through the console with the root username and password.
2. Change to the destination datastore and remove all .lck files from all destination datastores. For example, to remove the file store.lck file from destination SCSI-0\:1, run the commands:
   # cd /SCSI-0\:1/VMwareDataRecovery/BackupStore/# rm -rf store.lck
   
   Note: To find all *.lck files, run this command:# find / -name *.lck

原文出自此篇官方KB
此問題發生原因之解釋：

Symptoms

You are experiencing these issues with VMware Data Recovery (VDR):

• Cannot perform the backup
• Backup fails
• You see the error:
  
  failed, error -2241 (Catalog File invalid/damaged)

Resolution

This issue occurs if the VDR catalog files are corrupt.

To resolve the issue:

1. Recreate the catalog files:
   1. Verify that there are no operations running on the appliance.
   2. Delete these files:
      
      Note: The restore points will be temporarily deleted.
      
      • /{dedupe mount point}/VMwareDataRecovery/BackupStore.cat
      • /{dedupe mount point}/VMwareDataRecovery/BackupStore.cat.bak
        
   3. Run an integrity check on the destination directory to recreate the catalog files. The integrity check will rebuild the BackupStore.cat files, and repopulate VDR with all restore points.
      
      Note: The integrity check fails because the catalog files have just been deleted. This is an expected behavior. However, the catalog files will be created in the background, which might take a long time. You can monitor the rebuilding process by running the following command in the VDR command line:
      
      watch -n1 -d "ls -lt /{dedupe mount point}/VMwareDataRecovery/".
      
      If you cannot observe any progress, it is most likely due to the re-catalogue process not starting. This may be caused due to locks on the deduplication store. For more information, see VMware Data Recovery Integrity Check fails with the error: Trouble reading from destination volume, error -2241 (1018060).
      
2. Ensure that the VDR appliance does not have any hot-added vDisks from other virtual machines attached to it. To verify this, click Edit Settings in the VDR appliance.
3. Upgrade to the latest version of VDR. For more information, see VMware Data Recovery Documentation.

完整除錯KB可參考底下官方文件：

VMware KB: Troubleshooting VMware Data Recovery restore failures