CRC Error

[jcpro]

Hi,

My backups haven't been completing successfully for a few days now and I've been getting some errors that indicate a CRC error. I have run chkdsk against all volumes and not found any issues. I have run vssadmin list shadows /for=c: and cannot even find the "HarddiskVolumeShadowCopy13" that it is referring to. How can I troubleshoot this?

Code: Select all

18/07/2016 10:12:55 AM :: Error: Data error (cyclic redundancy check). Failed to read data from the file [\\?\GLOBALROOT\Device\HarddiskVolumeShadowCopy13]. The block at offset 89308266496 (length 1048576) is corrupt. Failed to upload disk. Agent failed to process method {DataTransfer.SyncDisk}. Exception from server: Data error (cyclic redundancy check). Failed to read data from the file [\\?\GLOBALROOT\Device\HarddiskVolumeShadowCopy13]. The block at offset 89308266496 (length 1048576) is corrupt. Unable to ret

[Dima P.]

Hi jcpro,

Unfortunately, it means you have some bad blocks on your source disk.

We have an experimental 'skip bad blocks' feature that you may want to try. Create the SkipBadBlocks (DWORD) = 1 registry value under the HKLM\SOFTWARE\Veeam\Veeam Endpoint Backup and re run the backup, then it should backup the volume passing by the corrupted blocks (files residing these blocks might be corrupted). Kindly, keep in mind that this does not solve the problem with bad blocks on your disk – it just allows creating a backup bypassing corrupted blocks.

[jcpro]

Is there any way to identify which source disk it's referring to? HarddiskVolumeShadowCopy13 isn't exactly descriptive and I can't find any reference to that when I list all Shadow Copies.
Also, how is Veeam able to identify a bad block but chkdsk reports all is well?

[Dima P.]

The error you got should be next to the line mentioning the specific volume being backed up. Check disk at some point may miss corrupted blocks, while VEB actually reads entire volume during backup, thus recognize the corruption.

[amoiseev]

Hello everyone

We have a problem with backing up a Windows Server 2008.

We can normally backup drives E and D, but if we try to backup drive C, we have the following error.

Code: Select all

 Error: Datenfehler (CRC-Prüfung). Error code: 23 Failed to read data from the file [\\?\GLOBALROOT\Device\HarddiskVolumeShadowCopy55]. The block at offset 210180751360 (length 1048576) is corrupt. Failed to upload disk. Agent failed to process method {DataTransfer.SyncDisk}. Exception from server: Datenfehler (CRC-Prüfung). Error code: 23 Failed to read data from the file [\\?\GLOBALROOT\Device\HarddiskVolumeShadowCopy55]. The block at offset 210180751360 (length 1048576) is corrupt. Una

The server is configured to use RAID 5 for all 3 logical drives, therefore, I assume that there is no corruption on the physical level.

[Dima P.]

Hi Anton,

Looks like you have a bad block allocated somewhere on volume C. You still can backup this volume using the regkey solution provided above but I’d strongly recommend you to check the said hard drive integrity.

[Bruce]

Hi Dima,

Hi jcpro,

Unfortunately, it means you have some bad blocks on your source disk.

We have an experimental 'skip bad blocks' feature that you may want to try. Create the SkipBadBlocks (DWORD) = 1 registry value under the HKLM\SOFTWARE\Veeam\Veeam Endpoint Backup and re run the backup, then it should backup the volume passing by the corrupted blocks (files residing these blocks might be corrupted). Kindly, keep in mind that this does not solve the problem with bad blocks on your disk – it just allows creating a backup bypassing corrupted blocks.

Would the use of the SkipBadBlocks registry entry in any way detract from the accuracy of the backup? In other words, could I use the backup file to overwrite the bad blocks during restore or would the restore simply reproduce the original problem?

Thank you

[Vitaliy S.]

If you plan to restore to the same faulty disk (hardware problem), then "skip bad blocks" option will not help. Also, since bad blocks that are skipped might contain some data, it will not be possible to perform restore without data loss.

[dadrianto]

Hi All,

does anyone in this forum have facing error like this?

Error: The request could not be performed because of an I/O device error. Failed to read data from the file [\\?\GLOBALROOT\Device\HarddiskVolumeShadowCopy7]. The block at offset 42991616 (length 1048576) is corrupt. Failed to upload disk. Agent failed to process method {DataTransfer.SyncDisk}. Exception from server: The request could not be performed because of an I/O device error. Failed to read data from the file [\\?\GLOBALROOT\Device\HarddiskVolumeShadowCopy7]. The block at offset 42

i have installed some notebook and perform entire backup, but one of the machine have error.
i already re-install veeam endpoint, but the issue still occurs.

anyone please tell me how to solve this error?

thanks,
dadrianto

[Dima P.]

dadrianto,

This is not something we can resolve - error code shows that your hard disk has some bad block which we cannot backup. As stated above we can skip bad blocks via specific kegkey but that would not solve the issue with your hardware.

Create the SkipBadBlocks (DWORD) = 1 registry value under the HKLM\SOFTWARE\Veeam\Veeam Endpoint Backup and re run the backup, then it should backup the volume passing by the corrupted blocks.

[dadrianto]

Hii Dima P,

I have created the DWORD Registry, the backup job running normaly until the error occurs "Error: ChannelError: ConnectionAborted"
is the error occurs cause the instability network connection ?

[Dima P.]

Hi Danang,

Need to check the application logs. Check with the support team please.

[dadrianto]

Hi Dima,

Create the SkipBadBlocks (DWORD) = 1 registry value under the HKLM\SOFTWARE\Veeam\Veeam Endpoint Backup and re run the backup, then it should backup the volume passing by the corrupted blocks.

I try your solution on two notebook, one notebook can backed up successful with warning "One or more bad blocks were detected and skipped."
thanks Dima

[Dima P.]

Danang,

You are welcome. Keep in mind that this workaround does not solve the hardware issue, it just allows to skip the corrupted blocks.

[techgirl]

I also have this error.

Code: Select all

2017-02-19 1:25:42 AM :: Error: Data error (cyclic redundancy check). Failed to read data from the file [\\?\GLOBALROOT\Device\HarddiskVolumeShadowCopy195]. The block at offset 9293529088 (length 1048576) is corrupt. Failed to upload disk. Agent failed to process method {DataTransfer.SyncDisk}. Exception from server: Data error (cyclic redundancy check). Failed to read data from the file [\\?\GLOBALROOT\Device\HarddiskVolumeShadowCopy195]. The block at offset 9293529088 (length 1048576) is corrupt. Unable to ret 

This error refers to the source (computer), right, and not the target backup device?

I wanted to back up my documents before running checkdisk in case checkdisk mangled anything. I created a new folder on the NAS and changed backup from "Entire Computer" to "File Level" and selected the 2 /Program File/ folders, /Program Data/ and /Users/ - this backed up OK.

Windows 10 - I ran Windows checkdisk which reported no errors, then defrag and checkdisk again which again found no issues.
QNAP 2 disks (mirror RAID) - SMART info is all good for both, I am currently in the process of running complete test on those disks nonetheless

After running checkdisk, I tried "Entire Computer" again, with the same CRC error. So something in the Windows system files is corrupt?

Is the error text accurate [truly bad/corrupt] or can it be something else (NAS is 'low' on space currently, my C; is 200GB, NAS had 466GB) or a permission error? Since the error references VolumeShadowCopy195 - if I make Windows delete that numbered shadowcopy or some or all other restore points, could that fix it? Does to have to do with space alotted to restore points? Is Veaam absolutely only reading in there or causing any writes in there?

Is there another utility you may suggest that could confirm bad blocks since Windows itself isn't detecting any issue ?

[Dima P.]

This error refers to the source (computer), right, and not the target backup device?

Correct.

I ran Windows checkdisk which reported no errors

Did you try CHKDSK VOLUME: /F /R ?

Is the error text accurate [truly bad/corrupt]

Unfortunately its true. Try the regkey suggested in this thread to backup the affected volume.

[techgirl]

UPDATE: Thank you Dima, yes, neither chkdisk F nor R report any issue - no errors. DISM found no issue. sfc /scnannow - nothing. Acer's diagnostics found no error. I have since tried Seagate Seatools on it, which did find an error but could not repair it. Then I have noticed in Windows Event Viewer ->Windows Logs->System: "The device, \Device\Harddisk0\DR0, has a bad block." This error was there starting exactly when Veeam started failing. EVERY windows tool or diagnostic shows the drives "healthy" and no errors except 'event viewer'! I am thankful that Veeam brought error this to my attention!

Asus wants to do a full wipe and restore, so I have done file level backup of my content as a safeguard, I am now off to tweak that registry key to force an "entire computer" backup to complete.

[Dima P.]

techgirl,

Most of third party check disk utilities simply reply on built-in windows checkdisk, that why they may show the same result. There is a tool called Victoria, our support team uses it from time to time and it’s very precise it terms of bad block detection.

[jezi55]

Hello,

Currently run Veeam Backup & Replication and Veeam Endpoint Protection to the same NAS box backup storage. The other day one of the hard drives failed in my finance server (raid 5) so replaced the drive everything was 'green' and the raid rebuilded fine. When i try to run Veeam endpoint on my server it copies about 40gb and then comes up with the following message

Code: Select all

"06/04/2017 19:59:41 :: Error: Data error (cyclic redundancy check). Failed to read data from the file [\\?\GLOBALROOT\Device\HarddiskVolumeShadowCopy7]. The block at offset 122037469184 (length 1048576) is corrupt. Failed to upload disk. Agent failed to process method {DataTransfer.SyncDisk}. Exception from server: Data error (cyclic redundancy check). Failed to read data from the file [\\?\GLOBALROOT\Device\HarddiskVolumeShadowCopy7]. The block at offset 122037469184 (length 1048576) is corrupt. "

I have tried doing a sfc and a checkdisk but no bad blocks are present, any ideas how i can fix this?

It is running on Server 2008 R2/Endpoint 1.5.0.306

Many thanks,

Dex

[Dima P.]

Hi Dex,

Unfortunately, this error refers to hard disk corruption. Kindly, review this thread and specifically this post.

[lrosa]

I just upgraded to 2.0 RTM on Windows 10 Pro Creators Update Release 1703. I choose not to install the license and proceed with free version.

I created a new full backup on a SMB share.

At the end of the job the log says

[Alert yellow icon] One or more bad blocks were detected and skipped
[Check green icon] Finalizing
[Alert yellow icon] Processing finished with warning at.....

My laptop has a 1Tb Samsung 840 EVO SSD and the Samsung diagnostic does not report any issue.

Is there a way to have more information about that warning message? I checked the logs in %ProgramData%\Veeam\Endpoint\Backup_Job_COMPUTERNME but I am not been able to find any reference to bad sectors.

luigi

[Vitaliy S.]

Please review this topic for some tips on how to detect these blocks, start with Windows Event log (for example).

[damon.theis]

I began receiving this error on a new windows 10 Lenovo pc last week that had previously backed up fine every night.

Code: Select all

Error: Data error (cyclic redundancy check). Failed to read data from the file [\\?\GLOBALROOT\Device\HarddiskVolumeShadowCopy4]. The block at offset 112056074240 (length 1048576) is corrupt. Failed to upload disk. Agent failed to process method {DataTransfer.SyncDisk}. Exception from server: Data error (cyclic redundancy check). Failed to read data from the file [\\?\GLOBALROOT\Device\HarddiskVolumeShadowCopy4]. The block at offset 112056074240 (length 1048576) is corrupt. Unable to ret

Windows reports disk is error free and not fragmented.

[Gostev]

Definitely not error free, as far as I can see

[Dima P.]

Damon,

To back this volume up use a regkey described in this post. By the way, in VAW 2.0 RTM code option to skip bad blocks is enabled by default, so worth checking.

[lrosa]

Please review this topic for some tips on how to detect these blocks, start with Windows Event log (for example).

i did a checkdisk then an incremental backup: same bad block skipped warning.

Event viewer reports the warning and nothing else

I grepped the word error in the Veeam logs and the only line I found is this in JobBackup_Job_COMPUTERNAME Backup.log:

<01> Error IS REPO RESCAN = FALSE

[Vitaliy S.]

Do you have any other native tools to check the data blocks? Maybe the post above made by techgirl might help too.

[lrosa]

The probles is exactly this: there are no disk related events in the System log.

Samsung Magician is enabled and says that the disk is good. S.M.A.R.T. metrics collected by Magician are all OK

lr

[Vitaliy S.]

Well... then you can try to ignore this message or contact our technical team to investigate this further.

[majones]

I got the bad blocks error with VEB so used "SkipBadBlocks (DWORD) = 1" edit in the registry and the backup then ran OK but with a warning. I then restored the backup to a new disk (without bad blocks) and did a backup of that disk, but still got a warning that there were bad blocks. Is this normal? - ie the restored backup retains the bad block characteristics of the original?