cancel
Showing results for 
Search instead for 
Did you mean: 

vSpehere 4.1 & NB 7.01: Blocklevel incremental writing as much data as full backup

SvenKessler
Level 4

Hello,

we have several servers doing backups with "Blocklevel incremental" using mapped full vm as backup method. All works fine, but one server has a problem since the last full backup last weekend.

The last two incremental backups wrote exactly as much data as the full backup to tape. It should normally not be more than 10 % of the full backup data.

If is open the "Backup, Archive and Restore" window i can see that on file level it is only a incremental backup. There are only changed files visible so this is normal. But why is the backup data on tape as much as a full backup? Also the backup time is as long as a full backup. Last week everything works fine.

 

Best regards

 

Sven

5 REPLIES 5

RiaanBadenhorst
Moderator
Moderator
Partner    VIP    Accredited Certified

Hi,

 

Is this vm in a policy by itself, did you change anything on that policy, any settings, renamed it?

 

If you change settings on a policy it basically becomes new, so if an incremental is scheduled to run, it will actually do a full.

 

Or did you maybe do any changes to the permissions on that client? The account used to access vCenter needs the role/rights as described in this note (which you probably have, since the rest are working)

 

http://www.symantec.com/business/support/index?page=content&id=TECH129738

SvenKessler
Level 4

Hello,

no, i didn't changed anything. The policy was untouched and the client also.

Best regards

Sven

Nicolai
Moderator
Moderator
Partner    VIP   

I am not sure this is the reason, but did some one turn off change block tracking ?. It seems you need to turn it on a disk per disk.

http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=1020128

SvenKessler
Level 4
Thank you for showing me the right direction. It seems that I was a little to fast saying nothing has changed. I talked again with our vmware admin and he remembered that he had changed the virtual scsi controller type last weekend. He also added two more scsi controllers so that every virtual hdd has now it's own controller. A test server showed that this was the problem. Now we have to look if the new controller type isn't compatible to block tracking or if block tracking won't work with more than one controller. But we know what caused the problem and that's the most important thing. Thank you very much. Sven

Nicolai
Moderator
Moderator
Partner    VIP   

Thanks for posting the root cause  - Very usefull pice of information.  I will bring it to my VMware admin to make sure they know.

best regards

Nicolai