We're using veeam as a backup / replication product on a esxi 5.5 host, I don't think this is a veeam issue but related to CBT
I'm struggeling with the amount of changed blocks within a replication job. Simply put I’ve got to many servers replicating over a narrow WAN link. I’m trying to keep the changes to certain servers as low as possible.
I’m testing with a small (disk: 33gb) Windows 2003 server with hardly any activity. (searching the local disk for changed files in the last week resulted into a hand full of changes) Transferred data (not the actual read) from the last week,
before making changes:
10-4: 3.9gb
14-4: 1.6gb
15-4: 3.0gb
16-4: 2.6gb
I’ve changed a few things on the 17th:
- added HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\FileSystem NtfsDisableLastAccessUpdate 1
- Ran sdelete
- Ran disk clean up
- disabled shadow copies
- disabled all scheduled tasks scripts
- defrag
- vietool c: --generate (we are running symantec endpoint protection 12)
- reboot
The next replica was: 17-4 7.4gb (expected since quite a few blocks have been changed due to the defrag)
18-4: 3,4gb
19-4: 2,8gb
The above changes made no big differences. Are there more things which I can do to reduce the amount of changed blocks?
Is there a tool which shows me whichs blocks which files are linked to those blocks are changed so I can narrow my search?
Thanks!
