02-02-2018 11:21 AM
Has anyone ever run into issues with there incremental jobs writing the same data as a full? I could understand if all the files were modified this would be the case. However, with this issue, it is not.
The backup source is a directory on a NetApp (Which is huge and takes BE several days to complete the full backup job. It's 10TB in size). The incremental was running normally up until the middle of last month. No changes were made to the BE server or the location being backed up.
Just now every time the incremental job kicks off it tries to write the full 10TB all over again.
Any help is appreciated.
Thanks,
02-02-2018 12:00 PM - edited 02-02-2018 12:01 PM
Double check all your clocks to make sure they match between all the NetApps and the Backup Exec server. I have heard of weird things when clocks & dates don't match.
02-02-2018 12:10 PM
Hi Larry,
I just checked our NetApp and BE server. Both times match.
02-02-2018 09:49 PM
If you are accessing your directory as a shared directory, then this is expected. For devices like your Netapp which does not have the BE agent loaded, incremental/differential backups are not supported.
If you attached your Netapp to the BE Server using iSCSI, then you can do incremental/differential backups.
02-12-2018 12:37 AM
to add to PKH's information - there is a further option of using NDMP Option Backups and configuring level based backups (which although slightly trickier to configure aand manage can run in an incremental manner) - As long as the NDMP device is supported for the NDMP option of course.
02-13-2018 07:51 AM
Go into your backup job and check which method it's using to determine if the file are modified. If it's set to modified time and doing this, there it probably something else that's hitting the files and changing the modified time. Alternatively you can try using archive bit if this is a Windows file system/share. If you switch to archive bit, you will need to run a full first to test if it works.
02-13-2018 11:29 PM
Do take note of this document
https://www.veritas.com/support/en_US/article.100004807.html
and Note 5 of this document
https://www.veritas.com/support/en_US/article.100019355.html
02-14-2018 12:18 AM - edited 02-14-2018 08:19 AM
As I just noticed that the second document posted by pkh was formatted a little oddly, in that the Note numbereing was out of sequence, I have arranged to get it edited. As such please be aware that Note 5 that pkh mentioned is now Note 3 (the note itself has only been renumbered and his point about support of incrementals over shares being unsupported remains the same)