Video Screencast Help
Symantec to Separate Into Two Focused, Industry-Leading Technology Companies. Learn more.

B2D2T - last x days to backup

Created: 19 Mar 2013 • Updated: 10 Apr 2013 | 5 comments
This issue has been solved. See solution.

Hello,

This is probably really easy for you experts.

I have a 5Tb E: drive on the Backup server.  Backupexec 2010 R3 is installed.

There are 20ish individual server backups that run from 16:00 - 04:00 - backing up to their own folder with E:\Backup.  (e.g. E:\Backup\SERVER1).  This runs fine.  Servers are a mix of physical and VM's, but this shouldn't matter.

Backup jobs are set to hold the last 2-3 backup copies of each server on disk for ease of restoration (acknowledged that this retention period is short).  So in essence, each servers folder on E: has 3 backup copies Day1, Day2 & Day3.  Say the backup for SERVER1 is 40Gb, it only ever uses upto 120Gb on the disk as overwrites take place.  This works fine.

I now wast to backup ONLY LAST NIGHTS (Day3) backup copy to tape.  Tapes are changed daily and removed from site.  It is preferred that FULL backups are taken.  I've tried backing up the E: drive using the 'General' setting of 'Backup Method for files = Working set - last accessed in x Days" and set this to 1 day.

My plan was to send just files changed in the last 24 hours to tape.  But this isn't happening.  Checking the job monitor, the total size of last nights D2D backup was 1.1Tb.  So I should only be sending 1.1Tb to tape.  what is happening is I am sending the whole E: drive to tape, some 3.3Gb - which fills the tape drive.

Is there a setting for this - or will I have to create a set of subfolders per server (MTWTF) and set some additional backup jobs and BTD locations?

Cheers in advance

Tony

Operating Systems:

Comments 5 CommentsJump to latest comment

Colin Weaver's picture

Do not flat file copy disk storage locations it is not a good idea

Please review this for why:

http://www.symantec.com/business/support/index?page=content&id=TECH176061

Run duplicate jobs in a policy linked to the original jobs to disk

Note: duplicate jobs do have advantages - if you flat file copy you have to restore the files to the media server, then sort out any problems relating to the above article then do another restore to the server that originally owned that data. Most types of duplicate job can be restored directly from the duplicate to the original servers/locations

SOLUTION
CraigV's picture

Hi,

Just add in a duplicate-to-tape job. This will duplicate the just-completed job's data that was backed up to tape.

Don't backup the data directly as it is going to mess up your catalogs. If you have a policy, then it's simple to go into the policy and then add in the duplicate job.

Thanks!

Alternative ways to access Backup Exec Technical Support:

https://www-secure.symantec.com/connect/blogs/alte...

AstonT's picture

Thanks Colin/Craig.  This is useful info. :)

I'm having a play with duplicate jobs - I hope I don't have to express my ignorance on you guys again!

Cheers

Tony

AstonT's picture

Update.

So far I'm running some test Duplicate jobs (to another disk in the interim).

These Duplicate jobs run immediately after the original backup job - I was hoping I would be able to schedule these for a specific time.  My backup to disk jobs start from about 16:00 and run until about 04:00 (having timed all the jobs so that they don't overlap).  I then have a backup to tape job backing up the E:\Backup folder, which finishes at about midday.  This gives the end user a 4 hour window to swap tapes (single LTO5 tape drive).

With these duplicate jobs, I would need the tape inserted no later than 16:00 as the first DuplicateToTape job would kick off straight after the first BTD job.  My concern is, if I do change all the Duplicate jobs to backup to tape (and not the interim disk location), it will cause overlapping of jobs which slows down disk writing.  Whilst I can amend backup times, I'm worried about the backup window.

I may have to rethink the whole thing here and have just one BTD job that details all servers, with the Duplicate to Tape running straight after.  The nice thing about separate jobs per server is when a backup fails or completes with errors, it is easier to see and rectify the failty job/server.  I guess that is a trade off I'll need to weigh up..

AstonT's picture

Thanks for your help Colin and Craig.

I have managed to solve the issue by creating a single backup job that contains all servers to backup.  Then by adding a duplicate job to tape has achieved my goal.  I would have liked to have kept the individual backup jobs, but this was not an option for me.

Cheers - this call can be closed now