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

backup to disk until full-then start overwriting

Created: 03 Jul 2011 | 8 comments
gkeramidas's picture

i couldn't really find an answer, so i'm posting a new thread. this shouldn't be that difficult, so i don't know what i'm missing.

i had this working until i upgraded to backup exec 2010, now i can't get it to work.

i have 2 separate backup to disk folders, differential to disk and full to disk, on a mirrored pair of 2tb drives . i have 2 separate jobs that run, the full on sunday  and the differential the other 6 days.

 i want to use all of the disk space up before i start overwriting any files, but right now the jobs are overwritten every time they run.

what am i missing and what do i have set incorrectly.

my current settings:

job properties is set to append to media, overwrite if no appendable media is available
then media set properties settings are: overwrite protection is set to 0 hours, infinite - allow append

tools/options media management
overwrite protection is set to none
overwrite options is overwrite scratch media before overwriting recyclable media contained in the targeted media 

thanks

Gary

Comments 8 CommentsJump to latest comment

pkh's picture
You should set the Overwrite Protection Level to at least Partial.  Otherwise, your OPP will not work.
 
There is no way to achieve what you want, i.e. fill up your disk and then overwrite.  Suppose your backup is 300Gb.  After 6 backups, you would have 200GB left on your disk which means that it cannot hold another one of your backup.  What you want is for BE to start overwriting your older .bkf files, but it does not work this way.  BE does not and cannot know that your backup is 300Gb.  What it will do is to write until your disk is full and then fail your job due to a lack of space.  If you have set the low disk space threshold, then BE will write until the threshold and then alert you to clear space on your disk for it to continue.  You would have to manually delete some files on the disk to let BE complete the job.
 
Assuming that you are backing up on a daily basis and your backups is 300GB, the best that you can do is to set BE to overwrite recycleable media before scratch and set your OPP to 5 days.  This way there are 6 backups of 300Gb on your disk.  If your backup is of a different size or different backup cycle, then you would have to adjust the OPP accordingly.
gkeramidas's picture

pkh:

like i mentioned, it was working in a previous version. it would just start overwriting the oldest bkf files when needed. my full backups are around 150gb. i just don't know how it was set up, because i couldn't move my jobs from the old version.

pkh's picture

I don't know how you managed to do it, but what I have described earlier is the expected behaviour.

VGalati's picture

I think you want "Automatically optimize storage" option.  It's in Manage Backup Destination\Task\Settings. It will delete the old recovery points as the drive starts to get full.

gkeramidas's picture

vg:

i have backup exec 2010 r3 and i could not find this setting anywhere.

Gary

Ken Putnam's picture

job properties is set to append to media, overwrite if no appendable media is available
then media set properties settings are: overwrite protection is set to 0 hours, infinite - allow append

tools/options media management
overwrite protection is set to none
overwrite options is overwrite scratch media before overwriting recyclable media contained in the targeted media 

As pkh said, from Tools\Options\Media Management verify that the Global Overwrite Protection is set to PARTIAL or FULL 

From Tools\Options\Media Management verify that the option to "use scratch media  before overwriteable media in the target media set" is selected

Set all your disk jobs to Overwrite, not Append

Set the Media Set Overwrite to whatever you need for your business needs (as long as that plus one day will fit on your B2D folder)

And don't set the BKF file size up too far.  With Overwrite only, any unused space in the last BKF of a job is just wasted

If this response answers your concern, please mark it as a "solution"

gkeramidas's picture

hi ken:

i've set this up like you have both mentioned and i will post back the results in a few days.

thanks

Colin Weaver's picture

Just for reference GRT (IMG folder) backups and when they were reclaimed did work differently in older versions of Backup Exec. As we had a lot of feedback from customers that  had job failures based on disk space issues against GRT locations we had to change it.

In then old version it ran until you either filled the disk or hit the disk space reserve and then tried to reclaim an overwriteable IMG folder except the logic for trying using a full device vs the timeout for reclaiming was a bit flawed and customers ended up with failed jobs.

We changed it to reclaiming/erasing one overwritable IMG folder at the start of every GRT set (which means if a given job backs up two Exchange Storage Groups as each of these is a GRT set it will reclaim 2 IMG folders - one at the start of the first storage group and one at the start of the second) This change means that the reclaim occurs well before the drive actually runs out of space thus avoiding the device paused / full scenario.

You cannot change the behaviour back. Also standard BKF files remain as they always did - i.e. controlled by the Overwrite protection and therefore also not waiting till the disk fills up before removal.