Our Off-Site backup drives are changed daily ie:

The drive used on Monday will next be used the following week on Tuesday.

Resulting in non-sequential numbering of the images.

Due to the non-sequential numbering of the images the retention policy fails.

Off-Site backups setup as follows:

7 x USB drive caddies fitted with 160GB drives set to do a 'Full Image' of both the OS and Data drives.

Retention policy set to 2.

Capacity of the drives will support up to 4 OS and 4 Data images.

Is it possible to resolve this issue by looking at the dates of the images?

OR

Dose anyone have a script that can be run pos-snapshot which can be configured to delete all but the latest number of specified images?

Thanks - Qbit

This is a tricky one because when it comes to the retention policy you have to keep in mind that it's a destructive action, and so it has to be very selective about the files that it deletes.  If it makes liberal decisions about the files that it deletes then there are many possible scenarios where it would delete files which the user intended to preserve, which could be catastrophic.  Therefore the retention policy implementation is intentionally very conservative.  It only assumes that it can safely delete older backup sets if their names indicate that they are indeed, in order, the backups that should be deleted.  I can see in your situation how this would cause the retention policy to fail because an individual drive (one of 7) rotated in once a week would end up with a set of image files like this:

C_VOL-b001.spf

C_VOL-b008.spf

C_VOL-b015.spf

C_VOL-b022.spf

Each time a new image is created on the drive, the retention policy would look at the existing contents and if it could safely determine that the drive contained consecutive backups then it could clean up the older ones.  In your case however it will see some older backups for the same drive but we don't know if those backups were perhaps from another job or even the same job from an earlier time (perhaps before the job's schedule was modified).  The retention policy therefore takes the safest route and doesn't delete the older files.  In your case this isn't what you want.  You can work around this by writing a little script to be called as a pre or post backup command which script deletes the files that you want deleted.  I'd probably just write a vbscript for this.  Admittedly it's inconvenient to write a script, and I apologize for that.  We're working some really cool new features that will resolve issues like this.

Uh -oh......that's my situation exactly.  I'm a new reseller for StorageCraft, and I've just purchased & installed ShadowProtect Server Edition along with a Dell removable cartridge system for 5 of my clients.  The plan has been to rotate cartridges on a weekly basis, keeping one cartridge 'live' during the week and the other offsite for safe storage (following standard best practices for secured backup).  Now I understand why the retention policy is failing on all of these systems, and as a result, the backups aren't being performed.  I don't have any experience writing VB scripts, and wouldn't know how to create a script that would somehow figure out how to delete the oldest backups....and now I'm feeling extremely exposed.  Any help would be EXTREMELY appreciated.

Apprarently there's an augmented retention policy scheme due in the next release, but does anyone have any scripts that can fix this problem in the interim?

Would a simple script to delete *.spf older than date x do the trick?

 Im thinking if you had a 160 usb holding 4 images less than 40 each, you could go ahead and delete any files older than 3 weeks before running the backup, which would then just delete the oldest image. Crude, but could work.

 I did find a utility around that does this in a batch file quite easily. If anyone is interested I'll dig it up.

 Regards,

Chris

If i may sugest why not buy an eight USB hard drive and use this as a fixed Hard Drive so that all images are stored on this drive.

Then as a post backup command use a robocopy script wih the purge switch to synchronise the removable USB with the Fixed USB and then your whole rentention policy works as expected and each Offsite USB drive is a replica of the Fixed drive for each day.

The whole idea of this software is to have a fixed image repository that allows for the local recovery of files and servers.

I just write a simple dos bat file that renames the files and deletes the older ones

But I have a sneaky suspiscion that if you have a system drive and data drive in the one backup job any post/pre commands are run each time for each drive which throws everything out for the script, but i haven't actually verify it to be so, just a gut feeling so i run separate backup jobs for the system and data drives and depending on if you have enough room or not I either delete the oldest image in pre command or in the post command if theres plenty of room for the current and previous backups. and i just run the script on the last job only. But had a problem on on site where the two jobs were clashing and the script wasnt executing cause the last job was terminating. which left the drive full after a few weeks. once i spaced out the backup times its all good.

Anyway enough of my rambling

My script just does this

del c_backup5.spf (No this command could be run in the pre backup command to allow for more backup images on the drive)

ren c_backup4.spf backup5.spf

ren c_backup3.spf backup4.spf

ren c_backup2.spf backup3.spf

ren c_backup1.spf backup2.spf

ren c_vol*.spf backup1.spf

do the same for any other Hard drive volume image sets

So who needs vbscripts anyway