BE 15 failing on GRT VM back up. (out of space but there is still space)

Firstly you are still backing up the VHD with GRT disabled it is just that if is disabled the VHD potentially ends up inide a BKF file and if you use GRT the VHDs have to be directly accessible on the disk hence you see them inside the IMG folder . Which brings me to a second point. Software Compression (and Encryption) don't really apply to GRT backups on disk, as  the VHD has to be directly mountable to get at the GRT data. So potentially a non-GRT backup that creates a BKF file will result in less disks space used than one that creates VHD copies (Note: As far as I know, both if them should not include the unused white space inside the VHD)

Now assuming you change your 'tape' (disk cartridge) daily, can I ask how you set the backup job definitions up? If you used separate job definitions per day instaed of a job that repeats daily then you will be creating a last recovery chain for every day of the week and not one just against the last day's backup and this will hinder DLM reclaims and means that you would need an RDX cartidge that has slightly more than double the capacity that a single full backup requires for your strategy to be successful. The way to fix this is to use a single backup definition that repeats daily and just change the cartridges

Are you doing full backups for the entire week or incremental/differential backups on some days?  If it is the latter, then you need to put the full + incremental/differential backups on one disk.

Hi All,

Thanks for the speedy reply. This is why I love a good forum.

So to cover some points off the top of my head.

I am doing a full back up everyday (mon-fri) with one Job.

I have GRT enabled (but none of the other GRT options like exchange)

I am backing up the host as well but have the folders that contain the VHD's set to ignore/skip

I have come in this morning to find it still backing up. very odd. I am waiting for it to finish (and presumably fail) and then I will grab all my job definitions.

Attachted screen shots of job settings. Wasnt sure the best way to show them in the thread. I have removed any domain names and such.

ignore that I have c:\Virtual disks ticked in one of those screen shots I ticked by mistake when expanding the lists for the screen shots.

Hmm well I can't see anything obviously wrong with your job settings

What should happen is as long as the existing backup full backup on the carttidge is expired and not the last full backup (and not linked to any incremental backups that are not expired) then at the start of the next job to write to the RDX cartridge it should reclaim (delete) the full backup giving you all the space back.  If you kept teh cartidge detached for longer than 2 weeks I think it would be marked as Read Only  but you are rotating against 1 week so this should not be the issue.

For a completely different reason I have been testing RDX cartridges and reclaims today, although not with a Hyper-V GRT backup set and my full backups have been reclaiming at the start of the job without issue. Note I am using FP3 for BE 15

You may need to log a formal support case for us to look at it in more detail

@paranoidsystems - I've had a case open with Veritas Support for 3 months for this exact same issue you are having.  I am hopeful that maybe you can make some progress here.  

Quite honestly, my company is ready to throw BE out the door and go with something else because the support is completely useless.  I actually worked for Symantec and supported another Veritas product up until last year and I can't even tell you how boggled I am at their complete ineptitude in supporting BE through their outsourced partner.  I'd have been fired for doing what they are doing.

In any case, to make this somewhat useful, I have a Windows Server 2008 Hyper-V server that I am backing up the guests with BE.  The scheduled backups complete except for the last one which fails because BE thinks there isn't enough room.  The particular Hyper-V guest, a file server, has 3xVHDs and two backup successfully, but the third always fails because of disk space dispite having more than enough room for it.  The VHD is roughly 40GB and the remaining space on the drive is over 300GB.

If I run the backup of the file server first, it completes without error but then other backups fails for other reasons.

The drives are changed out every day so the backup is always going to a wiped drive.

Again, I hope that you can get some assistance here because I'd really like to get the backup working and support has been worthless.

Hi Colin,

It's very strange. as a test yesterday I set my backups to only the guest VM's and nothing from the host. I have come in this morning and the job is still running on backing up my last VM. it says it is running but clearly it is not. the RDX cartridge is full and it has not deleted last weeks back up but has also not thrown an error so it's basically just sitting there doing nothing but thinking it is. screen shot attached. I have sat watching that screen do nothing since 9:00. there is litteraly now data transfer taking place on that RDX now now.

What I might try is wiping out tomorrows backup tape completely (is there a best practise for this? or do I just use erase tape function in BE) and seeing what that does tonight.

I am running BE 15 FP3 but this started happening before FP3 was installed (hoping it would resolve my issue maybe).

@beyond-done after tonights test I will try and swap the order the VM's in the backup job and see what happens for me.

If you can you should try and create another VHD for you file server. something incidental like 500mb and see if then it backup up everything but that final test VHD. not a solution but interesting proff of concept.

So erasing the tape first made no difference. I still get the failure even though the VHD is there.

Again the RDX has plenty of space. (422gb) and both VM's are back-up'd. I will open a ticket but will update this thread once a solution is discovered.

Is there a way to check for any flags on the media? for example if I had a cartridge that hadn't been used for 2 weeks I get that data is marked as read only but how could tell if any of these flags had been applied to media? Perhaps there is some sort of flag I am not aware of? I just don’t get how this works for months then stopped. The moment I turn GRT off then it all works (even though it's taking up the same space on the cartridge)

the 3 errors I still get are

Job ended: 20 January 2016 at 07:45:30
Completed status: Failed
Final error: 0xe00084f9 - A communications failure has occurred between the Backup Exec job engine and the Agent for Windows.
Final error category: Resource Errors

For additional information regarding this error refer to link V-79-57344-34041

The selections for this job have changed since the last time it ran.

Can't even raise a ticket. 

get as far as entering all the details.....

and have a choice of two buttons.... I've found an answer! or I still need help.

the I still need help button does nothing in any browser I have tried (presumiung as it doesnt like the version number I have entered?). The Veritas Support link goes to a 404. and I've found an answer! takes you back to the start.

As I have tried 3 different browers now this cannot be local to me? And a 404 on the support page is just crazy. Also the fact I can't actually select the product and version I have correctly is crazy.

Ok, Raised a support ticket and got a call back. Basically it seems to boild down to me needing 2tb RDX's to back up 490gb full jobs daily?

I am really unsure what the deal is now. I might just turn off GRT and live with it.