Link to home
Create AccountLog in
Avatar of youngslim
youngslim

asked on

Backup Exec 10.0 - Backup to Disk generating End of Media was Reached Error

I'm new to this office and am trouble shooting some errors that are occuring when we back up our Exchange information store. We currently have one Media Set whose properties are "Overwrite protection period = None" (this is greyed out and can't be changed) and an append period of 5 days.

We have one Backup to Disk Folder created and one Media Set created.

This week several jobs failed with a End of Media was Reached Error. I successfully completed a job about 3 days ago. I'm not sure why Backup Exec wouldn't create a new backup file (i.e. a new media) - iex B2D00000099.bkf - if there wasn't room to complete the job. Also, the maximum file size was set to 96GB and none of the files/media in the backup to disk folder were bigger than 41 GBs.

I'm not sure how to resolve this issue - would setting the append protection level to 12 hours cause Backup Exec to create a new file each time, then overwrite the oldest file once the NAS disk limit (125 GBs) was reached?

I realize that this isn't the optimal backup strategy, as I will only have three to five days of backups, but I want to figure out this problem first, before moving on to optimizing backup retention, adding brick by brick, etc.

Yours,
Avatar of honmapog
honmapog
Flag of Ireland image

Is the Exchange server a different server than the Backup Exec server?
If you expand the job log, is the "end of media reached" error during the verify phase?
If both answers are yes, the job may losing the connection to the remote agent during the backup phase, and then encountering the error you mention during the verify phase. In that case you should troubleshoot the error during the backup phase.

Also, check if you have a "Disk Space Reserve" set in the properties of the B2D device. Many people confuse disk space reserver for the amount of disk space Backup Exec is allowed to claim for itself. Instead, it is the opposite - it is the amount of disk space Backup Exec needs to keep free.
But I would expect a different error message if the disk space reserve value has been reached.
Avatar of youngslim
youngslim

ASKER

The Backup Exec Server is on the same machine as Exchange.

The end of media reached error is not in the verify phase.

The Disk space reserve is set to 5GB. Why would backup exec need 5GBs free? There are some "media" that are listed as not appendable (media full) with available capacity and total capacity of 5.GB.

The media - b2d000000033 - listed in the job that failed is listed as "as not appendable (media full) with available capacity 5.GB, and total capacity of 45gb and bytes written of 40gb. The allocated date was 2/26, the date of the job failure was 2/29.

Thoughts? Thanks
How much space is free on the complete disk at the moment? Backup Exec itself does not need 5GB free.
But, whoever set up BE thought it necessary that Backup Exec could not touch the last 5GB of free space on the disk. If you think that's not necessary you can drop disk space reserve down to 0GB.

What is the "Maximum size for backup-to-disk files" set to?

Check http://seer.entsupport.symantec.com/docs/292752.htm
Check all suggestions, but pay special attention to the last suggestion: delete b2d00000033.
I'm not sure if there is a reason to keep 5GB free, but i'm going to leave it alone until I know the NAS drive better.

There are 42 GBs free at this moment.
The Maximum size for backup to disk files is set to 96GB.
Those values look OK.

I didn't notice it in your orignal post, but when you say the NAS drive, I suppose Backup Exec's B2D device is on a disk on the network (at the other end of a network connection)?
In that case troublesome network communication to the NAS could be the cause.

To be sure: any chance of creating and testing a B2D device on a locally attached disk?
The device is locally attached, but has a nic card for me to manage it, so maybe it isn't technically NAS.
If you use network protocols to access the disk (SMB, iSCSI), check into the network side of things.
Otherwise try deleting the B2D*.bkf file as per the article.
Did i miss an article link? i.e. about deleting b2d files?

From what I can tell, there doesn't seem to be an issue on the network side, though never say never.
I just ran the job that failed last night right now, and it ran successfully. The local disk which contains the b2d folder now has 25 GBs free (makes sense, because the job wrote 18GBs). I'm thinking the issue must be in my append length of 5 days maybe? Tonight the backup will run again (full backup) and write another 18gbs, leaving just 7GB for tomorrow night, unless I've got a backup file older than 5 days so backup exec  can overwrite it?

My media set is set to have no overwrite protection and overwrite for five days. Again, I'd think that it would overwrite the oldest file before having a media end failure.

Thoughts and comments more than welcome. thanks

I know I can probably solve some problems by doing incremental backups along with a weekly full backup, but I'm trying to understand how it is working now before I start changing things.

-ys
ASKER CERTIFIED SOLUTION
Avatar of honmapog
honmapog
Flag of Ireland image

Link to home
membership
Create a free account to see this answer
Signing up is free and takes 30 seconds. No credit card required.
See answer
Oddly, when i right click the media set and choose properties, it says none for overwrite protection. It also says overwriteable in the overwrite protection column for each media. It says Overwrite protection period 5 days however in the properties window pane that is below the Media Tree window.

The job is setup for "append; overwrite if no appendable media is available."

I realize that the job seems to be acting odd - that it seems like it should overwrite the oldest file, since the settings seem to say that they are overwritable.

I'm going to read the articles you sent, and let the job run tonght and see what happens. I took a screen shot of the files that exist in that b2d folder so i can see what changes.

What do you think about changing the append to 3 days - to see if the shorter time period solves the problem. I even thought about making the append 12 hours. Then it should start a new file with each job, and overwrite the oldest files as needed.

I've just a thirty minutes left in the office, but can pick up again tomorrow. I appreciate all your help here so far.

-Pat

You can try lowering the append period - nothing against it.
Or you could lower the value of "Maximum size for backup-to-disk files". If you set it to for example 40GB, BE should skip to a new Backup to Disk media more often as well.
I was out of commission today. I will report back tomorrow. again, thanks for all the input.
You were very helpful. I didn't quite "solve" the problem, but you helped me get my arms around and understand better what was going on.
thanks
Part of the issue for me was to review how backup exec writes to (or creates new) files in a b2d folder. The ultimate solution for me might be to start using a full backup on Monday and incremental backups t-sat, to cut down on file size. I'm also going to try taking off the append option. I don't think for B2D files that makes sense. (Since I believe backup exec can just create a new file as needed, unlike creating a new physical tape.