Advice on backup method

[Moggyman]

Hi

I want to have a strategy whereby a disk-based full backup of my Exchange server, SQL server and data server is taken every Sunday PM and during the week incremental backups are appended to it.  After the last incremental runs on Saturday PM I want some means of 'archiving' off the full+incs backup for that week to another disk based folder. The aim (in theory) is to always keep the previous week's backup - fine if you have 2 tapes and use a tape drive to alternate between the tapes but I've been asked to implement this backing up to a NAS RAID box. Don't ask.

When the next full backup runs on the following Sunday it should delete/replace the files in the 'backup' folder so a new full+incs backup is started for the second week.  When archiving time comes around again after the final incremental backup on Saturday I want the contents of the 'archive' folder to be replaced with the latest data from the backup folder.

I find BExec a little confusing in the way it handles incrementals and BKF files, overwrite protection etc.  Now don't laugh - at the moment I've got the weekly full backup job set to Overwrite which I'm hoping will clear out the data from the backup folder and continue afresh.  The incrementals are set to Append.  I've then got another backup-to-disk target folder created and on Sunday AM another backup job (set to Overwrite) backs up the data in the backup folder and puts it in the 'previous week' folder.

Is there a more sensible/simple way to achieve my aim?

Thanks 

[pkh]

B2D is similar to tape in many ways, but also dissimilar in many ways.

When you say "backup to last weeks folder", I hope you didn't just back up the current folder just like any other folder.  This is the wrong way to backup a B2D folder.   If you want to backup a B2D folder, you would need to use a duplicate job to duplicate the backups.

Am I correct to assume that your "last week's" folder is also on the same box?  If this is so, you can simplify things by having just one folder.  If you want to have two weeks worth of data, then set the OPP to 13 days.  With this setting, when it comes to the second week, the first week's data would still be write protected and BE will create new .bkf files to keep the second week's data.  On the third week, the first week's media are recycleable and can be re-used.  Hence at any one time, you have 2 weeks worth of data on your disk and you don't have to do any copying.

All your jobs including incrementals should specify overwrite.  For B2D, append tends to complicate things. Set your AP to 1 hour.

You should also go to Tools ---> Options ---> Media Management and set the option to use recycleable media in the targeted media set before using scratch.

[Moggyman]

Thanks for the response - you'll have to excuse me.  I'm used to doing Full backups to tape so all this incremental stuff is quite new to me and I'm not fully up to speed on all the acronyms and where the options are as I never needed to bother before now.

I realised that doing a plain backup of the backup folder to an archive folder was a bit dumb but luckily I'm just testing at the moment while the full tape backup continues as per normal in the background.

Yes the folder for the previous week's data is on the same NAS RAID box.  Doing it in one folder sounds like the way to go but I'm not 100% on what you advise.  I take it that 'OPP' is overwrite protection period?  I've created a new media set and specified the OPP to be 13 days - what about the Append period?  Should that be the same or left as is is now (which is default None-Infinite Allow Append)?

And the weekly-full/daily-incr settings within the Policy should both be set to overwrite? Does this mean that when a full backup runs and creates a file 123.bkf that the next incremental will overwrite that file?  Sorry if it's a dumb question but I've probably confused myself with creating jobs when I'm not fully aware of what I should see in the backup folder at the end of it.  For example, I ran my test Full backup yesterday which created a 75Gb file B2D00043.bkf (size is in keeping with what gets backed up to tape).  My daily incremental backup ran overnight and claims it backed up just over 4Gb using media B2D00043.  The actual file size of B2D00043 this morning is 45Gb so I have no idea what happened!  I guess I need to tweak settings somewhere?

Sorry (I know I'll kick myself) but what is AP and where should it be set?  I've changed the Media Management setting as you advised.

Thanks for your help.

[Moggyman]

One further thing that I hope you can help me with.  When the current backups run (full or incr) I get a new folder each time created in the root of my B2D folder called IMG00001, IMG0002 etc which appears to contain Exchange data captured from my Exchange server which is part of the job.  How should I manage these when I get the 2 week backup strategy in place?  Do they have to stay there or can I just manually delete them?

Also, when it comes to use this system in anger and I want to clear out the B2D folder ready for action can I just delete any bkf files etc in there or do they need to be somehow removed by going through BE?

Many thanks

[Moggyman]

Please ignore the second part of my last post - got that sussed I think.

[pkh]

We use acronyms a lot in this forum.  If you are not sure what they are, check the Glossary diiscussion which I have started.  It it pinned to the top of the forum.

AP (append period) is a property of a media set.  On the Media tab, right-click on media set and you should see it.  Since you don't want any append you should set it to the minimum period, i.e. 1 hour.  It sounds like your incremental job is set to append.  You should change it it overwrite.

In the scenario that you have given.  123.bkf will still be write-protected when your next increment job begins because the OPP is 13 days.  The incremental job will create a new  .bkf file and write to that.

I would STRONGLY suggest that you familiarise yourself with the Media Management chapter of the Admin Guide.  You cannot hope to operate BE if you do not understand that chapter properly.

[Colin_Weaver]

Just for info IMGnnnnn folders are the backup storage GRT (Granular Recovery) enabled backup sets. BKF files are the storage locatiosn for data backed up using more traditional methods. Both the files and the folders will contain backup data.

[Moggyman]

Thanks all. True, I should RTFM.  Something I'll try and do later.

Like a few other people (judging by a search of the site) I seem to have fallen foul of the difference between the Overwrite Protection Period and the Global setting for Overwrite Protection LEVEL.  I did a test full backup yesterday followed by an Incremental job overnight and the Inc job completely overwrote the bkf file that the full backup created.  In Media Management I've now changed the level to Full and am crossing my fingers.  If that rings any alarm bells with you please let me know!

In the meantime I think you all again for your advice - it's much appreciated.

[Moggyman]

Coincedence?

I changed the OPL and the AP and did another full backup test.  That worked OK.  I then set an incremental job going and these appear to fail each time with these details -

Job ended: 07 July 2011 at 15:30:21
Completed status: Failed
Final error: 0xe00084af - The directory or file was not found, or could not be accessed.
Final error category: Job Errors

For additional information regarding this error refer to link V-79-57344-33967

Errors
Click an error below to locate it in the job log Backup- \\EXBOX\Microsoft Information Store\First Storage Group Database was not found, or could not be accessed. Database was not found, or could not be accessed.

Is this down to one of the setting changes or can anyone tell me why it's started moaning?  I haven't changed anything within the incremental backup job since it ran OK last night.

Thanks 

[pkh]

This is a separate problem for which you should start a new discussion.