Best Practice

Hello,

I'm wanting to sell a backup solution to my customers. I am currently selling an external USB HD and TI9. What I generally do is run a Full backup then schedule a differential backup. The primary purpose is disaster recovery, and only disaster recovery. Currently I have the entire external drive set as a Secure Zone.

First, how many differential backups wll it keep? Will it eventually fill up the drive? Second, should I ever need to use the backup (a failed primary HD), what is the quickest way to get the customer back up and running? And lastly, I am using the Secure Zone because I don't want the customer to be able to inadvertantly delete a backup image. Am I doing this the right way for the intended purpose? I would prefer a full clone, but I don't want my customers to have to interact with the software, reboot etc.

The end goal is to get the customer back up and running ASAP after a primary drive failure.

Thanks!
Brian

 

Hi Brian,

You will find many people will advise you not to use the SZ at all, but I think that your reasoning of keeping the backups 'safe' from the users is logical, all things considered. Do make sure though that you do not accept the default to activate the SRM.

I would however change tactics from taking single full backup and endless differentials. Strangely, Acronis have decided to make the SZ management of incrementals and differentials exactly the same. Ultimately, if all you have is a single full and a heap of differentials, there will come a day when there is insufficient room for the next differential and you will get an out of space error. It would make sense for the earliest differentials to be deleted automatically, but that is not how it works - not yet anyway.

You should set up a schedule where you start a new set i.e. a new full and then differentials, every few weeks - how frequently will depend on the size of the SZ. That way, when there is not enough room for a new differential, there will be an old set of backups which can be safely (and automatically) deleted.

In the event of a failed primary drive, you replace the drive if necessary, boot from the boot CD, restore the latest differential from the external and you should be back in business.

 

Hello Brian,

Ozmaniac has provided excellent advice. Perhaps I can expand a little on his statement below.

You may or may not know that activating the Startup Recovery Manager (SRM) overwrite the current Master Boot Record (MBR) with its own code. This will most likely cause any multiboot O/Ss or hidden bootable Recovery partitions to become non-accessable. In addition, the new MBR code also checks for the presence of the associated Secure Zone. Therefore, in your scenario, if the external USB drive (containing the SZ) happens to be disconnected or switched off whilst booting the computer the system will "hang".

Regards

 

Well, as I found out recently "One backup is no backup at all" (from one of the web forums I trawled solving my problem below).

Be warned!

I have just had two hard drives connected by firewire corrupted due to the issues described here:

http://support.microsoft.com/kb/885464

http://support.microsoft.com/kb/887170

The problem is well described here: http://www.bustrace.com/delayedwrite/

I had been using TI9 until it destroyed my backups setting up a secure zone partition, so until they release a stable TI9 I had reverted to using the WinXP backup utility...fortunately as it turns out.

Issue logged with file recovery sw company:
200gb backup drive contained .bkf of 3 partitions 24gb, 27bg, 29.7gb + 15gb laptop backup.
Had a system of chained external firewire drives that appeared to have been working fine, allowing me to reorganise and backup my data after a move from Australia to japan.
I turned them on early last sunday morning and promptly got a delayed write error (wtf?) which wiped one data drive and the 200gb drive...WinXP recognisd the drives but said file system is corrupted, drive undreadable.

AAARRRRRRRRRRRRRRRRRGHHHH!!!!

Searched web and found that this error is a serious nasty quite common problem with no easy solution, even from M$ (see links above).
Downloaded Active@filerecovery and Active@undelete to test and see if I could retrieve my life. Undelete looked slightly more functional, and looked liked it located 3 of the backups .bkf in a $MFT_1 folder. The data drive was also scanned, but all the folder names were unrecoverable, which made file retrieval possible but a real pain in the arse due to the extensive rebuilding necessary.
Bought Undelete and attempted recovery of a critical 29.7gb data backup as a .bkf. ~1.5 hours to scan 200gb drive and 4.5 hours to recover the 29.7gb .bkf file, which WinXP backup/restore catalogs and restores from perfectly. Success!!! ... Now looking at further backup redundancy, but Undelete is now in my core toolset.

Feedback from their techsupport (after thanking them) if I followed the right procedure was: "That was the correct way to do it. I do not think re-building the partiton would have helped and may have caused problems.
".

So in addition to backup/restore software I would also add some data recovery tools to your toolkit. Also choose your external HDD cases with care.

Now waiting for a TI9 version that works so I can safely and reliably take images (x2) of my laptop drive before gigs...just in case.

sv

 

and now trying to get some sleep after 3 sleepless days of data recovery, hdd testing (dumped two generic firewire/usb HDD cases and replaced with a USB ony -touch wood), and mutiple redundancy backups so I can get back to work...