Robocopy To The Rescue!
As a digital photographer, I keep all of my pictures organized into folders on my computer’s hard drive. Since getting my Canon Digital Rebel XTi last year, I have generated about 20 gigs of images. I’ve kept telling myself I really need to back them up somewhere. While I consider my main desktop system to be pretty stable and issue-free, there is no guarantee that it will always function smoothly.
Last month I had an issue with another of our computers in which one of the hard drives appeared to be corrupted (thankfully it was a minor problem with a driver). My greatest concern was the contents of that hard drive; there are a number of files on it that aren’t currently backed up. One of the unfortunate rules of computing is that you probably won’t have a decent backup strategy in place until after you’ve already lost a bunch of irreplaceable data. I got the system back up and running, and after breathing a sigh of relief I realized how close I had come to being a statistic.
The incident renewed my determination to be an exception to this data-loss rule. So this weekend I spent some quality time setting up robocopy scripts on my home computers. Robocopy is a powerful command-line utility that has been around for a long time. It requires some understanding of the command-line environment and can be a bit difficult to work with. Fortunately, there is now a robocopyGUI interface that acts as a standard windows front-end to the utility.
I generally set up our computers with at least 2 hard drives. One is dedicated to the OS and to programs that require installation on the “C” drive. The other drive is for user documents and other data. I’m a bit believer in having a separate drive or partition as a backup area, so I set up robocopy to keep a mirror of my data in this space. After running it on my digital images director, I worked on copying other documents from the computers in the network. I am even thinking of backing up the windows partitions just in case.
/MIR is really, really dangerous. I’ve whipped 100’s of gb’s before I could stop it. Luckily I backups were available. 🙂 I always, always, always use /E when i write commands, unless I explicitly want to mirror dirs.
Always consult the file management pros! Brad knows his stuff. He makes an excellent point that is worth calling out. When I said I “keep a mirror” of the data on the drive, I don’t use the /MIR option. I use robycopy to keep a synchronized copy of the data. As he points out, the /E option is greatly preferable and keeps the disk overhead to a minimum. Good catch!