[chasg]

Hi All, I've done a dumb thing.

I accidentally did a day's work on the destination drive for a SuperDuper duplication. It's all gone, and I really want that work back.

My working drive is duplicated every night to a backup drive by SuperDuper (using their "Smart Update" feature). I accidentally did yesterday's work on the destination drive, rather than on the source drive. This happened because I didn't navigate to the folder via the Finder, rather I did a quick spotlight search and just opened the first folder in the results, which were on the destination drive. Stupid.

A quick scan of the backup drive with Data Rescue 3 (which I've bought just now) has come up with no recovered files. I'm now running a "deep scan", but the software estimates that it'll take 13 hours.

The backup drive is quite a bit larger than the source drive. This gives me hope that the files I worked on weren't overwritten during the duplication (hope hope!). I have an email out to Dave Nanian (SuperDuper author), but no reply as of yet.

Can anyone give me some insight into whether or not I'll be successful in my attempts at recovery? Any hope at all?

Chas

[chasg]

Here is an update that might be useful to anyone else who encounters this (admittedly unusual) problem. I got an email back from SuperDuper's author, and he states:

"A Smart Update doesn't do anything to 'try' to delete files thoroughly - quite the opposite: we do a simple replace, which *should* give you the maximum chance of recovering the data, as long as you didn't fill the drive (OSX tries to keep around deleted files as long as possible)..."

After a "deep scan", Data Rescue 3 has found many gigs of files (many gigs!). To be safe, I'm recovering all PSD files (only about 100GB worth *phew*), hopefully my 15 photoshop files are in there.

Chas

[OreoCookie]

That's the reason cloning is a really bad backup strategy! IMHO you should really consider augmenting that with a proper incremental backup. Then going back would be pretty easy.

Anyway, I think you're already using the best strategy to recover your data: letting Data Rescue do its magic. Data Rescue has worked extremely well for me in the past (I had to use it twice).

[chasg]

Well, the clone really is the third tier to my backup strategy, but now I'm sounding defensive (the really bad thing about it was me using the wrong drive, and now I'm being reminded of how dumb I was making this mistake :-)

Sadly, there was no sign of the files after Data Rescue did its thing. I was pretty impressed with it though, some really ancient files were recovered. Worth the purchase. Now I need to get back to work, and do all that editing again *sigh*

[seanc]

I doubt it will find anything, but you could also give the free Photorec a try.

[chasg]

Thanks seanc, the suggestion is much appreciated, and I've got nothing to lose. I'll try it out.

[OreoCookie]

@chasg
I didn't want to come across as trying to lecture you on the topic either.

Sorry to hear you were unable to recover the files you were looking for.

[chasg]

Sorry, lost track of this discussion.

OreoCookie, thanks for the followup. You're not the lecturing type, so I certainly didn't take it that way :-)

Cheers!

Chas

[steve626]

Originally Posted by OreoCookie 

That's the reason cloning is a really bad backup strategy! IMHO you should really consider augmenting that with a proper incremental backup. Then going back would be pretty easy.

Actually, cloning is a reasonable thing to do as part of a comprehensive backup strategy. Such a strategy should include incremental backups that keep versions -- such as Time Machine or Retrospect. I do clones also, separately, but always dismount, power down and disconnect the clone when it completes and move it to a different location. I consider the clone simply another layer of backup above and beyond the Time Machine backup. I would also include offsite over the internet backups but at home do not have adequate bandwidth to make this feasible (my daughter has nearly 200 GB of photos, we get 10 Mbit/s downlink but just 1 Mbit/s uplink, which is roughly 0.1 MB/s uplink which would take ~ months).

The clone can be considered another layer of backup to guard against a catastrophe such as a power failure or power spike that occurs in the middle of Time Machine running its backup, that could in principle ruin both the primary and the backup Time Machine drive. Having a recent clone in a separate location, kept powered off except when the clone is being updated, can protect against this scenario. While unlikely, I have seen this scenario happen in fact.

Two types of backup should be used at a minimum, and a third (e.g. offsite over the internet) would be ideal. But leaving cloned drives AND backup drives all plugged in, connected and running all the time means that a single power spike could take out the primary drive plus all the backups all at once. Or, as in this case of the original poster, one can even unintentionally lose data by confusing the clone and the main drives.

[OreoCookie]

How is what you're writing different from the post of mine you quote: clones by themselves are a bad backup strategy, but clones can be part of a better backup strategy which IMO should involve incremental backups.