Possible to continue after I/O error?

[hankvt]

I am trying to get all the files off a malfunctioning disk, but SD! stops the backup process after encountering an I/O error. Is it possible to have SD! log the error and continue copying the remaining files? Currently I have to edit the copy script to tell SD! to ignore the file, and then restart the copy from the beginning. This is taking hours upon hours and my frustration level is high. I know that the resulting backup will not be a bonafide backup, but I just want to get as much off the original disk as possible before I replace it.

[dnanian]

I'm sorry, no. We're not really designed to scavenge files from a failing drive...

[hankvt]

New feature request then?

The bad files are the exception (12 files out of 672,978). I wouldn't call that scavenging

[dnanian]

Hard to know where to draw that line, of course.

Thanks for the suggestion. If you delete the file, Smart Update should resume pretty quickly...

[hankvt]

I would love to be able to do that but I'm working from a read-only directory provided by the Preview mode of DiskWarrior 4.

[dnanian]

Yikes. The line would be pretty close to there...

[hankvt]

Since it looks like you use ditto as a last resort, I wrote a replacement for ditto that eats all errors and returns no error to the calling process. The replacement program calls the original ditto (renamed to xditto) so all functionality is maintained.

[dnanian]

That'll certainly do it... but -- other readers -- don't do this kind of thing, if you can fix the problems. We really do stop for a reason, and Hank's case is unusual...

[xochi]

I found out recently that Apple's Migration Assistant does a wonderful job at skipping files on a failing disk -- I/O errors that stopped other backup programs did not affect it.

So, if you haven't already done so, give Migration Assistant a try.