Ignore failed files on Restore

Is there a way to automatically ignore failed files during a restore?

For example.....

You create an image of data files and validate the image. At some time for whatever reason you need to restore the image. Unknown to you the source contained files that won't restore e.g. "filename to long".

Currently when the restore process encounters such a file it stops and displays an error message with an option to "Ignore". This of course means that you have to babysit the restore so that you can click "Ignore" every time one of these files are encounted.

Does TI9 have a setting that will enable the restore to run and simply write the path of all the ignored files to a log? You could then let the restore run unattended. At the end of the restore TI could display a message that X number of files were not restored. You could then browse the log and see if any of these files were important.

In the real world these "failed" files are seldom important, 90% of them are URL's with very long filenames which inevitably end up in the users data folders.

The real irritation is that TI will happily backup and validate files which the restore process throws out. Consistency between TI backup, validate and restore would be really nice.

Or have I missed something really simple here?

 

You'll get such errors only when restoring files and directories, not in an image restore.

TI could handle such errors if they would modify their code as indicated in http://www.standards.com/index.html?CreateFileFailure.

In any case, to find out whether the files REALLY cannot be restored, run the CompareDrivesUsingFileAttibutes program at CompareDrives. That program uses the mechanism described in http://www.standards.com/index.html?CreateFileFailure to access allegedly bad files.

 

Hi Howard

Thanks for the links. Interesting info.

Now I have just got to figure out a way how to tell that to a customer

 

I've modified a program that will demonstrate the issue. Hopefully, I'll post it by tomorrow.

The program will allow you to try to open ANY file in two ways:

1. Using the usual API calls.

2. Using the method I described in http://www.standards.com/index.html?CreateFileFailure.

So, you can try both ways and see the difference.

 

I've posted the program at http://www.standards.com/index.html?ReadFile.

For each troublesome file, you can use the program to try to read the file two ways.

If you uncheck the box "Attempt to bypass errors", you will likely see the same error obtained when trying to restore the file from TI.

If you check the box, the program will attempt to overcome such errors.

On my systym, I have 5 files that cause a problem for some backup programs attempting to restore the files, ditto for some ZIP archive programs.

My program is able to open and read all 5 files.

Lemmee know whether this works on your system.

 

I posted another version that allows for three methods to open a file.

1, Checking the "Attempt to bypass errors" box will overcome the most problems.

2. Unchecking the box "Attempt to bypass errors", you can choose whether or not to use a Unicode open. More errors will be encounterd if you do not use the Unicode option.

 

Modified program to have 4 alternatives for OPENing a file.