Restore. I was wondering if anyone has a script that restores from an existing backup where the back up is created by script in a different directory and different file name such as filenamebac.fp

I've been finding that the backup works okay one or two times and then tells me the file is locked.... I assumed it was open/ so closing the file before starting the backup process worked okay but creating a restore script has proven elusive since it doesn't seem to find the file directory or if it does the file is locked or in use... even though the script is closing the file.

I have always just used windows explorer in the past and done this manually on the rare occations that repare didn't work. And guess it is okay to keep doing it that way. Just thought it would be nice to work out a script on this.

Restore is a "last resort" process: it should rarely need doing, and the data from the restored file should imported into a known-good clone of the original file. The restored file should not be used again as the restore process only fixed the data, not the complete file structure.

Are you using FMS to backup the files? backing up the shared files with any other backup utility is NOT the way to do it.

Actually I wasn't thinking in terms of a repare file situation but rather one where a record was accidently deleted by an end user and the restore would keep all current data and add the deleted record from the back-up.

It seems criptic to import each table of a single file when your working with up-teen tables.

SeaRanger:

Your best bet is to do one of the following:

1. not allow the deletion of records at all, but rather just "mark" them as deleted, and then omit the "deleted" records from any finds, etc... To "undelete" the record, you'd just unmark it.

2. When a record is "deleted", export it to a file that contains all of your "deleted" records. Then, when it comes time to undelete, you just find that record in the "deleted" file & send it back to your live file.

-Stanley

Thanks I will try that ... but I wonder how that works on very large files. Although that is going to require some thought with this application.

SeaRanger:

On a larger file, I'd use the second of the methods outlined in my previous post. The first method would begin to bog down after a while, if you had many many thousands of records.

-Stanley