DDP|ER Backup Policies

A list of file/path names with wildcards/regular expressions specifying what to exclude. Evaluated before include. DDP | ER has a (modifiable) default list of exclude strings for stuff like the swap file.

Including and Excluding Files

The most common include/exclude strategy is to list extensions to backup and leave exclude empty. Often we just leave both fields empty.

For example: enter *.doc,*.docx,*.pptx,*.xlsx to include office files. Or if your program creates unusual temporary files you may want to exclude *.mytmp,*.notdone (for example). You can exclude *\tmp\ to exclude the tmp subfolders from projects.

Path searches have a path separator (\) in them. Without a separator just the file name is compared. To exclude git and svn folders use *.git\,*.svn\. File name include and exclude values use the file name as a search. Path search values use the entire path in the search. So, for example, to exclude all folders named 'test' use the exclude string '*\test\*' to match all paths with \test\ in them. To exclude all paths which have a folder whose name starts with test use *\test*.

Unless you know what you're doing, leave Regular Expressions turned off. Default (Wildcard) expansion supports * (many characters match) and ? (one character match).

For sophisticated file and path name filtering please look at regular expressions.

Delimiters for include and exclude

Include and exclude both use , (comma) as a delimiter. If you want to use a comma in the include or exclude strings, you may use | as a delimiter. Include at least one so the interpreter knows that's the delimiter.

When

Pick a schedule for backing up. This may be advanced (back up every three hours only on tuesday between 9am and 5pm) or basic (back up daily at 9am).

Scheduling

When a backup is first scheduled to run we place it into the Job Queue. You'll see it in the dashboard when it's pending. When the job time hits, the job is checked and if it can run (destination available, no other backups to that destination, ...) the job is started. If the job can't run then it goes back into the queue until it can run.

When the job starts it loads the backup policy settings so if you change them before it starts it will use the at-start-time values.

When the job finishes (which may be a while) we check the backup policy to see when the next job for this policy should be scheduled. We schedule it and go through the process again.

So, if a job were to be scheduled daily, but it took more than a day to run you would find that it actually only runs every two days.

Where

Folders may be backed up to one or more destinations. These backups occur in parallel streams, so destinations, even if they come from the same backup, may not have the same data. For example, one of your destinations may be a network with congestion problems, while one is a local hard drive. The hard drive will back up sooner and possibly contain earlier files than the network.

How

Backups may use shadow copy or they may not. Use shadow copy to keep file sets together or for often-locked files. You may select either

Never - if locked files are encountered they will be skipped and the backup will show as a failure; unlocked files will backup fine
IfNeeded - if locked files are encountered they will be skipped and then at the end a Shadow Copy will be run and the locked files backed up automatically
Always - a Shadow Copy will always begin the backup and any locked files or file groups will be backed up in synch

You may also elect to only have backups happen when a specific user is logged in. Files that are encrypted and only available during a user's session may be backed up reliably in this case.