[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Bacula-devel] Bacula feature request:
Thanks for the Feature Request. This is a feature that I have wanted to see
for some time. I'm impressed to receive a patch that in addition to the code
contains the documentation -- a *really* nice surprise.
I'll take a careful look at the patch in the next few days and get back to
On Thursday 07 February 2008 23.32:03 Bailey, Scott wrote:
> Item 1: (Optionally) force backups to be upgraded more aggressively
> Origin: Scott Bailey <scott dot bailey at eds dot com>
> Date: 7 February 2008
> Status: "I can't believe he did that" draft patch attached
> What: Add "Max Full Age" specifier to job definitions
> Why: Bacula's current behavior is to force an incremental or
> differential backup to be upgraded to a full backup if
> no previous
> successful backup is found. In practice, this means that
> discovers no full backup exists and tries to create a
> *AFTER* you have eliminated your last good full backup.
> This proposal allows this behavior to be modified, i.e.
> of "Max Full Age = 1 month" to a job definition causes
> the next
> backup to become a full backup if no full has completed
> successfully during the previous month. This provides a
> safety net in the event jobs, retries, etc. have not
> succeeded in
> generating a good full backup within some expected
> It also simplifies scheduling for sites (such as my
> house) where
> precise scheduling of backups is unimportant but the
> general cycle
> is known. Presently, multiple systems on the same
> schedule have
> full backups that pile up; jobs at the end of the queue
> time out
> and are cancelled before they can run. They then tend
> not to be
> noticed, resulting in an endless chain of incrementals
> in the event
> a restore is required. The proposed behavior allows a
> simple daily
> incremental schedule with a suitable Max Full Age
> specifier to
> automatically generate full backups as required without
> attention or intervention.
> Notes: If this specifier is not used, existing behavior is preserved.
> It is anyone's guess if the logic in the attached patch
> works as I intended or described. :-) It compiles
> against head
> without errors.
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
Bacula-devel mailing list
This mailing list archive is a service of Copilot Consulting.