On Mon, Mar 13, 2006 at 04:05:00PM +0100, Thomas Widhalm wrote:
> On Mon, 2006-03-13 at 15:57 +0100, Paul Bijnens wrote:
> > On 2006-03-13 15:45, Thomas Widhalm wrote:
> > > Hi!
> > > 
> > > We have a rather old workstation running tao 1 (is RHEL 3) with amanda
> > > amanda-2.4.4p1-0.3E.1 . Since this is too old for vdisks, we just backup
> > > to the holding disk. But it makes lots of level 0 backups. Some
> > > filesystems have level 0 just every day!
> > 
...
> > > Is there a way to make level incemental backups more often? Making more
> > > fulls throughout the backup cycle is ok since we got enough space, but
> > > not that many!
> > 
> > What is your schedule?  (dumpcycle, etc.)
> 
> dumpcycle 8 days. amanda run from a script within crontab every day.

There was an "etc." up there.
Presumably your runspercycle is also 8?

> 
> > Are these dumps promoted frequently?
> 
> Yes. Every day
> 
> > How/when do you remove the older holdingdisk backups?
> 
> There is a procedure within our script, which deletes any directory
> named after a date, older than 9 days with "rm -fr"

Does this keep your index dir growing; or do you trim that too?

Have you tried resetting some parameters like bumpsize, bumpmult,
and bumpdays?  They may have an affect on your situation.

A workaround would be to use a dumptype that does no fulls, or
strategy incrmental only.  Then daily run a cron script that
selects the DLE's you want full dumps of and issues an amadmin
"force" command to cause a level 0 dump the next run.

-- 
Jon H. LaBadie                  jon AT jgcomp DOT com
 JG Computing
 4455 Province Line Road        (609) 252-0159
 Princeton, NJ  08540-4322      (609) 683-7220 (fax)