On Thu, May 15, 2003 at 08:12:03AM -0500, Randy Pearson wrote:
> 
> Now for my question.  Does anybody know why level 0 and level 1 dumps of my 
> disks would be the same approximate size always?
> 
> I thought a level 1 dump supposed to be an incremental backup, while level 0 
> is a full backup.  Here are some extracts for the Amanda reports of the dumps 
> of my /home disk.  Looks to me like Amanda is backing up every file, no 
> matter what the dump level is.
> 
> May 14
> - -------
> DUMP SUMMARY:
>                                      DUMPER STATS                TAPER STATS  
> HOST  DISK             L ORIG-KB  OUT-KB COMP%  MMM:SS    KB/s  MMM:SS    KB/s
> - -------------------------------- ---------------------    
> --------------------
> zev   /home            1 7436050 4881984  65.7   43:26  1873.6   90:42   897.1
> 
> May 13
> - ------
>                                      DUMPER STATS                TAPER STATS
> HOST  DISK             L ORIG-KB  OUT-KB COMP%  MMM:SS    KB/s  MMM:SS    KB/s
> - -------------------------------- ---------------------    
> --------------------
> zev   /home            1 7436050 4881984  65.7   43:26  1873.6   90:42   897.1
> 
> May 9
> - -----
>                                      DUMPER STATS                TAPER STATS
> HOST  DISK             L ORIG-KB  OUT-KB COMP%  MMM:SS    KB/s  MMM:SS    KB/s
> - -------------------------------- ---------------------    
> --------------------
> zev   /home            0 7518580 4895872  65.1   42:29  1920.4   90:59   896.8


I'm surprised the planner didn't say, level 1 and level 0 are the same size,
might as well do level 0.  The two level 1's should both be large since both
reflect changes since May 9.  Yet, it did not shift to level 2 on May 14.  

Others have noted the possibility of this being a PC file system and not
really having unix style timestamps.  And maybe the data is changing.
I'll just add one really farout one that happened to me a long time ago.
Some other task you run on your system affects the time stamps.  I forget
the details, but there was a cronjob that ran, scanning all the files on
the system.  For some reason it affected either the modification or inode
change time when it did so.

-- 
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)