Networker

[Networker] Too(?) much data in incremental savesets

2003-07-04 19:41:20
Subject: [Networker] Too(?) much data in incremental savesets
From: Oscar Olsson <spam1 AT QBRANCH DOT SE>
To: NETWORKER AT LISTMAIL.TEMPLE DOT EDU
Date: Sat, 5 Jul 2003 01:41:13 +0200
I have noticed that when we perform certain incremental backups, there is
a lot of data being backed up, even though a minimal part of the data has
been changed. This is also true for NDMP backups of a NetApp filer.

Basically, Out of 300GB data, I might back up 100-200GB data over one day,
when day1 is a level5 backup and the next day is a level6 backup. I know
that the data hasn't changed that much on the filer. What could the reson
be?

We're running data ontap 6.3.x and networker 6.1.3 on solaris 7. NDMP
protocol version is 3. The scedule works like this: day 1: full, day 2:
level 1, day 3: level 2, day 4: level 3, day 5: level 4, day 6: level 2
... and so on, in order to be able to do a more or less incremental backup
every day, even though the incr level isn't supported for NDMP (AFAIK).
We're doing a full backup on the 7:th of each month.

So what causes a file to be backed up again? Normal read file access
(atime)? Or is it just when it has been modified or some kind of archive
bit is set? We do scheduled anti-virus scanning of the filesystems, could
that affect what gets backed up, even though the scanner just reads the
files?

//Oscar

--
Note: To sign off this list, send a "signoff networker" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=

<Prev in Thread] Current Thread [Next in Thread>