Greetings.

I seem to be getting an awful lot of full (level-0) backups:

-rw------- 1 amanda backup 805,089,376 Apr  1 23:04 slot140/00020.alfred._var.0
-rw------- 1 amanda backup 805,100,793 Apr  2 23:04 slot141/00021.alfred._var.0
-rw------- 1 amanda backup 805,116,298 Apr  3 23:04 slot142/00021.alfred._var.0
-rw------- 1 amanda backup 805,124,968 Apr  4 23:05 slot143/00019.alfred._var.0
-rw------- 1 amanda backup 805,175,244 Apr  5 23:04 slot144/00020.alfred._var.0
-rw------- 1 amanda backup 805,240,006 Apr  6 23:04 slot145/00020.alfred._var.0
-rw------- 1 amanda backup 805,246,124 Apr  7 23:04 slot146/00019.alfred._var.0
-rw------- 1 amanda backup 805,285,370 Apr  8 23:04 slot147/00015.alfred._var.0
-rw------- 1 amanda backup 805,297,042 Apr  9 23:04 slot148/00021.alfred._var.0

It's not on all disks, but it's on a lot of them.

My "estimate" seems to be working correctly:

$ /usr/local/sbin/amadmin MyConfig estimate alfred /var
alfred /var 0 876790
alfred /var 1 1420

...and I haven't changed any of the default "bump" options. (I can post
my amanda.conf if you want me to.)

So I guess I'm missing something as to why my backups aren't bumping to
level 1.  My best guess is that "the level-0 backup is below some
threshhold that amanda decides that repeated 800-MB level-0 backups are
more efficient than requiring multiple tapes for a restore."  But I can't
find any obvious indication of such a threshhold.

It's not a big deal, but it's got me curious.  Can somebody toss a clue
my way?

--hymie!    http://lactose.homelinux.net/~hymie    [email protected]

Reply via email to