I recently noticed that Amanda wasn't using the holding area when dumping a 
large NT share via SAMBA.  I have 3 holding disks (partitions), which are fully 
available to Amanda, and have sizes of 30Gb, 15Gb and 2Gb for hd1, hd2 and hd3 
respectively.
I ran amstatus during a backup yesterday and noticed the following:

servb://mainserver/Users 1  941545K dumping 312512K
servb:/awcserver/Users   0 2508617K waiting for dump
7 dumpers idle: no-diskspace
holding space: 4955410

When I ran df against my holding disks I got:
hd1: 25303167K
hd2: 12670006K     (This area being used by the currently active dumper)
hd3:  1733404K

Can anyone tell me why the 2.5Gb level 0 waiting to dump didn't dump to 
hd1, which has 25Gb free space?

My hd definitions are:
holdingdisk hd1 {
    comment "holding disk dir1"
    directory "/data/hold/amanda1"      # where the holding disk is
    use -10 Mb          # how much space can we use on it
    chunksize 2000 Mb   # size of chunk if you want big dump to be
    }

holdingdisk hd2 {
    comment "holding disk dir2"
    directory "/data/hold/amanda2"      # where the holding disk is
    use -10 Mb          # how much space can we use on it
    chunksize 2000 Mb   # size of chunk if you want big dump to be
    }
     
holdingdisk hd3 {
    comment "holding disk dir3"
    directory "/data/hold/amanda3"      # where the holding disk is
    use -10 Mb          # how much space can we use on it
    chunksize 2000 Mb   # size of chunk if you want big dump to be
    }

The dumptype for the waiting share was:
define dumptype nocomp-user-gnutar {
    global
    program "GNUTAR"
    comment "Non-root partitions on reasonably fast machines"
    compress none
    dumpcycle 4
    priority high
}

I also have 
reserve 30 # for degraded mode backups

Can anyone spot where I'm going wrong?


_____________________________________________________________
Get Paid to Surf the Web!
http://www.alladvantage.com/home.asp?refid=AVZ855