Amanda-Users

Re: [amanda AT fractal.Mines DOT EDU: DDS4backup AMANDA PROBLEM: FIX BEFORE RUN, IF POSSIBLE]

2003-02-13 04:45:02
Subject: Re: [amanda AT fractal.Mines DOT EDU: DDS4backup AMANDA PROBLEM: FIX BEFORE RUN, IF POSSIBLE]
From: Michael Robbert <mrobbert AT mines DOT edu>
To: gene_heskett AT iolinc DOT net
Date: Thu, 13 Feb 2003 00:53:54 -0700
Gene Heskett wrote:

On Wednesday 12 February 2003 19:23, Michael Robbert wrote:
Here is an update on the problem that I sent yesterday. I had
entered my DDS4backup into cron and last night the backup
actually worked, sort of. It looks like it wrote to the tape in
slot one, which I labeled DDS4backup01, but the report says that
it was labeled DDS4backup05. Below you'll find the results of
todays amcheck, all tapes appear to be labeled DDS4backup05, but
that is not how I labeled them. I am guessing that tonights
backup will fail because it can't seem to find another tape.
A second question I have is about the number of slots to put into
my changer.conf. I am using the first 5 slots for backup tapes
and slot 6 is always a cleaning tape. If I configure lastslot=5
and cleanslot=6 then "amtape DDS4backup clean" gives the
following error: amtape: device 0 not clean: Slot 6 is out of
range (1 - 5) If I then change lastslot=6 then I get the problem
that you'll notice below. Amanda will cycle through the cleaning
tape thinking that it is a normal tape. I am using
tpchanger=chg-zd-mtx in my amanda.conf.

----- Forwarded message from Amanda user
<amanda AT fractal.Mines DOT EDU> -----

Date: Wed, 12 Feb 2003 16:04:15 -0700
From: Amanda user <amanda AT fractal.Mines DOT EDU>
To: amanda AT fractal.Mines DOT EDU, mrobbert AT mines DOT edu
Subject: DDS4backup AMANDA PROBLEM: FIX BEFORE RUN, IF POSSIBLE
X-Spam-Status: No, hits=-98.9 required=5.0
        tests=DOUBLE_CAPSWORD,USER_IN_WHITELIST
        version=2.31

Amanda Tape Server Host Check
-----------------------------
Holding disk /holding/amanda: 6128768 KB disk space available,
using 6026368 KB Holding disk /storage/amanda: 40446788 KB disk
space available, using 40344388 KB amcheck-server: slot 2: date 20030212 label DDS4backup05 (active tape)
amcheck-server: slot 3: date 20030212 label DDS4backup05 (active tape)
amcheck-server: slot 4: date 20030212 label DDS4backup05 (active tape)
amcheck-server: slot 5: date 20030212 label DDS4backup05 (active tape)
amcheck-server: fatal slot 6: Cleaning Cartridge Installed and Ejected

Gaack! Either its not advancing the magazine, or you've labeled all tapes
alike. If the latter, I'm fairly certain that amlabel would have had a
litter of kittens over that.

Which is it?

Neither, I have watched it advance the tapes and you can tell by the fact that it loaded a cleaning tape from slot 6 that different tapes are getting used. And I am sure that I labeled all these tapes properly, I had to use -f because amlabel thought they were labeled even before I labeled them. I also know that this isn't acurate because when I watch the drive it loads the tape, but never rewinds or reads the tape to see what label is actually on it.


ERROR: new tape not found in rack
     (expecting a new tape)
NOTE: skipping tape-writable test
NOTE: info dir /var/lib/amanda/DDS4backup/curinfo/ito/_usr_local:
does not exist NOTE: index dir
/var/lib/amanda/DDS4backup/index/ito/_usr_local: does not exist
NOTE: info dir /var/lib/amanda/DDS4backup/curinfo/ito/_u1: does
not exist NOTE: index dir
/var/lib/amanda/DDS4backup/index/ito/_u1: does not exist Server
check took 254.892 seconds

Amanda Backup Client Hosts Check
--------------------------------
Client check: 1 host checked in 0.510 seconds, 0 problems found

This settles it.

Its not advancing the tape then, and it obviously read all that
from the drives buffers. So something is amiss in the changer
script setup.  This amount of tape checking should have taken a time
measured in minutes, not just fractions of a second.

The tape check took 254.892 seconds(over 4 minutes), it was the client host check that only took 0.510 seconds