On 06/01/2009 02:22 AM, Hugh E Cruickshank wrote:
> From: Gene Heskett Sent: May 31, 2009 11:25
>> On Sunday 31 May 2009, Hugh E Cruickshank wrote:
>>> From: Jon LaBadie Sent: May 27, 2009 22:34
>>>> You got stuck on tape five because your tapecycle is 2.  It must
>>>> use another tape before reusing tape five and for some reason
>>>> it isn't getting to another tape.  That is why I question the
>>>> changer setup.
>>> What should tapecycle be set to?
>>>
>> The number of tapes available for amanda to use.
> 
> I guess that is where my confusion lies.
> 
> I have 5 tapes in slots 1-5 that I am attempting to use as set of
> "weekday" level 0 backups (i.e. a full backup each and every weekday
> night).
> 
> When I created the DailySet1 I used the command:
> 
>   amserverconfig DailySet1 --template tape-changer \
>   –-tapedev /dev/nst0 --changerdev /dev/sg1 --mailto pgiadm AT forsoft DOT 
> com \
>   --dumpcycle 0 --runspercycle 1 --tapecycle 2 --runtapes 1
> 
> However amanda.conf now has:

Let me try to explain how amanda understand what's in your config,
correct me if i'm wrong:

>   dumpcycle 1 week

amanda will try to do at least one full backup of every DLE (Disk List
Entry) once a week.

>   runspercycle 1

amanda runs 1 times in the dumpcycle. Means: amanda thinks you start
only one backup run in a week. This is probably not what you want.

>   tapecycle 2

amanda knows about 2 available tapes at all.

>   runtapes 1

amanda will use max. 1 tape once per run.

> and changer.conf has:
> 
>   firstslot=1
>   lastslot=5
>   driveslot=0
>   havereader=1
> 
> Should I change dumpcycle back to 0 and tapecycle to 1 or what?

What you want may differ, i haven't followed this thread from beginning.

Let's say you have 20 tapes, then set tapecycle to 20.

If you run amanda five times a week, set runspercycle to 5.

If all backup data for one run fits to one tape, leave runtapes at 1,
but it doesn't matter if you set it to 2. amanda will not use the second
tape for a run, if all data fits onto one tape.

However, it takes some experimenting and some weeks of manual monitoring
to learn and see how amanda works.

 Cheers,
 Peter