>> ... To me it looks like Amanda did a
>> level 1 dump on Wednesday am to the holding disk, then another level 1
>> dump on Thursday am, then another level 1 on Friday am.
>
>Yes, exactly. Your AMANDA works fine, where is the problem?
The problem is that off line of discussions on this list, I suggested
Denise change the various bump* parameters to zero to try and make Amanda
bump levels every day. The theory being that this would minimize the
size of each image in the holding disk and so she would get as much as
possible in there during the week.
Chris is correct that in a normal arrangement, doing the same level over
and over is expected behavior. But in Denise's case it didn't do what
she (or I) expected.
Since this was my bright idea :-), Denise, could you E-mail mail to me
(and not the list -- just to avoid the noise) the amdump.<NN> files that
go along with the reports you sent in this thread (7, 8, 9 of Nov)?
I'll look through what planner was doing and see if I can figure out
what went "wrong".
Note that nothing is really bad in this setup. The dumps you got are
perfectly valid. They just take up more holding disk space than you
might like.
One final question. You don't happen to have "record no" still set from
some testing, do you? Actually, why don't you go ahead and E-mail me
your amanda.conf and disklist, too.
And please E-mail them direct, i.e. attachments or stdin to your mailer.
If you use copy/paste, things get badly mangled. If that's a problem,
let me know and I'll set up a temporary ftp drop area.
>> Also what promoted from x days ahead?
>> (planner: Full dump of admin1.corp.walid.com:sda9 promoted from 4 days
>> ahead.)
Again, in the normal setup (definitely **not** Denise's :-), Amanda
would try to "balance" the full dumps across the dumpcycle so roughly
the same amount of work is done each day. It's just telling you that it
is trying to do that (move a full dump earlier than expected). In your
case it doesn't matter because you don't have a tape in the drive and
any full dumps planner schedules will be ignored in favor of the degraded
mode partial.
>>Denise Ives
>Chris Karakas
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
