On Wednesday 12 May 2010, Jason Frisvold wrote:
>Over time, the overall size of my backups has increased a bit and, in lieu
> of getting additional space, I decided to lower the number of backups we
> keep. As it ends up, we didn't really need quite that much... I edited
> the amanda.conf file and updated it with the new number of tapes in
> rotation. My config looks like this now :
>
>dumpcycle 7
>runspercycle 5
>tapecycle 20 tapes
>
>I also nuked the directories on disk for slots 21+. It appears that things
> are backing up properly using this new cycle, but there also seems to be
> some residual data in the indices. What else should I update to resolve
> this?
>
This is doing it both the hard way, and the leave nagging garbage laying
around way.
IIRC, and its been yonks since I last played with it, the proper command is
amrmtape, see the manpage for exact syntax. This not only removes the tape
from the active list, but also removes all the indice etc references to it,
leaving you with a clean system.
With all the rewrites amanda is getting in the last 2-3 years, I would hope
that it will not complain when it doesn't find something it should nuke, but
will go ahead and remove that which still exists, therefore cleaning up the
system for you.
>Also, amcheck seems to be complaining about labels and active tapes now.
> Is this a problem?
See above, amrmtape should clean that up too.
>[r...@example DailySet1]$ sudo -u amanda /usr/sbin/amcheck DailySet1
>Amanda Tape Server Host Check
>-----------------------------
>Holding disk /var/amanda: 9361724 KB disk space available, using 9259324 KB
>slot 17: read label `DailySet1-17', date `20100428004501'
>cannot overwrite active tape DailySet1-17
>slot 18: read label `DailySet1-18', date `20100429004501'
>cannot overwrite active tape DailySet1-18
>slot 19: read label `DailySet1-19', date `20100430004502'
>cannot overwrite active tape DailySet1-19
>slot 20: read label `DailySet1-20', date `20100501004502'
>cannot overwrite active tape DailySet1-20
>slot 1: read label `DailySet1-01', date `20100502004501'
>cannot overwrite active tape DailySet1-01
>slot 2: read label `DailySet1-02', date `20100503004501'
>cannot overwrite active tape DailySet1-02
>slot 3: read label `DailySet1-03', date `20100504004502'
>cannot overwrite active tape DailySet1-03
>slot 4: read label `DailySet1-04', date `20100505004501'
>cannot overwrite active tape DailySet1-04
>slot 5: read label `DailySet1-05', date `20100506004501'
>cannot overwrite active tape DailySet1-05
>slot 6: read label `DailySet1-06', date `20100507004501'
>cannot overwrite active tape DailySet1-06
>slot 7: read label `DailySet1-07', date `20100508004502'
>cannot overwrite active tape DailySet1-07
>slot 8: read label `DailySet1-08', date `20100509004502'
>cannot overwrite active tape DailySet1-08
>slot 9: read label `DailySet1-09', date `20100510004501'
>cannot overwrite active tape DailySet1-09
>slot 10: read label `DailySet1-10', date `20100511004502'
>cannot overwrite active tape DailySet1-10
>slot 11: read label `DailySet1-11', date `20100512004501'
>cannot overwrite active tape DailySet1-11
>slot 12: read label `DailySet1-12', date `20100423004502'
>slot 13: read label `DailySet1-13', date `20100424004501'
>cannot overwrite active tape DailySet1-13
>slot 14: read label `DailySet1-14', date `20100425004501'
>cannot overwrite active tape DailySet1-14
>slot 15: read label `DailySet1-15', date `20100426004501'
>cannot overwrite active tape DailySet1-15
>slot 16: read label `DailySet1-16', date `20100427004502'
>cannot overwrite active tape DailySet1-16
>read label `DailySet1-12', date `20100423004502'
>NOTE: skipping tape-writable test
>Tape DailySet1-12 label ok
>Server check took 6.050 seconds
Humm, I have never seen it continue after looking at -12 and finding it can
use it. Strange. OTOH, tape drive problems have made me wish that it would
actually check tapecycle+1, effectively rechecking the desired and correct
tape. I have had it fail to read the next tape, find no usable tape and
report the failure. But cycling the mechanism with a repeat of amcheck
always worked. 100% of the time.
>Amanda Backup Client Hosts Check
>--------------------------------
>Client check: 18 hosts checked in 4.062 seconds, 0 problems found
>
>(brought to you by Amanda 2.5.1p3)
Your elapsed times there indicate you may be using vtapes? When I was using
a changer, it was often over a minute per tape inspected. Slow, cheap,
seacrate DDS2 changer. For me, vtapes on a terrabyte HD have been dozens of
times more dependable, and since its random access, about 50 times faster
when doing a recovery. One drive failure in about 5 years now, and smartd
warned me about it in plenty of time to take corrective replacement action
which included rsync'ing the failing drive to the new one. Amanda never got
a hint there was a problem.
>Thanks,
>
>---------------------------
>Jason 'XenoPhage' Frisvold
>Engine / Technology Programmer
>f...@godshell.com
>RedHat Certified - RHCE # 803004140609871
>MySQL Pro Certified - ID# 207171862
>MySQL Core Certified - ID# 205982910
>---------------------------
>"Something mysterious is formed, born in the silent void. Waiting alone
>and unmoving, it is at once still and yet in constant motion. It is the
>source of all programs. I do not know its name, so I will call it the
>Tao of Programming."
--
Cheers, Gene
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Great American Axiom:
Some is good, more is better, too much is just right.
