--On Monday, April 05, 2004 16:45:19 -0400 Cory Visi <merlin AT phear.lwz DOT
org> wrote:
> I am having a problem with a specific backup set not making use of my
> holding disk. This is not a problem with my other backup sets.
>
> I have a particular archival backup set that is far from an ideal setup.
> In this archival set, the basic idea is to do a full backup of all
> partitions and fit as much as possible onto 1 tape. A full backup of all
> my partitions will not fit on 1 tape, though, but the idea is that they
> will be left in the holding disk so I can manually run amflush in the
> morning.
>
> I have used the following methods to ensure a full backup despite the tape
> size restriction:
>
> All my dump types have:
>
> priority high
> dumpcycle 0
>
> I execute the following before each archive set dump:
>
> amadmin archive force
> amadmin archive force-no-bump
>
> And finally, I have set:
>
> reserve 100
Here's your problem. You are telling Amanda to reserve 100% of your
holdingdisk space for incremental backups, but since you are doing
just full backups Amanda isn't using it.
Try changing it to a small number or zero.
Frank
>
> Amanda does an excellent job of getting as much as she can onto the first
> tape. The leftover partitions, however, seem to disappear. They are not
> left on the holding disk. Here are the relevant log messages:
>
> pondering server02.example.com://SERVER01/BACKUP-E... next_level0 0
> last_level -1 (due for level 0) (new disk, can't switch to degraded mode)
> curr level 0 size 1127294 total size 1127390 total_lev0 1127294
> balanced-lev0size 1127294
> pondering server02.example.com:sda1... next_level0 0 last_level -1 (due for
> level 0) (new disk, can't switch to degraded mode) curr level 0 size 1138527
> total size 5692392 total_lev0 5692168 balanced-lev0size 5692168
>
> DELAYING DUMPS IF NEEDED, total_size 5692392, tape length 3904512 mark 0
> planner: FAILED server02.example.com //SERVER01/BACKUP-E 20040330 0 [dumps
> too big, 1127294 KB, but cannot incremental dump new disk]
> planner: FAILED server02.example.com sda1 20040330 0 [dumps too big, 1138527
> KB, but cannot incremental dump new disk] delay: Total size now 3426507.
>
> These partitions do not get dumped to the first tape, nor do they get
> stored on the holding disk. I guess this may happen because Amanda knows
> they are not going to fit on the tape, so she does not even try to dump
> the two above partitions. If this is the case, what is the best way to
> handle this situation? Should I create a fake tape size with a much larger
> tape length? Perhaps I could even set the fake tape length to twice the
> tape length (i.e. 2 tapes) to add some redeaming amount of correctness to
> this sloppy setup.
>
> Any other ideas would be much appreciated.
> Thank you in advance to anyone that spends any of their valuable time
> reading or responding to this request for help.
>
> Thank you,
> Cory Visi
--
Frank Smith fsmith AT hoovers DOT com
Sr. Systems Administrator Voice: 512-374-4673
Hoover's Online Fax: 512-374-4501
|