Hi all.

I have a question about the retrial specifications of Amanda.

As far as I know, The amdump command retries a backup when
one of clients meets the following sudden troubles.

* Power down
* Death of Amanda processes by signals

I confirmed these behaviors in Amanda 2.6 and 3.3 and
I want to avoid any kind of retrial

To make the reason clear, I explain the situation I am in.

I am dealing with configuration of site wide backup by Amanda.
The site includes several DB-HA-clusters and so on.

In this backup plan, consistency of the backup data must be
needed. For instance, DBs must be backed up in ways that allow
PTR. To enable PTR, several steps of procedures are needed
before and after amdump.

I am afraid retrials to failure backups humper this process.
So I want amdump to exit with error status immediately whenever
any trouble happens during backup.

I have been reading Amanda documents to know what kind of
trouble causes retrials and how to avoid it, But not yet found
a clue.

If someone knows something about the infomation, please
point out it to me. I will be appreciated it.

Best regards.
-- 
Kazuhiro Takenaka <[email protected]>

Reply via email to