Take a last backup before termination

[majst01]

currently only during backup-intervall a backup is taken and stored. But if the pod is terminated there is a time gap between the last backup and the last commited transaction of the database. If then the disk dies, these last commits are lost.

At least in the case of normal pod termination we already get the termination signal. With this there is a chance to do one additional backup.

Downside: termination may take longer.

[Gerrit91]

Idea makes sense, but I would also see this as an additional feature, which can be turned on or off.

Reasoning: We have a special use case, in which we want two databases in the metal-stack (metal-db and ipam-db) to be in sync. This is why we take backups for those two databases with an identical time schedule. Your proposal works a bit against these "synchronous backup schedules".

[majst01]

Yes sure, must be configurable

[majst01]

Maybe also consider a dumpall as describe in #119