> Are you saying that?:
>
> 1) You ran pg_basebackup against a live cluster and sent the output to
> another location.
>
> 2) At the other location the cluster is not in use.
>
> 3) You want to grab the contents of the inactive cluster directly off the
> disk.
>
> If that is the case, then no it is not possible without making the cluster
> live.
>
> If you mean something else then more details are needed.
Sure.
1 - You ran pg_basebackup on node-1 against a live cluster and store
it on NFS or tape.
2 - Do a restore on node-2 from the backup taken on (1), but only for
a subset of the database
(schema/database)
3- Put the cluster live on node-2 after (2) completes. Essentially the
cluster will now be a small
subset of cluster on node-1.
Benefit: If I have to restore only 5% of entire db, it should be lot faster.