Nik wrote:
The reason I am running pg_dumpall is that I would like to create an
exact copy of my database on another server. I cannot use create
database with template command since there are connections to the
original database. The dumpall method seems somewhat cumbersome since
it generates only one file that is huge.
Is there another way I could do this that would be faster, more
reliable, and elegant? Could I start archiving the logs for the
original database and then do the pitr on another machine using that
backup and logs?
On the second machine
pg_dumpall -h first_machine | psql
Of course that is assuming the second machine is a unix clone. I don't
know if windows pipe works like unix or if it still works like the old
dos pipe.
---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match
