Mark Mikulec <mark(at)anticentertainment(dot)com> wrote:
> The strange thing is that this started after my database grew by
> about 25% after a large influx of data due to user load
In addition to the issues already mentioned, there is the fact that
to maintain consistency an entire database must be dumped in a
single database transaction with one snapshot. This means that
garbage collection can't run, which may lead to bloat under some
circumstances. This may be why your database grew by 25%. If that
bloat is concentrated in a small number of tables, you may want to
schedule aggressive maintenance (like CLUSTER) on those tables.
One other factor which can affect running applications is the table
locks which the dump must hold.
You might want to look into PITR backup techniques, or streaming
replication on 9.0
-Kevin