So I'm starting development on a project with multiple components which will
be talking to an Oracle 8 database, and I'm trying to understand what the
various options are.
The project will consist of HTML::Embperl pages, modperl CGIs, and random
Perl processes, all of which will connect to the database.
The questions I have mostly pertain to database connection caching. There
seem to be three different modules that all purport to cache the database
connection: DBI itself (using connect_cached), Apache::DBI, and DBD::Proxy.
What are folks out there using for the connection pooling? It's quite
tempting to use the DBI connect_cached method, so that I can use the same
pooling calls in all three places (Embperl, modperl, and other perl code),
but I'm worried by statements like "The exact behaviour of this method is
liable to change" in the documentation.
I'd really appreciate any tips/pointers folks could provide!
-jse
