I'm working on a web application which obtains data via a legacy system rather than DBI. Using DBI is (unfortunately) not an option. The machine where the web application runs cannot run the legacy system. Thus, the setup looks somewhat like this:
DB <=> DB Query (Legacy) <=> Apache/mod_perl <=> Client I was thinking to make the connection between DB Query and Apache/mod_perl using HTTP, so there will be an Apache running on the DB Query tier for access to the legacy system. What I'm interested in, is a persistent HTTP connection between the Apache/mod_perl tier and the DB Query tier. I'm thinking of a ChildInit handler which opens up a HTTP/1.1 connection to the apache sitting on the DB query tier. The connection would use keep-alive, which of course would be enabled and set to high timeout values on the DB Legacy tier. This is more or less what Apache::DBI does, just in this case it should be something like Apache::PersistentHTTPConnection. Unfortunately, I haven't found such a thing. Has anyone done such a thing before? Can someone point me to docs or modules which could help doing this? Or is this whole idea maybe just plain stupid? Are there obvious caveats I haven't thought of? Any comments are welcome.
