Hello,
I've a quick question regarding server-side interface implementation -
say servant - lifecycle.
If I have an interface INTF and an implementation IMPL, and call through
Resin into it from multiple hosts different operations, can someone
please describe accurately what happens ?
The way I see it is this:
- Resin creates a HessianServlet instance
- requests are called from different IO threads into the same servlet
instance
- which in turn calls into my object
How many servant objects will there be ?
The problem is that in a method in a service I'm acquiring an expensive
resource - a JMS producer wired to Oracle AQ. This is rather expensive
to come by, and I'd like to know the whole model before I start adding a
connection pool for the JMS producer JDBC connection needs (as I figure
re-connecting to the database is the most expensive thing it does).
I will look into the HessianServlet source code again, but I thought I'd
ask the experts (hint: Scott, please reply!)
Kind regards,
-- Radu-Adrian Popescu CSA, DBA, Developer Aldrapay MD Aldratech Ltd. +40213212243
This archive was generated by hypermail 2.1.8 : Thu Sep 28 2006 - 20:16:40 PDT