[DISCUSS] - Camel 2.0 - About Endpoints with lenient properties eating to much memory

2009-06-29 Thread Claus Ibsen
Hi Tickets == I am looking into two issues related to Camel eating memory https://issues.apache.org/activemq/browse/CAMEL-1771 https://issues.apache.org/activemq/browse/CAMEL-1738 The problem = It all boils down to using a lot of http endpoints with unique urls. So over time Camel ac

Re: [DISCUSS] - Camel 2.0 - About Endpoints with lenient properties eating to much memory

2009-06-30 Thread Willem Jiang
Hi Claus, I think we need to implements a method for get unique name from the endpoint. It will base on the endpoint's parameter not the customer parameter. With this method we could avoid to register to much temp endpoint with different customer parameter. In this case we need find a way to

Re: [DISCUSS] - Camel 2.0 - About Endpoints with lenient properties eating to much memory

2009-06-30 Thread James Strachan
I'm liking a) and allowing c) (a single mbean for all endpoints of a certain type) make sense to me. Using an LRU cache for non-singleton endpoints makes sense too; though I wonder if an endpoint might want to mark itself as not cachable if it knows its not really worth the trouble (as it has no s

Re: [DISCUSS] - Camel 2.0 - About Endpoints with lenient properties eating to much memory

2009-06-30 Thread Claus Ibsen
On Tue, Jun 30, 2009 at 11:15 AM, Willem Jiang wrote: > Hi Claus, > > I think we need to implements a method for get unique name from the > endpoint. It will base on the endpoint's parameter not the customer > parameter. With this method we could avoid to register to much temp endpoint > with diffe

Re: [DISCUSS] - Camel 2.0 - About Endpoints with lenient properties eating to much memory

2009-06-30 Thread Claus Ibsen
On Tue, Jun 30, 2009 at 1:50 PM, James Strachan wrote: > I'm liking a) and allowing c) (a single mbean for all endpoints of a > certain type) make sense to me. Just for the record. Option a + c is currently implemented and reported as fixed the issues by the end user. > > Using an LRU cache for n