In CastleServiceLocator.cs, the code which resolves message consumers looks
like this:
return (from h in container.Kernel.GetAssignableHandlers(type)
select (IHandler)new DefaultHandler(h.Service,
h.ComponentModel.Implementation, () =>
h.Resolve(CreationContext.CreateEmpty())));
The empty CreationContext references the NoTrackingReleasePolicy, which
prevents message consumers (and their dependencies) from being cleaned up
after message consumption (although RSB calls
container.Kernel.ReleaseComponent(item), it has no effect since the object
is not tracked).
In my scenario, a few message consumers need to call a WCF service, and
receive a proxy to this service via constructor injection. The consumers
themselves are simple classes that do not require cleanup. The proxy must
be closed, however, and I wanted to implement the logic as an
IDecommissionConcern. Unfortunately, the above problem prevents the cleanup
code from being invoked.
I have implemented a fix, which creates a CreationContext that references
the ReleasePolicy configured for the container:
() => h.Resolve(new CreationContext(h,
container.Kernel.ReleasePolicy, type, null, null, null))
It seems to be working and tests are green. I'm curious, however, what was
the original reason for using an empty CreationContext in the resolving
code and whether the reason still stands (and I'm missing something)?
--
You received this message because you are subscribed to the Google Groups
"Rhino Tools Dev" group.
To view this discussion on the web visit
https://groups.google.com/d/msg/rhino-tools-dev/-/4BmEA9QBD58J.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/rhino-tools-dev?hl=en.
