Niclas Hedhman wrote:
On Thursday 09 October 2003 02:50, Stephen McConnell wrote:
Under the current Startable model, the container has no way of
assuring service availablity. I.e. under the current interfaces its the
component to container communication between startup and shutdown that
remains undefined.
Back to the discussion surrounding dynamic service availability, asynchronous binding of services and allowing clients to handle intermittent disruption of service...
Exactly!
So if that was in place, for instance a thread crash would just allow the service to announce that it is not available, do a restart of itself (with help of container I guess) and then announce available later.
Its the service availablity question that *is* a container concerns becuase its the container that is taking on the responsibility of wiring up solutions and by inference - solution assurance.
Stephen.
p.s. What is the title of your recent aquisition?
--
Stephen J. McConnell mailto:[EMAIL PROTECTED]
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
