https://issues.apache.org/bugzilla/show_bug.cgi?id=50667
Filip Hanik <fha...@apache.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |NEEDINFO
--- Comment #3 from Filip Hanik <fha...@apache.org> 2011-01-27 11:33:32 EST ---
(In reply to comment #2)
> Created an attachment (id=26561)
--> (https://issues.apache.org/bugzilla/attachment.cgi?id=26561) [details]
> svn diff
>
> CAVEAT: this proposal modifies the RpcCallback interface. RpcCallback is
> currently implemented by at least four tomcat classes, EchoRpcTest in
> org.apache.catalina.tribes.demos, and AbstractReplicatedMap, ReplicatedMap ad
> LazyReplicatedMap in org.apache.catalina.tribes.tipis. The attached patch does
> not include the implementation of the additional interface method for those
> classes, as this coder did not feel qualified to tell how it should be
> implemented. They could probably simply be NOOPs, however.
I still don't understand this enhancement. What is the possibly use case for
implementing replyFailed.
The callback method replyFailed would get called to the receiver. And this is
backwards. There is nothing the receiver can do if reply fails. There are no
actions to be taken. It wont allow the receiver to retry the attempt. So the
method is at best a no-op, but really just a method one has to implement with
zero benefit. The sender still has to make a decision on what to do without a
reply.
best
Filip
--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org
