PengZheng commented on issue #515: URL: https://github.com/apache/celix/issues/515#issuecomment-2017176409
> The downside is that, from a remote perspective, the events are always treated as a postEvent. However, considering that remote calls can lead to long blocks and possible timeouts, this approach might be preferable. In some rare cases, where performance is not a concern, we still need sync semantics. For example, a master machine want to make sure all slave machines are ready to handle master shutdown before it actually performs the action. > Originally, the idea with PubSub was to build remote services on top of it (the C++ Remote Service Admin even had an integration test based on this). Yes, conventionally messaging middle-wire lies below RPC, i.e. RPC can be implemented with messaging middle-wire. Our EventAdmin is still not general enough to be used to implement RPC. For example it still lacks serialization mechanism. And our RSA lacks async interface, which makes it impossible to implement EventAdmin efficiently. Event if RSA has such support, whether it is a good idea to go that way is still debatable. I think the main motivation is to unify the service discovery of RSA/REA(RemoteEventAdmin). Currently an empty marking interface with its meta data can be easily advertised by service discovery of RSA. If the resulting coupling between RSA and REA is not desirable, we could instead try to make service discovery independent of RSA. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: dev-unsubscr...@celix.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
