As I understand from a previous post: most of the code in B4J is executed on a single thread (except for example the webserver implementation).
I would like to create a custom POP3 server (the protocol is very easy and I already have written the B4J code for it). To be production-safe it seems this kind of server should be written in a multithreaded way, so that a single connection can't block other connections.
Is there anyway way to do so in B4J ? I already though (as a workaround) to spawn different processes for each connection, but this is:
1. ugly
2. hard to implement as only a single process can listen on a specific TCP port.
Thanks for any hints.
I would like to create a custom POP3 server (the protocol is very easy and I already have written the B4J code for it). To be production-safe it seems this kind of server should be written in a multithreaded way, so that a single connection can't block other connections.
Is there anyway way to do so in B4J ? I already though (as a workaround) to spawn different processes for each connection, but this is:
1. ugly
2. hard to implement as only a single process can listen on a specific TCP port.
Thanks for any hints.