New candidate JEP: 407: Remove RMI Activation

[Peter Firmstone]

On 5/05/2021 5:55 am, Stuart Marks wrote:
> As near as we can tell, there is vanishingly small usage of RMI 
> Activation in the Java community in general. RMI Activation (and RMI 
> itself, for that matter) have not evolved in the past 15 years or so. 
> Furthermore, there has been virtually no demand for new features or 
> evolution of RMI-related technologies for most of that time. It's hard 
> to escape the conclusion that this it's all obsolete.


Yep, RMI isn't much chop by today's standard, I would also say that it's 
obsolete, old and clunky.   The only use we have for RMI is for RMI 
registry and maybe visualvm.

I did write TLS Socket factory's for RMI, that utilize the calling 
thread's subject for authentication so it can be used with JAAS.

We use something far better than RMI today, but it's not well known, 
which allows us to use Kerberos, or TLS1.3 (using session ticket 
extensions) and preserve the calling context between JVM's, so that 
multiple JVM's work together as one.

https://github.com/pfirmstone/JGDMS/tree/trunk/JGDMS/jgdms-jeri

This JEP doesn't affect us as much as JEP 411, we could have removed our 
activation implementation, which has continued to evolve, but with JEP 
411, Java's future migration path is unclear, I don't think this matters 
anymore.

-- 
Regards,
  
Peter Firmstone