Running JMXTrans without the scheduler (single run)

[Mika Arasola]

I've been running JMXTrans for ages with good results, but ended up with a new use case where I would need to:

- Collect JMX statistics on an external flexible schedule

- Control that schedule outside Jmxtrans

- Preferably make it very portable by directly calling a jar file and point to configs rather then have a static installation

As an example have an AWS Lambda that would start, do one round of collection and then exit. 

I've been trying to read through all the docs and examples but they are a bit sparse on the topic... I'm wondering if API Mode for example could support this kind of functionality without totally gutting the code?