CF11 Linux AWS JVM issue causing database connections spike and unresponsive server (fixed)

We had a troublesome intermittent ( 3x weekly ) fault over a six week period which usually occurred under peak load and was only temporarily "reset" by a CF restart.

1. First the AWS Linux box CPU spiked to 100%
2. Then CF struggled to make database connections and the database connections piled up
3. Requests delayed upto 20 seconds +
4. CF Server unresponsive and needed restart
5. CF Restart slow ( 5m + )

With help from a CF expert the problem was tracked to the inability of the JVM to get sufficient entropy

The fix is

For clarification we simply change the highlighted arg from securerandom.source=file:/dev/random

To -Djava.security.egd=file:/dev/./urandom

and restart Coldfusion!