Automatic and decentralized discovery and monitoring of nodejs instances with built in support for a variable number of master processes, service advertising and channel messaging.
In case if one node dies and two nodes remain, then almost always we have two masters. And we have these events:
1: promotion
2: promotion
2: master
1: master
In case if one node dies and two nodes remain, then almost always we have two masters. And we have these events: 1: promotion 2: promotion 2: master 1: master