Building an orchestrator and frontend for cluster health visualization
- A cluster state dashboard
Download activator from TypeSafe or use activator bash script.
Run each line in a new terminal.
activator "project orchestrator" "run 2551"
mvn hpi:run (ensure port is not clashing and change cluster system to HajpCluster from ClusterSystem)
HAJP orchestrator proxy requires HAProxy to be installed and available to running user. for Mac OS: $ brew update $ brew install haproxy
for CentOS: $ yum install haproxy
for Ubuntu: $ apt-get update $ apt-get install haproxy
Also it assumes the existence of an environmental variable called orchestrator_deploy which should point to absolute location of orchestrator/ folder.
Underneath orchestrator folder, there exists a folder named conf which contains two files:
haproxy.cfg which is the actively modified run-time ha proxy config file
haproxy.pid which is the actively modified run-time process id file
Both of these files need to be maintained to ensure userid etc. match system deployment settings
Proxy feature is fully implemented on port 5000 for testing purposes, modification of sample haproxy.conf is highly advised for production run.