Capstone - Study on High Availability and Fault Tolerance application

This repository is for Research Topic on High Availability and Fault Tolerance.

1. Paper
2. Poster
3. Presentation

Repository

#	Component	Usage
1	Load Agent	Record Resource Agent resource metrics and notify Recovery Agent if any resource agent crashed
2	Load Balancer	Customer facing application to redirect to resource agent based on Resource Awareness Routing Algorithm
3	Recovery Agent	Perform Resource Agent Eviction and Recovery once receive Load Agent recovery notification
4	Resource Agent	Sample Application to run a workload, with a embedded resource agent to connect to Load Agent for status update

Prerequisites docker image

1. Install Redis as caching server
2. Install Elasticsearch as log collector
3. Install Kibana as log viewer
4. Create a docker network for communication between components

Create Capstone Network

docker network create capstone --driver bridge

Install Redis Docker Image

docker pull redis docker run --network capstone --name redis -p 6379:6379 redis docker run --network capstone --name redis-cli --rm -it redis redis-cli -h redis

Install ELK Docker Image

1. Elastic Search

docker pull docker.elastic.co/elasticsearch/elasticsearch:8.3.1
docker run --name es01 --net capstone -p 9200:9200 -p 9300:9300 -it docker.elastic.co/elasticsearch/elasticsearch:8.3.1

2. Kibana

docker pull docker.elastic.co/kibana/kibana:8.3.1
docker run --name kib-01 --net capstone -p 5601:5601 docker.elastic.co/kibana/kibana:8.3.1

Setup the ELK Cluster

1. Reset Elastic user Password

docker exec -it es01 /usr/share/elasticsearch/bin/elasticsearch-reset-password -u elastic

2. Generate Kibana Connection Node

docker exec -it es01 /usr/share/elasticsearch/bin/elasticsearch-create-enrollment-token --scope kibana