Skip to content

K-OpenNet/K-ONE-M-CORD-Lite-Monitoring

Repository files navigation

K-ONE-M-CORD-Monitoring

Note that

  • containers of M-CORD Lite are implemented by K-ONE M-CORD Lite
  • influxDB container for time-series database and Grafana container for dashboard are used
  • Scapy for monitoring network traffic statistics
  • monitor contianer and shell scripts are implemented by Jibum Hong @ POSTECH

Specification

  • OS: Ubuntu 16.04 and 18.04
  • Docker version 18.09.7
  • Prometheus version 1.7.0
  • Grafana version 4.4.5
  • InfluxDB version 1.7.9
  • Scapy version 2.3.3

Preliminaries

1. Need to install Docker engine

node$ ./set_env.sh

NOTE: To make sure that the user is in the docker group, please log off and then log in again. You can check which groups belong to the user with the below command.

node$ groups

If you can see the group docker, everything looks good!

NOTE: If you install K-ONE M-CORD Lite, you don't need to this process.

Install

1. Download and run node-exporter and cAdvisor (Compute Nodes)

node$ ./set_each_node.sh
  • cAdvisor for container resource usage monitoring
  • node-exporter for PM resource usage monitoring

2. Configure prometheus configuration files (Control Node)

  • Go to conf/prometheus.yml and change <IP_number> to node IP addresses running node-exporter
  • Go to conf/prometheus-cadvisor.yml and change <IP_number> to node IP addresses running cAdvisor
  • Go to conf/prometheus-collectd.yml and change <IP_number> to node IP addresses running collectd

3. Download and run Prometheus for node-exporter, cAdvisor, and collectd (Control Node)

node$ ./set_prom.sh

The role of InfluxDB was replaced to Prometheus TSDB for compatibility.

4. Download and run Grafana (Control Node)

node$ ./set_grafana.sh

you can access to grafana web dashboard <IP_number>:3000

initial id : admin / pw : pass

  • Go to "Settings -> Data sources", and Add data sources for node-exporter, cadvisor, and collectd
Name: prom-cadvisor / prom-exporter / prom-collectd
URL: http://prometheus-cadvisor:9090 / http://prometheus:9090 / http://prometheus-collectd:9090
Scrape interval: 1s
HTTP Method: GET
  • Go to "Create -> Import", and paste node-exporter.json and cadvisor.json

Release information

  • Release 1 - Monitoring system resource metrics of each containers in EPC
  • Release 2 - Monitoring network statistics of internal link in EPC
  • Release 3 - Anomaly Detection using Monitoring System

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published