Network Monitoring System

⦁ The first step will involve using VMware to simulate a virtual environment where four virtual machines (VMs) will be used to create an architecture on which sensing operation will be performed using one of the network tools. For example: Nagios. The deliverables of this step will be web servers that will be load balanced by the third VM and the fourth will be a database server. ⦁ The second step will involve creating a fifth VM that will act as storage for the above architecture on which the metrics will be collected using the network tool Ganglia. ⦁ The third step will involve creating the sixth VM on which monitoring software will be installed. Analysis of the above collected metrics will be performed to create alerts and visualize the network anomalies. The tools used for analysis, alerting and visualization might be Sensu, Nagios or Graphite respectively. ⦁ Selection of the above network tools might change depending upon which are open-source, compatibility and which combination of the tools perform all the basic problems of monitoring. For testing the above said network monitoring system, I will be using software like SolarWinds Wan killer. I will be using it to generate network traffic on web servers, database servers and load balancing using the internet protocol (IP) address or the hostname. In addition, it will also allow me to set the packet size and the network load to the needed value and specific percent of the network bandwidth respectively. ⦁ Secondly, I will be using the network tool Nagios on the web servers, database server and load balancer to perform sensing. Nagios will allow me to gather data using metrics like type of network traffic (HTTP, DHCP), processor load, memory load, disk usage, different operating systems etc. from the traffic generated in the last step with the help of different plugins. ⦁ Thirdly, I will be using the network tool Sensu on my storage system to perform data collection from Nagios. Sensu will be collecting raw data by creating metrics collection check with the help of plugins for example; processor utilization check. In addition, I will be using Sensu to perform analysis as well on the raw data collected by creating a metrics analysis check. Nagios provides compatible plugins for Sensu, which will allow me to customize checks and directly analyze the output. ⦁ Fourthly, I will be using Nagios to implement alerting. The alerting system in Nagios will allow me to set an alert policy such that alerts are issued whenever there is any system failure, recovery, disk usage threshold, processor overload etc. through emails or messages. Also, the alerting system will allow me to customize alerts such that alerts get forwarded to other people if they are not handled on time, alerts are not issued when the network is intentionally brought down and timely reports are generated which gives detailed explanations on how alerts were handled. In addition, I will be using Nagios to perform visualization. Results generated will be represented in the form of graphs like which systems need to be upgrade soon, which disks need to be replaced or which services need to be improved to reduce the delay when users interact with them. So that management can easily understand them and they can take action before network failure occurs.