What Should You Monitor?
Anyone who works in IT has experienced the dreaded walk in the door to find that a key network component has gone down. Users and upper management are already hounding your desk asking when things will be back online and you are simply trying to figure out which component went down. Wouldn’t it be great if you knew what kind of storm was on the horizon and already be taking action to fix it? In my last blog post, I covered some valid business reasons for monitoring your IT systems. I am going to further explore monitoring by covering some key elements of the network to monitor.
In an ideal world, everything would be monitored; whether it is the server hosting your critical business apps or the wireless controller for the guest wireless network. Some organizations may have the budget and the staff to handle that, but that is not always going to be the case. What key elements need to be monitored? The answer is that it depends on the business and its critical business needs.
Let’s start with switching. The core is a good place to start. The core switch or switches should be monitored. If these go down, it will take down the rest of network. These are the heart. Without these, no traffic passes through your network internally or externally. As we make our way out of the core, we run into distribution switches. A little wiggle room may be had here. If there is a segment of the network that isn’t business critical, monitoring that distribution switch may not be as critical. This brings us to edge switching. These are probably the least critical. If one goes down or has an impending event, it will affect the least amount of the network. If that least amount happens to be the accounting department and it is payroll time, it may be necessary to know that ahead of time. The process here would be to start from the core and work out towards the edge and determine which items are most critical when deciding what to monitor.
Moving on to servers, an evaluation needs to be done on the criticality and available redundancy. It is important to understand the overall business impact of each server. There are some that will absolutely cripple the business if unavailable and others that will just be a minor nuisance. A good rule of thumb is to make sure you monitor any server that is going to keep you from doing business. If you are a point of sale business, whose cash registers connect back to a POS server, that server is critical to the business. The server that runs your email may not be critical. However, if you are a business that takes critical service tickets via email, the email server(s) just became a lot more important. The plan here is understand the critical business processes, know where those processes are being affected by a server, and monitor those servers.
Regarding firewalls and other security appliances, the instinct here is to say that these are critical and should absolutely be monitored. While I don’t disagree with that sentiment, it may not be practical. Once again, we come back to the needs of the business. The company firewall is most likely critical, but maybe the email and web filter can wait until the morning to tell of their travails.
What about other systems like video surveillance or telephone systems? Once again, it comes to the criticality of those components to the business. A video surveillance system is protecting a small administrative office may not be as important as a system watching over millions of dollars in inventory. If the business uses a call center to collect payments from clients, the phone system just became a lot more critical.
The below tips are a good starting point when evaluating what to monitor:
- Know which business functions are critical
- Know where there are redundancies in the network
- Know how each IT system affects the critical business functions
- Understand the impact of the loss of any given system to time and dollars
- Understand the budget constraints and where you can get the most bang for the monitoring dollar
Knowing what to monitor can be just as critical as the “why” of monitoring. There are some things that are going to be more critical than others. Choose wisely and your IT dollars can go further.
A proven leader, Jeremy Niedzwiecki has over 20 years in the IT industry. As the Director of Customer Support at ABS, Jeremy works to ensure that the ABS Customer Support team continuously provides the highest levels of support possible ABS clients.