How to respond to remote IT alerts
Thursday, May 30th 2013
The environmental monitoring systems are in place. Heating and cooling processes have been set up for maximum efficiency. Temperature monitoring systems track any fluctuations in the server room climate and generate alerts if any link in the chain goes awry. The data storage center is optimally protected, right? Not so fast. What happens when an alert occurs? If a company doesn't have well-defined support hierarchies and contingency plans, they risk having servers compromised by the same issues that remote monitoring systems are designed to protect. It's essential for organizations to implement a thorough, efficient response to potentially threatening incidents.
Establish formal procedures
If a remote sensor sends an alert, IT and management personnel should be able to spring immediately into action. Knowing what type of problem is affecting the server room can inform the method of action, whether it is something to be handled internally or whether outside emergency authorities need to be notified. According to Data Center Knowledge, power usage effectiveness is one area that most frequently causes headaches and IT downtime. Power outages are often caused by external factors, and so it's essential that steps be taken to manage power, like shifting data operations to a different storage facility in a separate geographic location. For incidents like these, there should be a chain of command that includes enough IT personnel and superiors that at least one is available at all times.
Contact third party providers
The unavoidable truth about data center monitoring is that in an optimal environment, servers are always running since data never sleeps. It's crucial to develop policies for responding to system failures at night or on weekends, when employees are less likely to be remotely connected. If there is a problem with the system, third party service providers would have to be contacted immediately. A remote power manager solution, like the ones ITWatchDogs offers, can escalate its warning systems based on the severity of the dilemma. For example, a small rise in temperature could enable alerts to be sent to some personnel, while an escalation could trigger alerts to more internal personnel and contractors as well.
Predictive environmental management is better than a more reactive approach. If an alert happens, make sure to diagnose the source of the problem and take steps to stop it from happening again. If any problem in procedure or chain of command hampered the resolution of a issue, re-evaluate contingency plans so that if the alarm goes off again, it can be responded to as efficiently and capably as possible.