By Mike Talon
Recently, while I was holding for the technical support staff of my mobile phone carrier, I finally understood what clients have been asking me for years. When cell networks drop and pagers aren't working properly, how do you get the word out that something has gone wrong? Nearly every tech shop has alerts and messages routed to wireless devices, but what happens when the person in charge can't make a call?
This question has many potential answers. Perhaps the most obvious solution is to ensure that you never have only one person responsible for delivering or receiving the alerts at any given time. While this method is fine for larger organizations, it isn't feasible for mom-and-pop shops that may have only one technical employee. However, there are other alternatives to make sure your data systems are able to reach out and touch someone.
For example, consider using multiple devices. While most IT pros don't want to look like Batman with a utility belt full of gadgets, carrying a pager to receive alerts and a mobile phone or other device to make and take calls can be a winning combination. Pagers are not expensive these days, and monthly fees are pretty low in most areas. It's a cost-effective method to ensure you don't inadvertently get cut out of the communications loop.
Another option is to utilize software that allows multiple paths for alerts based on urgency and response, such as IBM's Tivoli and HP's OpenView. There are a myriad of choices for this type of monitoring and alert software, and many that are tailored to specific types of data systems. These packages page backup technical staff in the event the employee who's on call doesn't respond.
In addition, this type of software pages nontechnical personnel when all else fails. Based on the feedback from my last column, many of you utilize nontechnical staff who are trained to handle minor issues in remote offices. These are the people who can be put into the pager rotation to respond if no one else answers the call. Again, this should be considered a last resort effort.
The solution that works best for your organization will be different than what works for others. However, events have consistently proven that single points of failure are never a good thing. That one cell phone you're counting on to send or receive a message could very well become a single point of failure—one that's avoidable with a little work.
Mike Talon is an IT consultant and freelance journalist who has worked for both traditional businesses and dot-com startups.