Skyen's Guide to Server Maintenance Best Practices

No Comments

Server Maintenance Best Practices Established from Working With Hundreds of Happy Customers

analogue-blur-business-159282From file servers to application servers to web servers, running a successful company requires a significant amount of technology. Like anything else in this life, though, these tools need maintenance. If appropriate upkeep of servers isn't performed internal resources or outsourced experts, they'll start to slow and eventually fail at some point. Fortunately, the following server maintenance best practices can help ensure your company stays up and running with minimal interruption.

Schedule Maintenance to Minimize User Impact

While servers continue doing their job regardless of whether users are utilizing them, it's important to schedule maintenance when it will have the least effect on those users. Whether this is between 9 p.m. and 5 a.m. because the company is closed or prior to noon due to fewer customers calling in, user impact should be a huge concern.

This is especially true when mission-critical functions will be down. Research has shown that a single hour of unplanned downtime costs around $82,864 on average. This is due to its huge impact on users, so even if you plan maintenance well in advance, make sure user impact is minimized.

Communicate Well in Advance

The worst thing that can happen during server maintenance is employees or clients running into issues they never saw coming. A little communication can go a long way. After all, would you be madder at McDonald's if they told you your food would take 15 minutes or if they just let you stand at the counter having no clue what's going on for 15 minutes?

Of course, how does one know how far in advance to communicate maintenance? The University of Northern Iowa has a smart plan in place based on how long the expected downtime will be:

  •          No anticipated outage: At least 24 hours warning.
  •          Outage for less than 60 minutes: At least 36 hours warning.
  •          Outage 1 to 6 hours: 48-72 hours warning.
  •          Outage more than 6 hours: 7 to 10 days warning.

If these guidelines are abided by at a minimum, the impact on your company will be minimized.

Utilize Automated Patching Tools

Intensive maintenance procedures require advanced warning and focus on users, but automated patching tools can be utilized much more simply. These tools, such as Windows Server Update Services (WSUS), allow administrators to distribute hotfixes and updates after they're downloaded from the web.

The simplicity of automated patching tools is far from their only benefit. One of the most advantageous is the ability to distribute patches across entire networks. This saves time and reduces the amount of necessary rework due to mistakes. Automated patching tools are easily one of the most efficient ways to keep servers running smoothly.

Mute Network Monitoring Software to Reduce Alerts

Your network monitoring software is typically meant to alert you when something has gone awry. During maintenance, though, false alarms can continuously come through. These repeated alerts will quickly stress out users, and repeated studies have shown that more stress leads to less productivity.

Additionally, "notification fatigue" can eventually occur when users keep receiving these false alarms. This makes them more likely to ignore actual threats that they may come across. It quickly turns into "The Boy Who Cried Wolf," so muting the network monitoring software is ideal.

Plan for Failure

One of the most important server maintenance best practices is simply realizing that failure is going to happen. The Canadian Medical Association Journal even said that medical record servers, something most people couldn't imagine being without, face "inevitable crash."

Even with great maintenance, your server is going to eventually crash. In fact, an unexpected crash could occur during routine maintenance. The best way to fight this is through mitigation. Have resources in place, such as manual credit card imprinters or data redundancy, that can reduce the effect of an eventual failure.

User Acceptance Testing

Regardless of how vigilant you are with server maintenance best practices, there's always the chance that something could have gone wrong. Maybe a new patch has an unknown vulnerability. Maybe the maintenance you undertook created a compatibility issue between different systems. Regardless of the potential issue, user acceptance testing is the best way to track it down.

User acceptance training allows the people who will be using your system to test its effectiveness and stability in real-world scenarios. If something no longer performs to expectation, there's a chance that something could've gone wrong. By utilizing user acceptance testing, you can identify and rectify the cause.

Communicate Success

One of the most overlooked aspects of server maintenance is the communication of success. Unfortunately, it's also one of the most important. Organizational leadership and stakeholders don't always realize how important proper care is for a server. This often results in server maintenance being an afterthought.

When stakeholders and organizations don't recognize the importance of server upkeep, entire networks can fall into disrepair and face security vulnerabilities. This makes it important for those "in the know" to close out the process and communicate the process's success to external and internal stakeholders. When they finally realize how important server maintenance is, this integral process will face less resistance in the future.

If you don't utilize outsourced IT, you'll still need to periodically perform server maintenance. Without it, you'll miss out on enhanced features, encounter security issues, reduce efficiency and even face costly downtime more frequently than necessary. They say an ounce of prevention is worth a pound of cure, and in the worlds of business and information technology, this adage couldn't ring truer.

Interested in learning how Skyen can take the risk out of managing servers?  Contact us for a free evaluation and consultation.