Prevent server downtime Do you know what to do?

Everyone with a car knows that regular oil changes and tire replacements help the car to travel more smoothly and last longer, thus protecting existing investments. Similarly, taking several precautionary measures can also help servers in the data center run more smoothly and last longer. More importantly, these measures help prevent server outages; in some cases, server outages can quickly lead to huge economic losses.

In order for your server to operate at its best, you need to maintain the right environment, proactively maintain and monitor it, and develop a plan so that if the server fails, the employee knows How to do it.

Maintaining a good and suitable computer room environment

Most large data centers have advanced equipment to monitor the physical condition of the data center, but many smaller companies have their server room not at the proper temperature. Or the humidity range, or not as clean as it should be. Joe Guenther, marketing specialist at TERiX Computer Services, points out that: “The system may still be able to operate in an environment that is too hot, too humid, or dusty, but they definitely do not reach the life expectancy and are more than recommended. Operating under environmental conditions, it is more susceptible to environmental conditions and is therefore more frequent. & rdquo;

Mark Gonsalves, Data Center Services Manager at Pollution Control Services, said the data center should schedule a quarterly cleaning schedule and clean the floor at least once a year. He said: "The cleaner the environment, the smoother the data center will run. “There is no need to replace costly equipment,” he added. He said that if your data center environment is particularly dirty, some hardware manufacturers will announce that the hardware warranty is invalid.

Be proactive. Prevent problems before they happen.

If you use preventive maintenance and appropriate monitoring software, you can avoid or reduce many other server problems. Chad Peters, Product Support Manager at SEI Services, said: "You can find out early before the problem has arisen." ”

Preventive maintenance consists of three main tasks: double check the error log, find an updated version of the firmware or software, and go to the machine room to check. Peters said his company checks the error log once or twice a year. He explained: "Do we have any hard drives that start logging errors?" Do we have a fan that does not have the required speed? Do we have any fiber NICs that start to have some problems, and some errors are recorded in the logs? If so, then we can proactively replace it to prevent the corresponding component from malfunctioning. ”

When you check the error log, you can take this opportunity to re-examine the firmware and software on the server to make sure that no important updates are missing. Larry Quinn, sales and marketing specialist at TERiX, said: “To clearly understand what kind of system you have, what patch version of the operating system you are running on. If you cut corners on asset and patch management, you will save money in the short term, but it will increase long-term costs. ”

Preventive maintenance should also include running to the data center to check to ensure that all cables are in place and that the air flow management equipment is still in place to ensure that the equipment room is as clean as it should be.

Before the next planned preventive maintenance arrives, you should use a monitoring solution so that when the server starts to have problems, you can receive the alert the first time. There are a wide variety of complex and cost-effective solutions on the market, so you need to make a comparison to find the solution that best suits your needs.

Developing a Disaster Recovery Support Plan

Finally, remember: problems will inevitably occur. If you have a disaster recovery plan and train your staff, knowing when to contact support and how to contact will help reduce the downtime caused by any server failure.

Christia Johnson, Marketing Specialist at TERiX, said: “It's easy for every IT employee to ensure that support is available. This may include preparing a copy of all maintenance contracts and contact numbers that are updated from time to time (in case your maintenance issues include power outages or disaster management incidents), clearly indicating which platform, brand name or original equipment manufacturer it is. (OEM) The product has failed and which party or contact should be found. & rdquo; Whether your support comes directly from the OEM or from a third party, you must contact the other person immediately, which will help you get the server back up and running as soon as possible.