Department running own server

Because many factors come into play, only your department can decide if it is right for you to run your own server. Be realistic about the personnel resources and costs involved in running your own server safely and securely. One of the most important and most costly components of successfully running a server is the hiring of properly trained personnel. This requirement is one that tends to get overlooked or grossly underestimated in terms of time and costs involved. Many times the responsibility is placed on a graduate student who may be gone in a year or added as a side task to the workload of a current employee with no server experience, with the assumption that once the server is setup it will run reliably forever. The reality is that a server requires constant attention throughout its lifecycle in order to keep it functional, secure and running with minimum downtime. A compromised server could have negative consequences for users across campus as well as for the reputation of the University of Wyoming.

The following is a brief overview of issues that should be considered when deciding whether to run your own server:

  • Hiring personnel that are trained in all aspects of server management and security to ensure that data is not compromised and that the University of Wyoming network and reputation are not compromised.
  • Installing the operating system safely and properly - some viruses can infect a vulnerable (un-patched) computer in as little as 10 seconds after being connected to the network.
  • Safely and properly installing and configuring any services the system may need.
  • Configuring the security on the system properly, including file-level access, account management, and service lockdown.
  • Evaluating and managing additional security concerns, including network protocol security and firewall configuration.
  • Installing upgrades, including motherboard BIOS, peripheral BIOS, hardware, software, operating system, etc.
  • Creating a backup system, including purchasing the backup hardware, planning a comprehensive strategy, defining a restore procedure, defining an offsite storage place, etc.
  • Researching, testing, and installing patches, including hot fixes and security releases. This includes monitoring discussion and mailing lists, contacting manufacturers for updates, etc. This is one of the most important ongoing functions of running a server on campus. If your server is vulnerable and not patched, it is possible (and common) for your server to negatively impact the entire campus network.
  • Providing a proper storage environment, including providing a climate-controlled, clean (the room should have air filters), secure room to store the server with networking installed, access to low distribution power units, access to emergency power supply, inclusion of a rack system that can make the addition or repair of hardware easier to accomplish (especially in multi-server environments), etc.
  • Training, including initial and ongoing training on the operating system as it is constantly upgraded, ongoing training on the software installed on the system (most companies upgrade their software annually), training on hardware as it is upgraded, etc.
  • Evaluating, installing, and configuring anti-virus software.
  • Installing and utilizing disk defragmenter software, which is a good idea to use on a server to keep it running as efficiently as possible.
  • Monitoring, including performance monitoring, monitoring Event Logs for possible security breaches or problems, 24-hour monitoring for server up-time, 24-hour monitoring of room climate, etc.
  • Alerting, including the implementation of a system to alert someone 24 hours a day if any of the systems being monitored are not working properly.
  • Creating redundancy, including the redundant drive arrays and power supplies, the most common problems for servers. Someone needs to know and understand how the redundancy works, how the hardware works, and how to best utilize what it offers.
  • Reporting, including creating a process to report prolonged out-of-service situations.
  • Keeping up with changing hardware and software.
  • Administering Maintenance Agreements, including the yearly purchase of Maintenance Contracts with both the OS vendor and the vendor of the software installed on the server.
Was this helpful?
0 reviews

Details

Article ID: 7985
Created
Wed 7/29/15 1:26 PM
Modified
Fri 8/14/15 9:57 AM