Simplicity and savings are only the beginning of what agentless IT (information technology) monitoring ought to provide. To help successfully deliver the performance and availability that end-users demand, the most desirable of agentless solutions will also have functionality that resembles higher-end, agent-based products. When considering an agentless product, determine whether proactive system and application management are realistic expectations with that package over the long-term. Organizations of every size need to monitor and manage their computer systems and applications for performance and availability, but few have ample information technology budgets and staff to devote to the job. A practical alternative to high-end monitoring solutions has emerged in the form of agentless monitoring.
Using these tips will help the IT pro understand how practical any given agentless approach is for the organization’s objectives.
1) Stop false alarms: An issues which we hear daily on our forum. End users who are sick of mindless alerting, which eventually develops into a “Boy Who Cried Wolf” situation. Too many false alarms causes those in charge of monitoring to habituate alerts in general. How do you curb false alarms when appropriate event thresholds will vary between different systems and applications?
Preset, manufacturer-recommended monitoring thresholds are useful and necessary. But to stop false alarms you need to tune thresholds that trigger alerts and alarms to be sure they are meaningful in context of the particular system or application.
2) Develop multiplatform over single platform coverage: Operating system support may span Microsoft Windows®, Red Hat® Linux®, SuSE® Linux®, Hewlett-Packard HP-UX®, IBM AIX®, Sun Solaris™, and VMware ® platforms. Application support may extend to a variety of application servers, Web servers, databases, messaging, and infrastructure products from different vendors. Apache™, Oracle®, Microsoft Exchange, and the like. In addition, look for the ability to specify and track a range of user and business metrics, including transactions and service level agreements (SLAs). Look for a broad scope of operating system and application support directly out of the box.
3) Become proactive: When there isn’t enough information to fix a minor issue until after it has mushroomed into a major problem, it’s too late; the service level and the user satisfaction have already taken the hit. To be proactive, the IT pro charged with monitoring needs a view into the health of the systems and applications from a real-time perspective in addition to trend and historical reporting. How does one do this? Enable a solution which fits into the organization’s budget, the timeline for implementation and about of hands-on work they wish to administer.
4) Remove legacy systems: By enabling a monitoring solution that can update itself automatically, there will be a serious reduction in troubleshooting. A monitoring tool should update itself automatically without anyone having to perform ongoing maintenance. The advantages are obvious: greater efficiency and lower cost of ownership.
5) Reduce the reporting headache: Reduce the burden of proof with agentless monitoring that integrates the required reporting and graphing. Avoid developing reports and graphs from scratch. By selecting a pre-packaged solution there will be the ability to produce documentation quickly and easily whenever an organization requires it.
Want to learn more?
Download our Best Practices for Server Monitoring Whitepaper and learn how to achieve a successful long-term server monitoring strategy by focusing on an approach that is lightweight, efficient, resilient, and automated.