Message and Event Monitoring

Any time an event happens on your system or in one of your applications, it generates a message that contains information more or less describing what has occurred. Sometimes the information is purely nice to know. Other times, it might indicate that you need to take action in order for processing to continue or to get system or application performance back to normal.

These messages can be sent to a kind of log called a message queue. Some message queues are built into the individual applications you may be running. Others like QSYSOPR or QSYSMSG are objects that you can create on the IBM i operating system to consolidate application and system messages into a central location.

Funneling your messages into a central location is all well and good, but now hundreds—more likely thousands!—of messages are flooding in with information about your business-critical applications, system resources, performance statuses, events, and even network components. How can you possibly filter through the noise to find the handful of urgent messages that actually need your attention?

You might be so overwhelmed by all the messages in your message queues that you’re operating in a completely reactive mode. You wait for end users to call with complaints about slow system performance, missing reports, or unavailable applications. When you can get away from the phone, you try manually looking for messages, but only infrequently.

You could also be checking components by running a query or command, keeping an eye out for a string of text, a status, or something that is suspiciously absent. Since it’s hard to know what you’re looking for—especially if it’s not there—you rely on runbooks, cheat sheets, and the tribal knowledge of staff with more experience.

Realizing how important monitoring is for keeping your applications available and ensuring that the environment where they run is healthy, you might have even cobbled together a kludgy series of CL commands, RPG programs, SQL statements, or Qshell scripts to relieve some of the burden. Unfortunately, this won’t keep you current with monitoring tools available in the operating system or the latest and greatest Java or web applications running on the system.

Robot Console is message and event monitoring software for the IBM i operating system. Robot Console filters incoming messages, monitors resources and system logs, and notifies an operator only when it can’t handle a situation automatically.