IBM i system monitoring packages such as my MessageFlex software, are some of the most popular and widely used packages in the IBM i community…with good reason. Many shops use these packages to provide 24×7 unattended monitoring, freeing up personnel from watching the system, and alerting on-call responders when a problem occurs.
Notice: Power System running IBM i is the name used today by IBM for identifying the current system hardware. Previous names were AS/400, iSeries, Systemi) so any of these terms may be used interchangeably in these document.
Being a monitoring vendor, we’ve helped a lot of people get started with system monitoring. Based on our experience, these are my steps for setting up and delivering an IBM i monitoring system:
- Create a Monitoring Strategy
- Determine who should be notified
- Determine how they should be notified (Text Msg or Email)
- Define Basic Monitors
- Define Critical Monitors
- Define Proactive Monitors
Here’s how each of these steps can help anyone create an IBM i monitoring environment.
1. Create a Monitoring Strategy
Before configuring software for 24×7 monitoring, you need to consider your monitoring strategy. A monitoring strategy should deal with the human elements of your monitoring system, such as how you set up a call tree; how many responders are needed; and how to escalate a problem. Your monitoring strategy defines the goals and processes for how your monitoring system will work. Read this article for more information on setting up a monitoring strategy using an IBM i system monitoring product like MessageFlex.
A monitoring strategy is your game plan for how system alerts will be handled. It’s your basic monitoring template and all other monitoring activities flow from that strategy.
2. Determine who should be notified
The next step is determining who should be notified for different issues. The user being notified for an issue with your backup may be different then the person notified when “Serious storage condition may exist” is encountered.
3. Determine how they should be notified (Text Message or Email)
Some users may prefer to be notified via Email others may prefer a Text Message on their phone or, you may decide to send notifications both as an Email and a Text Message.
4. Define Basic Monitors
The next step is to set up your basic monitors. These are the most important items your system will monitor for, including disk capacity, power outages, password violations, and error messages requiring a response.
My monitoring advice has always been to go slow. Don’t immediately try to set up monitors for every conceivable situation. Start with the four basic error messages you should never ignore on your IBM i box and get those monitors right first. This creates a stable monitoring environment that handles basic needs, an environment you can build on.
5. Define Critical Monitors
This step focuses on more critical monitoring items, the occasional but frightening issues that keep IT managers up at night….Disasters!!!
Disaster monitors include things like system hardware failures; network failures; security breaches; runaway jobs that may be corrupting your database; and damaged objects.
In this step, you add critical item monitoring to your monitoring strategy, items that must be immediately addressed or can cause excessive system damage.
6. Define Proactive Monitors
The next step is to set up monitoring to proactively look for hidden problems on your partitions.
Hidden problems are small system changes that if left unnoticed, can morph into larger issues affecting system and database health. These monitors alert on-call responders when the software sees a major issue in progress, such as disk utilization spikes; long-running jobs; or jobs that should be running but aren’t. You reconfigure your monitoring system to alert you when a problem is developing.
7. Repeat steps 1-6
It’s important to occasionally review and repeat the previous steps. On a regular basis, you should also perform these tasks to insure your monitoring system continues to meet your needs.
- Review existing monitors to see if they need to be change or removed – Eliminate monitors that have outlived their usefulness.
- Review your responder pager list to add new employees or remove terminated employees– Pager lists become obsolete if not maintained. Incorporate pager maintenance into your user provisioning process to keep your responder list current.
- Incorporate monitoring changes into your incident management plan – When an incident occurs, review your monitors to see if they should be changed in response to the incident. This could provide early detection the next time the same incident happens.
Using these steps, any shop can set up an efficient and effective IBM i system monitoring system. Feel free to contact us at DRVTech if you’d like to further discuss how to get started with IBM i system monitoring.
Our System monitoring software runs on AS/400, iSeries, System i and of course the current IBM system known as Power System running IBM i.
From small businesses to Fortune 500 companies, our clients in multiple industries get a good night’s sleep knowing our MessageFlex solution is on the job.