July 25, 2024 | IBM i

Seven Ways to Proactively Monitor IBM i Systems

image

Using IBM i message and resource monitoring software, IBM i servers can proactively analyze, predict, and respond to system and resource issues before they become major problems. The good news is you don’t have to wait for AI to use proactive IBM i monitoring. It’s available today in many third-party IBM i monitoring systems, including absMessage from SEA 

 

Here’s how proactive IBM i monitoring works. 

Reactive Versus Proactive IBM i Monitoring 

Some IBM i monitoring systems only use reactive monitoring (after the event occurs) to determine when a system or resource issue has happened.  

 

Reactive monitoring is generally activated by setting up message monitors (filters) inside your IBM i monitoring software. Software filters monitor QSYSOPR, QHST, and other message queues looking for informational, inquiry, or other messages relating to a specific process, job, user, subsystem, etc. When a message matches a reactive filter, it kicks off one or more pre-configured responses to automatically respond to or resolve the issue. 

 

Proactive monitoring (before or while the event occurs) complements reactive monitoring. Using proactive monitoring, IBM i monitoring solutions can spot unseen and developing system and resource issues. They can then automatically respond to or resolve a developing event. Proactive monitoring can detect invisible issues that do not generate system messages, such as when an expected job is not running, or a companion server is no longer attached to the system.  

 

Proactive monitoring detects, infers, and responds to system and resource issues by: 

  • Examining the current, expected, and future state of the environment 
  • Making decisions about any perceived system and resource issues that are occurring 
  • Triggering automated responses such as executing commands and scripts, notifying administrators, or generating a trouble ticket in a Help Desk system such as ServiceNow 

Seven Ways to Proactively Monitor IBM i Systems  

Using third-party IBM i solutions such as absMessage, here are seven ways you can proactively monitor your i servers to detect and respond to system events. 

 

1. Detecting active and inactive jobs/subsystems/job queues 

Many packages provide pre-configured filter templates for monitoring whether a job or subsystem is active or inactive during a specific processing window. If the job/subsystem isn’t running during its specified window or if it’s running outside that window, the filter can automatically restart or end the process, notify administrators, or perform other actions. 

 

2. Detecting jobs stuck in queue or behind a job in error 

Create a filter to count the number of jobs in a specific job queue and kick off a response when there are more than x jobs sitting in queue (ex., perform an action when there are more than 10 jobs sitting in an order processing queue). Actions could include moving the jobs to another queue or alerting DevOps.  

 

3. Detecting looping jobs 

Filter for jobs running longer than expected. For example, you can create a filter to alert DevOps when a customer order processing job is running longer than five minutes. This may indicate a job is looping. 

 

4. Disk utilization issues 

Some monitoring packages including absMessage now include dashboards to monitor disk performance and utilization. If the dashboard or a filter sees disk usage trending upwards (possibly indicating a runaway job) or trending significantly downward (possibly indicating file or object deletion), the solution can trigger a response. 

Filters or dashboards can also be used to determine if a disk drive has failed or if RAID protection has been suspended for a specific disk set. 

 

5. Monitor CPU utilization 

Create a filter or use a dashboard to detect when CPU utilization is running high for a specific time period, possibly indicating runaway jobs or inefficient processing. 

 

6. Send IBM i logs to Enterprise SIEM Servers 

Some solutions offer direct integration with Security Information and Event Management (SIEM) solutions, where IBM i log data can be sent to SIEM servers such as QRadar, Splunk, LogRhythm and others. SIEM Integration allows IBM i machine data to be evaluated alongside other network information for enterprise-wide forensics and analysis.  

 

7. Ping other devices and servers 

You can create filters to ping and check whether co-processing devices and servers are up and available. Device pings can quickly alert administrators about unseen communication issues that may affect production processing. 

More to come 

Proactive IBM i monitoring contains many AI-like qualities to detect and respond to detected issues. As AI matures, you can expect IBM i monitoring software to increasingly use AI technology for system monitoring and management.  

 

Please feel free to Contact SEA if you’d like more information on implementing IBM i system and resource monitoring.