How do you automate complex message-processing?

The System i QSYSOPR message queue, a chronological log of events from IBM i, contains a wealth of information. Application, storage, hardware, security, and systems management errors and other events are all posted to this queue. But more than 95 percent of these events may be unimportant "noise." And one of the most critical aspects of message automation is to "cut out the noise" by focusing on the important messages and ignoring the unimportant ones.

The problem is that a message, such as Job Ended Normally (CPF1241), may or may not be important based on many factors, including the content of the message and when it occurred. For example, if the job performing MQ Series on the system ends normally, you might want to know only if it ends during business hours. Or when certain backup jobs finish, you may need to notify the operations team with a confirmation message. And if certain jobs end abnormally, you might have to use SNMP to notify your enterprise management solution.

Robot/CONSOLE, the message management and resource monitoring software from Help/Systems, lets you "manage by exception" to automate your message processing on each System i or partition without programming. Robot/CONSOLE automates message events at the source of the problem. It uses your rules to suppress unimportant messages, change the color of messages, answer common messages, make messages response‑required, escalate messages, call programs, and execute commands.

OPAL Automates Conditional Processes

OPerator Assistance Language (OPAL) is Robot/CONSOLE's powerful operations language that helps you automate conditional message response procedures that usually require an operator's judgment. With OPAL, message automation is limited only by your imagination. Here are just a few examples of how you can respond automatically to messages using OPAL:

• File full--Respond based on which file is full and how many times the same file caused this message.
• Record lock--Respond based on the job that has the lock. After 10 failed attempts to release the lock, page an operator.
• Save-while-active--When this message arrives on the message queue, restart the interactive subsystem.
• User profile disabled--Activate the profile, but cause its password to expire so the user must enter a new password at the next signon attempt.
• Notify different application teams for the same message--Based on the user profile that generated the message, forward it to either the payroll group or the EDI (Electronic Data Interchange) team.
• Job ended normally--When a job ends normally, restart it during normal business hours, or send an email if it ends after hours.
• Job ends abnormally--When certain jobs end abnormally, use SNMP to notify your enterprise management software.

Using OPAL

Figure 1 shows an example of how to use Robot/CONSOLE OPAL to automate the response to a line-failure message. This response includes sending an SNMP trap for cross-enterprise notification.

Figure 1: OPAL lets you specify exactly how to handle a message. (Click images to enlarge.)

Figure 2 shows an example of using conditions--the time of day and the type of day (work day or other)--to control the rules that handle an event. Using the logic shown, the process for dealing with a failed signon attempt is handled automatically during normal business hours. But if the failed attempt occurs after hours, the system sends a pager message to indicate a potential security problem.

Figure 2: OPAL logic handles a failed signon attempt based on the time and day.

Learn more about Robot/CONSOLE by clicking here. You can even try it free for 30 days. And check out the other Help/Systems offerings in the MC Showcase Buyer's Guide.