Message automation eliminates the "noise" in managing your System i.
Do you monitor the QSYSOPR message queue manually? Do you want to develop a list of responses to messages in your system's message queues? Do you want to consolidate your console workstations? Do you need to monitor your HA product?
The System i (iSeries) is message-driven. Application failures, security errors, system management issues, hardware problems, print errors--just about everything generates error messages to QSYSOPR.
Management by exception is a key component of automated operations. The idea is to suppress unimportant events and concentrate on events that require attention. This is hard to do manually.
So how can you automate System i messages? The RCVMSG command allows you to read a message queue in real-time. When a message arrives, your program wakes up, and you can read and respond to the message. In newer releases of i5/OS and OS/400, you can use APIs to read the message queue. Your programmers can write CLPs to read, respond to, and execute commands to automate your processing. But is that what you want your programmers doing?
Another option is to use the Message Reply List Table. The WRKRPYLE command allows you to access the table and enter default replies to execute when an inquiry message arrives on a message queue. You can specify replies to messages that you want answered the same way every time. Unfortunately, this option lacks flexibility.
Robot/CONSOLE, Help/Systems' message management and resource monitoring software, provides automation to achieve management by exception on every system or partition-without programming. It focuses on automating message events at the source of the problem. Rules allow it to suppress unimportant messages, change message color, auto-answer common messages, call programs, execute commands, make a message response-required, and escalate the message.
One of Robot/CONSOLE's most powerful features is OPerator Assistance Language (OPAL), which that allows you to duplicate what an operator would do when a message arrives. OPAL is limited only by your imagination. For example, you can easily handle these situations:
•· File full--Respond differently, depending on which file is full.
•· Record lock--Respond differently, based on the locked job. After 10 attempts to release the lock, notify an operator.
•· Save-while-active--Restart the interactive subsystem when the message arrives on the message queue.
•· User profile disabled--Activate the profile, but cause the password to expire so the user needs to enter a new password at the next sign-on.
•· Notify different application teams for the same message--Forward the message to the payroll or EDI team, based on the user profile that generated the message.
These are just a few examples of what you can do with OPAL. Our free OPAL Cookbook provides sample code to show you the possibilities of message automation. OPAL offers unprecedented flexibility in message-handling (Figure 1).
Figure 1: OPAL allows you to specify how a message is handled. (Click images to enlarge.)
If you manage multiple instances of your OS, you're probably interested in console consolidation. For many data centers, console consolidation consists of constructing a command center or "war room" equipped with stackable monitor housing units, slide-out keyboard trays, toggle switches, and large-screen monitors. They look impressive but lack automation. The goal is to reduce the number of consoles you use to manage your systems. It's not unrealistic to have 50+ systems or partitions reporting into one console.
Robot/CONSOLE is very good at managing numerous systems by exception. It works with our networking software, Robot/NETWORK, which provides the conduit for consolidating the exceptions to one workstation. When Robot/CONSOLE identifies an exception, it redirects the event to Robot/NETWORK on the host (focal) system, so just one person can manage many systems by exception (Figure 2).
Figure 2: Robot/NETWORK displays the status of monitored systems.
Robot/NETWORK also allows you to set up Robot/CONSOLE message-handling rules on the host system and send them to remote systems. You can test your rules on one system and then apply them to multiple systems without having to sign on to each system. These two products also can convert any of these messages to an SNMP trap for interfacing into a non-System i monitoring tool.
Learn more about Robot/CONSOLE and Robot/NETWORK by clicking here. And check out Help/Systems' other offerings in the MC Showcase Buyer's Guide.
LATEST COMMENTS
MC Press Online