One of my slogans is “little problems become big problems if left alone.” When managing service level agreements (SLAs), we are concerned with notification and reporting that keeps us informed about the status of critical processes or an unknown event that can cause other processes to not finish on time.
A little problem can be a file that is late from an external or internal process, a batch process that has stopped due to a message wait or record lock, a service that has ended normally on a Windows server, or an operator who forgot to submit a process. If you know about these little problems in the first five minutes, they will have little or no impact on processing and will not become big problems.
Time can cause you to miss your SLA, but lack of notification about small problems is the real foe. Proper notification during critical times of processing is essential in a multi-platform/multi-system environment. Notifications can be e-mails, text messages, interactive bubbles on the operator’s workstation, sounds blaring in the data center, lights blinking, and maybe even an electrical shock! The notification process should be integrated into your scheduling tool.
Today, I am looking at a request for proposal (RFP) from a customer using a free scheduling program. Many of their needs reveal that they do not have a truly automated schedule with built-in notification. The downside of using a free scheduler is that you need to buy additional tools or write a lot of internal code to build dependencies and to monitor for processes that are late or not running.
Robot/SCHEDULE Enterprise is the answer for your SLA needs. It fixes many of the problems you need to solve, and it does much more with reporting, documentation, and control. Its built-in job monitors (see the screen shot on page 1) work with IBM i, AIX, Solaris, Linux, HP-UX, and Windows processes to notify you if a process is running too long, ran too quickly, or did not start on time. These monitors can use messaging software to notify your operator when there is a problem.
The Job Monitors report (shown below) provides online or text-based reports that list the events that might have been missed. Robot/SCHEDULE Enterprise can monitor for file arrival, services up/down, daemons up/down, the server being ready for batch, and batch scripts finishing or not finishing. Each script can have a different set of return codes that must be monitored. Every server that is defined to Robot/SCHEDULE Enterprise is monitored to make sure that the server is ready for processes.
We know controlling your enterprise scheduling is very important. Why not try Robot/SCHEDULE Enterprise free for 30 days in your environment?
Contributed by Tom Huntington, Vice President of Technical Services