Partner TechTip: Coordinating Events in the Network Is Error-Prone

System Administration
Typography
  • Smaller Small Medium Big Bigger
  • Default Helvetica Segoe Georgia Times

Eliminate lag time when scheduling across platforms.

 

What are some of the network events that make or break your nightly processing schedule? Is it a file arriving, a service starting, or a daemon ending? Have you missed a service-level agreement (SLA) because of one of these events? Does your team schedule the various events based on time?

 

Many organizations still handle multiple operating system events with timers. They might run a job on an AIX server at 7:15 and another job on a Windows server at 8:00 that depends on the AIX job completing successfully. Most of the time, this process works fine, but what if the AIX process doesn't run or finishes after 8:00? The Windows job results in an error, and the entire schedule is off. Timer jobs drive data center managers crazy because they can cause errors to critical processing schedules.

 

Another aspect of using a timed schedule is that there are usually no notification processes to warn the staff when there is an issue. For example, you might want a warning that the file transfer has not arrived and that you are only 15 minutes away from running the next critical step. If you aren't notified of this, the next job may not run correctly.

 

Since you have to allow time for each job to complete, timer jobs create a long nightly process window. If you eliminate these jobs, you have a fully event-driven schedule, which removes the lag times between jobs.

 

Robot/SCHEDULE Enterprise is Help/Systems' answer to these enterprise scheduling difficulties. This new offering coordinates the events amongst IBM i, Windows, UNIX, and Linux operating systems. Some of the benefits you will see are warnings if events are running late, a condensed schedule because you are eliminating timers, and advanced reporting. The user interface is accessed via the Robot/SCHEDULE Explorer (see Figure 1).

 

041709HelpMainInterface.gif

Figure 1: Robot/SCHEDULE Explorer provides access to all servers.

 

A common event in a network of systems is using file arrival as a dependency for an entire process. For example, a bank transfer or point-of-sale terminal sends data to a server. When the data gets there, you want a process to be launched. Robot/SCHEDULE Enterprise removes the guessing game by having a file event monitor that checks for the file every 30 seconds, for example. When the file arrives, the prerequisite is met, which in turn launches the script to process the data. Figure 2 shows a file event that monitors every 30 seconds. If the file has not grown in the last minute, it causes the dependent script to run.

 

041709Helptnt46filearrival.jpeg

Figure 2: Use the file event to launch a script.

 

Learn more about Robot/SCHEDULE Enterprise and even try it free for 30 days in your environment.

Eliminate lag time when scheduling across platforms.

 

What are some of the network events that make or break your nightly processing schedule? Is it a file arriving, a service starting, or a daemon ending? Have you missed a service-level agreement (SLA) because of one of these events? Does your team schedule the various events based on time?

 

Many organizations still handle multiple operating system events with timers. They might run a job on an AIX server at 7:15 and another job on a Windows server at 8:00 that depends on the AIX job completing successfully. Most of the time, this process works fine, but what if the AIX process doesn't run or finishes after 8:00? The Windows job results in an error, and the entire schedule is off. Timer jobs drive data center managers crazy because they can cause errors to critical processing schedules.

 

Another aspect of using a timed schedule is that there are usually no notification processes to warn the staff when there is an issue. For example, you might want a warning that the file transfer has not arrived and that you are only 15 minutes away from running the next critical step. If you aren't notified of this, the next job may not run correctly.

 

Since you have to allow time for each job to complete, timer jobs create a long nightly process window. If you eliminate these jobs, you have a fully event-driven schedule, which removes the lag times between jobs.

 

Robot/SCHEDULE Enterprise is Help/Systems' answer to these enterprise scheduling difficulties. This new offering coordinates the events amongst IBM i, Windows, UNIX, and Linux operating systems. Some of the benefits you will see are warnings if events are running late, a condensed schedule because you are eliminating timers, and advanced reporting. The user interface is accessed via the Robot/SCHEDULE Explorer (see Figure 1).

 

041709HelpMainInterface.gif

Figure 1: Robot/SCHEDULE Explorer provides access to all servers.

 

A common event in a network of systems is using file arrival as a dependency for an entire process. For example, a bank transfer or point-of-sale terminal sends data to a server. When the data gets there, you want a process to be launched. Robot/SCHEDULE Enterprise removes the guessing game by having a file event monitor that checks for the file every 30 seconds, for example. When the file arrives, the prerequisite is met, which in turn launches the script to process the data. Figure 2 shows a file event that monitors every 30 seconds. If the file has not grown in the last minute, it causes the dependent script to run.

 

041709Helptnt46filearrival.jpeg

Figure 2: Use the file event to launch a script.

 

Learn more about Robot/SCHEDULE Enterprise and even try it free for 30 days in your environment.

BLOG COMMENTS POWERED BY DISQUS