New Mainframe Monitors Help Staff Solve Problems Faster, Save Taxpayers Dollars
Industry: Government
Location: Arizona, USA
The state of Arizona is comprised of 15 different counties covering 113,642 square miles. The state boasts a population of approximately 5,130,632 residents and is home a number of noteworthy attractions including the Grand Canyon, the Petrified Forest, the Painted Desert, the Hoover Dam and the reconstructed London Bridge at Lake Havasu City.
The Challenge
Recently, the state of Arizona reorganized its computing operations with the Department
of Administration taking over operation and maintenance of computing systems for a
number of key departments, including the Department of Revenue and Department of
Transportation. These departments had previously used a number of different
monitoring tools.
The state of Arizona had long recognized the value of monitoring the enterprise systems that are used to manage the major share of its operations. With the computing operations consolidation, the Department of Administration began looking for a comprehensive solution that could replace all the different monitoring tools currently in use by various departments.
The Solution
When Rick Woods, technical support specialist for the Department of Revenue, transferred
to the Department of Administration, he suggested standardizing on
ASG-TMON™ monitoring
tools. Karl Eberhardt, technical support manager for the Department of Administration,
was receptive to the suggestion because he had just been subjected to an increase in
prices from the previous technology supplier after upgrading the CPU. Eberhardt
evaluated two tools and decided to go with the ASG-TMON suite, including monitors for
CICS, MVS®, IMS, VTAM®, and TCP/IP.
Various state departments previously used an earlier generation monitor whose less-than-intuitive, command-line-driven interface impeded troubleshooting activities. The new monitor from ASG provided a menu-driven interface that enables the user to simply tap on the screen to drill down for more information on any monitored system or parameter.
The new monitor not only made it easy to identify runaway programs, but also pinpointed the application program routine and the offset, which saves considerable debugging time. And it solves a wide range of unusual problems, such as making it possible to quickly correct terminals that were not set to perform upper case translation, which can cause applications to fail.
Woods said that the new tool soon became a regular part of his daily routine of identifying, isolating, and solving problems. The screen on the monitor that he uses most is the transaction monitor, which provides a quick summary of all running transactions. This allows him to monitor the applications, quickly identify problems, and respond to them faster.
“In our environment, a transaction that has been active for an hour and has 10 million I/Os is a sure sign of a problem,” he said. “But the software provides much more information to get to the root of the problem such as what section of the program is executing, what resources are in use, which datasets are being hit. This level of detail usually gives me a pretty good handle on what type of problem I am facing.”
Woods also uses the monitor to track the performance when accessing data stored in VSAM datasets. TMON quickly identifies datasets that are experiencing excessive control area (CA) and control interval (CI) splits, which can have a significant negative impact on application performance and the end user experience.
“TMON lets me quickly identify storage problems such as file fragmentation and insufficient buffer or string allocation,” Woods said. “When I see a problem, many times I can simply click on it and jump to a screen that lets me make real time adjustments such as resetting the control blocks. Being able to identify, isolate, and fix a problem within a single environment provides enormous time savings.”
ASG-TMON is especially valuable in identifying the frequent problems that occur in testing. “Programmers often come to me and say that they recompiled but CICS is not picking up their changes,” Woods said. “I can have CICS load the program, and then use TMON to inspect the storage into which the program was loaded. In this way, I can quickly see if I’m getting the programmer’s updated program. This feature saves time by quickly identifying problems such as compiling to the wrong library or a compile failure.”
Preventing unauthorized access to confidential information is one of the greatest concerns at the Department of Revenue. Using TMON, Woods configured the monitor to track and generate an audit trail for all access to certain critical systems. A batch job runs on a periodic basis to save the collection files to tape. TMON has a batch utility that can generate reports against those records. If a question arises, he can easily determine exactly who has read or edited sensitive records. Woods monitors the reports on a regular basis to look for suspicious access patterns to identify and prevent unauthorized access.
Measurable Results
The ASG TCP/IP monitor has helped to manage the increasing use of this communication
protocol throughout state government systems. Over half of the state’s users now access
the mainframe from their personal computers through TCP/IP and nearly half of the
printers also use TCP/IP.
Earl Timmerman, technical support specialist for the Department of Administration, said the new TMON for TCP/IP monitor has been provided to networking administrators in many of the departments that it services. “They tell me that they like its ease of use and the abundance of information that is available,” he said. “They can monitor user response time, packet errors, FTP failures, and timeouts. They can quickly identify problems and get the information they need to solve them, usually before the users even notice that something is wrong.”
Eberhardt said that standardizing on the monitoring tools saves additional time by allowing technical support staff to quickly jump from one module to another as they investigate different possible problem causes.
“All ASG-TMONs have a common user interface and are linked together so that you can jump from one to another very quickly. Suppose you notice a response time problem in the TCP/IP monitor. You can jump to the MVS monitor for a quick overview of which jobs are consuming resources. You might see that a CICS region is taking up 90% of the CPU so you jump to the monitor for that region and look at the all the individual jobs that are running. TMONs combine to function very much like one single monitor that covers the entire mainframe environment,” he concluded.
