In this lesson you see a series of videos with a business scenario relating to troubleshooting a system problem. Please watch the first one below.
Troubleshooting is one of the most important responsibilities of a System Administrator. When a system fails to operate properly, you, as System Administrator, are expected to find the fault, correct it, and put the system back in running order.
The nature of the troubleshooting effort depends on the problem experienced, the nature of the system where the problem occurred, and the availability of outside tools or help for the problem.
Troubleshooting Components
Troubleshooting is a process with three steps:
- Problem detection - Becoming aware of the problem.
- Diagnosis - Determining the cause of the problem.
- Resolution - Getting the system to work again.
1. Problem Detection
The most important aspect of troubleshooting is problem detection. If you, as a System Administrator, are unaware of a problem, you can’t fix it.
Problems may be obvious, such as a system crash, or ambiguous, such as a logic error in a program that delivers an incorrect result.
Most systems have predetermined means of detecting an error. These may include:
- Alerts - The system has a background process that tests error conditions and informs administrators of failures.
- Error/Trace logs - The system produces output files that display process results and errors.
- Monitoring tools - Interactive tools that display errors and other problems. These tools may be built into the system or third-party tools.
In addition, end users may have a role to play in error detection. If a problem occurs, users should have a means of reporting it to the administrators.
2. Diagnosis
Once you, as a System Administrator, detect a problem, you should be able to understand it. This includes knowing:
- How the affected part of the system is supposed to function
- What were the expected results
- What was the erroneous result
- What went wrong
Diagnosis is a logical, systematic search for the source of a problem. It is as much an art as a science. It may be a simple problem or require considerable examination of the system (and its components) to find the issue.
It may require additional resources, such as:
- System documentation
- The assistance of SAP Support
- Reproduction of the problem in a test system
3. Resolution
Resolution includes:
- Solving the problem
- Resolving any application/data/processing issues
- Getting the system back into normal operation
There may be multiple possible solutions to a given problem. It is the responsibility of the administrator to pick the solution which is:
- Most likely to succeed
- Minimize any additional problems
- Get the system back into normal operation in the shortest time
- Reduce the risk of the problem happening again
- Be maintainable over time
See the following video for information on resolving a system problem.
See the following video for more information on resolving a system problem.
You may think that by providing a resolution, troubleshooting is over. But what if there's more to it? What if, instead of having a problem that you need to detect, diagnose, and solve, you make sure the problem does not appear? In other words, what if you prevent it!
Prevention
The best way to deal with problems is to prevent them. This may include such methods as:
- Testing the system before it enters production.
SAP Best Practices typically recommend that all changes be thoroughly tested during implementation, updates, and customization before anything is allowed to run in production. SAP provides design, development, and testing guidelines to protect against inappropriate changes.
- Backups
The data in the system should be backed up regularly. These backups should be stored in a safe place and kept for a sufficient time. The backups should also be kept where they can be conveniently used for restores (if necessary).
Note
You will find out what it takes for a System Administrator to perform backups in unit 8 of this course. - High Availability/Failover/Disaster Recovery
The system may consist of duplicate instances, which can be made available in an emergency. These instances consist of duplicate hardware, duplicate software, and data that is in sync with the primary instance.
- Monitoring the System
Monitoring the system serves two purposes: detecting problems once they have occurred and detecting situations that may become problematic at some point. Typically, hardware, system software, and application software have monitoring tools that allow administrators to observe key performance indicators (KPIs).
- Training
To run the system properly, users, developers, and administrators must understand how it works and what is required to operate it. They must know what processes must be performed, who is responsible for each, and how they are done. No one should be allowed to interact with enterprise software without sufficient training.
And coincidentally or not, dear future SAP System Administrator, this is what you are doing now: training!
Summary
You have learned that problem detection, diagnosis and resolution are needed for troubleshooting SAP systems. But to reduce such problem occurrences, prevention is the key, including testing, backing up, monitoring, and training. You are now in a better position to explore some examples of such troubleshooting techniques in two SAP systems.