Contents
Troubleshooting methodology
When troubleshooting an issue with your Genesys Voice Platform (GVP) solution, it is important to take a methodical approach in order to quickly identify and resolve the cause of the issue. Drawing conclusions too quickly and making undocumented changes to the system(s) can result in making the issue worse.
This topic outlines a brief methodology that you can follow in order to troubleshoot issues that you might encounter with your GVP solution.
Describing the Problem
The first and most important step in troubleshooting any issue is to clearly define the problem. Your problem description should be as detailed as possible and include the following information:
- A clear indication of your system’s symptoms
- How you discovered the issue. For example, did you receive an alarm from the system? Did a caller identify the issue(s)?
- When did the symptom first start to occur?
- Was GVP previously running without issues, and the problem recently started to occur? If yes, what changes did you make to the system? For example, did you deploy a new VoiceXML application or make some configuration changes?
- How often does the symptom occur? For example, does it occur on every call or intermittently?
- Can you isolate the symptom to a particular site, system, voice application, or other component?
Gathering Relevant Information
Once you have a clear description of the issue, you can start to gather relevant information to isolate and identify the cause of the issue. This might include the following information:
- Recent changes that you have made to the system or environment. These can include operating system updates or patches, system or network configuration changes, or voice application changes.
- A more detailed description of the symptom. If callers are experiencing the symptom, can more specific information be gathered? For example, callers might report that their calls were dropped. In this case, it would be useful to know which voice application they were calling, where in the voice application they were dropped, and where they were calling from.
- Steps to reproduce the issue. If the symptom is reproducible, what are the detailed steps you can follow to make it occur?
- Whether the symptom can be isolated to a particular site, system, or voice application. If it can, you should review the particulars of that site, system, or voice application and compare them to that of ones that are not experiencing the issue.
- You may need to capture various log files for later analysis.
- Network traffic capture (via Wireshark or some other tool).
- You may need to export the configuration of the system.
Creating an Action Plan
You can now create an action plan to further isolate the issue based on the information you have gathered. Document the steps you are going to follow, and then check them off as you complete them.
Keep a record of any changes you make to the system as you go, as well as any observations you make. It is very difficult to remember what you did after the fact, and this information might be critical in preventing future issues.
Implement any changes one at a time, because you may not know which change corrected the issue if more than one change is implemented at a time.
Verifying the Resolution
Once you have taken measures to correct the problem, you must properly test the system to ensure that the symptoms are no longer occurring.
Document what you expect to happen, and then compare your expectations with your written observations of the system during your tests.
If the symptoms continue to occur, restart this process from the initial problem description.
Describe what is occurring now as it may not be the same as the initial problem, especially if you have made changes.