Table of Contents
Facing an issue on the Server's side is not that different that dealing with problems on the client's one. We are going to use the very same technique to troubleshoot the Server that we use to do the clients. Thus also in this case we are talking about identifying a problem starting from an incident, that is its sympton.
On the server it can be easier to spot incident as it's possible to read documentation about the expected behavior and get fast addressed to the wrong one the server is performing. Otherwise it can be difficultier to track down the problem which originated it.
When facing an issue on the server, please take care to collect as many data on it as possible, using the classical Kepner and Tregoe Help process, which lists 5 points:
Define the Incident
Describe the Incident
Try to solve the incident supposing the most probable cause
If failed at point 5, report the incident to the higher level of problem handling service desk
The server support is performed by PrivateWave only as the third level, so before addressing a new request make sure you:
collect all relevant information
first attempt to troubleshoot
deeper analysis and performs the troubleshooting
The communication between the second and the third line is carried out by a ticketing service. The second line has to report the issue by filling out all fields of the report properly. The third line has to report back using the same ticketing service updating the problem status. The ticketing service will report timings as well and will be used as an incidents/problems database to be consulted by the second level before escalating the problem.
This is the first phase of the troubleshooting process. The initial entry point is the call of the user reporting an Incident. In this phase the user describes autonomously the issue(s) experienced and the Help Desk guide him/her to:
This is a platform specific procedure to guide the user collecting the following information:
Server version number
Number of Network Interfaces
Network Segregation configuration
Possible route specific configurations
Short description of malfunction
Is the device supported?
Is the OS version supported?
Is the application installed?
Are the application edition and version OK?
Is data plan option available?
Is Internet connection functioning?
If any of this requirement checks are answered with "no" then the user will be informed about the requirements. This action closes the ticket. Otherwise we go to phase 2.
In this phase the Help Desk asks the User to retrieve as many useful information as possible, aiming to fill completely the following Incident table.
Similar systems/situations not failed
Where or in which network does it work?
When does it work?
Which parts and/or systems do work well?
This is a crucial phase as the Help Desk has to figure out the possible causes of the Incident with the help of the previously collected data. We provide a decision work flow that lists the most common complaints that are the entry points for the troubleshooting. The following complaints are examined:
Application local issues
User can't call
Can't Connect to Server
Application doesn't start
Server is unreachable
Bad Call Quality
Can't Register to Server
Application closes abruptly or crashes
Server is too slow answering
Application hangs at some point
Can't login to the Server
User can't receive call
Application disappeared or missing
Can't find the User/Callee on the Server
Application goes into an infinite loop
Using the information collected and with the help of the troubleshooting workflow the Help Desk should be able to list the most probable causes of the Incident and thus propose to the User the related procedure(s) to try the solutions in the provided order. If the first solution fails, it's possible to step to the second one and so on until either the Incident is solved or the list ends. In the later case the Incident should be escalated to the higher level of Service Desk.
Even if the first/second level has found a solution (and so it can close the Incident reporting success), the Incident must be reported as a Problem and passed to the proper team along with the Incident data collected during the process and the steps taken during the troubleshooting and it should be still tested for the solution. It would be up to the Problem team to eventually close the ticket or perform further investigations or perform proper further actions. The same practice should be executed if no workaround or solutions has been found and the Incident is still open.
After the Incident has been declared as solved, the answers of the third level can be:
not a problem: no further actions required by editor.
workaround identified and provided.
wait for the problem resolution: a software component upgrade will be provided.
Timings will be provided along with the answers. The second level has to report back to the User and ask for the acceptance of the answer.