Introduction
The connection issue is a specific Incident report, so we have to focus on connectivity issues, analyzing the entire end-to-end from the reporting User's network to the PBX's one.
Can't connect or register to the server
First Level
We start from the User's network. First we have to check the device can connect to an internet provider and if it doesn't close the incident as "Not an issue" and tell the user to solve the issue whit his(her) provider. If the internet connection is available check the client connection status. In any case we force a Force Manual Reconnection, so we are sure that the connection action is correctly triggered and thus we can collect answer by the client. If the connection proceeds fine, then the User experienced a Temporary Network Problem which is now over and then the Incident is closed, otherwise we write the exact error message (if any) about the connection problem. Using this message we Check the Configuration of the Application, looking for misconfigurations which might lead to the incident reported. Example Given: pbx wrong address or port.
If the Configuration of the Application is somehow wrong, we can try to fix it and test if the User can connect to the PrivateServer correctly. If the connection works fine, then we close the Incident, else we need to check that the PrivateGSM configurations are correct and that the User's account is actually present on the PrivateServer if it seems to be something wrong, try to fix the configurations and eventually close the incident.
If the configuration looks fine but the Application continue to complain about connecting, we need to escalate.
Second Level
The second level is in charge to fully evaluate the server configuration to understand if the connection problem is somehow related to some issues on the server.
First thing to do is to check the Network, as the connection problem is mostly network driven. If the check goes fine, then we check the server services, paying particular attention to the Asterisk server and to the Database backend, as they are potentially to blame for such behavior.
If the services are fine, we must investigate further what's really happening and so we go down into the client logs. If nothing arise from there, we can just escalate to the third level.