Media Server Failover Announcement Mechanism
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In communication systems, failover conditions can lead to interruptions in media streams, such as music on hold, resulting in silence for customers when the primary server fails, causing them to potentially hang up before a secondary server takes over and resumes the workflow.
Innovation Solution
A media server is designed to stream a first media message and, upon determining the primary server has failed, stream a second message during the failover period, ensuring continuous media playback until the secondary server reestablishes the communication session.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the primary server fails during media streaming, then system reliability deteriorates causing silence and customer hangups, but implementing failover detection and announcement mechanisms increases device complexity
Solution Approach 1:
The system sends status messages to check server health before failures occur and prepares failover announcements in advance. The media server is configured to detect failures and play backup announcements proactively, preventing silence gaps rather than reacting after complete failure.
Solution Approach 2:
The system implements a feedback mechanism where the media server sends status messages to application servers and waits for responses. If no response is received within a timeout period, the system detects the failure and triggers the failover announcement sequence, creating a closed-loop monitoring system.
2Measurement precision
If the media server waits for status message responses to detect failure, then detection precision improves, but loss of time increases during the waiting period
Solution Approach 1:
The system uses timeout-based failure detection where the media server sends status messages and waits for a predetermined time period. If no response is received within this timeout window, the system immediately assumes failure and triggers failover announcements, skipping prolonged waiting periods while maintaining reasonable detection accuracy.
Solution Approach 2:
The system pre-configures timeout thresholds and failover announcement sequences before failures occur. When status messages are not acknowledged within the predetermined time, the system immediately executes the pre-prepared failover procedure, reducing response time while maintaining detection precision through the timeout mechanism.
Data Source
AI summary
To handle a failover condition, a media server receives a request, from a first application server, to stream a first media message in a media channel of a communication session. The first media message is streamed in the media channel of the communication session by the media server. Once the first media message has ended, a status message can be sent to the first application server to determine if the first application server has failed. If a response to the status message is not received (i.e., because the first application server has failed), the media server can stream a second media message during a period where a second application server is failing over for the first application server. If a response to the status message is received, the second media message is not streamed.