Session Border Controller Message Blocking for Fault Avalanches
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In telephony applications, deterministic message or packet processing faults can lead to catastrophic failures and cascading crashes in Session Border Controller (SBC) clusters, particularly in cloud environments, due to uncorrelated faults causing message processing avalanches, which are difficult to predict and test, especially with text-based protocols like SIP, resulting in prolonged service outages.
Innovation Solution
Implementing a method to track key value fault counts and use key value-based message blocking for messages with fault counts exceeding a threshold, enabling message blocking at SBCs processing SIP messages, thereby reducing the impact of message processing failures and preventing cascading faults by dropping or blocking problematic messages.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If deterministic message processing faults are allowed to occur in SBC clusters, then fault detection and analysis can be performed, but message processing avalanches and cascading failures occur affecting multiple nodes
Solution Approach 1:
The system applies preliminary anti-action by blocking messages containing key values that have been identified as causing deterministic faults before they can trigger cascading failures. The SBC maintains a fault count table and blocks messages with key values exceeding a threshold, preventing the harmful effect from propagating to other nodes in the cluster.
Solution Approach 2:
The patent introduces an intermediary mechanism - the key value blocking table - that mediates between message processing and fault prevention. This table acts as a filter that intercepts potentially harmful messages before they reach the processing pipeline, allowing the system to prevent avalanches without completely stopping message flow.
2Reliability
If manual identification and elimination of offending entities is performed, then deterministic faults can be resolved, but service outages last for hours
Solution Approach 1:
The system implements self-service by automatically detecting, tracking, and blocking messages that cause deterministic faults. The SBC autonomously maintains fault count tables, identifies problematic key values, and blocks corresponding messages without requiring manual intervention, thereby eliminating hours-long service outages.
Solution Approach 2:
The patent employs feedback mechanisms where the SBC continuously monitors message processing outcomes, tracks fault counts by key value, and adjusts message blocking decisions accordingly. This closed-loop feedback enables real-time fault resolution without manual intervention.
3Reliability
If comprehensive testing for message processing faults is performed, then deterministic faults can be predicted, but the testing process is extremely difficult and time-consuming
Solution Approach 1:
The patent extracts the essential information needed for fault detection from complex SIP messages - specifically the key values (Call-ID, From, To, etc.) - and tracks their occurrence. This extraction simplifies the testing and monitoring process by focusing only on the critical fields that trigger deterministic faults rather than analyzing entire messages.
Solution Approach 2:
The system segments the message processing function by separating key value extraction and tracking from the main message processing logic. This segmentation allows independent analysis and testing of fault conditions without requiring comprehensive testing of the entire message processing pipeline.
4Adaptability or versatility
If cloud-based virtualized SBC services are deployed, then service scalability and flexibility improve, but deterministic faults can still cause cascading failures across multiple nodes
Solution Approach 1:
The patent applies universality by implementing a unified fault detection and blocking mechanism that works across all virtualized SBC nodes in the cloud environment. The key value blocking table and fault count tracking are implemented consistently across all nodes, enabling the system to handle deterministic faults uniformly regardless of which node processes the message first.
Data Source
AI summary
The present invention relates to methods and apparatus for detecting, minimizing and/or preventing message processing faults and in particular message processing fault avalanches at a single processing device and among a cluster of message processing devices. An exemplary method embodiment includes: enabling message blocking at a Session Border Controller (SBC) for Session Initiation Protocol (SIP) messages including a first key value in response to a first threshold number of SIP messages including the first key value having caused a SIP message processing failure; receiving, at the SBC, a SIP message including at least one key value; determining if message blocking is enabled for one or more key values included in the SIP message; and dropping by the SBC the SIP message in response to determining that message blocking is enabled for one or more key values included in the SIP message, but otherwise processing the SIP message by the SBC.