A packet analysis software provider found that their deep packet inspection application kept crashing intermittently. They could sometimes reproduce the crash with a debugger, but the call context information was lost, as the bug corrupted the stack itself, so the root cause of the crashes remained elusive.
For site-wide long-term coverage, CriticalBlue's memory error detection technology was customized and deployed at the customer's premises. This allowed them to prevent most software faults thereafter, even latent ones, before they had the chance to damage their company's installed base and reputation.
As the customer was one of the successful businesses that decided to rely on CriticalBlue's advanced memory access tracking technology, they had another option: request CriticalBlue to trace all memory-accessing instructions run by their application and to obtain a record of all call contexts visited at runtime. Using the record, they could finally link the visible effect of the crash (symptom) to the actual line of code that caused the corruption (root cause), which turned out to be an occasional buffer overrun.