A payments team has agreed that 99.9 percent of requests must succeed each month, and wants to know when it is spending its error budget too fast. Which tooling provides the signals it needs?
Show answer and explanation
Answer: B
SLOs are built on service level indicators, such as the proportion of successful requests and the share answered within a latency threshold. Application Insights records request outcomes and durations, which can be queried to compute these indicators. Alert rules over those measurements can fire when the error budget is being spent faster than planned, so the team can act before the objective is missed. Boards, commit history, and build durations describe delivery activity rather than service reliability.
Why the other options are wrong
- A. Boards and burndown charts track planned work and remaining effort in a sprint. They do not measure whether production requests succeed or how quickly they complete.
- C. Commit history shows who changed code and when. It says nothing about how the running service performed for its users.
- D. Build duration measures how long the pipeline takes. Faster builds do not indicate that customers are receiving reliable responses from the service.