What was the root cause and how did you prevent it from happening again?

Situation: In a data pipeline, we experienced a sudden spike in data latency that caused downstream dashboards to lag by 30 minutes. Task: Identify the cause and ensure it didn’t recur. Action: I performed a root‑cause analysis using the 5‑Whys technique. The first why revealed that a scheduled batch job was running concurrently with a nightly data load, causing resource contention. The second why showed that the job scheduler did not enforce exclusive resource locks. I implemented a lock‑file mechanism and updated the scheduler configuration to serialize the jobs. I also added a monitoring alert for CPU usage spikes and created a rollback plan for future incidents. Result: The latency issue was resolved within two hours, and since then, the pipeline has maintained SLA compliance. The new lock mechanism and monitoring have prevented any recurrence of the same problem.