Question

A SaaS company runs a customer-facing application on Cloud Run. After a recent release, support tickets spike with users reporting intermittent failures, but the operations team is struggling to identify which code defects are responsible because individual log entries are scattered across thousands of requests. The team wants a low-effort way to automatically group recurring application exceptions, see how often each distinct fault occurs, track whether a fault is new or regressed after a deploy, and get notified when a new error type appears. Which Google Cloud Observability capability should they adopt?

Accepted Answer

Enable Error Reporting, which automatically aggregates and deduplicates stack traces into distinct error groups with occurrence counts and new-error notifications.

Answer

Create a log-based metric that counts all log entries with severity ERROR and build a Cloud Monitoring dashboard from it.

Answer

Configure Cloud Trace with 100% sampling to capture every request span and inspect latency waterfalls for failures.

Answer

Export all Cloud Logging entries to BigQuery and write scheduled queries that GROUP BY the raw message text.