Common event contract
Fields that keep these queries reusable
- timestamp_utc, service, database_name, environment, and release
- query_fingerprint, transaction_id, pool_name, consumer_name, and migration_id
- duration_ms, wait_ms, lag_seconds, status, error_type, and capacity fields
Definitions before SQL
Decisions the query cannot make for you
- 1Normalize query operations before collection instead of storing raw SQL and parameters.
- 2Separate client pool wait, database execution, lock wait, and downstream replication delay.
- 3Tie database symptoms to an application service, release, and user-visible workflow.
Recommended sequence
Build detection first, then diagnosis
Analysis patterns
Make the result explain a decision
Instrument the client boundary
Record the safe operation fingerprint, service, final status, and phase durations where the application receives the database outcome.
Pair pressure with impact
A full pool or temporary lock becomes actionable when callers wait, time out, roll back, or fail a user-visible workflow.
Preserve rollout context
Keep release, migration, replica, and region fields so a regression can be isolated to the deployment boundary that introduced it.
Complete recipes
Copy the query, then validate the assumptions
Find Slow Database Queries by Fingerprint
Rank normalized database operations by slow-query rate, average duration, and worst observed duration without storing raw SQL or parameters.
Which database operations are consistently slow enough to investigate?
See SQL and resultMeasure Database Connection-Pool Saturation
Measure average pool utilization, queued acquisition waits, timeouts, and idle capacity by application service.
Which application pools are making callers wait for a database connection?
See SQL and resultCalculate Database Transaction Rollback Rate
Compare committed and rolled-back transactions by service while preserving error categories, duration, and affected accounts.
Which services roll back an unusual share of database transactions?
See SQL and resultFind Database Lock Waits and Deadlocks
Rank blocked database operation fingerprints by incident count, wait duration, unresolved locks, and detected deadlocks.
Which database operations create the most serious lock contention?
See SQL and resultMeasure Database Replication and CDC Lag
Compare replica and change-data-capture consumers by average lag, worst lag, bytes behind, and stale-sample rate.
Which replicas or CDC consumers are falling behind their source database?
See SQL and resultTrack Database Migration Failures by Release
Compare database migration success, failure, rollback, and duration by application release before completing a rollout.
Which releases contain failed or rolled-back database migrations?
See SQL and resultDetect N+1 Database Query Patterns
Detect database fingerprints repeated many times inside one application request.
Which route and query combinations repeat suspiciously within a request?
See SQL and resultAdapt the event contract before the threshold
Keep the analysis pattern, but validate table names, field types, business definitions, time windows, and minimum-volume rules against your own events. Every published query is also planned and executed against an empty typed table with the pinned engine.