See It Work

Watch an investigation

During a design partner session, a database engineer asked SIXTA to investigate a production anomaly. This is that investigation, replayed from the original Slack thread.

Aurora MySQL
Production cluster
15 min
Anomaly window
Datadog
Observability source
Read-only
SIXTA access level
# db-platform Production database operations
SIXTA is investigating...
0:00
Investigation Stats
Elapsed --:--
Metrics queried 0
Data sources 0
Events identified 0
Follow-up queries 0
Behind the Scenes
Query throughput (QPS)
Per-statement counters
Disk I/O + write IOPS
CPU + InnoDB buffer pool
Lock metrics + contention
Thread activity + connections
Replication lag
Slow query log
Handler statistics
7-day disk trend

Under 10 minutes. From alert to action.

SIXTA traced a production anomaly to a specific service account, identified the root cause, caught a misleading disk alert, and delivered four prioritised recommendations — all from a single Datadog connection.

< 2 min
To root cause
25+
Metrics analysed
4
Data sources
4
Recommendations

This investigation is available as a PDF on the Resources page.

This happens 24/7, across
every database on your team.

SIXTA investigates incidents, monitors capacity, and catches problems before they page your on-call. It gets better with every investigation.

Meet Your New DBRE
See a full day with SIXTA → How the platform works →