Playground (pgpg)

Incident tracks

Scenarios are organized into tracks — groups of related incidents that share a theme. There are 112 scenarios across 12 tracks, ranging from focused single-symptom drills to messy, compound outages.

The 12 tracks

  • Query Performance — slow queries, bad plans, missing indexes, and statistics gone stale.
  • Locks, Blocking & Transactions — lock contention, blocked sessions, deadlocks, and long-running transactions.
  • Connections, Pooling & App Integration — connection exhaustion, pooler misconfiguration, and application-side patterns.
  • Replication & WAL — replication lag, WAL growth, and standbys that fall behind or break.
  • Storage, Backup & Recovery — disk pressure, backup failures, and point-in-time recovery.
  • Multi-Database Operations — incidents spanning multiple databases and cross-database workflows.
  • Vacuum, Bloat & XID — autovacuum tuning, table and index bloat, and transaction ID wraparound risk.
  • HA & Failover — high-availability setups, failover handling, and split-brain avoidance.
  • Migrations & Releases — schema changes and deploys that go wrong under live traffic.
  • Security & Access — roles, privileges, and access-control misconfigurations.
  • Compound Production Incidents — multi-cause outages that combine symptoms from several tracks.
  • Incident Control — exercises in triage, prioritization, and steady incident response.

Scenario IDs

Each scenario ID encodes its track and stage. For example:

stage-04/02-replication-lag-spike

Here stage-04 is the track and 02 is the scenario's position within it. You can pass this ID straight to start:

pgpg> start stage-04/02-replication-lag-spike

A growing catalog

The scenario catalog grows over time as new incident patterns are added, and the full set is included in your subscription — new scenarios appear automatically, at no extra cost.