Skip to content
DevOps AI ToolKit
Newsletter
PostgreSQL Troubleshooting Toolkit

PostgreSQL Troubleshooting Toolkit

Diagnose slow queries and EXPLAIN plans, locks and deadlocks, bloat and vacuum, replication lag, and connection exhaustion with prompts and runbooks.

Start troubleshooting Browse Postgres prompts

Top PostgreSQL errors

Start with the most common production issues and troubleshooting paths.

canceling statement due to conflict with recovery

Fix PostgreSQL 'canceling statement due to conflict with recovery' on hot standbys: tune max_standby_delay, hot_standby_feedbac…

canceling statement due to lock timeout

Fix PostgreSQL 'canceling statement due to lock timeout': diagnose blocking sessions, long transactions, and tune lock_timeout…

could not serialize access due to concurrent update

Fix PostgreSQL 'could not serialize access due to concurrent update': understand REPEATABLE READ/SERIALIZABLE conflicts and add…

duplicate key value violates unique constraint

Fix PostgreSQL 'duplicate key value violates unique constraint': diagnose race conditions, out-of-sync sequences, and resolve w…

invalid page in block

Fix PostgreSQL 'invalid page in block': diagnose page-level corruption, identify the affected relation, and recover with checks…

out of shared memory

Fix PostgreSQL 'out of shared memory': diagnose max_locks_per_transaction exhaustion from many partitions and tables, and tune…

permission denied for table

Fix PostgreSQL 'permission denied for table': diagnose missing GRANTs, schema USAGE, role membership, and default privileges fo…

SSL connection has been closed unexpectedly

Fix PostgreSQL 'SSL connection has been closed unexpectedly': diagnose idle timeouts, server crashes, OOM kills, and network re…

Best PostgreSQL prompts

Use these prompts to turn symptoms, logs, and config into a structured troubleshooting plan.

Postgres Checkpoint & WAL Throughput Tuning

Smooth out checkpoint-driven I/O spikes and write stalls by tuning checkpoint, WAL, and full-page-write settings for the workload — without risking longer crash recovery than the RTO allows.

Postgres Disk-Full & pg_wal Growth Emergency Triage

Work through a Postgres disk-full or runaway pg_wal/pg_xlog incident under pressure — find what is consuming space, free it safely, and recover the instance without deleting WAL Postgres still needs.

Postgres Integer Primary-Key / Sequence Exhaustion Remediation

Plan the migration of an int4 (serial) primary key approaching its 2.1-billion ceiling to bigint, with a low-downtime backfill and a tested cutover before INSERTs start failing.

Postgres on Kubernetes Operator Troubleshooting

Diagnose a Postgres cluster managed by a Kubernetes operator (CloudNativePG, Crunchy PGO, Zalando) — stuck failovers, pods CrashLooping, PVC/storage issues, and split-brain risk — using operator status and pod logs.
All PostgreSQL prompts

Free PostgreSQL tool

Validate, troubleshoot, or analyze your configuration before production changes.

AI Incident Response Assistant

Paste a slow query, plan, or lock, get a structured investigation.

Start triage

PostgreSQL runbook

Use a repeatable checklist for production troubleshooting.

A first-response checklist for a struggling Postgres instance.

  1. 1 Check active queries and locks (pg_stat_activity, pg_locks)
  2. 2 Review slow queries and their EXPLAIN (ANALYZE) plans
  3. 3 Inspect bloat and autovacuum activity
  4. 4 Check replication lag and WAL status
  5. 5 Review connection limits and pooling (max_connections, PgBouncer)
Get the runbook checklist