The Requested Checkpoints alert notifies you when its configured condition is met on PostgreSQL instances so you can investigate and respond.
Mini DBA describes this alert as: Frequent requested checkpoints can indicate WAL pressure and may increase write latency. Mini DBA evaluates this alert once a minute so changes are detected quickly. Because duration is used, Mini DBA can avoid treating a single short spike as a full incident when the condition clears quickly.
This alert shows frequent requested checkpoints, which can indicate that WAL volume is reaching its configured limit and can create bursts of write latency.
Enable it on any monitored PostgreSQL instance where this condition would require an administrator to investigate or communicate with the application team. Tune noisy development systems after you have a short baseline.
Threshold meaning: Requested checkpoints per minute. Major threshold: 10 checkpoints/min. Minor threshold: 3 checkpoints/min. Comparison direction: over. Duration is used, so prefer requiring the condition to persist before paging people for transient spikes. For an over-threshold alert, decreasing a threshold makes that severity fire sooner; increasing it tolerates more load or pressure. Set the minor threshold as an early-warning level and the major threshold at the highest acceptable value for the service.
Compare requested and timed checkpoints with WAL generation, checkpoint write time, storage latency, and workload bursts. Reduce avoidable write amplification or adjust max_wal_size and checkpoint settings carefully after confirming disk capacity and recovery requirements.
Require a sustained requested-checkpoint rate and account for known bulk loads. Do not suppress a recurring pattern that coincides with write-latency spikes.