For AI agents: the complete documentation index is at llms.txt. Every page is also available as markdown by appending .md to its URL, or by sending an Accept: text/markdown request header.

Storage policy

note

Storage policy is Enterprise only.

Storage policies automate partition lifecycle management: they convert older partitions to Parquet, upload them to object storage, and drop local and remote copies on a schedule. These settings control the scan interval, retry behavior, and worker threads for the storage policy engine, which drives every lifecycle stage including the remote ones.

For details, see the storage policy concept page. The object store connection and the cold-read path are configured separately, under cold storage.

storage.policy.check.interval​

  • Default: 5m
  • Reloadable: no

How often QuestDB scans for partitions to process. This is a backstop full sweep: lifecycle transitions such as an applied partition seal or a manifest flush also trigger a recheck, so convergence is not bound to this interval.

storage.policy.recheck.drain.interval​

  • Default: 1s
  • Reloadable: no

How often coalesced recheck requests are drained, off the worker hot loop. Requests coalesce per table, so a burst affecting one table costs a single recheck. This paces convergence latency rather than storage.policy.check.interval.

storage.policy.max.reschedule.count​

  • Default: 20
  • Reloadable: no

Maximum number of retries before abandoning a storage policy task.

storage.policy.retry.interval​

  • Default: 1m
  • Reloadable: no

Retry interval for failed storage policy tasks.

storage.policy.worker.affinity​

  • Default: -1
  • Reloadable: no

CPU affinity for each storage policy worker thread (comma-separated list). -1 means no affinity.

storage.policy.worker.count​

  • Default: 4
  • Reloadable: no

Number of storage policy worker threads. Setting to 0 disables the feature, including the upload and cold-switch path used by cold storage, so it must be greater than 0 for remote stages to run.

storage.policy.worker.haltOnError​

  • Default: false
  • Reloadable: no

Whether a storage policy worker thread halts when it hits an unhandled error.

storage.policy.worker.nap.threshold​

  • Default: 100
  • Reloadable: no

Number of idle worker-loop iterations before a storage policy worker naps.

storage.policy.worker.sleep.threshold​

  • Default: 500
  • Reloadable: no

Number of idle worker-loop iterations before a storage policy worker sleeps.

storage.policy.worker.sleep.timeout​

  • Default: 100ms
  • Reloadable: no

Sleep duration when a storage policy worker has no tasks to process.

storage.policy.worker.yield.threshold​

  • Default: 10
  • Reloadable: no

Number of idle worker-loop iterations before a storage policy worker yields.