Skip to content
deedglassdocs

replay_too_large

POST /v1/policies/{id}/versions/{v}/replay counts the actions in the window before it queues anything, and the window holds more than 1,000,000 of them.

HTTP 400 · invalid_request_error · not retryable

What happened

Section titled “What happened”

POST /v1/policies/{id}/versions/{v}/replay counts the actions in the window before it queues anything, and the window holds more than 1,000,000 of them. Replay re-evaluates every stored decision input, so the limit keeps one job bounded. Nothing was queued, and the refusal doesn’t count toward the 10 replays an hour. A queued replay can also end failed with this code in its error object, when the window holds more than the limit by the time a runner counts it again.

What to do

Section titled “What to do”

Send a narrower window with from and to (RFC 3339). Split a long period into several replays if you need all of it.

Example

Section titled “Example”
HTTP 400 · replay_too_large
{
  "error": {
    "type": "invalid_request_error",
    "code": "replay_too_large",
    "message": "This replay covers 1,240,512 actions; the limit is 1,000,000. Narrow the window with from and to.",
    "param": null,
    "request_id": "req_01j9zr3m5k8q2w7x4t6v0b1c9d",
    "doc_url": "https://docs.deedglass.com/errors/replay_too_large"
  }
}

The message is for a person and can get more exact over time: branch on type and code, never on the message.

In the SDKs

Section titled “In the SDKs”
SDK Raised as
TypeScript (@deedglass/sdk) InvalidRequestError with code === "replay_too_large"
Python (deedglass) deedglass.InvalidRequestError with .code == "replay_too_large"
Section titled “Related”