{% extends "_layout.html" %} {% set title = "The job took too long to get ready" %} {% set status = "Not an HTTP response" %} {% set description = "Never an HTTP response. A type on a job’s error object: staging ran past the caller’s max_staging_seconds." %} {% block main %}

The job spent longer staging than your max_staging_seconds allows, and ends failed. Staging is everything between the submit and the job queueing: fetching the sources the server supplies, reading the manifest and installing the job’s Python packages. The limit counts all of it together, and starts again each time the job stages, so a job sent back for a source and submitted again gets the whole limit a second time.

It is the job’s own limit, not the server’s failure. The commonest cause is a Python package set too large to install in the time.

What to do

  1. Resubmitting it unchanged times out again. Send fewer Python packages, or ones that install faster.
  2. If the job needs the time, ask the deployment’s operator for a larger max_staging_seconds. GET /v1/me’s limits says what yours is.

The body

{
  "id":    "01J9...",
  "state": "failed",
  "error": {
    "type":   "https://sc.example.com/server-errors/staging-timed-out",
    "title":  "The job took too long to get ready",
    "limit":  "max_staging_seconds",
    "detail": "..."
  }
}
MemberPresenceMeaning
limitREQUIRED "max_staging_seconds": the key in limits that was exceeded.

detail is prose and may be reworded at any time. Branch on type — never parse detail.

Where it comes from

GET /v1/jobs/{id} As the job’s error.type, and the reason on its staging → failed transition. Never an HTTP response: staging runs after the submit’s 202, so there is no request left to answer.

Getting help with your specific failure

This type arrives inside a job, not on a request, so there is no trace_id on it. Quote the job’s id when you escalate. The job’s staging artifact is the server’s record of what it was doing when the time ran out.

This page documents the error type. It is the same page for everyone who hits it and it cannot see your job — the job id is what connects the two.

What this is not

staging-failed is the server failing for its own reasons: an index that did not answer, a build that was lost. That one is the server’s, and resubmitting later may work. This one is the job’s limit.

{% endblock %}