Run stuck queued? queue.state waiting vs runtime_unavailable

queue.state waiting is normal pickup; runtime_unavailable means nothing claimed the run. Position is always null. Back off, then contact support.

3 min readSume
All posts

Read queue.state on the status payload before you panic. Sume reports waiting while the run is inside the normal pickup window, and runtime_unavailable when it waited longer than that and nothing claimed it.

What each state asks of you

queue.state on a queued run (Sume docs, read 2026-10-05)
queue.stateMeaningDo
waitingNormal pickup windowKeep polling status_url with backoff
runtime_unavailableWaited past the window, unclaimedBack off for retry_after_seconds
(either)queue.position is always nullDo not show a place-in-line

Poll the small payload

status_url returns status, next_action, cancelable, expires_at, queue and the URLs, without output or artifacts. Use expires_at as your ceiling: the deadline is 90 minutes after created_at, or earlier when the run is older than 25 minutes and was silent for 10.

curl -sS "$STATUS_URL" -H "Authorization: Bearer $SUME_API_KEY" \
  | jq '.data | {status, expires_at, queue}'

If it lasts

If runtime_unavailable lasts more than a few minutes, send a support ticket with the request_id. In a bulk queue the child still occupies a slot, so one stuck run slows the rest of the window; cancel it with POST /v1/format-runs/{run_id}/cancel if you no longer want it.

Sources

Related posts

More in Developers

All Developers posts

Written by Sume