Monitor Publication Jobs¶
Use the job ID returned by a publish or remove request to inspect the job's status, result summary, and logs through EPT, or list all of your jobs at once.
List your jobs¶
Authentication: Authorization: Bearer <access_token>. The publishing service scopes job reads to the username carried by the access token, so this always lists your own jobs. Returns up to limit jobs, newest first (default 100, range 1..1000). No refresh token is required.
curl "${EPT_API_URL}/v1/edito/stac/jobs?limit=100" \
--header "Authorization: Bearer ${ACCESS_TOKEN}"
{
"jobs": [
{
"id": "8d7d5a93-ff52-4a8e-9eb6-0d76d874b670",
"type": "stac_publish",
"status": "succeeded",
"username": "alice",
"created_at": "2026-07-01T10:00:00Z",
"started_at": "2026-07-01T10:00:01Z",
"finished_at": "2026-07-01T10:00:03Z",
"message": "Job completed successfully."
}
]
}
The list covers publication, removal, validation, and visibility jobs, ordered from newest to oldest. Use each job's id with the polling endpoint below.
Monitor one job¶
Authentication: Authorization: Bearer <access_token>.
Usage¶
Poll the endpoint while the job has status queued or running. The terminal statuses are succeeded, failed, cancelled, and interrupted; use the publisher-provided message and, when needed, log views to diagnose a terminal failure.
Job summary¶
curl "${EPT_API_URL}/v1/edito/stac/jobs/${JOB_ID}" \
--header "Authorization: Bearer ${ACCESS_TOKEN}"
{
"id": "8d7d5a93-ff52-4a8e-9eb6-0d76d874b670",
"type": "stac_publish",
"status": "succeeded",
"username": "alice",
"created_at": "2026-07-01T10:00:00Z",
"started_at": "2026-07-01T10:00:01Z",
"finished_at": "2026-07-01T10:00:03Z",
"message": "Job completed successfully."
}
Log pages¶
curl "${EPT_API_URL}/v1/edito/stac/jobs/${JOB_ID}?view=detail&limit=100" \
--header "Authorization: Bearer ${ACCESS_TOKEN}"
The summary fields remain at the top level. Log views add logs, total, limit, next, and page_message:
{
"id": "8d7d5a93-ff52-4a8e-9eb6-0d76d874b670",
"type": "stac_publish",
"status": "succeeded",
"message": "Job completed successfully.",
"logs": [],
"total": 2400,
"limit": 100,
"page_message": "Returned the first page of job events.",
"next": "/v1/edito/stac/jobs/8d7d5a93-ff52-4a8e-9eb6-0d76d874b670?view=detail&limit=100&cursor=1071"
}
Follow next exactly as returned, using the same EPT base URL and access token. The final page has "next": null.
Query parameters¶
| Query | Result |
|---|---|
none or ?view=summary |
Minimal job lifecycle summary and publisher message |
?view=detail |
Summary plus one page of readable, normalized logs |
?view=raw |
Summary plus one page of canonical redacted raw logs |
limit=1..1000 |
Maximum records in a log page; default is 100 |
cursor=<value> |
Continue from the cursor in the previous response's next link |
Unknown query parameters are rejected. limit and cursor are relevant only to detail and raw views.
Good to know¶
- This is a bodyless
GET, soContent-Typeis not required. Do not sendX-EDITO-Refresh-Token. In interactive API documentation, authorize with the access token only; theBearerprefix is added automatically. - This page documents EPT's public API, not the Publisher API directly. EPT forwards the request to the Publisher, presents log pages in EPT's response shape, and rewrites Publisher continuation links as EPT
nextlinks. EPT translates its publiccursorto the Publisher'safter_idparameter — do not substituteafter_idfor EPT'scursor; follownextrather than constructing cursor values yourself.
Export all raw logs as NDJSON¶
The following examples follow every next link and produce one compact JSON object per line. Each raw event is tagged with its job_id, making the file self-describing when it is moved, combined, or ingested elsewhere. Both examples write to a temporary file first, so an incomplete export is never presented as a finished .ndjson file.
Bash¶
set -euo pipefail
output="publication-job-${JOB_ID}.raw.ndjson"
temporary="${output}.partial"
next="/v1/edito/stac/jobs/${JOB_ID}?view=raw&limit=100"
rm -f "${temporary}"
trap 'rm -f "${temporary}"' EXIT
while [ -n "${next}" ]; do
page=$(curl --fail --silent --show-error "${EPT_API_URL}${next}" \
--header "Authorization: Bearer ${ACCESS_TOKEN}")
printf '%s\n' "${page}" \
| jq -c --arg job_id "${JOB_ID}" '.logs[] | . + {job_id: $job_id}' \
>> "${temporary}"
next=$(printf '%s\n' "${page}" | jq -r '.next // empty')
done
mv "${temporary}" "${output}"
trap - EXIT
printf 'Exported raw events to %s\n' "${output}"
For a single page, the essential operation is:
curl --fail --silent --show-error \
"${EPT_API_URL}/v1/edito/stac/jobs/${JOB_ID}?view=raw" \
--header "Authorization: Bearer ${ACCESS_TOKEN}" \
| jq -c --arg job_id "${JOB_ID}" '.logs[] | . + {job_id: $job_id}' \
> "publication-job-${JOB_ID}.raw.ndjson"
Python (httpx)¶
import json
import os
from pathlib import Path
import httpx
workspace = Path("./workspace")
workspace.mkdir(parents=True, exist_ok=True)
access_token = os.environ["ACCESS_TOKEN"]
base_url = os.getenv("EPT_API_URL", "http://localhost:8000")
job_id = os.environ["JOB_ID"]
output_path = workspace / f"publication-job-{job_id}.raw.ndjson"
temporary_path = output_path.with_suffix(output_path.suffix + ".partial")
headers = {"Authorization": f"Bearer {access_token}"}
try:
with httpx.Client(base_url=base_url, headers=headers, timeout=30.0) as client:
summary = client.get(f"/v1/edito/stac/jobs/{job_id}")
summary.raise_for_status()
print(summary.json()["status"])
next_url = f"/v1/edito/stac/jobs/{job_id}?view=raw&limit=100"
with temporary_path.open("w", encoding="utf-8") as output:
while next_url:
response = client.get(next_url)
response.raise_for_status()
page = response.json()
for event in page["logs"]:
event["job_id"] = job_id
output.write(
json.dumps(
event,
ensure_ascii=False,
separators=(",", ":"),
)
+ "\n"
)
next_url = page["next"]
temporary_path.replace(output_path)
print(f"Exported raw events to {output_path}")
finally:
temporary_path.unlink(missing_ok=True)