API Documentation
Rerun Jobs
/api/rerun-jobThe POST API endpoint /rerun-job puts a Complete or Stopped job back in the queue, as the Rerun button on its page does: the same job, under the same name and with the same settings, its earlier outputs replaced as it runs. Send a batch name to rerun the batch's jobs. Rerunning is available on Premium plans.
Parameters:
Name of the job to rerun. For a batch, the batch name: its jobs are rerun.
For a batch only: "all" (the default) reruns every job in it, "incomplete" every job that is not Complete. The batch itself must be Complete or Stopped.
Note: a job that is still Running or In Queue can't be rerun; stop it first with /stop-job. A pipeline step's job can be rerun once it is Stopped. A rerun passes the checks a new submission passes: access to the job's tool, ownership of every file and molecule the job reads, your queue limit and your compute budgets. When a rerun is refused nothing is changed, and the JSON body's error says why; a plan without reruns answers 403 with "requiresPremium": true.
HTTP Response Status Codes
| Status code | Description |
|---|---|
| 200 | The job, or the batch's jobs, are back in the queue |
| 400 | Missing or invalid API key, missing "jobName", no job of yours has that name, the job can't be rerun in its status, a bad "only", nothing to rerun, a refused parent model, or your queue is full |
| 403 | Your plan doesn't include reruns ("requiresPremium": true), the tool or its licence is no longer available to you ("toolDenied": true), an input is no longer yours to read, or a compute budget would be exceeded |
| 409 | The job's inputs were never ingested or are still being ingested, or its settings are stored apart from it |
| 502 | A pipeline step's job: its pipeline state was reopened but the rerun could not be confirmed; check the job before retrying |
| 503 | A check could not be made right now and nothing was changed; send the same request again |
1import requests
2
3api_key = "insert-api-key-here"
4headers = {'x-api-key': api_key}
5base_url = "https://app.tamarind.bio/api/"
6
7# A job: runs again under the same name, with the same settings.
8response = requests.post(base_url + "rerun-job", headers=headers, json={"jobName": "myJobName"})
9print(response.status_code, response.json())
10
11# A batch: rerun only the jobs that did not complete.
12response = requests.post(base_url + "rerun-job", headers=headers,
13 json={"jobName": "myBatchName", "only": "incomplete"})
14print(response.status_code, response.json())Response Format
1{
2 "message": "myJobName rerun",
3 "rerunCount": 1
4}