/retry"
```
### Response
Returns the updated task object with `status` set to `INITIALIZING`.
Only **Dataset Creation** tasks in `FAILED` or `ABORTED` status can be retried. All previously uploaded media is preserved and reprocessed. A `400 Bad Request` is returned if the task is not in a retryable state.
***
## Task Types
The following task types appear in the `task_type` field:
| Task Type | Description |
| ------------------ | ------------------------------------------------------------------------------------------- |
| `DATASET_CREATION` | Creating a new dataset from uploaded media. |
| `MEDIA_ADDITION` | Adding new images or videos to an existing dataset. |
| `REINDEX` | Re-indexing a dataset after media addition to incorporate new items into the dataset index. |
| `ENRICHMENT` | Running an AI enrichment model on the dataset. |
| `CUSTOM_METADATA` | Importing or updating custom metadata fields. |
| `SNAPSHOT_RESTORE` | Restoring a dataset to a previously saved snapshot state. |
| `SNAPSHOT_CLONE` | Creating a new independent dataset from a snapshot. |
***
## Task Statuses
The following values appear in the `status` field:
| Status | Description |
| -------------- | --------------------------------------------------------------------------- |
| `INITIALIZING` | The task is performing initial setup and validation. |
| `RUNNING` | The task is actively processing. `progress_percentage` reflects completion. |
| `COMPLETED` | The task finished successfully. |
| `FAILED` | The task encountered an error. Check `error_message` for details. |
| `ABORTED` | The task was manually cancelled by a user. |
***
## Python Example
```python theme={"theme":"monokai"}
import requests
import time
VL_BASE_URL = "https://app.visual-layer.com"
JWT_TOKEN = ""
headers = {"Authorization": f"Bearer {JWT_TOKEN}"}
# List all failed tasks from the past week
resp = requests.get(
f"{VL_BASE_URL}/api/v1/tasks",
headers=headers,
params={
"task_status": "FAILED",
"start_time_from": "2026-02-24T00:00:00Z",
"sort_by": "start_time",
"sort_order": "desc",
},
)
resp.raise_for_status()
data = resp.json()
print(f"Found {data['total']} failed tasks")
for task in data["tasks"]:
print(f" {task['task_type']} on {task['dataset_name']}: {task['error_message']}")
# Retry failed dataset creation tasks
if task["task_type"] == "DATASET_CREATION":
retry_resp = requests.post(
f"{VL_BASE_URL}/api/v1/tasks/{task['task_id']}/retry",
headers=headers,
)
retry_resp.raise_for_status()
print(f" Retried task {task['task_id']}")
```
***
## Response Codes
See [Error Handling](/api-reference/errors) for the error response format and Python handling patterns.
| HTTP Code | Meaning |
| --------- | ---------------------------------------------------------------------------------- |
| **200** | Request successful. |
| **400** | Bad request — task is not in a retryable state (retry endpoint). |
| **401** | Unauthorized — check your JWT token. |
| **404** | Task or dataset not found. |
| **409** | Conflict — task cannot be aborted (already completed, failed, or in commit phase). |
| **500** | Internal Server Error — contact support if this persists. |
***
## Related Resources
Monitor and manage tasks from the Visual Layer interface.
Check the current processing state of a dataset.