Refresh a visualization
Chronos can refresh one Visualize chart on a schedule, as a chained step, or from a trigger flow. The Chronos task type is visualize_refresh. This task ships with Gluesync 2.3.1 and Chronos 0.2.5. See Visualize for what a refresh does to the snapshot.
|
HTTP 202 means Core Hub accepted the refresh. |
What Chronos sends
The action is not tied to a pipeline. pipeline_id may be omitted or empty. viz_id is required and must be a non-empty visualization id.
At execution time Chronos posts to Core Hub with the same module token it already uses for other Core Hub tasks (subject chronos). It does not send a separate user API key. Core Hub runs the refresh as the visualization owner.
POST /visualize/refresh
{
"task": "visualize.refresh",
"vizId": "viz-1",
"parameters": { "country": "IT" },
"scheduleId": 42
}
-
taskis always the exact stringvisualize.refresh. Chronos does not sendtaskTypeor a nestedpayload. -
vizIdis required. -
parametersis an optional map of strings. Chronos omits the field when the map is empty. Nested objects and arrays are rejected. A null value in the map is sent as an empty string. -
scheduleIdis the Chronos schedule id when the job has one. Chronos omits the field when there is no schedule, including a manual run. It is never sent as null.
Create the job
Use the same daily picker or cron expression as any other Chronos job. See Schedule jobs.
POST /api/jobs
{
"name": "Nightly viz",
"task_type": "visualize_refresh",
"cron_expression": "0 3 * * *",
"is_cron_expression": true,
"viz_id": "viz-1",
"visualize_parameters": { "country": "IT" }
}
A chained event or a trigger-flow step uses the same task_type, viz_id, and optional visualize_parameters. Each of those steps also requires its own non-empty viz_id.
An update can replace viz_id and visualize_parameters. Setting task_type to visualize_refresh on an update still requires a non-empty viz_id.
How Core Hub answers
| Response | What Chronos does |
|---|---|
|
Success. Chronos does not retry. |
|
The run fails. Chronos does not retry. |
|
The visualization is gone. Chronos disables that schedule and does not retry. A run that has no schedule id is not able to disable a schedule. |
|
Chronos re-authenticates once, the same way its other Core Hub calls do, then retries that call once. |
|
Chronos retries with the same HTTP backoff it uses for group calls: three retries, backoff factor 1. There is no separate backoff policy for Visualize. |
Other 4xx responses, including 429, fail the run and are not retried.