Skip to main content
For a dashboard focused on one Endpoint and its related Jobs, pass verified Job template IDs through Inputs as JSON, for example {"templateIds":["verified-template-id"]}. The following inline JavaScript makes a bounded read, groups the returned runs by UTC day and returns a heading, chart, table and intentional spacer. It labels the result as partial when the 100-row cap may have omitted runs:
Keep the chosen time window and timezone visible, bound pagination, and mark totals as partial if a cap omits rows. A Jobs Performance panel summarizes the instance, so use a custom chart for a trend restricted to one Endpoint’s related Jobs. Save the matching filters and sort separately on the server-side Jobs History panel; Trace needs the Endpoint’s own ID.
Use a pinned Component digest when an incident or executive dashboard must keep a reviewed revision. Use Latest when new active revisions should be picked up automatically.
The outer dashboard grid has 20 columns. In the Platform API, a dashboard preference’s data is one flat array of panel objects. Each panel stores its position and size as x, y, cols, and rows; the custom widget’s returned rows and cell widths control only the layout inside that panel. A server-side Jobs History panel can retain its column filter and sort order when the dashboard is saved. A custom TABLE uses the returned data in the browser; its live table filters are not saved as dashboard state. Trace stores its own applied filters, including an optional Endpoint selection. Next: query and refresh only the cells you need.
Last modified on November 8, 2026