Release control checks tasks against user-defined rules and policies before deployment. Non-compliant tasks are either blocked (strong control) or flagged with a warning (weak control) to help ensure platform stability.
Feature overview
Release control is part of change control, which includes change rules and change policies.
-
Create change rules: Define metadata based on object properties and set up validation rules using built-in system rules or custom code logic. Non-compliant objects are handled according to the associated change policy.
-
Create Change Policies: Define the scope and enforcement method for change rules. You can configure multiple rules within a single change policy, each with its own enforcement method.
Scenarios
Common scenarios for release control include:
-
Time Node Control: Restrict the release of new or updated tasks during critical periods, such as fiscal year-end, financial audits, or major events like Double 11, to reduce the risk of data calculation disruptions and protect cluster operations.
-
R&D Specification Control: Regulate releases based on task properties. For example, tasks in the ODS project must use the prefix
ods_to be eligible for release. -
Account Control: Restrict releases by submitter or publisher. For example, tasks submitted by former employees are blocked, and users whose Dataphin asset administration personal health score falls below a set threshold cannot perform releases.