Skip to content

Rules

Severity level

Rules are functions that evaluate sensors results and assigns them severity levels: low (1), medium (2), high (3), or information that the result is valid.

Severity level

3-degree alert system that assigns levels to the sensor's results as low, medium or high.

Rule configuration

Rules are split into the following categories:

Basic rules compare sensor results to the thresholds (numerical values) configured directly by the user. Configuration for such sensors is done by providing mode: current_value in rule configuration.

There is also mode: previous_redaings that enables calculating rules thresholds based on historical values. 

apiVersion: dqo/v1
kind: rule
spec:
  type: python 
  mode: current_value

apiVersion: dqo/v1
kind: rule
spec:
  type: python
  mode: previous_readings
  time_window:
    prediction_time_window: 14
    min_periods_with_reading: 7

How to configure a rule for check

Rules are configured under sensor name field. Below row_count sensor there is a configuration that sets up the rule min_count. 

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
# yaml-language-server: $schema=https://cloud.dqo.ai/dqo-yaml-schema/TableYaml-schema.json
apiVersion: dqo/v1
kind: table
spec:
  target:
    schema_name: test_data
    table_name: string_dates
  time_series:
    mode: current_time
    time_gradient: day
  checks:
    consistency:
      row_count:
        rules:
          min_count:
            low:
              min_value: 90.0
            medium:
              min_value: 80.0
            high:
              min_value: 70.0
  columns:
    dates:
      type_snapshot:
        column_type: STRING
        nullable: true