Monitors as Code

Overview

Monte Carlo developed a YAML-based monitors configuration to help teams deploy monitors as part of their CI/CD process. The following guide explains how to get started with monitors as code.

Prerequisites

  1. Install the CLI — https://docs.getmontecarlo.com/docs/using-the-cli
  2. When running montecarlo configure, provide your API key

Using code to define monitors

First, you will need to create a Monte Carlo project. A Monte Carlo project is simply a directory which contains a montecarlo.yml file, which contains project-level configuration options. If you are using DBT, we recommend placing montecarlo.yml in the same directory as dbt_project.yml.

The montecarlo.yml format:

version: 1
default_resource: <string>
include_file_patterns:
  - <string>
exclude_file_patterns:
  - <string>
namespace: <string - optional> 

Description of options inside the montecarlo.yml file:

  • version: The version of MC configuration. Set to 1
  • default_resource: The warehouse friendly name or UUID where YAML-defined monitors will be created. The warehouse UUID can be obtained via the getUser API as described here.
    • If your account only has a single warehouse configured, MC will use this warehouse by default, and this option does not need to be defined.
    • If you have multiple warehouses configured, you will need to (1) define default_resource, and (2) specify the warehouse friendly name or UUID for each monitor explicitly in the resource property to override default resource. (see YAML format for configuring monitors below).
  • include_file_patterns: List of file patterns to include when searching for monitor configuration files. By default, this is set to **/*.yaml and **/*.yml . With these defaults, MC will search recursively for all directories nested within the project directory for any files with a yaml or yml extension.
  • exclude_file_patterns: List of file patterns to exclude when searching for monitor configuration files. For example: directory_name/* , filename__* , *.format

Example montecarlo.yml configuration file, which should be sufficient for customers with a single warehouse:

version: 1

Example montecarlo.yml configuration file, for customers with multiple warehouses configured.

version: 1
default_resource: bigquery

Defining individual monitors

📘

Define monitors in separate YML files than montecarlo.yml

Your montecarlo.yml file should only be used to define project-level configuration options. Use separate YML files to define individual monitors.

Monitors are defined in YAML files within directories nested within the Monte Carlo project. Monitors can be configured in standalone YAML files, or embedded within DBT schema.yml files within the meta property of a DBT model definition.

Monitor definitions inside of yml files must conform to the expected format in order to be processed successfully by the CLI. Some example monitor configurations, defining the most basic options can be found below.

For an exhaustive list of configuration options and their definitions, refer to the Monitor configuration reference section below.

Example of a standalone monitor definition:

namespace: <string - optional>
montecarlo:
  field_health:
    - table: project:dataset.table_name
      timestamp_field: created
      resource: default warehouse override
      name: field_health_test_monitor
      comparisons:
        - type: threshold
          operator: AUTO
          metric: NULL_RATE
  dimension_tracking:
    - table: project:dataset.table_name
      timestamp_field: created
      field: order_status
      resource: default warehouse override
      name: dt_test_monitor

Example of monitor embedded within a DBT schema.yml file:

version: 2

models:
  - name: table_name
    description: My table
    meta:
      montecarlo:
        field_health:
          - table: project:dataset.table_name
            timestamp_field: created
            name: field_health_test_monitor
            comparisons:
              - type: threshold
                operator: AUTO
                metric: NULL_RATE
        dimension_tracking:
          - table: '[[ ref("table_name") ]]'
            name: dimension_tracking_for_table_name
            description: 'Tracks dimensions for [[ ref("table_name") ]]'
            timestamp_field: created
            field: order_status
            name: dt_test_monitor

dbt ref resolution

In the snippet above, you can see an example of resolving dbt ref("<model_name>") notation for any string field within Monte Carlo monitor configs. Just wrap the ref in [[ ]] and make sure to quote the string as in the example above. To resolve refs, you must pass the --dbt-manifest <path_to_dbt_manifest> argument to the CLI when applying the monitor config with the path to your dbt manifest.json file (by default created in the target/ directory after running dbt commands like compile or run). Each dbt ref will be looked up in the manifest and replaced with the full table name. To use this feature you must be on version 0.42.0 or newer of the Monte Carlo CLI.

📘

Tip: Using monitors as code with DBT

If your organization already uses DBT, you may find that embedding monitor configurations within DBT schema.yml files may make maintenance easier, as all configuration/metadata concerning a given table are maintained in the same location. For an example DBT repo with some basic monitor configuration, click here.

Shortcut to building monitor configuration

If you want to get a head start on setting up monitors as code, visit the monitor details page of an existing monitor where you can find the YAML definition for that monitor's configuration.

Below is a screenshot of the YAML definition for an existing monitor within Monte Carlo.

Developing and testing locally

Once the Monte Carlo project is setup with a montecarlo.yml and at least one monitor definition in a separate .yml file, you can use the Monte Carlo CLI to apply them.

To apply monitor configuration to MC run:

montecarlo monitors apply --namespace <namespace>

Note in the above command, a namespace parameter is supplied. Namespaces are required for all monitors configured via Monitors as Code as they make it easier to organize and manage monitors at scale. Namespaces can either be defined by passing a --namespace argument in the CLI command or defined within the .yml files (see above for more details).

The apply command behaves as follows:

  1. MC will search for all monitor configuration elements in the project, both in standalone and embedded in DBT schema files. All monitor configuration elements will be concatenated into a single configuration template.
  2. MC will apply the configuration template to your MC account:
    1. Any new monitors defined since last apply will be created
    2. Any previously defined monitors present in current configuration template will be updated if any attributes have changed
    3. Any previously defined monitors absent from current configuration template will be deleted

Namespaces

📘

You can think of namespaces like Cloudformation stack names. It’s a logical separation of a collection of resources that you can define. Monitors from different namespaces are isolated from each other.

Some examples of why this is useful -

  1. You have multiple people (or teams) working on managing monitors and don’t want to conflict or override configurations.
  2. You want to manage different groups monitors in different pipelines (e.g. dbt models in CI/CD x & non-dbt models in CI/CD y).

Namespaces can also be defined within the montecarlo.yml file or within individual monitor definitions.

Namespaces are currently limited to 2000 resources. Support for this starts in CLI version 0.52.0, so please upgrade your CLI version if you are running into timeouts while applying changes.

To delete (destroy) a namespace:

montecarlo monitors delete --namespace <namespace>

This will delete all monitors for a given namespace.

Dry Runs

👍

The apply command also supports a --dry-run argument which will dry run the configuration update and report each operation. Using this argument just shows planned changes but doesn't apply them.

Integrating into your CI/CD pipeline

Deploying monitors within a continuous integration pipeline is straightforward. Once changes are merged into your main production branch, configure your CI pipeline to install the montecarlodata CLI:

pip install montecarlodata

And run this command:

MCD_DEFAULT_API_ID=${MCD_DEFAULT_API_ID} \
	MCD_DEFAULT_API_TOKEN=${MCD_DEFAULT_API_TOKEN} \
	montecarlo monitors apply \
			--namespace ${MC_MONITORS_NAMESPACE} \
			--project-dir ${PROJECT_DIR} \
                         --auto-yes

📘

Skipping confirmations

The --auto-yes option will disable any confirmation prompts shown by the command and should only be used on CI, where there's no interaction. Otherwise we recommend not including the --auto-yes so you can review what changes will be applied before confirming.

These environment variables need to be populated:

  • MCD_DEFAULT_API_ID: Your Monte Carlo API ID
  • MCD_DEFAULT_API_TOKEN: Your Monte Carlo API Token
  • MC_MONITORS_NAMESPACE: Namespace to apply monitor configuration to
  • PROJECT_DIR: If the montecarlo command is not run within your Monte Carlo project directory, you can specify it here. Optional.

Example projects with CI/CD

For some examples of monitors as code projects with CI/CD setup, checkout these repositories:

Monitor configuration reference

📘

The term labels is used to define the audiences for each monitor. labels and audiences are equivalent, but the term labels must be used in monitor config. Notification config uses the term audiences. See Notifications as Code for more info.

montecarlo:
  field_health:
    - table: <string>  # required
      name: <string>  # required
      resource: <string> # optional -- override default warehouse
      description: <string> # required
      notes: <string> # optional
      segmented_expressions:
        - <string> # Can be a field or a SQL expression
      timestamp_field: <string> # optional
      timestamp_field_expression: <string> # optional
      where_condition: <string>
      lookback_days: <int> # optional
      aggregation_time_interval: <one of 'day' or 'hour'> # optional
      min_segment_size: <int> # optional -- by default, fetch all segment sizes
      comparisons:
      	- type: threshold
          metric: <string>  # metric name, see below for allowed values
          custom_metric: # custom metric (when a metric name is not used)
              display_name: <string>
              sql_expression: <string>
          operator: <string>  # AUTO, GT, LT, etc. see below
          fields:  # optional -- monitor specific field list instead of all fields
            - <string>  # field name
          threshold_value: <float>  # used for operators besides AUTO and range operators
          lower_threshold: <float>  # used only for range operators
          upper_threshold: <float>  # used only for range operators
      schedule:  # optional -- by default, fixed schedule with interval_minutes=720 (12h)
        type: <fixed, or dynamic>  # required
        interval_minutes: <integer>  # required if fixed
        start_time: <date as isoformatted string>
        timezone: <timezone> # optional - select regional timezone for daylight savings ex. America/Los_Angeles
      labels: # labels are equivalent to audiences
        - <string>
  dimension_tracking:
    - table: <string>  # required
      name: <string>  # required
      resource: <string> # optional -- override default warehouse
      description: <string> # required
      notes: <string> # optional
      field: <string>  # required
      timestamp_field: <string> # optional
      timestamp_field_expression: <string> # optional
      where_condition: <string>
      lookback_days: <int> # optional
      aggregation_time_interval: <one of 'day' or 'hour'> # optional
      schedule:  # optional -- by default, fixed schedule with interval_minutes=720 (12h)
        type: <fixed, or dynamic>  # required
        interval_minutes: <integer>  # required if fixed
        start_time: <date as isoformatted string>
        timezone: <timezone> # optional - select regional timezone for daylight savings ex. America/Los_Angeles
      labels: # labels are equivalent to audiences
        - <string>
  json_schema:
    - table: <string>  # required
      name: <string> # required
      resource: <string> # optional -- override default warehouse
      description: <string> # required
      notes: <string> # optional
      field: <string>  # required
      timestamp_field: <string>
      timestamp_field_expression: <string>
      where_condition: <string>
      schedule:  # optional -- by default, fixed schedule with interval_minutes=720 (12h)
        type: <fixed, or dynamic>  # required
        interval_minutes: <integer>  # required if fixed
        start_time: <date as isoformatted string>
        timezone: <timezone> # optional - select regional timezone for daylight savings ex. America/Los_Angeles
      labels: # labels are equivalent to audiences
        - <string>
  custom_sql:
    - sql: <string>  # required
      name: <string> # required
      resource: <string> # optional -- override default warehouse
      comparisons: <comparison>  # required
      variables: <variable values>
      description: <string> # required
      notes: <string> # optional
      schedule:
        type: <string>  # must be 'fixed' or 'manual'
        start_time: <date as isoformatted string>
        interval_minutes: <integer>	
        interval_crontab: 
          - <string>
        timezone: <timezone> # optional - select regional timezone for daylight savings ex. America/Los_Angeles
          - <string>
      event_rollup_count: <integer> # optional - enable to only send notifications every X incidents
      event_rollup_until_changed: <boolean> # optional - enable to send subsequent notifications if breach changes, default: false
      labels: # labels are equivalent to audiences
        - <string>
      severity: <string> # optional - only use SEV-0 through SEV-4 - custom values will be rejected
  field_quality:
    - table: <string> / tables: <list> # required
      name: <string> # requried
      field: <string> / fields: <list> # required      
      resource: <string> # optional -- override default warehouse
      metric_type: <metric_type> # required, see below
      comparisons: <comparison>  # required, see below
      filters: # optional, rows to consider when collecting the metric
        - field: <string> # required
          operator: <string> # required, one of EQ, NEQ, LT, LTE, GT, GTE
          value: <string> required
      description: <string> # required
      notes: <string> # optional
      schedule:
        type: <string>  # must be 'fixed' or 'manual'
        start_time: <date as isoformatted string>
        interval_minutes: <integer>	
        interval_crontab: 
          - <string>
        timezone: <timezone> # optional - select regional timezone for daylight savings ex. America/Los_Angeles
          - <string>
      event_rollup_count: <integer> # optional - enable to only send notifications every X incidents
      event_rollup_until_changed: <boolean> # optional - enable to send subsequent notifications if breach changes, default: false
      labels: # labels are equivalent to audiences
        - <string>
      severity: <string> # optional - only use SEV-0 through SEV-4 - custom values will be rejected  
  freshness:
    - table: <string> / tables: <list> # required
      name: <string> # requried
      resource: <string> # optional -- override default warehouse
      freshness_threshold: <integer>  # required
      description: <string>
      notes: <string> # optional
      schedule:
        type: fixed  # must be fixed
        start_time: <date as isoformatted string>
        interval_minutes: <integer>
        interval_crontab: 
          - <string>
        timezone: <timezone> # optional - select regional timezone for daylight savings ex. America/Los_Angeles
      event_rollup_count: <integer> # optional - enable to only send notifications every X incidents
      event_rollup_until_changed: <boolean> # optional - enable to send subsequent notifications if breach changes, default: false
      labels: # labels are equivalent to audiences
        - <string>
      severity: <string> # optional - only use SEV-0 through SEV-4 - custom values will be rejected
  volume:
    - table: <string> / tables: <list>   # required
      name: <string> # required
      resource: <string> # optional -- override default warehouse
      comparisons: <comparison>  # required
      volume_metric: <row_count or byte_count>  # row_count by default
      description: <string>
      notes: <string> # optional
      schedule:
        type: fixed  # must be fixed
        start_time: <date as isoformatted string>
        interval_minutes: <integer>
        interval_crontab: 
          - <string>
        timezone: <timezone> # optional - select regional timezone for daylight savings ex. America/Los_Angeles
      event_rollup_count: <integer> # optional - enable to only send notifications every X incidents
      event_rollup_until_changed: <boolean> # optional - enable to send subsequent notifications if breach changes, default: false
      labels: # labels are equivalent to audiences
        - <string>
      severity: <string> # optional - only use SEV-0 through SEV-4 - custom values will be rejected
  comparison:
    - description: <string>
      name: <string> # required
      notes: <string>
      notify_rule_run_failure: <boolean>
      event_rollup_until_changed: <boolean>
      schedule:
        type: fixed # must be fixed
        interval_minutes: <integer>
        start_time: <date as isoformatted string>
        timezone: <timezone> # optional - select regional timezone for daylight savings ex. America/Los_Angeles
      query_result_type: <string> # required, one of LABELED_NUMERICS, ROW_COUNT, SINGLE_NUMERIC
      source_sql: <string>
      target_sql: <string>
      source_resource: <string>
      target_resource: <string>
      comparisons:
        - type: <string>  # only accepts source_target_delta
          operator: <string> # required GT, as we check if absolute delta value is greater than threshold. 
          threshold_value: <float>
          is_threshold_relative: <boolean>

🚧

Lookback Limits

Where we allow you to specify a longer lookback period on some monitors (in case the data in your table has historical timestamps), you cannot pick a number larger than 7. This is because for each day we "lookback", an additional query against your table is run. This is a safeguard to prevent specifying a very large period, like 90 days, and then having 90 queries run against your warehouse each time the monitor runs. If you need help with these windows, please feel free to reach out to [email protected] or the chat bot in the lower right hand corner.

🚧

Tables

table and tables fields passed in the config should contain valid full table ids. If the table doesn't exist in the assets page the validation will fail. Sometimes the catalog service takes time to detect newer tables, you can wait for the tables to be detected and then create monitors for them or (not recommended) you can apply the monitor config with --create-non-ingested-tables (update montecarlo cli to latest version) option to force create the tables along with the monitor. These tables will be marked deleted if they are not detected later on.

Common properties

  • name: Name is now a required field for any new monitors created. We only have an exception for older monitors created via Monitors as Code. But its recommended to add a name to older monitors as well.
  • resource: Optional warehouse name can be set here to override the default warehouse set in montecarlo.yml. For comparison rules, you can use source_resource and target_resource instead.
  • description: Friendly description of rule.
  • notes: Additional context for the monitor.
  • labels: Optional list of audiences associated with the monitor (labels are equivalent to audiences).
  • severity: Optional, pre-set the severity of incidents generated by this monitor. Only use SEV-0 through SEV-4 as custom values will be rejected.

Field Health Monitor

  • table: MC global table ID (format <database>:<schema>.<table name>
  • comparisons: List of configurations for metrics, fields, and breach conditions. A field may only be included once in a monitor for a given metric. See section below for details.
  • segmented_expressions: List of fields or SQL expressions used to segment the field (must have exactly one field in fields). Enables Monitoring by Dimension.
  • timestamp_field: Timestamp field
  • timestamp_field_expression: Arbitrary SQL expression to be used as timestamp field, e.g. DATE(created). Must use either timestamp_field or timestamp_field_expression or neither.
  • where_condition: SQL snippet of where condition to add to field health query
  • lookback_days: Lookback period in days. Default: 3. This option is optional and will be ignored unless timestamp_field is set.
  • aggregation_time_interval: Aggregation bucket time interval, either hour (default) or day. This option is optional and will be ignored unless timestamp_field is set.
  • min_segment_size: Minimum number of rows for a segment to be fetched. Defaults to 1. Can be used to avoid cardinality limits for datasets with a long tail of less-relevant segments.
  • sensitivity_level: Set sensitivity level to One of high, medium or low.
  • schedule
    • type: One of fixed, or dynamic
    • interval_minutes: For fixed, how frequently to run the monitor
    • start_time: When to start the schedule. If unspecified, for fixed schedules, then start sometime within the next hour.
    • timezone: Optional - select regional timezone for daylight savings ex. America/Los_Angeles
  • connection_name: Specify the connection (also known as query-engine) to use. Obtain the warehouse UUID via the getUser API as described here. Then obtain names of the connections in the warehouse via the getWarehouseConnections API as described here. Use ["sql_query"] as the jobType parameter in the getWarehouseConnections API call.

The monitored fields cannot exceed 300 fields, including important fields and manually specified fields. If they do, we'll attempt to keep all the manually specified fields and as many important fields as possible (ordered by importance score) until we reach 300 fields.

Field Health comparisons

  • type: accepts only threshold
  • operator: accepts AUTO, GT, GTE, LT, LTE, EQ, NEQ, OUTSIDE_RANGE, INSIDE_RANGE. AUTO uses ML to detect breaches. All other operators require an explicit threshold.
  • fields: List of field names that should be monitored for this metric, operator, and threshold. This field is optional and defaults to all fields in the table. The field type must be compatible with the metric.
    • Each field may only be included once in the monitor for a given metric (including as part of an "all fields" metric).
    • When using a custom metric, fields should not be included; the fields used should be included directly in the SQL expression.
  • threshold_value: Breach threshold for the operator. Not used for AUTO, OUTSIDE_RANGE, or INSIDE_RANGE. For rate metrics, ensure the threshold is between 0 and 1 (not 0%-100%).
  • lower_threshold: Lower end of threshold range for operators OUTSIDE_RANGE and INSIDE_RANGE
  • upper_threshold: Upper end of threshold range for operators OUTSIDE_RANGE and INSIDE_RANGE
  • metric: See the list of available metrics for full details, including whether AUTO is supported for each metric and which field types are compatible. All operators besides AUTO are compatible with every metric. Use the value from the MaC Name column in this field.
  • custom_metric: Allows defining a custom metric based on a SQL expression instead of a predefined metric indicated in metric. It can use any operator except AUTO.

Dimension Tracking Monitor

  • table: MC global table ID (format <database>:<schema>.<table name>
  • field: Field in table to monitor or a valid SQL expression that returns the row's dimension value as a string
  • timestamp_field: Timestamp field
  • timestamp_field_expression: Arbitrary SQL expression to be used as timestamp field, e.g. DATE(created). Must use either timestamp_field or timestamp_field_expression or neither.
  • where_condition: SQL snippet of where condition to add to query
  • lookback_days: Lookback period in days. Default: 3. This option is optional and will be ignored unless timestamp_field is set.
  • aggregation_time_interval: Aggregation bucket time interval, either hour (default) or day. This option is optional and will be ignored unless timestamp_field is set.
  • schedule
    • type: One of fixed, or dynamic
    • interval_minutes: For fixed, how frequently to run the monitor
    • start_time: When to start the schedule. If unspecified, for fixed schedules, then start sometime within the next hour.
    • timezone: Optional - select regional timezone for daylight savings ex. America/Los_Angeles
  • connection_name: Specify the connection (also known as query-engine) to use. Obtain the warehouse UUID via the getUser API as described here. Then obtain names of the connections in the warehouse via the getWarehouseConnections API as described here. Use ["sql_query"] as the jobType parameter in the getWarehouseConnections API call.

JSON Schema Monitor

  • table: MC global table ID (format <database>:<schema>.<table name>
  • field: Field in table to monitor
  • timestamp_field: Timestamp field
  • timestamp_field_expression: Arbitrary SQL expression to be used as timestamp field, e.g. DATE(created). Must use either timestamp_field or timestamp_field_expression or neither.
  • where_condition: SQL snippet of where condition to add to query
  • schedule
    • type: One of fixed, or dynamic
    • interval_minutes: For fixed, how frequently to run the monitor
    • start_time: When to start the schedule. If unspecified, for fixed schedules, then start sometime within the next hour.
    • timezone: Optional - select regional timezone for daylight savings ex. America/Los_Angeles
  • connection_name: Specify the connection (also known as query-engine) to use. Obtain the warehouse UUID via the getUser API as described here. Then obtain names of the connections in the warehouse via the getWarehouseConnections API as described here. Use ["json_schema"] as the jobType parameter in the getWarehouseConnections API call.

SQL Rule

  • sql: SQL of rule
  • query_result_type: Optional, can be set to SINGLE_NUMERIC to make the rule use a value-based threshold
  • sampling_sql: Optional custom SQL query to be run on breach (results will be displayed in Incident IQ to help with investigation).
  • comparisons: See comparisons below
  • variables: See variables below
  • schedule
    • type: Can be fixed or manual. Manual would be for SQL rules implemented during processes like Circuit Breakers.
    • interval_minutes: How frequently to run the monitor (in minutes).
    • interval_crontab: How frequently to run the monitor (using a list of CRON expressions, check example below).
    • start_time: When to start the schedule. If unspecified, for fixed schedules, then start sometime within the next hour.
    • timezone: Optional - select regional timezone for daylight savings ex. America/Los_Angeles
  • event_rollup_count: Optional - a Reduce Noise option to only send notifications every X incidents
  • event_rollup_until_changed: Optional - a Reduce Noise option to send subsequent notifications if breach changes

SQL Rule comparisons

📘

comparisons are definitions of breaches, not expected return values. This section would be where you would define the logic for when to get alerted about anomalous behavior in your monitor. For example, if you make a custom SQL rule and pick:

  • type: threshold
  • operator: GT
  • threshold_value: 100

When Monte Carlo runs your monitor and the return results are greater than 100, we will fire an alert to any routes configured to be notified about breaches to this monitor.

  • type: threshold, dynamic_threshold or change. If threshold, threshold_value below is an absolute value. If dynamic_threshold no threshold is needed (it will be determined automatically). If change, threshold_value as change from the historical baseline
  • operator: One of EQ, NEQ, GT, GTE, LT, LTE, or AUTO. Operator of comparison, =, ≠, >, ≥, <, ≤ respectively. AUTO is used only for dynamic_threshold (used by default, so does not need to be provided).
  • threshold_value: Threshold value. Should not be provided for dynamic_threshold.
  • baseline_agg_function: If type = change, the aggregation function used to aggregate data points to calculate historical baseline. One of AVG, MAX, MIN.
  • baseline_interval_minutes: If type = change, the time interval in minutes (backwards from current time) to aggregate over to calculate historical baseline
  • is_threshold_relative: If type = change, whether or not threshold_value is a relative vs absolute threshold. is_threshold_relative: true would be a percentage measurement, is_threshold_relative: false would be a numerical measurement. Relative means the threshold_value will be treated as a percentage value, Absolute means the threshold_value will be treated as an actual count of rows.
  • connection_name: Specify the connection (also known as query-engine) to use. Obtain the warehouse UUID via the getUser API as described here. Then obtain names of the connections in the warehouse via the getWarehouseConnections API as described here. Use ["sql_query"] as the jobType parameter in the getWarehouseConnections API call.

SQL Rule variables

📘

When defining custom sql sentences, you can use variables to execute the same sentence for different combinations of values. Variables are defined as {{variable_name}}. Then, you can define one or more values for each variable, and all combinations will be tested.

Here is an example defining the same sentence for several tables and conditions (4 sentences will be executed):

custom_sql:
    - sql: |
         select foo from {{table}} where {{cond}}
      variables:
        table: 
            - project:dataset.table1 
            - project:dataset.table2
        cond: 
            - col1 > 1
            - col2 > 2 

Field Quality Rule

  • table: MC global table ID (format <database>:<schema>.<table name>)
  • tables: Instead of table, you can also use tables to define a list of tables (check example with Getting Started multiple tables below](https://docs.getmontecarlo.com/docs/monitors-as-code#example-with-multiple-tables)).
  • field: Field name
  • fields: Instead of field, you can also use fields to define a list of fields. All the fields must be present in the selected tables and have the same type if multiple tables are provided.
  • metric_type: What metric will be collected. Certain metrics are only available on specific data types:
    • Numeric: APPROX_DISTINCTNESS, APPROX_DISTINCT_COUNT, DUPLICATE_COUNT, NULL_RATE, NULL_COUNT, NON_NULL_COUNT, NUMERIC_MEAN, NUMERIC_MEDIAN, NUMERIC_MIN, NUMERIC_MAX, PERCENTILE_20, PERCENTILE_40, PERCENTILE_60, PERCENTILE_80, ZERO_RATE, ZERO_COUNT, NEGATIVE_RATE, NEGATIVE_COUNT, NUMERIC_STDDEV, SUM
    • Text: APPROX_DISTINCTNESS, APPROX_DISTINCT_COUNT, DUPLICATE_COUNT, NULL_RATE, NULL_COUNT, NON_NULL_COUNT, EMPTY_STRING_RATE, EMPTY_STRING_COUNT, TEXT_ALL_SPACES_RATE, TEXT_ALL_SPACES_COUNT, TEXT_NULL_KEYWORD_RATE, TEXT_NULL_KEYWORD_COUNT, TEXT_MAX_LENGTH, TEXT_MIN_LENGTH, TEXT_MEAN_LENGTH, TEXT_STD_LENGTH, TEXT_INT_RATE, TEXT_NOT_INT_COUNT, TEXT_NUMBER_RATE, TEXT_NOT_NUMBER_COUNT, TEXT_UUID_RATE, TEXT_NOT_UUID_COUNT, TEXT_SSN_RATE, TEXT_NOT_SSN_COUNT, TEXT_US_PHONE_RATE, TEXT_NOT_US_PHONE_COUNT, TEXT_US_STATE_CODE_RATE, TEXT_NOT_US_STATE_CODE_COUNT, TEXT_US_ZIP_CODE_RATE, TEXT_NOT_US_ZIP_CODE_COUNT, TEXT_EMAIL_ADDRESS_RATE, TEXT_NOT_EMAIL_ADDRESS_COUNT, TEXT_TIMESTAMP_RATE, TEXT_NOT_TIMESTAMP_COUNT
    • Time: APPROX_DISTINCTNESS, APPROX_DISTINCT_COUNT, DUPLICATE_COUNT, NULL_RATE, NULL_COUNT, NON_NULL_COUNT, FUTURE_TIMESTAMP_RATE, FUTURE_TIMESTAMP_COUNT, PAST_TIMESTAMP_RATE, PAST_TIMESTAMP_COUNT, UNIX_ZERO_RATE, UNIX_ZERO_COUNT
    • Date: APPROX_DISTINCTNESS, APPROX_DISTINCT_COUNT, DUPLICATE_COUNT,NULL_RATE, NULL_COUNT, NON_NULL_COUNT, FUTURE_TIMESTAMP_RATE, FUTURE_TIMESTAMP_COUNT, PAST_TIMESTAMP_RATE, PAST_TIMESTAMP_COUNT, UNIX_ZERO_RATE, UNIX_ZERO_COUNT
    • Boolean: APPROX_DISTINCTNESS, APPROX_DISTINCT_COUNT, DUPLICATE_COUNT, NULL_RATE, NULL_COUNT, NON_NULL_COUNT, TRUE_RATE, TRUE_COUNT, FALSE_RATE, FALSE_COUNT
  • comparisons: See comparisons below
  • filters: See filters below
  • schedule
    • type: Can be fixed or manual. Manual would be for rules used during processes like Circuit Breakers.
    • interval_minutes: How frequently to run the monitor (in minutes).
    • interval_crontab: How frequently to run the monitor (using a list of CRON expressions, check example below).
    • start_time: When to start the schedule. If unspecified, for fixed schedules, then start sometime within the next hour.
    • timezone: Optional - select regional timezone for daylight savings ex. America/Los_Angeles
  • event_rollup_count: Optional - a Reduce Noise option to only send notifications every X incidents
  • event_rollup_until_changed: Optional - a Reduce Noise option to send subsequent notifications if breach changes
  • connection_name: Specify the connection (also known as query-engine) to use. Obtain the warehouse UUID via the getUser API as described here. Then obtain names of the connections in the warehouse via the getWarehouseConnections API as described here. Use ["sql_query"] as the jobType parameter in the getWarehouseConnections API call.

Field Quality Rule comparisons

  • type: threshold
  • operator: One of EQ, NEQ, GT, GTE, LT, LTE. Operator of comparison, =, ≠, >, ≥, <, ≤ respectively.
  • threshold_value: Threshold value

Field Quality Rule filters

  • field: The field name
  • operator: One of EQ, NEQ, GT, GTE, LT, LTE. Operator of comparison, =, ≠, >, ≥, <, ≤ respectively.
  • value: The value to filter on

Freshness Rule

  • table: MC global table ID (format <database>:<schema>.<table name>)
  • tables: Instead of table, can also use tables to define a list of tables (check example with Getting Started multiple tables below](https://docs.getmontecarlo.com/docs/monitors-as-code#example-with-multiple-tables)).
  • freshness_threshold: Freshness breach threshold in minutes
  • schedule
    • type: Must be fixed
    • interval_minutes: How frequently to run the monitor (in minutes).
    • interval_crontab: How frequently to run the monitor (using a list of CRON expressions, check example below).
    • start_time: When to start the schedule. If unspecified, for fixed schedules, then start sometime within the next hour.
    • timezone: Optional - select regional timezone for daylight savings ex. America/Los_Angeles
  • event_rollup_count: Optional - a Reduce Noise option to only send notifications every X incidents
  • event_rollup_until_changed: Optional - a Reduce Noise option to send subsequent notifications if breach changes

Volume Rule

  • table: MC global table ID (format <database>:<schema>.<table name>
  • tables: Instead of table, can also use tables to define a list of tables (check example with multiple tables below).
  • volume_metric: Must be total_row_count or total_byte_count — defines which volume metric to monitor
  • comparisons: See comparisons below
  • schedule
    • type: Must be "fixed"
    • interval_minutes: How frequently to run the monitor (in minutes).
    • interval_crontab: How frequently to run the monitor (using a list of CRON expressions, check example below).
    • start_time: When to start the schedule. If unspecified, for fixed schedules, then start sometime within the next hour.
    • timezone: Optional - select regional timezone for daylight savings ex. America/Los_Angeles
  • event_rollup_count: Optional - a Reduce Noise option to only send notifications every X incidents
  • event_rollup_until_changed: Optional - a Reduce Noise option to send subsequent notifications if breach changes

Volume Rule comparisons

  • type: absolute_volume or growth_volume.

If absolute_volume:

  • operator: One of EQ, GTE, LTE. Operator of comparison, =, ≥, ≤ respectively.
  • threshold_lookback_minutes: if operator is EQ, the time to look back to compare with the current value.
  • threshold_value: If operator is GTE or LTE, the threshold value

If growth_volume:

  • operator: One of EQ, GT, GTE, LT, LTE. Operator of comparison, =, >, ≥, <, ≤ respectively.
  • baseline_agg_function: the aggregation function used to aggregate data points to calculate historical baseline. One of AVG, MAX, MIN.
  • number_of_agg_periods: the number of periods to use in the aggregate comparison.
  • baseline_interval_minutes: the aggregation period length.
  • min_buffer_value / max_buffer_value: the lower / upper bound buffer to modify the alert threshold.
  • min_buffer_modifier_type / max_buffer_modifier_type: the modifier type of min / max buffer, can be METRIC (absolute value) or PERCENTAGE.

Comparison Rule

  • comparisons: See comparisons below
  • query_result_type: Should be set to one of SINGLE_NUMERIC to make the rule use a value-based threshold or LABELED_NUMERICS to make the rule use a label-based-threshold or ROW_COUNT
  • source_sql: SQL for source
  • target_sql: SQL for target to compare with source.
  • source_resource: Required Source warehouse name
  • target_resource: Required Target warehouse name
  • schedule
    • type: Must be "fixed"
    • interval_minutes: How frequently to run the monitor (in minutes).
    • interval_crontab: How frequently to run the monitor (using a list of CRON expressions, check example below).
    • start_time: When to start the schedule. If unspecified, for fixed schedules, then start sometime within the next hour.
    • timezone: Optional - select regional timezone for daylight savings ex. America/Los_Angeles
  • event_rollup_count: Optional - a Reduce Noise option to only send notifications every X incidents
  • event_rollup_until_changed: Optional - a Reduce Noise option to send subsequent notifications if breach changes
  • source_connection_name: Specify the source connection (also known as query-engine) to use. Obtain the warehouse UUID via the getUser API as described here. Then obtain names of the connections in the warehouse via the getWarehouseConnections API as described here. Use ["sql_query"] as the jobType parameter in the getWarehouseConnections API call.
  • target_connection_name: Specify the target connection (also known as query-engine) to use. Obtain the warehouse UUID via the getUser API as described here. Then obtain names of the connections in the warehouse via the getWarehouseConnections API as described here. Use ["sql_query"] as the jobType parameter in the getWarehouseConnections API call.

Comparison Rule comparisons

  • type: accepts only source_target_delta.
  • operator: accepts only GT. We compare if the absolute delta value is greater than threshold and raise error.
  • threshold_value: Max acceptable delta between source and target sql
  • is_threshold_relative: Whether or not threshold_value is a relative vs absolute threshold. is_threshold_relative: true would be a percentage measurement, is_threshold_relative: false would be a absolute measurement. Relative means the threshold_value will be treated as a percentage value, Absolute means the threshold_value will be treated as an actual count.

Example

montecarlo:
  field_health:
    - name: field_health_1
      description: Test monitor
      table: project:dataset.table_name
      timestamp_field: created
      schedule:
        type: dynamic
      labels: # labels are equivalent to audiences
        - label_name1
      comparisons:
      	# Breach for anomalous unique rate (ML-based) on any field
      	- type: threshold
          operator: AUTO
          metric: UNIQUE_RATE
        # Breach if null rate exceeds 5% on any field
      	- type: threshold
          operator: GT
          metric: NULL_RATE
          threshold_value: 0.05
        # Breach if true rate <= 75% for these fields
      	- type: threshold
          operator: LTE
          metric: TRUE_RATE
          threshold_value: 0.75
          fields:
            - boolean_field_1
            - boolean_field_2
        # Breach if max value is outside range [0,1] for these fields
      	- type: threshold
          operator: OUTSIDE_RANGE
          metric: NUMERIC_MAX
          lower_threshold: 0.0
          upper_threshold: 1.0
          fields:
            - numeric_field_1
            - numeric_field_2
        # Using a custom metric
        - type: threshold
          operator: LTE
          custom_metric:
            display_name: Unique text combination
            sql_expression: COUNT(DISTINCT CONCAT(text_field_1, text_field_2))
          threshold_value: 2
    - name: field_health_2
    	description: Test monitor
      table: project:dataset.table_name
      timestamp_field: created
      segmented_expressions:
        - segmented_expression
      schedule:
        type: dynamic
      comparisons:
      	- type: threshold
          operator: AUTO
          metric: NULL_RATE
  dimension_tracking:
    - name: dimension_tracking_1
      description: Test monitor
      table: project:dataset.table_name
      timestamp_field: created
      field: order_status
      labels: # labels are equivalent to audiences
        - label_name2
  custom_sql:
    - name: custom_sql_1
      description: Test rule
      sql: |
         select foo from project.dataset.my_table
      comparisons:
        - type: threshold
          operator: GT
          threshold_value: 0
      schedule:
        type: fixed
        interval_minutes: 60
        start_time: "2021-07-27T19:00:00"
      severity: SEV-1
    - name: custom_sql_2
      description: Test rule dynamic
      sql: |
         select foo from project.dataset.my_table
      comparisons:
        - type: dynamic_threshold
      schedule:
        type: fixed
        interval_minutes: 60
        start_time: "2021-07-27T19:00:00"
  field_quality:
    - name: field_quality_1
      description: Test rule
      table: project.dataset.my_table
      field: my_field
      metric_type: NULL_RATE
      comparisons:
        - type: threshold
          operator: GT
          threshold_value: 0
      schedule:
        type: fixed
        interval_minutes: 60
        start_time: "2021-07-27T19:00:00"
      severity: SEV-1
  freshness:
    - name: freshness_1
      table: project:dataset.table_name
      freshness_threshold: 30
      schedule:
        type: fixed
        interval_minutes: 30
        start_time: "2021-07-27T19:00:00"
 comparison:
   - description: Rule 1
     name: comparison_rule_1
     schedule:
       type: fixed
       interval_minutes: 60
       start_time: '2020-08-01T01:00:00+00:00'
       timezone: UTC
     query_result_type: LABELED_NUMERICS
     source_sql: select * from t1_id;
     target_sql: select * from t2_id;
     source_resource: lake
     target_resource: bigquery
     comparisons:
       - type: source_target_delta
         operator: GT
         threshold_value: 2.0
         is_threshold_relative: false

Example with multiple tables

montecarlo:
  freshness:
    - name: freshness_1
      description: Test rule
      tables: 
        - project:dataset.table_name1
        - project:dataset.table_name2
      freshness_threshold: 30
      schedule:
        type: fixed
        interval_minutes: 30
        start_time: "2021-07-27T19:00:00"

Example with CRON expressions

montecarlo:
  custom_sql:
    - name: custom_sql_1
      description: Test rule
      sql: |
         select foo from project.dataset.my_table
      comparisons:
        - type: threshold
          operator: GT
          threshold_value: 0
      schedule:
        type: fixed
        interval_crontab:
          - "0 10,16 * * MON-FRI"
          - "0 12 * * SAT-SUN"   
        start_time: "2021-07-27T19:00:00"