# This is an example yaml template that can be used to define a service's SLIs and SLOs as code,
# which will be visible in Datadog
# Once you update and complete this yaml template, please add it to your service's repository
#
# Please be sure to edit/update each attribute to your applications specifications
#
# If an attribute requires specific values, all possibilities will be listed as a comment above it.
# If you are unsure what metrics your service supports or should be logging, please refer to the
# README for instructions on how to generate a list of all supported metrics for your service.
#
# Mustache templating can be used anywhere. NOTE: If a template is not present on the CLI, it will
# default to an empty string.

business_unit: devops # should match business_unit in services deployment.yaml
# Include any tags that would be valuable; tags here will be appended to all SLOs and SLIs below
# tags:
# - tag1
# - tag2

# First slo definition
slos:
- short_name: test-slo-converter-slo-1 # terraform resource name; must be unique within this file
  name: test slo converter slo 1 # name of SLO within Datadog
  # Default threshold target is "99.5" when omitted; however acceptable values are floats > 0 and < 100
  # threshold:
  #   target: 99.5

  # Monitor definition:
  monitors:
  - type: error_budget # Query type
    window: 7_day # The alerting window size [7_day, 30_day, 90_day]
    alert:
      critical: # A string or a list of strings specifying slack channels to notify for critical errors
      - "@slack-temp-channel-1"
      - "{{ critical_alert }}"
      warn: "{{ warn_alert }}"
    threshold: # Error threshold in percentage
      critical: 100
      warning: 50 # optional

  # First sli definition
  slis:
  # - short_name: test-slo-converter-sli-1 # terraform resource name; must be unique within this file
  #   # Available values are latency, error-rate, and external
  #   # external types accept an optional name and tags filter
  #   # External SLIs are references to existing SLIs that are managed in a separate service.
  #   type: external
  #   service: service-core-bank
  #   name_filter: devops-portal test slo converter sli 1 Latency # name of SLI (monitor) within Datadog
  - short_name: test-slo-converter-sli-1 # terraform resource name; must be unique within this file
    name: test slo converter sli 1 # SLI name in Datadog becomes "<service> <name> <type>"; or "<service> <type>" when omitted
    # Update slack message content with additional details or insight to sli.
    # This becomes the monitor message body.
    message: |-
      This is the message text for this monitor.

      This is an example of line two with a blank line between.
      @slack-temp-slo-monitoring
    # Service name will become a tag on all SLIs and SLOs, but this field should match
    # the FigureDeployment service name so SLOs are included on the Datadog APM page.
    service: devops-portal
    # Available values are latency, error-rate, and external
    # latency types accept a percentile and metric field
    type: latency
    # Default percentile is p95 when omitted; however other options are p50, p75, p90, p95, p99
    # percentile: p95
    # Source trace that contains the resources below; supported metrics include
    # trace.http.request.*, trace.servlet.request.*, trace.grpc.server.*, and trace.kafka.consume.*
    metric: trace.http.request
    # Enter/Update resources with those generated using service-resource-generator.sh described
    # in README
    # Resources get joined with "OR"
    resources:
    - post_/api/v1/deployments
    - get_/api/v1/deployments
    - post_/api/v1/deployments/
    # Optional, resources get negated and joined with "AND"
    excluded_resources:
    - get_/health
    # Datadog monitor trigger conditions, in seconds
    threshold: # in seconds
      warning: 0.8 # optional
      critical: 1
    # Include any tags that would be valuable for this SLI
    tags:
    - tag1
    - tag2

  # Second sli definition
  - short_name: test-slo-converter-sli-2
    # name: test slo converter sli 2
    # message: Here's another message for a second monitor. @slack-temp-slo-monitoring
    service: devops-portal
    # Available values are latency, error-rate, and external
    # error-rate types accept a metric block
    type: error-rate
    # Enter/Update resources with those generated using service-resource-generator.sh described in README
    resources:
    - post_/api/v1/deployments
    - get_/api/v1/deployments
    - post_/api/v1/deployments/
    # Error rate formula is #errors / #total_requests, i.e. 1 / 100 = 0.01% error rate
    metric:
      # Source traces that contains the resources below; supported metrics include
      # trace.http.request.*, trace.servlet.request.*, trace.grpc.server.*,
      # and trace.kafka.consume.*
      numerator: trace.http.request.errors
      denominator: trace.http.request.hits
    threshold:
      # warning: 0.8
      critical: 1
    # tags:
    # - tag1
    # - tag2

# First standalone sli definition - slis may exist outside the control of an slo.
# This would be useful for situations where a service contains a listing of slis and there's
# another service that will pull them into an slo as an external reference, like above.
slis:
- short_name: test-standalone-slo-converter-sli-1 # terraform resource name; must be unique within this file
  name: test standalone slo converter sli 1 # SLI name in Datadog becomes "<service> <name> <type>"; or "<service> <type>" when omitted
  # Update slack message content with additional details or insight to sli.
  # This becomes the monitor message body.
  message: |-
    This is the message text for this monitor.

    This is an example of line two with a blank line between.
    @slack-temp-slo-monitoring
  # Service name will become a tag on all SLIs and SLOs, but this field should match
  # the FigureDeployment service name so SLOs are included on the Datadog APM page.
  service: devops-portal
  # Available values are latency, error-rate, and external
  # latency types accept a percentile and metric field
  type: latency
  # Default percentile is p95 when omitted; however other options are p50, p75, p90, p95, p99
  # percentile: p95
  # Source trace that contains the resources below; supported metrics include
  # trace.http.request.*, trace.servlet.request.*, trace.grpc.server.*, and trace.kafka.consume.*
  metric: trace.http.request
  # Enter/Update resources with those generated using service-resource-generator.sh described
  # in README
  resources:
  - "!post_/api/v1/deployments"
  - "!get_/api/v1/deployments"
  - post_/api/v1/deployments/
  # Datadog monitor trigger conditions, in seconds
  threshold:
    # warning: 0.8
    critical: 1
  # Include any tags that would be valuable for this SLI
  tags:
  - standalonetag1
  - standalonetag2
---
business_unit: devops # should match business_unit in services deployment.yaml
# Include any tags that would be valuable; tags here will be appended to all SLOs and SLIs below
tags:
- tag1
- tag2

# Second slo definition
slos:
- short_name: test-slo-converter-slo-2
  name: test slo converter slo 2
  threshold:
    target: 99.9
  # Third sli definition
  slis:
  - short_name: test-slo-converter-sli-3
    name: test slo converter sli 3
    message: |-
      This is the message text for this monitor.

      This is an example of line two with a blank line between.
      @slack-temp-slo-monitoring
    service: devops-portal
    type: latency
    #percentile: p95
    resources:
    - post_/api/v1/deployments
    - get_/api/v1/deployments
    - post_/api/v1/deployments/
    metric: trace.http.request
    threshold:
      # warning: 0.8
      critical: 1

  # Fourth sli definition
  - short_name: test-slo-converter-sli-4
    name: test slo converter sli 4
    message: Here's another message for a second monitor. @slack-temp-slo-monitoring
    service: devops-portal
    type: error-rate
    resources:
    - post_/api/v1/deployments
    - get_/api/v1/deployments
    - post_/api/v1/deployments/
    metric:
      numerator: trace.http.request.errors
      denominator: trace.http.request.hits
    threshold:
      # warning: 0.8
      critical: 1