rabbitmq/meta/heka.yml

log_collector:
  decoder:
    rabbitmq_log:
      engine: sandbox
      module_file: /usr/share/lma_collector/decoders/rabbitmq.lua
      module_dir: /usr/share/lma_collector/common;/usr/share/heka/lua_modules
      adjust_timezone: true
  input:
    rabbitmq_log:
      engine: logstreamer
      log_directory: "/var/log/rabbitmq"
      file_match: 'rabbit@(?P<Node>.+)\.log$'
      differentiator: ["rabbitmq.", "Node"]
      decoder: "rabbitmq_log_decoder"
      splitter: "rabbitmq_log_splitter"
  splitter:
    rabbitmq_log:
      engine: regex
      delimiter: '\n\n(=[^=]+====)'
      delimiter_eol: false
metric_collector:
  trigger:
    rabbitmq_disk_limit_critical:
      description: 'RabbitMQ has reached the free disk threshold. All producers are blocked.'
      severity: 'critical'
      no_data_policy: 'okay'
      rules:
      - metric: rabbitmq_remaining_disk
        relational_operator: '<='
        threshold: 0
        window: 20
        periods: 0
        function: min
    rabbitmq_disk_limit_warning:
      description: 'RabbitMQ is getting close to the free disk threshold.'
      severity: 'warning'
      no_data_policy: 'okay'
      rules:
      - metric: rabbitmq_remaining_disk
        relational_operator: '<='
        threshold: 104857600 # 100MB
        window: 20
        periods: 0
        function: min
    rabbitmq_memory_limit_critical:
      description: 'RabbitMQ has reached the memory threshold. All producers are blocked.'
      severity: 'critical'
      no_data_policy: 'okay'
      rules:
      - metric: rabbitmq_remaining_memory
        relational_operator: '<='
        threshold: 0
        window: 20
        periods: 0
        function: min
    rabbitmq_memory_limit_warning:
      description: 'RabbitMQ is getting close to the memory threshold.'
      severity: warning
      no_data_policy: 'okay'
      rules:
      - metric: rabbitmq_remaining_memory
        relational_operator: '<='
        threshold: 104857600 # 100MB
        window: 20
        periods: 0
        function: min
    rabbitmq_queue_warning:
      description: 'The number of outstanding messages is too high.'
      severity: warning
      no_data_policy: 'okay'
      rules:
      - metric: rabbitmq_messages
        relational_operator: '>='
        threshold: 200
        window: 120
        periods: 0
        function: avg
    rabbitmq_check:
      description: 'RabbitMQ cannot be checked'
      severity: down
      rules:
      - metric: rabbitmq_check
        relational_operator: '=='
        threshold: 0
        window: 60
        periods: 0
        function: last
  alarm:
    rabbitmq_server_disk:
      alerting: enabled
      triggers:
      - rabbitmq_disk_limit_critical
      - rabbitmq_disk_limit_warning
      dimension:
        service: rabbitmq-cluster
    rabbitmq_server_memory:
      alerting: enabled
      triggers:
      - rabbitmq_memory_limit_critical
      - rabbitmq_memory_limit_warning
      dimension:
        service: rabbitmq-cluster
    rabbitmq_server_queue:
      alerting: enabled
      triggers:
      - rabbitmq_queue_warning
      dimension:
        service: rabbitmq-cluster
    rabbitmq_check:
      alerting: enabled
      triggers:
      - rabbitmq_check
aggregator:
  alarm_cluster:
    rabbitmq_cluster:
      alerting: enabled
      policy: highest_severity
      # A 'hostname' group_by is required because an alarm on a single node has
      # an impact on the whole cluster.
      group_by: hostname
      match:
        service: rabbitmq-cluster
      members:
      - rabbitmq_server_disk
      - rabbitmq_server_memory
      - rabbitmq_server_queue
      dimension:
        service: rabbitmq
        nagios_host: 01-service-clusters
    rabbitmq_service:
      # A check failure on a single node doesn't mean that the whole cluster
      # is down, this is why a 'hostname' group_by and 'availability_of_members'
      # policy are used here
      policy: availability_of_members
      alerting: enabled
      group_by: hostname
      match:
        member: rabbitmq_check
      members:
      - rabbitmq_check
      dimension:
        service: rabbitmq
        nagios_host: 01-service-clusters
    rabbitmq:
      policy: highest_severity
      alerting: enabled_with_notification
      match:
        service: rabbitmq
      members:
      - rabbitmq_cluster
      - rabbitmq_service
      dimension:
        cluster_name: rabbitmq
        nagios_host: 00-top-clusters