marbot-redshift.yml

---
# Copyright widdix GmbH
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
AWSTemplateFormatVersion: '2010-09-09'
Description: 'marbot.io: Redshift cluster monitoring (https://github.com/marbot-io/monitoring-jump-start)'
Metadata:
  'AWS::CloudFormation::Interface':
    ParameterGroups:
    - Label:
        default: 'marbot endpoint'
      Parameters:
      - EndpointId
      - Stage
    - Label:
        default: 'Redshift'
      Parameters:
      - ClusterIdentifier
      - NodeType
    - Label:
        default: 'Thresholds'
      Parameters:
      - CPUUtilizationThreshold
      - DiskSpaceThreshold
      - ConcurrencyScalingThreshold
Parameters:
  EndpointId:
    Description: 'Your marbot endpoint ID (to get this value: select a channel where marbot belongs to and send a message like this: "@marbot show me my endpoint id").'
    Type: String
  Stage:
    Description: 'marbot stage (never change this!).'
    Type: String
    Default: v1
    AllowedValues: [v1, dev]
  ClusterIdentifier:
    Description: 'The cluster ID of the Redshift cluster that you want to monitor.'
    Type: String
  NodeType:
    Description: 'The current Redshift node type.'
    Type: String
    AllowedValues:
    - 'dc2.large'
    - 'dc2.8xlarge'
    - 'ds2.xlarge'
    - 'ds2.8xlarge'
    - 'ra3.4xlarge'
    - 'ra3.16xlarge'
    - 'dc1.large'
    - 'dc1.8xlarge'
  CPUUtilizationThreshold:
    Description: 'The maximum percentage of CPU utilization (set to -1 to disable).'
    Type: Number
    Default: 80
    MinValue: -1
    MaxValue: 100
  DiskSpaceThreshold:
    Description: 'The maximum percentage of used disk space.'
    Type: Number
    Default: 90
    MinValue: 1
    MaxValue: 100
  ConcurrencyScalingThreshold:
    Description: 'The maximum number of concurrency scaling seconds per 24 hours (set -1 to disable).'
    Type: Number
    Default: 3600
    MinValue: -1
Mappings:
  NodeTypes:
    'dc2.large':
      IOCapacity: 0.54 # original value - 10%
    'dc2.8xlarge':
      IOCapacity: 6.75
    'ds2.xlarge':
      IOCapacity: 0.36
    'ds2.8xlarge':
      IOCapacity: 2.97
    'ra3.4xlarge':
      IOCapacity: 1.8
    'ra3.16xlarge':
      IOCapacity: 7.2
    'dc1.large':
      IOCapacity: 0.18
    'dc1.8xlarge':
      IOCapacity: 3.33
Conditions:
  HasCPUUtilizationThreshold: !Not [!Equals [!Ref CPUUtilizationThreshold, '-1']]
  HasConcurrencyScalingThreshold: !Not [!Equals [!Ref ConcurrencyScalingThreshold, '-1']]
Resources:
  ##########################################################################
  #                                                                        #
  #                                 TOPIC                                  #
  #                                                                        #
  ##########################################################################
  Topic:
    Type: 'AWS::SNS::Topic'
    Properties: {}
  TopicPolicy:
    Type: 'AWS::SNS::TopicPolicy'
    Properties:
      PolicyDocument:
        Id: Id1
        Version: '2012-10-17'
        Statement:
        - Sid: Sid1
          Effect: Allow
          Principal:
            AWS: '*' # Allow CloudWatch Alarms
          Action: 'sns:Publish'
          Resource: !Ref Topic
          Condition:
            StringEquals:
              'AWS:SourceOwner': !Ref 'AWS::AccountId'
      Topics:
      - !Ref Topic
  TopicEndpointSubscription:
    DependsOn: TopicPolicy
    Type: 'AWS::SNS::Subscription'
    Properties:
      DeliveryPolicy:
        healthyRetryPolicy:
          minDelayTarget: 1
          maxDelayTarget: 60
          numRetries: 100
          numNoDelayRetries: 0
          backoffFunction: exponential
        throttlePolicy:
          maxReceivesPerSecond: 1
      Endpoint: !Sub 'https://api.marbot.io/${Stage}/endpoint/${EndpointId}'
      Protocol: https
      TopicArn: !Ref Topic
  MonitoringJumpStartEvent:
    DependsOn: TopicEndpointSubscription
    Type: 'AWS::Events::Rule'
    Properties:
      Description: 'Monitoring Jump Start connection. (created by marbot)'
      ScheduleExpression: 'rate(30 days)'
      State: ENABLED
      Targets:
      - Arn: !Ref Topic
        Id: marbot
        Input: !Sub |
          {
            "Type": "monitoring-jump-start-connection",
            "StackTemplate": "marbot-redshift",
            "StackVersion": "1.1.1",
            "Partition": "${AWS::Partition}",
            "AccountId": "${AWS::AccountId}",
            "Region": "${AWS::Region}",
            "StackId": "${AWS::StackId}",
            "StackName": "${AWS::StackName}"
          }
  ##########################################################################
  #                                                                        #
  #                                 ALARMS                                 #
  #                                                                        #
  ##########################################################################
  HealthStatusAlarm:
    DependsOn: TopicEndpointSubscription
    Type: 'AWS::CloudWatch::Alarm'
    Properties:
      AlarmDescription: 'Redshift cluster is unhealty. (created by marbot)'
      Namespace: 'AWS/Redshift'
      MetricName: HealthStatus
      Dimensions:
      - Name: ClusterIdentifier
        Value: !Ref ClusterIdentifier
      Threshold: 1
      ComparisonOperator: LessThanThreshold
      Statistic: Minimum
      Period: 60
      EvaluationPeriods: 1
      AlarmActions:
      - !Ref Topic
      OKActions:
      - !Ref Topic
      TreatMissingData: notBreaching
  MaintenanceModeAlarm:
    DependsOn: TopicEndpointSubscription
    Type: 'AWS::CloudWatch::Alarm'
    Properties:
      AlarmDescription: 'Redshift cluster in maintenance mode. (created by marbot)'
      Namespace: 'AWS/Redshift'
      MetricName: MaintenanceMode
      Dimensions:
      - Name: ClusterIdentifier
        Value: !Ref ClusterIdentifier
      Threshold: 1
      ComparisonOperator: GreaterThanThreshold
      Statistic: Maximum
      Period: 60
      EvaluationPeriods: 1
      AlarmActions:
      - !Ref Topic
      OKActions:
      - !Ref Topic
      TreatMissingData: notBreaching
  DiskSpaceAlarm:
    DependsOn: TopicEndpointSubscription
    Type: 'AWS::CloudWatch::Alarm'
    Properties:
      AlarmDescription: 'Redshift cluster is running out of disk space. (created by marbot)'
      Namespace: 'AWS/Redshift'
      MetricName: PercentageDiskSpaceUsed
      Dimensions:
      - Name: ClusterIdentifier
        Value: !Ref ClusterIdentifier
      Threshold: !Ref DiskSpaceThreshold
      ComparisonOperator: GreaterThanThreshold
      Statistic: Average
      Period: 60
      EvaluationPeriods: 1
      AlarmActions:
      - !Ref Topic
      OKActions:
      - !Ref Topic
      TreatMissingData: notBreaching
  SchemaQuotasAlarm:
    DependsOn: TopicEndpointSubscription
    Type: 'AWS::CloudWatch::Alarm'
    Properties:
      AlarmDescription: 'More than one schema reached their storage quota. (created by marbot)'
      Namespace: 'AWS/Redshift'
      MetricName: NumExceededSchemaQuotas
      Dimensions:
      - Name: ClusterIdentifier
        Value: !Ref ClusterIdentifier
      Threshold: 0
      ComparisonOperator: GreaterThanThreshold
      Statistic: Maximum
      Period: 60
      EvaluationPeriods: 1
      AlarmActions:
      - !Ref Topic
      OKActions:
      - !Ref Topic
      TreatMissingData: notBreaching
  HighCPUUtilizationAlarm:
    Condition: HasCPUUtilizationThreshold
    DependsOn: TopicEndpointSubscription
    Type: 'AWS::CloudWatch::Alarm'
    Properties:
      AlarmDescription: 'Redshift cluster experiences high CPU load for more than 15 minutes. (created by marbot)'
      Namespace: 'AWS/Redshift'
      MetricName: CPUUtilization
      Dimensions:
      - Name: ClusterIdentifier
        Value: !Ref ClusterIdentifier
      Threshold: !Ref CPUUtilizationThreshold
      ComparisonOperator: GreaterThanThreshold
      Statistic: Average
      Period: 900
      EvaluationPeriods: 1
      AlarmActions:
      - !Ref Topic
      OKActions:
      - !Ref Topic
      TreatMissingData: notBreaching
  HighIOUtilizationAlarm:
    DependsOn: TopicEndpointSubscription
    Type: 'AWS::CloudWatch::Alarm'
    Properties:
      AlarmDescription: 'Redshift cluster experiences high IO load for more than 15 minutes. (created by marbot)'
      Metrics:
      - Id: read
        Label: ReadThroughput
        MetricStat:
          Metric:
            Namespace: 'AWS/Redshift'
            MetricName: ReadThroughput
            Dimensions:
            - Name: ClusterIdentifier
              Value: !Ref ClusterIdentifier
          Period: 300
          Stat: Average
          # Unit: Bytes
        ReturnData: false
      - Id: write
        Label: WriteThroughput
        MetricStat:
          Metric:
            Namespace: 'AWS/Redshift'
            MetricName: WriteThroughput
            Dimensions:
            - Name: ClusterIdentifier
              Value: !Ref ClusterIdentifier
          Period: 300
          Stat: Average
          # Unit: Bytes
        ReturnData: false
      - Id: total
        Label: 'TotalThroughput'
        Expression: '(read+write)/1000/1000/1000' # GB/s
        ReturnData: true
      Threshold: !FindInMap [NodeTypes, !Ref NodeType, IOCapacity]
      ComparisonOperator: GreaterThanThreshold
      EvaluationPeriods: 3
      AlarmActions:
      - !Ref Topic
      OKActions:
      - !Ref Topic
      TreatMissingData: notBreaching
  ConcurrencyScalingAlarm:
    Condition: HasConcurrencyScalingThreshold
    DependsOn: TopicEndpointSubscription
    Type: 'AWS::CloudWatch::Alarm'
    Properties:
      AlarmDescription: 'Redshift cluster causes additional costs due to concurrency scaling. (created by marbot)'
      Namespace: 'AWS/Redshift'
      MetricName: ConcurrencyScalingSeconds
      Dimensions:
      - Name: ClusterIdentifier
        Value: !Ref ClusterIdentifier
      Threshold: !Ref ConcurrencyScalingThreshold
      ComparisonOperator: GreaterThanThreshold
      Statistic: Sum
      Period: 86400
      EvaluationPeriods: 1
      AlarmActions:
      - !Ref Topic
      OKActions:
      - !Ref Topic
      TreatMissingData: notBreaching
Outputs:
  StackName:
    Description: 'Stack name.'
    Value: !Sub '${AWS::StackName}'
  StackTemplate:
    Description: 'Stack template.'
    Value: 'marbot-redshift'
  StackVersion:
    Description: 'Stack version.'
    Value: '1.1.1'