Docker Swarm Custom Metrics Example


#1

An example of an app that uses custom metrics with a swarm stack. The statser service is configured with the STATSD_xxx params, and sends custom metrics to the Replicated’s embedded statsd server.

# QA Application - Charlie - Swarm Scheduler
---
# kind: replicated

replicated_api_version: 2.8.0
name: "Compose Sample - AutoUpgrade"
version: "1.0.3"
release_notes: QA application meant for swarm scheduler installations.

###########################################################
# Host Requirements and Properties
###########################################################

host_requirements:
  docker_version: "17.03.1-ce"
  replicated_version: ">=2.9.0 <=2.11.0"
  cpu_cores: 1
  cpu_mhz: 1000
  memory: 2GB
  disk_space: 40GB
  
properties:
  app_url: http://{{repl ConfigOption "hostname" }}
  console_title: QA Application - Charlie UNstable Console
  logo_url: http://www.clipartbest.com/cliparts/aTq/ogz/aTqogze8c.jpg
  shell_alias: charliecli

###########################################################
# Swarm specific configurations
###########################################################

swarm:
  minimum_node_count: 1
  nodes:
  - role: manager
    labels:
      A: 
      B: C
    minimum_count: 1
    host_requirements:
      cpu_cores: 1
      cpu_mhz: 1000
      memory: 1GB
  - role: worker
    labels:
      D: 
      E: F
    minimum_count: 1
    host_requirements:
      cpu_cores: 1
      cpu_mhz: 1000
      memory: 1GB
  secrets:
  - name: secret_box
    value: '{{repl ConfigOption "secret_box" }}'
    labels:
      A: 
      B: C 
    

###########################################################
# Monitors and Custom Metrics
###########################################################

custom_metrics:
- target: stats.gauges.myapp100.disk.*.*.*
  retention: "1s:10m,1m:4h,1h:30d"
  aggregation_method: "average"
  xfiles_factor: 0.3
- target: stats.gauges.myapp100.ping.*
  retention: "1s:10m,1m:4h,1h:30d"
  aggregation_method: "average"
  xfiles_factor: 0.3
monitors:
  custom:
  - name: Disk Free (bytes) time 1 hour
    targets:
    - stats.gauges.myapp100.disk.*.free
    from: "-1hours"
    display:
      label_unit: B
      label_scale: metric
      label_count: 2
      fill_color: rgba(100, 0, 50, 0.5)
      stroke_color: "#ff1060"
  - name: Disk Free (%) time - default
    target: scale(divideSeries(stats.gauges.myapp100.disk.*.free,stats.gauges.myapp100.disk.*.total),100)
    display:
      label_unit: "%"
      label_scale: none
      label_count: 3
      label_min: 0
      label_max: 100
      label_range_override: true
  - name: Ping RTT time 30 minutes
    targets:
    - stats.gauges.myapp100.ping.*
    - movingAverage(stats.gauges.myapp100.ping.*,60)
    - movingAverage(stats.gauges.myapp100.ping.*,600)
    from: "-30minutes"
    display:
      label_unit: ms
      label_scale: none

###########################################################
# Admin Commands
###########################################################

admin_commands:
- alias: aliasenv
  command: [env]
  run_type: exec
  service: voting-app
- alias: redis-sadd
  command: [redis-cli, sadd]
  run_type: exec
  service: redis

###########################################################
# Config Settings
###########################################################

config:
- name: qaconfigs
  title: QA Test Configurations
  description: These are only meant to test configuration options.
  items:
  - name: secret_box
    title: Secret
    type: password
    help_text: What is typed in here will be a Docker secret in all app containers.
- name: hostname
  title: Hostname
  description: Ensure this domain name is routable on your network.
  items:
  - name: hostname
    title: Hostname
    value: '{{repl ConsoleSetting "tls.hostname"}}'
    type: text
    test_proc:
      display_name: Check DNS
      command: resolve_host

###########################################################
# Swarm specific section to allow airgap to build package
###########################################################

images:
- source: public
  name: redis
  tag: 3.2-alpine
- source: replicated
  name: statser
  tag: latest
- source: public
  name: postgres
  tag: 9.4
- source: replicated
  name: example-voting-app-vote
  tag: latest
- source: public
  name: gaiadocker/example-voting-app-result
  tag: latest
- source: public
  name: gaiadocker/example-voting-app-worker
  tag: latest

###########################################################
# Swarm services
###########################################################

---

# kind: scheduler-swarm
version: "3.1"

services:

  redis:
    image: redis:3.2-alpine
    ports:
      - "6379"
    networks:
      - voteapp
    deploy:
      mode: replicated
      placement:
        constraints: [node.role == manager]
      replicas: 1    
    secrets:
      - secret_box

  statser:
    image: registry.staging.replicated.com/charlie/statser:latest
    networks: 
      - voteapp
    deploy:
      placement:
        constraints: [node.role == worker]
    environment:
      REPLICATED_INTEGRATIONAPI: {{repl PremkitAPIAddress }}
      REPLICATED_STATSD_ADDRESS: {{repl StatsdAddress }}
    secrets:
      - secret_box  

  db:
    image: postgres:9.4
    volumes:
      - db-data:/var/lib/postgresql/data
    networks:
      - voteapp
    deploy:
      placement:
        constraints: [node.role == manager]

  voting-app:
    image: registry.staging.replicated.com/charlie/example-voting-app-vote:latest
    ports:
      - 5000:80
    networks:
      - voteapp
    depends_on:
      - redis
    deploy:
      mode: replicated
      replicas: 2
      labels: [APP=VOTING]
      placement:
        constraints: [node.role == worker]
      restart_policy:
        condition: none
        
  result-app:
    image: gaiadocker/example-voting-app-result:latest
    ports:
      - 5001:80
    networks:
      - voteapp
    depends_on:
      - db

  worker:
    image: gaiadocker/example-voting-app-worker:latest
    networks:
      voteapp:
        aliases:
          - workers
    depends_on:
      - db
      - redis
    # service deployment
    deploy:
      mode: replicated
      replicas: 2
      labels: [APP=VOTING]
      # service resource management
      resources:
        # Hard limit - Docker does not allow to allocate more
        limits:
          cpus: '0.25'
          memory: 512M
        # Soft limit - Docker makes best effort to return to it
        reservations:
          cpus: '0.25'
          memory: 256M
      # service restart policy
      restart_policy:
        condition: on-failure
        delay: 5s
        max_attempts: 3
        window: 120s
      # service update configuration
      update_config:
        parallelism: 1
        delay: 10s
        failure_action: continue
        monitor: 60s
        max_failure_ratio: 0.3
      # placement constraint - in this case on 'worker' nodes only
      placement:
        constraints: [node.role == worker]
        
secrets:
  secret_box:
    external: true        

networks:
    voteapp:
    
volumes:
  db-data: