Infrastructure & System Monitoring using Prometheus

1. Infrastructure & System Monitoring using Prometheus Marco Pas Philips Lighting Software geek, hands on Developer/Architect/DevOps Engineer @marcopas
2. Some stuff about me... • Mostly doing cloud related stuff ◦ Java, Groovy, Scala, Spring Boot, IOT, AWS, Terraform, Infrastructure • Enjoying the good things • Chef leuke dingen doen == “trying out cool and new stuff” • Currently involved in a big IOT project • Wannabe chef, movie & Netflix addict
3. Agenda • Monitoring ◦ Introducing you to a Scary Movie • Prometheus overview (demo’s) ◦ Running Prometheus ◦ Gathering host metrics ◦ Introducing Grafana ◦ Monitoring Docker containers ◦ Alerting ◦ Instrumenting your own code ◦ Service Discovery (Consul) integration
4. ..Quick Inventory..
5. I am going to introduce you to some bad movies
11. Commonality between these movies?
12. Monitoring
14. Our scary movie “The Happy Developer” • Lets push out features • I can demo so it works :) • It works with 1 user, so it will work with multiple • Don’t worry about performance we will just scale using multiple machines/processes • Logging is into place
15. Did anyone notice? Disaster Strikes
16. Logging “recording to diagnose a system” Monitoring “observation, checking and recording” http_requests_total{method="post",code="200"} 1027 1395066363000 127.0.0.1 - frank [10/Oct/2000:13:55:36 -0700] "GET /apache_pb.gif HTTP/1.0" 200 2326 Logging != Monitoring
17. Vital Signs
18. Why Monitoring? • Know when things go wrong ◦ Detection & Alerting • Be able to debug and gain insight • Detect changes over time and drive technical/business decisions • Feed into other systems/processes (e.g. security, automation)
19. What to monitor? IT Network Operating System Services Applications Capture Monitoring Information Functional Monitoring Operational Monitoring metric data
20. Houston we have Storage problem! Storage metric data metric data metric data metric data metric data metric data metric data metric data metric data How to store the mass amount of metrics and also making them easy to query?
21. Time Series - Database • Time series data is a sequence of data points collected at regular intervals over a period of time. (metrics) ◦ Examples: ▪ Device data ▪ Weather data ▪ Stock prices ▪ Tide measurements ▪ Solar flare tracking • The data requires aggregation and analysis Time Series Database metric data • High write performance • Data compaction • Fast, easy range queries
22. metric name and a set of key-value pairs, also known as labels <metric name>{<label name>=<label value>, ...} value [ timestamp ] http_requests_total{method="post",code="200"} 1027 1395066363000 Time Series - Data format
23. Source: http://db-engines.com/en/ranking/time+series+dbms http://db-engines.com/en/ranking/time+series+dbms
24. Prometheus Overview
25. Prometheus Prometheus is an open-source systems monitoring and alerting toolkit originally built at SoundCloud. It is now a standalone open source project and maintained independently of any company. https://prometheus.io Implemented using
26. Prometheus Components • The main Prometheus server which scrapes and stores time series data • Client libraries for instrumenting application code • A push gateway for supporting short-lived jobs • Special-purpose exporters (for HAProxy, StatsD, Graphite, etc.) • An alertmanager • Various support tools • WhiteBox Monitoring instead of probing [aka BlackBox Monitoring]
27. Prometheus Overview
28. List of Job Exporters • Prometheus managed: ◦ JMX ◦ Node ◦ Graphite ◦ Blackbox ◦ SNMP ◦ HAProxy ◦ Consul ◦ Memcached ◦ AWS Cloudwatch ◦ InfluxDB ◦ StatsD ◦ ... • Custom ones: ◦ Database ◦ Hardware related ◦ Messaging systems ◦ Storage ◦ HTTP ◦ APIs ◦ Logging ◦ … https://prometheus.io/docs/instrumenting/exporters/
29. Demo Structure
30. Demo: Run Prometheus (native)
31. # file: prometheus.yml global: scrape_interval: 15s # Set the scrape interval to every 15 seconds. Default is every 1 minute. # some settings intentionally removed!! # A scrape configuration containing exactly one endpoint to scrape: # Here it's Prometheus itself. scrape_configs: # The job name is added as a label `job=<job_name>` to any timeseries scraped from this config. - job_name: 'prometheus' static_configs: - targets: ['localhost:9090']
32. Code Demo “Running Prometheus Native”
33. Demo: Run Prometheus using Docker
34. 34 # file: docker-compose.yml version: '2' services: prometheus: image: prom/prometheus:latest → Using official prometheus container volumes: - $PWD:/etc/prometheus → Mount local directory used for config + data ports: - "9090:9090" → Port mapping used for this container host:container command: - "-config.file=/etc/prometheus/prometheus.yml" → Prometheus configuration
35. Code Demo “Running Prometheus Dockerized”
36. Demo: Add host metrics
37. # file: docker-compose.yml version: '2' services: prometheus: → Runnning prometheus as Docker container image: prom/prometheus:latest → Using official prometheus container volumes: - $PWD:/etc/prometheus → Mount local directory used for config + data ports: - "9090:9090" → Port mapping used for this container host:container command: - "-config.file=/etc/prometheus/prometheus.yml" → Prometheus configuration node-exporter: image: prom/node-exporter:latest → Using node exporter as an additional container ports: - '9100:9100' → Port mapping used for this container host:container
38. 38 # file: prometheus.yml global: scrape_interval: 15s # Set the scrape interval to every 15 seconds. Default is every 1 minute. # some settings intentionally removed!! # A scrape configuration containing exactly one endpoint to scrape: # Here it's Prometheus itself. scrape_configs: # The job name is added as a label `job=<job_name>` to any timeseries scraped from this config. - job_name: 'prometheus' static_configs: - targets: ['localhost:9090'] - job_name: 'node-exporter' static_configs: - targets: ['node-exporter:9100']
39. Code Demo “Add host metrics”
40. Demo: Grafana 40
41. # file: docker-compose.yml version: '2' services: # some code intentionally removed!! grafana: image: grafana/grafana:latest → Using official prometheus container ports: - "3000:3000" → Port mapping used for this container host:container You get the idea :)
42. Code Demo “Grafana”
43. Demo: Monitor Docker containers
44. Code Demo “cAdvisor”
45. Demo: Alerting
46. Alerting Configuration • Alert Rules ◦ What are the settings where we need to alert upon? • Alert Manager ◦ Where do we need to send the alert to?
47. # file: alert.rules ALERT serviceDownAlert IF absent(((time() - container_last_seen{name="<service_name>"}) < 5)) FOR 5s LABELS { severity = "critical", → setting the labels so we can use them in the AlertManager service = "backend" } ANNOTATIONS { → information used in the alert event SUMMARY = "Container Instance down", DESCRIPTION = "Container Instance is down for more than 15 sec." }
48. # file: alert-manager.yml global: → Global settings smtp_smarthost: 'mailslurper:2500' smtp_from: 'alertmanager@example.org' smtp_require_tls: false route: → Routing receiver: mail # Fallback → Fallback is there is no match routes: - match: severity: critical → Match on label! continue: true → Continue with other receivers if there is a match receiver: mail → Determine the receiver - match: severity: critical receiver: slack
49. # file: alert-manager.yml (continued) receivers: - name: mail → mail receiver email_configs: - to: 'team-X+alerts@example.org' - name: slack → slack receiver slack_configs: - send_resolved: true username: 'AlertManager' channel: '#alert' api_url: 'THIS IS A VERY SECRET URL :)’
50. # file: prometheus.yml global: scrape_interval: 15s # Set the scrape interval to every 15 seconds. Default is every 1 minute. # Load rules once and periodically evaluate them according to the global 'evaluation_interval'. rule_files: - "alert.rules" # some settings intentionally removed!!
51. Code Demo “Alerting -> The Alert Manager”
52. Instrumenting your own code! • Counter ◦ A cumulative metric that represents a single numerical value that only ever goes up • Gauge ◦ Single numerical value that can arbitrarily go up and down • Histogram ◦ Samples observations (usually things like request durations or response sizes) and counts them in configurable buckets. It also provides a sum of all observed values • Summary ◦ Histogram + total count of observations + sum of all observed values, it calculates configurable quantiles over a sliding time window
53. Available Languages • Official ◦ Go, Java or Scala, Python, Ruby • Unofficial ◦ Bash, C++, Common Lisp, Elixir, Erlang, Haskell, Lua for Nginx, Lua for Tarantool, .NET / C#, Node.js, PHP, Rust // Spring Boot example -> file: build.gradle dependencies { compile('org.springframework.boot:spring-boot-starter-web') testCompile('org.springframework.boot:spring-boot-starter-test') compile('io.prometheus:simpleclient_spring_boot:0.0.21') → Add dependency }
54. Prometheus Client Libaries: SpringBoot Example @EnablePrometheusEndpoint @EnableSpringBootMetricsCollector @RestController @SpringBootApplication public class DemoApplication { public static void main(String[] args) { SpringApplication.run(DemoApplication.class, args); } static final Counter requests = Counter.build() → create metric type counter .name("helloworld_requests_total") → set metric name .help("HelloWorld Total requests.").register(); → register the metric @RequestMapping("/helloworld") String home() { requests.inc(); → increment the counter with 1 (helloworld_requests_total) return "Hello World!"; } }
55. Demo: Application metrics
56. Code Demo “Application metrics”
57. Service Discovery (Consul) Integration
58. Demo: Consul Integration
59. Service Discovery
60. Demo: Consul integration Register the services with Consul and Monitor 1 2
61. Code Demo “Consul to the rescue”
63. That’s a wrap! Question? https://github.com/mpas/infrastructure-and-system-monitoring-using-prometheus Marco Pas Philips Lighting Software geek, hands on Developer/Architect/DevOps Engineer @marcopas