Observability 101

The talk was given at the Bangalore Observability meet-up on March, 13th. Covers the basics of observability relevant for beginners, and intermediate level practitioners of Observability. Also, covers some of the recent trends on Data observability, and monitoring.

Transcript

1. Observability ------ Metrics, Monitoring, Alerting Piyush Goel, March 2021 @pigol1
2. Release to Production is just the beginning! “40% to 90% of the total costs of software are incurred after launch.” • Facts and Fallacies of Software Engineering, Glass R (2002), Addison-Wesley, p-115 • Which Factors affect Software Projects maintenance costs more? Acta Informatica Medica
3. Systems Will Fail!
4. #failure points in a distributed system increase with each new component.
5. What to monitor?
6. What to monitor? Let’s take an example: HTTP REST API’s
7. What to monitor? Let’s take an example: HTTP REST API’s • API Latency (95th percentile, Avg, 99th percentile) • CPU, Load Avg • Memory • Swap • JMX Heap Size (Assuming Java implementation) • HTTP Error Codes (200, 400, 300) • Exceptions • External API call Latencies • … • And more…….
8. What to monitor? Let’s take an example: HTTP REST API’s • API Latency (95th percentile, Avg, 99th percentile) • CPU, Load Avg • Memory • Swap • JMX Heap Size • HTTP Error Codes (200, 400, 300) • Exceptions • External API call Latencies • And more……. X Servers X Clusters
9. What to monitor? Let’s take an example: HTTP REST API’s • API Latency (95th percentile, Avg, 99th percentile) • CPU, Load Avg • Memory • Swap • JMX Heap Size • HTTP Error Codes (200, 400, 300) • Exceptions • External API call Latencies • And more……. X Servers X Clusters
10. Let’s step back and understand monitoring.
11. Monitoring • Capturing the state of the system to determine its health. ◦ HealthChecks ▪ Is the service running? ▪ Can I send work? ◦ Metrics ▪ System, Application, Functional • Alerts ◦ Anomalous behaviors - How do you define an anomaly?
12. Monitoring • Alerts - Known Failures ◦ Knowledge Based ◦ Reactive (post-outages) What about the unknown failures?
13. Observability
14. Observability The bygone era!!
15. Observability https://medium.com/@copyconstruct/monitoring-and-observability-8417d1952e1c
16. Observability • https://www.vividcortex.com/blog/monitoring-isnt-observability
17. Observability • https://theagileadmin.com/2018/02/16/monitoring-and-observability/ Observability is a measure of how well the internal states of a system can be inferred from knowledge of its external outputs. Analogous to medical diagnosis.
18. Observability - Internal States • Context Specific ◦ Web Servers ▪ Availability ▪ Incoming Request Rate ▪ Latency ▪ HTTP Failures ◦ Application Services ▪ Success Rate ▪ Functionalities ◦ Message Queue ▪ Queue Length ▪ Consumer/Producer Count Needs Instrumentation! While writing code
19. Observability - External Outputs (3 Pillars) • Metrics ◦ Health Checks • Logging • Tracing
20. Observability - Metrics • External State at a broad scope (Time dimension) ◦ System ◦ Application ▪ Success Rate/Failure Rate ▪ Latency (internal/external) ▪ Error Codes ▪ Exceptions ◦ Business/Functional ◦ Order Management System - Order Rate (Regular, Cancel, Return) ◦ Payments - Forward payments, reverse payments, recon requests ◦ Coupons - Issued, Redeemed ◦ Promotion Engine - Success Evaluations, Failures.
21. Observability - Metrics • Meaningful Metrics* - Generous • Alerts - Judicious • Low Cardinality ◦ Keep a Watch! ◦ Don’t emit for users/orders. We use Logs for that! • Provide System Summary • Questions: ▪ How many transactions failed? ▪ How many logins succeeded? * Metrics That Matter - https://queue.acm.org/detail.cfm?id=3309571 - Must Read
22. Observability - Metrics • Tools ◦ Prometheus ◦ InfluxDB ◦ TimescaleDB ◦ Graphite ◦ OpenTSDB ◦ Scuba (Facebook) ◦ Apache Druid ◦ Grafana
23. Observability - Health Checks • HealthChecks ◦ Is the service running? (Liveness) ◦ Can I send work? (Readiness) • Methods ◦ Broadcast - Gossip Protocols (Cassandra, Riak) ◦ Register - Service Discovery ◦ Health endpoints - ELB, HAProxy, Nginx
24. Observability - Logging • Understanding at a smaller scope ◦ Request, customer, transaction • Ask Questions ◦ Why couldn’t the customer place an order? ◦ Why did the transaction fail? • Centralised - Log Collection, Aggregation • Searchable - Indexing • Correlatable - Common Key (Request Id) • Tools - ELK, EFK, Splunk, SumoLogic, Loki
25. Observability - Logging (Anatomy)
26. Observability - Tracing • Dissect a request into sub-paths. (Spans) • Profile system usage at a span level. • Extract Insights Tools: • Google Dapper (https://ai.google/research/pubs/pub36356) • Twitter Zipkin (https://zipkin.io/) • Open Jaeger (https://www.jaegertracing.io/) • New Relic
27. Observability Spectrum
28. We have Metrics. What to alert on?
29. Service Level Objectives (SLO)
30. Service Level Objectives (SLO) * https://landing.google.com/sre/sre-book/chapters/service-level-objectives/ * https://www.youtube.com/watch?v=tEylFyxbDLE • Defines a Quantifiable Goal for a service. • Measure the goal - Represents the User Experience/Delight Factor. • First step before writing a new service. Work backwards • Have as few SLO’s as possible. ◦ Represents the system behaviour.
31. Service Level Objectives (SLO) - Exercise • Cart Service • Authentication & Authorization Service • Communication Engine (SMS, Email, Push Notifications)
32. Why Observe? • Stable systems in the long run. ▪ Better Capacity & Load Planning • Lower MTTR. • Promotes a data-driven culture within the team. • Makes outcomes measurable and removes subjectivity. • Brings accountability across teams - Tech & Product. ▪ Transparency ▪ Less friction.
33. Observability - Standards • OpenMetrics • OpenTelemetry ◦ OpenCensus & OpenTracing ◦ API’s, SDK’s - Metrics, Tracing, Context. ◦ Open Source, Vendor Neutral ◦ Avoid lock-in
34. Observability - Recent Trends • Data Observability ◦ Freshness ◦ Volume ◦ Schema ◦ Distribution ◦ Lineage • AccelData, MonteCarlo, Soda * https://www.montecarlodata.com/what-is-data-observability/ * https://cloudedjudgement.substack.com/p/data-observability-the-next-monitoring
35. Happy Observing!
36. References • Debugging Production Systems • Pierre Vincent - How to build observable Distributed systems? • Charity Majors - Observability for Emerging Infra: What Got You Here Won't Get You There" • Caitee McAfree - Of the Order of Billions: Building Observability at Twitter • https://eng.uber.com/observability-at-scale/ • OpenTelemetry • OpenMetrics • Monitoring and Observability
37. Q & A Reach out to me @pigol1