Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Embracing Failure - Fault Injection and Service Resilience at Netflix


Published on

A presentation given at AWS re:Invent on how Netflix induces failure to validate and harden production systems. Technologies discussed include the Simian Army (Chaos Monkey, Gorilla, Kong) and our next gen Failure Injection Test framework (FIT).

Published in: Technology

Embracing Failure - Fault Injection and Service Resilience at Netflix

  1. 1. Embracing Failure Fault Injection and Service Resilience at Netflix Josh Evans – Director of Operations Engineering Naresh Gopalani – Software Engineer and Architect © 2014, Inc. and its affiliates. All rights reserved. May not be copied, modified, or distributed in whole or in part without the express consent of, Inc.
  2. 2. Netflix Ecosystem • ~50 million members, ~50 countries • > 1 billion hours per month • > 1000 device types • 3 AWS Regions, hundreds of services • Hundreds of thousands of requests/second • CDN serves petabytes of data at terabits/second Static Content Akamai Netflix CDN Service Partners AWS/Netfli x Control Plane Internet
  3. 3. Availability means that members can ● sign up ● activate a device ● browse ● watch
  4. 4. What keeps us up at night
  5. 5. Failures can happen any time • Disks fail • Power outages • Natural disasters • Software bugs • Human error
  6. 6. We design for failure • Exception handling • Fault tolerance and isolation • Fall-backs and degraded experiences • Auto-scaling clusters • Redundancy
  7. 7. Testing for failure is hard • Web-scale traffic • Massive, changing data sets • Complex interactions and request patterns • Asynchronous, concurrent requests • Complete and partial failure modes Constant innovation and change
  8. 8. What if we regularly inject failures into our systems under controlled circumstances?
  9. 9. Blast Radius • Unit of isolation • Scope of an outage • Scope a chaos exercise Instance Zone Region Global
  10. 10. An Instance Fails Edge Cluster Cluster A Cluster B Cluster C Cluster D
  11. 11. Chaos Monkey • Monkey loose in your DC • Run during business hours • What we learned – Auto-replacement works – State is problematic
  12. 12. A State of Xen - Chaos Monkey & Cassandra Out of our 2700+ Cassandra nodes • 218 rebooted • 22 did not reboot successfully • Automation replaced failed nodes • 0 downtime due to reboot
  13. 13. An Availability Zone Fails EU-West US-West US-East AZ1 AZ2
  14. 14. Chaos Gorilla Simulate an availability zone outage • 3-zone configuration • Eliminate one zone • Ensure that others can handle the load and nothing breaks Chaos Gorilla
  15. 15. Challenges • Rapidly shifting traffic – LBs must expire connections quickly – Lingering connections to caches must be addressed • Service configuration – Not all clusters auto-scaled or pinned – Services not configured for cross-zone calls – Mismatched timeouts – fallbacks prevented fail-over
  16. 16. A Region Fails US-West US-East EU-West
  17. 17. Regional Load Balancers Zuul – Traffic Shaping/Routing AZ1 AZ2 AZ3 Data Data Data Geo-located Chaos Kong Chaos Kong Regional Load Balancers Zuul – Traffic Shaping/Routing AZ1 AZ2 AZ3 Data Data Data Customer Device
  18. 18. Challenges ● Rapidly shifting traffic ○ Auto-scaling configuration ○ Static configuration/pinning ○ Instance start time ○ Cache fill time
  19. 19. Challenges ● Service Configuration ○ Timeout configurations ○ Fallbacks fail or don’t provide the desired experience ● No minimal (critical) stack ○ Any service may be critical!
  20. 20. A Service Fails Service Zone Region Global
  21. 21. Services Slow Down and Fail Simulate latent/failed service calls • Inject arbitrary latency and errors at the service level • Observe for effects Latency Monkey
  22. 22. Latency Monkey Device ELB Zuul Edge Service B Service C Internet Service A
  23. 23. Challenges • Startup resiliency is an issue • Services owners don’t know all dependencies • Fallbacks can fail too • Second order effects not easily tested • Dependencies are in constant flux • Latency Monkey tests function and scale – Not a staged approach – Lots of opt-outs
  24. 24. More Precise and Continuous
  25. 25. Service Failure Testing:FIT
  26. 26. Distributed Systems Fail ● Complex interactions at scale ● Variability across services ● Byzantine failures ● Combinatorial complexity
  27. 27. Any service can cause cascading failures ELB
  28. 28. Fault Injection Testing (FIT) Device Service B Service C Internet Zuul Edge Device or Account Override Service A Request-level simulations ELB
  29. 29. Failure Injection Points IPC Cassandra Client Memcached Client Service Container Fault Tolerance
  30. 30. FIT Details ● Common Simulation Syntax ● Single Simulation Interface ● Transported via Http Request header
  31. 31. Integrating Failure request [sendRequestHeader] >>fit.failure: 1|fit.Serializer| 2|[[{"name”:”failSocial, Filter Service Ribbon Filter Service Ribbon ServerRcv ClientSend ServerRcv Service A response Service B ”whitelist":false, "injectionPoints”: [“SocialService”]},{} ]], {"Id": "252c403b-7e34-4c0b-a28a-3606fcc38768"}]]
  32. 32. Failure Scenarios ● Set of injection points to fail ● Defined based on ○ Past outages ○ Specific dependency interactions ○ Whitelist of a set of critical services ○ Dynamic tracing of dependencies
  33. 33. FIT Insights : Salp ● Distributed tracing inspired by Dapper paper ● Provides insight into dependencies ● Helps define & visualize scenarios
  34. 34. Dialing Up Failure Functional Validation ● Isolated synthetic transactions ○ Set of devices Validation at Scale ● Dial up customer traffic - % based ● Simulation of full service failure Chaos!
  35. 35. Continuous Validation Critical Services Non-critical Services Synthetic Transactions
  36. 36. Don’t Fear The Monkeys
  37. 37. Take-aways • Don’t wait for random failures – Cause failure to validate resiliency – Remove uncertainty by forcing failures regularly – Better to fail at 2pm than 2am • Test design assumptions by stressing them Embrace Failure
  38. 38. The Simian Army is part of the Netflix open source cloud platform
  39. 39. Netflix talks at re:Invent Talk Time Title BDT-403 Wednesday, 2:15pm Next Generation Big Data Platform at Netflix PFC-306 Wednesday, 3:30pm Performance Tuning EC2 DEV-309 Wednesday, 3:30pm From Asgard to Zuul, How Netflix’s proven Open Source Tools can accelerate and scale your services ARC-317 Wednesday, 4:30pm Maintaining a Resilient Front-Door at Massive Scale PFC-304 Wednesday, 4:30pm Effective Inter-process Communications in the Cloud: The Pros and Cons of Micro Services Architectures ENT-209 Wednesday, 4:30pm Cloud Migration, Dev-Ops and Distributed Systems APP-310 Friday 9:00am Scheduling using Apache Mesos in the Cloud
  40. 40. Please give us your feedback on this presentation Josh Evans @josh_evans_nflx Naresh Gopalani Join the conversation on Twitter with #reinvent © 2014, Inc. and its affiliates. All rights reserved. May not be copied, modified, or distributed in whole or in part without the express consent of, Inc.