Creating Resiliency Through Destruction

With the wide adoption of micro-services and large-scale distributed systems, architectures have grown increasingly complex and hard to understand. Worse, the software systems running them have become extremely difficult to debug and test, increasing the risk of outages. With these new challenges, new tools are required and since failures have become more and more chaotic in nature, we must turn to chaos engineering in order to reveal failures before they become outages. In this talk, we will deep dive into chaos engineering, a discipline that promotes breaking things on purpose in order to learn how to build more robust systems.

Transcript

1. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
   S U M M I T
   Creating Resiliency Through Destruction
   Adrian Hornsby
   Sr. Technical Evangelist
   Amazon Web Services
   adhorn
   

   View Slide

2. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
   S U M M I T
   Can you guess what will happen?
   

   View Slide

3. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
   S U M M I T
   Distributed Systems are hard
   Amazon Twitter Netflix
   

   View Slide

4. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
   S U M M I T
   Failures are a given and
   everything will eventually fail
   over time.
   Werner Vogels
   CTO – Amazon.com
   “
   “
   

   View Slide

5. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
   S U M M I T
   Resiliency: Ability for a system to handle and
   eventually recover from unexpected conditions
   

   View Slide

6. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
   S U M M I T
   Resiliency at work
   

   View Slide

7. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
   S U M M I T
   How do we build resilient software
   systems?
   

   View Slide

8. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
   S U M M I T
   People
   Application
   Network & Data
   Infrastructure
   

   View Slide

9. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
   S U M M I T
   Building confidence through testing
   Unit testing of components:
   • Tested in isolation to ensure function meets expectations.
   Functional testing of integrations:
   • Each execution path tested to assure expected results.
   Is it enough???
   

   View Slide

10. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    GameDay at Amazon
    Creating Resiliency Through Destruction
    https://www.youtube.com/watch?v=zoz0ZjfrQ9s
    

    View Slide

11. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Chaos engineering
    https://github.com/Netflix/SimianArmy
    

    View Slide

12. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Failure injection
    • Start small & build confidence
    • Application level
    • Host failure
    • Resource attacks (CPU, memory, …)
    • Network attacks (dependencies, latency, …)
    • Region attack
    • Human attack
    

    View Slide

13. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Break your systems on purpose.
    Find out their weaknesses and fix
    them before they break when
    least expected.
    

    View Slide

14. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Chaos engineering is NOT about breaking
    things randomly without a purpose, chaos
    engineering is about breaking things in a
    controlled environment and through well-
    planned experiments in order to build
    confidence in your application to withstand
    turbulent conditions.
    

    View Slide

15. S U M M I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    Chaos Engineering
    

    View Slide

16. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Steady
    State
    Hypothesis
    Design & Run
    Experiment
    Fix
    Build Resilient
    Systems
    Verify & Learn
    

    View Slide

17. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Build Resilient
    Systems
    

    View Slide

18. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    

    View Slide

19. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Cascading and Overload Failures
    

    View Slide

20. Operations
    Infrastructure
    Application
    Software
    

    View Slide

21. Shameless plug - https://medium.com/@adhorn
    

    View Slide

22. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    https://aws.amazon.com/wellarchitected
    

    View Slide

23. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Steady
    State
    

    View Slide

24. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    What is steady state?
    • ”normal” behavior of your system
    https://www.elastic.co/blog/timelion-tutorial-from-zero-to-hero
    

    View Slide

25. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    What is steady state?
    • ”normal” behavior of your system
    • Business Metric
    https://medium.com/netflix-techblog/sps-the-pulse-of-netflix-streaming-ae4db0e05f8a
    

    View Slide

26. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Business metrics at work
    Amazon: 100 ms of extra load time caused a 1% drop in sales (Greg Linden).
    Google: 500 ms of extra load time caused 20% fewer searches (Marissa Mayer).
    Yahoo!: 400 ms of extra load time caused a 5–9% increase in the number
    of people who clicked “back” before the page even loaded (Nicole Sullivan).
    

    View Slide

27. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Hypothesis
    

    View Slide

28. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    What if…?
    “What if this load balancer breaks?”
    “What if Redis becomes slow?”
    “What if a host on Cassandra goes away?”
    ”What if latency increases by 300ms?”
    ”What if the database stops?”
    Make it everyone’s problem!
    

    View Slide

29. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Disclaimer!
    Don’t make an hypothesis that you know
    will break you!
    

    View Slide

30. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Design & Run
    Experiment
    

    View Slide

31. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Designing experiment
    • Pick hypothesis
    • Scope the experiment
    • Identify metrics
    • Notify the organization
    

    View Slide

32. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Rules of thumbs
    • Start with very small
    • As close as possible to production
    • Minimize the blast radius.
    • Have an emergency STOP!
    • Careful with state that can’t be rolled back
    (corrupt or incorrect data)
    

    View Slide

33. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Running Chaos Experiment
    Users
    Canary deployment
    Normal Version
    99%
    Users
    1%
    Users
    Start with ..
    

    View Slide

34. S U M M I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    

    View Slide

35. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Fault Injection Queries for Amazon Aurora
    SQL commands issued to simulate:
    • A crash of the master instance or an Aurora Replica
    • A failure of an Aurora Replica
    • A disk failure
    • Disk congestion
    https://docs.aws.amazon.com/AmazonRDS/latest/AuroraUserGuide/AuroraMySQL.Managing.FaultInjectionQueries.html
    

    View Slide

36. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Fault Injection Queries for Amazon Aurora
    SQL commands issued to simulate:
    • A crash of the master instance or an Aurora Replica
    • A failure of an Aurora Replica
    • A disk failure
    • Disk congestion
    ALTER SYSTEM SIMULATE percentage_of_failure PERCENT DISK FAILURE [ IN
    DISK index | NODE index ] FOR INTERVAL quantity { YEAR | QUARTER |
    MONTH | WEEK | DAY | HOUR | MINUTE | SECOND };
    

    View Slide

37. S U M M I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    How to DDoS yourself
    ~ wrk -t12 -c400 -d30s http://127.0.0.1/api/health
    

    View Slide

38. S U M M I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    Come late to work
    ~ tc qdisc add dev eth0 root netem delay 200ms
    

    View Slide

39. https://github.com/Netflix/SimianArmy
    Set of scheduled agent:
    • shuts down services randomly
    • slows down performances
    • checks conformity
    • breaks an entire region
    • More …
    

    View Slide

40. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    The Chaos Toolkit
    • Simplifying Adoption of Chaos Engineering
    • An Open API to Chaos Engineering
    • Open source extensions for
    • Infrastructure/Platform Fault Injections
    • Application Fault Injections
    • Observability
    • Integrates easily into CI/CD pipelines
    

    View Slide

41. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    

    View Slide

42. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    

    View Slide

43. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    

    View Slide

44. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    

    View Slide

45. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    ToxiProxy
    • HTTP API
    • Build for Automated testing in mind
    • Not for production environment
    • Fast
    • Toxics for:
    • Timeouts, latency, connections and bandwidth limitation, etc..
    • CLI
    • Stable and well tested (used for 3 years at Shopify)
    • Open Source: https://github.com/Shopify/toxiproxy
    

    View Slide

46. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    https://atscaleconference.com/videos/resiliency-testing-with-toxiproxy/
    

    View Slide

47. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    https://github.com/asobti/kube-monkey
    

    View Slide

48. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Pumba
    https://github.com/alexei-led/pumba/
    

    View Slide

49. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    https://blog.thundra.io/chaos-test-your-lambda-functions-with-thundra
    

    View Slide

50. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    https://medium.com/@adhorn/injecting-chaos-to-aws-lambda-functions-using-lambda-layers-2963f996e0ba
    Injecting Chaos to AWS Lambda functions using
    Lambda Layers
    

    View Slide

51. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Verify & Learn
    

    View Slide

52. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Quantifying the result of the experiment
    • Time to detect?
    • Time for notification? And escalation?
    • Time to public notification?
    • Time for graceful degradation to kick-in?
    • Time for self healing to happen?
    • Time to recovery – partial and full?
    • Time to all-clear and stable?
    

    View Slide

53. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    PostMortems – COE (Correction of Errors)
    The 5 WHYs
    Outage
    Because
    of …
    Because
    of …
    Because
    of …
    Because
    of …
    

    View Slide

54. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Rules to remember!
    There is no isolated ‘cause’ of an accident.
    

    View Slide

55. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    More questions to ask
    • Can you clarify if there were any preceding events?
    • Why would they believe acting in this way was the best course of action to
    deliver the desired outcome?
    • Is there another failure mode that could present here?
    • What decisions or events prior to this made this work before?
    • Why stop there – are there places to dig deeper that could shine a light more
    on this?
    • Did others step in to help, to advise, or to intercede?
    

    View Slide

56. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    DON’T blame that one person …
    

    View Slide

57. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Fix
    

    View Slide

58. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Fix
    

    View Slide

59. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Big challenges to chaos engineering
    Mostly Cultural
    • no time or flexibility to simulate disasters.
    • teams already spending all of its time fixing things.
    • can be very political.
    • might force deep conversations.
    • deeply invested in a specific technical roadmap (micro-services) that
    chaos engineering tests show is not as resilient to failures as originally
    predicted.
    

    View Slide

60. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Big challenges to chaos engineering
    • Chaos Engineering won’t make your system more robust,
    People will.
    • Chaos Engineering won’t replace __all__ the rest (test, quality, …)
    • Chaos Engineering is NOT the only way to learn from failure
    • Rollbacks are HARD because of state.
    • Your systems will continue to fail, sorry.
    

    View Slide

61. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Changing culture takes time!
    Be patient…
    

    View Slide

62. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    

    View Slide

63. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    More Resources
    • https://mvdirona.com/jrh/talksAndPapers/JamesRH_Lisa.pdf
    • https://www.gremlin.com
    • https://queue.acm.org/detail.cfm?id=2353017
    • https://softwareengineeringdaily.com/
    • https://github.com/dastergon/awesome-sre
    • https://www.usenix.org/system/files/conference/osdi14/osdi14-paper-yuan.pdf
    • https://medium.com/@NetflixTechBlog
    • http://principlesofchaos.org
    • https://speakerdeck.com/tammybutow/chaos-engineering-bootcamp
    • https://github.com/adhorn/awesome-chaos-engineering
    • https://www.infoq.com/presentations/netflix-chaos-microservices
    • http://royal.pingdom.com/wp-content/uploads/2015/04/pingdom_uptime_cheat_sheet.pdf
    • http://willgallego.com/2018/04/02/no-seriously-root-cause-is-a-fallacy
    • https://medium.com/@adhorn
    

    View Slide

64. S U M M I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    

    View Slide

65. Thank you!
    S U M M I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    Adrian Hornsby
    https://medium.com/@adhorn
    

    View Slide