Kubernetes: A Cautionary Tale (Minnebar)

The gap between Minikube and Kubernetes in production is vast and full of pain. Want to know what it's like to implement Kubernetes in production for the first time before you collect the bumps and bruises yourself? Join me to learn about the ups and downs of one company's initial experience - including who was involved, the scope, and what was intentionally deferred.

We will cover:
* Why even consider Kubernetes (Hint: you might not benefit from it!)
* How a real company approached their first proof-of-concept
* How goals changed during implementation of k8s
* What we learned so far
* What we still need to learn

Transcript

1. Kubernetes: A Cautionary Tale
   

   View Slide

2. C.H. ROBINSON
   Accelerating commerce through the worlds most powerful supply chain platform
   

   View Slide

3. Our people, processes, and technology improve the world's transportation and
   supply chains, delivering exceptional value to our customers and suppliers.
   OVERVIEW
   $16.6B
   in 2018 Gross 

   Revenues
   15,000+
   Employees
   Worldwide
   123K
   Active Customers
   Worldwide
   73K
   Active Carriers
   and Suppliers
   1100+
   IT Employees in
   14 Countries
   

   View Slide

4. AI USE CASES
   PRESCRIPTIVE &
   PREDICTIVE
   PERSONALIZED
   FLEET
   OPTIMIZATION
   MODEL BASED
   PRICING
   MAKE
   CONNECTIONS
   

   View Slide

5. View Slide

6. Kubernetes experience?
   

   View Slide

7. View Slide

8. View Slide

9. View Slide

10. View Slide

11. So what do you do?
    (A view from early 2018)
    

    View Slide

12. • Why K8s? (Hint: Might not be right for you)
    • One Approach
    • How Goals Changed
    • What We Learned
    • What Remained
    

    View Slide

13. Why Kubernetes (k8s)?
    

    View Slide

14. Why containers?
    

    View Slide

15. View Slide

16. When do I need Kubernetes (k8s)?
    

    View Slide

17. 1. Cloud Hosted (flexibility)
    2. Automation (CI/CD)
    3. Culture (people matter)
    4. Microservices
    5. Containers
    6. Orchestration (e.g. Kubernetes)
    https://container-solutions.com/when-is-the-wrong-time-to-use-kubernetes/
    

    View Slide

18. View Slide

19. View Slide

20. Why not Swarm?
    

    View Slide

21. How hard is K8s to learn?
    

    View Slide

22. “It depends.” ;)
    (Not polished in late 2017)
    

    View Slide

23. View Slide

24. View Slide

25. View Slide

26. In 2019?
    

    View Slide

27. View Slide

28. View Slide

29. View Slide

30. View Slide

31. View Slide

32. View Slide

33. View Slide

34. View Slide

35. View Slide

36. View Slide

37. View Slide

38. View Slide

39. View Slide

40. View Slide

41. Valuable for my team?
    – we learned a lot…
    … and apps stayed up.
    

    View Slide

42. Valuable for my company?
    Definitely.
    

    View Slide

43. View Slide

44. View Slide

45. View Slide

46. View Slide

47. View Slide

48. View Slide

49. View Slide

50. View Slide

51. View Slide

52. View Slide

53. View Slide

54. View Slide

55. What We Learned
    • Don’t roll your own
    • Certificates are important
    • Networking knowledge is critical
    • Secrets aren’t secret
    • Some apps don’t benefit
    

    View Slide

56. Don’t Roll Your Own
    If you don’t sell k8s, don’t build it
    • ~22 Hosted solutions
    • ~28 Cloud turnkey solutions
    • ~18 On-premises turnkey solutions
    https://kubernetes.io/docs/setup/pick-right-solution/
    

    View Slide

57. Certificates Are Important
    You need to “generate TLS certificates
    for the following components: etcd,
    kube-apiserver, kube-controller-
    manager, kube-scheduler, kubelet,
    and kube-proxy.”
    … And that’s before even considering your apps.
    

    View Slide

58. Networking Knowledge is Critical
    “Kubernetes cluster networking is perhaps one of the most
    complex components of the Kubernetes infrastructure because
    it involves so many layers and parts (e.g., container-to-container
    networking, Pod networking, services, ingress, load balancers),
    and many users are struggling to make sense of it all.”
    - Kirill Goltsman, Supergiant
    

    View Slide

59. Secrets Aren’t Secret
    • ENV vs volume mount
    • Base64 encoding in cluster
    • `kubeseal`
    • HashiCorp Vault
    

    View Slide

60. Some Apps Don’t Benefit
    • Where do they scale?
    • Are they “cluster-aware”?
    • What about failed deployments?
    

    View Slide

61. What Remains
    • CI/CD into K8s
    • How to get our nodes (vendor?)
    • Improved storage, external state
    • Namespaces and environments
    

    View Slide

62. Multiple Environments
    • Dev == Minikube?
    • Integration == Namespace? Cluster?
    • Production == 1+ clusters?
    • Testing apps vs testing k8s upgrades
    

    View Slide

63. Closing
    • Think before you k8s - do you need it?
    • Plan to learn - in stages, over time
    • People matter - especially networking folk ;)
    • Expect plans to evolve
    • Blaine? Eden Prairie?
    

    View Slide