Docker: From Dev to Prod [JavaSkop 2015]

A quick introduction to the docker technology followed by where the road to Docker may head to: Continuous Deployment, DevOps, High Availability Setups.

Transcript

1. Something until the talk starts, eg. Flickr slideshow
2. Docker from Dev to Prod for JavaSkop 2015
3. Chris @cimnine Disclaimer: The opinions expressed are those of the speaker and do not reflect opinions of any other person, company or agency.
4. Chris @cimnine Disclaimer: The opinions expressed are those of the speaker and do not reflect opinions of any other person, company or agency.
5. Chris @cimnine Disclaimer: The opinions expressed are those of the speaker and do not reflect opinions of any other person, company or agency.
6. Dev CI Integration Pre-Prod Prod
7. orange lines are data flow and blue lines are control flow VCS DEV DEV CI CI Integr. Integr. Pre Prod Pre Prod Prod Prod Dev
8. VCS DEV DEV CI CI Integr. Integr. Pre Prod Pre Prod Prod Prod Dev orange lines are data flow and blue lines are control flow
9. VCS DEV DEV CI CI Integr. Integr. Pre Prod Pre Prod Prod Prod Dev orange lines are data flow and blue lines are control flow
10. VCS DEV DEV CI CI Integr. Integr. Pre Prod Pre Prod Prod Prod Dev Ops orange lines are data flow and blue lines are control flow
12. VCS DEV DEV CI CI Integr. Integr. Dev Ops Pre- Prod Prod orange lines are data flow and blue lines are control flow
13. Dev CI Integration Pre-Prod Prod
14. Dev Dev CI Integration Pre-Prod Prod
15. Dev Dev CI Integration Pre-Prod Prod
17. jar Java
18. jar Java
19. jar Java
20. Dev Dev CI Integration Pre-Prod Prod
21. Dev CI Integration Pre-Prod Prod
22. VCS DEV DEV CI CI Dev orange lines are data flow and blue lines are control flow
23. VCS DEV DEV CI CI Dev git push orange lines are data flow and blue lines are control flow
24. VCS DEV DEV CI CI Dev post-commit hook orange lines are data flow and blue lines are control flow
25. Dev CI CI Integration Pre-Prod Prod
26. Dev CI CI Integration Pre-Prod Prod
27. Dev CI Integration Pre-Prod Prod
28. VCS DEV DEV CI CI Integr. Integr. Dev ? orange lines are data flow and blue lines are control flow
29. VCS DEV DEV CI CI Repository Integr. Integr. Dev orange lines are data flow and blue lines are control flow
30. VCS DEV DEV CI CI Docker Registry Integr. Integr. Dev orange lines are data flow and blue lines are control flow
31. VCS DEV DEV CI CI Integr. Integr. Dev Docker Registry docker push orange lines are data flow and blue lines are control flow
32. VCS DEV DEV CI CI Integr. Integr. Dev Docker Registry ssh + docker run orange lines are data flow and blue lines are control flow
33. VCS DEV DEV CI CI Integr. Integr. Helper Dev Docker Registry orange lines are data flow and blue lines are control flow
34. VCS DEV DEV CI CI Integr. Integr. Helper Dev Docker Registry install that thing orange lines are data flow and blue lines are control flow
35. VCS DEV DEV CI CI Integr. Integr. Helper Dev Docker Registry ssh + docker run orange lines are data flow and blue lines are control flow
36. Dev CI Integration Pre-Prod Prod
37. Dev CI Integration Pre-Prod Prod
38. VCS DEV DEV CI CI Integr. Integr. Helper Dev Docker Registry Pre Prod Pre Prod Prod Prod orange lines are data flow and blue lines are control flow
39. VCS DEV DEV CI CI Integr. Integr. Helper Dev Docker Registry Pre Prod Pre Prod Prod Prod Ops orange lines are data flow and blue lines are control flow
40. VCS DEV DEV CI CI Integr. Integr. Helper Dev Docker Registry Pre Prod Pre Prod Prod Prod Ops orange lines are data flow and blue lines are control flow
41. VCS DEV DEV CI CI Integr. Integr. Helper Dev Docker Registry Pre Prod Pre Prod Prod Prod Ops orange lines are data flow and blue lines are control flow
42. Easy Packaging Consistent Dependencies Repeatable Deployments Easy Automation
43. Logging Monitoring Scaling Continuous Delivery Automatic Recovery
44. VCS DEV DEV CI CI Integr. Integr. Helper Dev Docker Registry Pre Prod Pre Prod Prod Prod Ops orange lines are data flow and blue lines are control flow
45. VCS DEV DEV CI CI Integr. Integr. Helper Dev Docker Registry Pre Prod Pre Prod Prod Prod orange lines are data flow and blue lines are control flow
46. VCS DEV DEV CI CI Integr. Integr. Helper Dev Docker Registry Pre Prod Pre Prod Prod Prod orange lines are data flow and blue lines are control flow
47. VCS DEV DEV CI CI Integr. Integr. Scheduler (Master) Dev Docker Registry Pre Prod Pre Prod Prod Prod A A A orange lines are data flow and blue lines are control flow agent
48. VCS DEV DEV CI CI Integr. Integr. Scheduler (Master) Dev Docker Registry Pre Prod Pre Prod Prod Prod A A A orange lines are data flow and blue lines are control flow
49. Cluster State Master Worker Host w/ Agent Worker Host w/ Agent Worker Host w/ Agent Worker Host w/ Agent Worker Host w/ Agent
50. Worker Host w/ Agent Worker Host w/ Agent Worker Host w/ Agent Worker Host w/ Agent Worker Host w/ Agent Cluster State Master
51. Worker Host w/ Agent Worker Host w/ Agent Worker Host w/ Agent Worker Host w/ Agent Worker Host w/ Agent commands desired state vs. reports actual state Cluster State Master
52. Worker Host w/ Agent Worker Host w/ Agent Worker Host w/ Agent Worker Host w/ Agent Worker Host w/ Agent commands desired state vs. reports actual state Cluster State Master
53. Worker Host w/ Agent Worker Host w/ Agent Worker Host w/ Agent Worker Host w/ Agent Worker Host w/ Agent commands desired state vs. reports actual state Cluster State Master
54. Worker Host w/ Agent Worker Host w/ Agent Worker Host w/ Agent Worker Host w/ Agent Worker Host w/ Agent commands desired state vs. reports actual state Cluster State Master
55. Worker Worker Worker Worker Worker commands desired state vs. reports actual state Cluster State Master
56. Orchestration
57. Which container runs on which host? Orchestration
58. Which container runs on which host? How does the traffic come there? Orchestration
59. Which container runs on which host? How does the traffic come there? How do the services talk with each other? Orchestration
60. Which container runs on which host? How does the traffic come there? How do the services talk with each other? How to do updates w/o service interruption? Orchestration
61. Which container runs on which host? How does the traffic come there? How do the services talk with each other? How to scale out? How to do updates w/o service interruption? Orchestration
62. Which container runs on which host? How does the traffic come there? How do the services talk with each other? Where do the logs go? How to scale out? How to do updates w/o service interruption? Orchestration
63. Which container runs on which host? How does the traffic come there? How do the services talk with each other? Where do the logs go? How to scale out? How to do updates w/o service interruption? Orchestration
64. Orchestration Kubernetes Rancher Docker Swarm Apache Mesos with Marathon Spotify Helios Newrelic Centurion
65. Orchestration Kubernetes Rancher Docker Swarm Apache Mesos with Marathon Spotify Helios Newrelic Centurion
66. Cloud Provider Google Container Engine Amazon EC2 Container Service Microsoft Azure Redhat Openstack Cloudfoundry heroku
67. Dev CI Integration Pre-Prod Prod
68. http://12factor.net
69. I. Codebase One codebase tracked in revision control, many deploys II. Dependencies Explicitly declare and isolate dependencies III. Config Store config in the environment IV. Backing Services Treat backing services as attached resources V. Build, release, run Strictly separate build and run stages VI. Processes Execute the app as one or more stateless processes
70. VII. Port binding Export services via port binding VIII. Concurrency Scale out via the process model IX. Disposability Maximize robustness with fast startup and graceful shutdown X. Dev/prod parity Keep development, staging, and production as similar as possible XI. Logs Treat logs as event streams XII. Admin processes Run admin/management tasks as one-off processes
71. Dev CI Integration Pre-Prod Prod
73. Databases • Very own requirements: • State-full • Low CPU but high I/O • Distributed data • (usually) complex to scale • (usually) not built for “hot swap” •  not 12 Factor applications
74. Databases • Don’t dockerize databases • Keep them external to the system • DBA will love you not hate you • Easier to maintain • Very own cluster mechanisms • Easier routing in Datacenters