![@csanchez#scalingdocker
Pod
{
"id": "jenkins",
"kind": "Pod",
"apiVersion": "v1beta1",
"desiredState": {
"manifest": {
"version": "v1beta1",
"id": "jenkins",
"containers": [
{
"name": "jenkins",
"image": "csanchez/jenkins-kubernetes",
"ports": [
{
"containerPort": 8080,
"hostPort": 8080
},
{
"containerPort": 50000,
"hostPort": 50000
}
],
"volumeMounts": [
{
"name": "jenkins-data",
"mountPath": "/var/jenkins_home"
}
]
}
],](https://image.slidesharecdn.com/2015-04devoxxfr-150409115515-conversion-gate01/75/Scaling-Docker-with-Kubernetes-38-2048.jpg)
![@csanchez#scalingdocker
Pod
"volumes": [
{
"name": "jenkins-data",
"source": {
"persistentDisk": {
"pdName": "k8-jenkins",
"fsType": "ext4"
}
}
}
]
}
},
"labels": {
"name": “jenkins”,
"environment": “dev”
}
}](https://image.slidesharecdn.com/2015-04devoxxfr-150409115515-conversion-gate01/75/Scaling-Docker-with-Kubernetes-39-2048.jpg)
![@csanchez#scalingdocker
Replication
controller
{
"id": "jenkins-slave",
"apiVersion": "v1beta1",
"kind": "ReplicationController",
"desiredState": {
"replicas": 1,
"replicaSelector": {
"name": "jenkins-slave"
},
"podTemplate": {
"desiredState": {
"manifest": {
"version": "v1beta1",
"id": "jenkins-slave",
"containers": [
{"name": "jenkins-slave",
"image":"csanchez/jenkins-swarm-slave:1.20",
"command": [
“sh", "-c",
"/usr/local/bin/jenkins-slave.sh -master
http://$JENKINS_SERVICE_HOST:$JENKINS_SERVICE_PORT -
username jenkins -password jenkins -executors 1"
]
}]
}
},
"labels": {
"name": "jenkins-slave"
}
}
},
"labels": {
"name": "jenkins-slave"
}
}](https://image.slidesharecdn.com/2015-04devoxxfr-150409115515-conversion-gate01/75/Scaling-Docker-with-Kubernetes-43-2048.jpg)
Uploaded byCarlos Sanchez
Scaling Docker with Kubernetes
The document discusses scaling Docker containers using Kubernetes, emphasizing the benefits of container orchestration for high availability and efficiency in distributed applications. It outlines the architecture of Kubernetes, including pods, replication controllers, and services, while detailing networking, deployment, and provisioning in various cloud environments. Additionally, it highlights tools and integrations such as Docker Compose and Jenkins for managing multi-container applications.
More Related Content
Similar to Scaling Docker with Kubernetes
Scaling Docker with Kubernetes
- 1. @csanchez#scalingdocker Scaling Docker with Kubernetes CarlosSanchez @csanchez http://csanchez.org
- 2. Containers & microservices
- 3. @csanchez#scalingdocker Scale high availability distributed apps Googlestarts 20billion/week
- 4. @csanchez#scalingdocker But it isnot trivial
- 5. @csanchez#scalingdocker But it isnot trivial
- 6. @csanchez#scalingdocker But it isnot trivial
- 7. @csanchez#scalingdocker Service oriented think aboutservices, not machines
- 8. Unix processes -lightweight VMs
- 9. @YourTwitterHandle#DVXFR14{session hashtag} @csanchez#scalingdocker Docker Linuxcontainers Union File System File System Users Processes Network
- 10. @csanchez#scalingdocker Docker Linux required but Boot2Docker tothe rescue OS X Windows
- 11. @csanchez#scalingdocker Docker Build once, runanywhere (kind of) Bare metal Virtual Machines Cloud Docker
- 12.
- 13. @csanchez#scalingdocker developer oriented Dependency hell installationnightmares “it ran on my machine”
- 14. @csanchez#scalingdocker ops oriented no needto know internals of apps focus on OPs problems (scale, monitoring,…) clearer deliverables from dev
- 15. @YourTwitterHandle#DVXFR14{session hashtag} @csanchez#scalingdocker Dockerdelivery Docker image Docker infrastructure App System
- 16.
- 17.
- 18.
- 19.
- 20. @csanchez#scalingdocker Docker docker run -t-i ubuntu /bin/bash docker ps -a docker build -t csanchez/test .
- 21. @csanchez#scalingdocker Rocket App Container Image AppContainer Runtime App Container Discovery
- 22.
- 23. @csanchez#scalingdocker Container OS Minimal OS Containeroptimized No package manager Automatic atomic updates Easy rollbacks
- 24.
- 25. @csanchez#scalingdocker How would youdesign your infrastructure if you couldn't login? Ever. Kelsey Hightower CoreOS
- 26. @csanchez#scalingdocker Kubernetes Container cluster orchestration Dockercontainers across multiple hosts (nodes or minions) Higher level API Enforced state Monitoring of endpoints
- 27. @csanchez#scalingdocker Master Kubernetes API Server schedulingand synchronization etcd Kubernetes Controller Manager Server implements replication algorithm watching etcd
- 28. @csanchez#scalingdocker Node Docker Kubelet ensures state ofPods Kubernetes Proxy simple network proxy etcd SkyDNS ElasticSearch + Kibana
- 29.
- 30.
- 31.
- 32.
- 33. @csanchez#scalingdocker gcloud preview container --projectmy-project clusters create cluster-1 --zone us-central1-a --machine-type n1-standard-1 --no-scopes --num-nodes 2 --network default Google Container Engine
- 34.
- 35. @csanchez#scalingdocker Node { "id": "192.168.12.100", "kind": "Minion", "apiVersion":"v1beta1", "resources": { "capacity": { "cpu": 1000, "memory": 1073741824 } }, "labels": { "name": "node0", "role": "database" } }
- 36. @csanchez#scalingdocker Pod Group of colocatedcontainers Same network namespace/IP Environment variables Shared volumes host mounted empty volumes GCE data disks
- 37.
- 38. @csanchez#scalingdocker Pod { "id": "jenkins", "kind": "Pod", "apiVersion":"v1beta1", "desiredState": { "manifest": { "version": "v1beta1", "id": "jenkins", "containers": [ { "name": "jenkins", "image": "csanchez/jenkins-kubernetes", "ports": [ { "containerPort": 8080, "hostPort": 8080 }, { "containerPort": 50000, "hostPort": 50000 } ], "volumeMounts": [ { "name": "jenkins-data", "mountPath": "/var/jenkins_home" } ] } ],
- 39. @csanchez#scalingdocker Pod "volumes": [ { "name": "jenkins-data", "source":{ "persistentDisk": { "pdName": "k8-jenkins", "fsType": "ext4" } } } ] } }, "labels": { "name": “jenkins”, "environment": “dev” } }
- 40. @csanchez#scalingdocker Replication controller Ensure anumber of pods are running Pod templates Rolling update podpod pod container 1 container 2 container 3
- 41. @YourTwitterHandle#DVXFR14{session hashtag} @csanchez#scalingdocker Replicationcontrollers Node client podA Master Node podA Node podB podA podB
- 42.
- 43. @csanchez#scalingdocker Replication controller { "id": "jenkins-slave", "apiVersion": "v1beta1", "kind":"ReplicationController", "desiredState": { "replicas": 1, "replicaSelector": { "name": "jenkins-slave" }, "podTemplate": { "desiredState": { "manifest": { "version": "v1beta1", "id": "jenkins-slave", "containers": [ {"name": "jenkins-slave", "image":"csanchez/jenkins-swarm-slave:1.20", "command": [ “sh", "-c", "/usr/local/bin/jenkins-slave.sh -master http://$JENKINS_SERVICE_HOST:$JENKINS_SERVICE_PORT - username jenkins -password jenkins -executors 1" ] }] } }, "labels": { "name": "jenkins-slave" } } }, "labels": { "name": "jenkins-slave" } }
- 44. @csanchez#scalingdocker Services Pod discovery IP perservice Route to pods selected with labels Can create a load balancer in GCE
- 45.
- 46. @csanchez#scalingdocker Services { "id": "jenkins", "kind": "Service", "apiVersion":"v1beta1", "port": 8888, "containerPort": 8080, "selector": { "name": "jenkins" }, "createExternalLoadBalancer": true } { "id": "jenkins-slave", "kind": "Service", "apiVersion": "v1beta1", "port": 50000, "containerPort": 50000, "selector": { "name": "jenkins" } }
- 47. @csanchez#scalingdocker Networking all containers cancommunicate with all other containers without NAT all nodes can communicate with all containers (and vice-versa) without NAT the IP that a container sees itself as is the same IP that others see it as Containers in a Pod can talk using localhost
- 48. @csanchez#scalingdocker Networking Every machine inthe cluster is assigned a full subnet ie. node A 10.0.1.0/24 and node B 10.0.2.0/24 Simpler port mapping Only supported by GCE CoreOS flannel Creates an overlay network in other providers
- 49.
- 50.
- 51. @csanchez#scalingdocker Docker Machine Provision Dockerengines Amazon EC2 Microsoft Azure Google Compute Engine OpenStack Rackspace VirtualBox VMware …
- 52. @csanchez#scalingdocker Docker Swarm Clustering forDocker containers Using the same API Integrates with Mesos / Mesosphere And planned Amazon EC2 Container Service (ECS) Google Kubernetes IBM Bluemix Container Service Joyent Smart Data Center Microsoft Azure
- 53. @csanchez#scalingdocker Docker Compose Orchestration ofmulti-container apps Based on Fig Defined by: containers configuration links volumes
- 54. @csanchez#scalingdocker Apache Mesos A distributedsystems kernel Docker Containerizer Marathon & Chronos
- 55. @csanchez#scalingdocker Flocker data volume manager multi-hostDocker cluster management tool ZFS volumes https://github.com/ClusterHQ/flocker
- 56.
- 57.
- 58. @csanchez#scalingdocker Kubernetes and Jenkins Asa plugin a different approach https://github.com/jenkinsci/kubernetes-plugin
- 59.
- 60.
- 61. @csanchez#scalingdocker Example code andslides Available at http://slideshare.csanchez.org https://github.com/carlossg/kubernetes-jenkins http://blog.csanchez.org
- 62.