Welcome to Kubernetes Tasks’s documentation!¶
This Guide Structure
This guide is designed to complement instructor-led presentations by providing step-by-step instructions for hands-on exercises.
Kubernetes Installation¶
It’s expected, that you will install Kubernetes to 3 VMs / hosts - to have multinode installation. The installation part is taken from these two URLs:
- https://kubernetes.io/docs/setup/independent/install-kubeadm/
- https://kubernetes.io/docs/setup/independent/create-cluster-kubeadm/
Master node installation¶
SSH to the first VM which will be your Master node:
$ ssh root@node1
Enable packet forwarding:
$ sed -i 's/^#net.ipv4.ip_forward=1/net.ipv4.ip_forward=1/' /etc/sysctl.d/99-sysctl.conf
$ sysctl --quiet --system
Set the Kubernetes version which will be installed:
$ KUBERNETES_VERSION="1.10.3"
Set the proper CNI URL:
$ CNI_URL="https://raw.githubusercontent.com/coreos/flannel/v0.10.0/Documentation/kube-flannel.yml"
For Flannel installation you need to use proper “pod-network-cidr”:
$ POD_NETWORK_CIDR="10.244.0.0/16"
Add the Kubernetes repository (details):
$ apt-get update -qq && apt-get install -y -qq apt-transport-https curl
$ curl -s https://packages.cloud.google.com/apt/doc/apt-key.gpg | apt-key add -
$ tee /etc/apt/sources.list.d/kubernetes.list << EOF2
deb https://apt.kubernetes.io/ kubernetes-xenial main
EOF2
Install necessary packages:
$ apt-get update -qq
$ apt-get install -y -qq docker.io kubelet=${KUBERNETES_VERSION}-00 kubeadm=${KUBERNETES_VERSION}-00 kubectl=${KUBERNETES_VERSION}-00
Install Kubernetes Master:
$ kubeadm init --pod-network-cidr=$POD_NETWORK_CIDR --kubernetes-version v${KUBERNETES_VERSION}
Copy the “kubectl” config files to the home directory:
$ test -d $HOME/.kube || mkdir $HOME/.kube
$ cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
$ chown -R $USER:$USER $HOME/.kube
Install CNI:
$ export KUBECONFIG=/etc/kubernetes/admin.conf
$ kubectl apply -f $CNI_URL
Your Kuberenets Master node should be ready now. You can check it using this command:
$ kubectl get nodes
Worker nodes installation¶
Let’s connect the worker nodes now. SSH to the worker nodes and repeat these commands on all of them in paralel:
$ ssh root@node2
$ ssh root@node3
Set the Kubernetes version which will be installed:
$ KUBERNETES_VERSION="1.10.3"
Add the Kubernetes repository (details):
$ apt-get update -qq && apt-get install -y -qq apt-transport-https curl
$ curl -s https://packages.cloud.google.com/apt/doc/apt-key.gpg | apt-key add -
$ tee /etc/apt/sources.list.d/kubernetes.list << EOF2
deb https://apt.kubernetes.io/ kubernetes-xenial main
EOF2
Enable packet forwarding:
$ sed -i 's/^#net.ipv4.ip_forward=1/net.ipv4.ip_forward=1/' /etc/sysctl.d/99-sysctl.conf
$ sysctl --quiet --system
Install necessary packages:
$ apt-get update -qq
$ apt-get install -y -qq docker.io kubelet=${KUBERNETES_VERSION}-00 kubeadm=${KUBERNETES_VERSION}-00 kubectl=${KUBERNETES_VERSION}-00
All the woker nodes are prepared now - let’s connect them to master node. SSH to the master node again and generate the “joining” command:
$ ssh -t root@node1 "kubeadm token create --print-join-command"
You sould see something like:
$ kubeadm join <master-ip>:<master-port> --token <token> --discovery-token-ca-cert-hash sha256:<hash>
Execute the generated command on all worker nodes:
$ ssh -t root@node2 "kubeadm join --token ..."
$ ssh -t root@node3 "kubeadm join --token ..."
SSH back to the master nodes and check the cluster status - all the nodes should appear there in “Ready” status after while:
$ ssh root@node1
$ # Check nodes
$ kubectl get nodes
Allow pods to be scheduled on the master:
$ kubectl taint nodes node1 node-role.kubernetes.io/master-
Enable routing from local machine (host) to the kubernetes pods/services/etc. Adding routes (10.244.0.0/16, 10.96.0.0/12) -> [$NODE1_IP]:
$ sudo bash -c "ip route | grep -q 10.244.0.0/16 && ip route del 10.244.0.0/16; ip route add 10.244.0.0/16 via $NODE1_IP"
$ sudo bash -c "ip route | grep -q 10.96.0.0/12 && ip route del 10.96.0.0/12; ip route add 10.96.0.0/12 via $NODE1_IP"
Real installation example¶
Kubernetes Basics¶
Create directory where the files will be stored
$ mkdir files
Enable bash-completion for kubectl (bash-completion needs to be installed)
$ source <(kubectl completion bash)
Check the cluster status (if it is healthy)
$ kubectl get componentstatuses
List all namespaces
$ kubectl get namespaces
Create namespace ‘myns’
$ kubectl create namespace myns
Change default namespace for current context
$ kubectl config set-context $(kubectl config current-context) --namespace=myns
List out all of the nodes in our cluster
$ kubectl get pods -o wide --all-namespaces --show-labels --sort-by=.metadata.name
Get more details about a specific node
$ kubectl describe node $(kubectl get node --output=jsonpath="{.items[0].metadata.name}")
Helm Installation¶
Helm installation: https://github.com/kubernetes/helm/blob/master/docs/rbac.md
$ curl https://raw.githubusercontent.com/kubernetes/helm/master/scripts/get | bash
$ kubectl create serviceaccount tiller --namespace kube-system
$ kubectl create clusterrolebinding tiller-cluster-rule --clusterrole=cluster-admin --serviceaccount=kube-system:tiller
$ helm init --wait --service-account tiller
$ helm repo update
Install Traefik - Træfik is a modern HTTP reverse proxy and load balancer
$ helm install stable/traefik --wait --name my-traefik --namespace kube-system --set serviceType=NodePort,dashboard.enabled=true,accessLogs.enabled=true,rbac.enabled=true,metrics.prometheus.enabled=true
$ kubectl describe svc my-traefik --namespace kube-system
Install rook - File, Block, and Object Storage Services for your Cloud-Native Environment
$ helm repo add rook-stable https://charts.rook.io/stable
$ helm install --wait --name rook-ceph --namespace rook-ceph-system rook-stable/rook-ceph --version v0.9.3
$ sleep 60
Create your Rook cluster
$ kubectl create -f https://raw.githubusercontent.com/rook/rook/release-0.9/cluster/examples/kubernetes/ceph/cluster.yaml
Running the Toolbox with ceph commands
$ kubectl create -f https://raw.githubusercontent.com/rook/rook/release-0.9/cluster/examples/kubernetes/ceph/toolbox.yaml
Create a storage class based on the Ceph RBD volume plugin
$ kubectl create -f https://raw.githubusercontent.com/rook/rook/release-0.9/cluster/examples/kubernetes/ceph/storageclass.yaml
Create a shared file system which can be mounted read-write from multiple pods
$ kubectl create -f https://raw.githubusercontent.com/rook/rook/release-0.9/cluster/examples/kubernetes/ceph/filesystem.yaml
$ sleep 150
Check the status of your Ceph installation
$ export ROOK_CEPH_TOOLS_POD=$(kubectl -n rook-ceph get pod -l "app=rook-ceph-tools" -o jsonpath="{.items[0].metadata.name}")
$ kubectl -n rook-ceph exec ${ROOK_CEPH_TOOLS_POD} -- ceph status
$ kubectl -n rook-ceph exec ${ROOK_CEPH_TOOLS_POD} -- ceph osd status
Check health detail of Ceph cluster
$ kubectl -n rook-ceph exec ${ROOK_CEPH_TOOLS_POD} -- ceph health detail
Check monitor quorum status of Ceph
$ kubectl -n rook-ceph exec ${ROOK_CEPH_TOOLS_POD} -- ceph quorum_status --format json-pretty
Dump monitoring information from Ceph
$ kubectl -n rook-ceph exec ${ROOK_CEPH_TOOLS_POD} -- ceph mon dump
Check the cluster usage status
$ kubectl -n rook-ceph exec ${ROOK_CEPH_TOOLS_POD} -- ceph df
Check OSD usage of Ceph
$ kubectl -n rook-ceph exec ${ROOK_CEPH_TOOLS_POD} -- ceph osd df
Check the Ceph monitor, OSD, pool, and placement group stats
$ kubectl -n rook-ceph exec ${ROOK_CEPH_TOOLS_POD} -- ceph mon stat
$ kubectl -n rook-ceph exec ${ROOK_CEPH_TOOLS_POD} -- ceph osd stat
$ kubectl -n rook-ceph exec ${ROOK_CEPH_TOOLS_POD} -- ceph osd pool stats
$ kubectl -n rook-ceph exec ${ROOK_CEPH_TOOLS_POD} -- ceph pg stat
List the Ceph pools in detail
$ kubectl -n rook-ceph exec ${ROOK_CEPH_TOOLS_POD} -- ceph osd pool ls detail
Check the CRUSH map view of OSDs
$ kubectl -n rook-ceph exec ${ROOK_CEPH_TOOLS_POD} -- ceph osd tree
List the cluster authentication keys
$ kubectl -n rook-ceph exec ${ROOK_CEPH_TOOLS_POD} -- ceph auth list
Change the size of Ceph replica for “replicapool” pool
$ kubectl get cephblockpool --namespace=rook-ceph replicapool -o yaml | sed "s/size: 1/size: 3/" | kubectl replace -f -
List details for “replicapool”
$ kubectl describe cephblockpool --namespace=rook-ceph replicapool
See the manifest of the pod which should use rook/ceph
$ tee files/rook-ceph-test-job.yaml << EOF
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
name: rook-ceph-test-pv-claim
spec:
storageClassName: rook-ceph-block
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 1Gi
---
apiVersion: batch/v1
kind: Job
metadata:
name: rook-ceph-test
labels:
app: rook-ceph-test
spec:
template:
metadata:
labels:
app: rook-ceph-test
spec:
containers:
- name: rook-ceph-test
image: busybox
command: [ 'dd', 'if=/dev/zero', 'of=/data/zero_file', 'bs=1M', 'count=100' ]
volumeMounts:
- name: rook-ceph-test
mountPath: "/data"
restartPolicy: Never
volumes:
- name: rook-ceph-test
persistentVolumeClaim:
claimName: rook-ceph-test-pv-claim
EOF
Check the ceph usage
$ kubectl -n rook-ceph exec ${ROOK_CEPH_TOOLS_POD} -- ceph osd status
$ kubectl -n rook-ceph exec ${ROOK_CEPH_TOOLS_POD} -- ceph df
$ kubectl -n rook-ceph exec ${ROOK_CEPH_TOOLS_POD} -- ceph osd df
Apply the manifest
$ kubectl apply -f files/rook-ceph-test-job.yaml
$ sleep 10
Check the ceph usage again
$ kubectl -n rook-ceph exec ${ROOK_CEPH_TOOLS_POD} -- ceph osd status
$ kubectl -n rook-ceph exec ${ROOK_CEPH_TOOLS_POD} -- ceph df
$ kubectl -n rook-ceph exec ${ROOK_CEPH_TOOLS_POD} -- ceph osd df
List the Persistent Volume Claims
$ kubectl get pvc
Delete the job
$ kubectl delete job rook-ceph-test
Install Prometheus - Prometheus Operator creates/configures/manages Prometheus clusters atop Kubernetes
$ helm repo add coreos https://s3-eu-west-1.amazonaws.com/coreos-charts/stable/
$ helm install coreos/prometheus-operator --wait --name my-prometheus-operator --namespace monitoring
$ helm install coreos/kube-prometheus --name my-kube-prometheus --namespace monitoring --set alertmanager.ingress.enabled=true,alertmanager.ingress.hosts[0]=alertmanager.domain.com,alertmanager.storageSpec.volumeClaimTemplate.spec.storageClassName=rook-block,alertmanager.storageSpec.volumeClaimTemplate.spec.accessModes[0]=ReadWriteOnce,alertmanager.storageSpec.volumeClaimTemplate.spec.resources.requests.storage=20Gi,grafana.adminPassword=admin123,grafana.ingress.enabled=true,grafana.ingress.hosts[0]=grafana.domain.com,prometheus.ingress.enabled=true,prometheus.ingress.hosts[0]=prometheus.domain.com,prometheus.storageSpec.volumeClaimTemplate.spec.storageClassName=rook-block,prometheus.storageSpec.volumeClaimTemplate.spec.accessModes[0]=ReadWriteOnce,prometheus.storageSpec.volumeClaimTemplate.spec.resources.requests.storage=20Gi
$ GRAFANA_PASSWORD=$(kubectl get secret --namespace monitoring my-kube-prometheus-grafana -o jsonpath="{.data.password}" | base64 --decode ; echo)
$ echo "Grafana login: admin / $GRAFANA_PASSWORD"
Install Heapster - Compute Resource Usage Analysis and Monitoring of Container Clusters
$ helm install stable/heapster --name my-heapster --set rbac.create=true
Install Kubernetes Dashboard - General-purpose web UI for Kubernetes clusters
$ helm install stable/kubernetes-dashboard --name=my-kubernetes-dashboard --namespace monitoring --set ingress.enabled=true,rbac.clusterAdminRole=true
Pods¶
Check ‘kuard-pod.yaml’ manifest which will run kuard application once it is imported to Kubernetes
$ tee files/kuard-pod.yaml << EOF
apiVersion: v1
kind: Pod
metadata:
name: kuard
spec:
containers:
- image: gcr.io/kuar-demo/kuard-amd64:1
name: kuard
ports:
- containerPort: 8080
name: http
protocol: TCP
EOF
Start pod from the pod manifest via Kubernetes API (see the ‘ContainerCreating’ status)
$ kubectl apply --filename=files/kuard-pod.yaml; kubectl get pods
$ sleep 40
List pods (-o yaml will print all details)
$ kubectl get pods --namespace myns -o wide
Check pod details
$ kubectl describe pods kuard
Get IP for a kuard pod
$ kubectl get pods kuard -o jsonpath --template={.status.podIP}
Configure secure port-forwarding to access the specific pod exposed port using Kubernetes API Access the pod by opening the web browser with url: http://127.0.0.1:8080 and http://127.0.0.1:8080/fs/{etc,var,home}
$ kubectl port-forward kuard 8080:8080 &
Stop port forwarding
$ pkill -f "kubectl port-forward kuard 8080:8080"
Get the logs from pod (-f for tail) (–previous will get logs from a previous instance of the container)
$ kubectl logs kuard
Copy files to/from containers running in the pod
$ kubectl cp --container=kuard /etc/os-release kuard:/tmp/
Run commands in your container with exec (-it for interactive session). Check if I am in container
$ kubectl exec kuard -- cat /etc/os-release
Delete pod - see the status ‘Terminating’
$ kubectl delete pods/kuard; kubectl get pods
$ sleep 30
Check pods - the kuard should disappear form the ‘pod list’
$ kubectl get pods
Health Checks¶
Check ‘kuard-pod-health.yaml’ manifest which will start kuard and configure HTTP health check
$ tee files/kuard-pod-health.yaml << EOF
apiVersion: v1
kind: Pod
metadata:
name: kuard
spec:
volumes:
- name: "kuard-data"
hostPath:
path: "/var/lib/kuard"
containers:
- image: gcr.io/kuar-demo/kuard-amd64:1
name: kuard
volumeMounts:
- mountPath: "/data"
name: "kuard-data"
ports:
- containerPort: 8080
name: http
protocol: TCP
resources:
requests:
cpu: "100m"
memory: "128Mi"
limits:
cpu: "1000m"
memory: "256Mi"
# Pod must be ready, before Kubernetes start sending traffic to it
readinessProbe:
httpGet:
path: /ready
port: 8080
# Check is done every 2 seconds starting as soon as the pod comes up
periodSeconds: 2
# Start checking once pod is up
initialDelaySeconds: 0
# If three successive checks fail, then the pod will be considered not ready.
failureThreshold: 3
# If only one check succeeds, then the pod will again be considered ready.
successThreshold: 1
livenessProbe:
httpGet:
path: /healthy
port: 8080
# Start probe 5 seconds after all the containers in the Pod are created
initialDelaySeconds: 5
# The response must be max in 1 second and status HTTP code must be between 200 and 400
timeoutSeconds: 1
# Repeat every 10 seconds
periodSeconds: 10
# If more than 3 probes failed - the container will fail + restart
failureThreshold: 3
EOF
Create a Pod using this manifest and then port-forward to that pod
$ kubectl apply -f files/kuard-pod-health.yaml
$ sleep 30
Point your browser to http://127.0.0.1:8080 then click ‘Liveness Probe’ tab and then ‘fail’ link - it will cause to fail health checks
$ kubectl port-forward kuard 8080:8080 &
Stop port forwarding
$ pkill -f "kubectl port-forward kuard 8080:8080"
You will see ‘unhealthy’ messages in the in the following output
$ kubectl describe pods kuard | tail
Delete pod
$ kubectl delete pods/kuard
$ sleep 10
Labels, annotations, selectors¶
Create app1-prod deployment with labels (creates also Deployment)
$ kubectl run app1-prod --image=gcr.io/kuar-demo/kuard-amd64:1 --replicas=3 --port=8080 --labels="ver=1,myapp=app1,env=prod"
Create service (only routable inside cluster). The service is assigned Cluster IP (DNS record is automatically created) which load-balance across all of the pods that are identified by the selector
$ kubectl expose deployment app1-prod
Create app1-test deployment
$ kubectl run app1-test --image=gcr.io/kuar-demo/kuard-amd64:2 --replicas=1 --labels="ver=2,myapp=app1,env=test"
Create app2-prod
$ kubectl run app2-prod --image=gcr.io/kuar-demo/kuard-amd64:2 --replicas=2 --port=8080 --labels="ver=2,myapp=app2,env=prod"
$ sleep 5
Create service
$ kubectl expose deployment app2-prod
Check if the DNS record was properly created for the Cluster IPs. app2-prod [name of the service], myns [namespace that this service is in], svc [service], cluster.local. [base domain name for the cluster]
$ kubectl run nslookup --rm -it --restart=Never --image=busybox -- nslookup app2-prod
Create app2-staging
$ kubectl run app2-staging --image=gcr.io/kuar-demo/kuard-amd64:2 --replicas=1 --labels="ver=2,myapp=app2,env=staging"
Show deployments
$ kubectl get deployments -o wide --show-labels
Change labels
$ kubectl label deployments app1-test "canary=true"
Add annotation - usually longer than labels
$ kubectl annotate deployments app1-test description="My favorite deployment with my app"
List ‘canary’ deployments (with canary label)
$ kubectl get deployments -o wide --label-columns=canary
Remove label
$ kubectl label deployments app1-test "canary-"
List pods including labels
$ kubectl get pods --sort-by=.metadata.name --show-labels
List pods ver=2 using the –selector flag
$ kubectl get pods --selector="ver=2" --show-labels
List pods with 2 tags
$ kubectl get pods --selector="myapp=app2,ver=2" --show-labels
List pods where myapp=(app1 or app2)
$ kubectl get pods --selector="myapp in (app1,app2)" --show-labels
Label multiple pods
$ kubectl label pods -l canary=true my=testlabel
List all services
$ kubectl get services -o wide
Get service details
$ kubectl describe service app1-prod
Get service endpoints
$ kubectl describe endpoints app1-prod
List IPs belongs to specific pods
$ kubectl get pods -o wide --selector=myapp=app1,env=prod --show-labels
Cleanup all deployments
$ kubectl delete services,deployments -l myapp
ReplicaSet¶
Show minimal ReplicaSet definition
$ tee files/kuard-rs.yaml << EOF
apiVersion: extensions/v1beta1
kind: ReplicaSet
metadata:
name: kuard
spec:
replicas: 1
selector:
matchLabels:
app: kuard
version: "2"
template:
metadata:
labels:
app: kuard
version: "2"
spec:
containers:
- name: kuard
image: "gcr.io/kuar-demo/kuard-amd64:2"
EOF
Create ReplicaSet
$ kubectl apply -f files/kuard-rs.yaml
Check pods
$ kubectl get pods
Check ReplicaSet details
$ kubectl describe rs kuard
The pods have the same labels as ReplicaSet
$ kubectl get pods -l app=kuard,version=2 --show-labels
Check if pod is part of ReplicaSet
$ kubectl get pods -l app=kuard,version=2 -o json | jq ".items[].metadata"
Scale up ReplicaSet
$ kubectl scale replicasets kuard --replicas=4
New pods are beeing created
$ kubectl get pods -l app=kuard --show-labels
Delete ReplicaSet
$ kubectl delete rs kuard
DaemonSets and NodeSelector¶
Add labels to your nodes (hosts)
$ kubectl label nodes node2 ssd=true
Filter nodes based on labels
$ kubectl get nodes --selector ssd=true
Check ‘nginx-fast-storage.yaml’ which will provision nginx to ssd labeled nodes only. By default a DaemonSet will create a copy of a Pod on every node
$ tee files/nginx-fast-storage.yaml << EOF
apiVersion: extensions/v1beta1
kind: "DaemonSet"
metadata:
labels:
app: nginx
ssd: "true"
name: nginx-fast-storage
spec:
template:
metadata:
labels:
app: nginx
ssd: "true"
spec:
nodeSelector:
ssd: "true"
containers:
- name: nginx
image: nginx:1.10.0
EOF
Create daemonset from the nginx-fast-storage.yaml
$ kubectl apply -f files/nginx-fast-storage.yaml
Check the nodes where nginx was deployed
$ kubectl get pods -o wide
Add label ssd=true to the node3 - nginx should be deployed there automatically
$ kubectl label nodes node3 ssd=true
Check the nodes where nginx was deployed (it should be also on node3 with ssd=true label)
$ kubectl get pods -o wide
Check the nodes where nginx was deployed
$ kubectl delete ds nginx-fast-storage
Jobs¶
One-shot Jobs provide a way to run a single Pod once until successful termination. Pod is restarted in case of failure
$ kubectl run -it oneshot --image=gcr.io/kuar-demo/kuard-amd64:1 --restart=OnFailure -- --keygen-enable --keygen-exit-on-complete --keygen-num-to-gen 5
List all jobs
$ kubectl get jobs -o wide
Delete job
$ kubectl delete jobs oneshot
Show one-shot Job configuration file
$ tee files/job-oneshot.yaml << EOF
apiVersion: batch/v1
kind: Job
metadata:
name: oneshot
labels:
chapter: jobs
spec:
template:
metadata:
labels:
chapter: jobs
spec:
containers:
- name: kuard
image: gcr.io/kuar-demo/kuard-amd64:1
imagePullPolicy: Always
args:
- "--keygen-enable"
- "--keygen-exit-on-complete"
- "--keygen-num-to-gen=5"
restartPolicy: OnFailure
EOF
Create one-shot Job using a configuration file
$ kubectl apply -f files/job-oneshot.yaml
$ sleep 30
Print details about the job
$ kubectl describe jobs oneshot
Get pod name of a job called ‘oneshot’ and check the logs
$ POD_NAME=$(kubectl get pods --selector="job-name=oneshot" -o=jsonpath="{.items[0].metadata.name}")
$ kubectl logs ${POD_NAME}
Remove job oneshot
$ kubectl delete jobs oneshot
Show one-shot Job configuration file. See the keygen-exit-code parameter - nonzero exit code after generating three keys
$ tee files/job-oneshot-failure1.yaml << EOF
apiVersion: batch/v1
kind: Job
metadata:
name: oneshot
labels:
chapter: jobs
spec:
template:
metadata:
labels:
chapter: jobs
spec:
containers:
- name: kuard
image: gcr.io/kuar-demo/kuard-amd64:1
imagePullPolicy: Always
args:
- "--keygen-enable"
- "--keygen-exit-on-complete"
- "--keygen-exit-code=1"
- "--keygen-num-to-gen=3"
restartPolicy: OnFailure
EOF
Create one-shot Job using a configuration file
$ kubectl apply -f files/job-oneshot-failure1.yaml
$ sleep 60
Get pod status - look for CrashLoopBackOff/Error indicating pod restarts
$ kubectl get pod -l job-name=oneshot
Remove the job
$ kubectl delete jobs oneshot
Show Parallel Job configuration file - generate (5x10) keys generated in 5 containers
$ tee files/job-parallel.yaml << EOF
apiVersion: batch/v1
kind: Job
metadata:
name: parallel
labels:
chapter: jobs
spec:
# 5 pods simlutaneously
parallelism: 5
# repeat task 10 times
completions: 10
template:
metadata:
labels:
chapter: jobs
spec:
containers:
- name: kuard
image: gcr.io/kuar-demo/kuard-amd64:1
imagePullPolicy: Always
args:
- "--keygen-enable"
- "--keygen-exit-on-complete"
- "--keygen-num-to-gen=5"
restartPolicy: OnFailure
EOF
Create Parallel Job using a configuration file
$ kubectl apply -f files/job-parallel.yaml
Check the pods and list changes as they happen
$ kubectl get pods --watch -o wide &
$ sleep 10
Stop getting the pods
$ pkill -f "kubectl get pods --watch -o wide"
Remove the job
$ kubectl delete jobs parallel
Queue job example¶
Memory-based work queue system: Producer -> Work Queue -> Consumers diagram
$ tee /tmp/producer_queue_consumer-diagram.txt << EOF
+--------------+
| |
+-> | Consumer |
| | |
| +--------------+
|
+--------------+ +----------------+ | +--------------+
| | | | | | |
| Producer | +------> | Work Queue | +--+-> | Consumer |
| | | | | | |
+--------------+ +----------------+ | +--------------+
|
| +--------------+
| | |
+-> | Consumer |
| |
+--------------+
EOF
Create a simple ReplicaSet to manage a singleton work queue daemon
$ tee files/rs-queue.yaml << EOF
apiVersion: extensions/v1beta1
kind: ReplicaSet
metadata:
labels:
app: work-queue
component: queue
chapter: jobs
name: queue
spec:
replicas: 1
selector:
matchLabels:
app: work-queue
component: queue
chapter: jobs
template:
metadata:
labels:
app: work-queue
component: queue
chapter: jobs
spec:
containers:
- name: queue
image: "gcr.io/kuar-demo/kuard-amd64:1"
imagePullPolicy: Always
EOF
Create work queue using a configuration file
$ kubectl apply -f files/rs-queue.yaml
$ sleep 30
Configure port forwarding to connect to the ‘work queue daemon’ pod
$ QUEUE_POD=$(kubectl get pods -l app=work-queue,component=queue -o jsonpath="{.items[0].metadata.name}")
$ kubectl port-forward $QUEUE_POD 8080:8080 &
Expose work queue - this helps consumers+producers to locate the work queue via DNS
$ tee files/service-queue.yaml << EOF
apiVersion: v1
kind: Service
metadata:
labels:
app: work-queue
component: queue
chapter: jobs
name: queue
spec:
ports:
- port: 8080
protocol: TCP
targetPort: 8080
selector:
app: work-queue
component: queue
EOF
Create the service pod using a configuration file
$ kubectl apply -f files/service-queue.yaml
$ sleep 20
Create a work queue called ‘keygen’
$ curl -X PUT 127.0.0.1:8080/memq/server/queues/keygen
Create work items and load up the queue
$ for WORK in work-item-{0..20}; do curl -X POST 127.0.0.1:8080/memq/server/queues/keygen/enqueue -d "$WORK"; done
Queue should not be empty - check the queue by looking at the ‘MemQ Server’ tab in Web interface (http://127.0.0.1:8080/-/memq)
$ curl --silent 127.0.0.1:8080/memq/server/stats | jq
Show consumer job config file allowing start up five pods in parallel. Once the first pod exits with a zero exit code, the Job will not start any new pods (none of the workers should exit until the work is done)
$ tee files/job-consumers.yaml << EOF
apiVersion: batch/v1
kind: Job
metadata:
labels:
app: message-queue
component: consumer
chapter: jobs
name: consumers
spec:
parallelism: 5
template:
metadata:
labels:
app: message-queue
component: consumer
chapter: jobs
spec:
containers:
- name: worker
image: "gcr.io/kuar-demo/kuard-amd64:1"
imagePullPolicy: Always
args:
- "--keygen-enable"
- "--keygen-exit-on-complete"
- "--keygen-memq-server=http://queue:8080/memq/server"
- "--keygen-memq-queue=keygen"
restartPolicy: OnFailure
EOF
Create consumer job from config file
$ kubectl apply -f files/job-consumers.yaml
$ sleep 30
Five pods should be created to run until the work queue is empty. Open the web browser to see changing queue status (http://127.0.0.1:8080/-/memq)
$ kubectl get pods -o wide
Check the queue status - especially the ‘dequeued’ and ‘depth’ fields
$ curl --silent 127.0.0.1:8080/memq/server/stats | jq
Stop port-forwarding
$ pkill -f "kubectl port-forward $QUEUE_POD 8080:8080"
Clear the resources
$ kubectl delete rs,svc,job -l chapter=jobs
ConfigMaps¶
Show file with key/value pairs which will be available to the pod
$ tee files/my-config.txt << EOF
# This is a sample config file that I might use to configure an application
parameter1 = value1
parameter2 = value2
EOF
Create a ConfigMap with that file (environment variables are specified with a special valueFrom member)
$ kubectl create configmap my-config --from-file=files/my-config.txt --from-literal=extra-param=extra-value --from-literal=another-param=another-value
Show ConfigMaps
$ kubectl get configmaps
Show ConfigMap details
$ kubectl describe configmap my-config
See the YAML ConfigMap object
$ kubectl get configmaps my-config -o yaml
Prepare config file for ConfigMap usage
$ tee files/kuard-config.yaml << \EOF
apiVersion: v1
kind: Pod
metadata:
name: kuard-config
spec:
containers:
- name: test-container
image: gcr.io/kuar-demo/kuard-amd64:1
imagePullPolicy: Always
command:
- "/kuard"
- "$(EXTRA_PARAM)"
env:
- name: ANOTHER_PARAM
valueFrom:
configMapKeyRef:
name: my-config
key: another-param
# Define the environment variable
- name: EXTRA_PARAM
valueFrom:
configMapKeyRef:
# The ConfigMap containing the value you want to assign to ANOTHER_PARAM
name: my-config
# Specify the key associated with the value
key: extra-param
volumeMounts:
- name: config-volume
mountPath: /config
volumes:
- name: config-volume
configMap:
name: my-config
restartPolicy: Never
EOF
Apply the config file
$ kubectl apply -f files/kuard-config.yaml
$ sleep 20
{EXTRA_PARAM,ANOTHER_PARAM} variable has value from configmap my-config/{extra-param,another-param} and file /config/my-config.txt exists in container
$ kubectl exec kuard-config -- sh -xc "echo EXTRA_PARAM: \$EXTRA_PARAM; echo ANOTHER_PARAM: \$ANOTHER_PARAM && cat /config/my-config.txt"
Go to http://localhost:8080 and click on the ‘Server Env’ tab, then ‘File system browser’ tab (/config) and look for ANOTHER_PARAM and EXTRA_PARAM values
$ kubectl port-forward kuard-config 8080:8080 &
Stop port forwarding
$ pkill -f "kubectl port-forward kuard-config 8080:8080"
Remove pod”
$ kubectl delete pod kuard-config
Secrets¶
Download certificates
$ wget -q -c -P files https://storage.googleapis.com/kuar-demo/kuard.crt https://storage.googleapis.com/kuar-demo/kuard.key
Create a secret named kuard-tls
$ kubectl create secret generic kuard-tls --from-file=files/kuard.crt --from-file=files/kuard.key
Get details about created secret
$ kubectl describe secrets kuard-tls
Show secrets
$ kubectl get secrets
Update secrets - generate yaml and then edit the secret ‘kubectl edit configmap my-config’
$ kubectl create secret generic kuard-tls --from-file=files/kuard.crt --from-file=files/kuard.key --dry-run -o yaml | kubectl replace -f -
Create a new pod with secret attached
$ tee files/kuard-secret.yaml << EOF
apiVersion: v1
kind: Pod
metadata:
name: kuard-tls
spec:
containers:
- name: kuard-tls
image: gcr.io/kuar-demo/kuard-amd64:1
imagePullPolicy: Always
volumeMounts:
- name: tls-certs
mountPath: "/tls"
readOnly: true
volumes:
- name: tls-certs
secret:
secretName: kuard-tls
EOF
Apply the config file
$ kubectl apply -f files/kuard-secret.yaml
$ sleep 20
Set port-forwarding. Go to https://localhost:8080, check the certificate and click on “File system browser” tab (/tls)
$ kubectl port-forward kuard-tls 8443:8443 &
Stop port forwarding
$ pkill -f "kubectl port-forward kuard-tls 8443:8443"
Delete pod
$ kubectl delete pod kuard-tls
Deployments¶
Show nginx deployment definition
$ tee files/nginx-deployment.yaml << EOF
apiVersion: apps/v1
kind: Deployment
metadata:
name: nginx-deployment
labels:
app: nginx
spec:
selector:
matchLabels:
app: nginx
replicas: 3
template:
metadata:
labels:
app: nginx
spec:
containers:
- name: nginx
image: nginx:1.7.9
ports:
- containerPort: 80
EOF
Create nginx deployment
$ kubectl create -f files/nginx-deployment.yaml
List deployments
$ kubectl get deployments -o wide
Get deployment details
$ kubectl describe deployment nginx-deployment
Show deployment YAML file (look for: ‘nginx:1.7.9’)
$ kubectl get deployment nginx-deployment -o wide
Change deployment image (version 1.7.9 -> 1.8) - you can do the change by running ‘kubectl edit deployment nginx-deployment’ too…
$ kubectl set image deployment nginx-deployment nginx=nginx:1.8
See what is happening during the deployment change
$ kubectl rollout status deployment nginx-deployment
Get deployment details (see: ‘nginx:1.8’)
$ kubectl get deployment nginx-deployment -o wide
Show details for deployment
$ kubectl describe deployment nginx-deployment
See the deployment history (first there was version nginx:1.7.9, then nginx:1.8)
$ kubectl rollout history deployment nginx-deployment --revision=1
$ kubectl rollout history deployment nginx-deployment --revision=2
Rollback the deployment to previous version (1.7.9)
$ kubectl rollout undo deployment nginx-deployment
$ kubectl rollout status deployment nginx-deployment
Get deployment details - see the image is now again ‘nginx:1.7.9’
$ kubectl get deployment nginx-deployment -o wide
Rollback the deployment back to version (1.8)
$ kubectl rollout undo deployment nginx-deployment --to-revision=2
$ kubectl rollout status deployment nginx-deployment
Get deployment details - see the image is now again ‘nginx:1.8’
$ kubectl get deployment nginx-deployment -o wide
Check the utilization of pods
$ kubectl top pod --heapster-namespace=myns --all-namespaces --containers
Endpoints¶
Show external service DNS definition
$ tee files/dns-service.yaml << EOF
kind: Service
apiVersion: v1
metadata:
name: external-database
spec:
type: ExternalName
externalName: database.company.com
EOF
Create DNS name (CNAME) that points to the specific server running the database
$ kubectl create -f files/dns-service.yaml
Show services
$ kubectl get service
Remove service
$ kubectl delete service external-database
Self-Healing¶
Get pod details
$ kubectl get pods -o wide
Get first nginx pod and delete it - one of the nginx pods should be in ‘Terminating’ status
$ NGINX_POD=$(kubectl get pods -l app=nginx --output=jsonpath="{.items[0].metadata.name}")
$ kubectl delete pod $NGINX_POD; kubectl get pods -l app=nginx -o wide
$ sleep 10
Get pod details - one nginx pod should be freshly started
$ kubectl get pods -l app=nginx -o wide
Get deployement details and check the events for recent changes
$ kubectl describe deployment nginx-deployment
Halt one of the nodes (node2)
$ vagrant halt node2
$ sleep 30
Get node details - node2 Status=NotReady
$ kubectl get nodes
Get pod details - everything looks fine - you need to wait 5 minutes
$ kubectl get pods -o wide
Pod will not be evicted until it is 5 minutes old - (see Tolerations in ‘describe pod’ ). It prevents Kubernetes to spin up the new containers when it is not necessary
$ NGINX_POD=$(kubectl get pods -l app=nginx --output=jsonpath="{.items[0].metadata.name}")
$ kubectl describe pod $NGINX_POD | grep -A1 Tolerations
Sleeping for 5 minutes
$ sleep 300
Get pods details - Status=Unknown/NodeLost and new container was started
$ kubectl get pods -o wide
Get depoyment details - again AVAILABLE=3/3
$ kubectl get deployments -o wide
Power on the node2 node
$ vagrant up node2
$ sleep 70
Get node details - node2 should be Ready again
$ kubectl get nodes
Get pods details - ‘Unknown’ pods were removed
$ kubectl get pods -o wide
Persistent Storage¶
Install and configure NFS on node1
$ ssh $SSH_ARGS vagrant@node1 "sudo sh -xc \" apt-get update -qq; DEBIAN_FRONTEND=noninteractive apt-get install -y nfs-kernel-server > /dev/null; mkdir /nfs; chown nobody:nogroup /nfs; echo /nfs *\(rw,sync,no_subtree_check\) >> /etc/exports; systemctl restart nfs-kernel-server \""
Install NFS client to other nodes
$ for COUNT in {2..4}; do ssh $SSH_ARGS vagrant@node${COUNT} "sudo sh -xc \"apt-get update -qq; DEBIAN_FRONTEND=noninteractive apt-get install -y nfs-common > /dev/null\""; done
Show persistent volume object definition
$ tee files/nfs-volume.yaml << EOF
apiVersion: v1
kind: PersistentVolume
metadata:
name: nfs-pv
labels:
volume: nfs-volume
spec:
accessModes:
- ReadWriteMany
capacity:
storage: 1Gi
nfs:
server: node1
path: "/nfs"
EOF
Create persistent volume
$ kubectl create -f files/nfs-volume.yaml
Check persistent volumes
$ kubectl get persistentvolume
Show persistent volume claim object definition
$ tee files/nfs-volume-claim.yaml << EOF
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
name: nfs-pvc
spec:
accessModes:
- ReadWriteMany
resources:
requests:
storage: 1Gi
selector:
matchLabels:
volume: nfs-volume
EOF
Claim the persistent volume for our pod
$ kubectl create -f files/nfs-volume-claim.yaml
Check persistent volume claims
$ kubectl get persistentvolumeclaim
Show replicaset definition
$ tee files/nfs-test-replicaset.yaml << EOF
apiVersion: apps/v1
kind: ReplicaSet
metadata:
name: nfs-test
# labels so that we can bind a service to this pod
labels:
app: nfs-test
spec:
replicas: 2
selector:
matchLabels:
app: nfs-test
template:
metadata:
labels:
app: nfs-test
spec:
containers:
- name: nfs-test
image: busybox
command: [ 'sh', '-c', 'date >> /tmp/date && sleep 3600' ]
volumeMounts:
- name: nfs-test
mountPath: "/tmp"
volumes:
- name: nfs-test
persistentVolumeClaim:
claimName: nfs-pvc
securityContext:
runAsUser: 65534
fsGroup: 65534
EOF
Create replicaset
$ kubectl create -f files/nfs-test-replicaset.yaml
$ sleep 20
You can see the /tmp is mounted to both pods containing the same file ‘date’
$ NFS_TEST_POD2=$(kubectl get pods --no-headers -l app=nfs-test -o custom-columns=NAME:.metadata.name | head -1); echo $NFS_TEST_POD2
$ NFS_TEST_POD1=$(kubectl get pods --no-headers -l app=nfs-test -o custom-columns=NAME:.metadata.name | tail -1); echo $NFS_TEST_POD1
$ kubectl exec -it $NFS_TEST_POD1 -- sh -xc "hostname; echo $NFS_TEST_POD1 >> /tmp/date"
$ kubectl exec -it $NFS_TEST_POD2 -- sh -xc "hostname; echo $NFS_TEST_POD2 >> /tmp/date"
Show files on NFS server - there should be ‘nfs/date’ file with 2 dates
$ ssh $SSH_ARGS vagrant@node1 "set -x; ls -al /nfs -ls; ls -n /nfs; cat /nfs/date"
Node replacement¶
Move all pods away from node3
$ kubectl drain --delete-local-data --ignore-daemonsets node3
Get pod details
$ kubectl get pods -o wide --all-namespaces | grep node3
Destroy the node node3
$ vagrant destroy -f node3
Wait some time for Kubernetes to catch up…
$ sleep 40
The node3 shoult be in ‘NotReady’ state
$ kubectl get pods -o wide --all-namespaces
Remove the node3 from the cluster
$ kubectl delete node node3
Generate command which can add new node to Kubernetes cluster
$ KUBERNETES_JOIN_CMD=$(ssh $SSH_ARGS root@node1 "kubeadm token create --print-join-command"); echo $KUBERNETES_JOIN_CMD
Start new node
$ vagrant up node3
Install Kubernetes repository to new node
$ ssh $SSH_ARGS vagrant@node3 "sudo sh -xc \" apt-get update -qq; DEBIAN_FRONTEND=noninteractive apt-get install -y apt-transport-https curl > /dev/null; curl -s https://packages.cloud.google.com/apt/doc/apt-key.gpg | apt-key add -; echo deb https://apt.kubernetes.io/ kubernetes-xenial main > /etc/apt/sources.list.d/kubernetes.list \""
Install Kubernetes packages
$ ssh $SSH_ARGS vagrant@node3 "sudo sh -xc \" apt-get update -qq; DEBIAN_FRONTEND=noninteractive apt-get install -y docker.io kubelet=${KUBERNETES_VERSION}-00 kubeadm=${KUBERNETES_VERSION}-00 kubectl=${KUBERNETES_VERSION}-00 > /dev/null \""
Join node3 to the Kuberenets cluster
$ ssh $SSH_ARGS vagrant@node3 "sudo sh -xc \" $KUBERNETES_JOIN_CMD \""
$ sleep 40
Check the nodes - node3 should be there
$ kubectl get nodes
Notes¶
Show logs from specific docker container inside pod
$ kubectl logs --namespace=kube-system $(kubectl get pods -n kube-system -l k8s-app=kube-dns -o name) --container=dnsmasq --tail=10
$ kubectl logs --namespace=kube-system $(kubectl get pods -n kube-system -l k8s-app=kube-dns -o name) --container=kubedns --tail=10
See the logs directly on the Kubernetes node
$ ssh $SSH_ARGS vagrant@node1 "ls /var/log/containers/"
Show all
$ kubectl get all