Kubernetes 学習 Performing a Rolling Update を自前クラスタで復習
Performing a Rolling Update
https://kubernetes.io/docs/tutorials/kubernetes-basics/update/update-intro/
前回の続きを確認
$ kubectl get nodes NAME STATUS ROLES AGE VERSION kb1 Ready master 10d v1.13.2 kb2 Ready <none> 10d v1.13.2 kb3 Ready <none> 10d v1.13.2 $ kubectl get deployments NAME READY UP-TO-DATE AVAILABLE AGE kubernetes-bootcamp 2/2 2 2 18m $ kubectl get pods -o wide NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES kubernetes-bootcamp-6bf84cb898-6k94z 1/1 Running 0 16m 10.244.2.10 kb3 <none> <none> kubernetes-bootcamp-6bf84cb898-qswcf 1/1 Running 0 18m 10.244.2.9 kb3 <none> <none> $ kubectl get services NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE bootcamp-service LoadBalancer 10.108.87.251 <pending> 8080:31729/TCP 10m kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 10d
チュートリアルは replicas:4 でスタートするので スケールさせる
$ kubectl scale deployments/kubernetes-bootcamp --replicas=4 deployment.extensions/kubernetes-bootcamp scaled $ kubectl get pods -o wide NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES kubernetes-bootcamp-6bf84cb898-6k94z 1/1 Running 0 18m 10.244.2.10 kb3 <none> <none> kubernetes-bootcamp-6bf84cb898-gqms4 1/1 Running 0 4s 10.244.1.14 kb2 <none> <none> kubernetes-bootcamp-6bf84cb898-hnl46 1/1 Running 0 4s 10.244.1.13 kb2 <none> <none> kubernetes-bootcamp-6bf84cb898-qswcf 1/1 Running 0 21m 10.244.2.9 kb3 <none> <none>
Step 1: Update the version of the app
確認
$ kubectl get deployments NAME READY UP-TO-DATE AVAILABLE AGE kubernetes-bootcamp 4/4 4 4 22m $ kubectl get pods NAME READY STATUS RESTARTS AGE kubernetes-bootcamp-6bf84cb898-6k94z 1/1 Running 0 20m kubernetes-bootcamp-6bf84cb898-gqms4 1/1 Running 0 79s kubernetes-bootcamp-6bf84cb898-hnl46 1/1 Running 0 79s kubernetes-bootcamp-6bf84cb898-qswcf 1/1 Running 0 22m
pods 詳細確認。長いので2コ目以降は省略
$ kubectl describe pods
Name: kubernetes-bootcamp-6bf84cb898-6k94z
Namespace: default
Priority: 0
PriorityClassName: <none>
Node: kb3/192.168.0.103
Start Time: Tue, 05 Feb 2019 20:05:30 +0900
Labels: pod-template-hash=6bf84cb898
run=kubernetes-bootcamp
Annotations: cni.projectcalico.org/podIP: 10.244.2.10/32
Status: Running
IP: 10.244.2.10
Controlled By: ReplicaSet/kubernetes-bootcamp-6bf84cb898
Containers:
kubernetes-bootcamp:
Container ID: docker://81839f7a06215d94260e38103f735cd51625b555ec8242d24eecc5bf52ad6d05
Image: gcr.io/google-samples/kubernetes-bootcamp:v1
Image ID: docker-pullable://gcr.io/google-samples/kubernetes-bootcamp@sha256:0d6b8ee63bb57c5f5b6156f446b3bc3b3c143d233037f3a2f00e279c8fcc64af
Port: 8080/TCP
Host Port: 0/TCP
State: Running
Started: Tue, 05 Feb 2019 20:05:32 +0900
Ready: True
Restart Count: 0
Environment: <none>
Mounts:
/var/run/secrets/kubernetes.io/serviceaccount from default-token-88mmv (ro)
Conditions:
Type Status
Initialized True
Ready True
ContainersReady True
PodScheduled True
Volumes:
default-token-88mmv:
Type: Secret (a volume populated by a Secret)
SecretName: default-token-88mmv
Optional: false
QoS Class: BestEffort
Node-Selectors: <none>
Tolerations: node.kubernetes.io/not-ready:NoExecute for 300s
node.kubernetes.io/unreachable:NoExecute for 300s
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal Scheduled 20m default-scheduler Successfully assigned default/kubernetes-bootcamp-6bf84cb898-6k94z to kb3
Normal Pulled 20m kubelet, kb3 Container image "gcr.io/google-samples/kubernetes-bootcamp:v1" already present on machine
Normal Created 20m kubelet, kb3 Created container
Normal Started 20m kubelet, kb3 Started container
～略～
kubernetes-bootcamp のイメージを jocatalin/kubernetes-bootcamp:v2 にする。
$ kubectl set image deployments/kubernetes-bootcamp kubernetes-bootcamp=jocatalin/kubernetes-bootcamp:v2 deployment.extensions/kubernetes-bootcamp image updated
Step 2: Verify an update
イメージ更新を確認。v1 のpod がterminateして、v2のpodがrun してきた。
$ kubectl get pods NAME READY STATUS RESTARTS AGE kubernetes-bootcamp-5bf4d5689b-4gd7q 1/1 Running 0 36s kubernetes-bootcamp-5bf4d5689b-4xzxq 1/1 Running 0 31s kubernetes-bootcamp-5bf4d5689b-dsh7s 1/1 Running 0 36s kubernetes-bootcamp-5bf4d5689b-kpbvc 1/1 Running 0 31s kubernetes-bootcamp-6bf84cb898-6k94z 1/1 Terminating 0 22m kubernetes-bootcamp-6bf84cb898-gqms4 1/1 Terminating 0 3m40s kubernetes-bootcamp-6bf84cb898-hnl46 0/1 Terminating 0 3m40s kubernetes-bootcamp-6bf84cb898-qswcf 1/1 Terminating 0 25m $ kubectl get pods NAME READY STATUS RESTARTS AGE kubernetes-bootcamp-5bf4d5689b-4gd7q 1/1 Running 0 5m47s kubernetes-bootcamp-5bf4d5689b-4xzxq 1/1 Running 0 5m42s kubernetes-bootcamp-5bf4d5689b-dsh7s 1/1 Running 0 5m47s kubernetes-bootcamp-5bf4d5689b-kpbvc 1/1 Running 0 5m42s
deployments/kubernetes-bootcamp の詳細確認 image: jocatalin/kubernetes-bootcamp:v2 になっている。
$ kubectl describe deployments/kubernetes-bootcamp
Name: kubernetes-bootcamp
Namespace: default
CreationTimestamp: Tue, 05 Feb 2019 20:02:55 +0900
Labels: run=kubernetes-bootcamp
Annotations: deployment.kubernetes.io/revision: 2
Selector: run=kubernetes-bootcamp
Replicas: 4 desired | 4 updated | 4 total | 4 available | 0 unavailable
StrategyType: RollingUpdate
MinReadySeconds: 0
RollingUpdateStrategy: 25% max unavailable, 25% max surge
Pod Template:
Labels: run=kubernetes-bootcamp
Containers:
kubernetes-bootcamp:
Image: jocatalin/kubernetes-bootcamp:v2
Port: 8080/TCP
Host Port: 0/TCP
Environment: <none>
Mounts: <none>
Volumes: <none>
Conditions:
Type Status Reason
---- ------ ------
Available True MinimumReplicasAvailable
Progressing True NewReplicaSetAvailable
OldReplicaSets: <none>
NewReplicaSet: kubernetes-bootcamp-5bf4d5689b (4/4 replicas created)
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal ScalingReplicaSet 28m deployment-controller Scaled up replica set kubernetes-bootcamp-6bf84cb898 to 1
Normal ScalingReplicaSet 6m51s (x2 over 25m) deployment-controller Scaled up replica set kubernetes-bootcamp-6bf84cb898 to 4
Normal ScalingReplicaSet 3m47s deployment-controller Scaled up replica set kubernetes-bootcamp-5bf4d5689b to 1
Normal ScalingReplicaSet 3m47s deployment-controller Scaled down replica set kubernetes-bootcamp-6bf84cb898 to 3
Normal ScalingReplicaSet 3m47s deployment-controller Scaled up replica set kubernetes-bootcamp-5bf4d5689b to 2
Normal ScalingReplicaSet 3m42s (x2 over 15m) deployment-controller Scaled down replica set kubernetes-bootcamp-6bf84cb898 to 2
Normal ScalingReplicaSet 3m42s deployment-controller Scaled up replica set kubernetes-bootcamp-5bf4d5689b to 3
Normal ScalingReplicaSet 3m42s deployment-controller Scaled down replica set kubernetes-bootcamp-6bf84cb898 to 1
Normal ScalingReplicaSet 3m42s deployment-controller Scaled up replica set kubernetes-bootcamp-5bf4d5689b to 4
Normal ScalingReplicaSet 3m40s deployment-controller Scaled down replica set kubernetes-bootcamp-6bf84cb898 to 0
pods の詳細確認。ながいので略
$ kubectl describe pods
Name: kubernetes-bootcamp-5bf4d5689b-4gd7q
Namespace: default
Priority: 0
PriorityClassName: <none>
Node: kb2/192.168.0.102
Start Time: Tue, 05 Feb 2019 20:27:29 +0900
Labels: pod-template-hash=5bf4d5689b
run=kubernetes-bootcamp
Annotations: cni.projectcalico.org/podIP: 10.244.1.15/32
Status: Running
IP: 10.244.1.15
Controlled By: ReplicaSet/kubernetes-bootcamp-5bf4d5689b
Containers:
kubernetes-bootcamp:
Container ID: docker://1912a6a10c45f44ef1267ed7d585abed509fdc27ffcfcd8c8b85d22e087d87b6
Image: jocatalin/kubernetes-bootcamp:v2
Image ID: docker-pullable://jocatalin/kubernetes-bootcamp@sha256:fb1a3ced00cecfc1f83f18ab5cd14199e30adc1b49aa4244f5d65ad3f5feb2a5
Port: 8080/TCP
Host Port: 0/TCP
State: Running
Started: Tue, 05 Feb 2019 20:27:33 +0900
Ready: True
Restart Count: 0
Environment: <none>
Mounts:
/var/run/secrets/kubernetes.io/serviceaccount from default-token-88mmv (ro)
Conditions:
Type Status
Initialized True
Ready True
ContainersReady True
PodScheduled True
Volumes:
default-token-88mmv:
Type: Secret (a volume populated by a Secret)
SecretName: default-token-88mmv
Optional: false
QoS Class: BestEffort
Node-Selectors: <none>
Tolerations: node.kubernetes.io/not-ready:NoExecute for 300s
node.kubernetes.io/unreachable:NoExecute for 300s
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal Scheduled 5m1s default-scheduler Successfully assigned default/kubernetes-bootcamp-5bf4d5689b-4gd7q to kb2
Normal Pulling 5m kubelet, kb2 pulling image "jocatalin/kubernetes-bootcamp:v2"
Normal Pulled 4m57s kubelet, kb2 Successfully pulled image "jocatalin/kubernetes-bootcamp:v2"
Normal Created 4m57s kubelet, kb2 Created container
Normal Started 4m57s kubelet, kb2 Started container
～略～
curl 連打。各ポッドがお仕事しています。
$ curl http://10.108.87.251:8080 Hello Kubernetes bootcamp! | Running on: kubernetes-bootcamp-5bf4d5689b-dsh7s | v=2 $ curl http://10.108.87.251:8080 Hello Kubernetes bootcamp! | Running on: kubernetes-bootcamp-5bf4d5689b-4gd7q | v=2 $ curl http://10.108.87.251:8080 Hello Kubernetes bootcamp! | Running on: kubernetes-bootcamp-5bf4d5689b-kpbvc | v=2 $ curl http://10.108.87.251:8080 Hello Kubernetes bootcamp! | Running on: kubernetes-bootcamp-5bf4d5689b-4xzxq | v=2
ロールアウトステータスコマンドを実行しても、アップデートを確認できるそうだ。 successfully rolled out
$ kubectl rollout status deployments/kubernetes-bootcamp deployment "kubernetes-bootcamp" successfully rolled out
pods の詳細は略
Step 3: Rollback an update
チュートリアルやったときここ意味が分からなかった。
もう一度アップデートを実行して、v10としてタグ付けされたイメージをデプロイしましょう。
だそうです。
$ kubectl set image deployments/kubernetes-bootcamp kubernetes-bootcamp=gcr.io/google-samples/kubernetes-bootcamp:v10 deployment.extensions/kubernetes-bootcamp image updated
status:ImagePullBackOff <-> ErrImagePull を繰り返す。これは v10 のイメージがないから。
$ kubectl get pods NAME READY STATUS RESTARTS AGE kubernetes-bootcamp-597cfc5b76-flxjt 0/1 ImagePullBackOff 0 28s kubernetes-bootcamp-597cfc5b76-zrshs 0/1 ImagePullBackOff 0 28s kubernetes-bootcamp-5bf4d5689b-4gd7q 1/1 Running 0 17m kubernetes-bootcamp-5bf4d5689b-4xzxq 1/1 Running 0 17m kubernetes-bootcamp-5bf4d5689b-dsh7s 1/1 Running 0 17m kubernetes-bootcamp-5bf4d5689b-kpbvc 1/1 Terminating 0 17m $ kubectl get deployments NAME READY UP-TO-DATE AVAILABLE AGE kubernetes-bootcamp 3/4 2 3 42m @kb1 ~ $ kubectl get pods NAME READY STATUS RESTARTS AGE kubernetes-bootcamp-597cfc5b76-flxjt 0/1 ErrImagePull 0 39s kubernetes-bootcamp-597cfc5b76-zrshs 0/1 ErrImagePull 0 39s kubernetes-bootcamp-5bf4d5689b-4gd7q 1/1 Running 0 17m kubernetes-bootcamp-5bf4d5689b-4xzxq 1/1 Running 0 17m kubernetes-bootcamp-5bf4d5689b-dsh7s 1/1 Running 0 17m $ kubectl get pods NAME READY STATUS RESTARTS AGE kubernetes-bootcamp-597cfc5b76-flxjt 0/1 ImagePullBackOff 0 44s kubernetes-bootcamp-597cfc5b76-zrshs 0/1 ImagePullBackOff 0 44s kubernetes-bootcamp-5bf4d5689b-4gd7q 1/1 Running 0 17m kubernetes-bootcamp-5bf4d5689b-4xzxq 1/1 Running 0 17m kubernetes-bootcamp-5bf4d5689b-dsh7s 1/1 Running 0 17m $ kubectl get pods NAME READY STATUS RESTARTS AGE kubernetes-bootcamp-597cfc5b76-flxjt 0/1 ErrImagePull 0 64s kubernetes-bootcamp-597cfc5b76-zrshs 0/1 ErrImagePull 0 64s kubernetes-bootcamp-5bf4d5689b-4gd7q 1/1 Running 0 17m kubernetes-bootcamp-5bf4d5689b-4xzxq 1/1 Running 0 17m kubernetes-bootcamp-5bf4d5689b-dsh7s 1/1 Running 0 17m
rollout を undo 。なるほど rollout がここで生きてくると。
$ kubectl rollout undo deployments/kubernetes-bootcamp deployment.extensions/kubernetes-bootcamp rolled back
$ kubectl get pods NAME READY STATUS RESTARTS AGE kubernetes-bootcamp-5bf4d5689b-4gd7q 1/1 Running 0 21m kubernetes-bootcamp-5bf4d5689b-4xzxq 1/1 Running 0 21m kubernetes-bootcamp-5bf4d5689b-dsh7s 1/1 Running 0 21m kubernetes-bootcamp-5bf4d5689b-rtxj8 1/1 Running 0 5s
お掃除
$ kubectl delete services bootcamp-service service "bootcamp-service" deleted $ kubectl get services NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 10d $ kubectl delete deployment kubernetes-bootcamp deployment.extensions "kubernetes-bootcamp" deleted $ kubectl get deployment No resources found. $ kubectl get pods NAME READY STATUS RESTARTS AGE kubernetes-bootcamp-5bf4d5689b-4gd7q 1/1 Terminating 0 32m kubernetes-bootcamp-5bf4d5689b-4xzxq 1/1 Terminating 0 32m kubernetes-bootcamp-5bf4d5689b-dsh7s 1/1 Terminating 0 32m kubernetes-bootcamp-5bf4d5689b-rtxj8 1/1 Terminating 0 10m $ kubectl get pods No resources found.
