LFD259 Kubernetes开发者实验手册 (2019版)

需积分: 15 123 浏览量更新于2024-07-15 收藏 468KB PDF 举报

LFD259-labs_V2019-01-14 官方实验手册是一份由Linux Foundation（LF）发布的针对开发者的 Kubernetes 教程资料。该手册版本于2019年1月14日更新，包含了丰富的实践经验，旨在帮助开发者理解和掌握 Kubernetes 这一容器编排平台。Linux Foundation 保留了所有版权，并强调了这些培训材料的使用限制：参与者只能个人使用，不得复制、修改或分发给非参与者，也禁止用作他人的培训工具，除非得到 Linux Foundation 的书面许可。手册中包含开源代码，这些代码可能由多个版权持有者提供，并遵循相应的开源许可证。这确保了知识产权的合规性，同时也鼓励创新和社区协作。Linux Foundation 提供的培训材料仅限于学员在课程中的使用，任何未经授权的复制、传播或商业用途都是被禁止的。值得注意的是，Linux Foundation 不对本手册的内容或使用效果做出任何明确或隐含的保证，包括但不限于商品质量保证。这份出版物的完整性和准确性依赖于发布时的技术状态，可能会随着技术的发展而更新。此外，手册的发行者是 Linux Foundation，访问其官方网站（<https://www.linuxfoundation.org>）可以获取更多关于 Linux 和 Kubernetes 相关的信息和服务。 LFD259-labs_V2019-01-14 官方实验手册是开发者学习 Kubernetes 的重要资源，提供了实践操作的机会，对于理解和应用容器化技术具有很高的价值。然而，使用时务必遵守版权和许可协议，尊重知识产权。

10 CHAPTER 2. KUBERNETES ARCHITECTURE

container and verify it is running. In the Deployment Conﬁguration chapter we will continue to work on this pod by

adding persistent storage and conﬁgure ﬂuentd via a configMap.

Edit the YAML ﬁle and add a ﬂuentd container. The dash should line up with the previous container dash. At this point

a name and image should be enough to start the second container.

student@ckad-1:~$ vim basic.yaml

basic.yaml

1 ....

2 containers:

3 - name: webcont

4 image: nginx

5 ports:

6 - containerPort: 80

7 - name: fdlogger

8 image: k8s.gcr.io/fluentd-gcp:1.30

2. Delete and create the pod again. The commands can be typed on a single line, separated by a semicolon. The backslash

is only used to ﬁt on the page. This time you should see 2/2 under the READY column. You should also ﬁnd information

on the ﬂuentd container inside of the kubectl describe output.

student@ckad-1:~$ kubectl delete pod basicpod ; \

kubectl create -f basic.yaml

pod "basicpod" deleted

pod/basicpod created

student@ckad-1:~$ kubectl get pod

NAME READY STATUS RESTARTS AGE

basicpod 2/2 Running 0 2m8s

student@ckad-1:~$ kubectl describe pod basicpod

Name: basicpod

Namespace: default

Priority: 0

PriorityClassName: <none>

<output_omitted>

3. For now shut down the pod. We will use it again in a future exercise.

student@ckad-1:~$ kubectl delete pod basicpod

pod "basicpod" deleted

Exercise 2.5: Create a Simple Deployment

Creating a pod does not take advantage of orchestration abilities of Kubernetes. We will now create a Deployment

which gives use scalability reliability and updates.

1. Now run a containerized webserver nginx. Use kubectl create to create a simple, single replica deployment running

the nginx web server. It will create a single pod as we did previously but with new controllers to ensure it runs as well as

other features.

student@ckad-1:~$ kubectl create deployment firstpod --image=nginx

deployment.apps/firstpod created

2. Verify the new deployment exists and the desired number of pods matches the current number. Using a comma, you

can request two resource types at once. The Tab key can be helpful. Type enough of the word to be unique and press

the Tab key, it should complete the word. The deployment should show a number 1 for each value, such that the desired

number of pods matches the up-to-date and running number. The pod should show zero restarts.

student@ckad-1:~$ kubectl get deployment,pod

NAME READY UP-TO-DATE AVAILABLE AGE

deployment.extensions/firstpod 1/1 1 1 2m42s

NAME READY STATUS RESTARTS AGE

pod/firstpod-7d88d7b6cf-lrsbk 1/1 Running 0 2m42s

3. View the details of the deployment, then the pod. Work through the output slowly. Knowing what a healthy deployment

and looks like can be helpful when troubleshooting issues. Again the Tab key can be helpful when using long auto-

generated object names. You should be able to type ﬁrstpodTab and the name will complete when viewing the pod.

student@ckad-1:~$ kubectl describe deployment firstpod

Name: firstpod

Namespace: default

CreationTimestamp: Fri, 25 Jul 2018 16:46:57 +0000

Labels: run=firstpod

Annotations: deployment.kubernetes.io/revision=1

Selector: run=firstpod

Replicas: 1 desired | 1 updated | 1 total | 1 available....

StrategyType: RollingUpdate

MinReadySeconds: 0

<output_omitted>

student@ckad-1:~$ kubectl describe pod firstpod-6bb4574d94-rqk76

Name: firstpod-6bb4574d94-rqk76

Namespace: default

Priority: 0

PriorityClassName: <none>

Node: ckad-1/10.128.0.2

Start Time: Wed, 25 Jul 2018 06:13:18 +0000

Labels: pod-template-hash=2660130850

run=firstpod

Annotations: <none>

Status: Running

IP: 192.168.200.65

Controlled By: ReplicaSet/firstpod-6bb4574d94

<output_omitted>

4. Note that the resources are in the default namespace. Get a list of available namespaces.

student@ckad-1:~$ kubectl get namespaces

NAME STATUS AGE

default Active 20m

kube-public Active 20m

kube-system Active 20m

5. There are two other namespaces. Look at the pods in the kube-system namespace.

student@ckad-1:~$ kubectl get pod -n kube-system

NAME READY STATUS RESTARTS AGE

calico-etcd-rvrpk 1/1 Running 1 20m

calico-kube-controllers-d554689d5-lm687 1/1 Running 1 20m

calico-node-2ck9g 2/2 Running 4 19m

calico-node-kkxvl 2/2 Running 3 20m

etcd-ckad-1 1/1 Running 1 20m

<output_omitted>

6. Now look at the pods in a namespace that does not exist. Note you do not receive an error.

student@ckad-1:~$ kubectl get pod -n fakenamespace

No resources found.

12 CHAPTER 2. KUBERNETES ARCHITECTURE

7. You can also view resources in all namespaces at once. Use the --all-namespaces options to select objects in all

namespaces at once.

student@ckad-1:~$ kubectl get pod --all-namespaces

NAMESPACE NAME READY STATUS RESTARTS AGE

default firstpod-6bb4574d94-rqk76 1/1 Running 0 5m

kube-system calico-etcd-z49kx 1/1 Running 0 31m

kube-system calico-kube-controllers-d554689d5-pfszw 1/1 Running 0 31m

<output_omitted>

8. View several resources at once. Note that most resources have a short name such as rs for ReplicaSet, po for Pod,

svc for Service, and ep for endpoint.

student@ckad-1:~$ kubectl get deploy,rs,po,svc,ep

NAME READY UP-TO-DATE AVAILABLE AGE

deployment.extensions/firstpod 1/1 1 1 4m

NAME DESIRED CURRENT READY....

replicaset.extensions/firstpod-6bb4574d94-rqk76 1 1 1 ....

NAME READY STATUS RESTARTS AGE

pod/firstpod-6bb4574d94-rqk76 1/1 Running 0 4m

NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE

service/basicservice NodePort 10.108.147.76 <none> 80:31601/TCP 21m

service/kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 21m

NAME ENDPOINTS AGE

endpoints/basicservice <none> 21m

endpoints/kubernetes 10.128.0.3:6443 21m

9. Delete the ReplicaSet and view the resources again. Note that the age on the ReplicaSet and the pod it controls is

now less than a minute. The deployment controller started a new ReplicaSet when we deleted the existing one, which

started another pod when the desired conﬁguration did not match the current status.

student@ckad-1:~$ kubectl delete rs firstpod-6bb4574d94-rqk76

replicaset.extensions "firstpod-6bb4574d94-rqk76" deleted

student@ckad-1:~$ kubectl get deployment,rs,po,svc,ep

NAME READY UP-TO-DATE AVAILABLE AGE

deployment.extensions/firstpod 1/1 1 1 7m

NAME DESIRED CURRENT....

replicaset.extensions/firstpod-6bb4574d94-rqk76 1 1 ....

NAME READY STATUS RESTARTS AGE

pod/firstpod-7d99ffc75-p9hbw 1/1 Running 0 12s

NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE

service/kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 24m

NAME ENDPOINTS AGE

endpoints/kubernetes 10.128.0.2:6443 80m

endpoints/basicservice <none> 21m

10. This time delete the top-level controller. After about 30 seconds for everything to shut down you should only see the

cluster service and endpoint remain for the cluster and the service we created.

student@ckad-1:~$ kubectl delete deployment firstpod

deployment.extensions "firstpod" deleted

student@ckad-1:~$ kubectl get deployment,rs,po,svc,ep

NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE

service/basicservice NodePort 10.108.147.76 <none> 80:31601/TCP 35m

剩余85页未读，继续阅读

gaoyidong

粉丝: 2
资源: 29

LFD259 Kubernetes开发者实验手册 (2019版)

LFD259-Labs_V2018-10-21.官方实验手册

LFD259-labs_V2019-06-27 官方实验手册

如何下载pycuda-2022.1+cuda116-cp38-cp38-win_amd64.whl

opencv_python-4.5.3.56-cp38-cp38-win_amd64.whl下载地址

ERROR: pygame-2.5.2-cp310-cp310-win_amd64.whl is not a supported wheel on this platform.

GDAL-3.4.3-cp39-cp39-win_amd64.whl下载

pycuda-2022.1 (3)没有pycuda-2022.1+cuda116-cp38-cp38-win_amd64.whl这个文件

python3.7安装pyqt5报错 Could not build wheels for PyQt5-sip, which is required to install pyproject.toml-based projects

能不能提供预编译好的pysqlcipher3安装包，通过本地方式进行安装，如果没有下载地址，那应该如何找到靠谱的下载链接

怎么下载scikit-learn

最新资源