# 5. Pod Design (20%) ### 1. Defining and Querying Labels and Annotations 1. Create three different Pods with the names `frontend`, `backend` and `database` that use the image `nginx`. 2. Declare labels for those Pods as follows: ``` frontend: env=prod, team=shiny backend: env=prod, team=legacy, app=v1.2.4 database: env=prod, team=storage ``` 1. Declare annotations for those Pods as follows: ``` frontend: contact=John Doe, commit=2d3mg3 backend: contact=Mary Harris ``` 1. Render the list of all Pods and their labels. 2. Use label selectors on the command line to query for all production Pods that belong to the teams shiny and legacy. 3. Remove the label env from the backend Pod and rerun the selection. 4. Render the surrounding 3 lines of YAML of all Pods that have\ annotations. You can assign labels upon Pod creation with the --labels option. ``` $ kubectl run frontend --image=nginx --restart=Never --labels=env=prod,team=shiny pod/frontend created $ kubectl run backend --image=nginx --restart=Never --labels=env=prod,team=legacy,app=v1.2.4 pod/backend created $ kubectl run database --image=nginx --restart=Never --labels=env=prod,team=storage pod/database created ``` Edit the existing Pods with the edit command and add the annotations as follows: ``` apiVersion: v1 kind: Pod metadata: annotations: commit: 2d3mg3 contact: John Doe name: frontend ... apiVersion: v1 kind: Pod metadata: annotations: contact: 'Mary Harris' name: backend ... ``` Render all Pods and their Pods including their assigned labels. ``` $ kubectl get pods --show-labels NAME READY STATUS RESTARTS AGE LABELS backend 1/1 Running 0 41s app=v1.2.4,env=prod,team=legacy database 1/1 Running 0 8s env=prod,team=storage frontend 1/1 Running 0 1m env=prod,team=shiny ``` You can combine the selector rules into one expression. ``` $ kubectl get pods -l 'team in (shiny, legacy)',env=prod --show-labels NAME READY STATUS RESTARTS AGE LABELS backend 1/1 Running 0 19m app=v1.2.4,env=prod,team=legacy frontend 1/1 Running 0 20m env=prod,team=shiny ``` You can add and remove labels with the label command. The selection now doesn’t match for the backend Pod anymore. ``` $ kubectl label pods backend env- pod/backend labeled $ kubectl get pods -l 'team in (shiny, legacy)',env==prod --show-labels NAME READY STATUS RESTARTS AGE LABELS frontend 1/1 Running 0 23m env=prod,team=shiny ``` The grep command can help with rendering any YAML code around the identified search term. ``` $ kubectl get pods -o yaml | grep -C 3 'annotations:' apiVersion: v1 kind: Pod metadata: annotations: cni.projectcalico.org/podIP: 192.168.60.163/32 contact: Mary Harris creationTimestamp: 2019-05-10T17:57:38Z -- -- apiVersion: v1 kind: Pod metadata: annotations: cni.projectcalico.org/podIP: 192.168.60.147/32 creationTimestamp: 2019-05-10T17:58:11Z labels: -- -- apiVersion: v1 kind: Pod metadata: annotations: cni.projectcalico.org/podIP: 192.168.60.159/32 commit: 2d3mg3 contact: John Doe ``` ### 2. Performing Rolling Updates for a Deployment 1. Create a Deployment named `deploy` with `3 replicas.` The Pods should use the `nginx` image and the name `nginx`. The Deployment uses the label `tier=backend`. The Pods should use the label `app=v1`. 2. List the Deployment and ensure that the correct number of replicas\ is running. 3. Update the image to nginx:latest. 4. Verify that the change has been rolled out to all replicas. 5. Scale the Deployment to 5 replicas. 6. Have a look at the Deployment rollout history. 7. Revert the Deployment to revision 1. 8. Ensure that the Pods use the image nginx. Generate the YAML for a Deployment plus Pod for further editing. ``` $ kubectl create deployment deploy --image=nginx --dry-run -o yaml > deploy.yaml ``` Edit the labels. The selector should match the `labels of the Pods`. Change the replicas from `1 to 3`. ``` apiVersion: apps/v1 kind: Deployment metadata: creationTimestamp: null labels: tier: backend name: deploy spec: replicas: 1 selector: matchLabels: app: v1 strategy: {} template: metadata: creationTimestamp: null labels: app: v1 spec: containers: - image: nginx name: nginx resources: {} status: {} ``` Create the deployment by pointing it to the YAML file. ``` $ kubectl create -f deploy.yaml deployment.apps/deploy created $ kubectl get deployments NAME DESIRED CURRENT UP-TO-DATE AVAILABLE AGE deploy 3 3 3 1 4s ``` Set the new image and check the revision history. ``` $ kubectl set image deployment/deploy nginx=nginx:latest deployment.extensions/deploy image updated $ kubectl rollout history deploy deployment.extensions/deploy REVISION CHANGE-CAUSE 1 2 $ kubectl rollout history deploy --revision=2 deployment.extensions/deploy with revision #2 Pod Template: Labels: app=v1 pod-template-hash=1370799740 Containers: nginx: Image: nginx:latest Port: Host Port: Environment: Mounts: Volumes: ``` Now scale the Deployment to `5 replicas`. ``` $ kubectl scale deployments deploy --replicas=5 deployment.extensions/deploy scaled ``` Roll back to `revision 1`. You will see the new revision. Inspecting the revision should show the image nginx. ``` $ kubectl rollout undo deployment/deploy --to-revision=1 deployment.extensions/deploy $ kubectl rollout history deploy deployment.extensions/deploy REVISION CHANGE-CAUSE 2 3 $ kubectl rollout history deploy --revision=3 deployment.extensions/deploy with revision #3 Pod Template: Labels: app=v1 pod-template-hash=454670702 Containers: nginx: Image: nginx Port: Host Port: Environment: Mounts: Volumes: ``` ### 3. Creating a Scheduled Container Operation 1. Create a `CronJob` named `current-date` that runs `every minute` and executes the shell command `echo "Current date: $(date)"`. 2. Watch the jobs as they are being scheduled. 3. Identify one of the Pods that ran the CronJob and render the logs. 4. Determine the number of successful executions the CronJob will keep\ in its history. Delete the Job. 答案：\ The run command is deprecated but it provides a good shortcut for creating a CronJob with a single command.\ **注意**：**此命令kubectl v1.8并不会显示创建cronjob成功，仅是pod** ``` $ kubectl run current-date --schedule="* * * * *" --restart=OnFailure --image=nginx -- /bin/sh -c 'echo "Current date: $(date)"' kubectl run --generator=cronjob/v1beta1 is DEPRECATED and will be removed in a future version. Use kubectl create instead. cronjob.batch/hello created ``` **正确命令**： ``` kubectl create cronjob current-update --image=nginx --schedule="* * * * *" --restart=OnFailure -- /bin/sh -c 'echo "Current date: $(date)"' ``` Watch the Jobs as they are executed. ``` $ kubectl get cronjobs --watch NAME COMPLETIONS DURATION AGE current-date-1557522540 1/1 3s 103s current-date-1557522600 1/1 4s 43s ``` Identify one of the Pods (the label indicates the Job name) and render its logs. ``` $ kubectl get pods --show-labels NAME READY STATUS RESTARTS AGE LABELS current-date-1557522540-dp8l9 0/1 Completed 0 1m controller-uid=3aaabf96-7369-11e9-96c6-025000000001,job-name=current-date-1557523140,run=current-date $ kubectl logs current-date-1557522540-dp8l9 Current date: Fri May 10 21:09:12 UTC 2019 ``` The value of the attribute successfulJobsHistoryLimit defines how many executions are kept in the history. ``` $ kubectl get cronjobs current-date -o yaml | grep successfulJobsHistoryLimit: successfulJobsHistoryLimit: 3 ``` Finally, delete the CronJob. ``` $ kubectl delete cronjob current-date ``` --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://ghostwritten.gitbook.io/kubernetes-exam-in-action/ckad-kao-shi/5.-pod-design-20.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.