How can i scale a deployment using an YAML file?

Question

I don't understand about kubectl.

I'm trying to scale a pod using Cronjob in Huawei Cloud. I've resolved how to scale then using Linux Terminal and it works, but when i try to use the kubectl, it brings me the following error: `error:

I have the following YAML code from a kubectl in Huawei Cloud:

kind: CronJob
apiVersion: batch/v1
metadata:
  name: resize-app01-test
  namespace: my-namespace-hmg
  uid: <my-private-uid>
  resourceVersion: '<xxx>'
  generation: 19
  creationTimestamp: '2024-07-30T13:38:49Z'
  annotations:
    description: ''
  managedFields:
    - manager: cfe-apiserver
      operation: Update
      apiVersion: batch/v1
      time: '2024-07-30T16:41:37Z'
      fieldsType: FieldsV1
      fieldsV1:
        f:metadata:
          f:annotations:
            .: {}
            f:description: {}
        f:spec:
          f:concurrencyPolicy: {}
          f:failedJobsHistoryLimit: {}
          f:jobTemplate:
            f:spec:
              f:template:
                f:metadata:
                  f:labels:
                    .: {}
                    f:app: {}
                    f:version: {}
                f:spec:
                  f:containers:
                    k:{"name":"container-teste"}:
                      .: {}
                      f:args: {}
                      f:command: {}
                      f:env:
                        .: {}
                        k:{"name":"PAAS_APP_NAME"}:
                          .: {}
                          f:name: {}
                          f:value: {}
                        k:{"name":"PAAS_NAMESPACE"}:
                          .: {}
                          f:name: {}
                          f:value: {}
                        k:{"name":"PAAS_PROJECT_ID"}:
                          .: {}
                          f:name: {}
                          f:value: {}
                      f:image: {}
                      f:imagePullPolicy: {}
                      f:name: {}
                      f:resources:
                        .: {}
                        f:limits:
                          .: {}
                          f:cpu: {}
                          f:memory: {}
                        f:requests:
                          .: {}
                          f:cpu: {}
                          f:memory: {}
                      f:terminationMessagePath: {}
                      f:terminationMessagePolicy: {}
                  f:dnsConfig:
                    .: {}
                    f:options: {}
                  f:dnsPolicy: {}
                  f:imagePullSecrets:
                    .: {}
                    k:{"name":"default-secret"}: {}
                  f:restartPolicy: {}
                  f:schedulerName: {}
                  f:securityContext: {}
                  f:terminationGracePeriodSeconds: {}
          f:schedule: {}
          f:successfulJobsHistoryLimit: {}
          f:suspend: {}
          f:timeZone: {}
    - manager: kube-controller-manager
      operation: Update
      apiVersion: batch/v1
      time: '2024-07-30T16:43:38Z'
      fieldsType: FieldsV1
      fieldsV1:
        f:status:
          f:lastScheduleTime: {}
          f:lastSuccessfulTime: {}
      subresource: status
spec:
  schedule: "43 13 * * *"
  timeZone: America/Sao_Paulo
  concurrencyPolicy: Forbid
  suspend: false
  jobTemplate:
    metadata:
      creationTimestamp: null
    spec:
      template:
        metadata:
          creationTimestamp: null
          labels:
            app: resize-app01-teste
            version: v1
        spec:
          containers:
            - name: container-teste
              image: bitnami/kubectl:latest
              command:
                - /bin/sh
              args:
                - '-c'
                - kubectl scale --replicas=3 deployment/workload-v2-app01-hmg -n app01-hmg
              env:
                - name: PAAS_APP_NAME
                  value: resize-app01-teste
                - name: PAAS_NAMESPACE
                  value: app01-hmg
                - name: PAAS_PROJECT_ID
                  value: <my-project-id>
              resources:
                limits:
                  cpu: 250m
                  memory: 512Mi
                requests:
                  cpu: 250m
                  memory: 512Mi
              terminationMessagePath: /dev/termination-log
              terminationMessagePolicy: File
              imagePullPolicy: IfNotPresent
          restartPolicy: OnFailure
          terminationGracePeriodSeconds: 60
          dnsPolicy: ClusterFirst
          securityContext: {}
          imagePullSecrets:
            - name: default-secret
          schedulerName: default-scheduler
          dnsConfig:
            options:
              - name: single-request-reopen
  successfulJobsHistoryLimit: 3
  failedJobsHistoryLimit: 1
status:
  lastScheduleTime: '2024-07-30T16:43:00Z'
  lastSuccessfulTime: '2024-07-30T13:40:03Z'

when the Cronjob is executed, the console brings me this error:

Error - no objects passed to scale

How can i resolve this?

I've tried to split the command: and args: like that ways:

1° try:

command:
- /bin/sh
- '-c'
- kubectl
- scale --replicas=3 
- deployment/workload-v2-app01-hmg -n app01-hmg

2° try:

command:
- /bin/sh
args:
- '-c'
- kubectl
- scale --replicas=3 
- deployment/workload-v2-app01-hmg -n app01-hmg

3° try:

command:
- /bin/sh
args:
- '-c'
- kubectl scale --replicas=3 deployment/workload-v2-app01-hmg -n app01-hmg

Answer 1

This isn't directly an answer, but your 1st and 2nd tries are doomed to failure because of how you've structured them. When you're using sh -c '...some script...', note that the -c option only takes a single argument. That means that when you do this:

command:
- /bin/sh
- '-c'
- kubectl
- scale --replicas=3 
- deployment/workload-v2-app01-hmg -n app01-hmg

Or this:

command:
- /bin/sh
args:
- '-c'
- kubectl
- scale --replicas=3 
- deployment/workload-v2-app01-hmg -n app01-hmg

In both cases the argument to sh -c is kubectl, and nothing else. The remaining arguments become positional parameters to the script passed to -c.

It's critical to understand this, because you're going to run into similar situations many times as you work with kubernetes.

The canonical way of embedding a shell script in a pod/deployment is to do something like this is:

command:
- /bin/sh
- -c
- |
  echo your shell script
  echo goes here

Using the | literal block quote operator allows you to format a shell script as you normally would.

---

Your third try appears correct, assuming that your code is running under a serviceaccount with appropriate privileges.