Console

To create a new job:

1. In the Google Cloud console, go to the Cloud Run page:

   Go to Cloud Run

2. SelectJobs from the Cloud Run navigation menu, and clickDeploy container to display theCreate job form.

   1. In the form, specify the container image containing the job code orselect from a list of containers previously deployed.
   2. The job name is automatically generated from the container image. Youcan edit or change the job name as needed. A job name cannot bechanged after the job is created.
   3. Select theregion where you want your job located.The region selector highlights regions withthelowest carbon impact.
   4. Specify the number of tasks that you want to run in the job. All ofthe tasks must succeed for the job to succeed. By default, the tasksexecute in parallel.

3. ClickContainer(s), Volumes, Networking, Security to set additional job properties.

   • Under Task capacity:
   1. In theMemory menu, specify the amount of memory required.The default is the minimum required, 512MiB.
   2. In theCPU menu, specify the amount of CPU required. Thedefault is the minimum required, 1 CPU.

   3. UnderTask timeout, specify the maximum amount of time in secondsthat the task can run, up to 168 hours (7 days). For tasksusing GPUs, the maximum available timeout is1 hour. Each task must complete withinthis time. The default is 10 minutes.

   4. UnderNumber of retries per failed task, specify the number ofretries in case of task failures. The defaultis 3 retries.

   • Under Parallelism:

     1. In most cases you can selectRun as many tasks concurrently as possible.
     2. If you need to set a lower limit due to scaling constraints onresources your job accesses, selectLimit the maximum number of concurrent tasksand specify the number of concurrent tasks in theCustom parallelism limitfield.

4. Optionally, configure other settings in the appropriate tabs:

   • Container configuration
   • CPU limits
   • Memory limits
   • Secrets
   • Environment variables
   • Labels
   • Service accounts
   • Cloud SQL connections
   • VPC connectivity

5. When you are finished configuring your job, clickCreate tocreate the job in Cloud Run.

6. To execute the job, seeexecute jobs orexecute jobs on a schedule.

gcloud

To use the command line, you need to have alreadyset up the gcloud CLI.

To create a new job:

1. Run the command:

   gcloudrunjobscreateJOB_NAME--imageIMAGE_URLOPTIONS

   Alternatively, use the deploy command:

   gcloudrunjobsdeployJOB_NAME--imageIMAGE_URLOPTIONS

   Replace the following:

   • JOB_NAME: the name of the job youwant to create. If you omit this parameter, you will be prompted for thejob name when you run the command.
   • IMAGE_URL: a reference to thecontainer image—forexample,us-docker.pkg.dev/cloudrun/container/job:latest.

   • Optionally, replaceOPTIONSwith any of the following options:

     Option	Description
     --tasks	Accepts integers greater or equal to 1. Defaults to 1; maximum is 10,000. Each task is provided the environment variablesCLOUD_RUN_TASK_INDEX with a value between 0 and the number of tasks minus 1, along withCLOUD_RUN_TASK_COUNT, which is the number of tasks
     --max-retries	The number of times a failed task is retried. Once any task fails beyond this limit, the entire job is marked asfailed. For example, if set to 1, a failed task will be retried once, for a total of two attempts. The default is 3. Accepts integers from 0 to 10.
     --task-timeout	Accepts a duration like "2s". Defaults to 10 minutes; maximum is 168 hours (7 days). For tasks using GPUs, the maximum available timeout is 1 hour.
     --parallelism	The maximum number of tasks that can execute in parallel. By default, tasks will be started as quickly as possible in parallel. Refer toParallelism for the range of values.
     --execute-now	If set, immediately after the job is created, ajob execution is started. Equivalent to callinggcloud run jobs create followed bygcloud run jobs execute.

     In addition to the preceding options, you also specify moreconfiguration such as environment variables or memory limits.

     For a full list of available options when creating a job, refer to thegcloud run jobs createcommand line documentation.

2. Wait for the job creation to finish. You'll see a success message upona successful completion.

3. To execute the job, seeexecute jobs orexecute jobs on a schedule.

YAML

You can store your job specification in aYAML file and then deploy itusing the gcloud CLI.

1. Create a newjob.yaml file with this content:

   apiVersion:run.googleapis.com/v1kind:Jobmetadata:name:JOBspec:template:spec:template:spec:containers:-image:IMAGE_URL

   Replace the following:

   • JOB: the name of your Cloud Run job.Job names must be 49 charactersor less and must be unique per region and project.
   • IMAGE_URL: a reference to thecontainer image—forexample,us-docker.pkg.dev/cloudrun/container/job:latest.

   You can also specify more configuration such as environment variables ormemory limits.

2. Deploy the new job using the following command:

   gcloudrunjobsreplacejob.yaml

Terraform

To learn how to apply or remove a Terraform configuration, seeBasic Terraform commands.

resource"google_cloud_run_v2_job""default"{name="cloud-run-job"location="us-central1"deletion_protection=false # set to "true" in productiontemplate{template{containers{image="us-docker.pkg.dev/cloudrun/container/job:latest"}}}}

Client libraries

To create a job from code:

• Go
• Java
• Node.js
• Python
• Ruby
• PHP
• .NET

REST API

To create a job, send aPOST HTTP request to request tothe Cloud Run Admin APIjobs endpoint.

For example, usingcurl:

curl-H"Content-Type: application/json"\-H"Authorization: BearerACCESS_TOKEN"\-XPOST\-d'{template: {template: {containers: [{image: "IMAGE_URL"}]}}}'\https://run.googleapis.com/v2/projects/PROJECT_ID/locations/REGION/jobs?jobId=JOB_NAME

Replace the following:

• ACCESS_TOKEN: a valid access token for an account thathas theIAM permissions to create jobs.For example, if you are logged into gcloud, you can retrieve anaccess token usinggcloud auth print-access-token.From within a Cloud Run container instance, you can retrievean access token using thecontainer instance metadata server.
• JOB_NAME: the name of the job you want to create.
• IMAGE_URL: a reference to thecontainer image—forexample,us-docker.pkg.dev/cloudrun/container/job:latest.
• REGION: the Google Cloud region of the job.
• PROJECT_ID: the Google Cloud project ID.

Console

To update an existing job:

1. In the Google Cloud console, go to the Cloud Run page:

   Go to Cloud Run

2. ClickJobs from the Cloud Run navigation menu to display the list of jobs.

3. Click the job to display theJob details page.

4. ClickEdit.

5. If you made changes to your job code, specify the new containerimage digest.

6. Optionally, change the number of tasks that are in the job if needed.

7. Optionally, clickContainer(s), Volumes, Networking, Security to update any additionaljob properties:

   • Under Task capacity:
   1. In theMemory menu, specify the amount of memory required.The default is the minimum required, 512MiB.
   2. In theCPU menu, specify the amount of CPU required. Thedefault is the minimum required, 1 CPU.
   3. UnderTask timeout, specify the maximum amount of time in secondsthat the task can run, up to 168 hours (7 days). Each taskmust complete within this time. The default is 10 minutes (600 seconds).
   4. UnderNumber of retries per failed task, specify the number ofretries in case of task failures. The default is3 retries.

   • Under Parallelism:

     1. In most cases you can selectRun as many tasks concurrently as possible.
     2. If you need to set a lower limit due to scaling constraints onresources your job accesses, selectLimit the number of concurrent tasks and specify the maximumnumber of concurrent tasks in theCustom parallelism limit field.

8. Optionally, configure other settings in the appropriate tabs:

   • Container configuration
   • CPU limits
   • Memory limits
   • Secrets
   • Environment variables
   • Labels
   • Service accounts
   • Cloud SQL connections
   • VPC connectivity

9. When you are finished configuring your job, clickSave tocreate the job in Cloud Run and wait for the job creation to finish.

10. To execute the job, seeexecute jobs orexecute jobs on a schedule.

gcloud

1. In the Google Cloud console, activate Cloud Shell.

   Activate Cloud Shell

   At the bottom of the Google Cloud console, aCloud Shell session starts and displays a command-line prompt. Cloud Shell is a shell environment with the Google Cloud CLI already installed and with values already set for your current project. It can take a few seconds for the session to initialize.

2. Run the command:

   gcloudrunjobsupdateJOB_NAME

   Replace the following:

   • JOB_NAME: the name of the job youwant to update.

   • Optionally, replaceOPTIONS with thefollowing options:

     Option	Description
     --tasks	Accepts integers equal or greater than 1. Defaults to 1; maximum is 10,000. Each task is provided the environment variablesCLOUD_RUN_TASK_INDEX with a value between 0 and the number of tasks minus 1, along withCLOUD_RUN_TASK_COUNT, which is the number of tasks
     --max-retries	The number of times a failed task is retried. Once any task fails beyond this limit, the entire job is marked asfailed. For example, if set to 1, a failed task will be retried once, for a total of two attempts. The default is3. Accepts integers from 0 to 10.
     --task-timeout	Accepts a duration like "2s". Defaults to 10 minutes; maximum is 168 hours (7 days).
     --parallelism	The maximum number of tasks that can execute in parallel. By default, tasks will be started as quickly as possible, in parallel. Refer toParallelism for the range of values.

   In addition to the previous options, you can set other optional configurationsettings:

   • Container configuration
   • CPU limits
   • Memory limits
   • Secrets
   • Environment variables
   • Labels
   • Service accounts
   • Cloud SQL connections
   • VPC connection

   For a full list of available options when creating a job, refer to thegcloud run jobs createcommand line documentation.

3. Wait for the job update to finish. Upon successful completion, a successmessage is displayed, similar to the following:

   Job[JOB_NAME]hasbeensuccessfullyupdated.Viewdetailsaboutthisjobbyrunning`gcloudrunjobsdescribeJOB_NAME`.Seelogsforthisexecutionat:https://console.cloud.google.com/logs/viewer?project=PROJECT_ID&resource=cloud_run_revision/service_name/JOB_NAME

4. To execute the job, seeexecute jobs orexecute jobs on a schedule.

YAML

gcloudrunjobsdescribeJOB--formatexport>job.yaml

1. Update the existing job configuration:

   gcloudrunjobsreplacejob.yaml

2. To execute the job, seeexecute jobs orexecute jobs on a schedule.

Terraform

Make changes to your job configuration in yourmain.tf file using theterraform apply command. Detailed Terraform instructions are available for:

• Parallelism
• Max retries
• Timeouts

For more information, refer to theterraform apply command line options.

Client libraries

To update an existing job from code:

• Go
• Java
• Node.js
• Python
• Ruby
• PHP
• .NET

REST API

To update a job, send aPATCH HTTP request to request tothe Cloud Run Admin APIjobs endpoint.

For example, usingcurl:

curl-H"Content-Type: application/json"\-H"Authorization: BearerACCESS_TOKEN"\-XPATCH\-d'{template: {template: {containers: [{image: "IMAGE_URL"}]}}}'\https://run.googleapis.com/v2/projects/PROJECT_ID/locations/REGION/jobs/JOB_NAME

Replace the following:

• ACCESS_TOKEN: a valid access token for an account thathas theIAM permissions to update jobs.For example, if you are logged into gcloud, you can retrieve anaccess token usinggcloud auth print-access-token.From within a Cloud Run container instance, you can retrievean access token using thecontainer instance metadata server.
• JOB_NAME: the name of your job you want to update.
• IMAGE_URL: a reference to thecontainer image—forexample,us-docker.pkg.dev/cloudrun/container/job:latest.
• REGION: the Google Cloud region of the job.
• PROJECT_ID: the Google Cloud project ID.

Console

1. In the Google Cloud console, go to the Cloud Run page:

   Go to Cloud Run

2. To deploy to an existing job, clickJobs, locate the job in the jobs list, and click to open, then clickView and edit configuration to display the edit job form.

3. For a new job, clickDeploy container, then clickJob. Supply the job name and the URL to the main job container image you want to deploy.

4. ClickContainer(s), Volumes, Networking, Security

5. In theEdit container card, configure the main job container as needed.

6. ClickAdd container and configure a sidecar container you want toadd alongside the main job container. If the sidecardepends on another container in the service, indicate this in theContainer start-up order drop-down menu. Repeat this step for each sidecar container you are deploying.

7. ClickCreate for a new service orUpdate for an existing job,then wait for the deployment to finish.

gcloud

1. In the Google Cloud console, activate Cloud Shell.

   Activate Cloud Shell

   At the bottom of the Google Cloud console, aCloud Shell session starts and displays a command-line prompt. Cloud Shell is a shell environment with the Google Cloud CLI already installed and with values already set for your current project. It can take a few seconds for the session to initialize.

2. To deploy multiple containers to a job, run the following command:

   gcloudrunjobscreateJOB\--containerJOB_CONTAINER_NAME\--image='IMAGE_URL'\--containerSIDECAR_CONTAINER_NAME\--image='SIDECAR_IMAGE'

   Replace the following:

   • JOB: the name of the job you aredeploying to. You can omit this parameter entirely, but you will beprompted for the job name if you omit it.
   • JOB_CONTAINER_NAME: a name for the main job container.
   • IMAGE_URL: a reference to thecontainer image—forexample,us-docker.pkg.dev/cloudrun/container/job:latest.
   • SIDECAR_CONTAINER_NAME: a name for the sidecarcontainer—for examplesidecar.
   • SIDECAR_IMAGE: a reference to the sidecar container image.

   If you want to configure each container in the deploy command, supply eachcontainer's configuration after thecontainer parameters, for example:

   gcloudrunjobscreateJOB\--containerCONTAINER_1_NAME\--image='IMAGE_URL'\--set-env-vars=KEY=VALUE\--containerSIDECAR_CONTAINER_NAME\--image='SIDECAR_IMAGE'\--set-env-vars=KEY_N=VALUE_N

   Important: When you use the--container flag, you must specify all non-container-level flags before the container-level flags, otherwise the deploy command fails with an error message to that effect.

3. Wait for the jobs deployment to finish. Upon successful completion, a success message is displayed.

YAML

These instructions show a basic YAML file for your Cloud Run job with sidecars.Create a file namedjob.yaml containing the following:

apiVersion:run.googleapis.com/v1kind:Jobmetadata:name:JOBspec:template:spec:containers:-image:IMAGE_URL-image:SIDECAR_IMAGE

• JOB with the name of your Cloud Run main job container.Job names must be 49 characters or less.
• IMAGE_URL: a reference to thecontainer image—forexample,us-docker.pkg.dev/cloudrun/container/job:latest.
• SIDECAR_IMAGE: a reference to the sidecar container image.

After you update the YAML to include the ingress and sidecar containers, deployto Cloud Run using the command:

gcloud run jobs replace job.yaml

Terraform

To learn how to apply or remove a Terraform configuration, seeBasic Terraform commands.

resource"google_cloud_run_v2_job""default"{name="JOB_NAME"location="europe-west1"template{template{containers{name="CONTAINER_NAME"image="IMAGE_URL"}containers{name="SIDECAR_CONTAINER_NAME"image="SIDECAR_IMAGE_URL"}}}}

Replace the following:

• JOB_NAME: the name of your Cloud Run job.
• CONTAINER_NAME: the name of the container.
• IMAGE_URL: a reference to thecontainer image—forexample,us-docker.pkg.dev/cloudrun/container/job:latest.
• SIDECAR_CONTAINER_NAME: the name of the sidecar container.
• SIDECAR_IMAGE_URL: a reference to the sidecar containerimage.