📘
Note
While using ServiceFoundry python SDK type is not a required field in any of the imported classes

Job

Description

Describes the configuration for the job

Schema

{
  "type": "string",
  "name": "string",
  "image": {},
  "resources": {
    "cpu_request": 0.2,
    "cpu_limit": 0.5,
    "memory_request": 200,
    "memory_limit": 500,
    "ephemeral_storage_request": 1000,
    "ephemeral_storage_limit": 2000,
    "instance_family": [
      "string"
    ]
  },
  "trigger": {
    "type": "manual"
  },
  "retries": 1,
  "timeout": 1000,
  "successful_jobs_history_limit": 20,
  "failed_jobs_history_limit": 20,
  "restart": "Never",
  "params": [
    {
      "name": "string",
      "description": "string",
      "default": "string"
    }
  ],
  "env": null,
  "service_account": "string",
  "file_mounts": [
    {
      "mount_dir": "string",
      "data": {
        "property1": "string",
        "property2": "string"
      }
    }
  ]
}

Properties

Name	Type	Required	Description
type	string	true	none
name	string	true	Name of the job
image	object	true	Specify whether you want to deploy a Docker image or build and deploy from source code
resources	Resources	false	Describes the resource constraints for the application so that it can be deployed accordingly on the cluster To learn more you can go here
trigger	object	true	Specify the trigger
retries	integer	false	Specify the maximum number of attempts to retry a job before it is marked as failed.
timeout	integer	false	Job timeout in seconds.
restart	string	true	When to restart the job
params	[Param]	false	Configure params and pass it to create different job runs
env	object¦null	false	Configure environment variables to be injected in the service. Docs
service_account	string	false	Service account that this workload should use
file_mounts	[FileMount]	false	Files to be mounted +ignore

Enumerate Values

Python Examples

Building using a Python Build pack

from servicefoundry import Job, Build, PythonBuild, Port

job = Job(
    name="my-job",
  	image=Build(
      build_spec=PythonBuild(
          python_version="3.9",
          requirements_path="requirements.txt",
          command="python train.py", # or as a list ["python", "train.py"]
      )
    )
)

deployment = job.deploy(workspace_fqn='...')

from servicefoundry import (
    Job, Build, PythonBuild, LocalSource, Resources,
    Manual, 
)

job = Job(
    name="my-job",
  	image=Build(
        docker_registry="admin-truefoundry:production-euwe1-ecr",  # optional, FQN registry to push the built image to
        build_source=LocalSource(
            project_root_path="./"
        ),
        build_spec=PythonBuild(
            python_version="3.9",
            build_context_path="./",
            requirements_path="requirements.txt",
            pip_packages=["scikit-learn==1.0.2", "numpy==1.20.0"],
            apt_packages=["ffmpeg", "curl", "wget"],
            command="python train.py", # or as a list ["python", "train.py"]
        ),
    ),
    command="python train.py", # or as a list ["python", "train.py"]
  	# Note: This command has precedence over the one in build_spec
    resources=Resources(
        cpu_request=1,  
        memory_request=1000, # in Megabytes
        ephemeral_storage_request=1000, # in Megabytes
        cpu_limit=4,
        memory_limit=4000,
        ephemeral_storage_limit=10000,
        instance_family=["c6i", "t3", "m4"],
    ),
    env={
        "VERY_SECRET_API_KEY": "tfy-secret://my-user:my-secret-group:VERY_SECRET_API_KEY"
    },
    trigger=Manual(run=False),
  	retries=1,
  	timeout=100,
  	successful_jobs_history_limit=20,
  	failed_jobs_history_limit=20,
  	restart="Never",
    service_account="my-k8s-service-account"

)

deployment = job.deploy(workspace_fqn='...')

Building using a Dockerfile

This is useful if you have an already written Dockerfile

from servicefoundry import Job, Build, DockerFileBuild

job = Job(
    name="my-job",
  	image=Build(
      build_spec=DockerFileBuild(),
    ),
)

deployment = job.deploy(workspace_fqn='...')

from servicefoundry import (
    Job, Build, PythonBuild, LocalSource, Resources,
    Manual, 
)

job = Job(
    name="my-job",
  	image=Build(
        docker_registry="admin-truefoundry:production-euwe1-ecr",  # optional, FQN registry to push the built image to
        build_source=LocalSource(
            project_root_path="./"
        ),
        build_spec=PythonBuild(
            python_version="3.9",
            build_context_path="./",
            requirements_path="requirements.txt",
            pip_packages=["scikit-learn==1.0.2", "numpy==1.20.0"],
            apt_packages=["ffmpeg", "curl", "wget"],
            command="python train.py", # or as a list ["python", "train.py"]
        ),
    ),
    resources=Resources(
        cpu_request=1,  
        memory_request=1000, # in Megabytes
        ephemeral_storage_request=1000, # in Megabytes
        cpu_limit=4,
        memory_limit=4000,
        ephemeral_storage_limit=10000,
        instance_family=["c6i", "t3", "m4"],
    ),
    env={
        "VERY_SECRET_API_KEY": "tfy-secret://my-user:my-secret-group:VERY_SECRET_API_KEY"
    },
    trigger=Manual(run=False),
  	retries=1,
  	timeout=100,
  	successful_jobs_history_limit=20,
  	failed_jobs_history_limit=20,
  	restart="Never",
    service_account="my-k8s-service-account"

)

deployment = job.deploy(workspace_fqn='...')

Using an already existing Image from an Image Registry

Note: The remote Image Registry should either be public or added to Truefoundry Platform via the Integrations Tab

from servicefoundry import Job, Image

job = Job(
    name="my-job",
  	image=Image(
      image_uri="docker/whalesay:latest",
    ),
)

deployment = job.deploy(workspace_fqn='...')

from servicefoundry import (
    Job, Image, Resources, Manual
)
 
job = Job(
    name="my-job",
  	image=Image(
      image_uri="docker/whalesay:latest",
      docker_registry="admin-truefoundry:production-euwe1-ecr",  # optional, FQN registry to pull the built image from
      command=["cowsay", "boo"],
    ),    
    resources=Resources(
        cpu_request=1,  
        memory_request=1000, # in Megabytes
        ephemeral_storage_request=1000, # in Megabytes
        cpu_limit=4,
        memory_limit=4000,
        ephemeral_storage_limit=10000,
        instance_family=["c6i", "t3", "m4"],
    ),
    env={
        "VERY_SECRET_API_KEY": "tfy-secret://my-user:my-secret-group:VERY_SECRET_API_KEY"
    },
    trigger=Manual(run=False),
  	retries=1,
  	timeout=100,
  	successful_jobs_history_limit=20,
  	failed_jobs_history_limit=20,
  	restart="Never",
    service_account="my-k8s-service-account"
)

deployment = job.deploy(workspace_fqn='...')

Deploying from a Git Repository

Note: The remote Git Server (Github or Bitbucket) needs to be connected via the Integrations Tab

from servicefoundry import (
    Job, Build, GitSource, PythonBuild, Resources,
    Manual
)

job = Job(
    name="my-job",
  	image=Build(
        docker_registry="admin-truefoundry:production-euwe1-ecr",  # optional, FQN registry to push the built image to
        build_source=GitSource(
            repo_url="https://github.com/myorg/myrepo",
            branch_name="main", # Optional, by default the default branch configured on the repo
            ref="547ea1423267902a6ff98517d4ae205d8f7c47f8", # Optional, by default latest commit on the branch
        ),
        build_spec=PythonBuild(
            python_version="3.9",
            build_context_path="./",
            requirements_path="requirements.txt",
            pip_packages=["scikit-learn==1.0.2", "numpy==1.20.0"],
            apt_packages=["ffmpeg", "curl", "wget"],
            command="python train.py", # or as a list ["python", "train.py"]
        ),
    ),
    resources=Resources(
        cpu_request=1,  
        memory_request=1000, # in Megabytes
        ephemeral_storage_request=1000, # in Megabytes
        cpu_limit=4,
        memory_limit=4000,
        ephemeral_storage_limit=10000,
        instance_family=["c6i", "t3", "m4"],
    ),
    env={
        "VERY_SECRET_API_KEY": "tfy-secret://my-user:my-secret-group:VERY_SECRET_API_KEY"
    },
    trigger=Manual(run=False),
  	retries=1,
  	timeout=100,
    service_account="my-k8s-service-account"
    
)

deployment = job.deploy(workspace_fqn='...')

from servicefoundry import (
    Job, Build, DockerFileBuild, GitSource, Resources,
    Manual
)

job = Job(
    name="my-job",
  	image=Build(
        docker_registry="admin-truefoundry:production-euwe1-ecr",  # optional, FQN registry to push the built image to
        build_source=GitSource(
            repo_url="https://github.com/myorg/myrepo",
            branch_name="main", # Optional, by default the default branch configured on the repo
            ref="547ea1423267902a6ff98517d4ae205d8f7c47f8", # Optional, by default latest commit on the branch
        ),
        build_spec=DockerFileBuild(
            dockerfile_path='./Dockerfile',
            build_context_path='./',
            command="python train.py",
        ),
    ),    
    resources=Resources(
        cpu_request=1,  
        memory_request=1000, # in Megabytes
        ephemeral_storage_request=1000, # in Megabytes
        cpu_limit=4,
        memory_limit=4000,
        ephemeral_storage_limit=10000,
        instance_family=["c6i", "t3", "m4"],
    ),
    env={
        "VERY_SECRET_API_KEY": "tfy-secret://my-user:my-secret-group:VERY_SECRET_API_KEY"
    },
    trigger=Manual(run=False),
  	retries=1,
  	timeout=100,
  	successful_jobs_history_limit=20,
  	failed_jobs_history_limit=20,
  	restart="Never",
    service_account="my-k8s-service-account"
    
)

deployment = job.deploy(workspace_fqn='...')

Triggers

The modules below help configure the Triggers for the Job

Manual

Description

Describes that we are going to manually trigger our job.

Schema

{
  "type": "string"
}

Properties

Name	Type	Required	Description
type	string	true	none

Python Examples

from servicefoundry import Job, Manual

job = Job(
    ...
    trigger=Manual(
      run=False
      ),
    ...
    )
)

Schedule

Description

Describes that we are going to schedule our job to run at a schedule, making our job a cron job.

Schema

{
  "type": "string",
  "schedule": "string",
  "concurrency_policy": "Forbid"
}

Properties

Name	Type	Required	Description
type	string	true	none
schedule	string	true	Specify the schedule for this job to be run periodically in cron format. Learn more
concurrency_policy	string	false	Choose whether to allow this job to run while another instance of the job is running, or to replace the currently running instance. Allow will enable multiple instances of this job to run. Forbid will keep the current instance of the job running and stop a new instance from being run. Replace will terminate any currently running instance of the job and start a new one.

Enumerate Values

Property	Value
concurrency_policy	Forbid
concurrency_policy	Allow
concurrency_policy	Replace

Python Examples

from servicefoundry import Job, Schedule

job = Job(
    ...
    trigger=Schedule(
      schedule="0 8 1 * *", # Cron format
      concurrency_policy="Forbid" # Any one of ["Forbid", "Allow", "Replace"]
      ),
    ...
    )
)

Param

Description

Describes what params are going to be given to the Deployment

Schema

{
  "name": "string",
  "description": "string",
  "default": "string"
}

Properties

Name	Type	Required	Description
name	string	true	Name of the param
description	string	false	Description of param
default	string	false	Default value or placeholder