DeploymentConfig

OpenShift DeploymentConfig resource for managing application deployments

This resource wrapper does not currently exist in the openshift-python-wrapper library.

You can use the class generator to create this wrapper for your cluster:

class-generator --kind DeploymentConfig

Or work with DeploymentConfig resources using the generic Resource class with kind="DeploymentConfig" and api_group="apps.openshift.io".

Overview

DeploymentConfig is an OpenShift-specific resource that defines the template for a pod and manages deploying new images or configuration changes. DeploymentConfigs provide more advanced deployment strategies and triggers compared to standard Kubernetes Deployments.

API Group: apps.openshift.io/v1

While still supported, OpenShift recommends using standard Kubernetes Deployments for new applications. DeploymentConfigs are maintained for backward compatibility and specific OpenShift features.

Class Definition

from ocp_resources.deployment_config import DeploymentConfig

class DeploymentConfig(NamespacedResource):
    api_group = NamespacedResource.ApiGroup.APPS_OPENSHIFT_IO

Constructor

DeploymentConfig(
    name=None,
    namespace=None,
    client=None,
    replicas=None,
    selector=None,
    template=None,
    strategy=None,
    triggers=None,
    revision_history_limit=None,
    min_ready_seconds=None,
    paused=None,
    teardown=True,
    yaml_file=None,
    delete_timeout=TIMEOUT_4MINUTES,
    **kwargs
)

Parameters

name · str · default: "None"

Name of the DeploymentConfig resource. If not provided, it must be specified in the YAML file.

namespace · str · default: "None"

Namespace where the DeploymentConfig will be created. Required for namespaced resources.

client · DynamicClient · default: "None"

Kubernetes dynamic client for API communication. If not provided, the default client will be used.

replicas · int · default: "None"

Number of desired pod replicas. Defaults to 1 if not specified.

selector · dict[str, Any] · default: "None"

Label selector for pods. Must match the labels in the pod template.

Example:

{"app": "my-app", "deploymentconfig": "my-app"}

template · dict[str, Any] · default: "None"

Pod template specification. Describes the pods that will be created.

Example:

{
    "metadata": {
        "labels": {"app": "my-app"}
    },
    "spec": {
        "containers": [
            {
                "name": "app",
                "image": "my-app:latest",
                "ports": [{"containerPort": 8080}]
            }
        ]
    }
}

strategy · dict[str, Any] · default: "None"

Deployment strategy specification. Defines how the deployment should be executed (Rolling, Recreate, or Custom).

Example for Rolling strategy:

{
    "type": "Rolling",
    "rollingParams": {
        "updatePeriodSeconds": 1,
        "intervalSeconds": 1,
        "timeoutSeconds": 600,
        "maxUnavailable": "25%",
        "maxSurge": "25%"
    }
}

triggers · list[dict[str, Any]] · default: "None"

List of triggers that automatically start a new deployment. Common trigger types include ConfigChange and ImageChange.

Example:

[
    {"type": "ConfigChange"},
    {
        "type": "ImageChange",
        "imageChangeParams": {
            "automatic": True,
            "containerNames": ["app"],
            "from": {
                "kind": "ImageStreamTag",
                "name": "my-app:latest"
            }
        }
    }
]

revision_history_limit · int · default: "None"

Number of old ReplicationControllers to retain for rollback. Defaults to 10.

min_ready_seconds · int · default: "None"

Minimum number of seconds for which a newly created pod should be ready without any containers crashing. Defaults to 0.

paused · bool · default: "None"

Indicates whether the deployment is paused. Paused deployments do not automatically deploy new versions.

teardown · bool · default: "True"

If True, the resource will be automatically deleted when used as a context manager.

yaml_file · str · default: "None"

Path to a YAML file containing the DeploymentConfig definition. If provided, the DeploymentConfig will be created from this file.

delete_timeout · int · default: "TIMEOUT_4MINUTES"

Timeout in seconds for delete operations. Defaults to 4 minutes.

**kwargs · Any

Additional keyword arguments passed to the parent NamespacedResource class. Common options include label and annotations.

Usage Examples

Basic DeploymentConfig

Create a simple DeploymentConfig:

from ocp_resources.deployment_config import DeploymentConfig

dc = DeploymentConfig(
    name="my-app",
    namespace="default",
    replicas=3,
    selector={"app": "my-app"},
    template={
        "metadata": {
            "labels": {"app": "my-app"}
        },
        "spec": {
            "containers": [
                {
                    "name": "app",
                    "image": "my-registry/my-app:v1.0",
                    "ports": [{"containerPort": 8080}]
                }
            ]
        }
    },
    triggers=[
        {"type": "ConfigChange"}
    ]
)
dc.deploy()

DeploymentConfig with ImageStream Trigger

Create a DeploymentConfig that automatically deploys when the ImageStream changes:

from ocp_resources.deployment_config import DeploymentConfig

dc = DeploymentConfig(
    name="auto-deploy-app",
    namespace="default",
    replicas=2,
    selector={"app": "auto-deploy-app"},
    template={
        "metadata": {
            "labels": {"app": "auto-deploy-app"}
        },
        "spec": {
            "containers": [
                {
                    "name": "app",
                    "image": "auto-deploy-app:latest",
                    "ports": [{"containerPort": 8080}]
                }
            ]
        }
    },
    triggers=[
        {"type": "ConfigChange"},
        {
            "type": "ImageChange",
            "imageChangeParams": {
                "automatic": True,
                "containerNames": ["app"],
                "from": {
                    "kind": "ImageStreamTag",
                    "name": "auto-deploy-app:latest"
                }
            }
        }
    ]
)
dc.deploy()

Rolling Deployment Strategy

Create a DeploymentConfig with a custom rolling deployment strategy:

from ocp_resources.deployment_config import DeploymentConfig

dc = DeploymentConfig(
    name="rolling-app",
    namespace="default",
    replicas=5,
    selector={"app": "rolling-app"},
    template={
        "metadata": {
            "labels": {"app": "rolling-app"}
        },
        "spec": {
            "containers": [
                {
                    "name": "app",
                    "image": "rolling-app:v2.0",
                    "ports": [{"containerPort": 8080}],
                    "readinessProbe": {
                        "httpGet": {
                            "path": "/health",
                            "port": 8080
                        },
                        "initialDelaySeconds": 5,
                        "periodSeconds": 10
                    }
                }
            ]
        }
    },
    strategy={
        "type": "Rolling",
        "rollingParams": {
            "updatePeriodSeconds": 1,
            "intervalSeconds": 1,
            "timeoutSeconds": 600,
            "maxUnavailable": "25%",
            "maxSurge": "25%",
            "pre": {
                "failurePolicy": "Abort",
                "execNewPod": {
                    "command": ["/bin/sh", "-c", "echo pre-deployment hook"],
                    "containerName": "app"
                }
            }
        }
    }
)
dc.deploy()

Recreate Deployment Strategy

Create a DeploymentConfig with Recreate strategy for applications that cannot run multiple versions:

from ocp_resources.deployment_config import DeploymentConfig

dc = DeploymentConfig(
    name="database-app",
    namespace="default",
    replicas=1,
    selector={"app": "database-app"},
    template={
        "metadata": {
            "labels": {"app": "database-app"}
        },
        "spec": {
            "containers": [
                {
                    "name": "db",
                    "image": "postgres:13",
                    "env": [
                        {"name": "POSTGRES_PASSWORD", "value": "secret"}
                    ],
                    "ports": [{"containerPort": 5432}]
                }
            ]
        }
    },
    strategy={
        "type": "Recreate",
        "recreateParams": {
            "timeoutSeconds": 600
        }
    }
)
dc.deploy()

Creating from YAML

Create a DeploymentConfig from a YAML file:

from ocp_resources.deployment_config import DeploymentConfig

dc = DeploymentConfig(
    namespace="default",
    yaml_file="deploymentconfig.yaml"
)
dc.deploy()

Using Context Manager

Automatically clean up the DeploymentConfig after use:

from ocp_resources.deployment_config import DeploymentConfig

with DeploymentConfig(
    name="temp-dc",
    namespace="default",
    yaml_file="temp-deploymentconfig.yaml"
) as dc:
    # Wait for rollout
    dc.wait_for_replicas()
    # DeploymentConfig is automatically deleted when exiting

Deployment Strategies

Rolling Strategy

Gradually replaces old pods with new ones, maintaining availability:

strategy={
    "type": "Rolling",
    "rollingParams": {
        "updatePeriodSeconds": 1,      # Time between pod updates
        "intervalSeconds": 1,           # Time between checks
        "timeoutSeconds": 600,          # Maximum deployment time
        "maxUnavailable": "25%",        # Max pods that can be unavailable
        "maxSurge": "25%"               # Max additional pods during update
    }
}

Recreate Strategy

Terminates all old pods before creating new ones:

strategy={
    "type": "Recreate",
    "recreateParams": {
        "timeoutSeconds": 600,
        "mid": {                        # Mid-lifecycle hook
            "failurePolicy": "Abort",
            "execNewPod": {
                "command": ["/bin/migrate"],
                "containerName": "app"
            }
        }
    }
}

Custom Strategy

Uses a custom deployment process:

strategy={
    "type": "Custom",
    "customParams": {
        "image": "custom-deployer:latest",
        "command": ["/bin/deploy"],
        "environment": [
            {"name": "CUSTOM_VAR", "value": "value"}
        ]
    }
}

Lifecycle Hooks

DeploymentConfigs support lifecycle hooks for executing tasks during deployment:

Pre Hook

Executes before the deployment starts:

"pre": {
    "failurePolicy": "Abort",
    "execNewPod": {
        "command": ["/bin/sh", "-c", "echo Starting deployment"],
        "containerName": "app",
        "env": [{"name": "HOOK_TYPE", "value": "pre"}]
    }
}

Mid Hook (Recreate only)

Executes after old pods are scaled down but before new pods are created:

"mid": {
    "failurePolicy": "Abort",
    "execNewPod": {
        "command": ["/bin/migrate"],
        "containerName": "app"
    }
}

Post Hook

Executes after the deployment completes:

"post": {
    "failurePolicy": "Ignore",
    "execNewPod": {
        "command": ["/bin/sh", "-c", "echo Deployment complete"],
        "containerName": "app"
    }
}

Triggers

Configuration Change Trigger

Automatically starts a new deployment when the DeploymentConfig changes:

triggers=[{"type": "ConfigChange"}]

Image Change Trigger

Automatically starts a new deployment when the specified image changes:

triggers=[
    {
        "type": "ImageChange",
        "imageChangeParams": {
            "automatic": True,
            "containerNames": ["app"],
            "from": {
                "kind": "ImageStreamTag",
                "name": "my-app:latest",
                "namespace": "default"
            },
            "lastTriggeredImage": ""  # Updated automatically
        }
    }
]

DeploymentConfig vs Deployment

Feature	DeploymentConfig	Deployment
API	OpenShift-specific	Kubernetes standard
Strategies	Rolling, Recreate, Custom	RollingUpdate, Recreate
Lifecycle Hooks	Yes (pre, mid, post)	No
Image Triggers	Yes	No (use external tools)
Custom Deployer	Yes	No
Recommendation	Legacy/specific features	Preferred for new apps

Related Resources

• Deployment - Kubernetes standard deployment
• ReplicationController - Created by DeploymentConfig
• ImageStream - Used with ImageChange triggers
• OpenShift DeploymentConfig Documentation

See Also

• Working with Deployments Guide
• Core Concepts: Resources
• NamespacedResource Base Class