Page last updated:
This documentation is intended for end users of Cloud Foundry and covers provisioning of service instances and integrating them with apps that have been pushed to Cloud Foundry. If you are interested in building services for Pivotal Platform and making them available to end users, see Services.
Cloud Foundry offers a marketplace of services, from which users can provision reserved resources on-demand. Examples of resources services provide include databases on a shared or dedicated server, or accounts on a SaaS app. These resources are known as service instances and the systems that deliver and operate these resources are known as Services. Think of a service as a factory that delivers service instances.
For more information about provisioning service instances and other lifecycle operations, see Managing Service Instances.
Note: For a service to be available in the marketplace, it must be integrated with Cloud Foundry by way of APIs. If you are interested in building services for Pivotal Platform and making them available to end users, see [Services](https://docs.pivotal.io/platform/services/).
Cloud Foundry enables users to leverage services that are not available in the marketplace using user-provided service instances (UPSI). For more information, see User-Provided Service Instances
Cloud Foundry enables users to provision credentials needed to interface with a service instance. You can use app binding to automatically deliver these credentials to your Cloud Foundry app. For external and local clients, you can use service keys to generate credentials to communicate directly with a service instance.
Service instance credentials can be delivered automatically to apps running on Cloud Foundry in an environment variable. For more information, see Delivering Service Credentials to an App.
For information about binding to a specific app development framework, see Buildpacks.
Credentials managed manually are known as service keys. Use service keys when you want a set of credentials for use by clients other than the app in the same space. For instance, you can use service keys to connect to a service instance from a local client, or from an app in another space, or even from outside of Cloud Foundry.
For more information about creating a user-provided service instance with service keys, see the User-Provided Service Instances topic. For more information about service keys, see the Managing Service Keys topic.
Note: Not all services support service keys. Some services support credentials through app binding only.
To allow an app to communicate with a service external to PAS, you may need to configure the service to accept connections from your app based on its outbound IP address.
In your external service configuration, you must do one of the following:
- Add the entire IP range for the Diego Cell where the app is deployed to your allow list.
- Derive the app IP address from its DNS name using a command-line tool such as
nslookup. In your external service configuration, add the IP address or range of the app instance to your allow list.
To learn how your app logs can be streamed to third-party log management services, see Streaming App Logs to Log Management Services.
User-provided service instances can be used to drain app logs to a service not available in the marketplace. This is also known as setting up a syslog drain. For guidance on configuring some third-party log management services, see Service-Specific Instructions for Streaming App Logs.
To learn how marketplace services (and user-provided service instances) can be used to perform preprocessing on app requests, see Manage App Requests with Route Services.
If your app relies on a relational database, you must apply schema changes periodically. To perform database schema migrations on Cloud Foundry-managed services, run a database migration task with the Cloud Foundry Command Line Interface (cf CLI) tool.
For more information about running cf CLI tasks, see Running Tasks.
Note: To run tasks with the cf CLI, you must install cf CLI v6.23.0 or later. For information about downloading, installing, and uninstalling the cf CLI, see the Installing the cf CLI.
To perform a database schema migration with the cf CLI:
Push the app:
$ cf push APP-NAME
APP-NAMEis the name of the app.
Note: To run a task without starting the app, push the app with
cf push -i 0and then run the task. You can run the app later by scaling up its instance count.
Perform a database schema migration as a task on the app:
$ cf run-task APP-NAME "bin/rails db:migrate" --name TASK-NAME Creating task for app APP-NAME in org jdoe-org / space development as firstname.lastname@example.org... OK Task 1 has been submitted successfully for execution.
APP-NAMEis the name of the app.
TASK-NAMEis the name of the task.