Java Infrastructure Endpoints

This library deliveries a simple way to give your spring-boot app two restful endpoints for obtain data about application's info and health routes.

Routes

Application Info

Response Example

Body

{
  "applicationName": "my-application",
  "createdBy": "My Build Tool vX.Y.Z",
  "version": "Major.minor.release",
  "buildNumber": "20180301-25",
  "buildJdk": "1.8",
  "springBootVersion": "1.5.9.RELEASE"
}

Status

Application info always return 200 Ok.

Objective

When called, this route should return information about the application's package construction.

Response body explained

The response object have this following fields:

• applicationName: your application name (replacing blank spaces for dashes, accented characters for unaccented characters and omitting special charactes).
• createdBy: name and version of your build tool. If no one are used, you can specify the CI server name and version.
• version: your application version, may be '{major}.{minor}.{release}' pattern or a git hash from repository.
• buildNumber: the date and the CI build number where itś possible to find the artifacts of this deployment. Preferable that it be in the following pattern: 'yyyyMMdd-{CI-BUILD_NUMBER}'.

Pro Tip: You can configure your gradle (or maven) to configure your MANIFEST.MF to contains those keys above.

Application Health

Response Example

Body

{
  "status": "UP",
  "message": "All dependencies are up",
  "dependencies": [
    {
      "name": "My Database",
      "isCritical": true,
      "status": "UP"
    },
    {
      "name": "My 1st Http Dependency",
      "isCritical": true,
      "status": "UP"
    },
    {
      "name": "My 2nd Http Dependency",
      "isCritical": false,
      "status": "UP"
    }
  ]
}

Status

Application health returns 200 for UP or PARTIAL statuses and 503 for the FAIL status.

Objective

When called, this route should ping all the dependencies you've declared in your application: databases, filesystem directory, http dependencies, a queue, a topic, etc. Basically, these pings should garantee that application connections to dependencies are established or, at least, establishables.

Response body explained

The response object have this following fields:

• status: The overall health of your application. It's an enumeration with UP, PARTIAL or FAIL as possible values
• message: An explanation about the status field
• dependencies: Dependencies declared in your application. Each dependency is a complex object that has in its structure the following fields:
  • name: The declared dependency name
  • is_critical: dependency's criticity. It should be determined by its impact when out of service: if your application gets only partially impacted, it should be false, it should be true otherwise
  • status: the health of your application connectivity with this dependency. It's an enumeration with UP or FAIL as possible values.