Spring Boot Healthchecks in Google Cloud Run

When trying to get my Spring Boot application to run on Google Cloud Run the health checks were returning 

{
  "status": "DOWN"
}

Which is odd when you consider the fact that I could get to the health check itself.  How can it be running yet return a status of down?

In my Cloud Run logs I saw the error:

o.s.b.a.system.DiskSpaceHealthIndicator : Free disk space below threshold. Available: 0 bytes (threshold: 10485760B)

After doing some digging, since I've never really looker too deeply into the Spring Health Indicators, I realized that spring checks a bunch of different items as part of their status.  Disk Space being one of them.  

Since I'm running this as a jibbified Spring Boot application I didn't reserve any diskspace and the application is running in a OCI without an underlying OS - which totally makes sense why it was failing.

The first thing I did was enable showing the details of the health check (/actuator/health).  In my application.yaml I added:

spring.endpoint.health.show-details=always

which returned me a bunch more information - notice the bold line below

{
  "status": "DOWN",
  "details": {
    "diskSpace": {
      "status": "DOWN",
      "details": {
        "total": 0,
        "free": 0,
        "threshold": 10485760
      }
    },
    "db": {
      "status": "UP",
      "details": {
        "database": "MySQL",
        "hello": 1
      }
    },
    "refreshScope": {
      "status": "UP"
    },
    "discoveryComposite": {
      "status": "UP",
      "details": {
        "discoveryClient": {
          "status": "UP",
          "details": {
            "services": ["api-gateway", "catalog-service", "cart-service"]
          }
        },
        "eureka": {
          "description": "Eureka discovery client has not yet successfully connected to a Eureka server",
          "status": "UP",
          "details": {
            "applications": {
              "API-GATEWAY": 1,
              "CATALOG-SERVICE": 1,
              "CART-SERVICE": 1
            }
          }
        }
      }
    },
    "hystrix": {
      "status": "UP"
    }
  }
}

When one of the health indicators is "down" the status of the service returns "down" - which in turn removes it completely from Eureka.

By addinng another simple config to my application.yaml I was able to fix this problem 

health.diskspace.enabled=false

Which removed the disck check from my status, and listed the entire service a healthy  and allowed it to register with Eureka   

  "status": "UP",
  "details": {
    "db": {
      "status": "UP",
      "details": {
        "database": "MySQL",
        "hello": 1
      }
    },
    "refreshScope": {
      "status": "UP"
    },
    "discoveryComposite": {
      "status": "UP",
      "details": {
        "discoveryClient": {
          "status": "UP",
          "details": {
            "services": ["api-gateway", "cart-service", "catalog-service"]
          }
        },
        "eureka": {
          "description": "Remote status from Eureka server",
          "status": "UP",
          "details": {
            "applications": {
              "API-GATEWAY": 1,
              "CATALOG-SERVICE": 1,
              "CART-SERVICE": 1
            }
          }
        }
      }
    },
    "hystrix": {
      "status": "UP"
    }
  }
}