Fix the health checks not to be executed before the server and agent are running

[ryysud]

Signed-off-by: Ryuma Yoshida ryumyosh@zlab.co.jp

Pull Request check list

• Commit conforms to CONTRIBUTING.md?
• Proper tests/regressions included?
• Documentation updated?

Affected functionality

the health check system.

Description of change

I received #2015 (review) after merging the implementation of the readiness endpoint for the health check.
To address the comments in that, I fixed the health checks not to be executed before the server and agent are running.

Which issue this PR fixes

This PR fixes #2063.

[ryysud]

For example, the response of the readiness endpoint in the server:

Before the server is running

$ curl -s localhost:8080/ready | jq .
{
  "details": {
    "server": {
      "name": "server",
      "status": "ok",
      "fatal": true,
      "details": {
        "message": "skip the health check because the server is not running yet"
      },
      "check_time": "2021-01-26T15:09:13.928089+09:00",
      "num_failures": 0,
      "first_failure_at": "0001-01-01T00:00:00Z"
    }
  },
  "status": "ok"
}

After the server is running

$ curl -s localhost:8080/ready | jq .
{
  "details": {
    "server": {
      "name": "server",
      "status": "ok",
      "fatal": true,
      "details": {
        "message": "successfully fetched bundle"
      },
      "check_time": "2021-01-26T15:09:21.933612+09:00",
      "num_failures": 0,
      "first_failure_at": "0001-01-01T00:00:00Z"
    }
  },
  "status": "ok"
}

[ryysud]

With this change, the server and agent will be ready even if they are stopped, so I close this PR...