Fix empty host list for instance stoppable response

[huizhilu]

Issues

• My PR addresses the following Helix issues and references them in the PR description:

Fixed #1232

Description

• Here are some details about my PR, including screenshots of any UI changes:

Health check stoppable responds with empty hostname.

This PR adds error message of instance check to API response so users have a clear idea of the reason if PERSIST_INTERMEDIATE_ASSIGNMENT is not turned on.

curl -XPOST http://localhost:8100/admin/v2/namespaces/namespacce/clusters/cluster/instances\?command\=stoppable -d ' {"instances": ["instance"], "selection_base": "zone_based", "max_instance": "2", "customized_values": "{}"}' -H "Content-Type: application/json"
{
  "instance_stoppable_parallel" : [ ],
  "instance_not_stoppable_with_reasons" : {
    "instance" : [ "HELIX:INVALID_CONFIG" ]
  }
}

Server log will have the reason logged: Cluster config PERSIST_INTERMEDIATE_ASSIGNMENT is not turned on, which is required for instance stability check.

Tests

• The following tests are written for this issue:

• testBatchGetInstancesStoppableChecksWhenException()

• The following is the result of the "mvn test" command on the appropriate module:

[INFO] Tests run: 166, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 163.995 s - in TestSuite
[INFO]
[INFO] Results:
[INFO]
[INFO] Tests run: 166, Failures: 0, Errors: 0, Skipped: 0
[INFO]
[INFO] ------------------------------------------------------------------------
[INFO] BUILD SUCCESS
[INFO] ------------------------------------------------------------------------
[INFO] Total time:  02:49 min
[INFO] Finished at: 2020-08-07T21:27:36-07:00
[INFO] ------------------------------------------------------------------------

Documentation (Optional)

• In case of new functionality, my PR adds documentation in the following wiki page:

(Link the GitHub wiki you added)

Commits

• My commits all reference appropriate Apache Helix GitHub issues in their subject lines. In addition, my commits follow the guidelines from "How to write a good git commit message":
  1. Subject is separated from body by a blank line
  2. Subject is limited to 50 characters (not including Jira issue reference)
  3. Subject does not end with a period
  4. Subject uses the imperative mood ("add", not "adding")
  5. Body wraps at 72 characters
  6. Body explains "what" and "why", not "how"

Code Quality

• My diff has been formatted using helix-style.xml
  (helix-style-intellij.xml if IntelliJ IDE is used)

[junkaixue]

Error message for server side is clear to find out the fail reason. The more important is let user to understand what is the real fail reason. Do we have that if persist assignment turned off, at least, we gave user info EV and IS does not match.

[huizhilu]

Error message for server side is clear to find out the fail reason. The more important is let user to understand what is the real fail reason. Do we have that if persist assignment turned off, at least, we gave user info EV and IS does not match.

@dasahcc This PR is to provide the info to let users know the reason. The root cause is exception/error is not returned as response. This PR adds the exception/error message to not stoppable reason in the return response.
https://github.com/apache/helix/pull/1237/files#diff-c98c0908866d29b64f625e799c198d15R224-R227

The result will be what is the description

curl -XPOST http://localhost:8100/admin/v2/namespaces/namespacce/clusters/cluster/instances\?command\=stoppable -d ' {"instances": ["instance"], "selection_base": "zone_based", "max_instance": "2", "customized_values": "{}"}' -H "Content-Type: application/json"
{
  "instance_stoppable_parallel" : [ ],
  "instance_not_stoppable_with_reasons" : {
    "instance" : [ "HELIX:Instance stability check needs persist assignment cluster config turned on: PERSIST_INTERMEDIATE_ASSIGNMENT = true" ]
  }
}

[huizhilu]

Error message for server side is clear to find out the fail reason. The more important is let user to understand what is the real fail reason. Do we have that if persist assignment turned off, at least, we gave user info EV and IS does not match.

@dasahcc I think it is more appropriate to put this config check in INVALID_CONFIG health check, which is the first check in health status check. If configs are invalid, we don't need to do remaining checks like INSTANCE_NOT_STABLE.

The response should be like this

curl -XPOST http://localhost:8100/admin/v2/namespaces/namespacce/clusters/cluster/instances\?command\=stoppable -d ' {"instances": ["instance"], "selection_base": "zone_based", "max_instance": "2", "customized_values": "{}"}' -H "Content-Type: application/json"
{
  "instance_stoppable_parallel" : [ ],
  "instance_not_stoppable_with_reasons" : {
    "instance" : [ "HELIX:INVALID_CONFIG" ]
  }
}

I actually would like to add more info for users to know why. But with current implementation, it is not easy/straightforward to add the message. Maybe ` "HELIX:INVALID_CONFIG:Cluster config PERSIST_INTERMEDIATE_ASSIGNMENT is not turned on"

I think the better one would be a map response for each instance:

{
  "instance_stoppable_parallel" : [ ],
  "instance_not_stoppable_with_reasons" : [{
    "instance" : "hostname",
    "health-check": [ "HELIX:INVALID_CONFIG" ],
    "message": "Cluster config PERSIST_INTERMEDIATE_ASSIGNMENT is not turned on"
  }]
}

[huizhilu]

Thanks, @dasahcc @jiajunwang, for the review.

This PR is ready to be merged, approved by @dasahcc