[Ingest] Allow to reassign agent to a new config

[nchaulet]

Summary

Resolves #63021

This PR allow a fleet user to assign an already enrolled agent to a new configuration.

Done in this PR

• UI Create a new flyout to reassign an agent
• Create a new API to reassign an agent
• Updated unenroll API to not support bulk operations

UI Changes

API Change

Create a new route similar to the one used for unenrolling agent

POST /api/ingest_manager/fleet/agents/{agentId}reassign
{
  "config_id": "25f8b9f0-7fe4-11ea-b993-1723d0399bd5"
}

// Response 
{
  "success": true
}

[elasticmachine]

Pinging @elastic/ingest-management (Team:Ingest Management)

[ruflin]

Lets focus this for now on just reassigning a single agent to a new config. This will make also error handling easier so we don't have to think about what if 2 out of 4 reassignements worked?

[nchaulet]

Updated the API to not support bulk operations.

[neptunian]

Testing this locally. When I reassign my agent to a new config I get this error in the console where the agent is running:

Is this expected?

[nchaulet]

@neptunian no the error is not expected, just tested locally it's working fine for me, how do you get the error?

[nchaulet]

@elasticmachine merge upstream

[jen-huang]

When the reassignment flyout is open on the agent list page with many agents, the context of which agent is being reassigned gets lost since we don't have a visual indicator of which agent is selected (when it was a bulk action, the indicator is selected rows, but we are not doing bulk at the moment). Maybe we can include the selected agent host as part of the flyout info. @hbharding could you suggest a design for this? (Not a blocker for this PR; enhancement can be done in follow up PR.)

I have some other small copy suggestions below, otherwise tested locally and LGTM 👍

[jen-huang]

Btw I also see the same errors in agent sdout that @neptunian saw:

2020-04-22T10:52:42-07:00 DEBUG	action_dispatcher.go:96	Successfully dispatched action: 'action_id: 15396920-84c1-11ea-9a94-79f7dbd8dc26, type: CONFIG_CHANGE'
2020-04-22T10:52:42-07:00 DEBUG	client.go:170	Request method: POST, path: /api/ingest_manager/fleet/agents/0b78c480-84c1-11ea-9a94-79f7dbd8dc26/acks
2020-04-22T10:52:42-07:00 ERROR	fleet_gateway.go:143	failed to dispatch actions, error: acknowledge 132 actions '[action_id: 256098e0-8427-11ea-b57d-d145f25d3ece, type: CONFIG_CHANGE action_id: 15396920-84c1-11ea-9a94-79f7dbd8dc26, type: CONFIG_CHANGE action_id: 15396920-84c1-11ea-9a94-79f7dbd8dc26, type: CONFIG_CHANGE action_id: 15396920-84c1-11ea-9a94-79f7dbd8dc26, type: CONFIG_CHANGE action_id: 15396920-84c1-11ea-9a94-79f7dbd8dc26, type: CONFIG_CHANGE action_id: 15396920-84c1-11ea-9a94-79f7dbd8dc26, type: CONFIG_CHANGE action_id: 15396920-84c1-11ea-9a94-79f7dbd8dc26, type: CONFIG_CHANGE action_id: 15396920-84c1-11ea-9a94-79f7dbd8dc26, type: CONFIG_CHANGE action_id: 15396920-84c1-11ea-9a94-79f7dbd8dc26, type: CONFIG_CHANGE action_id: 15396920-84c1-11ea-9a94-79f7dbd8dc26,
...

I think I have seen the above ^ outside of this PR, though.

I also have a bunch of errors like this:

2020-04-22T10:53:28-07:00 ERROR	fleet_gateway.go:162	Could not communicate with Checking API will retry, error: Status code: 500, Kibana returned an error: Internal Server Error, message: Unexpected token o in JSON at position 1
	Status code: 500, Kibana returned an error: Internal Server Error, message: Unexpected token o in JSON at position 1
2020-04-22T10:53:37-07:00 DEBUG	client.go:170	Request method: POST, path: /api/ingest_manager/fleet/agents/0b78c480-84c1-11ea-9a94-79f7dbd8dc26/checkin

The only thing I did was rapidly reassign an agent between two configs.

[nchaulet]

@jen-huang did you re-enroll multiple time an agent from the same folder? if yes it's probably related to this error elastic/beats#17389

[jen-huang]

Interesting, that may be the case. Thanks for the link.

[kibanamachine]

💚 Build Succeeded

• continuous-integration/kibana-ci/pull-request
• Commit: 4284181

History

• 💚 Build #42315 succeeded ddc2188
• 💔 Build #42260 failed 683fc46
• 💚 Build #42160 succeeded 4ba3fd6
• 💔 Build #42030 failed bdc7514
• 💔 Build #41847 failed 22817a7

To update your PR or re-run it, just comment with:
@elasticmachine merge upstream