Fix application name in API key

[mtojek]

This PR fixes issue spotted in: elastic/elastic-package#650

{"log.level":"info","ecs.version":"1.6.0","service.name":"fleet-server","http.request.id":"01FSMA8YRG2G2DDJESQ6ZCVW42","mod":"enroll","fleet.enroll.apikey.id":"wm2kaH4BMg40PYJ8YCOR","error.message":"fail CreateAPIKey: [400 Bad Request] {\"error\":{\"root_cause\":[{\"type\":\"action_request_validation_exception\",\"reason\":\"Validation Failed: 1: An application name prefix must match the pattern ^[a-z][A-Za-z0-9]*$ (found '.fleet');\"}],\"type\":\"action_request_validation_exception\",\"reason\":\"Validation Failed: 1: An application name prefix must match the pattern ^[a-z][A-Za-z0-9]*$ (found '.fleet');\"},\"status\":400}","http.response.status_code":400,"event.duration":306104815,"@timestamp":"2022-01-17T15:25:17.251Z","message":"fail enroll"}
{"log.level":"error","ecs.version":"1.6.0","service.name":"fleet-server","http.request.id":"01FSMA8YRG2G2DDJESQ6ZCVW42","mod":"enroll","fleet.enroll.apikey.id":"wm2kaH4BMg40PYJ8YCOR","error.message":"fail CreateAPIKey: [400 Bad Request] {\"error\":{\"root_cause\":[{\"type\":\"action_request_validation_exception\",\"reason\":\"Validation Failed: 1: An application name prefix must match the pattern ^[a-z][A-Za-z0-9]*$ (found '.fleet');\"}],\"type\":\"action_request_validation_exception\",\"reason\":\"Validation Failed: 1: An application name prefix must match the pattern ^[a-z][A-Za-z0-9]*$ (found '.fleet');\"},\"status\":400}","@timestamp":"2022-01-17T15:25:17.251Z","message":"perform rollback on enrollment failure"}

[elasticmachine]

💚 Build Succeeded

the below badges are clickable and redirect to their specific view in the CI or DOCS

Expand to view the summary

Build stats

• Start Time: 2022-01-17T16:57:43.078+0000

• Duration: 11 min 51 sec

• Commit: aa949ec

Test stats 🧪

Test	Results
Failed	0
Passed	229
Skipped	0
Total	229

🤖 GitHub comments

To re-run your PR in the CI, just comment with:

• /test : Re-trigger the build.

[michel-laterman]

thanks for fixing this

[ruflin]

I wonder if we need to get this also into 7.17 as a 7.17 fleet-server could talk to an Elasticsearch 8.0 (x?) instance.

@ph @blakerouse

[ruflin]

I'm also wondering if this was catched by our e2e tests? I assume also we should have seen this in the CI with the integration tests but we did not have any PR's open. If not, we should adjust the tests.

@mtojek Thanks for the PR, I expect the control plane team to follow up from here. @jlind23

[ruflin]

Why do we have a copy here instead of using the one from the actual implementation? Using a copy means we could have a passing test but the actual implementation does not work.

[jlind23]

@ruflin it seems to be the same issue mentioned there elastic/beats#29811

And it's causing this e2e-tests to fail.

cc @ph

[mtojek]

I can confirm on our side that this fix solved the issue for elastic-package (fresh Docker images baked). The PR was green, didn't see any failing tests.

[jlind23]

@mtojek i think that @ruflin question was, were they any failing tests before that had been triggered by this issue?

[blakerouse]

@ruflin Yes this needs to go to 7.17. Fleet Server needs to be able to communicate with an 8.0 elasticsearch.