[Upgrade Assistant] Remove "boom" from reindex service

[jloleysens]

Summary

This contribution replaces the error handling inside reindex service with custom error handling (which was prone to errors could either be Boom HTTP errors or actual Errors which led to complications in handling). Custom Errors are also platform prescribed: #57458. The mapping to HTTP specific codes is moved to just the handlers (see mapAnyErrorToKibanaHttpResponse inside reindex_indices).

How to test

The existing Jest test coverage tests that all throws are still occurring when expected.

This was integration tested locally using this setup:

1. Paste the code snippet into cluster_checkup.ts over the handler
2. Inside the reindex_service, remove the if statement around the throw statement on line 287. This will demonstrate an integration test.

Hack cluster_checkup.ts, ln 32

versionCheckHandlerWrapper(async (
        {
          core: {
            elasticsearch: { dataClient },
          },
        },
        request,
        response
      )
 => {
      // post migration
      const body = await getUpgradeAssistantStatus(
        dataClient,
        isCloudEnabled,
      );

      body.indices = body.indices.concat([{
        level: 'none',
        message: 'none level test',
        url: 'test/url',
        details: 'test details',
        index: 'test1',
        reindex: true,
        needsDefaultFields: false,
      }, {
        level: 'info',
        message: 'info level test',
        url: 'test/url',
        details: 'test details',
        index: 'test2',
        reindex: true,
        needsDefaultFields: false,
      }, {
        level: 'warning',
        message: 'warning level test2',
        url: 'test/url',
        details: 'test details',
        index: 'test3',
        reindex: true,
        needsDefaultFields: false,
      }, {
        level: 'critical',
        message: 'metricbeat-1 successful fix',
        url: 'test/url',
        details: 'test details',
        index: 'metricbeat-1',
        reindex: false,
        needsDefaultFields: true,
      }, {
        level: 'critical',
        message: 'metricbeat-2 failed fix',
        url: 'test/url',
        details: 'test details',
        index: 'metricbeat-2',
        reindex: false,
        needsDefaultFields: true,
      }]);


      body.cluster = body.cluster.concat([{
        level: 'none',
        message: 'none level test message',
        url: 'test/url',
        details: 'test details',
      }, {
        level: 'info',
        message: 'info level test message',
        url: 'test/url',
        details: 'test details',
      }, {
        level: 'warning',
        message: 'warning level test message',
        url: 'test/url',
        details: 'test details',
      }, {
        level: 'critical',
        message: 'critical level test message',
        url: 'test/url',
        details: 'test details',
      }]);

      try {
        return response.ok({
          body,
        });
      } catch (e) {
        if (e.status === 403) {
          return response.forbidden(e.message);
        }

        return response.internalError({ body: e });
      }
    })

To reviewers

1. Currently, reindex service is only used in the reindex routes and inside of reindex worker. Specifically, the processNextStep function is used inside of reindex worker. These cases are all covered and tested, flag it here if you can find any instance of reindex service being used (and errors being caught) that I may have missed.
2. I am using MyError extends Error with symbol to uniquely identify errors. Open to discussing this approach here too.

Additional

• Currently there is only one place that a boom check is in server side UA - reindex actions. This is an isBoom check against an error that could be emitted from Saved Objects client.

For maintainers

• This was checked for breaking API changes and was labeled appropriately

[elasticmachine]

Pinging @elastic/es-ui (Team:Elasticsearch UI)

[jloleysens]

@elasticmachine merge upstream

[alisonelizabeth]

Code changes LGTM. Tested locally and all error handling working as expected.

[jloleysens]

@elasticmachine merge upstream

[jloleysens]

@elasticmachine merge upstream

[kibanamachine]

💚 Build Succeeded

• continuous-integration/kibana-ci/pull-request
• Commit: ff35dc8

History

• 💚 Build #29654 succeeded c32744c

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