[Fleet] Fix secrets exception when installing CSPM or other integrations

[criamico]

Closes #173718

Summary

Fix secrets exception when installing CSPM or other integrations

Steps to reproduce:

• Install cloud_security_posture-1.8.0-preview02 (note that a licence is needed to install CSPM)
• Select Setup access: manual and Preferred manual method: Direct access keys
• Add some test values a secrets and try to install
• The integration should install correctly with no exceptions.

Checklist

• Unit or functional tests were updated or added to match the most common scenarios
• Flaky Test Runner was used on any tests changed

[apmmachine]

🤖 GitHub comments

Expand to view the GitHub comments

Just comment with:

• /oblt-deploy : Deploy a Kibana instance using the Observability test environments.
• /oblt-deploy-serverless : Deploy a serverless Kibana instance using the Observability test environments.
• run elasticsearch-ci/docs : Re-trigger the docs validation. (use unformatted text in the comment!)

[criamico]

@kpollich I found the reason of the original bug, the index used to get the secret in this block was not correct. In the specific case of CSPM we have values of this type:

 secrets:  [{ id: 'secret_id_0001' }];

secretPaths: [
  {
    path: [ 'inputs', '1', 'streams', '0', 'vars', 'secret_access_key' ],
    value: { type: 'text' }
  },
  {
    path: [ 'inputs', '2', 'streams', '0', 'vars', 'secret_access_key' ],
    value: { value: 'xfgdgh' }
  },
...
]

The issue was that secretPathIndex in this case was 1, hence causing the error. I fixed it by adding a second index that is incremented over secrets.

I also wrote this test for cases where there are two valid values in secretPaths (which it should be a feasible occurrence) and two secrets. I'd expect the code to update both values, but instead the policy returns unchanged.

My guess is that in this function acc, that gets the initial value from result, gets reset between iterations of forEach. In fact I printed the values between each iteration and it gets updated then reset again. I'm not really sure what could be the solution for this, since the spread operator is too expensive to use here and lodash set was ruled out before. I'd like to have your opinion here.

[criamico]

Thanks @kpollich for the commit, I'm going to open the PR with this change.

[elasticmachine]

Pinging @elastic/fleet (Team:Fleet)

[criamico]

@elasticmachine merge upstream

[kibana-ci]

💚 Build Succeeded

• Buildkite Build
• Commit: 362da9f

Metrics [docs]

✅ unchanged

History

• 💔 Build #186155 failed c6880dd

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

cc @criamico

[kibanamachine]

💚 All backports created successfully

Status	Branch	Result
✅	8.12

Note: Successful backport PRs will be merged automatically after passing CI.

Questions ?

Please refer to the Backport tool documentation