break out okta endpoint from health check

[fzhao99]

DEVOPS PULL REQUEST

Related Issue

• Fixes Add health endpoint to handle degraded Okta status #7806

Changes Proposed

• Splits out the Okta health logic away from the backend health indicator
• Creates a specific okta degraded health state that's configured to return a 204 status code on trigger
• Modifies the frontend status display to indicate differential status
  • I added in the Okta check to flag out a failure in the deploy alerting script, which wasn't explicitly called out in the ticket but that duplicates what used to happen (thanks @mehansen for flagging!). Worth talking if we want to break this out to a separate alert but leaving it for now for status quo reasons.

Testing

I've deployed this test branch that hardcodes a "down" status from the Okta check. Double check that the frontend status display looks ok and that the health check in Azure looks ok

Can also pull the value down locally and hard code return values for getApplicationStatusForHealthCheck in the demo okta repository to observe changes on the backend API return on port 8080 or on the frontend status page

---

[mehansen]

left a couple questions and nit comments but overall looks good! fun to see how you could set a custom status code

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
94.1% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarCloud

[DanielSass]

The pedantic nerd inside me debates the correctness of returning a 204/NO CONTENT. That's not to say that there is an existing 2xx status code that would be correct. But I'd almost be tempted to use a 2xx that doesn't have an existing status (like 299 for example).

At the end of the day I don't think it matters.

[DanielSass]

Awesome work, thanks Bob!

[mehansen]

lgtm! thanks Bob

[emyl3]

LGTM! 😸