Add note about '/liveness' endpoint for standalone agent

[kilfoyle]

This updates the "Configure monitoring for standalone Elastic Agents" docs page to include mention of the /liveness endpoint added for 8.15. Rel: elastic/elastic-agent#107 (comment)

ADD: Please see my comment below for an additional change to the Fleet-managed agent docs.

---

Closes: #1090

[github-actions]

A documentation preview will be available soon.

• 🔨 Buildkite builds
• 📚 HTML diff
• 📙 Preview page

Request a new doc build by commenting

• Rebuild this PR: run docs-build
• Rebuild this PR and all Elastic docs: run docs-build rebuild

_{run docs-build is much faster than run docs-build rebuild. A rebuild should only be needed in rare situations.}

_{If your PR continues to fail for an unknown reason, the doc build pipeline may be broken. Elastic employees can check the pipeline status here.}

[shainaraskas]

approved w/ suggest!

[kilfoyle]

ADD: The "Enable agent monitoring" setting in Fleet wasn't documented at all, so I've added it to the Elastic Agent policies page that applies to Fleet-managed agents.

@cmacknz @nimarezainia If the /liveness endpoint should be mentioned in the Fleet docs, maybe we should provide an example of how to call it? As Shaina mentioned, the "where that server is configured to be exposed" part might not be clear, so an example could help.

---

[karenzone]

LGTM

[cmacknz]

It's actually agent.monitoring.enabled.http that needs to be enabled, agent.monitoring.enabled controls whether we collect monitoring data from agent to be shipped to Fleet.

The liveness probe is an HTTP feature. Sorry for the late review.

[kilfoyle]

Thanks for spotting this @cmacknz!

A couple of questions:

• Just to make sure I get it right, should it be agent.monitoring.enabled.http or agent.monitoring.http.enabled? In your comment you mentioned the latter.
• I don't think the setting is available to in Fleet, so I think I should mention the setting only in the standalone agent docs (here). Does that seem right to you?

ADD: Here's a PR with the changes: #1100

[zez3]

The current documentation still does not reflect the new liveness option

@kilfoyle can you please fix that?

[nimarezainia]

@zez3 seems to be here: https://www.elastic.co/guide/en/fleet/8.15/elastic-agent-monitoring-configuration.html

[kilfoyle]

Thanks @nimarezainia!

[zez3]

So this will work only for standalone Agents?

[zez3]

It can be configured to also monitor the component states and return an error if anything is degraded or has failed.

How?

[zez3]

So this will work only for standalone Agents?

I see elastic/kibana#180922 (comment) , so this is not yet fully functional in fleet

[cmacknz]

It isn't in the Fleet UI yet, but the Fleet overrides API will let you control it. We have tests proving Fleet can control the parameters, there just isn't a UI box for it yet.

The request to turn on the HTTP endpoint, which also enables the /liveness endpoint would look something like the following from dev tools:

PUT kbn:/api/fleet/agent_policies/65df03d7-5d26-490c-a24a-ae05095c1a4a
{
    "name": "Agent policy 1",
    "namespace": "default",
    "overrides": {
      "agent": {
          "monitoring": map[string]interface{}{
            "http": map[string]interface{}{
              "enabled": true,
              "host":    "localhost",
              "port":    6792,
            },
        }
    }
}

There is a failon query parameter that controls what it considers "live" to be: https://github.com/elastic/elastic-agent/blob/44528c49507c18e71a4cfd2a6ba4f0904bd32009/elastic-agent.reference.yml#L155-L168

[nimarezainia]

Fleet UI changes: elastic/kibana#153950