Add health_check extension to bundled otel.yml file

[ycombinator]

What does this PR do?

This PR adds the health_check extension to the otel.yml file that's bundled with the Elastic Agent package.

Why is it important?

Without this addition, if a user tries to install the open-telemetry/opentelemetry-collector Helm chart with Elastic's distribution of the OTel Collector (i.e. Elastic Agent), the installation eventually fails. Specifically, as noted by @andrzej-stencel in #5092 (comment):

One problem I'm seeing when running the chart with this configuration is that the Health Check extension is not used in the otel.yml file, which results in Kubernetes restarting the collector container and eventually resulting in a CrashLoopBackOff. We should add the health_check extension to service::extensions in the otel.yml file that's bundled with the image.

Checklist

• My code follows the style guidelines of this project
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• I have made corresponding change to the default configuration files
• I have added tests that prove my fix is effective or that my feature works
• I have added an entry in ./changelog/fragments using the changelog tool
• I have added an integration test or an E2E test

Disruptive User Impact

If users are already running something that's listening on localhost:13133, there will be a port conflict.

How to test this PR locally

See #5092 (comment) and #5092 (comment)

[ycombinator]

I chose this value based on the documentation in https://github.com/open-telemetry/opentelemetry-collector-contrib/blob/main/extension/healthcheckextension/README.md#health-check. I'm happy to change it to another value if others have a better suggestion.

[elasticmachine]

Pinging @elastic/elastic-agent-control-plane (Team:Elastic-Agent-Control-Plane)

[rogercoll]

Sounds good to me on adding the health_check extension in the bundled otel.yml. 👍

Without this addition, if a user tries to install the open-telemetry/opentelemetry-collector Helm chart with Elastic's distribution of the OTel Collector (i.e. Elastic Agent), the installation eventually fails.

I think that this will only happen if the user tries to use the otel.yml file as it overrides the configuration set by the Chart:

$ kubectl describe pod  elastic-otelcol-opentelemetry-collector-agent-d9s4k
...
Command:
      //usr/share/elastic-agent/elastic-agent
    Args:
      --config=/conf/relay.yaml
      otel
      --config=otel.yml

Note that the Chart added another default config --config=/conf/relay.yaml (which contains the health_check extension). A workaround is to not use the otel.yml config and rely on the one configured by the Chart:

helm install elastic-otelcol open-telemetry/opentelemetry-collector --version=0.102.1 --set mode=daemonset --set image.repository="docker.elastic.co/beats/elastic-agent" --set image.tag="8.16.0-SNAPSHOT" --set image.pullPolicy=Always --set command.name="/usr/share/elastic-agent/elastic-agent" --set command.extraArgs="{otel}"

The recommened approach to provide a custom configuration for the collector is to manully create a configuration configMap. There is a PR upstream that will allow overriding the default configuration with a config option: open-telemetry/opentelemetry-helm-charts#1301

[elastic-sonarqube]

Quality Gate passed

Issues
0 New issues
0 Fixed issues
0 Accepted issues

Measures
0 Security Hotspots
No data about Coverage
No data about Duplication

See analysis details on SonarQube

[blakerouse]

Looks good, thanks for the updating the path.