engine: use 0.0.0.0 for metrics-addr

[dvdksn]

• Closes metrics-addr should be 0.0.0.0:9323 for a dockerized prometheus to reach it #21632

[netlify]

✅ Deploy Preview for docsdocker ready!

Name	Link
🔨 Latest commit	7e4fd0f
🔍 Latest deploy log	https://app.netlify.com/sites/docsdocker/deploys/67614686d57fd70008e8d8b3
😎 Deploy Preview	https://deploy-preview-21643--docsdocker.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify site configuration.

[thaJeztah]

We need to fix some of this; there was a proposal to expose it through the Engine API instead of separate, but I also need to dust off the work I started to expose listeners in docker info

• Prometheus Metrics API exposure moby/moby#31107
• api: add "Listeners" field to /info, to help discover how the API is exposed moby/moby#43459

So... I just recalled there may be a different way; as part of docker plugin, there was a PR to allow "metrics plugins";

• Add support for metrics plugins moby/moby#32874

And the daemon also actually starts the metrics endpoint listening on a socket; https://github.com/moby/moby/blob/8d3ecc28c1e886c59f4d23b9d1e4b35aa40696ba/daemon/metrics_unix.go#L24-L46

docker run --rm -v /var/run/docker/metrics.sock:/var/run/docker/metrics.sock alpine sh -c 'apk add --quiet --no-cache curl && curl -s --unix-socket /var/run/docker/metrics.sock http://localhost/metrics | head -n10'
# HELP builder_builds_failed_total Number of failed image builds
# TYPE builder_builds_failed_total counter
builder_builds_failed_total{reason="build_canceled"} 0
builder_builds_failed_total{reason="build_target_not_reachable_error"} 0
builder_builds_failed_total{reason="command_not_supported_error"} 0
builder_builds_failed_total{reason="dockerfile_empty_error"} 0
builder_builds_failed_total{reason="dockerfile_syntax_error"} 0
builder_builds_failed_total{reason="error_processing_commands_error"} 0
builder_builds_failed_total{reason="missing_onbuild_arguments_error"} 0
builder_builds_failed_total{reason="unknown_instruction_error"} 0

Given that that one is inside /var/run/docker I'm not sure if we should document that though (although it's probably OK(ish)). The alternative may be to configure the daemon to use a socket instead of tcp

We should probably look if we can make the config allow for a unix:/// socket to be set instead of just a tcp address.

[thaJeztah]

LGTM for this change though

[dvdksn]

What's the state of engine plugins anyway?