[Breaking] Remove experimental_prometheus and disable_metrics (#1686)

These options were never useful as the metrics exposed didn't follow
usable Prometheus structure. This change removes the public-facing
configuration options related to the prometheus service. We keep the
internals of the metric gathering infrastructure around as that's being
migrated to a new OpenTelemetry setup.

Fixes GHSA-2gh3-rmm4-6rq5

Author: aaronmondal

BUILD.bazel

@@ -32,10 +32,8 @@ rust_binary(
         "@crates//:hyper-util",
         "@crates//:mimalloc",
         "@crates//:opentelemetry",
-        "@crates//:opentelemetry-prometheus",
         "@crates//:opentelemetry_sdk",
         "@crates//:parking_lot",
-        "@crates//:prometheus",
         "@crates//:rustls-pemfile",
         "@crates//:scopeguard",
         "@crates//:serde_json",

Cargo.lock

@@ -57,9 +57,6 @@ tracing = { version = "0.1.41", default-features = false }
 opentelemetry_sdk = { version = "0.29.0", default-features = false }
 tracing-subscriber = { version = "0.3.19", default-features = false }
 opentelemetry = { version = "0.29.1", default-features = false }
-prometheus = { version = "0.13.4", default-features = false }
-opentelemetry-prometheus = "0.29.0"
-serde_json = "1.0.140"
 
 [workspace.cargo-features-manager.keep]
 async-lock = ["std"]

Cargo.toml

@@ -59,19 +59,6 @@
       }
     }
   }, {
-    // Only publish metrics on a private port.
-    "listener": {
-      "http": {
-        "socket_address": "0.0.0.0:50061"
-      }
-    },
-    "services": {
-      "experimental_prometheus": {
-        "path": "/metrics"
-      }
-    }
-  },
-  {
     "listener": {
       "http": {
         "socket_address": "0.0.0.0:50071",

deployment-examples/docker-compose/local-storage-cas.json5

@@ -72,19 +72,6 @@
     }
   ],
   "servers": [{
-    // Only publish metrics on a private port.
-    "listener": {
-      "http": {
-        "socket_address": "0.0.0.0:9090"
-      }
-    },
-    "services": {
-      "experimental_prometheus": {
-        "path": "/metrics"
-      }
-    }
-  },
-  {
     "listener": {
       "http": {
         "socket_address": "0.0.0.0:50051"

integration_tests/simple_prometheus_test.sh

@@ -312,7 +312,7 @@ The `public` server consists of a `listener` object and a `services` object. The
 
 > ⚠️ _WARNING_: A private server shouldn't be exposed to the public. ⚠️
 
-The `private` server consists of a `listener` object and a `services` object. The `listener` object is one level and includes an `http` with a `socket address`. The `services` server consists of an `experimental_prometheus` object with a `path` field, a `worker_api` object with `scheduler_field`, and an `admin` object.
+The `private` server consists of a `listener` object and a `services` object. The `listener` object is one level and includes an `http` with a `socket address`. The `services` server consists of a `worker_api` object with `scheduler_field` and an `admin` object.
 
 ```json5
  {
@@ -364,9 +364,6 @@ The `private` server consists of a `listener` object and a `services` object. Th
       }
     },
     "services": {
-      "experimental_prometheus": {
-        "path": "/metrics"
-      },
       // Note: This should be served on a different port, because it has
       // a different permission set than the other services.
       // In other words, this service is a backend api. The ones above
@@ -548,9 +545,6 @@ Below, you will find a fully tested example that you can also find in [basic_cas
       }
     },
     "services": {
-      "experimental_prometheus": {
-        "path": "/metrics"
-      },
       // Note: This should be served on a different port, because it has
       // a different permission set than the other services.
       // In other words, this service is a backend api. The ones above

kubernetes/nativelink/nativelink-config.json5

@@ -142,9 +142,6 @@
       }
     },
     "services": {
-      "experimental_prometheus": {
-        "path": "/metrics"
-      },
       // Note: This should be served on a different port, because it has
       // a different permission set than the other services.
       // In other words, this service is a backend api. The ones above

nativelink-config/examples/README.md

@@ -69,18 +69,6 @@
           }
         }
       }
-    },
-    {
-      "listener": {
-        "http": {
-          "socket_address": "0.0.0.0:50061"
-        }
-      },
-      "services": {
-        "experimental_prometheus": {
-          "path": "/metrics"
-        }
-      }
     }
   ]
 }

nativelink-config/examples/basic_cas.json5

@@ -167,18 +167,6 @@ pub struct WorkerApiConfig {
     pub scheduler: SchedulerRefName,
 }
 
-#[derive(Deserialize, Debug, Default)]
-#[serde(deny_unknown_fields)]
-pub struct PrometheusConfig {
-    /// Path to register prometheus metrics. If path is "/metrics", and your
-    /// domain is "example.com", you can reach the endpoint with:
-    /// <http://example.com/metrics>.
-    ///
-    /// Default: "/metrics"
-    #[serde(default)]
-    pub path: String,
-}
-
 #[derive(Deserialize, Debug, Default)]
 #[serde(deny_unknown_fields)]
 pub struct AdminConfig {
@@ -287,10 +275,6 @@ pub struct ServicesConfig {
     /// publish them to a store for processing by an external service.
     pub experimental_bep: Option<BepConfig>,
 
-    /// Experimental - Prometheus metrics configuration. Metrics are gathered
-    /// as a singleton but may be served on multiple endpoints.
-    pub experimental_prometheus: Option<PrometheusConfig>,
-
     /// This is the service for any administrative tasks.
     /// It provides a REST API endpoint for administrative purposes.
     pub admin: Option<AdminConfig>,
@@ -716,19 +700,6 @@ pub struct GlobalConfig {
     #[serde(deserialize_with = "convert_numeric_with_shellexpand")]
     pub max_open_files: usize,
 
-    /// This flag can be used to prevent metrics from being collected at runtime.
-    /// Metrics are still able to be collected, but this flag prevents metrics that
-    /// are collected at runtime (performance metrics) from being tallied. The
-    /// overhead of collecting metrics is very low, so this flag should only be
-    /// used if there is a very good reason to disable metrics.
-    /// This flag can be forcibly set using the `NATIVELINK_DISABLE_METRICS` variable.
-    /// If the variable is set it will always disable metrics regardless of what
-    /// this flag is set to.
-    ///
-    /// Default: <true (disabled) if no prometheus service enabled, false otherwise>
-    #[serde(default)]
-    pub disable_metrics: bool,
-
     /// Default hash function to use while uploading blobs to the CAS when not set
     /// by client.
     ///