Nr 438909 multi binary health (#1615)

* feat: Add new Exec Health repository

* feat: Add id field to executables

* feat: Save supervisor healthiness into its repository

* feat: Add exec health checker

* feat: add ID to EBPF agent type executables

* feat: set exec health as the default

* feat: fix test failing because now there is always a health_checker spawned (default exec)

* feat: Modify exec checker to work with a buffered channel instead of repository

* feat: Modify healthiness variable to is_healthy for a more meaningful name.

* feat: Executable id can't be duplicated + Health is calculated adding Exec health and File/http health if present.

* feat: Modify the publishing of health events to not transform the error

* feat: Format code

* feat: Correct leftover from rebase

* feat: Add default to exectuables

* feat: Make exec id non templateable + add the non empty health status.

* feat: Fix test_file_health_without_supervisor.

Author: alvarocabanas

agent-control/agent-type-registry/newrelic/com.newrelic.ebpf-0.1.0.yaml

@@ -247,7 +247,8 @@ deployment:
   on_host:
     enable_file_logging: ${nr-var:enable_file_logging}
     executables:
-      - path: /usr/bin/nr-ebpf-agent-client
+      - id: nr-ebpf-agent-client
+        path: /usr/bin/nr-ebpf-agent-client
         env:
           NEW_RELIC_LICENSE_KEY: "${nr-env:NEW_RELIC_LICENSE_KEY}"
           DEPLOYMENT_NAME: "${nr-var:config_agent.DEPLOYMENT_NAME}"
@@ -292,7 +293,8 @@ deployment:
           backoff_strategy:
             type: fixed
             backoff_delay: ${nr-var:backoff_delay}
-      - path: /usr/bin/nr-ebpf-agent
+      - id: nr-ebpf-agent
+        path: /usr/bin/nr-ebpf-agent
         env:
           NEW_RELIC_LICENSE_KEY: "${nr-env:NEW_RELIC_LICENSE_KEY}"
           DEPLOYMENT_NAME: "${nr-var:config_agent.DEPLOYMENT_NAME}"

agent-control/agent-type-registry/newrelic/com.newrelic.infrastructure-0.1.0.yaml

@@ -117,7 +117,8 @@ deployment:
       args: --version
       regex: \d+\.\d+\.\d+
     executables:
-      - path: /usr/bin/newrelic-infra
+      - id: newrelic-infra
+        path: /usr/bin/newrelic-infra
         args: >-
           --config=${nr-var:config_agent}
         env:

agent-control/agent-type-registry/newrelic/io.opentelemetry.collector-0.1.0.yaml

@@ -83,6 +83,7 @@ deployment:
       regex: \d+\.\d+\.\d+
     executables:
       - # Important to note the binary name is nrdot-collector-host matching the new nrdot binary
+        id: nrdot-collector-host
         path: /usr/bin/nrdot-collector-host
         args: >-
           --config=${nr-var:config}

agent-control/src/agent_type/README.md

@@ -142,7 +142,8 @@ deployment:
       args: --version
       regex: \d+\.\d+\.\d+
     executables:
-      - path: /usr/bin/newrelic-infra
+      - id: newrelic-infra
+        path: /usr/bin/newrelic-infra
         args: "--config=${nr-var:config_agent}"
         env: "NRIA_PLUGIN_DIR=${nr-var:config_integrations} NRIA_STATUS_SERVER_ENABLED=true"
         restart_policy:
@@ -157,7 +158,8 @@ In this section:
 * `health`: The measures used to check the health status of the agent.
 * `version`: The command used to check the version of the binary.
 * `executables`: This outlines the list of binaries the agent supervisor runs. Developers can define:
-  - * `path`: The location of the binary required.
+  - * `id`: Unique identifier for the exec used by the health checker.
+    * `path`: The location of the binary required.
     * `args`: The command-line arguments needed by the binary.
     * `env`: Specifies the required environment variables.
     * `restart_policy`: The guidelines for if or when the process should be restarted.
@@ -402,7 +404,8 @@ variables:
 deployment:
   on_host:
     executables:
-      - path: /usr/bin/telegraf
+      - id: telegraf
+        path: /usr/bin/telegraf
         args: "--config ${nr-var:config_file}"
         env: ""
         restart_policy:

agent-control/src/agent_type/definition.rs

@@ -245,7 +245,8 @@ deployment:
         path: /healthz
         port: 8080
     executables:
-      - path: ${nr-var:bin}/otelcol
+      - id: otelcol
+        path: ${nr-var:bin}/otelcol
         args: "-c ${nr-var:deployment.k8s.image}"
         restart_policy:
           backoff_strategy:
@@ -265,7 +266,8 @@ spec:
 deployment:
   on_host:
     executables:
-      - path: ${nr-var:bin}/otelcol
+      - id: otelcol
+        path: ${nr-var:bin}/otelcol
         args: "-c ${nr-var:deployment.k8s.image}"
 "#;
 
@@ -379,7 +381,8 @@ deployment:
         path: /v1/status
         port: "${nr-var:status_server_port}"
     executables:
-      - path: /usr/bin/newrelic-infra
+      - id: newrelic-infra
+        path: /usr/bin/newrelic-infra
         args: "--config ${nr-var:config} --config2 ${nr-var:config2}"
 "#;
 
@@ -509,7 +512,8 @@ variables:
 deployment:
   on_host:
       executables:
-        - path: /bin/echo
+        - id: echo
+          path: /bin/echo
           args: "${nr-var:restart_policy.type}"
 "#;
 

agent-control/src/agent_type/render/persister/config_persister_file.rs

@@ -729,7 +729,8 @@ variables:
 deployment:
   on_host:
     executables:
-      - path: /usr/bin/newrelic-infra
+      - id: newrelic-infra
+        path: /usr/bin/newrelic-infra
         args: "--config=${config_file}"
         restart_policy:
           backoff_strategy:
@@ -774,7 +775,8 @@ variables:
 deployment:
   on_host:
     executables:
-      - path: /usr/bin/newrelic-infra
+      - id: newrelic-infra
+        path: /usr/bin/newrelic-infra
         args: "--config=${config_file1} --config=${config_file2} --config=${config_file3}"
         restart_policy:
           backoff_strategy:
@@ -808,7 +810,8 @@ variables:
 deployment:
   on_host:
     executables:
-      - path: /usr/bin/newrelic-infra
+      - id: newrelic-infra
+        path: /usr/bin/newrelic-infra
         args: "--config=${integrations}"
         restart_policy:
           backoff_strategy:
@@ -855,7 +858,8 @@ variables:
 deployment:
   on_host:
     executables:
-      - path: /usr/bin/newrelic-infra
+      - id: newrelic-infra
+        path: /usr/bin/newrelic-infra
         args: "--config=${integrations} --logging=${logging}"
         restart_policy:
           backoff_strategy:

agent-control/src/agent_type/render/renderer.rs

@@ -804,7 +804,8 @@ variables: {}
 deployment:
   on_host:
     executables:
-      - path: /opt/first
+      - id: first
+        path: /opt/first
         args: "${nr-ac:sa-fake-var}"
 "#,
             &Environment::OnHost,
@@ -864,9 +865,11 @@ variables:
 deployment:
   on_host:
     executables:
-      - path: /opt/first
+      - id: first
+        path: /opt/first
         args: "--config_path=${nr-var:config_path} --foo=${nr-var:config_argument}"
-      - path: /opt/second
+      - id: second
+        path: /opt/second
         args: "--config_path=${nr-var:config_path} --foo=${nr-var:config_argument}"
 "#;
 
@@ -897,7 +900,8 @@ variables:
 deployment:
   on_host:
     executables:
-      - path: /usr/bin/newrelic-infra
+      - id: newrelic-infra
+        path: /usr/bin/newrelic-infra
         args: "--config1 ${nr-var:config1} --config2 ${nr-var:config2}"
 "#;
 
@@ -941,7 +945,8 @@ variables:
 deployment:
   on_host:
     executables:
-      - path: /bin/otelcol
+      - id: otelcol
+        path: /bin/otelcol
         args: "-c some-arg"
         restart_policy:
           backoff_strategy:

agent-control/src/agent_type/runtime_config/health_config.rs

@@ -16,7 +16,7 @@ const DEFAULT_HEALTH_CHECK_TIMEOUT: Duration = Duration::from_secs(15);
 ///
 /// This structure includes parameters to define intervals between health checks,
 /// timeouts for checks, and the specific health check method—either HTTP or execute command.
-#[derive(Debug, Deserialize, Clone, PartialEq)]
+#[derive(Debug, Default, Deserialize, Clone, PartialEq)]
 pub struct OnHostHealthConfig {
     /// The duration to wait between health checks.
     #[serde(default)]
@@ -31,8 +31,8 @@ pub struct OnHostHealthConfig {
     pub(crate) timeout: HealthCheckTimeout,
 
     /// Details on the type of health check. Defined by the `HealthCheck` enumeration.
-    #[serde(flatten)]
-    pub(crate) check: OnHostHealthCheck,
+    #[serde(default, flatten)]
+    pub(crate) check: Option<OnHostHealthCheck>,
 }
 
 #[derive(Debug, Deserialize, Clone, Copy, PartialEq, WrapperWithDefault)]
@@ -171,7 +171,10 @@ impl Templateable for HttpPath {
 impl Templateable for OnHostHealthConfig {
     fn template_with(self, variables: &Variables) -> Result<Self, AgentTypeError> {
         Ok(Self {
-            check: self.check.template_with(variables)?,
+            check: self
+                .check
+                .map(|check| check.template_with(variables))
+                .transpose()?,
             ..self
         })
     }