wip:added alertreceiver and alert rules which are built and added to the yaml before deploying prometheus, added a few dashboards to grafana. Trying to fix prometheus-server clusterrole/role/serviceaccount so that it can discover targets and kubernetes in a namespaced release where it does not have access to clusterrole

Co-authored-by: tahahawa <tahahawa@gmail.com>
Reviewed-on: #80

examples/rust/src/main.rs

@@ -3,8 +3,18 @@ use std::{path::PathBuf, sync::Arc};
 use harmony::{
     inventory::Inventory,
     maestro::Maestro,
-    modules::application::{
-        RustWebFramework, RustWebapp, RustWebappScore, features::ContinuousDelivery,
+    modules::{
+        application::{
+            ApplicationScore, RustWebFramework, RustWebapp,
+            features::{ContinuousDelivery, PrometheusMonitoring},
+        },
+        monitoring::{
+            alert_channel::discord_alert_channel::DiscordWebhook,
+            alert_rule::prometheus_alert_rule::AlertManagerRuleGroup,
+        },
+        prometheus::alerts::k8s::{
+            pod::pod_in_failed_state, pvc::high_pvc_fill_rate_over_two_days,
+        },
     },
     topology::{K8sAnywhereTopology, Url},
 };
@@ -12,18 +22,34 @@ use harmony::{
 #[tokio::main]
 async fn main() {
     env_logger::init();
-    let application = RustWebapp {
+    let application = Arc::new(RustWebapp {
         name: "harmony-example-rust-webapp".to_string(),
+        domain: Url::Url(url::Url::parse("https://rustapp.harmony.example.com").unwrap()),
         project_root: PathBuf::from("./examples/rust/webapp"),
         framework: Some(RustWebFramework::Leptos),
-    };
-    // TODO RustWebappScore should simply take a RustWebApp as config
-    let app = RustWebappScore {
-        name: "Example Rust Webapp".to_string(),
-        domain: Url::Url(url::Url::parse("https://rustapp.harmony.example.com").unwrap()),
-        features: vec![Box::new(ContinuousDelivery {
-            application: Arc::new(application.clone()),
-        })],
+    });
+
+    let pod_failed = pod_in_failed_state();
+    let pod_failed_2 = pod_in_failed_state();
+    let pod_failed_3 = pod_in_failed_state();
+
+    let additional_rules = AlertManagerRuleGroup::new("pod-alerts", vec![pod_failed]);
+    let additional_rules_2 = AlertManagerRuleGroup::new("pod-alerts-2", vec![pod_failed_2, pod_failed_3]);
+    let app = ApplicationScore {
+        features: vec![
+            //Box::new(ContinuousDelivery {
+            //    application: application.clone(),
+            //}),
+            Box::new(PrometheusMonitoring {
+                application: application.clone(),
+                alert_receivers: vec![Box::new(DiscordWebhook {
+                    name: "dummy-discord".to_string(),
+                    url: Url::Url(url::Url::parse("https://discord.doesnt.exist.com").unwrap()),
+                })],
+                alert_rules: vec![Box::new(additional_rules), Box::new(additional_rules_2)],
+            }),
+            // TODO add monitoring, backups, multisite ha, etc
+        ],
         application,
     };
 

harmony/Cargo.toml

@@ -13,7 +13,7 @@ reqwest = { version = "0.11", features = ["blocking", "json"] }
 russh = "0.45.0"
 rust-ipmi = "0.1.1"
 semver = "1.0.23"
-serde = { version = "1.0.209", features = ["derive"] }
+serde = { version = "1.0.209", features = ["derive", "rc"] }
 serde_json = "1.0.127"
 tokio.workspace = true
 derive-new.workspace = true

harmony/src/domain/interpret/mod.rs

@@ -22,6 +22,7 @@ pub enum InterpretName {
     K3dInstallation,
     TenantInterpret,
     Application,
+    ArgoCD,
 }
 
 impl std::fmt::Display for InterpretName {
@@ -39,6 +40,7 @@ impl std::fmt::Display for InterpretName {
             InterpretName::K3dInstallation => f.write_str("K3dInstallation"),
             InterpretName::TenantInterpret => f.write_str("Tenant"),
             InterpretName::Application => f.write_str("Application"),
+            InterpretName::ArgoCD => f.write_str("ArgoCD"),
         }
     }
 }

harmony/src/domain/score_with_dep.rs

@@ -0,0 +1,59 @@
+////////////////////
+/// Working idea 
+///
+///
+trait ScoreWithDep<T> {
+    fn create_interpret(&self) -> Box<dyn Interpret<T>>;
+    fn name(&self) -> String;
+    fn get_dependencies(&self) -> Vec<TypeId>; // Force T to impl Installer<TypeId> or something
+                                               // like that
+}
+
+struct PrometheusAlertScore;
+
+impl <T> ScoreWithDep<T> for PrometheusAlertScore {
+    fn create_interpret(&self) -> Box<dyn Interpret<T>> {
+        todo!()
+    }
+
+    fn name(&self) -> String {
+        todo!()
+    }
+
+    fn get_dependencies(&self) -> Vec<TypeId> {
+        // We have to find a way to constrait here so at compile time we are only allowed to return
+        // TypeId for types which can be installed by T 
+        //
+        // This means, for example that T must implement HelmCommand if the impl <T: HelmCommand> Installable<T> for
+        // KubePrometheus calls for HelmCommand.
+        vec![TypeId::of::<KubePrometheus>()]
+    }
+}
+
+trait Installable{}
+
+struct KubePrometheus;
+
+impl Installable for KubePrometheus;
+ 
+
+struct Maestro<T> {
+    topology: T
+}
+
+impl <T>Maestro<T> {
+  fn execute_store(&self, score: ScoreWithDep<T>) {
+      score.get_dependencies().iter().for_each(|dep| {
+          self.topology.ensure_dependency_ready(dep);
+      });
+  }
+}
+
+struct TopologyWithDep {
+}
+
+impl TopologyWithDep {
+    fn ensure_dependency_ready(&self, type_id: TypeId) -> Result<(), String> {
+        self.installer
+    }
+}

harmony/src/domain/topology/k8s.rs

@@ -4,8 +4,6 @@ use k8s_openapi::{
     ClusterResourceScope, NamespaceResourceScope,
     api::{apps::v1::Deployment, core::v1::Pod},
 };
-use kube::runtime::conditions;
-use kube::runtime::wait::await_condition;
 use kube::{
     Client, Config, Error, Resource,
     api::{Api, AttachParams, ListParams, Patch, PatchParams, ResourceExt},
@@ -13,6 +11,11 @@ use kube::{
     core::ErrorResponse,
     runtime::reflector::Lookup,
 };
+use kube::{api::DynamicObject, runtime::conditions};
+use kube::{
+    api::{ApiResource, GroupVersionKind},
+    runtime::wait::await_condition,
+};
 use log::{debug, error, trace};
 use serde::de::DeserializeOwned;
 use similar::{DiffableStr, TextDiff};
@@ -239,6 +242,54 @@ impl K8sClient {
         Ok(result)
     }
 
+    pub async fn apply_yaml_many(
+        &self,
+        yaml: &Vec<serde_yaml::Value>,
+        ns: Option<&str>,
+    ) -> Result<(), Error> {
+        for y in yaml.iter() {
+            self.apply_yaml(y, ns).await?;
+        }
+        Ok(())
+    }
+
+    pub async fn apply_yaml(
+        &self,
+        yaml: &serde_yaml::Value,
+        ns: Option<&str>,
+    ) -> Result<(), Error> {
+        let obj: DynamicObject = serde_yaml::from_value(yaml.clone()).expect("TODO do not unwrap");
+        let name = obj.metadata.name.as_ref().expect("YAML must have a name");
+        let namespace = obj
+            .metadata
+            .namespace
+            .as_ref()
+            .expect("YAML must have a namespace");
+
+        // 4. Define the API resource type using the GVK from the object.
+        //    The plural name 'applications' is taken from your CRD definition.
+        error!("This only supports argocd application harcoded, very rrrong");
+        let gvk = GroupVersionKind::gvk("argoproj.io", "v1alpha1", "Application");
+        let api_resource = ApiResource::from_gvk_with_plural(&gvk, "applications");
+
+        // 5. Create a dynamic API client for this resource type.
+        let api: Api<DynamicObject> =
+            Api::namespaced_with(self.client.clone(), namespace, &api_resource);
+
+        // 6. Apply the object to the cluster using Server-Side Apply.
+        //    This will create the resource if it doesn't exist, or update it if it does.
+        println!(
+            "Applying Argo Application '{}' in namespace '{}'...",
+            name, namespace
+        );
+        let patch_params = PatchParams::apply("harmony"); // Use a unique field manager name
+        let result = api.patch(name, &patch_params, &Patch::Apply(&obj)).await?;
+
+        println!("Successfully applied '{}'.", result.name_any());
+
+        Ok(())
+    }
+
     pub(crate) async fn from_kubeconfig(path: &str) -> Option<K8sClient> {
         let k = match Kubeconfig::read_from(path) {
             Ok(k) => k,

harmony/src/domain/topology/k8s_anywhere.rs

@@ -246,7 +246,7 @@ pub struct K8sAnywhereConfig {
     ///
     /// default: true
     pub use_local_k3d: bool,
-    harmony_profile: String,
+    pub harmony_profile: String,
 }
 
 impl K8sAnywhereConfig {

harmony/src/domain/topology/oberservability/monitoring.rs

@@ -1,3 +1,5 @@
+use std::any::Any;
+
 use async_trait::async_trait;
 use log::debug;
 
@@ -9,7 +11,7 @@ use crate::{
 };
 
 #[async_trait]
-pub trait AlertSender: Send + Sync + std::fmt::Debug {
+pub trait AlertSender: Any + Send + Sync + std::fmt::Debug {
     fn name(&self) -> String;
 }
 

harmony/src/modules/application/feature.rs

@@ -3,7 +3,6 @@ use serde::Serialize;
 
 use crate::topology::Topology;
 
-use super::Application;
 /// An ApplicationFeature provided by harmony, such as Backups, Monitoring, MultisiteAvailability,
 /// ContinuousIntegration, ContinuousDelivery
 #[async_trait]

harmony/src/modules/application/features/argo_types.rs

@@ -0,0 +1,357 @@
+use std::{backtrace, collections::HashMap};
+
+use k8s_openapi::{Metadata, NamespaceResourceScope, Resource};
+use log::debug;
+use serde::Serialize;
+use serde_yaml::Value;
+use url::Url;
+
+use crate::modules::application::features::CDApplicationConfig;
+
+#[derive(Clone, Debug, Serialize)]
+#[serde(rename_all = "camelCase")]
+pub struct Helm {
+    pub pass_credentials: Option<bool>,
+    pub parameters: Vec<Value>,
+    pub file_parameters: Vec<Value>,
+    pub release_name: Option<String>,
+    pub value_files: Vec<String>,
+    pub ignore_missing_value_files: Option<bool>,
+    pub values: Option<String>,
+    pub values_object: Option<Value>,
+    pub skip_crds: Option<bool>,
+    pub skip_schema_validation: Option<bool>,
+    pub version: Option<String>,
+    pub kube_version: Option<String>,
+    pub api_versions: Vec<String>,
+    pub namespace: Option<String>,
+}
+
+#[derive(Clone, Debug, Serialize)]
+#[serde(rename_all = "camelCase")]
+pub struct Source {
+    pub repo_url: Url,
+    pub target_revision: Option<String>,
+    pub chart: String,
+    pub helm: Helm,
+}
+
+#[derive(Clone, Debug, Serialize)]
+#[serde(rename_all = "camelCase")]
+pub struct Automated {
+    pub prune: bool,
+    pub self_heal: bool,
+    pub allow_empty: bool,
+}
+
+#[derive(Clone, Debug, Serialize)]
+#[serde(rename_all = "camelCase")]
+pub struct Backoff {
+    pub duration: String,
+    pub factor: u32,
+    pub max_duration: String,
+}
+
+#[derive(Clone, Debug, Serialize)]
+#[serde(rename_all = "camelCase")]
+pub struct Retry {
+    pub limit: u32,
+    pub backoff: Backoff,
+}
+
+#[derive(Clone, Debug, Serialize)]
+#[serde(rename_all = "camelCase")]
+pub struct SyncPolicy {
+    pub automated: Automated,
+    pub sync_options: Vec<String>,
+    pub retry: Retry,
+}
+
+#[derive(Clone, Debug, Serialize)]
+#[serde(rename_all = "camelCase")]
+pub struct ArgoApplication {
+    pub name: String,
+    pub namespace: Option<String>,
+    pub project: String,
+    pub source: Source,
+    pub sync_policy: SyncPolicy,
+    pub revision_history_limit: u32,
+}
+
+impl Default for ArgoApplication {
+    fn default() -> Self {
+        Self {
+            name: Default::default(),
+            namespace: Default::default(),
+            project: Default::default(),
+            source: Source {
+                repo_url: Url::parse("http://asdf").expect("Couldn't parse to URL"),
+                target_revision: None,
+                chart: "".to_string(),
+                helm: Helm {
+                    pass_credentials: None,
+                    parameters: vec![],
+                    file_parameters: vec![],
+                    release_name: None,
+                    value_files: vec![],
+                    ignore_missing_value_files: None,
+                    values: None,
+                    values_object: None,
+                    skip_crds: None,
+                    skip_schema_validation: None,
+                    version: None,
+                    kube_version: None,
+                    api_versions: vec![],
+                    namespace: None,
+                },
+            },
+            sync_policy: SyncPolicy {
+                automated: Automated {
+                    prune: false,
+                    self_heal: false,
+                    allow_empty: false,
+                },
+                sync_options: vec![],
+                retry: Retry {
+                    limit: 5,
+                    backoff: Backoff {
+                        duration: "5s".to_string(),
+                        factor: 2,
+                        max_duration: "3m".to_string(),
+                    },
+                },
+            },
+            revision_history_limit: 10,
+        }
+    }
+}
+
+impl From<CDApplicationConfig> for ArgoApplication {
+    fn from(value: CDApplicationConfig) -> Self {
+        Self {
+            name: value.name,
+            namespace: Some(value.namespace),
+            project: "default".to_string(),
+            source: Source {
+                repo_url: Url::parse(value.helm_chart_repo_url.to_string().as_str())
+                    .expect("couldn't convert to URL"),
+                target_revision: None,
+                chart: value.helm_chart_name,
+                helm: Helm {
+                    pass_credentials: None,
+                    parameters: vec![],
+                    file_parameters: vec![],
+                    release_name: None,
+                    value_files: vec![],
+                    ignore_missing_value_files: None,
+                    values: None,
+                    values_object: Some(value.values_overrides),
+                    skip_crds: None,
+                    skip_schema_validation: None,
+                    version: None,
+                    kube_version: None,
+                    api_versions: vec![],
+                    namespace: None,
+                },
+            },
+            sync_policy: SyncPolicy {
+                automated: Automated {
+                    prune: false,
+                    self_heal: false,
+                    allow_empty: true,
+                },
+                sync_options: vec![],
+                retry: Retry {
+                    limit: 5,
+                    backoff: Backoff {
+                        duration: "5s".to_string(),
+                        factor: 2,
+                        max_duration: "3m".to_string(),
+                    },
+                },
+            },
+            ..Self::default()
+        }
+    }
+}
+
+impl ArgoApplication {
+    pub fn to_yaml(&self) -> serde_yaml::Value {
+        let name = &self.name;
+        let namespace = if let Some(ns) = self.namespace.as_ref() {
+            &ns
+        } else {
+            "argocd"
+        };
+        let project = &self.project;
+        let source = &self.source;
+
+        let yaml_str = format!(
+            r#"
+apiVersion: argoproj.io/v1alpha1
+kind: Application
+metadata:
+  name: {name}
+  # You'll usually want to add your resources to the argocd namespace.
+  namespace: {namespace}
+spec:
+  # The project the application belongs to.
+  project: {project}
+
+  # Destination cluster and namespace to deploy the application
+  destination:
+    # cluster API URL
+    server: https://kubernetes.default.svc
+    # or cluster name
+    # name: in-cluster
+    # The namespace will only be set for namespace-scoped resources that have not set a value for .metadata.namespace
+    namespace: {namespace}
+
+"#
+        );
+
+        let mut yaml_value: Value =
+            serde_yaml::from_str(yaml_str.as_str()).expect("couldn't parse string to YAML");
+
+        let mut spec = yaml_value
+            .get_mut("spec")
+            .expect("couldn't get spec from yaml")
+            .as_mapping_mut()
+            .expect("couldn't unwrap spec as mutable mapping");
+
+        let source =
+            serde_yaml::to_value(&self.source).expect("couldn't serialize source to value");
+        let sync_policy = serde_yaml::to_value(&self.sync_policy)
+            .expect("couldn't serialize sync_policy to value");
+        let revision_history_limit = serde_yaml::to_value(&self.revision_history_limit)
+            .expect("couldn't serialize revision_history_limit to value");
+
+        spec.insert(
+            serde_yaml::to_value("source").expect("string to value failed"),
+            source,
+        );
+        spec.insert(
+            serde_yaml::to_value("syncPolicy").expect("string to value failed"),
+            sync_policy,
+        );
+        spec.insert(
+            serde_yaml::to_value("revisionHistoryLimit")
+                .expect("couldn't convert str to yaml value"),
+            revision_history_limit,
+        );
+
+        debug!("spec: {}", serde_yaml::to_string(spec).unwrap());
+        debug!(
+            "entire yaml_value: {}",
+            serde_yaml::to_string(&yaml_value).unwrap()
+        );
+
+        yaml_value
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use url::Url;
+
+    use crate::modules::application::features::{
+        ArgoApplication, Automated, Backoff, Helm, Retry, Source, SyncPolicy,
+    };
+
+    #[test]
+    fn test_argo_application_to_yaml_happy_path() {
+        let app = ArgoApplication {
+            name: "test".to_string(),
+            namespace: Some("test-ns".to_string()),
+            project: "test-project".to_string(),
+            source: Source {
+                repo_url: Url::parse("http://test").unwrap(),
+                target_revision: None,
+                chart: "test-chart".to_string(),
+                helm: Helm {
+                    pass_credentials: None,
+                    parameters: vec![],
+                    file_parameters: vec![],
+                    release_name: Some("test-release-neame".to_string()),
+                    value_files: vec![],
+                    ignore_missing_value_files: None,
+                    values: None,
+                    values_object: None,
+                    skip_crds: None,
+                    skip_schema_validation: None,
+                    version: None,
+                    kube_version: None,
+                    api_versions: vec![],
+                    namespace: None,
+                },
+            },
+            sync_policy: SyncPolicy {
+                automated: Automated {
+                    prune: false,
+                    self_heal: false,
+                    allow_empty: false,
+                },
+                sync_options: vec![],
+                retry: Retry {
+                    limit: 5,
+                    backoff: Backoff {
+                        duration: "5s".to_string(),
+                        factor: 2,
+                        max_duration: "3m".to_string(),
+                    },
+                },
+            },
+            revision_history_limit: 10,
+        };
+
+        let expected_yaml_output = r#"apiVersion: argoproj.io/v1alpha1
+kind: Application
+metadata:
+  name: test
+  namespace: test-ns
+spec:
+  project: test-project
+  destination:
+    server: https://kubernetes.default.svc
+    namespace: test-ns
+  source:
+    repoUrl: http://test/
+    targetRevision: null
+    chart: test-chart
+    helm:
+      passCredentials: null
+      parameters: []
+      fileParameters: []
+      releaseName: test-release-neame
+      valueFiles: []
+      ignoreMissingValueFiles: null
+      values: null
+      valuesObject: null
+      skipCrds: null
+      skipSchemaValidation: null
+      version: null
+      kubeVersion: null
+      apiVersions: []
+      namespace: null
+  syncPolicy:
+    automated:
+      prune: false
+      selfHeal: false
+      allowEmpty: false
+    syncOptions: []
+    retry:
+      limit: 5
+      backoff:
+        duration: 5s
+        factor: 2
+        maxDuration: 3m
+  revisionHistoryLimit: 10"#;
+
+        assert_eq!(
+            expected_yaml_output.trim(),
+            serde_yaml::to_string(&app.clone().to_yaml())
+                .unwrap()
+                .trim()
+        );
+    }
+}

harmony/src/modules/application/features/continuous_delivery.rs

@@ -2,7 +2,7 @@ use std::{io::Write, process::Command, sync::Arc};
 
 use async_trait::async_trait;
 use log::{error, info};
-use serde_json::Value;
+use serde_yaml::Value;
 use tempfile::NamedTempFile;
 
 use crate::{
@@ -10,11 +10,14 @@ use crate::{
     data::Version,
     inventory::Inventory,
     modules::{
-        application::{Application, ApplicationFeature, HelmPackage, OCICompliant},
+        application::{
+            Application, ApplicationFeature, HelmPackage, OCICompliant,
+            features::{ArgoApplication, ArgoHelmScore},
+        },
         helm::chart::HelmChartScore,
     },
     score::Score,
-    topology::{DeploymentTarget, HelmCommand, MultiTargetTopology, Topology, Url},
+    topology::{DeploymentTarget, HelmCommand, K8sclient, MultiTargetTopology, Topology, Url},
 };
 
 /// ContinuousDelivery in Harmony provides this functionality :
@@ -139,7 +142,7 @@ impl<A: OCICompliant + HelmPackage> ContinuousDelivery<A> {
 #[async_trait]
 impl<
     A: OCICompliant + HelmPackage + Clone + 'static,
-    T: Topology + HelmCommand + MultiTargetTopology + 'static,
+    T: Topology + HelmCommand + MultiTargetTopology + K8sclient + 'static,
 > ApplicationFeature<T> for ContinuousDelivery<A>
 {
     async fn ensure_installed(&self, topology: &T) -> Result<(), String> {
@@ -150,12 +153,16 @@ impl<
             "TODO reverse helm chart packaging and docker image build. I put helm package first for faster iterations"
         );
 
+        // TODO Write CI/CD workflow files
+        // we can autotedect the CI type using the remote url (default to github action for github
+        // url, etc..)
+        // Or ask for it when unknown
+
         let helm_chart = self.application.build_push_helm_package(&image).await?;
         info!("Pushed new helm chart {helm_chart}");
 
-        // let image = self.application.build_push_oci_image().await?;
-        // info!("Pushed new docker image {image}");
-        error!("uncomment above");
+        let image = self.application.build_push_oci_image().await?;
+        info!("Pushed new docker image {image}");
 
         info!("Installing ContinuousDelivery feature");
         // TODO this is a temporary hack for demo purposes, the deployment target should be driven
@@ -178,29 +185,33 @@ impl<
             }
             target => {
                 info!("Deploying to target {target:?}");
-                let cd_server = HelmChartScore {
-                    namespace: todo!(
-                        "ArgoCD Helm chart with proper understanding of Tenant, see how Will did it for Monitoring for now"
-                    ),
-                    release_name: todo!("argocd helm chart whatever"),
-                    chart_name: todo!(),
-                    chart_version: todo!(),
-                    values_overrides: todo!(),
-                    values_yaml: todo!(),
-                    create_namespace: todo!(),
-                    install_only: todo!(),
-                    repository: todo!(),
+                let score = ArgoHelmScore {
+                    namespace: "harmonydemo-staging".to_string(),
+                    openshift: true,
+                    domain: "argo.harmonydemo.apps.st.mcd".to_string(),
+                    argo_apps: vec![ArgoApplication::from(CDApplicationConfig {
+                        // helm pull oci://hub.nationtech.io/harmony/harmony-example-rust-webapp-chart/harmony-example-rust-webapp-chart --version 0.1.0
+                        version: Version::from("0.1.0").unwrap(),
+                        helm_chart_repo_url: Url::Url(url::Url::parse("oci://hub.nationtech.io/harmony/harmony-example-rust-webapp-chart/harmony-example-rust-webapp-chart").unwrap()),
+                        helm_chart_name: "harmony-example-rust-webapp-chart".to_string(),
+                        values_overrides: Value::Null,
+                        name: "harmony-demo-rust-webapp".to_string(),
+                        namespace: "harmonydemo-staging".to_string(),
+                    })],
                 };
-                let interpret = cd_server.create_interpret();
-                interpret.execute(&Inventory::empty(), topology);
+                score
+                    .create_interpret()
+                    .execute(&Inventory::empty(), topology)
+                    .await
+                    .unwrap();
             }
         };
 
         todo!("1. Create ArgoCD score that installs argo using helm chart, see if Taha's already done it
             - [X] Package app (docker image, helm chart)
             - [X] Push to registry
-            - [ ] Push only if staging or prod
-            - [ ] Deploy to local k3d when target is local
+            - [X] Push only if staging or prod
+            - [X] Deploy to local k3d when target is local
             - [ ] Poke Argo
             - [ ] Ensure app is up")
     }
@@ -212,9 +223,12 @@ impl<
 /// For now this is entirely bound to K8s / ArgoCD, will have to be revisited when we support
 /// more CD systems
 pub struct CDApplicationConfig {
-    version: Version,
-    helm_chart_url: Url,
-    values_overrides: Value,
+    pub version: Version,
+    pub helm_chart_repo_url: Url,
+    pub helm_chart_name: String,
+    pub values_overrides: Value,
+    pub name: String,
+    pub namespace: String,
 }
 
 pub trait ContinuousDeliveryApplication {

harmony/src/modules/application/features/endpoint.rs

@@ -2,7 +2,7 @@ use async_trait::async_trait;
 use log::info;
 
 use crate::{
-    modules::application::{Application, ApplicationFeature},
+    modules::application::ApplicationFeature,
     topology::{K8sclient, Topology},
 };
 

harmony/src/modules/application/features/mod.rs

@@ -6,3 +6,9 @@ pub use monitoring::*;
 
 mod continuous_delivery;
 pub use continuous_delivery::*;
+
+mod helm_argocd_score;
+pub use helm_argocd_score::*;
+
+mod argo_types;
+pub use argo_types::*;

harmony/src/modules/application/features/monitoring.rs

@@ -1,19 +1,53 @@
+use std::sync::Arc;
+
 use async_trait::async_trait;
 use log::info;
 
 use crate::{
-    modules::application::{Application, ApplicationFeature},
-    topology::{HelmCommand, Topology},
+    inventory::Inventory,
+    modules::{
+        application::{Application, ApplicationFeature},
+        monitoring::{
+            application_monitoring::k8s_application_monitoring_score::ApplicationPrometheusMonitoringScore,
+            kube_prometheus::types::{NamespaceSelector, ServiceMonitor}, prometheus::prometheus::Prometheus,
+        },
+    },
+    score::Score,
+    topology::{oberservability::monitoring::{AlertReceiver, AlertRule, AlertSender}, tenant::TenantManager, HelmCommand, K8sclient, Topology},
 };
 
-#[derive(Debug, Default, Clone)]
-pub struct Monitoring {}
+#[derive(Debug, Clone)]
+pub struct PrometheusMonitoring {
+    pub application: Arc<dyn Application>,
+    pub alert_receivers: Vec<Box<dyn AlertReceiver<Prometheus>>>,
+    pub alert_rules: Vec<Box<dyn AlertRule<Prometheus>>>,
+}
 
 #[async_trait]
-impl<T: Topology + HelmCommand + 'static> ApplicationFeature<T> for Monitoring {
-    async fn ensure_installed(&self, _topology: &T) -> Result<(), String> {
+impl<T: Topology + HelmCommand + 'static + TenantManager> ApplicationFeature<T> for PrometheusMonitoring {
+    async fn ensure_installed(&self, topology: &T) -> Result<(), String> {
         info!("Ensuring monitoring is available for application");
-        todo!("create and execute k8s prometheus score, depends on Will's work")
+        let ns = self.application.name();
+        let mut service_monitor = ServiceMonitor::default();
+        service_monitor.name = ns.clone();
+        service_monitor.namespace = ns.clone();
+        service_monitor.namespace_selector = Some(NamespaceSelector {
+            any: true,
+            match_names: vec![ns.clone()],
+        });
+        let alerting_score = ApplicationPrometheusMonitoringScore {
+            namespace: ns,
+            receivers: self.alert_receivers.clone(),
+            rules: self.alert_rules.clone(),
+            service_monitors: vec![service_monitor],
+        };
+
+        alerting_score
+            .create_interpret()
+            .execute(&Inventory::empty(), topology)
+            .await
+            .unwrap();
+        Ok(())
     }
     fn name(&self) -> String {
         "Monitoring".to_string()

harmony/src/modules/application/mod.rs

@@ -23,13 +23,13 @@ pub trait Application: std::fmt::Debug + Send + Sync {
 }
 
 #[derive(Debug)]
-pub struct ApplicationInterpret<T: Topology + std::fmt::Debug> {
+pub struct ApplicationInterpret<A: Application, T: Topology + std::fmt::Debug> {
     features: Vec<Box<dyn ApplicationFeature<T>>>,
-    application: Arc<Box<dyn Application>>,
+    application: Arc<A>,
 }
 
 #[async_trait]
-impl<T: Topology + std::fmt::Debug> Interpret<T> for ApplicationInterpret<T> {
+impl<A: Application, T: Topology + std::fmt::Debug> Interpret<T> for ApplicationInterpret<A, T> {
     async fn execute(
         &self,
         _inventory: &Inventory,

harmony/src/modules/application/rust.rs

@@ -19,23 +19,30 @@ use crate::{
 use super::{Application, ApplicationFeature, ApplicationInterpret, HelmPackage, OCICompliant};
 
 #[derive(Debug, Serialize, Clone)]
-pub struct RustWebappScore<T: Topology + Clone + Serialize> {
-    pub name: String,
-    pub domain: Url,
+pub struct ApplicationScore<A: Application + Serialize, T: Topology + Clone + Serialize>
+where
+    Arc<A>: Serialize + Clone,
+{
     pub features: Vec<Box<dyn ApplicationFeature<T>>>,
-    pub application: RustWebapp,
+    pub application: Arc<A>,
 }
 
-impl<T: Topology + std::fmt::Debug + Clone + Serialize + 'static> Score<T> for RustWebappScore<T> {
+impl<
+    A: Application + Serialize + Clone + 'static,
+    T: Topology + std::fmt::Debug + Clone + Serialize + 'static,
+> Score<T> for ApplicationScore<A, T>
+where
+    Arc<A>: Serialize,
+{
     fn create_interpret(&self) -> Box<dyn crate::interpret::Interpret<T>> {
         Box::new(ApplicationInterpret {
             features: self.features.clone(),
-            application: Arc::new(Box::new(self.application.clone())),
+            application: self.application.clone(),
         })
     }
 
     fn name(&self) -> String {
-        format!("{}-RustWebapp", self.name)
+        format!("Application: {}", self.application.name())
     }
 }
 
@@ -47,6 +54,7 @@ pub enum RustWebFramework {
 #[derive(Debug, Clone, Serialize)]
 pub struct RustWebapp {
     pub name: String,
+    pub domain: Url,
     /// The path to the root of the Rust project to be containerized.
     pub project_root: PathBuf,
     pub framework: Option<RustWebFramework>,

harmony/src/modules/helm/chart.rs

@@ -220,7 +220,6 @@ impl<T: Topology + HelmCommand> Interpret<T> for HelmChartInterpret {
             yaml_path,
             Some(&helm_options),
         );
-
         let status = match res {
             Ok(status) => status,
             Err(err) => return Err(InterpretError::new(err.to_string())),

harmony/src/modules/monitoring/alert_channel/discord_alert_channel.rs

@@ -1,14 +1,22 @@
+use std::any::Any;
+
 use async_trait::async_trait;
 use serde::Serialize;
 use serde_yaml::{Mapping, Value};
 
 use crate::{
     interpret::{InterpretError, Outcome},
-    modules::monitoring::kube_prometheus::{
-        prometheus::{Prometheus, PrometheusReceiver},
-        types::{AlertChannelConfig, AlertManagerChannelConfig},
+    modules::monitoring::{
+        kube_prometheus::{
+            prometheus::{KubePrometheus, KubePrometheusReceiver},
+            types::{AlertChannelConfig, AlertManagerChannelConfig},
+        },
+        prometheus::prometheus::{Prometheus, PrometheusReceiver},
+    },
+    topology::{
+        Url,
+        oberservability::monitoring::{AlertReceiver, AlertSender},
     },
-    topology::{Url, oberservability::monitoring::AlertReceiver},
 };
 
 #[derive(Debug, Clone, Serialize)]
@@ -37,6 +45,26 @@ impl PrometheusReceiver for DiscordWebhook {
     }
 }
 
+#[async_trait]
+impl AlertReceiver<KubePrometheus> for DiscordWebhook {
+    async fn install(&self, sender: &KubePrometheus) -> Result<Outcome, InterpretError> {
+        sender.install_receiver(self).await
+    }
+    fn clone_box(&self) -> Box<dyn AlertReceiver<KubePrometheus>> {
+        Box::new(self.clone())
+    }
+}
+
+#[async_trait]
+impl KubePrometheusReceiver for DiscordWebhook {
+    fn name(&self) -> String {
+        self.name.clone()
+    }
+    async fn configure_receiver(&self) -> AlertManagerChannelConfig {
+        self.get_config().await
+    }
+}
+
 #[async_trait]
 impl AlertChannelConfig for DiscordWebhook {
     async fn get_config(&self) -> AlertManagerChannelConfig {

harmony/src/modules/monitoring/alert_channel/webhook_receiver.rs

@@ -4,9 +4,12 @@ use serde_yaml::{Mapping, Value};
 
 use crate::{
     interpret::{InterpretError, Outcome},
-    modules::monitoring::kube_prometheus::{
-        prometheus::{Prometheus, PrometheusReceiver},
-        types::{AlertChannelConfig, AlertManagerChannelConfig},
+    modules::monitoring::{
+        kube_prometheus::{
+            prometheus::{KubePrometheus, KubePrometheusReceiver},
+            types::{AlertChannelConfig, AlertManagerChannelConfig},
+        },
+        prometheus::prometheus::{Prometheus, PrometheusReceiver},
     },
     topology::{Url, oberservability::monitoring::AlertReceiver},
 };
@@ -36,6 +39,25 @@ impl PrometheusReceiver for WebhookReceiver {
         self.get_config().await
     }
 }
+#[async_trait]
+impl AlertReceiver<KubePrometheus> for WebhookReceiver {
+    async fn install(&self, sender: &KubePrometheus) -> Result<Outcome, InterpretError> {
+        sender.install_receiver(self).await
+    }
+    fn clone_box(&self) -> Box<dyn AlertReceiver<KubePrometheus>> {
+        Box::new(self.clone())
+    }
+}
+
+#[async_trait]
+impl KubePrometheusReceiver for WebhookReceiver {
+    fn name(&self) -> String {
+        self.name.clone()
+    }
+    async fn configure_receiver(&self) -> AlertManagerChannelConfig {
+        self.get_config().await
+    }
+}
 
 #[async_trait]
 impl AlertChannelConfig for WebhookReceiver {

harmony/src/modules/monitoring/alert_rule/prometheus_alert_rule.rs

@@ -5,13 +5,26 @@ use serde::Serialize;
 
 use crate::{
     interpret::{InterpretError, Outcome},
-    modules::monitoring::kube_prometheus::{
-        prometheus::{Prometheus, PrometheusRule},
-        types::{AlertGroup, AlertManagerAdditionalPromRules},
+    modules::monitoring::{
+        kube_prometheus::{
+            prometheus::{KubePrometheus, KubePrometheusRule},
+            types::{AlertGroup, AlertManagerAdditionalPromRules},
+        },
+        prometheus::prometheus::{Prometheus, PrometheusRule},
     },
     topology::oberservability::monitoring::AlertRule,
 };
 
+#[async_trait]
+impl AlertRule<KubePrometheus> for AlertManagerRuleGroup {
+    async fn install(&self, sender: &KubePrometheus) -> Result<Outcome, InterpretError> {
+        sender.install_rule(&self).await
+    }
+    fn clone_box(&self) -> Box<dyn AlertRule<KubePrometheus>> {
+        Box::new(self.clone())
+    }
+}
+
 #[async_trait]
 impl AlertRule<Prometheus> for AlertManagerRuleGroup {
     async fn install(&self, sender: &Prometheus) -> Result<Outcome, InterpretError> {
@@ -41,6 +54,25 @@ impl PrometheusRule for AlertManagerRuleGroup {
         }
     }
 }
+#[async_trait]
+impl KubePrometheusRule for AlertManagerRuleGroup {
+    fn name(&self) -> String {
+        self.name.clone()
+    }
+    async fn configure_rule(&self) -> AlertManagerAdditionalPromRules {
+        let mut additional_prom_rules = BTreeMap::new();
+
+        additional_prom_rules.insert(
+            self.name.clone(),
+            AlertGroup {
+                groups: vec![self.clone()],
+            },
+        );
+        AlertManagerAdditionalPromRules {
+            rules: additional_prom_rules,
+        }
+    }
+}
 
 impl AlertManagerRuleGroup {
     pub fn new(name: &str, rules: Vec<PrometheusAlertRule>) -> AlertManagerRuleGroup {

harmony/src/modules/monitoring/application_monitoring/k8s_application_monitoring_score.rs

@@ -0,0 +1,45 @@
+use std::sync::{Arc, Mutex};
+
+use log::debug;
+use serde::Serialize;
+
+use crate::{
+    modules::monitoring::{
+        kube_prometheus::types::ServiceMonitor,
+        prometheus::{prometheus::Prometheus, prometheus_config::HelmPrometheusConfig},
+    },
+    score::Score,
+    topology::{
+        oberservability::monitoring::{AlertReceiver, AlertRule, AlertingInterpret}, tenant::TenantManager, HelmCommand, K8sclient, Topology
+    },
+};
+
+#[derive(Clone, Debug, Serialize)]
+pub struct ApplicationPrometheusMonitoringScore {
+    pub namespace: String,
+    pub receivers: Vec<Box<dyn AlertReceiver<Prometheus>>>,
+    pub rules: Vec<Box<dyn AlertRule<Prometheus>>>,
+    pub service_monitors: Vec<ServiceMonitor>,
+}
+
+impl<T: Topology + HelmCommand + TenantManager> Score<T> for ApplicationPrometheusMonitoringScore {
+    fn create_interpret(&self) -> Box<dyn crate::interpret::Interpret<T>> {
+        let config = Arc::new(Mutex::new(HelmPrometheusConfig::new()));
+        config
+            .try_lock()
+            .expect("couldn't lock config")
+            .additional_service_monitors = self.service_monitors.clone();
+        let ns = self.namespace.clone();
+
+        config.try_lock().expect("couldn't lock config").namespace = Some(ns.clone());
+        debug!("set namespace to {}", ns);
+        Box::new(AlertingInterpret {
+            sender: Prometheus { config },
+            receivers: self.receivers.clone(),
+            rules: self.rules.clone(),
+        })
+    }
+    fn name(&self) -> String {
+        "ApplicationPrometheusMonitoringScore".to_string()
+    }
+}

harmony/src/modules/monitoring/application_monitoring/mod.rs

@@ -0,0 +1 @@
+pub mod k8s_application_monitoring_score;

harmony/src/modules/monitoring/grafana/helm/helm_grafana.rs

@@ -0,0 +1,64 @@
+use std::str::FromStr;
+use non_blank_string_rs::NonBlankString;
+use crate::modules::helm::chart::HelmChartScore;
+
+pub fn grafana_helm_chart_score(ns: &str) -> HelmChartScore {
+    let values = format!(
+        r#"
+rbac:
+  namespaced: true
+
+datasources:
+  datasources.yaml:
+    apiVersion: 1
+    datasources:
+    - name: Prometheus
+      type: prometheus
+      access: proxy
+      url: http://prometheus-server.{ns}.svc.cluster.local
+      isDefault: true
+
+dashboardProviders:
+  dashboardproviders.yaml:
+    apiVersion: 1
+    providers:
+    - name: 'default'
+      orgId: 1
+      folder: ''
+      type: file
+      disableDeletion: false
+      updateIntervalSeconds: 10 
+      allowUiUpdates: true
+      editable: true
+      options:
+        path: /var/lib/grafana/dashboards/default
+
+dashboards:
+  default:
+    compute-usage:
+      url: https://grafana.com/api/dashboards/315/revisions/1/download
+    pod-health:
+      url: https://grafana.com/api/dashboards/15758/revisions/1/download
+    namespace-resources:
+      url: https://grafana.com/api/dashboards/9809/revisions/1/download
+    namespace-resources-vs-quotas:
+      url: https://grafana.com/api/dashboards/17044/revisions/1/download
+    persistent-volume-usage:
+      url: https://grafana.com/api/dashboards/7685/revisions/1/download
+"#,
+        ns = ns
+    );
+
+    HelmChartScore {
+        namespace: Some(NonBlankString::from_str(ns).unwrap()),
+        release_name: NonBlankString::from_str("grafana").unwrap(),
+        chart_name: NonBlankString::from_str("oci://ghcr.io/grafana/helm-charts/grafana").unwrap(),
+        chart_version: None,
+        values_overrides: None,
+        values_yaml: Some(values),
+        create_namespace: true,
+        install_only: false,
+        repository: None,
+    }
+}
+

harmony/src/modules/monitoring/grafana/helm/mod.rs

@@ -0,0 +1 @@
+pub mod helm_grafana;

harmony/src/modules/monitoring/grafana/mod.rs

@@ -0,0 +1 @@
+pub mod helm;

harmony/src/modules/monitoring/kube_prometheus/helm/kube_prometheus_helm_chart.rs

@@ -68,6 +68,9 @@ pub fn kube_prometheus_helm_chart_score(
 
     let mut values = format!(
         r#"
+global:
+  rbac:
+    create: false
 prometheus:
   enabled: {prometheus}
   prometheusSpec:
@@ -242,7 +245,7 @@ prometheus-node-exporter:
       cpu: 200m
       memory: 250Mi
 prometheusOperator:
-  enabled: {prometheus_operator}
+  enabled: false
   resources:
     requests: 
       cpu: 100m

harmony/src/modules/monitoring/kube_prometheus/helm_prometheus_alert_score.rs

@@ -2,7 +2,7 @@ use std::sync::{Arc, Mutex};
 
 use serde::Serialize;
 
-use super::{helm::config::KubePrometheusConfig, prometheus::Prometheus};
+use super::{helm::config::KubePrometheusConfig, prometheus::KubePrometheus};
 use crate::{
     modules::monitoring::kube_prometheus::types::ServiceMonitor,
     score::Score,
@@ -15,8 +15,8 @@ use crate::{
 
 #[derive(Clone, Debug, Serialize)]
 pub struct HelmPrometheusAlertingScore {
-    pub receivers: Vec<Box<dyn AlertReceiver<Prometheus>>>,
-    pub rules: Vec<Box<dyn AlertRule<Prometheus>>>,
+    pub receivers: Vec<Box<dyn AlertReceiver<KubePrometheus>>>,
+    pub rules: Vec<Box<dyn AlertRule<KubePrometheus>>>,
     pub service_monitors: Vec<ServiceMonitor>,
 }
 
@@ -28,7 +28,7 @@ impl<T: Topology + HelmCommand + TenantManager> Score<T> for HelmPrometheusAlert
             .expect("couldn't lock config")
             .additional_service_monitors = self.service_monitors.clone();
         Box::new(AlertingInterpret {
-            sender: Prometheus::new(),
+            sender: KubePrometheus::new(),
             receivers: self.receivers.clone(),
             rules: self.rules.clone(),
         })

harmony/src/modules/monitoring/kube_prometheus/prometheus.rs

@@ -10,7 +10,7 @@ use crate::{
     modules::monitoring::alert_rule::prometheus_alert_rule::AlertManagerRuleGroup,
     score,
     topology::{
-        HelmCommand, K8sAnywhereTopology, Topology,
+        HelmCommand, Topology,
         installable::Installable,
         oberservability::monitoring::{AlertReceiver, AlertRule, AlertSender},
         tenant::TenantManager,
@@ -27,14 +27,14 @@ use super::{
 };
 
 #[async_trait]
-impl AlertSender for Prometheus {
+impl AlertSender for KubePrometheus {
     fn name(&self) -> String {
         "HelmKubePrometheus".to_string()
     }
 }
 
 #[async_trait]
-impl<T: Topology + HelmCommand + TenantManager> Installable<T> for Prometheus {
+impl<T: Topology + HelmCommand + TenantManager> Installable<T> for KubePrometheus {
     async fn configure(&self, _inventory: &Inventory, topology: &T) -> Result<(), InterpretError> {
         self.configure_with_topology(topology).await;
         Ok(())
@@ -51,11 +51,11 @@ impl<T: Topology + HelmCommand + TenantManager> Installable<T> for Prometheus {
 }
 
 #[derive(Debug)]
-pub struct Prometheus {
+pub struct KubePrometheus {
     pub config: Arc<Mutex<KubePrometheusConfig>>,
 }
 
-impl Prometheus {
+impl KubePrometheus {
     pub fn new() -> Self {
         Self {
             config: Arc::new(Mutex::new(KubePrometheusConfig::new())),
@@ -75,7 +75,7 @@ impl Prometheus {
 
     pub async fn install_receiver(
         &self,
-        prometheus_receiver: &dyn PrometheusReceiver,
+        prometheus_receiver: &dyn KubePrometheusReceiver,
     ) -> Result<Outcome, InterpretError> {
         let prom_receiver = prometheus_receiver.configure_receiver().await;
         debug!(
@@ -120,12 +120,12 @@ impl Prometheus {
 }
 
 #[async_trait]
-pub trait PrometheusReceiver: Send + Sync + std::fmt::Debug {
+pub trait KubePrometheusReceiver: Send + Sync + std::fmt::Debug {
     fn name(&self) -> String;
     async fn configure_receiver(&self) -> AlertManagerChannelConfig;
 }
 
-impl Serialize for Box<dyn AlertReceiver<Prometheus>> {
+impl Serialize for Box<dyn AlertReceiver<KubePrometheus>> {
     fn serialize<S>(&self, _serializer: S) -> Result<S::Ok, S::Error>
     where
         S: serde::Serializer,
@@ -134,19 +134,19 @@ impl Serialize for Box<dyn AlertReceiver<Prometheus>> {
     }
 }
 
-impl Clone for Box<dyn AlertReceiver<Prometheus>> {
+impl Clone for Box<dyn AlertReceiver<KubePrometheus>> {
     fn clone(&self) -> Self {
         self.clone_box()
     }
 }
 
 #[async_trait]
-pub trait PrometheusRule: Send + Sync + std::fmt::Debug {
+pub trait KubePrometheusRule: Send + Sync + std::fmt::Debug {
     fn name(&self) -> String;
     async fn configure_rule(&self) -> AlertManagerAdditionalPromRules;
 }
 
-impl Serialize for Box<dyn AlertRule<Prometheus>> {
+impl Serialize for Box<dyn AlertRule<KubePrometheus>> {
     fn serialize<S>(&self, _serializer: S) -> Result<S::Ok, S::Error>
     where
         S: serde::Serializer,
@@ -155,7 +155,7 @@ impl Serialize for Box<dyn AlertRule<Prometheus>> {
     }
 }
 
-impl Clone for Box<dyn AlertRule<Prometheus>> {
+impl Clone for Box<dyn AlertRule<KubePrometheus>> {
     fn clone(&self) -> Self {
         self.clone_box()
     }

harmony/src/modules/monitoring/kube_prometheus/types.rs

@@ -211,8 +211,10 @@ pub struct Selector {
 pub struct ServiceMonitor {
     pub name: String,
 
+    pub namespace: String,
+
     // # Additional labels to set used for the ServiceMonitorSelector. Together with standard labels from the chart
-    pub additional_labels: Option<Mapping>,
+    pub additional_labels: Option<HashMap<String, String>>,
 
     // # Service label for use in assembling a job name of the form <label value>-<port>
     // # If no label is specified, the service name is used.
@@ -240,7 +242,7 @@ pub struct ServiceMonitor {
     //   any: bool,
     //   # Explicit list of namespace names to select
     //   matchNames: Vec,
-    pub namespace_selector: Option<Mapping>,
+    pub namespace_selector: Option<NamespaceSelector>,
 
     // # Endpoints of the selected service to be monitored
     pub endpoints: Vec<ServiceMonitorEndpoint>,
@@ -250,10 +252,18 @@ pub struct ServiceMonitor {
     pub fallback_scrape_protocol: Option<String>,
 }
 
+#[derive(Debug, Serialize, Clone)]
+#[serde(rename_all = "camelCase")]
+pub struct NamespaceSelector {
+    pub any: bool,
+    pub match_names: Vec<String>,
+}
+
 impl Default for ServiceMonitor {
     fn default() -> Self {
         Self {
             name: Default::default(),
+            namespace: Default::default(),
             additional_labels: Default::default(),
             job_label: Default::default(),
             target_labels: Default::default(),

harmony/src/modules/monitoring/mod.rs

@@ -1,4 +1,7 @@
 pub mod alert_channel;
 pub mod alert_rule;
+pub mod application_monitoring;
+pub mod grafana;
 pub mod kube_prometheus;
 pub mod ntfy;
+pub mod prometheus;

harmony/src/modules/monitoring/prometheus/helm/mod.rs

@@ -0,0 +1,2 @@
+pub mod prometheus_helm;
+pub mod types;

harmony/src/modules/monitoring/prometheus/helm/prometheus_helm.rs

@@ -0,0 +1,155 @@
+use std::collections::BTreeMap;
+use std::str::FromStr;
+use std::sync::{Arc, Mutex};
+
+use log::debug;
+use non_blank_string_rs::NonBlankString;
+use serde_yaml::{Mapping, Value};
+
+use crate::modules::helm::chart::HelmChartScore;
+use crate::modules::monitoring::kube_prometheus::types::{
+    AlertGroup, AlertManager, AlertManagerConfig, AlertManagerRoute, AlertManagerSpec,
+    ConfigReloader, Limits, Requests, Resources,
+};
+use crate::modules::monitoring::prometheus::helm::types::{
+    AlertFile, EnabledConfig, KsmRbacConfig, KubeStateMetricsConfig, LabelSelector, Monitor,
+    Prometheus, PrometheusHelmValues, RbacConfig, ServerConfig, ServerRbacConfig,
+};
+use crate::modules::monitoring::prometheus::prometheus_config::HelmPrometheusConfig;
+
+pub fn prometheus_helm_chart_score(config: Arc<Mutex<HelmPrometheusConfig>>) -> HelmChartScore {
+    let config = config.lock().unwrap();
+    let ns = config.namespace.clone().unwrap();
+
+    let rbac_config = RbacConfig { create: false };
+
+    let ksm_config = KubeStateMetricsConfig {
+        enabled: true,
+        rbac: KsmRbacConfig {
+            use_cluster_role: false,
+        },
+        prometheus: Prometheus {
+            monitor: Monitor { enabled: true },
+        },
+    };
+
+    let mut selector_labels = BTreeMap::new();
+    selector_labels.insert("kubernetes.io/metadata.name".to_string(), ns.clone());
+    let mut kube_state_metrics_labels = BTreeMap::new();
+    kube_state_metrics_labels.insert(
+        "app.kubernetes.io/name".to_string(),
+        "kube-state-metrics".to_string(),
+    );
+    let selector = LabelSelector {
+        match_labels: selector_labels,
+    };
+
+    let server_config = ServerConfig {
+        namespaces: vec![ns.clone()],
+        use_existing_cluster_role_name: false,
+    };
+
+    let mut null_receiver = Mapping::new();
+    null_receiver.insert(
+        Value::String("receiver".to_string()),
+        Value::String("default-receiver".to_string()),
+    );
+    null_receiver.insert(
+        Value::String("matchers".to_string()),
+        Value::Sequence(vec![Value::String("alertname!=Watchdog".to_string())]),
+    );
+    null_receiver.insert(Value::String("continue".to_string()), Value::Bool(true));
+
+    let mut alert_manager_channel_config = AlertManagerConfig {
+        global: Mapping::new(),
+        route: AlertManagerRoute {
+            routes: vec![Value::Mapping(null_receiver)],
+        },
+        receivers: vec![serde_yaml::from_str("name: 'default-receiver'").unwrap()],
+    };
+    for receiver in config.alert_receiver_configs.iter() {
+        if let Some(global) = receiver.channel_global_config.clone() {
+            alert_manager_channel_config
+                .global
+                .insert(global.0, global.1);
+        }
+        alert_manager_channel_config
+            .route
+            .routes
+            .push(receiver.channel_route.clone());
+        alert_manager_channel_config
+            .receivers
+            .push(receiver.channel_receiver.clone());
+    }
+    let alert_manager_values = AlertManager {
+        enabled: config.alert_manager,
+        config: alert_manager_channel_config,
+        alertmanager_spec: AlertManagerSpec {
+            resources: Resources {
+                limits: Limits {
+                    memory: "100Mi".to_string(),
+                    cpu: "100m".to_string(),
+                },
+                requests: Requests {
+                    memory: "100Mi".to_string(),
+                    cpu: "100m".to_string(),
+                },
+            },
+        },
+        init_config_reloader: ConfigReloader {
+            resources: Resources {
+                limits: Limits {
+                    memory: "100Mi".to_string(),
+                    cpu: "100m".to_string(),
+                },
+                requests: Requests {
+                    memory: "100Mi".to_string(),
+                    cpu: "100m".to_string(),
+                },
+            },
+        },
+    };
+
+    let mut result: BTreeMap<String, AlertFile> = BTreeMap::new();
+    for rule in config.alert_rules.clone().iter() {
+        for (name, group) in &rule.rules {
+            result
+                .entry("alerting_rules.yml".to_string())
+                .and_modify(|e| e.groups.extend(group.groups.clone()))
+                .or_insert(AlertFile {
+                    groups: group.groups.clone(),
+                });
+        }
+    }
+
+    let final_values = PrometheusHelmValues {
+        rbac: rbac_config,
+        kube_state_metrics: ksm_config,
+        server: server_config,
+        alertmanager: alert_manager_values,
+        server_files: result,
+        additional_service_monitors: config.additional_service_monitors.clone(),
+        prometheus_node_exporter: EnabledConfig { enabled: false },
+        prometheus_pushgateway: EnabledConfig { enabled: false },
+    };
+
+    let values_yaml =
+        serde_yaml::to_string(&final_values).expect("Failed to serialize final Helm values");
+
+    debug!("full values.yaml: \n{}", values_yaml);
+
+    HelmChartScore {
+        namespace: Some(NonBlankString::from_str(&ns).unwrap()),
+        release_name: NonBlankString::from_str("prometheus").unwrap(),
+        chart_name: NonBlankString::from_str(
+            "oci://ghcr.io/prometheus-community/charts/prometheus",
+        )
+        .unwrap(),
+        chart_version: None,
+        values_overrides: None,
+        values_yaml: Some(values_yaml),
+        create_namespace: true,
+        install_only: true,
+        repository: None,
+    }
+}

harmony/src/modules/monitoring/prometheus/helm/types.rs

@@ -0,0 +1,94 @@
+use std::collections::BTreeMap;
+
+use serde::Serialize;
+
+use crate::modules::monitoring::{alert_rule::prometheus_alert_rule::AlertManagerRuleGroup, kube_prometheus::types::{
+    AlertGroup, AlertManager, AlertManagerAdditionalPromRules, AlertManagerValues, ServiceMonitor
+}};
+
+#[derive(Debug, Clone, Serialize)]
+pub struct RuleFilesConfig {
+    #[serde(rename = "ruleFiles")]
+    pub files: BTreeMap<String, AlertGroup>,
+}
+
+#[derive(Serialize, Debug)]
+#[serde(rename_all = "camelCase")]
+pub struct PrometheusHelmValues {
+    pub rbac: RbacConfig,
+    #[serde(rename = "kube-state-metrics")]
+    pub kube_state_metrics: KubeStateMetricsConfig,
+    pub server: ServerConfig,
+    pub alertmanager: AlertManager, // You already have this
+    #[serde(rename = "serverFiles")]
+    pub server_files: BTreeMap<String, AlertFile>, // You already have this
+    pub additional_service_monitors: Vec<ServiceMonitor>, // You already have this
+    #[serde(rename = "prometheus-node-exporter")]
+    pub prometheus_node_exporter: EnabledConfig,
+    #[serde(rename = "prometheus-pushgateway")]
+    pub prometheus_pushgateway: EnabledConfig,
+}
+
+#[derive(Serialize, Debug, Clone)]
+pub struct AlertFile {
+    pub groups: Vec<AlertManagerRuleGroup>,
+}
+
+#[derive(Serialize, Debug)]
+#[serde(rename_all = "camelCase")]
+pub struct RbacConfig {
+    pub create: bool,
+}
+
+#[derive(Serialize, Debug)]
+#[serde(rename_all = "camelCase")]
+pub struct KubeStateMetricsConfig {
+    pub enabled: bool,
+    pub rbac: KsmRbacConfig,
+    pub prometheus: Prometheus,
+}
+
+#[derive(Serialize, Debug)]
+#[serde(rename_all = "camelCase")]
+pub struct Prometheus {
+    pub monitor: Monitor
+}
+
+#[derive(Serialize, Debug)]
+#[serde(rename_all = "camelCase")]
+pub struct Monitor{
+    pub enabled: bool
+}
+
+#[derive(Serialize, Debug)]
+#[serde(rename_all = "camelCase")]
+pub struct KsmRbacConfig {
+    pub use_cluster_role: bool,
+}
+
+#[derive(Serialize, Debug)]
+#[serde(rename_all = "camelCase")]
+pub struct ServerConfig {
+    pub namespaces: Vec<String>,
+    pub use_existing_cluster_role_name: bool,
+}
+
+#[derive(Serialize, Debug)]
+#[serde(rename_all = "camelCase")]
+pub struct ServerRbacConfig {
+    pub create: bool,
+    pub use_cluster_role: bool,
+    pub namespaced: bool,
+}
+
+#[derive(Serialize, Debug, Clone)]
+#[serde(rename_all = "camelCase")]
+pub struct LabelSelector {
+    #[serde(rename = "matchLabels")]
+    pub match_labels: BTreeMap<String, String>,
+}
+
+#[derive(Serialize, Debug)]
+pub struct EnabledConfig {
+    pub enabled: bool,
+}

harmony/src/modules/monitoring/prometheus/mod.rs

@@ -0,0 +1,3 @@
+pub mod helm;
+pub mod prometheus;
+pub mod prometheus_config;

harmony/src/modules/monitoring/prometheus/prometheus.rs

@@ -0,0 +1,189 @@
+use std::sync::{Arc, Mutex};
+
+use async_trait::async_trait;
+use log::{debug, error};
+use serde::Serialize;
+
+use crate::{
+    interpret::{InterpretError, Outcome},
+    inventory::Inventory,
+    modules::monitoring::{
+        alert_rule::prometheus_alert_rule::AlertManagerRuleGroup,
+        grafana::helm::helm_grafana::grafana_helm_chart_score,
+        kube_prometheus::types::{AlertManagerAdditionalPromRules, AlertManagerChannelConfig},
+    },
+    score::Score,
+    topology::{
+        HelmCommand, K8sclient, Topology,
+        installable::Installable,
+        oberservability::monitoring::{AlertReceiver, AlertRule, AlertSender},
+        tenant::TenantManager,
+    },
+};
+
+use super::{
+    helm::prometheus_helm::prometheus_helm_chart_score, prometheus_config::HelmPrometheusConfig,
+};
+
+#[derive(Debug)]
+pub struct Prometheus {
+    pub config: Arc<Mutex<HelmPrometheusConfig>>,
+}
+
+#[async_trait]
+impl AlertSender for Prometheus {
+    fn name(&self) -> String {
+        "Prometheus".to_string()
+    }
+}
+
+impl Prometheus {
+    pub fn new() -> Self {
+        Self {
+            config: Arc::new(Mutex::new(HelmPrometheusConfig::new())),
+        }
+    }
+    pub async fn configure_with_topology<T: TenantManager>(&self, topology: &T) {
+        if let Some(cfg) = topology.get_tenant_config().await {
+            debug!("Overriding namespace with tenant config: {}", cfg.name);
+            self.config.lock().unwrap().namespace = Some(cfg.name.clone());
+        } else {
+            debug!("No tenant config found; keeping existing namespace.");
+        }
+        error!("This must be refactored, see comments in pr #74");
+    }
+
+    pub async fn install_receiver(
+        &self,
+        prometheus_receiver: &dyn PrometheusReceiver,
+    ) -> Result<Outcome, InterpretError> {
+        let prom_receiver = prometheus_receiver.configure_receiver().await;
+        debug!(
+            "adding alert receiver to prometheus config: {:#?}",
+            &prom_receiver
+        );
+        let mut config = self.config.lock().unwrap();
+
+        config.alert_receiver_configs.push(prom_receiver);
+        let prom_receiver_name = prometheus_receiver.name();
+        debug!("installed alert receiver {}", &prom_receiver_name);
+        Ok(Outcome::success(format!(
+            "Sucessfully installed receiver {}",
+            prom_receiver_name
+        )))
+    }
+
+    pub async fn install_rule(
+        &self,
+        prometheus_rule: &AlertManagerRuleGroup,
+    ) -> Result<Outcome, InterpretError> {
+        let prometheus_rule = prometheus_rule.configure_rule().await;
+        let mut config = self.config.lock().unwrap();
+
+        config.alert_rules.push(prometheus_rule.clone());
+        Ok(Outcome::success(format!(
+            "Successfully installed alert rule: {:#?},",
+            prometheus_rule
+        )))
+    }
+
+    pub async fn install_prometheus<T: Topology + HelmCommand + Send + Sync>(
+        &self,
+        inventory: &Inventory,
+        topology: &T,
+    ) -> Result<Outcome, InterpretError> {
+        prometheus_helm_chart_score(self.config.clone())
+            .create_interpret()
+            .execute(inventory, topology)
+            .await
+    }
+    pub async fn install_grafana<T: Topology + HelmCommand + Send + Sync>(
+        &self,
+        inventory: &Inventory,
+        topology: &T,
+    ) -> Result<Outcome, InterpretError> {
+        let namespace = {
+            let config = self.config.lock().unwrap();
+            config.namespace.clone()
+        };
+
+        if let Some(ns) = namespace.as_deref() {
+            grafana_helm_chart_score(ns)
+                .create_interpret()
+                .execute(inventory, topology)
+                .await
+        } else {
+            Err(InterpretError::new(format!(
+                "could not install grafana, missing namespace",
+            )))
+        }
+    }
+}
+#[async_trait]
+impl<T: Topology + HelmCommand + TenantManager> Installable<T> for Prometheus {
+    async fn configure(&self, _inventory: &Inventory, topology: &T) -> Result<(), InterpretError> {
+        self.configure_with_topology(topology).await;
+        Ok(())
+    }
+
+    async fn ensure_installed(
+        &self,
+        inventory: &Inventory,
+        topology: &T,
+    ) -> Result<(), InterpretError> {
+        self.install_prometheus(inventory, topology).await?;
+
+        let install_grafana = {
+            let config = self.config.lock().unwrap();
+            config.grafana
+        };
+
+        if install_grafana {
+            self.install_grafana(inventory, topology).await?;
+        }
+
+        Ok(())
+    }
+}
+
+#[async_trait]
+pub trait PrometheusReceiver: Send + Sync + std::fmt::Debug {
+    fn name(&self) -> String;
+    async fn configure_receiver(&self) -> AlertManagerChannelConfig;
+}
+
+impl Serialize for Box<dyn AlertReceiver<Prometheus>> {
+    fn serialize<S>(&self, _serializer: S) -> Result<S::Ok, S::Error>
+    where
+        S: serde::Serializer,
+    {
+        todo!()
+    }
+}
+
+impl Clone for Box<dyn AlertReceiver<Prometheus>> {
+    fn clone(&self) -> Self {
+        self.clone_box()
+    }
+}
+
+#[async_trait]
+pub trait PrometheusRule: Send + Sync + std::fmt::Debug {
+    fn name(&self) -> String;
+    async fn configure_rule(&self) -> AlertManagerAdditionalPromRules;
+}
+
+impl Serialize for Box<dyn AlertRule<Prometheus>> {
+    fn serialize<S>(&self, _serializer: S) -> Result<S::Ok, S::Error>
+    where
+        S: serde::Serializer,
+    {
+        todo!()
+    }
+}
+
+impl Clone for Box<dyn AlertRule<Prometheus>> {
+    fn clone(&self) -> Self {
+        self.clone_box()
+    }
+}

harmony/src/modules/monitoring/prometheus/prometheus_config.rs

@@ -0,0 +1,32 @@
+use crate::modules::monitoring::kube_prometheus::types::{
+    AlertManagerAdditionalPromRules, AlertManagerChannelConfig, ServiceMonitor,
+};
+
+#[derive(Debug)]
+pub struct HelmPrometheusConfig {
+    pub namespace: Option<String>,
+    pub alert_manager: bool,
+    pub node_exporter: bool,
+    pub kube_state_metrics: bool,
+    pub grafana: bool,
+    pub prometheus_pushgateway: bool,
+    pub alert_receiver_configs: Vec<AlertManagerChannelConfig>,
+    pub alert_rules: Vec<AlertManagerAdditionalPromRules>,
+    pub additional_service_monitors: Vec<ServiceMonitor>,
+}
+
+impl HelmPrometheusConfig {
+    pub fn new() -> Self {
+        Self {
+            namespace: None,
+            alert_manager: true,
+            node_exporter: false,
+            kube_state_metrics: false,
+            grafana: true,
+            prometheus_pushgateway: false,
+            alert_receiver_configs: vec![],
+            alert_rules: vec![],
+            additional_service_monitors: vec![],
+        }
+    }
+}

harmony/src/modules/prometheus/alerts/k8s/mod.rs

@@ -1 +1,2 @@
 pub mod pvc;
+pub mod pod;

harmony/src/modules/prometheus/alerts/k8s/pod.rs

@@ -0,0 +1,38 @@
+use crate::modules::monitoring::alert_rule::prometheus_alert_rule::PrometheusAlertRule;
+
+pub fn pod_in_failed_state() -> PrometheusAlertRule {
+    PrometheusAlertRule::new(
+        "PodInFailedState",
+        // This expression checks for any pod where the status phase is 'Failed' and the value is 1 (true).
+        "kube_pod_status_phase{phase=\"Failed\"} == 1",
+    )
+    .for_duration("1m") // Fire if the pod is in this state for 1 minute.
+    .label("severity", "critical") // A failed pod is a critical issue.
+    .annotation(
+        "summary",
+        "Pod {{ $labels.pod }} in namespace {{ $labels.namespace }} has failed.",
+    )
+    .annotation(
+        "description",
+        "The pod {{ $labels.pod }} in namespace {{ $labels.namespace }} has entered the 'Failed' state. This is a terminal error and the pod will not be automatically restarted. Please check the pod logs to diagnose the issue.",
+    )
+}
+
+pub fn pod_restarting_frequently() -> PrometheusAlertRule {
+    PrometheusAlertRule::new(
+        "PodRestartingFrequently",
+        // This expression calculates the increase in the restart count over the last 30 minutes.
+        // Alert if a container has restarted more than 5 times.
+        "increase(kube_pod_container_status_restarts_total[30m]) > 5",
+    )
+    .for_duration("15m") // The condition must persist for 15 minutes to avoid alerts for minor flaps.
+    .label("severity", "critical") // A crash-looping pod is effectively down.
+    .annotation(
+        "summary",
+        "Container {{ $labels.container }} in pod {{ $labels.pod }} is restarting frequently.",
+    )
+    .annotation(
+        "description",
+        "The container '{{ $labels.container }}' in pod '{{ $labels.pod }}' (namespace '{{ $labels.namespace }}') has restarted more than 5 times in the last 30 minutes. The pod is likely in a CrashLoopBackOff state.",
+    )
+}