worked with will to get monitoring + ntfy demoed/tested

examples/ntfy/src/main.rs

@@ -14,6 +14,7 @@ async fn main() {
 
     maestro.register_all(vec![Box::new(NtfyScore {
         namespace: "monitoring".to_string(),
+        host: "localhost".to_string(),
     })]);
     harmony_cli::init(maestro, None).await.unwrap();
 }

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

@@ -159,8 +159,8 @@ impl<
         info!("Pushed new helm chart {helm_chart}");
 
         error!("TODO Make building image configurable/skippable");
-        let image = self.application.build_push_oci_image().await?;
-        info!("Pushed new docker image {image}");
+        // 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

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

@@ -1,6 +1,6 @@
 use async_trait::async_trait;
 use base64::{Engine as _, engine::general_purpose};
-use log::info;
+use log::{debug, info};
 
 use crate::{
     inventory::Inventory,
@@ -9,7 +9,10 @@ use crate::{
         monitoring::{
             alert_channel::webhook_receiver::WebhookReceiver,
             application_monitoring::k8s_application_monitoring_score::ApplicationPrometheusMonitoringScore,
-            kube_prometheus::types::{NamespaceSelector, ServiceMonitor},
+            kube_prometheus::{
+                helm_prometheus_alert_score::HelmPrometheusAlertingScore,
+                types::{NamespaceSelector, ServiceMonitor},
+            },
             ntfy::ntfy::NtfyScore,
         },
     },
@@ -28,11 +31,13 @@ impl<T: Topology + HelmCommand + K8sclient + 'static + TenantManager> Applicatio
         info!("Ensuring monitoring is available for application");
 
         let ntfy = NtfyScore {
-            namespace: topology
-                .get_tenant_config()
-                .await
-                .expect("couldn't get tenant config")
-                .name,
+            // namespace: topology
+            //     .get_tenant_config()
+            //     .await
+            //     .expect("couldn't get tenant config")
+            //     .name,
+            namespace: "harmonydemo-staging".to_string(),
+            host: "localhost".to_string(),
         };
         ntfy.create_interpret()
             .execute(&Inventory::empty(), topology)
@@ -48,11 +53,13 @@ impl<T: Topology + HelmCommand + K8sclient + 'static + TenantManager> Applicatio
             ))
         );
 
+        debug!("ntfy_default_auth_header: {ntfy_default_auth_header}");
+
         let ntfy_default_auth_param = general_purpose::STANDARD
             .encode(ntfy_default_auth_header)
-            .rsplit("=")
-            .collect::<Vec<&str>>()[0]
-            .to_string();
+            .replace("=", "");
+
+        debug!("ntfy_default_auth_param: {ntfy_default_auth_param}");
 
         let ntfy_receiver = WebhookReceiver {
             name: "ntfy-webhook".to_string(),
@@ -60,7 +67,7 @@ impl<T: Topology + HelmCommand + K8sclient + 'static + TenantManager> Applicatio
                 url::Url::parse(
                     format!(
                         "http://ntfy.{}.svc.cluster.local/rust-web-app?auth={ntfy_default_auth_param}",
-                        topology.get_tenant_config().await.expect("couldn't get tenant config").name
+                        "harmonydemo-staging".to_string()
                     )
                     .as_str(),
                 )
@@ -73,7 +80,16 @@ impl<T: Topology + HelmCommand + K8sclient + 'static + TenantManager> Applicatio
             any: true,
             match_names: vec![],
         });
-        let alerting_score = ApplicationPrometheusMonitoringScore {
+
+        service_monitor.name = "rust-webapp".to_string();
+
+        // let alerting_score = ApplicationPrometheusMonitoringScore {
+        //     receivers: vec![Box::new(ntfy_receiver)],
+        //     rules: vec![],
+        //     service_monitors: vec![service_monitor],
+        // };
+
+        let alerting_score = HelmPrometheusAlertingScore {
             receivers: vec![Box::new(ntfy_receiver)],
             rules: vec![],
             service_monitors: vec![service_monitor],

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

@@ -24,7 +24,10 @@ pub struct ApplicationPrometheusMonitoringScore {
 
 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(PrometheusConfig::new()));
+        let mut prom_config = PrometheusConfig::new();
+        prom_config.alert_manager = true;
+
+        let config = Arc::new(Mutex::new(prom_config));
         config
             .try_lock()
             .expect("couldn't lock config")

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

@@ -38,15 +38,15 @@ impl KubePrometheusConfig {
             node_exporter: false,
             prometheus: true,
             kubernetes_service_monitors: true,
-            kubernetes_api_server: false,
+            kubernetes_api_server: true,
             kubelet: true,
-            kube_controller_manager: false,
-            kube_etcd: false,
-            kube_proxy: false,
+            kube_controller_manager: true,
+            kube_etcd: true,
+            kube_proxy: true,
             kube_state_metrics: true,
             prometheus_operator: true,
-            core_dns: false,
-            kube_scheduler: false,
+            core_dns: true,
+            kube_scheduler: true,
             alert_receiver_configs: vec![],
             alert_rules: vec![],
             additional_service_monitors: vec![],

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

@@ -70,12 +70,12 @@ pub fn kube_prometheus_helm_chart_score(
         r#"
 global:
   rbac:
-    create: false
+    create: true
 prometheus:
   enabled: {prometheus}
   prometheusSpec:
     resources:
-      requests: 
+      requests:
         cpu: 100m
         memory: 500Mi
       limits:
@@ -121,7 +121,7 @@ defaultRules:
 windowsMonitoring:
   enabled: {windows_monitoring}
   resources:
-    requests: 
+    requests:
       cpu: 100m
       memory: 150Mi
     limits:
@@ -130,13 +130,13 @@ windowsMonitoring:
 grafana:
   enabled: {grafana}
   resources:
-    requests: 
+    requests:
       cpu: 100m
       memory: 150Mi
     limits:
       cpu: 200m
       memory: 250Mi
-  initChownData: 
+  initChownData:
     resources:
       requests:
         cpu: 10m
@@ -157,7 +157,7 @@ kubernetesServiceMonitors:
 kubeApiServer:
   enabled: {kubernetes_api_server}
   resources:
-    requests: 
+    requests:
       cpu: 100m
       memory: 150Mi
     limits:
@@ -166,7 +166,7 @@ kubeApiServer:
 kubelet:
   enabled: {kubelet}
   resources:
-    requests: 
+    requests:
       cpu: 100m
       memory: 150Mi
     limits:
@@ -175,7 +175,7 @@ kubelet:
 kubeControllerManager:
   enabled: {kube_controller_manager}
   resources:
-    requests: 
+    requests:
       cpu: 100m
       memory: 150Mi
     limits:
@@ -184,7 +184,7 @@ kubeControllerManager:
 coreDns:
   enabled: {core_dns}
   resources:
-    requests: 
+    requests:
       cpu: 100m
       memory: 150Mi
     limits:
@@ -193,7 +193,7 @@ coreDns:
 kubeEtcd:
   enabled: {kube_etcd}
   resources:
-    requests: 
+    requests:
       cpu: 100m
       memory: 150Mi
     limits:
@@ -202,7 +202,7 @@ kubeEtcd:
 kubeScheduler:
   enabled: {kube_scheduler}
   resources:
-    requests: 
+    requests:
       cpu: 100m
       memory: 150Mi
     limits:
@@ -211,7 +211,7 @@ kubeScheduler:
 kubeProxy:
   enabled: {kube_proxy}
   resources:
-    requests: 
+    requests:
       cpu: 100m
       memory: 150Mi
     limits:
@@ -221,7 +221,7 @@ kubeStateMetrics:
   enabled: {kube_state_metrics}
 kube-state-metrics:
   resources:
-    requests: 
+    requests:
       cpu: 100m
       memory: 150Mi
     limits:
@@ -230,7 +230,7 @@ kube-state-metrics:
 nodeExporter:
   enabled: {node_exporter}
   resources:
-    requests: 
+    requests:
       cpu: 100m
       memory: 150Mi
     limits:
@@ -238,16 +238,16 @@ nodeExporter:
       memory: 250Mi
 prometheus-node-exporter:
   resources:
-    requests: 
+    requests:
       cpu: 100m
       memory: 150Mi
     limits:
       cpu: 200m
       memory: 250Mi
 prometheusOperator:
-  enabled: false
+  enabled: true
   resources:
-    requests: 
+    requests:
       cpu: 100m
       memory: 150Mi
     limits:
@@ -255,7 +255,7 @@ prometheusOperator:
       memory: 200Mi
   prometheusConfigReloader:
     resources:
-      requests: 
+      requests:
         cpu: 100m
         memory: 150Mi
       limits:
@@ -267,7 +267,7 @@ prometheusOperator:
         limits:
           cpu: 10m
           memory: 100Mi
-        requests: 
+        requests:
           cpu: 10m
           memory: 100Mi
     patch:
@@ -275,7 +275,7 @@ prometheusOperator:
         limits:
           cpu: 10m
           memory: 100Mi
-        requests: 
+        requests:
           cpu: 10m
           memory: 100Mi
 "#,

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

@@ -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: KubePrometheus::new(),
+            sender: KubePrometheus { config },
             receivers: self.receivers.clone(),
             rules: self.rules.clone(),
         })

harmony/src/modules/monitoring/ntfy/helm/ntfy_helm_chart.rs

@@ -3,7 +3,7 @@ use std::str::FromStr;
 
 use crate::modules::helm::chart::{HelmChartScore, HelmRepository};
 
-pub fn ntfy_helm_chart_score(namespace: String) -> HelmChartScore {
+pub fn ntfy_helm_chart_score(namespace: String, host: String) -> HelmChartScore {
     let values = format!(
         r#"
 replicaCount: 1
@@ -28,12 +28,12 @@ service:
   port: 80
 
 ingress:
-  enabled: false
+  enabled: true
 #  annotations:
     # kubernetes.io/ingress.class: nginx
     # kubernetes.io/tls-acme: "true"
   hosts:
-    - host: ntfy.host.com
+    - host: {host}
       paths:
         - path: /
           pathType: ImplementationSpecific

harmony/src/modules/monitoring/ntfy/ntfy.rs

@@ -17,6 +17,7 @@ use crate::{
 #[derive(Debug, Clone, Serialize)]
 pub struct NtfyScore {
     pub namespace: String,
+    pub host: String,
 }
 
 impl<T: Topology + HelmCommand + K8sclient> Score<T> for NtfyScore {
@@ -126,7 +127,7 @@ impl<T: Topology + HelmCommand + K8sclient> Interpret<T> for NtfyInterpret {
         inventory: &Inventory,
         topology: &T,
     ) -> Result<Outcome, InterpretError> {
-        ntfy_helm_chart_score(self.score.namespace.clone())
+        ntfy_helm_chart_score(self.score.namespace.clone(), self.score.host.clone())
             .create_interpret()
             .execute(inventory, topology)
             .await?;