From 58b62689899f4fd450bbae9f41499a0c6ffaa9b6 Mon Sep 17 00:00:00 2001 From: Willem Date: Mon, 29 Sep 2025 10:46:29 -0400 Subject: [PATCH 01/11] wip: moving the install steps for grafana and prometheus into the trait installable --- harmony/src/domain/topology/k8s_anywhere.rs | 116 ++++++++++++++++-- .../application/features/monitoring.rs | 5 +- .../application/features/rhob_monitoring.rs | 4 +- .../application_monitoring_score.rs | 8 +- .../rhobs_application_monitoring_score.rs | 6 +- .../src/modules/monitoring/grafana/grafana.rs | 15 +++ .../monitoring/grafana/helm/helm_grafana.rs | 25 ++-- harmony/src/modules/monitoring/grafana/mod.rs | 1 + .../crd/crd_alertmanager_config.rs | 41 ++++++- .../monitoring/prometheus/prometheus.rs | 2 +- .../k8s_prometheus_alerting_score.rs | 6 +- harmony/src/modules/prometheus/prometheus.rs | 8 +- .../modules/prometheus/rhob_alerting_score.rs | 10 +- 13 files changed, 195 insertions(+), 52 deletions(-) create mode 100644 harmony/src/modules/monitoring/grafana/grafana.rs diff --git a/harmony/src/domain/topology/k8s_anywhere.rs b/harmony/src/domain/topology/k8s_anywhere.rs index e6c37ea..6dfb1a8 100644 --- a/harmony/src/domain/topology/k8s_anywhere.rs +++ b/harmony/src/domain/topology/k8s_anywhere.rs @@ -12,14 +12,17 @@ use crate::{ inventory::Inventory, modules::{ k3d::K3DInstallationScore, - monitoring::kube_prometheus::crd::{ - crd_alertmanager_config::CRDPrometheus, - prometheus_operator::prometheus_operator_helm_chart_score, - rhob_alertmanager_config::RHOBObservability, + monitoring::{ + grafana::{grafana::Grafana, helm::helm_grafana::grafana_helm_chart_score}, + kube_prometheus::crd::{ + crd_alertmanager_config::CRDPrometheus, + prometheus_operator::prometheus_operator_helm_chart_score, + rhob_alertmanager_config::RHOBObservability, service_monitor::ServiceMonitor, + }, }, prometheus::{ k8s_prometheus_alerting_score::K8sPrometheusCRDAlertingScore, - prometheus::PrometheusApplicationMonitoring, rhob_alerting_score::RHOBAlertingScore, + prometheus::PrometheusMonitoring, rhob_alerting_score::RHOBAlertingScore, }, }, score::Score, @@ -86,7 +89,43 @@ impl K8sclient for K8sAnywhereTopology { } #[async_trait] -impl PrometheusApplicationMonitoring for K8sAnywhereTopology { +impl Grafana for K8sAnywhereTopology { + async fn ensure_grafana_operator_ready( + &self, + inventory: &Inventory, + ) -> Result { + let client = self.k8s_client().await.unwrap(); + let grafana_gvk = GroupVersionKind { + group: "grafana.integreatly.org".to_string(), + version: "v1beta1".to_string(), + kind: "Grafana".to_string(), + }; + let name = "grafanas.grafana.integreatly.org"; + let ns = "grafana"; + + let grafana_crd = client + .get_resource_json_value(name, Some(ns), &grafana_gvk) + .await; + match grafana_crd { + Ok(_) => { + return Ok(PreparationOutcome::Success { + details: "Found grafana CRDs in cluster".to_string(), + }); + } + Err(_) => { + return self + .install_grafana_operator(inventory, Some("grafana")) + .await; + } + }; + } + async fn install_grafana(&self) -> Result { + todo!() + } +} + +#[async_trait] +impl PrometheusMonitoring for K8sAnywhereTopology { async fn install_prometheus( &self, sender: &CRDPrometheus, @@ -101,7 +140,11 @@ impl PrometheusApplicationMonitoring for K8sAnywhereTopology { } let result = self - .get_k8s_prometheus_application_score(sender.clone(), receivers) + .get_k8s_prometheus_application_score( + sender.clone(), + receivers, + Some(sender.service_monitor.clone()), + ) .await .interpret(inventory, self) .await; @@ -117,10 +160,24 @@ impl PrometheusApplicationMonitoring for K8sAnywhereTopology { Err(err) => Err(PreparationError::new(err.to_string())), } } + async fn ensure_prometheus_operator( + &self, + sender: &CRDPrometheus, + inventory: &Inventory, + ) -> Result { + let po_result = self.ensure_prometheus_operator(sender).await?; + + if po_result == PreparationOutcome::Noop { + debug!("Skipping Prometheus CR installation due to missing operator."); + return Ok(po_result); + } else { + todo!() + } + } } #[async_trait] -impl PrometheusApplicationMonitoring for K8sAnywhereTopology { +impl PrometheusMonitoring for K8sAnywhereTopology { async fn install_prometheus( &self, sender: &RHOBObservability, @@ -154,6 +211,13 @@ impl PrometheusApplicationMonitoring for K8sAnywhereTopology Err(err) => Err(PreparationError::new(err.to_string())), } } + async fn ensure_prometheus_operator( + &self, + sender: &RHOBObservability, + inventory: &Inventory, + ) -> Result { + todo!() + } } impl Serialize for K8sAnywhereTopology { @@ -253,12 +317,22 @@ impl K8sAnywhereTopology { &self, sender: CRDPrometheus, receivers: Option>>>, + service_monitors: Option>, ) -> K8sPrometheusCRDAlertingScore { - K8sPrometheusCRDAlertingScore { - sender, - receivers: receivers.unwrap_or_default(), - service_monitors: vec![], - prometheus_rules: vec![], + if let Some(sm) = service_monitors { + return K8sPrometheusCRDAlertingScore { + sender, + receivers: receivers.unwrap_or_default(), + service_monitors: sm, + prometheus_rules: vec![], + }; + } else { + return K8sPrometheusCRDAlertingScore { + sender, + receivers: receivers.unwrap_or_default(), + service_monitors: vec![], + prometheus_rules: vec![], + }; } } @@ -527,6 +601,22 @@ impl K8sAnywhereTopology { details: "prometheus operator present in cluster".into(), }) } + + async fn install_grafana_operator( + &self, + inventory: &Inventory, + ns: Option<&str>, + ) -> Result { + let _grafana_operator_score = grafana_helm_chart_score(ns.unwrap(), true) + .interpret(inventory, self) + .await; + Ok(PreparationOutcome::Success { + details: format!( + "Successfully installed grafana operator in ns {}", + ns.unwrap() + ), + }) + } } #[derive(Clone, Debug)] diff --git a/harmony/src/modules/application/features/monitoring.rs b/harmony/src/modules/application/features/monitoring.rs index 1a60d00..0fd155d 100644 --- a/harmony/src/modules/application/features/monitoring.rs +++ b/harmony/src/modules/application/features/monitoring.rs @@ -14,7 +14,7 @@ use crate::{ topology::{HelmCommand, K8sclient, Topology, tenant::TenantManager}, }; use crate::{ - modules::prometheus::prometheus::PrometheusApplicationMonitoring, + modules::prometheus::prometheus::PrometheusMonitoring, topology::oberservability::monitoring::AlertReceiver, }; use async_trait::async_trait; @@ -40,7 +40,7 @@ impl< + TenantManager + K8sclient + MultiTargetTopology - + PrometheusApplicationMonitoring + + PrometheusMonitoring + Ingress + std::fmt::Debug, > ApplicationFeature for Monitoring @@ -61,6 +61,7 @@ impl< sender: CRDPrometheus { namespace: namespace.clone(), client: topology.k8s_client().await.unwrap(), + service_monitor: vec![], }, application: self.application.clone(), receivers: self.alert_receiver.clone(), diff --git a/harmony/src/modules/application/features/rhob_monitoring.rs b/harmony/src/modules/application/features/rhob_monitoring.rs index d87ef61..876dba9 100644 --- a/harmony/src/modules/application/features/rhob_monitoring.rs +++ b/harmony/src/modules/application/features/rhob_monitoring.rs @@ -18,7 +18,7 @@ use crate::{ topology::{HelmCommand, K8sclient, Topology, tenant::TenantManager}, }; use crate::{ - modules::prometheus::prometheus::PrometheusApplicationMonitoring, + modules::prometheus::prometheus::PrometheusMonitoring, topology::oberservability::monitoring::AlertReceiver, }; use async_trait::async_trait; @@ -42,7 +42,7 @@ impl< + MultiTargetTopology + Ingress + std::fmt::Debug - + PrometheusApplicationMonitoring, + + PrometheusMonitoring, > ApplicationFeature for Monitoring { async fn ensure_installed( diff --git a/harmony/src/modules/monitoring/application_monitoring/application_monitoring_score.rs b/harmony/src/modules/monitoring/application_monitoring/application_monitoring_score.rs index 8246d15..2780edd 100644 --- a/harmony/src/modules/monitoring/application_monitoring/application_monitoring_score.rs +++ b/harmony/src/modules/monitoring/application_monitoring/application_monitoring_score.rs @@ -10,7 +10,7 @@ use crate::{ modules::{ application::Application, monitoring::kube_prometheus::crd::crd_alertmanager_config::CRDPrometheus, - prometheus::prometheus::PrometheusApplicationMonitoring, + prometheus::prometheus::PrometheusMonitoring, }, score::Score, topology::{PreparationOutcome, Topology, oberservability::monitoring::AlertReceiver}, @@ -24,9 +24,7 @@ pub struct ApplicationMonitoringScore { pub receivers: Vec>>, } -impl> Score - for ApplicationMonitoringScore -{ +impl> Score for ApplicationMonitoringScore { fn create_interpret(&self) -> Box> { Box::new(ApplicationMonitoringInterpret { score: self.clone(), @@ -47,7 +45,7 @@ pub struct ApplicationMonitoringInterpret { } #[async_trait] -impl> Interpret +impl> Interpret for ApplicationMonitoringInterpret { async fn execute( diff --git a/harmony/src/modules/monitoring/application_monitoring/rhobs_application_monitoring_score.rs b/harmony/src/modules/monitoring/application_monitoring/rhobs_application_monitoring_score.rs index 5f5127f..6f45c88 100644 --- a/harmony/src/modules/monitoring/application_monitoring/rhobs_application_monitoring_score.rs +++ b/harmony/src/modules/monitoring/application_monitoring/rhobs_application_monitoring_score.rs @@ -12,7 +12,7 @@ use crate::{ monitoring::kube_prometheus::crd::{ crd_alertmanager_config::CRDPrometheus, rhob_alertmanager_config::RHOBObservability, }, - prometheus::prometheus::PrometheusApplicationMonitoring, + prometheus::prometheus::PrometheusMonitoring, }, score::Score, topology::{PreparationOutcome, Topology, oberservability::monitoring::AlertReceiver}, @@ -26,7 +26,7 @@ pub struct ApplicationRHOBMonitoringScore { pub receivers: Vec>>, } -impl> Score +impl> Score for ApplicationRHOBMonitoringScore { fn create_interpret(&self) -> Box> { @@ -49,7 +49,7 @@ pub struct ApplicationRHOBMonitoringInterpret { } #[async_trait] -impl> Interpret +impl> Interpret for ApplicationRHOBMonitoringInterpret { async fn execute( diff --git a/harmony/src/modules/monitoring/grafana/grafana.rs b/harmony/src/modules/monitoring/grafana/grafana.rs new file mode 100644 index 0000000..411d7a6 --- /dev/null +++ b/harmony/src/modules/monitoring/grafana/grafana.rs @@ -0,0 +1,15 @@ +use async_trait::async_trait; + +use crate::{ + inventory::Inventory, + topology::{PreparationError, PreparationOutcome}, +}; + +#[async_trait] +pub trait Grafana { + async fn ensure_grafana_operator_ready( + &self, + inventory: &Inventory, + ) -> Result; + async fn install_grafana(&self) -> Result; +} diff --git a/harmony/src/modules/monitoring/grafana/helm/helm_grafana.rs b/harmony/src/modules/monitoring/grafana/helm/helm_grafana.rs index 3af6550..094beca 100644 --- a/harmony/src/modules/monitoring/grafana/helm/helm_grafana.rs +++ b/harmony/src/modules/monitoring/grafana/helm/helm_grafana.rs @@ -1,25 +1,22 @@ use non_blank_string_rs::NonBlankString; -use std::str::FromStr; +use std::{collections::HashMap, str::FromStr}; use crate::modules::helm::chart::HelmChartScore; -pub fn grafana_helm_chart_score(ns: &str) -> HelmChartScore { - let values = r#" -rbac: - namespaced: true -sidecar: - dashboards: - enabled: true - "# - .to_string(); - +pub fn grafana_helm_chart_score(ns: &str, scope: bool) -> HelmChartScore { + let mut values_overrides = HashMap::new(); + values_overrides.insert( + NonBlankString::from_str("namespaceScope").unwrap(), + scope.to_string(), + ); 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_name: NonBlankString::from_str("oci://ghcr.io/grafana/helm-charts/grafana-operator") + .unwrap(), chart_version: None, - values_overrides: None, - values_yaml: Some(values.to_string()), + values_overrides: Some(values_overrides), + values_yaml: None, create_namespace: true, install_only: true, repository: None, diff --git a/harmony/src/modules/monitoring/grafana/mod.rs b/harmony/src/modules/monitoring/grafana/mod.rs index c821bcb..8dccab1 100644 --- a/harmony/src/modules/monitoring/grafana/mod.rs +++ b/harmony/src/modules/monitoring/grafana/mod.rs @@ -1 +1,2 @@ +pub mod grafana; pub mod helm; diff --git a/harmony/src/modules/monitoring/kube_prometheus/crd/crd_alertmanager_config.rs b/harmony/src/modules/monitoring/kube_prometheus/crd/crd_alertmanager_config.rs index 2165a4a..0ac8fc7 100644 --- a/harmony/src/modules/monitoring/kube_prometheus/crd/crd_alertmanager_config.rs +++ b/harmony/src/modules/monitoring/kube_prometheus/crd/crd_alertmanager_config.rs @@ -1,12 +1,25 @@ use std::sync::Arc; +use async_trait::async_trait; use kube::CustomResource; use schemars::JsonSchema; use serde::{Deserialize, Serialize}; -use crate::topology::{ - k8s::K8sClient, - oberservability::monitoring::{AlertReceiver, AlertSender}, +use crate::{ + interpret::{InterpretError, Outcome}, + inventory::Inventory, + modules::{ + monitoring::{ + grafana::grafana::Grafana, kube_prometheus::crd::service_monitor::ServiceMonitor, + }, + prometheus::prometheus::PrometheusMonitoring, + }, + topology::{ + K8sclient, Topology, + installable::Installable, + k8s::K8sClient, + oberservability::monitoring::{AlertReceiver, AlertSender}, + }, }; #[derive(CustomResource, Serialize, Deserialize, Debug, Clone, JsonSchema)] @@ -26,6 +39,7 @@ pub struct AlertmanagerConfigSpec { pub struct CRDPrometheus { pub namespace: String, pub client: Arc, + pub service_monitor: Vec, } impl AlertSender for CRDPrometheus { @@ -48,3 +62,24 @@ impl Serialize for Box> { todo!() } } + +#[async_trait] +impl + Grafana> Installable + for CRDPrometheus +{ + async fn configure(&self, inventory: &Inventory, topology: &T) -> Result<(), InterpretError> { + topology.ensure_grafana_operator_ready(inventory).await?; + topology.ensure_prometheus_operator(self, inventory).await?; + Ok(()) + } + + async fn ensure_installed( + &self, + inventory: &Inventory, + topology: &T, + ) -> Result<(), InterpretError> { + topology.install_grafana().await?; + topology.install_prometheus(&self, inventory, None).await?; + Ok(()) + } +} diff --git a/harmony/src/modules/monitoring/prometheus/prometheus.rs b/harmony/src/modules/monitoring/prometheus/prometheus.rs index a207d5a..2fe0d06 100644 --- a/harmony/src/modules/monitoring/prometheus/prometheus.rs +++ b/harmony/src/modules/monitoring/prometheus/prometheus.rs @@ -114,7 +114,7 @@ impl Prometheus { }; if let Some(ns) = namespace.as_deref() { - grafana_helm_chart_score(ns) + grafana_helm_chart_score(ns, false) .interpret(inventory, topology) .await } else { diff --git a/harmony/src/modules/prometheus/k8s_prometheus_alerting_score.rs b/harmony/src/modules/prometheus/k8s_prometheus_alerting_score.rs index 24ca918..2cb4ffb 100644 --- a/harmony/src/modules/prometheus/k8s_prometheus_alerting_score.rs +++ b/harmony/src/modules/prometheus/k8s_prometheus_alerting_score.rs @@ -39,7 +39,7 @@ use crate::{ }; use harmony_types::id::Id; -use super::prometheus::PrometheusApplicationMonitoring; +use super::prometheus::PrometheusMonitoring; #[derive(Clone, Debug, Serialize)] pub struct K8sPrometheusCRDAlertingScore { @@ -49,7 +49,7 @@ pub struct K8sPrometheusCRDAlertingScore { pub prometheus_rules: Vec, } -impl> Score +impl> Score for K8sPrometheusCRDAlertingScore { fn create_interpret(&self) -> Box> { @@ -75,7 +75,7 @@ pub struct K8sPrometheusCRDAlertingInterpret { } #[async_trait] -impl> Interpret +impl> Interpret for K8sPrometheusCRDAlertingInterpret { async fn execute( diff --git a/harmony/src/modules/prometheus/prometheus.rs b/harmony/src/modules/prometheus/prometheus.rs index d3940c7..efb89da 100644 --- a/harmony/src/modules/prometheus/prometheus.rs +++ b/harmony/src/modules/prometheus/prometheus.rs @@ -9,11 +9,17 @@ use crate::{ }; #[async_trait] -pub trait PrometheusApplicationMonitoring { +pub trait PrometheusMonitoring { async fn install_prometheus( &self, sender: &S, inventory: &Inventory, receivers: Option>>>, ) -> Result; + + async fn ensure_prometheus_operator( + &self, + sender: &S, + inventory: &Inventory, + ) -> Result; } diff --git a/harmony/src/modules/prometheus/rhob_alerting_score.rs b/harmony/src/modules/prometheus/rhob_alerting_score.rs index 95908d5..644e6f9 100644 --- a/harmony/src/modules/prometheus/rhob_alerting_score.rs +++ b/harmony/src/modules/prometheus/rhob_alerting_score.rs @@ -38,7 +38,7 @@ use crate::{ }; use harmony_types::id::Id; -use super::prometheus::PrometheusApplicationMonitoring; +use super::prometheus::PrometheusMonitoring; #[derive(Clone, Debug, Serialize)] pub struct RHOBAlertingScore { @@ -48,8 +48,8 @@ pub struct RHOBAlertingScore { pub prometheus_rules: Vec, } -impl> - Score for RHOBAlertingScore +impl> Score + for RHOBAlertingScore { fn create_interpret(&self) -> Box> { Box::new(RHOBAlertingInterpret { @@ -74,8 +74,8 @@ pub struct RHOBAlertingInterpret { } #[async_trait] -impl> - Interpret for RHOBAlertingInterpret +impl> Interpret + for RHOBAlertingInterpret { async fn execute( &self, From 1f3796f50301b38746366c4d5e4909332db203dd Mon Sep 17 00:00:00 2001 From: Willem Date: Thu, 9 Oct 2025 12:26:05 -0400 Subject: [PATCH 02/11] refactor(prometheus): modified crd prometheus to impl the installable trait --- examples/try_rust_webapp/src/main.rs | 2 +- harmony/src/domain/topology/k8s_anywhere.rs | 208 ++++++++++++++---- .../topology/oberservability/monitoring.rs | 2 + .../application/features/monitoring.rs | 17 +- .../application_monitoring_score.rs | 79 ++----- .../src/modules/monitoring/grafana/grafana.rs | 4 +- .../monitoring/grafana/helm/helm_grafana.rs | 17 +- .../crd/crd_alertmanager_config.rs | 2 +- 8 files changed, 219 insertions(+), 112 deletions(-) diff --git a/examples/try_rust_webapp/src/main.rs b/examples/try_rust_webapp/src/main.rs index 56a058d..7bfdf57 100644 --- a/examples/try_rust_webapp/src/main.rs +++ b/examples/try_rust_webapp/src/main.rs @@ -3,7 +3,7 @@ use harmony::{ modules::{ application::{ ApplicationScore, RustWebFramework, RustWebapp, - features::{PackagingDeployment, rhob_monitoring::Monitoring}, + features::{Monitoring, PackagingDeployment}, }, monitoring::alert_channel::discord_alert_channel::DiscordWebhook, }, diff --git a/harmony/src/domain/topology/k8s_anywhere.rs b/harmony/src/domain/topology/k8s_anywhere.rs index 6dfb1a8..895f7da 100644 --- a/harmony/src/domain/topology/k8s_anywhere.rs +++ b/harmony/src/domain/topology/k8s_anywhere.rs @@ -1,7 +1,7 @@ -use std::{process::Command, sync::Arc}; +use std::{collections::BTreeMap, process::Command, sync::Arc}; use async_trait::async_trait; -use kube::api::GroupVersionKind; +use kube::api::{GroupVersionKind, ObjectMeta}; use log::{debug, info, warn}; use serde::Serialize; use tokio::sync::OnceCell; @@ -12,12 +12,20 @@ use crate::{ inventory::Inventory, modules::{ k3d::K3DInstallationScore, + k8s::ingress::{K8sIngressScore, PathType}, monitoring::{ grafana::{grafana::Grafana, helm::helm_grafana::grafana_helm_chart_score}, kube_prometheus::crd::{ crd_alertmanager_config::CRDPrometheus, + crd_grafana::{ + Grafana as GrafanaCRD, GrafanaDashboard, GrafanaDashboardSpec, + GrafanaDatasource, GrafanaDatasourceConfig, GrafanaDatasourceSpec, GrafanaSpec, + }, + crd_prometheuses::LabelSelector, + grafana_default_dashboard::build_default_dashboard, prometheus_operator::prometheus_operator_helm_chart_score, - rhob_alertmanager_config::RHOBObservability, service_monitor::ServiceMonitor, + rhob_alertmanager_config::RHOBObservability, + service_monitor::ServiceMonitor, }, }, prometheus::{ @@ -90,10 +98,11 @@ impl K8sclient for K8sAnywhereTopology { #[async_trait] impl Grafana for K8sAnywhereTopology { - async fn ensure_grafana_operator_ready( + async fn ensure_grafana_operator( &self, inventory: &Inventory, ) -> Result { + debug!("ensure grafana operator"); let client = self.k8s_client().await.unwrap(); let grafana_gvk = GroupVersionKind { group: "grafana.integreatly.org".to_string(), @@ -112,6 +121,7 @@ impl Grafana for K8sAnywhereTopology { details: "Found grafana CRDs in cluster".to_string(), }); } + Err(_) => { return self .install_grafana_operator(inventory, Some("grafana")) @@ -120,7 +130,41 @@ impl Grafana for K8sAnywhereTopology { }; } async fn install_grafana(&self) -> Result { - todo!() + debug!("install grafana"); + let ns = "grafana"; + + let mut label = BTreeMap::new(); + + label.insert("dashboards".to_string(), "grafana".to_string()); + let label_selector = LabelSelector { + match_labels: label.clone(), + match_expressions: vec![], + }; + + let client = self.k8s_client().await?; + + let datasource = self.build_grafana_datasource(ns, &label_selector); + + client.apply(&datasource, Some(ns)).await?; + + let dashboard = self.build_grafana_dashboard(ns, &label_selector); + + client.apply(&dashboard, Some(ns)).await?; + + let grafana = self.build_grafana(ns, &label); + + client.apply(&grafana, Some(ns)).await?; + + let grafana_ingress = self.build_grafana_ingress(ns).await; + + grafana_ingress + .interpret(&Inventory::empty(), self) + .await + .map_err(|e| PreparationError::new(e.to_string()))?; + + Ok(PreparationOutcome::Success { + details: "Installed grafana composants".to_string(), + }) } } @@ -129,49 +173,38 @@ impl PrometheusMonitoring for K8sAnywhereTopology { async fn install_prometheus( &self, sender: &CRDPrometheus, - inventory: &Inventory, - receivers: Option>>>, + _inventory: &Inventory, + _receivers: Option>>>, ) -> Result { - let po_result = self.ensure_prometheus_operator(sender).await?; + let client = self.k8s_client().await?; - if po_result == PreparationOutcome::Noop { - debug!("Skipping Prometheus CR installation due to missing operator."); - return Ok(po_result); - } - - let result = self - .get_k8s_prometheus_application_score( - sender.clone(), - receivers, - Some(sender.service_monitor.clone()), - ) - .await - .interpret(inventory, self) - .await; - - match result { - Ok(outcome) => match outcome.status { - InterpretStatus::SUCCESS => Ok(PreparationOutcome::Success { - details: outcome.message, - }), - InterpretStatus::NOOP => Ok(PreparationOutcome::Noop), - _ => Err(PreparationError::new(outcome.message)), - }, - Err(err) => Err(PreparationError::new(err.to_string())), + for monitor in sender.service_monitor.iter() { + client + .apply(monitor, Some(&sender.namespace)) + .await + .map_err(|e| PreparationError::new(e.to_string()))?; } + Ok(PreparationOutcome::Success { + details: "successfuly installed prometheus components".to_string(), + }) } + async fn ensure_prometheus_operator( &self, sender: &CRDPrometheus, - inventory: &Inventory, + _inventory: &Inventory, ) -> Result { let po_result = self.ensure_prometheus_operator(sender).await?; - if po_result == PreparationOutcome::Noop { - debug!("Skipping Prometheus CR installation due to missing operator."); - return Ok(po_result); - } else { - todo!() + match po_result { + PreparationOutcome::Success { details: _ } => { + debug!("Detected prometheus crds operator present in cluster."); + return Ok(po_result); + } + PreparationOutcome::Noop => { + debug!("Skipping Prometheus CR installation due to missing operator."); + return Ok(po_result); + } } } } @@ -211,6 +244,7 @@ impl PrometheusMonitoring for K8sAnywhereTopology { Err(err) => Err(PreparationError::new(err.to_string())), } } + async fn ensure_prometheus_operator( &self, sender: &RHOBObservability, @@ -300,6 +334,95 @@ impl K8sAnywhereTopology { .clone() } + fn build_grafana_datasource( + &self, + ns: &str, + label_selector: &LabelSelector, + ) -> GrafanaDatasource { + let mut json_data = BTreeMap::new(); + json_data.insert("timeInterval".to_string(), "5s".to_string()); + + let graf_data_source = GrafanaDatasource { + metadata: ObjectMeta { + name: Some(format!("grafana-datasource-{}", ns)), + namespace: Some(ns.to_string()), + ..Default::default() + }, + spec: GrafanaDatasourceSpec { + instance_selector: label_selector.clone(), + allow_cross_namespace_import: Some(false), + datasource: GrafanaDatasourceConfig { + access: "proxy".to_string(), + database: Some("prometheus".to_string()), + json_data: Some(json_data), + //this is fragile + name: format!("prometheus-{}-0", ns), + r#type: "prometheus".to_string(), + url: format!("http://prometheus-operated.{}.svc.cluster.local:9090", ns), + }, + }, + }; + graf_data_source + } + + fn build_grafana_dashboard( + &self, + ns: &str, + label_selector: &LabelSelector, + ) -> GrafanaDashboard { + let json = build_default_dashboard(ns); + let graf_dashboard = GrafanaDashboard { + metadata: ObjectMeta { + name: Some(format!("grafana-dashboard-{}", ns)), + namespace: Some(ns.to_string()), + ..Default::default() + }, + spec: GrafanaDashboardSpec { + resync_period: Some("30s".to_string()), + instance_selector: label_selector.clone(), + json, + }, + }; + graf_dashboard + } + + fn build_grafana(&self, ns: &str, labels: &BTreeMap) -> GrafanaCRD { + let grafana = GrafanaCRD { + metadata: ObjectMeta { + name: Some(format!("grafana-{}", ns)), + namespace: Some(ns.to_string()), + labels: Some(labels.clone()), + ..Default::default() + }, + spec: GrafanaSpec { + config: None, + admin_user: None, + admin_password: None, + ingress: None, + persistence: None, + resources: None, + }, + }; + grafana + } + + async fn build_grafana_ingress(&self, ns: &str) -> K8sIngressScore { + let domain = self.get_domain(&format!("grafana-{}", ns)).await.unwrap(); + let name = format!("{}-grafana", ns); + let backend_service = format!("grafana-{}-service", ns); + + K8sIngressScore { + name: fqdn::fqdn!(&name), + host: fqdn::fqdn!(&domain), + backend_service: fqdn::fqdn!(&backend_service), + port: 3000, + path: Some("/".to_string()), + path_type: Some(PathType::Prefix), + namespace: Some(fqdn::fqdn!(&ns)), + ingress_class_name: Some("openshift-default".to_string()), + } + } + async fn get_cluster_observability_operator_prometheus_application_score( &self, sender: RHOBObservability, @@ -607,7 +730,14 @@ impl K8sAnywhereTopology { inventory: &Inventory, ns: Option<&str>, ) -> Result { - let _grafana_operator_score = grafana_helm_chart_score(ns.unwrap(), true) + let namespace = ns.unwrap_or("grafana"); + info!("installing grafana operator in ns {namespace}"); + let tenant = self.get_k8s_tenant_manager()?.get_tenant_config().await; + let mut namespace_scope = false; + if tenant.is_some() { + namespace_scope = true; + } + let _grafana_operator_score = grafana_helm_chart_score(namespace, namespace_scope) .interpret(inventory, self) .await; Ok(PreparationOutcome::Success { diff --git a/harmony/src/domain/topology/oberservability/monitoring.rs b/harmony/src/domain/topology/oberservability/monitoring.rs index 1489e83..0c57ea4 100644 --- a/harmony/src/domain/topology/oberservability/monitoring.rs +++ b/harmony/src/domain/topology/oberservability/monitoring.rs @@ -30,6 +30,7 @@ impl, T: Topology> Interpret for AlertingInte inventory: &Inventory, topology: &T, ) -> Result { + debug!("hit sender configure for AlertingInterpret"); self.sender.configure(inventory, topology).await?; for receiver in self.receivers.iter() { receiver.install(&self.sender).await?; @@ -38,6 +39,7 @@ impl, T: Topology> Interpret for AlertingInte debug!("installing rule: {:#?}", rule); rule.install(&self.sender).await?; } + debug!("hit sender ensure installed for AlertingInterpret"); self.sender.ensure_installed(inventory, topology).await?; Ok(Outcome::success(format!( "successfully installed alert sender {}", diff --git a/harmony/src/modules/application/features/monitoring.rs b/harmony/src/modules/application/features/monitoring.rs index 0fd155d..fd6ae2a 100644 --- a/harmony/src/modules/application/features/monitoring.rs +++ b/harmony/src/modules/application/features/monitoring.rs @@ -2,7 +2,11 @@ use crate::modules::application::{ Application, ApplicationFeature, InstallationError, InstallationOutcome, }; use crate::modules::monitoring::application_monitoring::application_monitoring_score::ApplicationMonitoringScore; +use crate::modules::monitoring::grafana::grafana::Grafana; use crate::modules::monitoring::kube_prometheus::crd::crd_alertmanager_config::CRDPrometheus; +use crate::modules::monitoring::kube_prometheus::crd::service_monitor::{ + ServiceMonitor, ServiceMonitorSpec, +}; use crate::topology::MultiTargetTopology; use crate::topology::ingress::Ingress; use crate::{ @@ -22,6 +26,7 @@ use base64::{Engine as _, engine::general_purpose}; use harmony_secret::SecretManager; use harmony_secret_derive::Secret; use harmony_types::net::Url; +use kube::api::ObjectMeta; use log::{debug, info}; use serde::{Deserialize, Serialize}; use std::sync::Arc; @@ -41,6 +46,7 @@ impl< + K8sclient + MultiTargetTopology + PrometheusMonitoring + + Grafana + Ingress + std::fmt::Debug, > ApplicationFeature for Monitoring @@ -57,11 +63,20 @@ impl< .unwrap_or_else(|| self.application.name()); let domain = topology.get_domain("ntfy").await.unwrap(); + let app_service_monitor = ServiceMonitor { + metadata: ObjectMeta { + name: Some(self.application.name()), + namespace: Some(namespace.clone()), + ..Default::default() + }, + spec: ServiceMonitorSpec::default(), + }; + let mut alerting_score = ApplicationMonitoringScore { sender: CRDPrometheus { namespace: namespace.clone(), client: topology.k8s_client().await.unwrap(), - service_monitor: vec![], + service_monitor: vec![app_service_monitor], }, application: self.application.clone(), receivers: self.alert_receiver.clone(), diff --git a/harmony/src/modules/monitoring/application_monitoring/application_monitoring_score.rs b/harmony/src/modules/monitoring/application_monitoring/application_monitoring_score.rs index 2780edd..0f6e0ec 100644 --- a/harmony/src/modules/monitoring/application_monitoring/application_monitoring_score.rs +++ b/harmony/src/modules/monitoring/application_monitoring/application_monitoring_score.rs @@ -1,21 +1,23 @@ use std::sync::Arc; -use async_trait::async_trait; +use log::debug; use serde::Serialize; use crate::{ - data::Version, - interpret::{Interpret, InterpretError, InterpretName, InterpretStatus, Outcome}, - inventory::Inventory, + interpret::Interpret, modules::{ application::Application, - monitoring::kube_prometheus::crd::crd_alertmanager_config::CRDPrometheus, + monitoring::{ + grafana::grafana::Grafana, kube_prometheus::crd::crd_alertmanager_config::CRDPrometheus, + }, prometheus::prometheus::PrometheusMonitoring, }, score::Score, - topology::{PreparationOutcome, Topology, oberservability::monitoring::AlertReceiver}, + topology::{ + K8sclient, Topology, + oberservability::monitoring::{AlertReceiver, AlertingInterpret}, + }, }; -use harmony_types::id::Id; #[derive(Debug, Clone, Serialize)] pub struct ApplicationMonitoringScore { @@ -24,10 +26,15 @@ pub struct ApplicationMonitoringScore { pub receivers: Vec>>, } -impl> Score for ApplicationMonitoringScore { +impl + K8sclient + Grafana> Score + for ApplicationMonitoringScore +{ fn create_interpret(&self) -> Box> { - Box::new(ApplicationMonitoringInterpret { - score: self.clone(), + debug!("creating alerting interpret"); + Box::new(AlertingInterpret { + sender: self.sender.clone(), + receivers: self.receivers.clone(), + rules: vec![], }) } @@ -38,55 +45,3 @@ impl> Score for Application ) } } - -#[derive(Debug)] -pub struct ApplicationMonitoringInterpret { - score: ApplicationMonitoringScore, -} - -#[async_trait] -impl> Interpret - for ApplicationMonitoringInterpret -{ - async fn execute( - &self, - inventory: &Inventory, - topology: &T, - ) -> Result { - let result = topology - .install_prometheus( - &self.score.sender, - inventory, - Some(self.score.receivers.clone()), - ) - .await; - - match result { - Ok(outcome) => match outcome { - PreparationOutcome::Success { details: _ } => { - Ok(Outcome::success("Prometheus installed".into())) - } - PreparationOutcome::Noop => { - Ok(Outcome::noop("Prometheus installation skipped".into())) - } - }, - Err(err) => Err(InterpretError::from(err)), - } - } - - fn get_name(&self) -> InterpretName { - InterpretName::ApplicationMonitoring - } - - fn get_version(&self) -> Version { - todo!() - } - - fn get_status(&self) -> InterpretStatus { - todo!() - } - - fn get_children(&self) -> Vec { - todo!() - } -} diff --git a/harmony/src/modules/monitoring/grafana/grafana.rs b/harmony/src/modules/monitoring/grafana/grafana.rs index 411d7a6..5ab57c2 100644 --- a/harmony/src/modules/monitoring/grafana/grafana.rs +++ b/harmony/src/modules/monitoring/grafana/grafana.rs @@ -1,4 +1,5 @@ use async_trait::async_trait; +use k8s_openapi::Resource; use crate::{ inventory::Inventory, @@ -7,9 +8,10 @@ use crate::{ #[async_trait] pub trait Grafana { - async fn ensure_grafana_operator_ready( + async fn ensure_grafana_operator( &self, inventory: &Inventory, ) -> Result; + async fn install_grafana(&self) -> Result; } diff --git a/harmony/src/modules/monitoring/grafana/helm/helm_grafana.rs b/harmony/src/modules/monitoring/grafana/helm/helm_grafana.rs index 094beca..2965ada 100644 --- a/harmony/src/modules/monitoring/grafana/helm/helm_grafana.rs +++ b/harmony/src/modules/monitoring/grafana/helm/helm_grafana.rs @@ -1,24 +1,27 @@ use non_blank_string_rs::NonBlankString; use std::{collections::HashMap, str::FromStr}; -use crate::modules::helm::chart::HelmChartScore; +use crate::modules::helm::chart::{HelmChartScore, HelmRepository}; -pub fn grafana_helm_chart_score(ns: &str, scope: bool) -> HelmChartScore { +pub fn grafana_helm_chart_score(ns: &str, namespace_scope: bool) -> HelmChartScore { let mut values_overrides = HashMap::new(); values_overrides.insert( NonBlankString::from_str("namespaceScope").unwrap(), - scope.to_string(), + namespace_scope.to_string(), ); 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-operator") - .unwrap(), + release_name: NonBlankString::from_str("grafana-operator").unwrap(), + chart_name: NonBlankString::from_str("grafana/grafana-operator").unwrap(), chart_version: None, values_overrides: Some(values_overrides), values_yaml: None, create_namespace: true, install_only: true, - repository: None, + repository: Some(HelmRepository::new( + "grafana".to_string(), + url::Url::parse("https://grafana.github.io/helm-charts").unwrap(), + true, + )), } } diff --git a/harmony/src/modules/monitoring/kube_prometheus/crd/crd_alertmanager_config.rs b/harmony/src/modules/monitoring/kube_prometheus/crd/crd_alertmanager_config.rs index 0ac8fc7..ceeca41 100644 --- a/harmony/src/modules/monitoring/kube_prometheus/crd/crd_alertmanager_config.rs +++ b/harmony/src/modules/monitoring/kube_prometheus/crd/crd_alertmanager_config.rs @@ -68,7 +68,7 @@ impl + Grafana> In for CRDPrometheus { async fn configure(&self, inventory: &Inventory, topology: &T) -> Result<(), InterpretError> { - topology.ensure_grafana_operator_ready(inventory).await?; + topology.ensure_grafana_operator(inventory).await?; topology.ensure_prometheus_operator(self, inventory).await?; Ok(()) } From 85bec66e5878f0eb9b496b37be614ab7d2f904e3 Mon Sep 17 00:00:00 2001 From: Willem Date: Fri, 10 Oct 2025 12:09:26 -0400 Subject: [PATCH 03/11] wip: fixing grafana datasource for openshift which requires creating a token, sa, secret and inserting them into the grafanadatasource --- harmony/src/domain/topology/k8s_anywhere.rs | 140 ++++++++++++++++-- .../kube_prometheus/crd/crd_grafana.rs | 34 ++++- 2 files changed, 159 insertions(+), 15 deletions(-) diff --git a/harmony/src/domain/topology/k8s_anywhere.rs b/harmony/src/domain/topology/k8s_anywhere.rs index 895f7da..efbe33f 100644 --- a/harmony/src/domain/topology/k8s_anywhere.rs +++ b/harmony/src/domain/topology/k8s_anywhere.rs @@ -1,7 +1,15 @@ use std::{collections::BTreeMap, process::Command, sync::Arc}; use async_trait::async_trait; -use kube::api::{GroupVersionKind, ObjectMeta}; +use k8s_openapi::api::{ + authentication::v1::{TokenRequest, TokenRequestSpec}, + core::v1::{Secret, ServiceAccount}, + rbac::v1::{ClusterRoleBinding, RoleRef, Subject}, +}; +use kube::{ + Api, + api::{GroupVersionKind, ObjectMeta, PostParams}, +}; use log::{debug, info, warn}; use serde::Serialize; use tokio::sync::OnceCell; @@ -19,12 +27,14 @@ use crate::{ crd_alertmanager_config::CRDPrometheus, crd_grafana::{ Grafana as GrafanaCRD, GrafanaDashboard, GrafanaDashboardSpec, - GrafanaDatasource, GrafanaDatasourceConfig, GrafanaDatasourceSpec, GrafanaSpec, + GrafanaDatasource, GrafanaDatasourceConfig, GrafanaDatasourceJsonData, + GrafanaDatasourceSecureJsonData, GrafanaDatasourceSpec, GrafanaSpec, }, crd_prometheuses::LabelSelector, grafana_default_dashboard::build_default_dashboard, prometheus_operator::prometheus_operator_helm_chart_score, rhob_alertmanager_config::RHOBObservability, + role::build_prom_service_account, service_monitor::ServiceMonitor, }, }, @@ -142,8 +152,26 @@ impl Grafana for K8sAnywhereTopology { }; let client = self.k8s_client().await?; + let url = format!("{}:9091", self.get_domain("thanos-querier").await.unwrap()); + + let sa = self.build_service_account(); + //TODO finish this section + //needs apply Api or something + client.apply(&sa, Some(ns)).await?; - let datasource = self.build_grafana_datasource(ns, &label_selector); + let token_request =self.get_token_request(); + //this wont work needs a new function for apply secret + client.apply(&token_request, Some(ns)).await?; + + let clusterrolebinding = self.build_cluster_rolebinding(); + + client.apply(&clusterrolebinding, Some(ns)).await?; + + let secret = self.build_token_secret(); + + client.apply(&secret, Some(ns)).await?; + + let datasource = self.build_grafana_datasource(ns, &label_selector, &url); client.apply(&datasource, Some(ns)).await?; @@ -334,35 +362,121 @@ impl K8sAnywhereTopology { .clone() } + pub fn build_service_account(&self, name: &str, namespace: &str) -> ServiceAccount { + build_prom_service_account(name.to_string(), namespace.to_string()) + } + + pub fn build_cluster_rolebinding( + &self, + ns: &str, + account_name: &str, + role: &str, + ) -> ClusterRoleBinding { + ClusterRoleBinding { + metadata: ObjectMeta { + name: Some(format!("{}-view-binding", account_name)), + ..Default::default() + }, + role_ref: RoleRef { + api_group: "rbac.authorization.k8s.io".into(), + kind: "ClusterRole".into(), + name: role.into(), + }, + subjects: Some(vec![Subject { + kind: "ServiceAccount".into(), + name: account_name.into(), + namespace: Some(ns.into()), + ..Default::default() + }]), + } + } + + pub fn get_token_request(&self) -> TokenRequest { + TokenRequest { + spec: TokenRequestSpec { + audiences: vec!["https://kubernetes.default.svc".to_string()], + expiration_seconds: Some(3600), + ..Default::default() + }, + ..Default::default() + } + } + + pub fn build_token_secret(&self, token: &str, ns: &str) -> Secret { + Secret { + metadata: ObjectMeta { + name: Some("grafana-credentials".into()), + namespace: Some(ns.into()), + ..Default::default() + }, + string_data: Some(std::collections::BTreeMap::from([( + "PROMETHEUS_TOKEN".into(), + format!("Bearer {}", token), + )])), + ..Default::default() + } + } + fn build_grafana_datasource( &self, ns: &str, label_selector: &LabelSelector, + url: &str, ) -> GrafanaDatasource { let mut json_data = BTreeMap::new(); json_data.insert("timeInterval".to_string(), "5s".to_string()); + // + // let graf_data_source = GrafanaDatasource { + // metadata: ObjectMeta { + // name: Some(format!("grafana-datasource-{}", ns)), + // namespace: Some(ns.to_string()), + // ..Default::default() + // }, + // spec: GrafanaDatasourceSpec { + // instance_selector: label_selector.clone(), + // allow_cross_namespace_import: Some(false), + // datasource: GrafanaDatasourceConfig { + // access: "proxy".to_string(), + // database: Some("prometheus".to_string()), + // json_data: Some(json_data), + // //this is fragile + // name: format!("prometheus-{}-0", ns), + // r#type: "prometheus".to_string(), + // url: url.to_string(), + // //url: format!("http://prometheus-operated.{}.svc.cluster.local:9090", ns), + // }, + // }, + // }; + // graf_data_source - let graf_data_source = GrafanaDatasource { + GrafanaDatasource { metadata: ObjectMeta { - name: Some(format!("grafana-datasource-{}", ns)), + name: Some("thanos-prometheus".to_string()), namespace: Some(ns.to_string()), ..Default::default() }, spec: GrafanaDatasourceSpec { instance_selector: label_selector.clone(), - allow_cross_namespace_import: Some(false), + allow_cross_namespace_import: Some(true), datasource: GrafanaDatasourceConfig { access: "proxy".to_string(), - database: Some("prometheus".to_string()), - json_data: Some(json_data), - //this is fragile - name: format!("prometheus-{}-0", ns), + name: "OpenShift-Thanos".to_string(), r#type: "prometheus".to_string(), - url: format!("http://prometheus-operated.{}.svc.cluster.local:9090", ns), + url: url.to_string(), + database: None, + json_data: Some(GrafanaDatasourceJsonData { + time_interval: Some("60s".to_string()), + http_header_name1: Some("Authorization".to_string()), + }), + secure_json_data: Some(GrafanaDatasourceSecureJsonData { + http_header_value1: Some("Bearer eyJhbGc...".to_string()), + }), + is_default: Some(false), + editable: Some(true), + version: Some(1), }, }, - }; - graf_data_source + } } fn build_grafana_dashboard( diff --git a/harmony/src/modules/monitoring/kube_prometheus/crd/crd_grafana.rs b/harmony/src/modules/monitoring/kube_prometheus/crd/crd_grafana.rs index 793f639..4134670 100644 --- a/harmony/src/modules/monitoring/kube_prometheus/crd/crd_grafana.rs +++ b/harmony/src/modules/monitoring/kube_prometheus/crd/crd_grafana.rs @@ -133,13 +133,43 @@ pub struct GrafanaDatasourceSpec { pub struct GrafanaDatasourceConfig { pub access: String, pub database: Option, - #[serde(default, skip_serializing_if = "Option::is_none")] - pub json_data: Option>, pub name: String, pub r#type: String, pub url: String, + /// Represents jsonData in the GrafanaDatasource spec + #[serde(default, skip_serializing_if = "Option::is_none")] + pub json_data: Option, + + /// Represents secureJsonData (secrets) + #[serde(default, skip_serializing_if = "Option::is_none")] + pub secure_json_data: Option, + + #[serde(default, skip_serializing_if = "Option::is_none")] + pub is_default: Option, + + #[serde(default, skip_serializing_if = "Option::is_none")] + pub editable: Option, + + #[serde(default, skip_serializing_if = "Option::is_none")] + pub version: Option, } +#[derive(Serialize, Deserialize, Debug, Clone, JsonSchema)] +#[serde(rename_all = "camelCase")] +pub struct GrafanaDatasourceJsonData { + #[serde(default, skip_serializing_if = "Option::is_none")] + pub time_interval: Option, + + #[serde(default, skip_serializing_if = "Option::is_none")] + pub http_header_name1: Option, +} + +#[derive(Serialize, Deserialize, Debug, Clone, JsonSchema)] +#[serde(rename_all = "camelCase")] +pub struct GrafanaDatasourceSecureJsonData { + #[serde(default, skip_serializing_if = "Option::is_none")] + pub http_header_value1: Option, +} // ------------------------------------------------------------------------------------------------ #[derive(Serialize, Deserialize, Debug, Clone, JsonSchema, Default)] From 06a0c44c3cfa982f0ce585ad815caec4270208f7 Mon Sep 17 00:00:00 2001 From: Willem Date: Tue, 14 Oct 2025 15:53:42 -0400 Subject: [PATCH 04/11] wip: connected the thanos-datasource to grafana, need to complete connecting the openshift-userworkload-monitoring as well --- harmony/src/domain/topology/k8s.rs | 17 +- harmony/src/domain/topology/k8s_anywhere.rs | 194 ++++++++++++------ .../kube_prometheus/crd/crd_grafana.rs | 12 +- .../k8s_prometheus_alerting_score.rs | 16 +- 4 files changed, 165 insertions(+), 74 deletions(-) diff --git a/harmony/src/domain/topology/k8s.rs b/harmony/src/domain/topology/k8s.rs index 144533c..f1a783f 100644 --- a/harmony/src/domain/topology/k8s.rs +++ b/harmony/src/domain/topology/k8s.rs @@ -1,12 +1,20 @@ use derive_new::new; +use http::StatusCode; use k8s_openapi::{ ClusterResourceScope, NamespaceResourceScope, - api::{apps::v1::Deployment, core::v1::Pod}, + api::{ + apps::v1::Deployment, + authentication::v1::{TokenRequest, TokenRequestSpec, TokenRequestStatus}, + core::v1::{Pod, ServiceAccount}, + }, apimachinery::pkg::version::Info, }; use kube::{ Client, Config, Discovery, Error, Resource, - api::{Api, AttachParams, DeleteParams, ListParams, Patch, PatchParams, ResourceExt}, + api::{ + Api, AttachParams, DeleteParams, ListParams, ObjectMeta, Patch, PatchParams, PostParams, + ResourceExt, + }, config::{KubeConfigOptions, Kubeconfig}, core::ErrorResponse, runtime::reflector::Lookup, @@ -54,6 +62,11 @@ impl K8sClient { }) } + pub async fn service_account_api(&self, namespace: &str) -> Api { + let api: Api = Api::namespaced(self.client.clone(), namespace); + api + } + pub async fn get_apiserver_version(&self) -> Result { let client: Client = self.client.clone(); let version_info: Info = client.apiserver_version().await?; diff --git a/harmony/src/domain/topology/k8s_anywhere.rs b/harmony/src/domain/topology/k8s_anywhere.rs index efbe33f..cb37ece 100644 --- a/harmony/src/domain/topology/k8s_anywhere.rs +++ b/harmony/src/domain/topology/k8s_anywhere.rs @@ -1,8 +1,12 @@ -use std::{collections::BTreeMap, process::Command, sync::Arc}; +use std::{ + collections::{BTreeMap, HashMap}, + process::Command, + sync::Arc, +}; use async_trait::async_trait; use k8s_openapi::api::{ - authentication::v1::{TokenRequest, TokenRequestSpec}, + authentication::v1::{TokenRequest, TokenRequestSpec, TokenRequestStatus}, core::v1::{Secret, ServiceAccount}, rbac::v1::{ClusterRoleBinding, RoleRef, Subject}, }; @@ -150,39 +154,90 @@ impl Grafana for K8sAnywhereTopology { match_labels: label.clone(), match_expressions: vec![], }; - + debug!("getting client"); let client = self.k8s_client().await?; - let url = format!("{}:9091", self.get_domain("thanos-querier").await.unwrap()); - - let sa = self.build_service_account(); - //TODO finish this section - //needs apply Api or something - client.apply(&sa, Some(ns)).await?; - - let token_request =self.get_token_request(); - //this wont work needs a new function for apply secret - client.apply(&token_request, Some(ns)).await?; - - let clusterrolebinding = self.build_cluster_rolebinding(); - - client.apply(&clusterrolebinding, Some(ns)).await?; - - let secret = self.build_token_secret(); - - client.apply(&secret, Some(ns)).await?; - - let datasource = self.build_grafana_datasource(ns, &label_selector, &url); - - client.apply(&datasource, Some(ns)).await?; - - let dashboard = self.build_grafana_dashboard(ns, &label_selector); - - client.apply(&dashboard, Some(ns)).await?; + info!("creating grafanas crd"); let grafana = self.build_grafana(ns, &label); client.apply(&grafana, Some(ns)).await?; + client + .wait_until_deployment_ready( + "grafana-grafana-deployment".to_string(), + Some("grafana"), + Some(15), + ) + .await?; + + let sa_name = "grafana-grafana-sa"; + + debug!("creating token for sevice account {sa_name}"); + let token = self.create_service_account_token(sa_name, ns).await?; + + debug!("creating secret"); + let secret_name = "grafana-sa-secret"; + let secret = self.build_token_secret(secret_name, &token.token, ns).await; + + client.apply(&secret, Some(ns)).await?; + + debug!("creating grafana clusterrole binding"); + let clusterrolebinding = + self.build_cluster_rolebinding(sa_name, "cluster-monitoring-view", ns); + + client.apply(&clusterrolebinding, Some(ns)).await?; + + debug!("creating grafana datasource crd"); + + let token_str = format!("Bearer {}", token.token); + + let thanos_url = format!( + "https://{}", + self.get_domain("thanos-querier-openshift-monitoring") + .await + .unwrap() + ); + + let thanos_openshift_datasource = self.build_grafana_datasource( + "thanos-openshift-monitoring", + ns, + &label_selector, + &thanos_url, + token_str.clone(), + ); + + client.apply(&thanos_openshift_datasource, Some(ns)).await?; + + //TODO user workload datasource returns 503 -> need to figure out how to correctly add the + //userworkload thanos-ruler or prometheus-federate to the grafana datasource + //it may alrady be included in the overall monitoring stack + + let user_thanos_url = format!( + "https://{}", + self.get_domain( + "thanos-ruler-openshift-user-workload-monitoring.apps.ncd0.harmony.mcd" + ) + .await + .unwrap() + ); + + let thanos_openshift_userworkload_datasource = self.build_grafana_datasource( + "thanos-openshift-userworkload-monitoring", + ns, + &label_selector, + &user_thanos_url, + token_str.clone(), + ); + + client + .apply(&thanos_openshift_userworkload_datasource, Some(ns)) + .await?; + + debug!("creating grafana dashboard crd"); + let dashboard = self.build_grafana_dashboard(ns, &label_selector); + + client.apply(&dashboard, Some(ns)).await?; + debug!("creating grafana ingress"); let grafana_ingress = self.build_grafana_ingress(ns).await; grafana_ingress @@ -368,31 +423,36 @@ impl K8sAnywhereTopology { pub fn build_cluster_rolebinding( &self, + service_account_name: &str, + clusterrole_name: &str, ns: &str, - account_name: &str, - role: &str, ) -> ClusterRoleBinding { ClusterRoleBinding { metadata: ObjectMeta { - name: Some(format!("{}-view-binding", account_name)), + name: Some(format!("{}-view-binding", service_account_name)), ..Default::default() }, role_ref: RoleRef { api_group: "rbac.authorization.k8s.io".into(), kind: "ClusterRole".into(), - name: role.into(), + name: clusterrole_name.into(), }, subjects: Some(vec![Subject { kind: "ServiceAccount".into(), - name: account_name.into(), + name: service_account_name.into(), namespace: Some(ns.into()), ..Default::default() }]), } } - pub fn get_token_request(&self) -> TokenRequest { + pub fn get_token_request(&self, ns: &str) -> TokenRequest { + debug!("building token request"); TokenRequest { + metadata: ObjectMeta { + namespace: Some(ns.to_string()), + ..Default::default() + }, spec: TokenRequestSpec { audiences: vec!["https://kubernetes.default.svc".to_string()], expiration_seconds: Some(3600), @@ -402,15 +462,39 @@ impl K8sAnywhereTopology { } } - pub fn build_token_secret(&self, token: &str, ns: &str) -> Secret { + pub async fn create_service_account_token( + &self, + service_account_name: &str, + ns: &str, + ) -> Result { + debug!("creating service account token"); + let token_request = self.get_token_request(ns); + let client = self.k8s_client().await?; + let pp = PostParams::default(); + let token_requests_api = client.service_account_api(ns).await; + + let data = serde_json::to_vec(&token_request).unwrap(); + + let created_token_request = token_requests_api + .create_subresource::("token", service_account_name, &pp, data) + .await?; + + let status = created_token_request + .status + .ok_or_else(|| PreparationError::new("missing token request status".to_string()))?; + + Ok(status) + } + + pub async fn build_token_secret(&self, secret_name: &str, token: &str, ns: &str) -> Secret { Secret { metadata: ObjectMeta { - name: Some("grafana-credentials".into()), + name: Some(secret_name.into()), namespace: Some(ns.into()), ..Default::default() }, string_data: Some(std::collections::BTreeMap::from([( - "PROMETHEUS_TOKEN".into(), + secret_name.into(), format!("Bearer {}", token), )])), ..Default::default() @@ -419,39 +503,18 @@ impl K8sAnywhereTopology { fn build_grafana_datasource( &self, + name: &str, ns: &str, label_selector: &LabelSelector, url: &str, + token: String, ) -> GrafanaDatasource { let mut json_data = BTreeMap::new(); json_data.insert("timeInterval".to_string(), "5s".to_string()); - // - // let graf_data_source = GrafanaDatasource { - // metadata: ObjectMeta { - // name: Some(format!("grafana-datasource-{}", ns)), - // namespace: Some(ns.to_string()), - // ..Default::default() - // }, - // spec: GrafanaDatasourceSpec { - // instance_selector: label_selector.clone(), - // allow_cross_namespace_import: Some(false), - // datasource: GrafanaDatasourceConfig { - // access: "proxy".to_string(), - // database: Some("prometheus".to_string()), - // json_data: Some(json_data), - // //this is fragile - // name: format!("prometheus-{}-0", ns), - // r#type: "prometheus".to_string(), - // url: url.to_string(), - // //url: format!("http://prometheus-operated.{}.svc.cluster.local:9090", ns), - // }, - // }, - // }; - // graf_data_source GrafanaDatasource { metadata: ObjectMeta { - name: Some("thanos-prometheus".to_string()), + name: Some(name.to_string()), namespace: Some(ns.to_string()), ..Default::default() }, @@ -460,20 +523,21 @@ impl K8sAnywhereTopology { allow_cross_namespace_import: Some(true), datasource: GrafanaDatasourceConfig { access: "proxy".to_string(), - name: "OpenShift-Thanos".to_string(), + name: name.to_string(), r#type: "prometheus".to_string(), url: url.to_string(), database: None, json_data: Some(GrafanaDatasourceJsonData { time_interval: Some("60s".to_string()), http_header_name1: Some("Authorization".to_string()), + tls_skip_verify: Some(true), + oauth_pass_thru: Some(true), }), secure_json_data: Some(GrafanaDatasourceSecureJsonData { - http_header_value1: Some("Bearer eyJhbGc...".to_string()), + http_header_value1: Some(token), }), is_default: Some(false), editable: Some(true), - version: Some(1), }, }, } diff --git a/harmony/src/modules/monitoring/kube_prometheus/crd/crd_grafana.rs b/harmony/src/modules/monitoring/kube_prometheus/crd/crd_grafana.rs index 4134670..e58f4ca 100644 --- a/harmony/src/modules/monitoring/kube_prometheus/crd/crd_grafana.rs +++ b/harmony/src/modules/monitoring/kube_prometheus/crd/crd_grafana.rs @@ -132,6 +132,7 @@ pub struct GrafanaDatasourceSpec { #[serde(rename_all = "camelCase")] pub struct GrafanaDatasourceConfig { pub access: String, + #[serde(default, skip_serializing_if = "Option::is_none")] pub database: Option, pub name: String, pub r#type: String, @@ -149,9 +150,6 @@ pub struct GrafanaDatasourceConfig { #[serde(default, skip_serializing_if = "Option::is_none")] pub editable: Option, - - #[serde(default, skip_serializing_if = "Option::is_none")] - pub version: Option, } #[derive(Serialize, Deserialize, Debug, Clone, JsonSchema)] @@ -162,6 +160,14 @@ pub struct GrafanaDatasourceJsonData { #[serde(default, skip_serializing_if = "Option::is_none")] pub http_header_name1: Option, + + /// Disable TLS skip verification (false = verify) + #[serde(default, skip_serializing_if = "Option::is_none")] + pub tls_skip_verify: Option, + + /// Auth type - set to "forward" for OpenShift OAuth identity + #[serde(default, skip_serializing_if = "Option::is_none")] + pub oauth_pass_thru: Option, } #[derive(Serialize, Deserialize, Debug, Clone, JsonSchema)] diff --git a/harmony/src/modules/prometheus/k8s_prometheus_alerting_score.rs b/harmony/src/modules/prometheus/k8s_prometheus_alerting_score.rs index 2cb4ffb..f9e8531 100644 --- a/harmony/src/modules/prometheus/k8s_prometheus_alerting_score.rs +++ b/harmony/src/modules/prometheus/k8s_prometheus_alerting_score.rs @@ -12,7 +12,7 @@ use crate::modules::monitoring::kube_prometheus::crd::crd_alertmanager_config::C use crate::modules::monitoring::kube_prometheus::crd::crd_default_rules::build_default_application_rules; use crate::modules::monitoring::kube_prometheus::crd::crd_grafana::{ Grafana, GrafanaDashboard, GrafanaDashboardSpec, GrafanaDatasource, GrafanaDatasourceConfig, - GrafanaDatasourceSpec, GrafanaSpec, + GrafanaDatasourceJsonData, GrafanaDatasourceSpec, GrafanaSpec, }; use crate::modules::monitoring::kube_prometheus::crd::crd_prometheus_rules::{ PrometheusRule, PrometheusRuleSpec, RuleGroup, @@ -466,10 +466,15 @@ impl K8sPrometheusCRDAlertingInterpret { match_labels: label.clone(), match_expressions: vec![], }; - let mut json_data = BTreeMap::new(); - json_data.insert("timeInterval".to_string(), "5s".to_string()); + // let mut json_data = BTreeMap::new(); + // json_data.insert("timeInterval".to_string(), "5s".to_string()); let namespace = self.sender.namespace.clone(); - + let json_data = GrafanaDatasourceJsonData { + time_interval: Some("5s".to_string()), + http_header_name1: None, + tls_skip_verify: Some(true), + oauth_pass_thru: Some(true), + }; let json = build_default_dashboard(&namespace); let graf_data_source = GrafanaDatasource { @@ -495,6 +500,9 @@ impl K8sPrometheusCRDAlertingInterpret { "http://prometheus-operated.{}.svc.cluster.local:9090", self.sender.namespace.clone() ), + secure_json_data: None, + is_default: None, + editable: None, }, }, }; From 7dff70edcf459751b9656184bccc157ebc88ce2a Mon Sep 17 00:00:00 2001 From: Willem Date: Wed, 15 Oct 2025 15:26:36 -0400 Subject: [PATCH 05/11] wip: fixed token expiration and configured grafana dashboard --- harmony/src/domain/topology/k8s_anywhere.rs | 108 ++++++++++-------- .../kube_prometheus/crd/crd_grafana.rs | 51 ++++++++- .../k8s_prometheus_alerting_score.rs | 8 +- 3 files changed, 116 insertions(+), 51 deletions(-) diff --git a/harmony/src/domain/topology/k8s_anywhere.rs b/harmony/src/domain/topology/k8s_anywhere.rs index cb37ece..cb4ab2d 100644 --- a/harmony/src/domain/topology/k8s_anywhere.rs +++ b/harmony/src/domain/topology/k8s_anywhere.rs @@ -6,7 +6,9 @@ use std::{ use async_trait::async_trait; use k8s_openapi::api::{ - authentication::v1::{TokenRequest, TokenRequestSpec, TokenRequestStatus}, + authentication::v1::{ + BoundObjectReference, TokenRequest, TokenRequestSpec, TokenRequestStatus, + }, core::v1::{Secret, ServiceAccount}, rbac::v1::{ClusterRoleBinding, RoleRef, Subject}, }; @@ -30,9 +32,11 @@ use crate::{ kube_prometheus::crd::{ crd_alertmanager_config::CRDPrometheus, crd_grafana::{ - Grafana as GrafanaCRD, GrafanaDashboard, GrafanaDashboardSpec, - GrafanaDatasource, GrafanaDatasourceConfig, GrafanaDatasourceJsonData, - GrafanaDatasourceSecureJsonData, GrafanaDatasourceSpec, GrafanaSpec, + Grafana as GrafanaCRD, GrafanaCom, GrafanaDashboard, + GrafanaDashboardDatasource, GrafanaDashboardSpec, GrafanaDatasource, + GrafanaDatasourceConfig, GrafanaDatasourceJsonData, + GrafanaDatasourceSecureJsonData, GrafanaDatasourceSpec, GrafanaSecretKeyRef, + GrafanaSpec, GrafanaValueFrom, GrafanaValueSource, }, crd_prometheuses::LabelSelector, grafana_default_dashboard::build_default_dashboard, @@ -166,22 +170,24 @@ impl Grafana for K8sAnywhereTopology { .wait_until_deployment_ready( "grafana-grafana-deployment".to_string(), Some("grafana"), - Some(15), + Some(30), ) .await?; let sa_name = "grafana-grafana-sa"; - debug!("creating token for sevice account {sa_name}"); - let token = self.create_service_account_token(sa_name, ns).await?; + let token_secret_name = "grafana-sa-token-secret"; - debug!("creating secret"); - let secret_name = "grafana-sa-secret"; - let secret = self.build_token_secret(secret_name, &token.token, ns).await; + // let sa_token_secret = self.build_sa_token_secret(token_secret_name, sa_name, ns); + // + // client.apply(&sa_token_secret, Some(ns)).await?; + let secret = self.build_token_secret(token_secret_name, ns).await; client.apply(&secret, Some(ns)).await?; + let token_request_status = self.create_service_account_token(sa_name, ns).await?; debug!("creating grafana clusterrole binding"); + let clusterrolebinding = self.build_cluster_rolebinding(sa_name, "cluster-monitoring-view", ns); @@ -189,7 +195,7 @@ impl Grafana for K8sAnywhereTopology { debug!("creating grafana datasource crd"); - let token_str = format!("Bearer {}", token.token); + // let token_str = format!("Bearer {}", token.token); let thanos_url = format!( "https://{}", @@ -203,36 +209,11 @@ impl Grafana for K8sAnywhereTopology { ns, &label_selector, &thanos_url, - token_str.clone(), + &token_request_status.token, // Pass the secret name here ); client.apply(&thanos_openshift_datasource, Some(ns)).await?; - //TODO user workload datasource returns 503 -> need to figure out how to correctly add the - //userworkload thanos-ruler or prometheus-federate to the grafana datasource - //it may alrady be included in the overall monitoring stack - - let user_thanos_url = format!( - "https://{}", - self.get_domain( - "thanos-ruler-openshift-user-workload-monitoring.apps.ncd0.harmony.mcd" - ) - .await - .unwrap() - ); - - let thanos_openshift_userworkload_datasource = self.build_grafana_datasource( - "thanos-openshift-userworkload-monitoring", - ns, - &label_selector, - &user_thanos_url, - token_str.clone(), - ); - - client - .apply(&thanos_openshift_userworkload_datasource, Some(ns)) - .await?; - debug!("creating grafana dashboard crd"); let dashboard = self.build_grafana_dashboard(ns, &label_selector); @@ -446,6 +427,30 @@ impl K8sAnywhereTopology { } } + pub fn build_sa_token_secret( + &self, + secret_name: &str, + service_account_name: &str, + ns: &str, + ) -> Secret { + let mut annotations = BTreeMap::new(); + annotations.insert( + "kubernetes.io/service-account.name".to_string(), + service_account_name.to_string(), + ); + + Secret { + metadata: ObjectMeta { + name: Some(secret_name.into()), + namespace: Some(ns.into()), + annotations: Some(annotations), + ..Default::default() + }, + type_: Some("kubernetes.io/service-account-token".to_string()), + ..Default::default() + } + } + pub fn get_token_request(&self, ns: &str) -> TokenRequest { debug!("building token request"); TokenRequest { @@ -456,7 +461,11 @@ impl K8sAnywhereTopology { spec: TokenRequestSpec { audiences: vec!["https://kubernetes.default.svc".to_string()], expiration_seconds: Some(3600), - ..Default::default() + bound_object_ref: Some(BoundObjectReference { + kind: Some("Secret".to_string()), + name: Some("grafana-sa-token-secret".to_string()), + ..Default::default() + }), }, ..Default::default() } @@ -486,17 +495,14 @@ impl K8sAnywhereTopology { Ok(status) } - pub async fn build_token_secret(&self, secret_name: &str, token: &str, ns: &str) -> Secret { + pub async fn build_token_secret(&self, secret_name: &str, ns: &str) -> Secret { Secret { metadata: ObjectMeta { name: Some(secret_name.into()), namespace: Some(ns.into()), ..Default::default() }, - string_data: Some(std::collections::BTreeMap::from([( - secret_name.into(), - format!("Bearer {}", token), - )])), + string_data: None, ..Default::default() } } @@ -507,7 +513,7 @@ impl K8sAnywhereTopology { ns: &str, label_selector: &LabelSelector, url: &str, - token: String, + token: &str, // Pass in the secret name ) -> GrafanaDatasource { let mut json_data = BTreeMap::new(); json_data.insert("timeInterval".to_string(), "5s".to_string()); @@ -521,6 +527,7 @@ impl K8sAnywhereTopology { spec: GrafanaDatasourceSpec { instance_selector: label_selector.clone(), allow_cross_namespace_import: Some(true), + values_from: None, datasource: GrafanaDatasourceConfig { access: "proxy".to_string(), name: name.to_string(), @@ -534,7 +541,7 @@ impl K8sAnywhereTopology { oauth_pass_thru: Some(true), }), secure_json_data: Some(GrafanaDatasourceSecureJsonData { - http_header_value1: Some(token), + http_header_value1: Some(format!("Bearer {token}")), }), is_default: Some(false), editable: Some(true), @@ -548,7 +555,6 @@ impl K8sAnywhereTopology { ns: &str, label_selector: &LabelSelector, ) -> GrafanaDashboard { - let json = build_default_dashboard(ns); let graf_dashboard = GrafanaDashboard { metadata: ObjectMeta { name: Some(format!("grafana-dashboard-{}", ns)), @@ -558,7 +564,15 @@ impl K8sAnywhereTopology { spec: GrafanaDashboardSpec { resync_period: Some("30s".to_string()), instance_selector: label_selector.clone(), - json, + datasources: Some(vec![GrafanaDashboardDatasource { + input_name: "DS_PROMETHEUS".to_string(), + datasource_name: "thanos-openshift-monitoring".to_string(), + }]), + json: None, + grafana_com: Some(GrafanaCom { + id: 17406, + revision: None, + }), }, }; graf_dashboard diff --git a/harmony/src/modules/monitoring/kube_prometheus/crd/crd_grafana.rs b/harmony/src/modules/monitoring/kube_prometheus/crd/crd_grafana.rs index e58f4ca..c99adc1 100644 --- a/harmony/src/modules/monitoring/kube_prometheus/crd/crd_grafana.rs +++ b/harmony/src/modules/monitoring/kube_prometheus/crd/crd_grafana.rs @@ -103,9 +103,34 @@ pub struct GrafanaDashboardSpec { #[serde(default, skip_serializing_if = "Option::is_none")] pub resync_period: Option, + #[serde(default, skip_serializing_if = "Option::is_none")] + pub datasources: Option>, + pub instance_selector: LabelSelector, - pub json: String, + #[serde(default, skip_serializing_if = "Option::is_none")] + pub json: Option, + + #[serde(default, skip_serializing_if = "Option::is_none")] + pub grafana_com: Option, +} + +#[derive(Serialize, Deserialize, Debug, Clone, JsonSchema)] +#[serde(rename_all = "camelCase")] +pub struct GrafanaDashboardDatasource { + pub input_name: String, + pub datasource_name: String, +} + +// ------------------------------------------------------------------------------------------------ + +#[derive(Serialize, Deserialize, Debug, Clone, JsonSchema)] +#[serde(rename_all = "camelCase")] +pub struct GrafanaCom { + pub id: u32, + + #[serde(default, skip_serializing_if = "Option::is_none")] + pub revision: Option, } // ------------------------------------------------------------------------------------------------ @@ -126,6 +151,30 @@ pub struct GrafanaDatasourceSpec { pub allow_cross_namespace_import: Option, pub datasource: GrafanaDatasourceConfig, + + #[serde(default, skip_serializing_if = "Option::is_none")] + pub values_from: Option>, +} + + +#[derive(Serialize, Deserialize, Debug, Clone, JsonSchema)] +#[serde(rename_all = "camelCase")] +pub struct GrafanaValueFrom { + pub target_path: String, + pub value_from: GrafanaValueSource, +} + +#[derive(Serialize, Deserialize, Debug, Clone, JsonSchema)] +#[serde(rename_all = "camelCase")] +pub struct GrafanaValueSource { + pub secret_key_ref: GrafanaSecretKeyRef, +} + +#[derive(Serialize, Deserialize, Debug, Clone, JsonSchema)] +#[serde(rename_all = "camelCase")] +pub struct GrafanaSecretKeyRef { + pub name: String, + pub key: String, } #[derive(Serialize, Deserialize, Debug, Clone, JsonSchema)] diff --git a/harmony/src/modules/prometheus/k8s_prometheus_alerting_score.rs b/harmony/src/modules/prometheus/k8s_prometheus_alerting_score.rs index f9e8531..7873235 100644 --- a/harmony/src/modules/prometheus/k8s_prometheus_alerting_score.rs +++ b/harmony/src/modules/prometheus/k8s_prometheus_alerting_score.rs @@ -11,8 +11,7 @@ use std::process::Command; use crate::modules::monitoring::kube_prometheus::crd::crd_alertmanager_config::CRDPrometheus; use crate::modules::monitoring::kube_prometheus::crd::crd_default_rules::build_default_application_rules; use crate::modules::monitoring::kube_prometheus::crd::crd_grafana::{ - Grafana, GrafanaDashboard, GrafanaDashboardSpec, GrafanaDatasource, GrafanaDatasourceConfig, - GrafanaDatasourceJsonData, GrafanaDatasourceSpec, GrafanaSpec, + Grafana, GrafanaDashboard, GrafanaDashboardSpec, GrafanaDatasource, GrafanaDatasourceConfig, GrafanaDatasourceJsonData, GrafanaDatasourceSpec, GrafanaSecretKeyRef, GrafanaSpec, GrafanaValueFrom, GrafanaValueSource }; use crate::modules::monitoring::kube_prometheus::crd::crd_prometheus_rules::{ PrometheusRule, PrometheusRuleSpec, RuleGroup, @@ -504,6 +503,7 @@ impl K8sPrometheusCRDAlertingInterpret { is_default: None, editable: None, }, + values_from: None, }, }; @@ -524,7 +524,9 @@ impl K8sPrometheusCRDAlertingInterpret { spec: GrafanaDashboardSpec { resync_period: Some("30s".to_string()), instance_selector: labels.clone(), - json, + json: Some(json), + grafana_com: None, + datasources: None, }, }; From fc384599a1a31d2678cc11c1f9b067b146afae46 Mon Sep 17 00:00:00 2001 From: Willem Date: Thu, 16 Oct 2025 14:07:23 -0400 Subject: [PATCH 06/11] feat: implementation of Installable for CRDPrometheusIntroduction of Grafana trait and its impl for k8sanywhereallows for CRDPrometheus to be installed via AlertingInterpret which standardizes the installation of alert receivers, alerting rules, and alert senders --- harmony/src/domain/topology/k8s_anywhere.rs | 131 ++++++------------ .../kube_prometheus/crd/crd_grafana.rs | 3 +- .../k8s_prometheus_alerting_score.rs | 4 +- 3 files changed, 45 insertions(+), 93 deletions(-) diff --git a/harmony/src/domain/topology/k8s_anywhere.rs b/harmony/src/domain/topology/k8s_anywhere.rs index cb4ab2d..e45b65f 100644 --- a/harmony/src/domain/topology/k8s_anywhere.rs +++ b/harmony/src/domain/topology/k8s_anywhere.rs @@ -1,21 +1,12 @@ -use std::{ - collections::{BTreeMap, HashMap}, - process::Command, - sync::Arc, -}; +use std::{collections::BTreeMap, process::Command, sync::Arc}; use async_trait::async_trait; +use base64::{Engine, engine::general_purpose}; use k8s_openapi::api::{ - authentication::v1::{ - BoundObjectReference, TokenRequest, TokenRequestSpec, TokenRequestStatus, - }, - core::v1::{Secret, ServiceAccount}, + core::v1::Secret, rbac::v1::{ClusterRoleBinding, RoleRef, Subject}, }; -use kube::{ - Api, - api::{GroupVersionKind, ObjectMeta, PostParams}, -}; +use kube::api::{DynamicObject, GroupVersionKind, ObjectMeta}; use log::{debug, info, warn}; use serde::Serialize; use tokio::sync::OnceCell; @@ -35,14 +26,11 @@ use crate::{ Grafana as GrafanaCRD, GrafanaCom, GrafanaDashboard, GrafanaDashboardDatasource, GrafanaDashboardSpec, GrafanaDatasource, GrafanaDatasourceConfig, GrafanaDatasourceJsonData, - GrafanaDatasourceSecureJsonData, GrafanaDatasourceSpec, GrafanaSecretKeyRef, - GrafanaSpec, GrafanaValueFrom, GrafanaValueSource, + GrafanaDatasourceSecureJsonData, GrafanaDatasourceSpec, GrafanaSpec, }, crd_prometheuses::LabelSelector, - grafana_default_dashboard::build_default_dashboard, prometheus_operator::prometheus_operator_helm_chart_score, rhob_alertmanager_config::RHOBObservability, - role::build_prom_service_account, service_monitor::ServiceMonitor, }, }, @@ -148,24 +136,23 @@ impl Grafana for K8sAnywhereTopology { }; } async fn install_grafana(&self) -> Result { - debug!("install grafana"); let ns = "grafana"; let mut label = BTreeMap::new(); label.insert("dashboards".to_string(), "grafana".to_string()); + let label_selector = LabelSelector { match_labels: label.clone(), match_expressions: vec![], }; - debug!("getting client"); + let client = self.k8s_client().await?; - info!("creating grafanas crd"); let grafana = self.build_grafana(ns, &label); client.apply(&grafana, Some(ns)).await?; - + //TODO change this to a ensure ready or something better than just a timeout client .wait_until_deployment_ready( "grafana-grafana-deployment".to_string(), @@ -175,16 +162,25 @@ impl Grafana for K8sAnywhereTopology { .await?; let sa_name = "grafana-grafana-sa"; - let token_secret_name = "grafana-sa-token-secret"; - // let sa_token_secret = self.build_sa_token_secret(token_secret_name, sa_name, ns); - // - // client.apply(&sa_token_secret, Some(ns)).await?; + let sa_token_secret = self.build_sa_token_secret(token_secret_name, sa_name, ns); - let secret = self.build_token_secret(token_secret_name, ns).await; - client.apply(&secret, Some(ns)).await?; - let token_request_status = self.create_service_account_token(sa_name, ns).await?; + client.apply(&sa_token_secret, Some(ns)).await?; + let secret_gvk = GroupVersionKind { + group: "".to_string(), + version: "v1".to_string(), + kind: "Secret".to_string(), + }; + + let secret = client + .get_resource_json_value(token_secret_name, Some(ns), &secret_gvk) + .await?; + + let token = format!( + "Bearer {}", + self.extract_and_normalize_token(&secret).unwrap() + ); debug!("creating grafana clusterrole binding"); @@ -195,8 +191,6 @@ impl Grafana for K8sAnywhereTopology { debug!("creating grafana datasource crd"); - // let token_str = format!("Bearer {}", token.token); - let thanos_url = format!( "https://{}", self.get_domain("thanos-querier-openshift-monitoring") @@ -209,7 +203,7 @@ impl Grafana for K8sAnywhereTopology { ns, &label_selector, &thanos_url, - &token_request_status.token, // Pass the secret name here + &token, ); client.apply(&thanos_openshift_datasource, Some(ns)).await?; @@ -398,8 +392,21 @@ impl K8sAnywhereTopology { .clone() } - pub fn build_service_account(&self, name: &str, namespace: &str) -> ServiceAccount { - build_prom_service_account(name.to_string(), namespace.to_string()) + fn extract_and_normalize_token(&self, secret: &DynamicObject) -> Option { + let token_b64 = secret + .data + .get("token") + .or_else(|| secret.data.get("data").and_then(|d| d.get("token"))) + .and_then(|v| v.as_str())?; + + let bytes = general_purpose::STANDARD.decode(token_b64).ok()?; + + let s = String::from_utf8(bytes).ok()?; + + let cleaned = s + .trim_matches(|c: char| c.is_whitespace() || c == '\0') + .to_string(); + Some(cleaned) } pub fn build_cluster_rolebinding( @@ -451,69 +458,13 @@ impl K8sAnywhereTopology { } } - pub fn get_token_request(&self, ns: &str) -> TokenRequest { - debug!("building token request"); - TokenRequest { - metadata: ObjectMeta { - namespace: Some(ns.to_string()), - ..Default::default() - }, - spec: TokenRequestSpec { - audiences: vec!["https://kubernetes.default.svc".to_string()], - expiration_seconds: Some(3600), - bound_object_ref: Some(BoundObjectReference { - kind: Some("Secret".to_string()), - name: Some("grafana-sa-token-secret".to_string()), - ..Default::default() - }), - }, - ..Default::default() - } - } - - pub async fn create_service_account_token( - &self, - service_account_name: &str, - ns: &str, - ) -> Result { - debug!("creating service account token"); - let token_request = self.get_token_request(ns); - let client = self.k8s_client().await?; - let pp = PostParams::default(); - let token_requests_api = client.service_account_api(ns).await; - - let data = serde_json::to_vec(&token_request).unwrap(); - - let created_token_request = token_requests_api - .create_subresource::("token", service_account_name, &pp, data) - .await?; - - let status = created_token_request - .status - .ok_or_else(|| PreparationError::new("missing token request status".to_string()))?; - - Ok(status) - } - - pub async fn build_token_secret(&self, secret_name: &str, ns: &str) -> Secret { - Secret { - metadata: ObjectMeta { - name: Some(secret_name.into()), - namespace: Some(ns.into()), - ..Default::default() - }, - string_data: None, - ..Default::default() - } - } - fn build_grafana_datasource( &self, name: &str, ns: &str, label_selector: &LabelSelector, url: &str, - token: &str, // Pass in the secret name + token: &str, ) -> GrafanaDatasource { let mut json_data = BTreeMap::new(); json_data.insert("timeInterval".to_string(), "5s".to_string()); diff --git a/harmony/src/modules/monitoring/kube_prometheus/crd/crd_grafana.rs b/harmony/src/modules/monitoring/kube_prometheus/crd/crd_grafana.rs index c99adc1..386890e 100644 --- a/harmony/src/modules/monitoring/kube_prometheus/crd/crd_grafana.rs +++ b/harmony/src/modules/monitoring/kube_prometheus/crd/crd_grafana.rs @@ -105,7 +105,7 @@ pub struct GrafanaDashboardSpec { #[serde(default, skip_serializing_if = "Option::is_none")] pub datasources: Option>, - + pub instance_selector: LabelSelector, #[serde(default, skip_serializing_if = "Option::is_none")] @@ -156,7 +156,6 @@ pub struct GrafanaDatasourceSpec { pub values_from: Option>, } - #[derive(Serialize, Deserialize, Debug, Clone, JsonSchema)] #[serde(rename_all = "camelCase")] pub struct GrafanaValueFrom { diff --git a/harmony/src/modules/prometheus/k8s_prometheus_alerting_score.rs b/harmony/src/modules/prometheus/k8s_prometheus_alerting_score.rs index 7873235..d7dca5e 100644 --- a/harmony/src/modules/prometheus/k8s_prometheus_alerting_score.rs +++ b/harmony/src/modules/prometheus/k8s_prometheus_alerting_score.rs @@ -11,7 +11,9 @@ use std::process::Command; use crate::modules::monitoring::kube_prometheus::crd::crd_alertmanager_config::CRDPrometheus; use crate::modules::monitoring::kube_prometheus::crd::crd_default_rules::build_default_application_rules; use crate::modules::monitoring::kube_prometheus::crd::crd_grafana::{ - Grafana, GrafanaDashboard, GrafanaDashboardSpec, GrafanaDatasource, GrafanaDatasourceConfig, GrafanaDatasourceJsonData, GrafanaDatasourceSpec, GrafanaSecretKeyRef, GrafanaSpec, GrafanaValueFrom, GrafanaValueSource + Grafana, GrafanaDashboard, GrafanaDashboardSpec, GrafanaDatasource, GrafanaDatasourceConfig, + GrafanaDatasourceJsonData, GrafanaDatasourceSpec, GrafanaSecretKeyRef, GrafanaSpec, + GrafanaValueFrom, GrafanaValueSource, }; use crate::modules::monitoring::kube_prometheus::crd::crd_prometheus_rules::{ PrometheusRule, PrometheusRuleSpec, RuleGroup, From a815f6ac9c06a1a65ba9f6e101e6c12248fedd14 Mon Sep 17 00:00:00 2001 From: Willem Date: Mon, 20 Oct 2025 11:44:11 -0400 Subject: [PATCH 07/11] feat: scrape targets to be able to get snmp alerts from machines to prometheus --- .../topology/oberservability/monitoring.rs | 9 +- .../kube_prometheus/crd/crd_scrape_config.rs | 187 ++++++++++++++++++ .../monitoring/kube_prometheus/crd/mod.rs | 1 + .../helm_prometheus_alert_score.rs | 1 + harmony/src/modules/monitoring/mod.rs | 1 + .../modules/monitoring/scrape_target/mod.rs | 1 + .../monitoring/scrape_target/server.rs | 76 +++++++ 7 files changed, 274 insertions(+), 2 deletions(-) create mode 100644 harmony/src/modules/monitoring/kube_prometheus/crd/crd_scrape_config.rs create mode 100644 harmony/src/modules/monitoring/scrape_target/mod.rs create mode 100644 harmony/src/modules/monitoring/scrape_target/server.rs diff --git a/harmony/src/domain/topology/oberservability/monitoring.rs b/harmony/src/domain/topology/oberservability/monitoring.rs index 1489e83..d9fe4d0 100644 --- a/harmony/src/domain/topology/oberservability/monitoring.rs +++ b/harmony/src/domain/topology/oberservability/monitoring.rs @@ -21,6 +21,7 @@ pub struct AlertingInterpret { pub sender: S, pub receivers: Vec>>, pub rules: Vec>>, + pub scrape_target: Vec>>, } #[async_trait] @@ -38,6 +39,10 @@ impl, T: Topology> Interpret for AlertingInte debug!("installing rule: {:#?}", rule); rule.install(&self.sender).await?; } + for target in self.scrape_target.iter() { + debug!("installing scrape_target: {:#?}", target); + target.install(&self.sender).await?; + } self.sender.ensure_installed(inventory, topology).await?; Ok(Outcome::success(format!( "successfully installed alert sender {}", @@ -77,6 +82,6 @@ pub trait AlertRule: std::fmt::Debug + Send + Sync { } #[async_trait] -pub trait ScrapeTarget { - async fn install(&self, sender: &S) -> Result<(), InterpretError>; +pub trait ScrapeTarget: std::fmt::Debug + Send + Sync { + async fn install(&self, sender: &S) -> Result; } diff --git a/harmony/src/modules/monitoring/kube_prometheus/crd/crd_scrape_config.rs b/harmony/src/modules/monitoring/kube_prometheus/crd/crd_scrape_config.rs new file mode 100644 index 0000000..24a2833 --- /dev/null +++ b/harmony/src/modules/monitoring/kube_prometheus/crd/crd_scrape_config.rs @@ -0,0 +1,187 @@ +use std::net::IpAddr; + +use async_trait::async_trait; +use kube::CustomResource; +use schemars::JsonSchema; +use serde::{Deserialize, Serialize}; + +use crate::{ + modules::monitoring::kube_prometheus::crd::{ + crd_alertmanager_config::CRDPrometheus, crd_prometheuses::LabelSelector, + }, + topology::oberservability::monitoring::ScrapeTarget, +}; + +#[derive(CustomResource, Serialize, Deserialize, Debug, Clone, JsonSchema)] +#[kube( + group = "monitoring.coreos.com", + version = "v1alpha1", + kind = "ScrapeConfig", + plural = "scrapeconfigs", + namespaced +)] +#[serde(rename_all = "camelCase")] +pub struct ScrapeConfigSpec { + /// List of static configurations. + pub static_configs: Option>, + + /// Kubernetes service discovery. + pub kubernetes_sd_configs: Option>, + + /// HTTP-based service discovery. + pub http_sd_configs: Option>, + + /// File-based service discovery. + pub file_sd_configs: Option>, + + /// DNS-based service discovery. + pub dns_sd_configs: Option>, + + /// Consul service discovery. + pub consul_sd_configs: Option>, + + /// Relabeling configuration applied to discovered targets. + pub relabel_configs: Option>, + + /// Metric relabeling configuration applied to scraped samples. + pub metric_relabel_configs: Option>, + + /// Path to scrape metrics from (defaults to `/metrics`). + pub metrics_path: Option, + + /// Interval at which Prometheus scrapes targets (e.g., "30s"). + pub scrape_interval: Option, + + /// Timeout for scraping (e.g., "10s"). + pub scrape_timeout: Option, + + /// Optional job name override. + pub job_name: Option, + + /// Optional scheme (http or https). + pub scheme: Option, + + /// Authorization paramaters for snmp walk + pub params: Option, +} + +/// Static configuration section of a ScrapeConfig. +#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)] +#[serde(rename_all = "camelCase")] +pub struct StaticConfig { + pub targets: Vec, + + pub labels: Option, +} + +/// Relabeling configuration for target or metric relabeling. +#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)] +#[serde(rename_all = "camelCase")] +pub struct RelabelConfig { + pub source_labels: Option>, + pub separator: Option, + pub target_label: Option, + pub regex: Option, + pub modulus: Option, + pub replacement: Option, + pub action: Option, +} + +/// Kubernetes service discovery configuration. +#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)] +#[serde(rename_all = "camelCase")] +pub struct KubernetesSDConfig { + ///"pod", "service", "endpoints"pub role: String, + pub namespaces: Option, + pub selectors: Option>, + pub api_server: Option, + pub bearer_token_file: Option, + pub tls_config: Option, +} + +/// Namespace selector for Kubernetes service discovery. +#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)] +#[serde(rename_all = "camelCase")] +pub struct NamespaceSelector { + pub any: Option, + pub match_names: Option>, +} + +/// HTTP-based service discovery configuration. +#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)] +#[serde(rename_all = "camelCase")] +pub struct HttpSDConfig { + pub url: String, + pub refresh_interval: Option, + pub basic_auth: Option, + pub authorization: Option, + pub tls_config: Option, +} + +/// File-based service discovery configuration. +#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)] +#[serde(rename_all = "camelCase")] +pub struct FileSDConfig { + pub files: Vec, + pub refresh_interval: Option, +} + +/// DNS-based service discovery configuration. +#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)] +#[serde(rename_all = "camelCase")] +pub struct DnsSDConfig { + pub names: Vec, + pub refresh_interval: Option, + pub type_: Option, // SRV, A, AAAA + pub port: Option, +} + +/// Consul service discovery configuration. +#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)] +#[serde(rename_all = "camelCase")] +pub struct ConsulSDConfig { + pub server: String, + pub services: Option>, + pub scheme: Option, + pub datacenter: Option, + pub tag_separator: Option, + pub refresh_interval: Option, + pub tls_config: Option, +} + +/// Basic authentication credentials. +#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)] +#[serde(rename_all = "camelCase")] +pub struct BasicAuth { + pub username: String, + pub password: Option, + pub password_file: Option, +} + +/// Bearer token or other auth mechanisms. +#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)] +#[serde(rename_all = "camelCase")] +pub struct Authorization { + pub credentials: Option, + pub credentials_file: Option, + pub type_: Option, +} + +/// TLS configuration for secure scraping. +#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)] +#[serde(rename_all = "camelCase")] +pub struct TLSConfig { + pub ca_file: Option, + pub cert_file: Option, + pub key_file: Option, + pub server_name: Option, + pub insecure_skip_verify: Option, +} + +/// Authorization parameters for SNMP walk. +#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)] +#[serde(rename_all = "camelCase")] +pub struct Params { + pub auth: Option>, + pub module: Option>, +} diff --git a/harmony/src/modules/monitoring/kube_prometheus/crd/mod.rs b/harmony/src/modules/monitoring/kube_prometheus/crd/mod.rs index 4dbea74..c8cb854 100644 --- a/harmony/src/modules/monitoring/kube_prometheus/crd/mod.rs +++ b/harmony/src/modules/monitoring/kube_prometheus/crd/mod.rs @@ -4,6 +4,7 @@ pub mod crd_default_rules; pub mod crd_grafana; pub mod crd_prometheus_rules; pub mod crd_prometheuses; +pub mod crd_scrape_config; pub mod grafana_default_dashboard; pub mod grafana_operator; pub mod prometheus_operator; diff --git a/harmony/src/modules/monitoring/kube_prometheus/helm_prometheus_alert_score.rs b/harmony/src/modules/monitoring/kube_prometheus/helm_prometheus_alert_score.rs index c9a0c04..da26b03 100644 --- a/harmony/src/modules/monitoring/kube_prometheus/helm_prometheus_alert_score.rs +++ b/harmony/src/modules/monitoring/kube_prometheus/helm_prometheus_alert_score.rs @@ -31,6 +31,7 @@ impl Score for HelmPrometheusAlert sender: KubePrometheus { config }, receivers: self.receivers.clone(), rules: self.rules.clone(), + scrape_target: vec![], }) } fn name(&self) -> String { diff --git a/harmony/src/modules/monitoring/mod.rs b/harmony/src/modules/monitoring/mod.rs index edda516..7f07d5a 100644 --- a/harmony/src/modules/monitoring/mod.rs +++ b/harmony/src/modules/monitoring/mod.rs @@ -6,3 +6,4 @@ pub mod kube_prometheus; pub mod ntfy; pub mod okd; pub mod prometheus; +pub mod scrape_target; diff --git a/harmony/src/modules/monitoring/scrape_target/mod.rs b/harmony/src/modules/monitoring/scrape_target/mod.rs new file mode 100644 index 0000000..74f47ad --- /dev/null +++ b/harmony/src/modules/monitoring/scrape_target/mod.rs @@ -0,0 +1 @@ +pub mod server; diff --git a/harmony/src/modules/monitoring/scrape_target/server.rs b/harmony/src/modules/monitoring/scrape_target/server.rs new file mode 100644 index 0000000..ba41f49 --- /dev/null +++ b/harmony/src/modules/monitoring/scrape_target/server.rs @@ -0,0 +1,76 @@ +use std::net::IpAddr; + +use async_trait::async_trait; +use kube::api::ObjectMeta; +use serde::Serialize; + +use crate::{ + interpret::{InterpretError, Outcome}, + modules::monitoring::kube_prometheus::crd::{ + crd_alertmanager_config::CRDPrometheus, + crd_scrape_config::{Params, RelabelConfig, ScrapeConfig, ScrapeConfigSpec, StaticConfig}, + }, + topology::oberservability::monitoring::ScrapeTarget, +}; + +#[derive(Debug, Clone, Serialize)] +pub struct Server { + pub name: String, + pub ip: IpAddr, + pub auth: String, + pub module: String, + pub domain: String, +} + +#[async_trait] +impl ScrapeTarget for Server { + async fn install(&self, sender: &CRDPrometheus) -> Result { + let scrape_config_spec = ScrapeConfigSpec { + static_configs: Some(vec![StaticConfig { + targets: vec![self.ip.to_string()], + labels: None, + }]), + scrape_interval: Some("2m".to_string()), + kubernetes_sd_configs: None, + http_sd_configs: None, + file_sd_configs: None, + dns_sd_configs: None, + params: Some(Params { + auth: Some(vec![self.auth.clone()]), + module: Some(vec![self.module.clone()]), + }), + consul_sd_configs: None, + relabel_configs: Some(vec![RelabelConfig { + action: None, + source_labels: Some(vec!["__address__".to_string()]), + separator: None, + target_label: Some("__param_target".to_string()), + regex: None, + replacement: Some(format!("snmp.{}:31080", self.domain.clone())), + modulus: None, + }]), + metric_relabel_configs: None, + metrics_path: Some("/snmp".to_string()), + scrape_timeout: Some("2m".to_string()), + job_name: Some(format!("snmp_exporter/cloud/{}", self.name.clone())), + scheme: None, + }; + + let scrape_config = ScrapeConfig { + metadata: ObjectMeta { + name: Some(self.name.clone()), + namespace: Some(sender.namespace.clone()), + ..Default::default() + }, + spec: scrape_config_spec, + }; + sender + .client + .apply(&scrape_config, Some(&sender.namespace.clone())) + .await?; + Ok(Outcome::success(format!( + "installed scrape target {}", + self.name.clone() + ))) + } +} From cb66b7592ec52ae14f95284669e4aaa159a8dabd Mon Sep 17 00:00:00 2001 From: Willem Date: Mon, 20 Oct 2025 14:44:37 -0400 Subject: [PATCH 08/11] fix: made targets plural and changed scrape targets to option in AlertingInterpret --- .../src/domain/topology/oberservability/monitoring.rs | 10 ++++++---- .../kube_prometheus/helm_prometheus_alert_score.rs | 2 +- 2 files changed, 7 insertions(+), 5 deletions(-) diff --git a/harmony/src/domain/topology/oberservability/monitoring.rs b/harmony/src/domain/topology/oberservability/monitoring.rs index d9fe4d0..8a1368f 100644 --- a/harmony/src/domain/topology/oberservability/monitoring.rs +++ b/harmony/src/domain/topology/oberservability/monitoring.rs @@ -21,7 +21,7 @@ pub struct AlertingInterpret { pub sender: S, pub receivers: Vec>>, pub rules: Vec>>, - pub scrape_target: Vec>>, + pub scrape_targets: Option>>>, } #[async_trait] @@ -39,9 +39,11 @@ impl, T: Topology> Interpret for AlertingInte debug!("installing rule: {:#?}", rule); rule.install(&self.sender).await?; } - for target in self.scrape_target.iter() { - debug!("installing scrape_target: {:#?}", target); - target.install(&self.sender).await?; + if let Some(targets) = &self.scrape_targets { + for target in targets.iter() { + debug!("installing scrape_target: {:#?}", target); + target.install(&self.sender).await?; + } } self.sender.ensure_installed(inventory, topology).await?; Ok(Outcome::success(format!( diff --git a/harmony/src/modules/monitoring/kube_prometheus/helm_prometheus_alert_score.rs b/harmony/src/modules/monitoring/kube_prometheus/helm_prometheus_alert_score.rs index da26b03..468d308 100644 --- a/harmony/src/modules/monitoring/kube_prometheus/helm_prometheus_alert_score.rs +++ b/harmony/src/modules/monitoring/kube_prometheus/helm_prometheus_alert_score.rs @@ -31,7 +31,7 @@ impl Score for HelmPrometheusAlert sender: KubePrometheus { config }, receivers: self.receivers.clone(), rules: self.rules.clone(), - scrape_target: vec![], + scrape_targets: None, }) } fn name(&self) -> String { From ce91ee01685c48b81e46440e171440c92723577b Mon Sep 17 00:00:00 2001 From: Willem Date: Mon, 20 Oct 2025 15:31:06 -0400 Subject: [PATCH 09/11] fix: removed dead code, mapped error from grafana operator to preparation error rather than ignoring it, modified k8sprometheus score to unwrap_or_default() service monitors --- harmony/src/domain/topology/k8s_anywhere.rs | 24 +++++++------------ .../monitoring/grafana/helm/helm_grafana.rs | 3 ++- .../k8s_prometheus_alerting_score.rs | 2 -- 3 files changed, 10 insertions(+), 19 deletions(-) diff --git a/harmony/src/domain/topology/k8s_anywhere.rs b/harmony/src/domain/topology/k8s_anywhere.rs index e45b65f..cf56333 100644 --- a/harmony/src/domain/topology/k8s_anywhere.rs +++ b/harmony/src/domain/topology/k8s_anywhere.rs @@ -585,21 +585,12 @@ impl K8sAnywhereTopology { receivers: Option>>>, service_monitors: Option>, ) -> K8sPrometheusCRDAlertingScore { - if let Some(sm) = service_monitors { - return K8sPrometheusCRDAlertingScore { - sender, - receivers: receivers.unwrap_or_default(), - service_monitors: sm, - prometheus_rules: vec![], - }; - } else { - return K8sPrometheusCRDAlertingScore { - sender, - receivers: receivers.unwrap_or_default(), - service_monitors: vec![], - prometheus_rules: vec![], - }; - } + return K8sPrometheusCRDAlertingScore { + sender, + receivers: receivers.unwrap_or_default(), + service_monitors: service_monitors.unwrap_or_default(), + prometheus_rules: vec![], + }; } async fn openshift_ingress_operator_available(&self) -> Result<(), PreparationError> { @@ -882,7 +873,8 @@ impl K8sAnywhereTopology { } let _grafana_operator_score = grafana_helm_chart_score(namespace, namespace_scope) .interpret(inventory, self) - .await; + .await + .map_err(|e| PreparationError::new(e.to_string())); Ok(PreparationOutcome::Success { details: format!( "Successfully installed grafana operator in ns {}", diff --git a/harmony/src/modules/monitoring/grafana/helm/helm_grafana.rs b/harmony/src/modules/monitoring/grafana/helm/helm_grafana.rs index 2965ada..c9ccacb 100644 --- a/harmony/src/modules/monitoring/grafana/helm/helm_grafana.rs +++ b/harmony/src/modules/monitoring/grafana/helm/helm_grafana.rs @@ -1,3 +1,4 @@ +use harmony_macros::hurl; use non_blank_string_rs::NonBlankString; use std::{collections::HashMap, str::FromStr}; @@ -20,7 +21,7 @@ pub fn grafana_helm_chart_score(ns: &str, namespace_scope: bool) -> HelmChartSco install_only: true, repository: Some(HelmRepository::new( "grafana".to_string(), - url::Url::parse("https://grafana.github.io/helm-charts").unwrap(), + hurl!("https://grafana.github.io/helm-charts"), true, )), } diff --git a/harmony/src/modules/prometheus/k8s_prometheus_alerting_score.rs b/harmony/src/modules/prometheus/k8s_prometheus_alerting_score.rs index d7dca5e..7093ee8 100644 --- a/harmony/src/modules/prometheus/k8s_prometheus_alerting_score.rs +++ b/harmony/src/modules/prometheus/k8s_prometheus_alerting_score.rs @@ -467,8 +467,6 @@ impl K8sPrometheusCRDAlertingInterpret { match_labels: label.clone(), match_expressions: vec![], }; - // let mut json_data = BTreeMap::new(); - // json_data.insert("timeInterval".to_string(), "5s".to_string()); let namespace = self.sender.namespace.clone(); let json_data = GrafanaDatasourceJsonData { time_interval: Some("5s".to_string()), From 73681849175c439dbeb7bd044e3896434ccb7e47 Mon Sep 17 00:00:00 2001 From: Ian Letourneau Date: Wed, 22 Oct 2025 15:12:53 -0400 Subject: [PATCH 10/11] fix(ha_cluster): inject switch client for better testability --- Cargo.lock | 27 +++++----- brocade/src/fast_iron.rs | 1 + brocade/src/lib.rs | 2 +- brocade/src/network_operating_system.rs | 1 + brocade/src/shell.rs | 1 + examples/nanodc/Cargo.toml | 2 + examples/nanodc/src/main.rs | 34 +++++++++++-- examples/okd_installation/Cargo.toml | 1 + examples/okd_installation/src/topology.rs | 31 +++++++++++- examples/okd_pxe/Cargo.toml | 1 + examples/okd_pxe/src/topology.rs | 34 +++++++++++-- examples/opnsense/Cargo.toml | 3 ++ examples/opnsense/src/main.rs | 33 ++++++++++++- harmony/src/domain/topology/ha_cluster.rs | 60 ++++++++++------------- harmony/src/domain/topology/network.rs | 32 +++++++----- harmony/src/infra/brocade.rs | 11 +---- 16 files changed, 197 insertions(+), 77 deletions(-) diff --git a/Cargo.lock b/Cargo.lock index 666fe3a..7d9cdcf 100644 --- a/Cargo.lock +++ b/Cargo.lock @@ -1780,6 +1780,7 @@ dependencies = [ name = "example-nanodc" version = "0.1.0" dependencies = [ + "brocade", "cidr", "env_logger", "harmony", @@ -1788,6 +1789,7 @@ dependencies = [ "harmony_tui", "harmony_types", "log", + "serde", "tokio", "url", ] @@ -1806,6 +1808,7 @@ dependencies = [ name = "example-okd-install" version = "0.1.0" dependencies = [ + "brocade", "cidr", "env_logger", "harmony", @@ -1836,13 +1839,16 @@ dependencies = [ name = "example-opnsense" version = "0.1.0" dependencies = [ + "brocade", "cidr", "env_logger", "harmony", "harmony_macros", + "harmony_secret", "harmony_tui", "harmony_types", "log", + "serde", "tokio", "url", ] @@ -1851,6 +1857,7 @@ dependencies = [ name = "example-pxe" version = "0.1.0" dependencies = [ + "brocade", "cidr", "env_logger", "harmony", @@ -1865,6 +1872,15 @@ dependencies = [ "url", ] +[[package]] +name = "example-remove-rook-osd" +version = "0.1.0" +dependencies = [ + "harmony", + "harmony_cli", + "tokio", +] + [[package]] name = "example-rust" version = "0.1.0" @@ -1918,8 +1934,6 @@ dependencies = [ "env_logger", "harmony", "harmony_macros", - "harmony_secret", - "harmony_secret_derive", "harmony_tui", "harmony_types", "log", @@ -4613,15 +4627,6 @@ version = "0.8.6" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "caf4aa5b0f434c91fe5c7f1ecb6a5ece2130b02ad2a590589dda5146df959001" -[[package]] -name = "remove_rook_osd" -version = "0.1.0" -dependencies = [ - "harmony", - "harmony_cli", - "tokio", -] - [[package]] name = "reqwest" version = "0.11.27" diff --git a/brocade/src/fast_iron.rs b/brocade/src/fast_iron.rs index a1a2478..5a3474e 100644 --- a/brocade/src/fast_iron.rs +++ b/brocade/src/fast_iron.rs @@ -10,6 +10,7 @@ use log::{debug, info}; use regex::Regex; use std::{collections::HashSet, str::FromStr}; +#[derive(Debug)] pub struct FastIronClient { shell: BrocadeShell, version: BrocadeInfo, diff --git a/brocade/src/lib.rs b/brocade/src/lib.rs index 3822abd..57b464a 100644 --- a/brocade/src/lib.rs +++ b/brocade/src/lib.rs @@ -162,7 +162,7 @@ pub async fn init( } #[async_trait] -pub trait BrocadeClient { +pub trait BrocadeClient: std::fmt::Debug { /// Retrieves the operating system and version details from the connected Brocade switch. /// /// This is typically the first call made after establishing a connection to determine diff --git a/brocade/src/network_operating_system.rs b/brocade/src/network_operating_system.rs index b14bc08..0ee4a88 100644 --- a/brocade/src/network_operating_system.rs +++ b/brocade/src/network_operating_system.rs @@ -10,6 +10,7 @@ use crate::{ parse_brocade_mac_address, shell::BrocadeShell, }; +#[derive(Debug)] pub struct NetworkOperatingSystemClient { shell: BrocadeShell, version: BrocadeInfo, diff --git a/brocade/src/shell.rs b/brocade/src/shell.rs index cfa672d..28eceb8 100644 --- a/brocade/src/shell.rs +++ b/brocade/src/shell.rs @@ -13,6 +13,7 @@ use log::info; use russh::ChannelMsg; use tokio::time::timeout; +#[derive(Debug)] pub struct BrocadeShell { ip: IpAddr, port: u16, diff --git a/examples/nanodc/Cargo.toml b/examples/nanodc/Cargo.toml index 889c24d..3072ddf 100644 --- a/examples/nanodc/Cargo.toml +++ b/examples/nanodc/Cargo.toml @@ -17,3 +17,5 @@ harmony_secret = { path = "../../harmony_secret" } log = { workspace = true } env_logger = { workspace = true } url = { workspace = true } +serde = { workspace = true } +brocade = { path = "../../brocade" } diff --git a/examples/nanodc/src/main.rs b/examples/nanodc/src/main.rs index 57574d2..d00503f 100644 --- a/examples/nanodc/src/main.rs +++ b/examples/nanodc/src/main.rs @@ -3,12 +3,13 @@ use std::{ sync::Arc, }; +use brocade::BrocadeOptions; use cidr::Ipv4Cidr; use harmony::{ config::secret::SshKeyPair, data::{FileContent, FilePath}, hardware::{HostCategory, Location, PhysicalHost, SwitchGroup}, - infra::opnsense::OPNSenseManagementInterface, + infra::{brocade::BrocadeSwitchClient, opnsense::OPNSenseManagementInterface}, inventory::Inventory, modules::{ http::StaticFilesHttpScore, @@ -22,8 +23,9 @@ use harmony::{ topology::{LogicalHost, UnmanagedRouter}, }; use harmony_macros::{ip, mac_address}; -use harmony_secret::SecretManager; +use harmony_secret::{Secret, SecretManager}; use harmony_types::net::Url; +use serde::{Deserialize, Serialize}; #[tokio::main] async fn main() { @@ -32,6 +34,26 @@ async fn main() { name: String::from("fw0"), }; + let switch_auth = SecretManager::get_or_prompt::() + .await + .expect("Failed to get credentials"); + + let switches: Vec = vec![ip!("192.168.33.101")]; + let brocade_options = Some(BrocadeOptions { + dry_run: *harmony::config::DRY_RUN, + ..Default::default() + }); + let switch_client = BrocadeSwitchClient::init( + &switches, + &switch_auth.username, + &switch_auth.password, + brocade_options, + ) + .await + .expect("Failed to connect to switch"); + + let switch_client = Arc::new(switch_client); + let opnsense = Arc::new( harmony::infra::opnsense::OPNSenseFirewall::new(firewall, None, "root", "opnsense").await, ); @@ -83,7 +105,7 @@ async fn main() { name: "wk2".to_string(), }, ], - switch: vec![], + switch_client: switch_client.clone(), }; let inventory = Inventory { @@ -166,3 +188,9 @@ async fn main() { .await .unwrap(); } + +#[derive(Secret, Serialize, Deserialize, Debug)] +pub struct BrocadeSwitchAuth { + pub username: String, + pub password: String, +} diff --git a/examples/okd_installation/Cargo.toml b/examples/okd_installation/Cargo.toml index 7314e4f..dfbe944 100644 --- a/examples/okd_installation/Cargo.toml +++ b/examples/okd_installation/Cargo.toml @@ -19,3 +19,4 @@ log = { workspace = true } env_logger = { workspace = true } url = { workspace = true } serde.workspace = true +brocade = { path = "../../brocade" } diff --git a/examples/okd_installation/src/topology.rs b/examples/okd_installation/src/topology.rs index 31062f5..617a3a8 100644 --- a/examples/okd_installation/src/topology.rs +++ b/examples/okd_installation/src/topology.rs @@ -1,7 +1,8 @@ +use brocade::BrocadeOptions; use cidr::Ipv4Cidr; use harmony::{ hardware::{Location, SwitchGroup}, - infra::opnsense::OPNSenseManagementInterface, + infra::{brocade::BrocadeSwitchClient, opnsense::OPNSenseManagementInterface}, inventory::Inventory, topology::{HAClusterTopology, LogicalHost, UnmanagedRouter}, }; @@ -22,6 +23,26 @@ pub async fn get_topology() -> HAClusterTopology { name: String::from("opnsense-1"), }; + let switch_auth = SecretManager::get_or_prompt::() + .await + .expect("Failed to get credentials"); + + let switches: Vec = vec![ip!("192.168.1.101")]; // TODO: Adjust me + let brocade_options = Some(BrocadeOptions { + dry_run: *harmony::config::DRY_RUN, + ..Default::default() + }); + let switch_client = BrocadeSwitchClient::init( + &switches, + &switch_auth.username, + &switch_auth.password, + brocade_options, + ) + .await + .expect("Failed to connect to switch"); + + let switch_client = Arc::new(switch_client); + let config = SecretManager::get_or_prompt::().await; let config = config.unwrap(); @@ -58,7 +79,7 @@ pub async fn get_topology() -> HAClusterTopology { name: "bootstrap".to_string(), }, workers: vec![], - switch: vec![], + switch_client: switch_client.clone(), } } @@ -75,3 +96,9 @@ pub fn get_inventory() -> Inventory { control_plane_host: vec![], } } + +#[derive(Secret, Serialize, Deserialize, Debug)] +pub struct BrocadeSwitchAuth { + pub username: String, + pub password: String, +} diff --git a/examples/okd_pxe/Cargo.toml b/examples/okd_pxe/Cargo.toml index f75f42b..133b2f9 100644 --- a/examples/okd_pxe/Cargo.toml +++ b/examples/okd_pxe/Cargo.toml @@ -19,3 +19,4 @@ log = { workspace = true } env_logger = { workspace = true } url = { workspace = true } serde.workspace = true +brocade = { path = "../../brocade" } diff --git a/examples/okd_pxe/src/topology.rs b/examples/okd_pxe/src/topology.rs index 707969a..0cf4b72 100644 --- a/examples/okd_pxe/src/topology.rs +++ b/examples/okd_pxe/src/topology.rs @@ -1,13 +1,15 @@ +use brocade::BrocadeOptions; use cidr::Ipv4Cidr; use harmony::{ config::secret::OPNSenseFirewallCredentials, hardware::{Location, SwitchGroup}, - infra::opnsense::OPNSenseManagementInterface, + infra::{brocade::BrocadeSwitchClient, opnsense::OPNSenseManagementInterface}, inventory::Inventory, topology::{HAClusterTopology, LogicalHost, UnmanagedRouter}, }; use harmony_macros::{ip, ipv4}; -use harmony_secret::SecretManager; +use harmony_secret::{Secret, SecretManager}; +use serde::{Deserialize, Serialize}; use std::{net::IpAddr, sync::Arc}; pub async fn get_topology() -> HAClusterTopology { @@ -16,6 +18,26 @@ pub async fn get_topology() -> HAClusterTopology { name: String::from("opnsense-1"), }; + let switch_auth = SecretManager::get_or_prompt::() + .await + .expect("Failed to get credentials"); + + let switches: Vec = vec![ip!("192.168.1.101")]; // TODO: Adjust me + let brocade_options = Some(BrocadeOptions { + dry_run: *harmony::config::DRY_RUN, + ..Default::default() + }); + let switch_client = BrocadeSwitchClient::init( + &switches, + &switch_auth.username, + &switch_auth.password, + brocade_options, + ) + .await + .expect("Failed to connect to switch"); + + let switch_client = Arc::new(switch_client); + let config = SecretManager::get_or_prompt::().await; let config = config.unwrap(); @@ -52,7 +74,7 @@ pub async fn get_topology() -> HAClusterTopology { name: "cp0".to_string(), }, workers: vec![], - switch: vec![], + switch_client: switch_client.clone(), } } @@ -69,3 +91,9 @@ pub fn get_inventory() -> Inventory { control_plane_host: vec![], } } + +#[derive(Secret, Serialize, Deserialize, Debug)] +pub struct BrocadeSwitchAuth { + pub username: String, + pub password: String, +} diff --git a/examples/opnsense/Cargo.toml b/examples/opnsense/Cargo.toml index 60986d3..1574f29 100644 --- a/examples/opnsense/Cargo.toml +++ b/examples/opnsense/Cargo.toml @@ -16,3 +16,6 @@ harmony_macros = { path = "../../harmony_macros" } log = { workspace = true } env_logger = { workspace = true } url = { workspace = true } +harmony_secret = { path = "../../harmony_secret" } +brocade = { path = "../../brocade" } +serde = { workspace = true } diff --git a/examples/opnsense/src/main.rs b/examples/opnsense/src/main.rs index fcfaf09..d03643b 100644 --- a/examples/opnsense/src/main.rs +++ b/examples/opnsense/src/main.rs @@ -3,10 +3,11 @@ use std::{ sync::Arc, }; +use brocade::BrocadeOptions; use cidr::Ipv4Cidr; use harmony::{ hardware::{HostCategory, Location, PhysicalHost, SwitchGroup}, - infra::opnsense::OPNSenseManagementInterface, + infra::{brocade::BrocadeSwitchClient, opnsense::OPNSenseManagementInterface}, inventory::Inventory, modules::{ dummy::{ErrorScore, PanicScore, SuccessScore}, @@ -18,7 +19,9 @@ use harmony::{ topology::{LogicalHost, UnmanagedRouter}, }; use harmony_macros::{ip, mac_address}; +use harmony_secret::{Secret, SecretManager}; use harmony_types::net::Url; +use serde::{Deserialize, Serialize}; #[tokio::main] async fn main() { @@ -27,6 +30,26 @@ async fn main() { name: String::from("opnsense-1"), }; + let switch_auth = SecretManager::get_or_prompt::() + .await + .expect("Failed to get credentials"); + + let switches: Vec = vec![ip!("192.168.5.101")]; // TODO: Adjust me + let brocade_options = Some(BrocadeOptions { + dry_run: *harmony::config::DRY_RUN, + ..Default::default() + }); + let switch_client = BrocadeSwitchClient::init( + &switches, + &switch_auth.username, + &switch_auth.password, + brocade_options, + ) + .await + .expect("Failed to connect to switch"); + + let switch_client = Arc::new(switch_client); + let opnsense = Arc::new( harmony::infra::opnsense::OPNSenseFirewall::new(firewall, None, "root", "opnsense").await, ); @@ -54,7 +77,7 @@ async fn main() { name: "cp0".to_string(), }, workers: vec![], - switch: vec![], + switch_client: switch_client.clone(), }; let inventory = Inventory { @@ -109,3 +132,9 @@ async fn main() { .await .unwrap(); } + +#[derive(Secret, Serialize, Deserialize, Debug)] +pub struct BrocadeSwitchAuth { + pub username: String, + pub password: String, +} diff --git a/harmony/src/domain/topology/ha_cluster.rs b/harmony/src/domain/topology/ha_cluster.rs index 7be2725..59787a1 100644 --- a/harmony/src/domain/topology/ha_cluster.rs +++ b/harmony/src/domain/topology/ha_cluster.rs @@ -1,7 +1,5 @@ use async_trait::async_trait; -use brocade::BrocadeOptions; use harmony_macros::ip; -use harmony_secret::SecretManager; use harmony_types::{ net::{MacAddress, Url}, switch::PortLocation, @@ -14,8 +12,6 @@ use log::info; use crate::data::FileContent; use crate::executors::ExecutorError; use crate::hardware::PhysicalHost; -use crate::infra::brocade::BrocadeSwitchAuth; -use crate::infra::brocade::BrocadeSwitchClient; use crate::modules::okd::crd::{ InstallPlanApproval, OperatorGroup, OperatorGroupSpec, Subscription, SubscriptionSpec, nmstate::{self, NMState, NodeNetworkConfigurationPolicy, NodeNetworkConfigurationPolicySpec}, @@ -30,7 +26,6 @@ use super::{ }; use std::collections::BTreeMap; -use std::net::IpAddr; use std::sync::Arc; #[derive(Debug, Clone)] @@ -43,10 +38,10 @@ pub struct HAClusterTopology { pub tftp_server: Arc, pub http_server: Arc, pub dns_server: Arc, + pub switch_client: Arc, pub bootstrap_host: LogicalHost, pub control_plane: Vec, pub workers: Vec, - pub switch: Vec, } #[async_trait] @@ -280,36 +275,15 @@ impl HAClusterTopology { } } - async fn get_switch_client(&self) -> Result, SwitchError> { - let auth = SecretManager::get_or_prompt::() - .await - .map_err(|e| SwitchError::new(format!("Failed to get credentials: {e}")))?; - - // FIXME: We assume Brocade switches - let switches: Vec = self.switch.iter().map(|s| s.ip).collect(); - let brocade_options = Some(BrocadeOptions { - dry_run: *crate::config::DRY_RUN, - ..Default::default() - }); - let client = - BrocadeSwitchClient::init(&switches, &auth.username, &auth.password, brocade_options) - .await - .map_err(|e| SwitchError::new(format!("Failed to connect to switch: {e}")))?; - - Ok(Box::new(client)) - } - async fn configure_port_channel( &self, host: &PhysicalHost, config: &HostNetworkConfig, ) -> Result<(), SwitchError> { debug!("Configuring port channel: {config:#?}"); - let client = self.get_switch_client().await?; - let switch_ports = config.switch_ports.iter().map(|s| s.port.clone()).collect(); - client + self.switch_client .configure_port_channel(&format!("Harmony_{}", host.id), switch_ports) .await .map_err(|e| SwitchError::new(format!("Failed to configure switch: {e}")))?; @@ -333,10 +307,10 @@ impl HAClusterTopology { tftp_server: dummy_infra.clone(), http_server: dummy_infra.clone(), dns_server: dummy_infra.clone(), + switch_client: dummy_infra.clone(), bootstrap_host: dummy_host, control_plane: vec![], workers: vec![], - switch: vec![], } } } @@ -494,8 +468,7 @@ impl HttpServer for HAClusterTopology { #[async_trait] impl Switch for HAClusterTopology { async fn setup_switch(&self) -> Result<(), SwitchError> { - let client = self.get_switch_client().await?; - client.setup().await?; + self.switch_client.setup().await?; Ok(()) } @@ -503,8 +476,7 @@ impl Switch for HAClusterTopology { &self, mac_address: &MacAddress, ) -> Result, SwitchError> { - let client = self.get_switch_client().await?; - let port = client.find_port(mac_address).await?; + let port = self.switch_client.find_port(mac_address).await?; Ok(port) } @@ -704,3 +676,25 @@ impl DnsServer for DummyInfra { unimplemented!("{}", UNIMPLEMENTED_DUMMY_INFRA) } } + +#[async_trait] +impl SwitchClient for DummyInfra { + async fn setup(&self) -> Result<(), SwitchError> { + unimplemented!("{}", UNIMPLEMENTED_DUMMY_INFRA) + } + + async fn find_port( + &self, + _mac_address: &MacAddress, + ) -> Result, SwitchError> { + unimplemented!("{}", UNIMPLEMENTED_DUMMY_INFRA) + } + + async fn configure_port_channel( + &self, + _channel_name: &str, + _switch_ports: Vec, + ) -> Result { + unimplemented!("{}", UNIMPLEMENTED_DUMMY_INFRA) + } +} diff --git a/harmony/src/domain/topology/network.rs b/harmony/src/domain/topology/network.rs index 99db03a..b78f1a0 100644 --- a/harmony/src/domain/topology/network.rs +++ b/harmony/src/domain/topology/network.rs @@ -1,4 +1,10 @@ -use std::{error::Error, net::Ipv4Addr, str::FromStr, sync::Arc}; +use std::{ + error::Error, + fmt::{self, Debug}, + net::Ipv4Addr, + str::FromStr, + sync::Arc, +}; use async_trait::async_trait; use derive_new::new; @@ -19,8 +25,8 @@ pub struct DHCPStaticEntry { pub ip: Ipv4Addr, } -impl std::fmt::Display for DHCPStaticEntry { - fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result { +impl fmt::Display for DHCPStaticEntry { + fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { let mac = self .mac .iter() @@ -42,8 +48,8 @@ pub trait Firewall: Send + Sync { fn get_host(&self) -> LogicalHost; } -impl std::fmt::Debug for dyn Firewall { - fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result { +impl Debug for dyn Firewall { + fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { f.write_fmt(format_args!("Firewall {}", self.get_ip())) } } @@ -65,7 +71,7 @@ pub struct PxeOptions { } #[async_trait] -pub trait DhcpServer: Send + Sync + std::fmt::Debug { +pub trait DhcpServer: Send + Sync + Debug { async fn add_static_mapping(&self, entry: &DHCPStaticEntry) -> Result<(), ExecutorError>; async fn remove_static_mapping(&self, mac: &MacAddress) -> Result<(), ExecutorError>; async fn list_static_mappings(&self) -> Vec<(MacAddress, IpAddress)>; @@ -104,8 +110,8 @@ pub trait DnsServer: Send + Sync { } } -impl std::fmt::Debug for dyn DnsServer { - fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result { +impl Debug for dyn DnsServer { + fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { f.write_fmt(format_args!("DnsServer {}", self.get_ip())) } } @@ -141,8 +147,8 @@ pub enum DnsRecordType { TXT, } -impl std::fmt::Display for DnsRecordType { - fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result { +impl fmt::Display for DnsRecordType { + fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { match self { DnsRecordType::A => write!(f, "A"), DnsRecordType::AAAA => write!(f, "AAAA"), @@ -216,8 +222,8 @@ pub struct SwitchError { msg: String, } -impl std::fmt::Display for SwitchError { - fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result { +impl fmt::Display for SwitchError { + fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { f.write_str(&self.msg) } } @@ -225,7 +231,7 @@ impl std::fmt::Display for SwitchError { impl Error for SwitchError {} #[async_trait] -pub trait SwitchClient: Send + Sync { +pub trait SwitchClient: Debug + Send + Sync { /// Executes essential, idempotent, one-time initial configuration steps. /// /// This is an opiniated procedure that setups a switch to provide high availability diff --git a/harmony/src/infra/brocade.rs b/harmony/src/infra/brocade.rs index f721328..774c8f8 100644 --- a/harmony/src/infra/brocade.rs +++ b/harmony/src/infra/brocade.rs @@ -1,15 +1,14 @@ use async_trait::async_trait; use brocade::{BrocadeClient, BrocadeOptions, InterSwitchLink, InterfaceStatus, PortOperatingMode}; -use harmony_secret::Secret; use harmony_types::{ net::{IpAddress, MacAddress}, switch::{PortDeclaration, PortLocation}, }; use option_ext::OptionExt; -use serde::{Deserialize, Serialize}; use crate::topology::{SwitchClient, SwitchError}; +#[derive(Debug)] pub struct BrocadeSwitchClient { brocade: Box, } @@ -114,12 +113,6 @@ impl SwitchClient for BrocadeSwitchClient { } } -#[derive(Secret, Serialize, Deserialize, Debug)] -pub struct BrocadeSwitchAuth { - pub username: String, - pub password: String, -} - #[cfg(test)] mod tests { use std::sync::{Arc, Mutex}; @@ -235,7 +228,7 @@ mod tests { assert_that!(*configured_interfaces).is_empty(); } - #[derive(Clone)] + #[derive(Debug, Clone)] struct FakeBrocadeClient { stack_topology: Vec, interfaces: Vec, From 609d7acb5d888a3c949159fd94d889eb94f9a0eb Mon Sep 17 00:00:00 2001 From: Jean-Gabriel Gill-Couture Date: Fri, 24 Oct 2025 11:39:27 -0400 Subject: [PATCH 11/11] feat: impl clone_box for ScrapeTarget --- harmony/src/domain/topology/oberservability/monitoring.rs | 1 + .../application_monitoring_score.rs | 3 ++- .../kube_prometheus/crd/crd_alertmanager_config.rs | 8 +++++++- harmony/src/modules/monitoring/scrape_target/server.rs | 4 ++++ 4 files changed, 14 insertions(+), 2 deletions(-) diff --git a/harmony/src/domain/topology/oberservability/monitoring.rs b/harmony/src/domain/topology/oberservability/monitoring.rs index 951d061..6d7411c 100644 --- a/harmony/src/domain/topology/oberservability/monitoring.rs +++ b/harmony/src/domain/topology/oberservability/monitoring.rs @@ -87,4 +87,5 @@ pub trait AlertRule: std::fmt::Debug + Send + Sync { #[async_trait] pub trait ScrapeTarget: std::fmt::Debug + Send + Sync { async fn install(&self, sender: &S) -> Result; + fn clone_box(&self) -> Box>; } diff --git a/harmony/src/modules/monitoring/application_monitoring/application_monitoring_score.rs b/harmony/src/modules/monitoring/application_monitoring/application_monitoring_score.rs index 0f6e0ec..8f6b624 100644 --- a/harmony/src/modules/monitoring/application_monitoring/application_monitoring_score.rs +++ b/harmony/src/modules/monitoring/application_monitoring/application_monitoring_score.rs @@ -15,7 +15,7 @@ use crate::{ score::Score, topology::{ K8sclient, Topology, - oberservability::monitoring::{AlertReceiver, AlertingInterpret}, + oberservability::monitoring::{AlertReceiver, AlertingInterpret, ScrapeTarget}, }, }; @@ -35,6 +35,7 @@ impl + K8sclient + Grafana> Sc sender: self.sender.clone(), receivers: self.receivers.clone(), rules: vec![], + scrape_targets: None, }) } diff --git a/harmony/src/modules/monitoring/kube_prometheus/crd/crd_alertmanager_config.rs b/harmony/src/modules/monitoring/kube_prometheus/crd/crd_alertmanager_config.rs index ceeca41..88ec745 100644 --- a/harmony/src/modules/monitoring/kube_prometheus/crd/crd_alertmanager_config.rs +++ b/harmony/src/modules/monitoring/kube_prometheus/crd/crd_alertmanager_config.rs @@ -18,7 +18,7 @@ use crate::{ K8sclient, Topology, installable::Installable, k8s::K8sClient, - oberservability::monitoring::{AlertReceiver, AlertSender}, + oberservability::monitoring::{AlertReceiver, AlertSender, ScrapeTarget}, }, }; @@ -54,6 +54,12 @@ impl Clone for Box> { } } +impl Clone for Box> { + fn clone(&self) -> Self { + self.clone_box() + } +} + impl Serialize for Box> { fn serialize(&self, _serializer: S) -> Result where diff --git a/harmony/src/modules/monitoring/scrape_target/server.rs b/harmony/src/modules/monitoring/scrape_target/server.rs index ba41f49..178e914 100644 --- a/harmony/src/modules/monitoring/scrape_target/server.rs +++ b/harmony/src/modules/monitoring/scrape_target/server.rs @@ -73,4 +73,8 @@ impl ScrapeTarget for Server { self.name.clone() ))) } + + fn clone_box(&self) -> Box> { + Box::new(self.clone()) + } }