Compare commits

...

17 Commits

Author SHA1 Message Date
b2825ec1ef Merge pull request 'feat/impl_installable_crd_prometheus' (#170) from feat/impl_installable_crd_prometheus into master
Some checks failed
Run Check Script / check (push) Successful in 1m25s
Compile and package harmony_composer / package_harmony_composer (push) Failing after 2m42s
Reviewed-on: https://git.nationtech.io/NationTech/harmony/pulls/170
2025-10-24 16:42:54 +00:00
609d7acb5d feat: impl clone_box for ScrapeTarget<CRDPrometheus>
All checks were successful
Run Check Script / check (pull_request) Successful in 1m25s
2025-10-24 12:05:54 -04:00
de761cf538 Merge branch 'master' into feat/impl_installable_crd_prometheus 2025-10-24 11:23:56 -04:00
c069207f12 Merge pull request 'refactor(ha_cluster): inject switch client for better testability' (#174) from switch-client into master
Some checks failed
Run Check Script / check (push) Successful in 1m44s
Compile and package harmony_composer / package_harmony_composer (push) Failing after 2m43s
Reviewed-on: https://git.nationtech.io/NationTech/harmony/pulls/174
2025-10-23 15:05:17 +00:00
Ian Letourneau
7368184917 fix(ha_cluster): inject switch client for better testability
All checks were successful
Run Check Script / check (pull_request) Successful in 1m30s
2025-10-22 15:12:53 -04:00
05205f4ac1 Merge pull request 'feat: scrape targets to be able to get snmp alerts from machines to prometheus' (#171) from feat/scrape_target into master
Some checks are pending
Run Check Script / check (push) Waiting to run
Compile and package harmony_composer / package_harmony_composer (push) Waiting to run
Reviewed-on: https://git.nationtech.io/NationTech/harmony/pulls/171
2025-10-22 15:33:24 +00:00
3174645c97 Merge branch 'master' into feat/scrape_target
All checks were successful
Run Check Script / check (pull_request) Successful in 1m32s
2025-10-22 15:33:01 +00:00
ce91ee0168 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
Some checks failed
Run Check Script / check (pull_request) Has been cancelled
2025-10-20 15:31:06 -04:00
cb66b7592e fix: made targets plural and changed scrape targets to option in AlertingInterpret
Some checks failed
Run Check Script / check (pull_request) Has been cancelled
2025-10-20 14:44:37 -04:00
a815f6ac9c feat: scrape targets to be able to get snmp alerts from machines to prometheus
Some checks failed
Run Check Script / check (pull_request) Has been cancelled
2025-10-20 11:44:11 -04:00
c0d54a4466 Merge remote-tracking branch 'origin/master' into feat/impl_installable_crd_prometheus
Some checks failed
Run Check Script / check (pull_request) Has been cancelled
2025-10-16 14:17:32 -04:00
fc384599a1 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 2025-10-16 14:07:23 -04:00
7dff70edcf wip: fixed token expiration and configured grafana dashboard 2025-10-15 15:26:36 -04:00
06a0c44c3c wip: connected the thanos-datasource to grafana, need to complete connecting the openshift-userworkload-monitoring as well 2025-10-14 15:53:42 -04:00
85bec66e58 wip: fixing grafana datasource for openshift which requires creating a token, sa, secret and inserting them into the grafanadatasource 2025-10-10 12:09:26 -04:00
1f3796f503 refactor(prometheus): modified crd prometheus to impl the installable trait 2025-10-09 12:26:05 -04:00
58b6268989 wip: moving the install steps for grafana and prometheus into the trait installable<T> 2025-09-29 10:46:29 -04:00
39 changed files with 1117 additions and 224 deletions

27
Cargo.lock generated
View File

@ -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"

View File

@ -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,

View File

@ -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

View File

@ -10,6 +10,7 @@ use crate::{
parse_brocade_mac_address, shell::BrocadeShell,
};
#[derive(Debug)]
pub struct NetworkOperatingSystemClient {
shell: BrocadeShell,
version: BrocadeInfo,

View File

@ -13,6 +13,7 @@ use log::info;
use russh::ChannelMsg;
use tokio::time::timeout;
#[derive(Debug)]
pub struct BrocadeShell {
ip: IpAddr,
port: u16,

View File

@ -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" }

View File

@ -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::<BrocadeSwitchAuth>()
.await
.expect("Failed to get credentials");
let switches: Vec<IpAddr> = 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,
}

View File

@ -19,3 +19,4 @@ log = { workspace = true }
env_logger = { workspace = true }
url = { workspace = true }
serde.workspace = true
brocade = { path = "../../brocade" }

View File

@ -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::<BrocadeSwitchAuth>()
.await
.expect("Failed to get credentials");
let switches: Vec<IpAddr> = 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::<OPNSenseFirewallConfig>().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,
}

View File

@ -19,3 +19,4 @@ log = { workspace = true }
env_logger = { workspace = true }
url = { workspace = true }
serde.workspace = true
brocade = { path = "../../brocade" }

View File

@ -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::<BrocadeSwitchAuth>()
.await
.expect("Failed to get credentials");
let switches: Vec<IpAddr> = 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::<OPNSenseFirewallCredentials>().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,
}

View File

@ -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 }

View File

@ -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::<BrocadeSwitchAuth>()
.await
.expect("Failed to get credentials");
let switches: Vec<IpAddr> = 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,
}

View File

@ -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,
},

View File

@ -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<dyn TftpServer>,
pub http_server: Arc<dyn HttpServer>,
pub dns_server: Arc<dyn DnsServer>,
pub switch_client: Arc<dyn SwitchClient>,
pub bootstrap_host: LogicalHost,
pub control_plane: Vec<LogicalHost>,
pub workers: Vec<LogicalHost>,
pub switch: Vec<LogicalHost>,
}
#[async_trait]
@ -280,36 +275,15 @@ impl HAClusterTopology {
}
}
async fn get_switch_client(&self) -> Result<Box<dyn SwitchClient>, SwitchError> {
let auth = SecretManager::get_or_prompt::<BrocadeSwitchAuth>()
.await
.map_err(|e| SwitchError::new(format!("Failed to get credentials: {e}")))?;
// FIXME: We assume Brocade switches
let switches: Vec<IpAddr> = 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<Option<PortLocation>, 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<Option<PortLocation>, SwitchError> {
unimplemented!("{}", UNIMPLEMENTED_DUMMY_INFRA)
}
async fn configure_port_channel(
&self,
_channel_name: &str,
_switch_ports: Vec<PortLocation>,
) -> Result<u8, SwitchError> {
unimplemented!("{}", UNIMPLEMENTED_DUMMY_INFRA)
}
}

View File

@ -3,7 +3,10 @@ use std::time::Duration;
use derive_new::new;
use k8s_openapi::{
ClusterResourceScope, NamespaceResourceScope,
api::{apps::v1::Deployment, core::v1::Pod},
api::{
apps::v1::Deployment,
core::v1::{Pod, ServiceAccount},
},
apimachinery::pkg::version::Info,
};
use kube::{
@ -21,7 +24,7 @@ use kube::{
};
use log::{debug, error, info, trace};
use serde::{Serialize, de::DeserializeOwned};
use serde_json::{Value, json};
use serde_json::json;
use similar::TextDiff;
use tokio::{io::AsyncReadExt, time::sleep};
@ -57,6 +60,11 @@ impl K8sClient {
})
}
pub async fn service_account_api(&self, namespace: &str) -> Api<ServiceAccount> {
let api: Api<ServiceAccount> = Api::namespaced(self.client.clone(), namespace);
api
}
pub async fn get_apiserver_version(&self) -> Result<Info, Error> {
let client: Client = self.client.clone();
let version_info: Info = client.apiserver_version().await?;

View File

@ -1,7 +1,12 @@
use std::{process::Command, sync::Arc};
use std::{collections::BTreeMap, process::Command, sync::Arc};
use async_trait::async_trait;
use kube::api::GroupVersionKind;
use base64::{Engine, engine::general_purpose};
use k8s_openapi::api::{
core::v1::Secret,
rbac::v1::{ClusterRoleBinding, RoleRef, Subject},
};
use kube::api::{DynamicObject, GroupVersionKind, ObjectMeta};
use log::{debug, info, warn};
use serde::Serialize;
use tokio::sync::OnceCell;
@ -12,14 +17,26 @@ use crate::{
inventory::Inventory,
modules::{
k3d::K3DInstallationScore,
monitoring::kube_prometheus::crd::{
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, GrafanaCom, GrafanaDashboard,
GrafanaDashboardDatasource, GrafanaDashboardSpec, GrafanaDatasource,
GrafanaDatasourceConfig, GrafanaDatasourceJsonData,
GrafanaDatasourceSecureJsonData, GrafanaDatasourceSpec, GrafanaSpec,
},
crd_prometheuses::LabelSelector,
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,
@ -86,41 +103,172 @@ impl K8sclient for K8sAnywhereTopology {
}
#[async_trait]
impl PrometheusApplicationMonitoring<CRDPrometheus> for K8sAnywhereTopology {
impl Grafana for K8sAnywhereTopology {
async fn ensure_grafana_operator(
&self,
inventory: &Inventory,
) -> Result<PreparationOutcome, PreparationError> {
debug!("ensure grafana operator");
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<PreparationOutcome, PreparationError> {
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 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(),
Some("grafana"),
Some(30),
)
.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 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");
let clusterrolebinding =
self.build_cluster_rolebinding(sa_name, "cluster-monitoring-view", ns);
client.apply(&clusterrolebinding, Some(ns)).await?;
debug!("creating grafana datasource crd");
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,
);
client.apply(&thanos_openshift_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
.interpret(&Inventory::empty(), self)
.await
.map_err(|e| PreparationError::new(e.to_string()))?;
Ok(PreparationOutcome::Success {
details: "Installed grafana composants".to_string(),
})
}
}
#[async_trait]
impl PrometheusMonitoring<CRDPrometheus> for K8sAnywhereTopology {
async fn install_prometheus(
&self,
sender: &CRDPrometheus,
inventory: &Inventory,
receivers: Option<Vec<Box<dyn AlertReceiver<CRDPrometheus>>>>,
_inventory: &Inventory,
_receivers: Option<Vec<Box<dyn AlertReceiver<CRDPrometheus>>>>,
) -> Result<PreparationOutcome, PreparationError> {
let client = self.k8s_client().await?;
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,
) -> Result<PreparationOutcome, PreparationError> {
let po_result = self.ensure_prometheus_operator(sender).await?;
if po_result == PreparationOutcome::Noop {
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);
}
let result = self
.get_k8s_prometheus_application_score(sender.clone(), receivers)
.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())),
}
}
}
#[async_trait]
impl PrometheusApplicationMonitoring<RHOBObservability> for K8sAnywhereTopology {
impl PrometheusMonitoring<RHOBObservability> for K8sAnywhereTopology {
async fn install_prometheus(
&self,
sender: &RHOBObservability,
@ -154,6 +302,14 @@ impl PrometheusApplicationMonitoring<RHOBObservability> for K8sAnywhereTopology
Err(err) => Err(PreparationError::new(err.to_string())),
}
}
async fn ensure_prometheus_operator(
&self,
sender: &RHOBObservability,
inventory: &Inventory,
) -> Result<PreparationOutcome, PreparationError> {
todo!()
}
}
impl Serialize for K8sAnywhereTopology {
@ -215,6 +371,180 @@ impl K8sAnywhereTopology {
.await
}
fn extract_and_normalize_token(&self, secret: &DynamicObject) -> Option<String> {
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(
&self,
service_account_name: &str,
clusterrole_name: &str,
ns: &str,
) -> ClusterRoleBinding {
ClusterRoleBinding {
metadata: ObjectMeta {
name: Some(format!("{}-view-binding", service_account_name)),
..Default::default()
},
role_ref: RoleRef {
api_group: "rbac.authorization.k8s.io".into(),
kind: "ClusterRole".into(),
name: clusterrole_name.into(),
},
subjects: Some(vec![Subject {
kind: "ServiceAccount".into(),
name: service_account_name.into(),
namespace: Some(ns.into()),
..Default::default()
}]),
}
}
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()
}
}
fn build_grafana_datasource(
&self,
name: &str,
ns: &str,
label_selector: &LabelSelector,
url: &str,
token: &str,
) -> GrafanaDatasource {
let mut json_data = BTreeMap::new();
json_data.insert("timeInterval".to_string(), "5s".to_string());
GrafanaDatasource {
metadata: ObjectMeta {
name: Some(name.to_string()),
namespace: Some(ns.to_string()),
..Default::default()
},
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(),
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(format!("Bearer {token}")),
}),
is_default: Some(false),
editable: Some(true),
},
},
}
}
fn build_grafana_dashboard(
&self,
ns: &str,
label_selector: &LabelSelector,
) -> GrafanaDashboard {
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(),
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
}
fn build_grafana(&self, ns: &str, labels: &BTreeMap<String, String>) -> 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,
@ -232,13 +562,14 @@ impl K8sAnywhereTopology {
&self,
sender: CRDPrometheus,
receivers: Option<Vec<Box<dyn AlertReceiver<CRDPrometheus>>>>,
service_monitors: Option<Vec<ServiceMonitor>>,
) -> K8sPrometheusCRDAlertingScore {
K8sPrometheusCRDAlertingScore {
return K8sPrometheusCRDAlertingScore {
sender,
receivers: receivers.unwrap_or_default(),
service_monitors: vec![],
service_monitors: service_monitors.unwrap_or_default(),
prometheus_rules: vec![],
}
};
}
async fn openshift_ingress_operator_available(&self) -> Result<(), PreparationError> {
@ -506,6 +837,30 @@ impl K8sAnywhereTopology {
details: "prometheus operator present in cluster".into(),
})
}
async fn install_grafana_operator(
&self,
inventory: &Inventory,
ns: Option<&str>,
) -> Result<PreparationOutcome, PreparationError> {
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
.map_err(|e| PreparationError::new(e.to_string()));
Ok(PreparationOutcome::Success {
details: format!(
"Successfully installed grafana operator in ns {}",
ns.unwrap()
),
})
}
}
#[derive(Clone, Debug)]

View File

@ -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

View File

@ -21,6 +21,7 @@ pub struct AlertingInterpret<S: AlertSender> {
pub sender: S,
pub receivers: Vec<Box<dyn AlertReceiver<S>>>,
pub rules: Vec<Box<dyn AlertRule<S>>>,
pub scrape_targets: Option<Vec<Box<dyn ScrapeTarget<S>>>>,
}
#[async_trait]
@ -30,6 +31,7 @@ impl<S: AlertSender + Installable<T>, T: Topology> Interpret<T> for AlertingInte
inventory: &Inventory,
topology: &T,
) -> Result<Outcome, InterpretError> {
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 +40,12 @@ impl<S: AlertSender + Installable<T>, T: Topology> Interpret<T> for AlertingInte
debug!("installing rule: {:#?}", rule);
rule.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!(
"successfully installed alert sender {}",
@ -77,6 +85,7 @@ pub trait AlertRule<S: AlertSender>: std::fmt::Debug + Send + Sync {
}
#[async_trait]
pub trait ScrapeTarget<S: AlertSender> {
async fn install(&self, sender: &S) -> Result<(), InterpretError>;
pub trait ScrapeTarget<S: AlertSender>: std::fmt::Debug + Send + Sync {
async fn install(&self, sender: &S) -> Result<Outcome, InterpretError>;
fn clone_box(&self) -> Box<dyn ScrapeTarget<S>>;
}

View File

@ -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<dyn BrocadeClient + Send + Sync>,
}
@ -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<InterSwitchLink>,
interfaces: Vec<InterfaceInfo>,

View File

@ -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::{
@ -14,7 +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;
@ -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;
@ -40,7 +45,8 @@ impl<
+ TenantManager
+ K8sclient
+ MultiTargetTopology
+ PrometheusApplicationMonitoring<CRDPrometheus>
+ PrometheusMonitoring<CRDPrometheus>
+ Grafana
+ Ingress
+ std::fmt::Debug,
> ApplicationFeature<T> for Monitoring
@ -57,10 +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![app_service_monitor],
},
application: self.application.clone(),
receivers: self.alert_receiver.clone(),

View File

@ -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<RHOBObservability>,
+ PrometheusMonitoring<RHOBObservability>,
> ApplicationFeature<T> for Monitoring
{
async fn ensure_installed(

View File

@ -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,
prometheus::prometheus::PrometheusApplicationMonitoring,
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, ScrapeTarget},
},
};
use harmony_types::id::Id;
#[derive(Debug, Clone, Serialize)]
pub struct ApplicationMonitoringScore {
@ -24,12 +26,16 @@ pub struct ApplicationMonitoringScore {
pub receivers: Vec<Box<dyn AlertReceiver<CRDPrometheus>>>,
}
impl<T: Topology + PrometheusApplicationMonitoring<CRDPrometheus>> Score<T>
impl<T: Topology + PrometheusMonitoring<CRDPrometheus> + K8sclient + Grafana> Score<T>
for ApplicationMonitoringScore
{
fn create_interpret(&self) -> Box<dyn Interpret<T>> {
Box::new(ApplicationMonitoringInterpret {
score: self.clone(),
debug!("creating alerting interpret");
Box::new(AlertingInterpret {
sender: self.sender.clone(),
receivers: self.receivers.clone(),
rules: vec![],
scrape_targets: None,
})
}
@ -40,55 +46,3 @@ impl<T: Topology + PrometheusApplicationMonitoring<CRDPrometheus>> Score<T>
)
}
}
#[derive(Debug)]
pub struct ApplicationMonitoringInterpret {
score: ApplicationMonitoringScore,
}
#[async_trait]
impl<T: Topology + PrometheusApplicationMonitoring<CRDPrometheus>> Interpret<T>
for ApplicationMonitoringInterpret
{
async fn execute(
&self,
inventory: &Inventory,
topology: &T,
) -> Result<Outcome, InterpretError> {
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<Id> {
todo!()
}
}

View File

@ -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<Box<dyn AlertReceiver<RHOBObservability>>>,
}
impl<T: Topology + PrometheusApplicationMonitoring<RHOBObservability>> Score<T>
impl<T: Topology + PrometheusMonitoring<RHOBObservability>> Score<T>
for ApplicationRHOBMonitoringScore
{
fn create_interpret(&self) -> Box<dyn Interpret<T>> {
@ -49,7 +49,7 @@ pub struct ApplicationRHOBMonitoringInterpret {
}
#[async_trait]
impl<T: Topology + PrometheusApplicationMonitoring<RHOBObservability>> Interpret<T>
impl<T: Topology + PrometheusMonitoring<RHOBObservability>> Interpret<T>
for ApplicationRHOBMonitoringInterpret
{
async fn execute(

View File

@ -0,0 +1,17 @@
use async_trait::async_trait;
use k8s_openapi::Resource;
use crate::{
inventory::Inventory,
topology::{PreparationError, PreparationOutcome},
};
#[async_trait]
pub trait Grafana {
async fn ensure_grafana_operator(
&self,
inventory: &Inventory,
) -> Result<PreparationOutcome, PreparationError>;
async fn install_grafana(&self) -> Result<PreparationOutcome, PreparationError>;
}

View File

@ -1,27 +1,28 @@
use harmony_macros::hurl;
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();
use crate::modules::helm::chart::{HelmChartScore, HelmRepository};
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(),
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").unwrap(),
release_name: NonBlankString::from_str("grafana-operator").unwrap(),
chart_name: NonBlankString::from_str("grafana/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,
repository: Some(HelmRepository::new(
"grafana".to_string(),
hurl!("https://grafana.github.io/helm-charts"),
true,
)),
}
}

View File

@ -1 +1,2 @@
pub mod grafana;
pub mod helm;

View File

@ -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::{
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},
oberservability::monitoring::{AlertReceiver, AlertSender, ScrapeTarget},
},
};
#[derive(CustomResource, Serialize, Deserialize, Debug, Clone, JsonSchema)]
@ -26,6 +39,7 @@ pub struct AlertmanagerConfigSpec {
pub struct CRDPrometheus {
pub namespace: String,
pub client: Arc<K8sClient>,
pub service_monitor: Vec<ServiceMonitor>,
}
impl AlertSender for CRDPrometheus {
@ -40,6 +54,12 @@ impl Clone for Box<dyn AlertReceiver<CRDPrometheus>> {
}
}
impl Clone for Box<dyn ScrapeTarget<CRDPrometheus>> {
fn clone(&self) -> Self {
self.clone_box()
}
}
impl Serialize for Box<dyn AlertReceiver<CRDPrometheus>> {
fn serialize<S>(&self, _serializer: S) -> Result<S::Ok, S::Error>
where
@ -48,3 +68,24 @@ impl Serialize for Box<dyn AlertReceiver<CRDPrometheus>> {
todo!()
}
}
#[async_trait]
impl<T: Topology + K8sclient + PrometheusMonitoring<CRDPrometheus> + Grafana> Installable<T>
for CRDPrometheus
{
async fn configure(&self, inventory: &Inventory, topology: &T) -> Result<(), InterpretError> {
topology.ensure_grafana_operator(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(())
}
}

View File

@ -103,9 +103,34 @@ pub struct GrafanaDashboardSpec {
#[serde(default, skip_serializing_if = "Option::is_none")]
pub resync_period: Option<String>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub datasources: Option<Vec<GrafanaDashboardDatasource>>,
pub instance_selector: LabelSelector,
pub json: String,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub json: Option<String>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub grafana_com: Option<GrafanaCom>,
}
#[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<u32>,
}
// ------------------------------------------------------------------------------------------------
@ -126,20 +151,79 @@ pub struct GrafanaDatasourceSpec {
pub allow_cross_namespace_import: Option<bool>,
pub datasource: GrafanaDatasourceConfig,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub values_from: Option<Vec<GrafanaValueFrom>>,
}
#[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)]
#[serde(rename_all = "camelCase")]
pub struct GrafanaDatasourceConfig {
pub access: String,
pub database: Option<String>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub json_data: Option<BTreeMap<String, String>>,
pub database: Option<String>,
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<GrafanaDatasourceJsonData>,
/// Represents secureJsonData (secrets)
#[serde(default, skip_serializing_if = "Option::is_none")]
pub secure_json_data: Option<GrafanaDatasourceSecureJsonData>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub is_default: Option<bool>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub editable: Option<bool>,
}
#[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<String>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub http_header_name1: Option<String>,
/// Disable TLS skip verification (false = verify)
#[serde(default, skip_serializing_if = "Option::is_none")]
pub tls_skip_verify: Option<bool>,
/// Auth type - set to "forward" for OpenShift OAuth identity
#[serde(default, skip_serializing_if = "Option::is_none")]
pub oauth_pass_thru: Option<bool>,
}
#[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<String>,
}
// ------------------------------------------------------------------------------------------------
#[derive(Serialize, Deserialize, Debug, Clone, JsonSchema, Default)]

View File

@ -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<Vec<StaticConfig>>,
/// Kubernetes service discovery.
pub kubernetes_sd_configs: Option<Vec<KubernetesSDConfig>>,
/// HTTP-based service discovery.
pub http_sd_configs: Option<Vec<HttpSDConfig>>,
/// File-based service discovery.
pub file_sd_configs: Option<Vec<FileSDConfig>>,
/// DNS-based service discovery.
pub dns_sd_configs: Option<Vec<DnsSDConfig>>,
/// Consul service discovery.
pub consul_sd_configs: Option<Vec<ConsulSDConfig>>,
/// Relabeling configuration applied to discovered targets.
pub relabel_configs: Option<Vec<RelabelConfig>>,
/// Metric relabeling configuration applied to scraped samples.
pub metric_relabel_configs: Option<Vec<RelabelConfig>>,
/// Path to scrape metrics from (defaults to `/metrics`).
pub metrics_path: Option<String>,
/// Interval at which Prometheus scrapes targets (e.g., "30s").
pub scrape_interval: Option<String>,
/// Timeout for scraping (e.g., "10s").
pub scrape_timeout: Option<String>,
/// Optional job name override.
pub job_name: Option<String>,
/// Optional scheme (http or https).
pub scheme: Option<String>,
/// Authorization paramaters for snmp walk
pub params: Option<Params>,
}
/// Static configuration section of a ScrapeConfig.
#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)]
#[serde(rename_all = "camelCase")]
pub struct StaticConfig {
pub targets: Vec<String>,
pub labels: Option<LabelSelector>,
}
/// Relabeling configuration for target or metric relabeling.
#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)]
#[serde(rename_all = "camelCase")]
pub struct RelabelConfig {
pub source_labels: Option<Vec<String>>,
pub separator: Option<String>,
pub target_label: Option<String>,
pub regex: Option<String>,
pub modulus: Option<u64>,
pub replacement: Option<String>,
pub action: Option<String>,
}
/// 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<NamespaceSelector>,
pub selectors: Option<Vec<LabelSelector>>,
pub api_server: Option<String>,
pub bearer_token_file: Option<String>,
pub tls_config: Option<TLSConfig>,
}
/// Namespace selector for Kubernetes service discovery.
#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)]
#[serde(rename_all = "camelCase")]
pub struct NamespaceSelector {
pub any: Option<bool>,
pub match_names: Option<Vec<String>>,
}
/// 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<String>,
pub basic_auth: Option<BasicAuth>,
pub authorization: Option<Authorization>,
pub tls_config: Option<TLSConfig>,
}
/// File-based service discovery configuration.
#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)]
#[serde(rename_all = "camelCase")]
pub struct FileSDConfig {
pub files: Vec<String>,
pub refresh_interval: Option<String>,
}
/// DNS-based service discovery configuration.
#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)]
#[serde(rename_all = "camelCase")]
pub struct DnsSDConfig {
pub names: Vec<String>,
pub refresh_interval: Option<String>,
pub type_: Option<String>, // SRV, A, AAAA
pub port: Option<u16>,
}
/// Consul service discovery configuration.
#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)]
#[serde(rename_all = "camelCase")]
pub struct ConsulSDConfig {
pub server: String,
pub services: Option<Vec<String>>,
pub scheme: Option<String>,
pub datacenter: Option<String>,
pub tag_separator: Option<String>,
pub refresh_interval: Option<String>,
pub tls_config: Option<TLSConfig>,
}
/// Basic authentication credentials.
#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)]
#[serde(rename_all = "camelCase")]
pub struct BasicAuth {
pub username: String,
pub password: Option<String>,
pub password_file: Option<String>,
}
/// Bearer token or other auth mechanisms.
#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)]
#[serde(rename_all = "camelCase")]
pub struct Authorization {
pub credentials: Option<String>,
pub credentials_file: Option<String>,
pub type_: Option<String>,
}
/// TLS configuration for secure scraping.
#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)]
#[serde(rename_all = "camelCase")]
pub struct TLSConfig {
pub ca_file: Option<String>,
pub cert_file: Option<String>,
pub key_file: Option<String>,
pub server_name: Option<String>,
pub insecure_skip_verify: Option<bool>,
}
/// Authorization parameters for SNMP walk.
#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)]
#[serde(rename_all = "camelCase")]
pub struct Params {
pub auth: Option<Vec<String>>,
pub module: Option<Vec<String>>,
}

View File

@ -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;

View File

@ -31,6 +31,7 @@ impl<T: Topology + HelmCommand + TenantManager> Score<T> for HelmPrometheusAlert
sender: KubePrometheus { config },
receivers: self.receivers.clone(),
rules: self.rules.clone(),
scrape_targets: None,
})
}
fn name(&self) -> String {

View File

@ -6,3 +6,4 @@ pub mod kube_prometheus;
pub mod ntfy;
pub mod okd;
pub mod prometheus;
pub mod scrape_target;

View File

@ -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 {

View File

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

View File

@ -0,0 +1,80 @@
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<CRDPrometheus> for Server {
async fn install(&self, sender: &CRDPrometheus) -> Result<Outcome, InterpretError> {
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()
)))
}
fn clone_box(&self) -> Box<dyn ScrapeTarget<CRDPrometheus>> {
Box::new(self.clone())
}
}

View File

@ -12,7 +12,8 @@ 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, GrafanaSecretKeyRef, GrafanaSpec,
GrafanaValueFrom, GrafanaValueSource,
};
use crate::modules::monitoring::kube_prometheus::crd::crd_prometheus_rules::{
PrometheusRule, PrometheusRuleSpec, RuleGroup,
@ -39,7 +40,7 @@ use crate::{
};
use harmony_types::id::Id;
use super::prometheus::PrometheusApplicationMonitoring;
use super::prometheus::PrometheusMonitoring;
#[derive(Clone, Debug, Serialize)]
pub struct K8sPrometheusCRDAlertingScore {
@ -49,7 +50,7 @@ pub struct K8sPrometheusCRDAlertingScore {
pub prometheus_rules: Vec<RuleGroup>,
}
impl<T: Topology + K8sclient + PrometheusApplicationMonitoring<CRDPrometheus>> Score<T>
impl<T: Topology + K8sclient + PrometheusMonitoring<CRDPrometheus>> Score<T>
for K8sPrometheusCRDAlertingScore
{
fn create_interpret(&self) -> Box<dyn crate::interpret::Interpret<T>> {
@ -75,7 +76,7 @@ pub struct K8sPrometheusCRDAlertingInterpret {
}
#[async_trait]
impl<T: Topology + K8sclient + PrometheusApplicationMonitoring<CRDPrometheus>> Interpret<T>
impl<T: Topology + K8sclient + PrometheusMonitoring<CRDPrometheus>> Interpret<T>
for K8sPrometheusCRDAlertingInterpret
{
async fn execute(
@ -466,10 +467,13 @@ 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()),
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 {
@ -495,7 +499,11 @@ impl K8sPrometheusCRDAlertingInterpret {
"http://prometheus-operated.{}.svc.cluster.local:9090",
self.sender.namespace.clone()
),
secure_json_data: None,
is_default: None,
editable: None,
},
values_from: None,
},
};
@ -516,7 +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,
},
};

View File

@ -9,11 +9,17 @@ use crate::{
};
#[async_trait]
pub trait PrometheusApplicationMonitoring<S: AlertSender> {
pub trait PrometheusMonitoring<S: AlertSender> {
async fn install_prometheus(
&self,
sender: &S,
inventory: &Inventory,
receivers: Option<Vec<Box<dyn AlertReceiver<S>>>>,
) -> Result<PreparationOutcome, PreparationError>;
async fn ensure_prometheus_operator(
&self,
sender: &S,
inventory: &Inventory,
) -> Result<PreparationOutcome, PreparationError>;
}

View File

@ -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<RuleGroup>,
}
impl<T: Topology + K8sclient + Ingress + PrometheusApplicationMonitoring<RHOBObservability>>
Score<T> for RHOBAlertingScore
impl<T: Topology + K8sclient + Ingress + PrometheusMonitoring<RHOBObservability>> Score<T>
for RHOBAlertingScore
{
fn create_interpret(&self) -> Box<dyn crate::interpret::Interpret<T>> {
Box::new(RHOBAlertingInterpret {
@ -74,8 +74,8 @@ pub struct RHOBAlertingInterpret {
}
#[async_trait]
impl<T: Topology + K8sclient + Ingress + PrometheusApplicationMonitoring<RHOBObservability>>
Interpret<T> for RHOBAlertingInterpret
impl<T: Topology + K8sclient + Ingress + PrometheusMonitoring<RHOBObservability>> Interpret<T>
for RHOBAlertingInterpret
{
async fn execute(
&self,