feat(argocd): Can now detect argocd instance when already installed and write crd accordingly. One major caveat though is that crd versions are not managed properly yet

This commit is contained in:
2025-10-15 22:27:15 -04:00
parent dc70266b5a
commit 9f7b90d182
7 changed files with 296 additions and 73 deletions

View File

@@ -137,7 +137,7 @@ impl K8sClient {
// Get the application-controller ServiceAccount name (fallback to default)
pub async fn get_argocd_controller_sa_name(&self, ns: &str) -> Result<String, Error> {
let api: Api<Deployment> = Api::namespaced(self.client.clone(), ns);
let lp = ListParams::default().labels("app.kubernetes.io/name=argocd-application-controller");
let lp = ListParams::default().labels("app.kubernetes.io/component=controller");
let list = api.list(&lp).await?;
if let Some(dep) = list.items.get(0) {
if let Some(sa) = dep

View File

@@ -2,7 +2,7 @@ use std::{process::Command, sync::Arc};
use async_trait::async_trait;
use kube::api::GroupVersionKind;
use log::{debug, info, warn};
use log::{debug, info, trace, warn};
use serde::Serialize;
use tokio::sync::OnceCell;
@@ -71,6 +71,7 @@ pub struct K8sAnywhereTopology {
#[async_trait]
impl K8sclient for K8sAnywhereTopology {
async fn k8s_client(&self) -> Result<Arc<K8sClient>, String> {
trace!("getting k8s client");
let state = match self.k8s_state.get() {
Some(state) => state,
None => return Err("K8s state not initialized yet".to_string()),
@@ -620,36 +621,56 @@ impl TenantManager for K8sAnywhereTopology {
#[async_trait]
impl Ingress for K8sAnywhereTopology {
//TODO this is specifically for openshift/okd which violates the k8sanywhere idea
async fn get_domain(&self, service: &str) -> Result<String, PreparationError> {
use log::{trace, debug, warn};
let client = self.k8s_client().await?;
if let Some(Some(k8s_state)) = self.k8s_state.get() {
match k8s_state.source {
K8sSource::LocalK3d => Ok(format!("{service}.local.k3d")),
K8sSource::LocalK3d => {
// Local developer UX
return Ok(format!("{service}.local.k3d"));
}
K8sSource::Kubeconfig => {
self.openshift_ingress_operator_available().await?;
trace!("K8sSource is kubeconfig; attempting to detect domain");
let gvk = GroupVersionKind {
group: "operator.openshift.io".into(),
version: "v1".into(),
kind: "IngressController".into(),
};
let ic = client
.get_resource_json_value(
"default",
Some("openshift-ingress-operator"),
&gvk,
)
.await
.map_err(|_| {
PreparationError::new("Failed to fetch IngressController".to_string())
})?;
// 1) Try OpenShift IngressController domain (backward compatible)
if self.openshift_ingress_operator_available().await.is_ok() {
trace!("OpenShift ingress operator detected; using IngressController");
let gvk = GroupVersionKind {
group: "operator.openshift.io".into(),
version: "v1".into(),
kind: "IngressController".into(),
};
let ic = client
.get_resource_json_value("default", Some("openshift-ingress-operator"), &gvk)
.await
.map_err(|_| PreparationError::new("Failed to fetch IngressController".to_string()))?;
match ic.data["status"]["domain"].as_str() {
Some(domain) => Ok(format!("{service}.{domain}")),
None => Err(PreparationError::new("Could not find domain".to_string())),
if let Some(domain) = ic.data["status"]["domain"].as_str() {
return Ok(format!("{service}.{domain}"));
} else {
warn!("OpenShift IngressController present but no status.domain set");
}
} else {
trace!("OpenShift ingress operator not detected; trying generic Kubernetes");
}
// 2) Try NGINX Ingress Controller common setups
// 2.a) Well-known namespace/name for the controller Service
// - upstream default: namespace "ingress-nginx", service "ingress-nginx-controller"
// - some distros: "ingress-nginx-controller" svc in "ingress-nginx" ns
// If found with LoadBalancer ingress hostname, use its base domain.
if let Some(domain) = try_nginx_lb_domain(&client).await? {
return Ok(format!("{service}.{domain}"));
}
// 3) Fallback: internal cluster DNS suffix (service.namespace.svc.cluster.local)
// We don't have tenant namespace here, so we fallback to 'default' with a warning.
warn!("Could not determine external ingress domain; falling back to internal-only DNS");
let internal = format!("{service}.default.svc.cluster.local");
Ok(internal)
}
}
} else {
@@ -659,3 +680,57 @@ impl Ingress for K8sAnywhereTopology {
}
}
}
async fn try_nginx_lb_domain(client: &K8sClient) -> Result<Option<String>, PreparationError> {
use log::{trace, debug};
// Try common service path: svc/ingress-nginx-controller in ns/ingress-nginx
let svc_gvk = GroupVersionKind {
group: "".into(), // core
version: "v1".into(),
kind: "Service".into(),
};
let candidates = [
("ingress-nginx", "ingress-nginx-controller"),
("ingress-nginx", "ingress-nginx-controller-internal"),
("ingress-nginx", "ingress-nginx"), // some charts name the svc like this
("kube-system", "ingress-nginx-controller"), // less common but seen
];
for (ns, name) in candidates {
trace!("Checking NGINX Service {ns}/{name} for LoadBalancer hostname");
if let Ok(svc) = client.get_resource_json_value(ns, Some(name), &svc_gvk).await {
let lb_hosts = svc.data["status"]["loadBalancer"]["ingress"].as_array().cloned().unwrap_or_default();
for entry in lb_hosts {
if let Some(host) = entry.get("hostname").and_then(|v| v.as_str()) {
debug!("Found NGINX LB hostname: {host}");
if let Some(domain) = extract_base_domain(host) {
return Ok(Some(domain.to_string()));
} else {
return Ok(Some(host.to_string())); // already a domain
}
}
if let Some(ip) = entry.get("ip").and_then(|v| v.as_str()) {
// If only an IP is exposed, we can't create a hostname; return None to keep searching
debug!("NGINX LB exposes IP {ip} (no hostname); skipping");
}
}
}
}
Ok(None)
}
fn extract_base_domain(host: &str) -> Option<String> {
// For a host like a1b2c3d4e5f6abcdef.elb.amazonaws.com -> base domain elb.amazonaws.com
// For a managed DNS like xyz.example.com -> base domain example.com (keep 2+ labels)
// Heuristic: keep last 2 labels by default; special-case known multi-label TLDs if needed.
let parts: Vec<&str> = host.split('.').collect();
if parts.len() >= 2 {
// Very conservative: last 2 labels
Some(parts[parts.len() - 2..].join("."))
} else {
None
}
}

View File

@@ -21,7 +21,7 @@ pub struct Helm {
pub skip_schema_validation: Option<bool>,
pub version: Option<String>,
pub kube_version: Option<String>,
pub api_versions: Vec<String>,
// pub api_versions: Vec<String>,
pub namespace: Option<String>,
}
@@ -105,7 +105,7 @@ impl Default for ArgoApplication {
skip_schema_validation: None,
version: None,
kube_version: None,
api_versions: vec![],
// api_versions: vec![],
namespace: None,
},
path: "".to_string(),
@@ -155,7 +155,7 @@ impl From<CDApplicationConfig> for ArgoApplication {
skip_schema_validation: None,
version: None,
kube_version: None,
api_versions: vec![],
// api_versions: vec![],
namespace: None,
},
},
@@ -181,13 +181,11 @@ impl From<CDApplicationConfig> for ArgoApplication {
}
impl ArgoApplication {
pub fn to_yaml(&self) -> serde_yaml::Value {
pub fn to_yaml(&self, target_namespace: Option<&str>) -> serde_yaml::Value {
let name = &self.name;
let namespace = if let Some(ns) = self.namespace.as_ref() {
ns
} else {
"argocd"
};
let default_ns = "argocd".to_string();
let namespace: &str =
target_namespace.unwrap_or(self.namespace.as_ref().unwrap_or(&default_ns));
let project = &self.project;
let yaml_str = format!(
@@ -285,7 +283,7 @@ mod tests {
skip_schema_validation: None,
version: None,
kube_version: None,
api_versions: vec![],
// api_versions: vec![],
namespace: None,
},
path: "".to_string(),
@@ -345,7 +343,7 @@ spec:
assert_eq!(
expected_yaml_output.trim(),
serde_yaml::to_string(&app.clone().to_yaml())
serde_yaml::to_string(&app.clone().to_yaml(None))
.unwrap()
.trim()
);

View File

@@ -1,8 +1,7 @@
use async_trait::async_trait;
<<<<<<< HEAD
use harmony_macros::hurl;
use kube::{Api, api::GroupVersionKind};
use log::{debug, warn};
use log::{debug, info, trace, warn};
use non_blank_string_rs::NonBlankString;
use serde::Serialize;
use std::{str::FromStr, sync::Arc};
@@ -11,9 +10,12 @@ use crate::{
data::Version,
interpret::{Interpret, InterpretError, InterpretName, InterpretStatus, Outcome},
inventory::Inventory,
modules::{argocd::{detect_argo_deployment_type, ArgoDeploymentType}, helm::chart::{HelmChartScore, HelmRepository}},
modules::{
argocd::{ArgoDeploymentType, detect_argo_deployment_type},
helm::chart::{HelmChartScore, HelmRepository},
},
score::Score,
topology::{ingress::Ingress, k8s::K8sClient, HelmCommand, K8sclient, Topology},
topology::{HelmCommand, K8sclient, Topology, ingress::Ingress, k8s::K8sClient},
};
use harmony_types::id::Id;
@@ -22,7 +24,7 @@ use super::ArgoApplication;
#[derive(Debug, Serialize, Clone)]
pub struct ArgoHelmScore {
pub namespace: String,
// TODO remove this field and rely on topology, it can now know what flavor it is running
// TODO: remove and rely on topology (it now knows the flavor)
pub openshift: bool,
pub argo_apps: Vec<ArgoApplication>,
}
@@ -53,38 +55,101 @@ impl<T: Topology + K8sclient + HelmCommand + Ingress> Interpret<T> for ArgoInter
inventory: &Inventory,
topology: &T,
) -> Result<Outcome, InterpretError> {
let k8s_client = topology.k8s_client().await?;
let svc = format!("argo-{}", self.score.namespace.clone());
trace!("Starting ArgoInterpret execution {self:?}");
let k8s_client: Arc<K8sClient> = topology.k8s_client().await?;
trace!("Got k8s client");
let desired_ns = self.score.namespace.clone();
debug!("ArgoInterpret detecting cluster configuration");
let svc = format!("argo-{}", desired_ns);
let domain = topology.get_domain(&svc).await?;
debug!("Resolved Argo service domain for '{}': {}", svc, domain);
let current_argo_deployment = detect_argo_deployment_type(&k8s_client, &self.score.namespace).await?;
// Detect current Argo deployment type
let current = detect_argo_deployment_type(&k8s_client, &desired_ns).await?;
info!("Detected Argo deployment type: {:?}", current);
match current_argo_deployment {
ArgoDeploymentType::NotInstalled => todo!(),
ArgoDeploymentType::AvailableInDesiredNamespace(_) => todo!(),
ArgoDeploymentType::InstalledClusterWide(_) => todo!(),
ArgoDeploymentType::InstalledNamespaceScoped(_) => todo!(),
// Decide control namespace and whether we must install
let (control_ns, must_install) = match current.clone() {
ArgoDeploymentType::NotInstalled => {
info!(
"Argo CD not installed. Will install via Helm into namespace '{}'.",
desired_ns
);
(desired_ns.clone(), true)
}
ArgoDeploymentType::AvailableInDesiredNamespace(ns) => {
info!(
"Argo CD already installed by Harmony in '{}'. Skipping install.",
ns
);
(ns, false)
}
ArgoDeploymentType::InstalledClusterWide(ns) => {
info!(
"Argo CD installed cluster-wide in namespace '{}'.",
ns
);
(ns, false)
}
ArgoDeploymentType::InstalledNamespaceScoped(ns) => {
// TODO we could support this use case by installing a new argo instance. But that
// means handling a few cases that are out of scope for now :
// - Wether argo operator is installed
// - Managing CRD versions compatibility
// - Potentially handling the various k8s flavors and setups we might encounter
//
// There is a possibility that the helm chart already handles most or even all of these use cases but they are out of scope for now.
let msg = format!(
"Argo CD found in '{}' but it is namespace-scoped and not supported for attachment yet.",
ns
);
warn!("{}", msg);
return Err(InterpretError::new(msg));
}
};
let helm_score =
argo_helm_chart_score(&self.score.namespace, self.score.openshift, &domain);
helm_score.interpret(inventory, topology).await?;
info!("ArgoCD will be installed : {must_install} . Current argocd status : {current:?} ");
if must_install {
let helm_score = argo_helm_chart_score(&desired_ns, self.score.openshift, &domain);
info!(
"Installing Argo CD via Helm into namespace '{}' ...",
desired_ns
);
helm_score.interpret(inventory, topology).await?;
info!("Argo CD install complete in '{}'.", desired_ns);
}
let yamls: Vec<serde_yaml::Value> = self
.argo_apps
.iter()
.map(|a| a.to_yaml(Some(&control_ns)))
.collect();
info!(
"Applying {} Argo application object(s) into control namespace '{}'.",
yamls.len(),
control_ns
);
k8s_client
.apply_yaml_many(&self.argo_apps.iter().map(|a| a.to_yaml()).collect(), None)
.apply_yaml_many(&yamls, Some(control_ns.as_str()))
.await
.unwrap();
.map_err(|e| InterpretError::new(format!("Failed applying Argo CRs: {e}")))?;
Ok(Outcome::success_with_details(
format!(
"ArgoCD {} {}",
self.argo_apps.len(),
match self.argo_apps.len() {
1 => "application",
_ => "applications",
if self.argo_apps.len() == 1 {
"application"
} else {
"applications"
}
),
vec![format!("argo application: http://{}", domain)],
vec![
format!("control_namespace={}", control_ns),
format!("argo ui: http://{}", domain),
],
))
}
@@ -93,7 +158,7 @@ impl<T: Topology + K8sclient + HelmCommand + Ingress> Interpret<T> for ArgoInter
}
fn get_version(&self) -> Version {
todo!()
Version::from("0.1.0").unwrap()
}
fn get_status(&self) -> InterpretStatus {
@@ -101,7 +166,7 @@ impl<T: Topology + K8sclient + HelmCommand + Ingress> Interpret<T> for ArgoInter
}
fn get_children(&self) -> Vec<Id> {
todo!()
vec![]
}
}

View File

@@ -198,7 +198,7 @@ impl<
openshift: true,
argo_apps: vec![ArgoApplication::from(CDApplicationConfig {
// helm pull oci://hub.nationtech.io/harmony/harmony-example-rust-webapp-chart --version 0.1.0
version: Version::from("0.1.0").unwrap(),
version: Version::from("0.2.1").unwrap(),
helm_chart_repo_url: "hub.nationtech.io/harmony".to_string(),
helm_chart_name: format!("{}-chart", self.application.name()),
values_overrides: None,

View File

@@ -480,7 +480,7 @@ apiVersion: v2
name: {chart_name}
description: A Helm chart for the {app_name} web application.
type: application
version: 0.2.0
version: 0.2.1
appVersion: "{image_tag}"
"#,
);

View File

@@ -29,58 +29,142 @@ pub enum ArgoDeploymentType {
pub async fn discover_argo_all(
k8s: &Arc<K8sClient>,
) -> Result<Vec<DiscoveredArgo>, InterpretError> {
use log::{debug, info, trace, warn};
trace!("Starting Argo discovery");
// CRDs
let mut has_crds = true;
for crd in vec!["applications.argoproj.io", "appprojects.argoproj.io"] {
let required_crds = vec!["applications.argoproj.io", "appprojects.argoproj.io"];
trace!("Checking required Argo CRDs: {:?}", required_crds);
for crd in required_crds {
trace!("Verifying CRD presence: {crd}");
let crd_exists = k8s.has_crd(crd).await.map_err(|e| {
InterpretError::new(format!("Failed to verify existence of CRD {crd}: {e}"))
})?;
debug!("CRD {crd} exists: {crd_exists}");
if !crd_exists {
info!("Missing argo CRD {crd}, looks like ArgoCD is not installed");
info!(
"Missing Argo CRD {crd}, looks like Argo CD is not installed (or partially installed)"
);
has_crds = false;
break;
}
}
// Namespaces that have healthy argocd deployments
let candidate_namespaces = k8s
trace!(
"Listing namespaces with healthy Argo CD deployments using selector app.kubernetes.io/part-of=argocd"
);
let mut candidate_namespaces = k8s
.list_namespaces_with_healthy_deployments("app.kubernetes.io/part-of=argocd")
.await
.map_err(|e| InterpretError::new(format!("List healthy argocd deployments: {e}")))?;
trace!(
"Listing namespaces with healthy Argo CD deployments using selector app.kubernetes.io/name=argo-cd"
);
candidate_namespaces.append(
&mut k8s
.list_namespaces_with_healthy_deployments("app.kubernetes.io/name=argo-cd")
.await
.map_err(|e| InterpretError::new(format!("List healthy argocd deployments: {e}")))?,
);
debug!(
"Discovered {} candidate namespace(s) for Argo CD: {:?}",
candidate_namespaces.len(),
candidate_namespaces
);
let mut found = Vec::new();
for ns in candidate_namespaces {
trace!("Evaluating namespace '{ns}' for Argo CD instance");
// Require the application-controller to be healthy (sanity check)
trace!(
"Checking healthy deployment with label app.kubernetes.io/name=argocd-application-controller in namespace '{ns}'"
);
let controller_ok = k8s
.has_healthy_deployment_with_label(
&ns,
"app.kubernetes.io/name=argocd-application-controller",
)
.await
.unwrap_or(false);
.unwrap_or_else(|e| {
warn!(
"Error while checking application-controller health in namespace '{ns}': {e}"
);
false
}) || k8s
.has_healthy_deployment_with_label(
&ns,
"app.kubernetes.io/component=controller",
)
.await
.unwrap_or_else(|e| {
warn!(
"Error while checking application-controller health in namespace '{ns}': {e}"
);
false
});
debug!("Namespace '{ns}': application-controller healthy = {controller_ok}");
if !controller_ok {
trace!("Skipping namespace '{ns}' because application-controller is not healthy");
continue;
}
let scope = if k8s.is_argocd_cluster_wide(&ns).await? {
ArgoScope::ClusterWide(ns.to_string())
} else {
ArgoScope::NamespaceScoped(ns.to_string())
trace!("Determining Argo CD scope for namespace '{ns}' (cluster-wide vs namespace-scoped)");
let scope = match k8s.is_argocd_cluster_wide(&ns).await {
Ok(true) => {
debug!("Namespace '{ns}' identified as cluster-wide Argo CD control plane");
ArgoScope::ClusterWide(ns.to_string())
}
Ok(false) => {
debug!("Namespace '{ns}' identified as namespace-scoped Argo CD control plane");
ArgoScope::NamespaceScoped(ns.to_string())
}
Err(e) => {
warn!(
"Failed to determine Argo CD scope for namespace '{ns}': {e}. Assuming namespace-scoped."
);
ArgoScope::NamespaceScoped(ns.to_string())
}
};
trace!("Checking optional ApplicationSet CRD (applicationsets.argoproj.io)");
let has_applicationset = match k8s.has_crd("applicationsets.argoproj.io").await {
Ok(v) => {
debug!("applicationsets.argoproj.io present: {v}");
v
}
Err(e) => {
warn!("Failed to check applicationsets.argoproj.io CRD: {e}. Assuming absent.");
false
}
};
let argo = DiscoveredArgo {
control_namespace: ns,
control_namespace: ns.clone(),
scope,
has_crds,
has_applicationset: k8s.has_crd("applicationsets.argoproj.io").await?,
has_applicationset,
};
debug!("Found argo instance {argo:?}");
debug!("Discovered Argo instance in '{ns}': {argo:?}");
found.push(argo);
}
if found.is_empty() {
info!("No Argo CD installations discovered");
} else {
info!(
"Argo CD discovery complete: {} instance(s) found",
found.len()
);
}
Ok(found)
}
@@ -89,6 +173,7 @@ pub async fn detect_argo_deployment_type(
desired_namespace: &str,
) -> Result<ArgoDeploymentType, InterpretError> {
let discovered = discover_argo_all(k8s).await?;
debug!("Discovered argo instances {discovered:?}");
if discovered.is_empty() {
return Ok(ArgoDeploymentType::NotInstalled);