feat:added Slack notifications support #38

Merged
wjro merged 2 commits from feat/slack-notifs into master 2025-05-22 20:04:52 +00:00
4 changed files with 91 additions and 81 deletions
Showing only changes of commit 44b2b092a8 - Show all commits

View File

@ -1,8 +1,7 @@
use email_address::EmailAddress;
use serde::Serialize; use serde::Serialize;
use url::Url;
use super::monitoring_alerting::WebhookServiceType; use super::monitoring_alerting::AlertChannel;
#[derive(Debug, Clone, Serialize)] #[derive(Debug, Clone, Serialize)]
pub struct KubePrometheusConfig { pub struct KubePrometheusConfig {
@ -23,9 +22,7 @@ pub struct KubePrometheusConfig {
pub kube_proxy: bool, pub kube_proxy: bool,
pub kube_state_metrics: bool, pub kube_state_metrics: bool,
pub prometheus_operator: bool, pub prometheus_operator: bool,
pub webhook_url: Option<Url>, pub alert_channel: Option<AlertChannel>,
pub webhook_service_type: Option<WebhookServiceType>,
pub discord_alert_manager_release_name: String,
} }
impl KubePrometheusConfig { impl KubePrometheusConfig {
pub fn new() -> Self { pub fn new() -> Self {
@ -34,8 +31,7 @@ impl KubePrometheusConfig {
default_rules: true, default_rules: true,
windows_monitoring: false, windows_monitoring: false,
alert_manager: true, alert_manager: true,
webhook_service_type: None, alert_channel: None,
webhook_url: None,
grafana: true, grafana: true,
node_exporter: false, node_exporter: false,
prometheus: true, prometheus: true,
@ -49,7 +45,6 @@ impl KubePrometheusConfig {
prometheus_operator: true, prometheus_operator: true,
core_dns: false, core_dns: false,
kube_scheduler: false, kube_scheduler: false,
discord_alert_manager_release_name: "discord-alert-manager".into(),
} }
} }
} }

View File

@ -4,13 +4,14 @@ use non_blank_string_rs::NonBlankString;
use crate::modules::helm::chart::HelmChartScore; use crate::modules::helm::chart::HelmChartScore;
use super::config::KubePrometheusConfig; use super::{config::KubePrometheusConfig, monitoring_alerting::AlertChannel};
pub fn discord_alert_manager_score(config: &KubePrometheusConfig) -> HelmChartScore { pub fn discord_alert_manager_score(config: &KubePrometheusConfig) -> HelmChartScore {
let url = if let Some(url) = &config.webhook_url { let (url, release_name) = match &config.alert_channel {
Review

Since we're supporting a list, shouldn't we use filter_map instead so we can handle all the instances at once?

Also this is a smell to me. All the handlers should implement the same trait. For example :

trait AlertEndpoint {
  fn register(&self, ...);
}

Ce serait une facon de simplifier la gestion de chacun des types ensuite. C'est plus type safe et moins error prone.

Since we're supporting a list, shouldn't we use filter_map instead so we can handle all the instances at once? Also this is a smell to me. All the handlers should implement the same trait. For example : ``` trait AlertEndpoint { fn register(&self, ...); } ``` Ce serait une facon de simplifier la gestion de chacun des types ensuite. C'est plus type safe et moins error prone.
url.to_string() Some(AlertChannel::Discord { webhook_url, name }) => {
} else { (webhook_url.to_string(), name.to_string())
"None".to_string() }
_ => panic!("Expected Discord alert channel configuration"),
}; };
let values = format!( let values = format!(
@ -24,7 +25,7 @@ environment:
HelmChartScore { HelmChartScore {
namespace: Some(NonBlankString::from_str(&config.namespace).unwrap()), namespace: Some(NonBlankString::from_str(&config.namespace).unwrap()),
release_name: NonBlankString::from_str(&config.discord_alert_manager_release_name).unwrap(), release_name: NonBlankString::from_str(&release_name).unwrap(),
chart_name: NonBlankString::from_str("oci://hub.nationtech.io/nt/alertmanager-discord") chart_name: NonBlankString::from_str("oci://hub.nationtech.io/nt/alertmanager-discord")
.unwrap(), .unwrap(),
chart_version: None, chart_version: None,

View File

@ -1,5 +1,7 @@
use super::{config::KubePrometheusConfig, monitoring_alerting::WebhookServiceType}; use super::{config::KubePrometheusConfig, monitoring_alerting::AlertChannel};
use log::info;
use non_blank_string_rs::NonBlankString; use non_blank_string_rs::NonBlankString;
use url::Url;
use std::{collections::HashMap, str::FromStr}; use std::{collections::HashMap, str::FromStr};
use crate::modules::helm::chart::HelmChartScore; use crate::modules::helm::chart::HelmChartScore;
@ -10,14 +12,6 @@ pub fn kube_prometheus_helm_chart_score(config: &KubePrometheusConfig) -> HelmCh
let default_rules = config.default_rules.to_string(); let default_rules = config.default_rules.to_string();
let windows_monitoring = config.windows_monitoring.to_string(); let windows_monitoring = config.windows_monitoring.to_string();
let alert_manager = config.alert_manager.to_string(); let alert_manager = config.alert_manager.to_string();
let webhook_service_type = if let Some(service) = &config.webhook_service_type {
match service {
WebhookServiceType::Discord => "Discord".to_string(),
WebhookServiceType::Slack => "Slack".to_string(),
}
} else {
"None".to_string()
};
let grafana = config.grafana.to_string(); let grafana = config.grafana.to_string();
let kubernetes_service_monitors = config.kubernetes_service_monitors.to_string(); let kubernetes_service_monitors = config.kubernetes_service_monitors.to_string();
let kubernetes_api_server = config.kubernetes_api_server.to_string(); let kubernetes_api_server = config.kubernetes_api_server.to_string();
@ -31,8 +25,7 @@ pub fn kube_prometheus_helm_chart_score(config: &KubePrometheusConfig) -> HelmCh
let node_exporter = config.node_exporter.to_string(); let node_exporter = config.node_exporter.to_string();
let prometheus_operator = config.prometheus_operator.to_string(); let prometheus_operator = config.prometheus_operator.to_string();
let prometheus = config.prometheus.to_string(); let prometheus = config.prometheus.to_string();
let discord_alert_manager_release_name = config.discord_alert_manager_release_name.to_string(); let mut values = format!(
let values = format!(
r#" r#"
additionalPrometheusRulesMap: additionalPrometheusRulesMap:
pods-status-alerts: pods-status-alerts:
@ -45,23 +38,23 @@ additionalPrometheusRulesMap:
labels: labels:
severity: critical severity: critical
annotations: annotations:
title: "[CRIT] POD not healthy : {{ $labels.pod }}" title: "[CRIT] POD not healthy : {{{{ $labels.pod }}}}"
description: | description: |
A POD is in a non-ready state! A POD is in a non-ready state!
- **Pod**: {{ $labels.pod }} - **Pod**: {{{{ $labels.pod }}}}
- **Namespace**: {{ $labels.namespace }} - **Namespace**: {{{{ $labels.namespace }}}}
- alert: "[CRIT] POD crash looping" - alert: "[CRIT] POD crash looping"
expr: increase(kube_pod_container_status_restarts_total[5m]) > 3 expr: increase(kube_pod_container_status_restarts_total[5m]) > 3
for: 0m for: 0m
labels: labels:
severity: critical severity: critical
annotations: annotations:
title: "[CRIT] POD crash looping : {{ $labels.pod }}" title: "[CRIT] POD crash looping : {{{{ $labels.pod }}}}"
description: | description: |
A POD is drowning in a crash loop! A POD is drowning in a crash loop!
- **Pod**: {{ $labels.pod }} - **Pod**: {{{{ $labels.pod }}}}
- **Namespace**: {{ $labels.namespace }} - **Namespace**: {{{{ $labels.namespace }}}}
- **Instance**: {{ $labels.instance }} - **Instance**: {{{{ $labels.instance }}}}
pvc-alerts: pvc-alerts:
groups: groups:
- name: pvc-alerts - name: pvc-alerts
@ -82,8 +75,8 @@ additionalPrometheusRulesMap:
labels: labels:
severity: warning severity: warning
annotations: annotations:
description: The PVC {{ $labels.persistentvolumeclaim }} in namespace {{ $labels.namespace }} is predicted to fill over 95% in less than 2 days. description: The PVC {{{{ $labels.persistentvolumeclaim }}}} in namespace {{{{ $labels.namespace }}}} is predicted to fill over 95% in less than 2 days.
title: PVC {{ $labels.persistentvolumeclaim }} in namespace {{ $labels.namespace }} will fill over 95% in less than 2 days title: PVC {{{{ $labels.persistentvolumeclaim }}}} in namespace {{{{ $labels.namespace }}}} will fill over 95% in less than 2 days
defaultRules: defaultRules:
create: {default_rules} create: {default_rules}
rules: rules:
@ -131,7 +124,6 @@ alertmanager:
group_wait: 30s group_wait: 30s
group_interval: 5m group_interval: 5m
repeat_interval: 12h repeat_interval: 12h
receiver: '{webhook_service_type}'
routes: routes:
- receiver: 'null' - receiver: 'null'
matchers: matchers:
@ -139,10 +131,6 @@ alertmanager:
continue: false continue: false
receivers: receivers:
- name: 'null' - name: 'null'
- name: '{webhook_service_type}'
webhook_configs:
- url: 'http://{discord_alert_manager_release_name}-alertmanager-discord:9094'
send_resolved: true
grafana: grafana:
enabled: {grafana} enabled: {grafana}
kubernetesServiceMonitors: kubernetesServiceMonitors:
@ -172,6 +160,16 @@ prometheus:
"#, "#,
); );
if let Some(alert_channel) = &config.alert_channel {
match alert_channel {
AlertChannel::Discord { name, .. } => {
values.push_str(&discord_alert_builder(name).to_string());}
AlertChannel::Slack { slack_channel, webhook_url } => {
values.push_str(&slack_alert_builder(slack_channel, webhook_url).to_string());}
AlertChannel::Smpt { .. } => todo!(),
}
};
HelmChartScore { HelmChartScore {
namespace: Some(NonBlankString::from_str(&config.namespace).unwrap()), namespace: Some(NonBlankString::from_str(&config.namespace).unwrap()),
release_name: NonBlankString::from_str("kube-prometheus").unwrap(), release_name: NonBlankString::from_str("kube-prometheus").unwrap(),
@ -187,3 +185,44 @@ prometheus:
repository: None, repository: None,
} }
} }
fn discord_alert_builder(release_name: &String) -> String {
let discord_alert_manager_release_name = release_name;
let discord_alert_values = format!(
r#"
alertmanager:
config:
route:
receiver: 'Discord'
receivers:
- name: 'null'
- name: 'Discord'
webhook_configs:
- url: 'http://{discord_alert_manager_release_name}-alertmanager-discord:9094'
send_resolved: true
"#,
);
discord_alert_values
}
fn slack_alert_builder(slack_channel: &String, webhook_url: &Url) -> String {
let slack_alert_values = format!(
r#"
alertmanager:
config:
global:
slack_api_url: {webhook_url}
route:
receiver: 'Slack'
receivers:
- name: 'null'
- name: 'Slack'
slack_configs:
- channel: '{slack_channel}'
send_resolved: true
title: '{{{{ .CommonAnnotations.title }}}}'
text: '{{{{ .CommonAnnotations.description }}}}'
Review

Yeah, these two functions here discord_alert_builder and slack_alert_builder here should be implementations of a trait.

The fact that their names share the same semantics is a great givaway.

Also, returning (String, String) is pretty weak. I guess there should be a type or you should build a custom type for that. Then this type can Derive Serialize.

Yeah, these two functions here `discord_alert_builder` and `slack_alert_builder` here should be implementations of a trait. The fact that their names share the same semantics is a great givaway. Also, returning (String, String) is pretty weak. I guess there should be a type or you should build a custom type for that. Then this type can Derive Serialize.
"#,
);
slack_alert_values
}

View File

@ -20,9 +20,13 @@ use super::{
#[derive(Debug, Clone, Serialize)] #[derive(Debug, Clone, Serialize)]
pub enum AlertChannel { pub enum AlertChannel {
WebHookUrl { Discord {
url: Url, name: String,
webhook_service_type: WebhookServiceType, webhook_url: Url,
},
Slack {
slack_channel: String,
webhook_url: Url,
}, },
//TODO test and implement in helm chart //TODO test and implement in helm chart
//currently does not work //currently does not work
@ -32,13 +36,6 @@ pub enum AlertChannel {
}, },
} }
#[derive(Debug, Clone, Serialize)]
pub enum WebhookServiceType {
Discord,
//TODO test slack notifications
Slack,
}
#[derive(Debug, Clone, Serialize)] #[derive(Debug, Clone, Serialize)]
pub struct MonitoringAlertingStackScore { pub struct MonitoringAlertingStackScore {
pub alert_channel: Option<AlertChannel>, pub alert_channel: Option<AlertChannel>,
@ -52,26 +49,6 @@ impl MonitoringAlertingStackScore {
namespace: None, namespace: None,
} }
} }
fn set_alert_channel(&self, config: &mut KubePrometheusConfig) {
if let Some(alert_channel) = &self.alert_channel {
match alert_channel {
AlertChannel::WebHookUrl {
url,
webhook_service_type,
} => {
config.webhook_url = Some(url.clone());
config.webhook_service_type = Some(webhook_service_type.clone());
}
AlertChannel::Smpt {
//TODO setup smpt alerts
email_address,
service_name,
} => {
todo!()
}
}
}
}
} }
impl<T: Topology + HelmCommand> Score<T> for MonitoringAlertingStackScore { impl<T: Topology + HelmCommand> Score<T> for MonitoringAlertingStackScore {
@ -93,10 +70,10 @@ struct MonitoringAlertingStackInterpret {
impl MonitoringAlertingStackInterpret { impl MonitoringAlertingStackInterpret {
async fn build_kube_prometheus_helm_chart_config(&self) -> KubePrometheusConfig { async fn build_kube_prometheus_helm_chart_config(&self) -> KubePrometheusConfig {
let mut config = KubePrometheusConfig::new(); let mut config = KubePrometheusConfig::new();
self.score.set_alert_channel(&mut config);
if let Some(ns) = &self.score.namespace { if let Some(ns) = &self.score.namespace {
config.namespace = ns.clone(); config.namespace = ns.clone();
} }
config.alert_channel = self.score.alert_channel.clone();
config config
} }
@ -120,23 +97,20 @@ impl MonitoringAlertingStackInterpret {
config: &KubePrometheusConfig, config: &KubePrometheusConfig,
) -> Result<Outcome, InterpretError> { ) -> Result<Outcome, InterpretError> {
match &self.score.alert_channel { match &self.score.alert_channel {
Some(AlertChannel::WebHookUrl { Some(channel) => match channel {
webhook_service_type, AlertChannel::Discord { .. } => {
..
}) => match webhook_service_type {
WebhookServiceType::Discord => {
discord_alert_manager_score(config) discord_alert_manager_score(config)
.create_interpret() .create_interpret()
.execute(inventory, topology) .execute(inventory, topology)
.await .await
} }
WebhookServiceType::Slack => Ok(Outcome::success( AlertChannel::Slack { .. } => Ok(Outcome::success(
"No extra configs for slack alerting".to_string(), "No extra configs for slack alerting".to_string(),
)), )),
AlertChannel::Smpt { .. } => {
todo!()
}
}, },
Some(AlertChannel::Smpt { .. }) => {
todo!()
}
None => Ok(Outcome::success("No alert channel configured".to_string())), None => Ok(Outcome::success("No alert channel configured".to_string())),
} }
} }
@ -155,7 +129,8 @@ impl<T: Topology + HelmCommand> Interpret<T> for MonitoringAlertingStackInterpre
self.deploy_kube_prometheus_helm_chart_score(inventory, topology, &config) self.deploy_kube_prometheus_helm_chart_score(inventory, topology, &config)
.await?; .await?;
info!("Installing alert channel service"); info!("Installing alert channel service");
self.deploy_alert_channel_service(inventory, topology, &config).await?; self.deploy_alert_channel_service(inventory, topology, &config)
.await?;
Ok(Outcome::success(format!( Ok(Outcome::success(format!(
"succesfully deployed monitoring and alerting stack" "succesfully deployed monitoring and alerting stack"
))) )))