This is an automated email from the ASF dual-hosted git repository. gongchao pushed a commit to branch integte-extern-alarm in repository https://gitbox.apache.org/repos/asf/hertzbeat.git
commit c55e91054408d1a7ec06812068c3e4f306b8a7b7 Author: tomsun28 <[email protected]> AuthorDate: Fri Jan 10 23:17:19 2025 +0800 [improve] init Signed-off-by: tomsun28 <[email protected]> --- .../alert-integration.component.ts | 13 ++-- .../{webhook.en-US.md => alertmanager.en-US.md} | 0 .../doc/alert-integration/alertmanager.zh-CN.md | 44 ++++++++++++ .../doc/alert-integration/alertmanager.zh-TW.md | 0 .../doc/alert-integration/prometheus.zh-CN.md | 83 ++++++++++++++++------ .../assets/doc/alert-integration/webhook.en-US.md | 53 ++++++++++++-- .../assets/doc/alert-integration/webhook.zh-CN.md | 81 +++++++++++++++------ web-app/src/assets/i18n/zh-CN.json | 1 + 8 files changed, 224 insertions(+), 51 deletions(-) diff --git a/web-app/src/app/routes/alert/alert-integration/alert-integration.component.ts b/web-app/src/app/routes/alert/alert-integration/alert-integration.component.ts index 0a4802b007..aef7601346 100644 --- a/web-app/src/app/routes/alert/alert-integration/alert-integration.component.ts +++ b/web-app/src/app/routes/alert/alert-integration/alert-integration.component.ts @@ -35,10 +35,15 @@ export class AlertIntegrationComponent implements OnInit { icon: 'assets/img/integration/prometheus.svg' }, { - id: 'tencent', - name: this.i18nSvc.fanyi('alert.integration.source.tencent'), - icon: 'assets/img/integration/tencent.svg' - } + id: 'alertmanager', + name: this.i18nSvc.fanyi('alert.integration.source.alertmanager'), + icon: 'assets/img/integration/prometheus.svg' + }, + // { + // id: 'tencent', + // name: this.i18nSvc.fanyi('alert.integration.source.tencent'), + // icon: 'assets/img/integration/tencent.svg' + // } ]; selectedSource: DataSource | null = null; diff --git a/web-app/src/assets/doc/alert-integration/webhook.en-US.md b/web-app/src/assets/doc/alert-integration/alertmanager.en-US.md similarity index 100% copy from web-app/src/assets/doc/alert-integration/webhook.en-US.md copy to web-app/src/assets/doc/alert-integration/alertmanager.en-US.md diff --git a/web-app/src/assets/doc/alert-integration/alertmanager.zh-CN.md b/web-app/src/assets/doc/alert-integration/alertmanager.zh-CN.md new file mode 100644 index 0000000000..a345e69d4f --- /dev/null +++ b/web-app/src/assets/doc/alert-integration/alertmanager.zh-CN.md @@ -0,0 +1,44 @@ +> 将 Prometheus AlertManager 的告警发送到 HertzBeat 告警平台。 + +### Alertmanager 配置 Webhook + +1. 在 Alertmanager 配置文件中添加 webhook 配置。 + + ```yaml + receivers: + - name: 'webhook' + webhook_configs: + - url: 'http://${hertzbeat_host}:1157/api/alerts/report/alertmanager' + http_config: + headers: + Authorization: 'Bearer {token}' + send_resolved: true + ``` +- `http://${hertzbeat_ip}:8080/api/alerts/report` 为 HertzBeat 提供的 webhook 接口地址。 +- `send_resolved: true` 表示发送告警恢复信息。 +- `Authorization` 内的 {token} 为 HertzBeat 提供的 token。 + +2. 重启 Alertmanager 服务。 + +### 配置验证 + +- 触发 Prometheus AlertManager 告警。 +- 在 HertzBeat 告警平台中对告警数据处理查看，验证告警数据是否正确。 + +### 数据流转: + +```mermaid +graph LR + A[Prometheus Alertmanager] --> B[Webhook] + B --> C[HertzBeat告警平台] + C --> D[分组收敛] + D --> E[告警抑制] + E --> F[告警静默] + F --> G[告警中心] + F --> H[消息分发] +``` + +### 常见问题 + +- 确保 Alertmanager 配置文件中的 webhook 地址正确，且网络通畅。 +- 检查 Alertmanager 的告警是否触发，是否发送到 HertzBeat 告警平台。 diff --git a/web-app/src/assets/doc/alert-integration/alertmanager.zh-TW.md b/web-app/src/assets/doc/alert-integration/alertmanager.zh-TW.md new file mode 100644 index 0000000000..e69de29bb2 diff --git a/web-app/src/assets/doc/alert-integration/prometheus.zh-CN.md b/web-app/src/assets/doc/alert-integration/prometheus.zh-CN.md index 50880bd187..8ae23270d2 100644 --- a/web-app/src/assets/doc/alert-integration/prometheus.zh-CN.md +++ b/web-app/src/assets/doc/alert-integration/prometheus.zh-CN.md @@ -1,32 +1,69 @@ -## 告警字段說明 +> 由于 Prometheus Server 本身并不支持 HTTP API 的告警发送，因此需要借助外部脚本或者 Alertmanager 来实现告警发送。 +> 若使用 Alertmanager，可参考 **Alertmanager 集成文档**。这里提供非部署 Alertmanager 的 Prometheus Server 的告警配置方法。 -- `alert`: 告警規則名稱 -- `expr`: 告警觸發條件表達式 -- `for`: 告警持續時間閾值 -- `labels`: 告警標籤 - - `severity`: 告警級別 (warning, critical) -- `annotations`: 告警註釋信息 - - `summary`: 告警摘要 - - `description`: 告警詳細描述 +### Prometheus 告警配置 -## 驗證配置 +1. 編輯 Prometheus 配置文件 `prometheus.yml`，添加告警規則配置 + ```yaml + rule_files: + - "rules/*.rules.yml" + ``` + > `rules/*.rules.yml` 為告警規則文件路徑，可以根據實際情況修改 +> 2. 創建告警規則文件夾 `rules`，並創建告警規則文件 `rules/*.rules.yml` +> 3. 編輯告警規則文件，添加告警規則配置 +> 4. 重新加載 Prometheus 配置 -1. 確保 Prometheus 配置正確並重新加載配置 - ```bash - curl -X POST http://localhost:9090/-/reload - ``` +### 编写脚本自动发送告警 -2. 檢查 Prometheus 告警規則狀態 - ```bash - curl http://localhost:9090/api/v1/rules - ``` +> 由于 Prometheus Server 本身并不支持 HTTP API 的告警发送，这里我们使用 Python 脚本来实现告警发送。 -3. 觸發測試告警並在 HertzBeat 告警中心查看 +1. 安装 Python requests 库 + ```bash + pip install requests + ``` +2. 编写 Python 脚本 `send_alerts.py` +```python +import requests -## 常見問題 +PROMETHEUS_URL = "http://<prometheus-host>:9090/api/v1/alerts" +WEBHOOK_URL = "http://<hertzbeat-host>:1157/api/alerts/report/prometheus" -1. 確保 HertzBeat URL 可以被 Prometheus 服務器訪問 -2. 檢查 Prometheus 日誌中是否有告警發送失敗的錯誤信息 -3. 驗證告警規則表達式的正確性 +def get_prometheus_alerts(): + response = requests.get(PROMETHEUS_URL) + alerts = response.json()["data"]["alerts"] + return alerts + +def send_to_webhook(alert): + requests.post(WEBHOOK_URL, json=alert) + +if __name__ == "__main__": + alerts = get_prometheus_alerts() + for alert in alerts: + send_to_webhook(alert) + +``` +3. 运行 Python 脚本 + ```bash + python send_alerts.py + ``` + > 该脚本会从 Prometheus Server 获取告警数据，并通过 Webhook 推送到 HertzBeat 告警平台。 + +## 验证配置 + +1. 确保 Prometheus 配置正确并重新加载配置 + ```bash + curl -X POST http://localhost:9090/-/reload + ``` +2. 检查 Prometheus 告警规则状态 + ```bash + curl http://localhost:9090/api/v1/rules + ``` +3. 触发测试告警并在 HertzBeat 告警中心查看 + +## 常见问题 + +- 确保 HertzBeat URL 可以被 Prometheus 服务器访问 +- 检查 Prometheus 日志中是否有告警发送失败的错误信息 +- 验证告警规则表达式的正确性 更多信息請參考 [Prometheus 告警配置文檔](https://prometheus.io/docs/alerting/latest/configuration/) diff --git a/web-app/src/assets/doc/alert-integration/webhook.en-US.md b/web-app/src/assets/doc/alert-integration/webhook.en-US.md index b876d8a96d..d80c736cde 100644 --- a/web-app/src/assets/doc/alert-integration/webhook.en-US.md +++ b/web-app/src/assets/doc/alert-integration/webhook.en-US.md @@ -1,7 +1,52 @@ -# Prometheus Alert Integration +HertzBeat provides an external API interface that allows external systems to push alert data to the HertzBeat alert platform via Webhook. -HertzBeat is fully compatible with Prometheus alert data format. You can configure Prometheus alerting rules to send alerts to HertzBeat. +--- -## Prometheus Alert Configuration +### API Endpoint -Add the following configuration to your Prometheus configuration file: +`POST /api/alerts/report` + +--- + +### Request Headers + +- `Content-Type`: `application/json` +- `Authorization`: `Bearer ${token}` + +--- + +### Request Body + +```json +{ + "labels": { + "alertname": "HighCPUUsage", + "priority": "critical", + "instance": "343483943" + }, + "annotations": { + "summary": "High CPU usage detected" + }, + "content": "The CPU usage on instance 343483943 is critically high.", + "status": "firing", + "triggerTimes": 3, + "startAt": "2025-01-10T12:00:00.000Z", + "activeAt": "2025-01-10T12:05:00.000Z", + "endAt": null +} +``` + +Field Description +• labels: Alert labels +• alertname: Name of the alert rule +• priority: Alert priority (warning, critical) +• instance: Alert instance +• annotations: Alert annotation information +• summary: Alert summary +• description: Detailed description of the alert +• content: Alert content +• status: Alert status (firing, resolved) +• triggerTimes: Number of times the alert was triggered +• startAt: Start time of the alert +• activeAt: Time when the alert became active +• endAt: End time of the alert (if resolved) diff --git a/web-app/src/assets/doc/alert-integration/webhook.zh-CN.md b/web-app/src/assets/doc/alert-integration/webhook.zh-CN.md index 50880bd187..ecc9d13abe 100644 --- a/web-app/src/assets/doc/alert-integration/webhook.zh-CN.md +++ b/web-app/src/assets/doc/alert-integration/webhook.zh-CN.md @@ -1,32 +1,73 @@ -## 告警字段說明 +> HertzBeat 对外提供 api 接口，外部系统可以通过 Webhook 方式调用此接口将告警数据推送到 HertzBeat 告警平台。 + +### 接口端点 + +`POST /api/alerts/report` + +### 请求头 + +- `Content-Type`: `application/json` +- `Authorization`: `Bearer ${token}` + +### 请求体 + +```json +{ + "labels": { + "alertname": "HighCPUUsage", + "priority": "critical", + "instance": "343483943" + }, + "annotations": { + "summary": "High CPU usage detected" + }, + "content": "The CPU usage on instance 343483943 is critically high.", + "status": "firing", + "triggerTimes": 3, + "startAt": "2025-01-10T12:00:00.000Z", + "activeAt": "2025-01-10T12:05:00.000Z", + "endAt": null +} +``` + +字段說明 -- `alert`: 告警規則名稱 -- `expr`: 告警觸發條件表達式 -- `for`: 告警持續時間閾值 - `labels`: 告警標籤 - - `severity`: 告警級別 (warning, critical) + - `alertname`: 告警規則名稱 + - `priority`: 告警級別 (warning, critical) + - `instance`: 告警實例 - `annotations`: 告警註釋信息 - `summary`: 告警摘要 - `description`: 告警詳細描述 +- `content`: 告警內容 +- `status`: 告警狀態 (firing, resolved) +- `triggerTimes`: 告警觸發次數 +- `startAt`: 告警開始時間 +- `activeAt`: 告警激活時間 +- `endAt`: 告警結束時間 + + +### 配置验证 -## 驗證配置 +- 第三方系统触发告警后通过 webhook 回调 HertzBeat 的 `/api/alerts/report` 接口，将告警数据推送到 HertzBeat 告警平台。 +- 在 HertzBeat 告警平台中对告警数据处理查看，验证告警数据是否正确。 -1. 確保 Prometheus 配置正確並重新加載配置 - ```bash - curl -X POST http://localhost:9090/-/reload - ``` -2. 檢查 Prometheus 告警規則狀態 - ```bash - curl http://localhost:9090/api/v1/rules - ``` +### 数据流转: -3. 觸發測試告警並在 HertzBeat 告警中心查看 +```mermaid +graph LR + A[外部系统告警] --> B[Webhook] + B --> C[HertzBeat告警平台] + C --> D[分组收敛] + D --> E[告警抑制] + E --> F[告警静默] + F --> G[告警中心] + F --> H[消息分发] +``` -## 常見問題 -1. 確保 HertzBeat URL 可以被 Prometheus 服務器訪問 -2. 檢查 Prometheus 日誌中是否有告警發送失敗的錯誤信息 -3. 驗證告警規則表達式的正確性 +### 常见问题 -更多信息請參考 [Prometheus 告警配置文檔](https://prometheus.io/docs/alerting/latest/configuration/) +- 确保 HertzBeat URL 可以被第三方系统服务器访问。 +- 检查第三方系统日志中是否有告警发送成功失败的消息。 diff --git a/web-app/src/assets/i18n/zh-CN.json b/web-app/src/assets/i18n/zh-CN.json index 5320cecaa1..b65bfa25bd 100644 --- a/web-app/src/assets/i18n/zh-CN.json +++ b/web-app/src/assets/i18n/zh-CN.json @@ -382,6 +382,7 @@ "alert.integration.source": "集成告警源", "alert.integration.source.webhook": "默认Webhook", "alert.integration.source.prometheus": "Prometheus", + "alert.integration.source.alertmanager": "Alertmanager", "alert.integration.source.tencent": "腾讯云监控", "dashboard.alerts.title": "最近告警列表", "dashboard.alerts.title-no": "最近未处理告警", --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
