Feat: Airflow uses kubernetes_executor as default executor, git sync …

…dags and add tutorials (#109)

Signed-off-by: xingcan-ltc <[email protected]>

catalog/airflow/apps/airflow.app/README.md

@@ -12,56 +12,34 @@ Airflow 通过KDP web部署。
 
 ##### 2.2.1 创建DAG文件
 
-以下为一个简单的 DAG 文件，将以下内容复制到文件 demo.py 中。
-
-```python
-from datetime import datetime
-
-from airflow import DAG
-from airflow.decorators import task
-from airflow.operators.bash import BashOperator
-
-with DAG(dag_id="demo", start_date=datetime(2024, 5, 20), schedule="0 0 * * *") as dag:
-
-    hello = BashOperator(task_id="hello", bash_command="echo hello")
-
-    @task()
-    def airflow():
-        print("airflow")
-
-    hello >> airflow()
-```
-
-##### 2.2.2 将 demo.py 文件通过kubectl 拷贝到airflow scheduler和worker pod 中
-
-```shell
-kubectl cp demo.py airflow-scheduler-7cf5ddb9c5-fql6x:/opt/airflow/dags -n kdp-data
-kubectl cp demo.py airflow-worker-584b97f6cb-8gxq8:/opt/airflow/dags -n kdp-data
-```
-
-注意：使用时请修改 `kdp-data` 为你的 namespace，修改 `airflow-scheduler-7cf5ddb9c5-fql6x` 为你的 scheduler pod 名称，修改 `airflow-worker-584b97f6cb-8gxq8` 为你的 worker pod 名称。
-
 ##### 2.2.3 浏览器访问airflow web
 
-可通过配置的ingress地址访问airflow web，或者通过kubectl port-forward 访问，以下为port-forward命令。
+可通过配置的ingress(http://airflow-web-kdp-data.kdp-e2e.io/home)地址访问airflow web，或者通过kubectl port-forward 访问，以下为port-forward命令。
 
 ```shell
 kubectl port-forward svc/airflow-webserver -n kdp-data 8080:8080
 ```
 
 默认登陆用户/密码为 `admin/admin`
 
-#### 2.2.4 查看DAG以及任务执行
+#### 2.2.4 配置DAG文件
+
+DAG文件存放在git仓库中，默认安装配置的dag文件存放在git仓库中，你可以通过修改该文件来修改DAG文件。
+你也可以自己fork该仓库，然后修改dags文件，然后提交到git仓库中, 在 KDP 页面安装配置修改 dag repo， branch 等，然后更新即可。
+
+#### 2.2.4 运行DAG
+
+DAG 默认是暂停状态，需要手动启动。手动激活（点击名称旁边的开关即可）名称`hello_airflow`的DAG即可，该DAG是每天运行一次，激活后会自动补跑昨天的任务。
+也可以手动触发：点击`hello_airflow`DAG右边`Trigger DAG`按钮即可。
 
-当前配置的scheduler扫描频率为1分钟，在web页面可见demo DAG，点击右侧 `Trigger DAG` 按钮，即可触发DAG执行。
 
 ### 3. 常见问题自检
 
 #### 3.1. DAG执行失败
 
 原因与排查：
 
-1. 检查scheduler和worker pod 中 `/opt/airflow/dags` 目录下是否均存在 demo.py 文件；
+1. 检查 dag 代码同步是否成功，检查日志： `kubectl logs -l component=scheduler,release=airflow -c git-sync -n kdp-data`
 2. 查看scheduler和worker pod 日志输出信息。
 
 ### 4. 附录

catalog/airflow/apps/airflow.app/app.yaml

@@ -10,8 +10,6 @@ metadata:
 spec:
   type: airflow
   properties:
-    chartVersion: v1.1.0-1.14.0
-    images:
-      airflow:
-        tag: 2.9.1
+    scheduler:
+      replicas: 1
 

catalog/airflow/apps/airflow.app/i18n/en/README.md

@@ -1,74 +1,49 @@
 ### 1. Application Description
 
-Airflow™ is a platform created by the community to programmatically author, schedule and monitor workflows.
+Airflow™ is a platform for programmatically authoring, scheduling, and monitoring workflows.
 
 ### 2. Quick Start
 
-#### 2.1 Deploy
+#### 2.1 Deployment
 
-Airflow is deployed by using KDP web.
+Airflow is deployed via the KDP web interface.
 
-#### 2.2. Practice
+#### 2.2 Practical Usage
 
-##### 2.2.1 Create DAG file
+##### 2.2.1 Creating DAG Files
 
-Here is a simple DAG file, copy the following content to the demo.py file.
+##### 2.2.3 Accessing Airflow Web via Browser
 
-```python
-from datetime import datetime
-
-from airflow import DAG
-from airflow.decorators import task
-from airflow.operators.bash import BashOperator
-
-with DAG(dag_id="demo", start_date=datetime(2024, 5, 20), schedule="0 0 * * *") as dag:
-
-    hello = BashOperator(task_id="hello", bash_command="echo hello")
-
-    @task()
-    def airflow():
-        print("airflow")
-
-    hello >> airflow()
-```
-
-##### 2.2.2 Copy DAG file to airflow scheduler and worker pod
+You can access the Airflow web interface through the configured ingress (http://airflow-web-kdp-data.kdp-e2e.io/home) or by using kubectl port-forward, as shown in the following command:
 
 ```shell
-kubectl cp demo.py airflow-scheduler-7cf5ddb9c5-fql6x:/opt/airflow/dags -n kdp-data
-kubectl cp demo.py airflow-worker-584b97f6cb-8gxq8:/opt/airflow/dags -n kdp-data
+kubectl port-forward svc/airflow-webserver -n kdp-data 8080:8080
 ```
 
-Note: When using this example, please modify `kdp-data` to your namespace, modify `airflow-scheduler-7cf5ddb9c5-fql6x` to your scheduler pod name, modify `airflow-worker-584b97f6cb-8gxq8` to your worker pod name.
-
-##### 2.2.3 Visit airflow web
+The default login username/password is admin/admin.
 
-Visit airflow web by ingress address, or using kubectl port-forward, as follows:
 
-```shell
-kubectl port-forward svc/airflow-webserver -n kdp-data 8080:8080
-```
+### 2.2.4 Configuring DAG Files
 
-Default login user/password is `admin/admin`
+DAG files are stored in a Git repository. The default installation configuration places the DAG files in a Git repository, which you can modify to change the DAG files. Alternatively, you can fork the repository, modify the DAG files, and then commit them to the Git repository. You can also install and configure the DAG repository, branch, etc., on the KDP page and then update it.
 
-#### 2.2.4 View DAG and task execution
+### 2.2.4 Running DAGs
 
-The current configured scheduler scanning frequency is 1 minute, and the demo DAG can be seen on the web page. Click the 'Trigger DAG' button on the right to trigger DAG execution.
+DAGs are set to a paused state by default and need to be manually started. Manually activate the DAG named `hello_airflow` by clicking the switch next to its name. This DAG runs once a day and will automatically catch up on yesterday's tasks after activation. You can also manually trigger it by clicking the `Trigger DAG` button on the right side of the `hello_airflow` DAG.
 
 ### 3. FAQ
 
-#### 3.1. DAG execution failed
+#### 3.1. DAG Execution Failure
 
-Reasons and results:
+Causes and Troubleshooting:
 
-1. Check whether the demo.py file exists in the `/opt/airflow/dags` directory of the scheduler and worker pod;
-2. View the output information of scheduler and worker pod logs.
+- Check if the DAG code synchronization is successful by checking the logs: `kubectl logs -l component=scheduler,release=airflow -c git-sync -n kdp-data`
+- Review the log output information for the scheduler and worker pods.
 
 ### 4. Appendix
 
-#### 4.1. Concept
+#### 4.1. Concept Introduction
 
 **DAG**
 
-A DAG (Directed Acyclic Graph) is the core concept of Airflow, collecting Tasks together, organized with dependencies and relationships to say how they should run.
-
+Directed Acyclic Graph, which is the basic unit for describing workflows in Airflow.

catalog/airflow/x-definitions/app-airflow.cue

@@ -1,9 +1,4 @@
-import (
-	"strconv"
-	"strings"
-)
-
-import ("encoding/json")
+import "strings"
 
 airflow: {
 	annotations: {}
@@ -65,6 +60,16 @@ template: {
 						values: {
 							defaultAirflowRepository: _imageRegistry + "apache/airflow"
 							defaultAirflowTag:        parameter.images.airflow.tag
+							executor:                 "KubernetesExecutor"
+							dags: {
+								gitSync: {
+									enabled: true
+									repo:    parameter.dags.gitSync.repo
+									branch:  parameter.dags.gitSync.branch
+									rev:     parameter.dags.gitSync.rev
+									subPath: parameter.dags.gitSync.subPath
+								}
+							}
 							config: {
 								core: {
 									"default_timezone": "Asia/Shanghai"
@@ -198,10 +203,6 @@ template: {
 									enabled: false
 								}
 							}
-							redis: {
-								enabled:                       true
-								terminationGracePeriodSeconds: 30
-							}
 							ingress: {
 								web: {
 									enabled: true
@@ -210,18 +211,17 @@ template: {
 											name: "airflow-web-" + context.namespace + "." + context["ingress.root_domain"]
 										},
 									]
-									ingressClassName: "kong"
 								}
 							}
 							images: {
 								statsd: {
 									repository: _imageRegistry + "prometheus/statsd-exporter"
-									tag:        "v0.26.1"
 								}
-								redis: {
-									repository: _imageRegistry + "redis"
-									tag:        "7-bookworm"
+								gitSync: {
+									repository: _imageRegistry + "git-sync/git-sync"
+									tag:        "v4.2.3"
 								}
+
 							}
 						}
 					}
@@ -246,8 +246,29 @@ template: {
 				mysqlSecret: string
 			}
 		}
-		// +ui:title=Webserver
+
+		// ui:title=DAG 配置
 		// +ui:order=2
+		dags: {
+			// +ui:description=git 同步配置 (必须配置)
+			gitSync: {
+				// +ui:description=git 仓库地址。 备用镜像仓库： https://gitee.com/linktime-cloud/example-datasets.git
+				// +ui:order=1
+				repo: *"https://github.com/linktimecloud/example-datasets.git" | string
+				// +ui:description=git 仓库分支
+				// +ui:order=2
+				branch: *"airflow" | string
+				// +ui:description=git 仓库提交 ID 
+				// +ui:order=3
+				rev: *"HEAD" | string
+				// +ui:description= DAG 代码目录 (根目录请使用空字符串)
+				// +ui:order=4
+				subPath: *"dags" | string
+			}
+		}
+
+		// +ui:title=Webserver
+		// +ui:order=3
 		webserver: {
 			// +ui:description=资源规格
 			// +ui:order=1
@@ -288,7 +309,7 @@ template: {
 			replicas: *1 | int
 		}
 		// +ui:title=Scheduler
-		// +ui:order=3
+		// +ui:order=4
 		scheduler: {
 			// +ui:description=资源规格
 			// +ui:order=1
@@ -329,7 +350,7 @@ template: {
 			replicas: *1 | int
 		}
 		// +ui:title=Workers
-		// +ui:order=4
+		// +ui:order=5
 		workers: {
 			// +ui:description=资源规格
 			// +ui:order=1
@@ -380,7 +401,7 @@ template: {
 		images: {
 			airflow: {
 				// +ui:options={"disabled":true}
-				tag: *"2.9.1" | string
+				tag: *"v1.0.0-2.9.1" | string
 			}
 		}
 	}

docs/en/user-tutorials/batch-job-scheduling-for-hive-sql-with-apache-airflow.md

@@ -0,0 +1,51 @@
+
+# Batch Job Scheduling for Hive SQL with Apache Airflow
+
+# 1. Introduction
+Apache Airflow is an open-source platform for orchestrating and automating batch jobs, allowing for the easy creation, scheduling, monitoring, and management of workflows. Airflow supports Hive SQL, enabling the effortless execution of Hive SQL tasks.
+
+Apache Airflow utilizes Directed Acyclic Graphs (DAGs) to represent workflows, which consist of task nodes and dependencies. Task nodes can be Python operations, Shell operations, SQL operations, and more. Airflow supports various schedulers, including the Local scheduler, Celery scheduler, Kubernetes scheduler, and others.
+
+This article introduces how to write Hive SQL tasks using `pyhive` and execute them with the Apache Airflow Kubernetes scheduler.
+
+# 2. Writing a Hive SQL DAG
+
+The specific code implementation can be accessed on [Github](https://github.com/linktimecloud/example-datasets/blob/airflow/dags/hive-sql-example.py) or [Gitee](https://gitee.com/linktime-cloud/example-datasets/blob/airflow/dags/hive-sql-example.py).
+
+This code is a DAG (Directed Acyclic Graph) written using the Apache Airflow framework, designed for automating data processing tasks. It primarily performs two tasks: creating a Hive table and inserting data, followed by identifying the top-scoring students in each subject.
+
+# 3. Running the DAG
+## 3.1 Component Dependencies
+The following components need to be installed in KDP:
+- mysql
+- airflow
+- zookeeper
+- hdfs
+- hive (hive metastore, hive server)
+- hue, httpfs-gateway (optional)
+
+## 3.2 Scheduling Jobs
+After installing Airflow with default parameters in KDP, log in to the Airflow Web interface using the username `admin` and password `admin`.
+
+Start the DAG named `hive-sql-example`.
+
+![Airflow Web Interface](./images/airflow01.png)
+
+Upon successful execution, the results can be viewed through the Hue Web interface. Alternatively, you can refer to the `hive-server2` Quick Start guide to connect to Hive Server2 using beeline and view the results.
+
+![Hue Web Interface](./images/airflow02.png)
+
+
+运行成功后，可以通过Hue Web界面查看结果。也可以参考 `hive-server2` Qucick Start 使用beeline 连接 Hive Server2 查看结果。
+
+![](./images/airflow02.png)
+
+
+
+
+
+
+
+
+
+

docs/en/user-tutorials/tutorials.md

@@ -19,4 +19,6 @@ Users can refer to the following scenario tutorials to practice how to do data i
   * [How to integrate with Kafka on KDP](./integration-kafka-with-int-ext-comps.md)
   * [Data Import from External HDFS to KDP HDFS](./import-from-hdfs-to-hdfs.md)
   * [Exploring data using Airbyte/ClickHouse/Superset](./exploring-data-using-airbyte-clickhouse-superset.md)
-  * More...
+  * [Batch Job Scheduling for Hive SQL with Apache Airflow](./batch-job-scheduling-for-hive-sql-with-apache-airflow.md)
+  * More...
+