Update 01-08-23. Fix some bugs (377, 378)

README.md

@@ -456,8 +456,36 @@ Listo. Con esto se ha configurado todos los servicios requeridos para levantar n
 
 ### Desplegando Airflow con GKE
 
+⚠️ Ojo: Antes de desplegar nuestros componentes, hay que asegurarnos que cumple con los
+recursos de nuestro clúster (en cuestión de vCPU y memoria RAM), para esto hay que validar
+los templates que son de tipo despliegue (deployment), estos son los que en su nombre termina
+con la palabra _"deployment"_: "airflow-scheduler-deployment,yaml", "airflow-webserver-deployment.yaml",
+"postgres-deployment.yaml" y "sync-dags-deployment-gcs" (para este caso que estamos en GCP).
 
-1.- 
+Además del archivo "airflow-envvars-configmap.yaml" que contiene los recursos para los workers.
 
+Todos los templates mencionados tienen sus parámetros "containers.resources.limits" o 
+"containers.resources.requests" las cuales indican el máximo y mínimo, respectivamente.
 
 
+1.- Luego de tener ya configurado nuestro Bucket y NFS (Filestore), podemos desplegar nuestro sistema
+Airflow con Kubernetes. Para esto nos situamos dentro de la carpeta **"deploy-k8"** y ejecutamos
+el siguiente comando:
+
+```shell
+sh script-apply.sh
+```
+
+Esto arrancará todos los componentes y dentro de unos segundos o minutos estará todo arriba,
+listo para su uso.
+
+2.- Para validar que están todos los componentes arriba, se ejecuta el siguiente comando:
+
+```shell
+kubectl get pods -n bcom-airflow
+```
+
+Deberá mostrarse algo similar a esto:
+
+![image](readme-images/componentes-ready1.png)
+

dags/components/Extractor.py

@@ -4,6 +4,7 @@ import json
 import numpy as np
 import pandas as pd
 from enums.ProcessStatusEnum import ProcessStatusEnum
+from enums.DatabaseTypeEnum import DatabaseTypeEnum
 from components.Utils import select_multiple, generateModel
 from components.DatabaseOperation.DatabaseExtraction import get_iterator, get_steps
 from components.DatabaseOperation.DatabaseLoad import save_from_dataframe

dags/components/Generation.py

@@ -8,6 +8,7 @@ from airflow.decorators import task
 from airflow.exceptions import AirflowSkipException
 
 from enums.ProcessStatusEnum import ProcessStatusEnum
+from enums.DatabaseTypeEnum import DatabaseTypeEnum
 from components.S3Route import save_df_to_s3, load_control_to_s3
 from components.Utils import select_multiple, create_temp_file, delete_temp_dir
 from components.Control import get_tasks_from_control, update_new_process

dags/components/S3Route.py

@@ -4,6 +4,7 @@ from typing import Any, Dict, List, Tuple
 import pytz
 from io import BytesIO, StringIO
 import pandas as pd
+import re
 
 from components.Utils import get_type_file
 from enums.FileTypeEnum import FileTypeEnum
@@ -181,6 +182,7 @@ def get_file_from_prefix(conn: str, bucket: str, key: str, provider: str, timezo
                          frequency: str = "montly") -> Any:
     result, key_result = BytesIO(), ''
     try:
+        format_re_pattern = '[0-9]{4}-[0-9]{2}\.'
         format_date = "%Y-%m" if frequency == "montly" else "%Y-%W"
         period = str(datetime_by_tzone(timezone, format_date))[:7]
         logger.info(f"Periodo actual: {period}.")
@@ -197,7 +199,7 @@ def get_file_from_prefix(conn: str, bucket: str, key: str, provider: str, timezo
             files = gcp_hook.list(bucket, prefix=key)
         files_with_period = []
         for file in files:
-            if file.endswith("/"):
+            if file.endswith("/") or not re.search(format_re_pattern, file):
                 continue
             file_period = file[file.rfind("_") + 1:file.rfind(".")]
             files_with_period.append((file, file_period))

dags/components/Utils.py

@@ -107,7 +107,7 @@ def update_sql_commands(dataset: List[Tuple[str, str]], label_tablename: str) ->
                     final_data[-1] = final_data[-1] + "; end;"
                 final_item = item
                 if item.lower().strip().find(label_tablename.lower().strip() + ":") != -1:
-                    init_index = item.lower().strip().index(label_tablename.lower().strip() + ":")
+                    init_index = item.replace(" ", "").lower().strip().index(label_tablename.lower().strip() + ":")
                     table_name = item.replace(" ", "").strip()[init_index + len(label_tablename + ":"):].strip()
                     add_next = True
                 elif item != "":

dags/dag_conf.yml

@@ -10,7 +10,7 @@ app:
         port: 3306
         username: admin
         password: adminadmin
-        database: prueba_bcom
+        database: prueba_ca_1
         service: ORCLPDB1
         schema: sources
     transformation:
@@ -19,7 +19,7 @@ app:
       port: 3306
       username: admin
       password: adminadmin
-      database: prueba_bcom2
+      database: prueba_ca_2
       service:
       schema: intern_db
   chunksize: 8000
@@ -28,16 +28,16 @@ app:
   procedure_mask: procedure  # S
   transformation_mask: transform  # S
   prefix_order_delimiter: .
-  cloud_provider: google
+  cloud_provider: aws
   scripts:
     s3_params:
-      bucket: prueba-airflow3
+      bucket: prueba-airflow13
       prefix: bcom_scripts
       connection_id: conn_script
   control:
     s3_params:
       connection_id: conn_script
-      bucket: prueba-airflow3
+      bucket: prueba1234568
       prefix: bcom_control
       filename: control_<period>.json
   timezone: 'GMT-5'
@@ -48,12 +48,12 @@ app:
     delimiter: '|'
     tmp_path: /tmp
     s3_params:
-      bucket: prueba-airflow3
+      bucket: prueba-airflow13
       prefix: bcom_results
       connection_id: conn_script
   report:
     s3_params:
-      bucket: prueba-airflow3
+      bucket: prueba1234568
       prefix: bcom_report
       connection_id: conn_script
       filename: report_<datetime>.xlsx

dags/dag_inform_process.py

@@ -76,7 +76,7 @@ def create_report(tmp_path: str, **kwargs) -> None:
             title_format.set_font_size(20)
             title_format.set_font_color("#333333")
 
-            header = f"Reporte ejecutado el día {execution_date}"
+            header = f"Proceso ejecutado el día {execution_date}"
             if status == ProcessStatusEnum.SUCCESS.value:
                 status = "EXITOSO"
             elif status == ProcessStatusEnum.FAIL.value:
@@ -95,7 +95,7 @@ def create_report(tmp_path: str, **kwargs) -> None:
             row_format = workbook.add_format()
             row_format.set_font_size(8)
             row_format.set_font_color("#000000")
-
+            if status != ProcessStatusEnum.RESET.value:
                 base_index = 5
                 for index, key in enumerate(data.keys()):
                     index = base_index + index
@@ -125,7 +125,8 @@ def get_data_report(**kwargs) -> None:
         else:
             last_process = control[-1]
             if "reset_by_user" in last_process.keys():
-                report_data["PROCESS_EXECUTION"] = ProcessStatusEnum.RESET.value
+                report_data["PROCESS_STATUS"] = ProcessStatusEnum.RESET.value
+                report_data["PROCESS_EXECUTION"] = last_process["date"]
             else:
                 total_tasks = [last_process["tasks"]]
                 current_status = last_process["status"]

deploy-k8/airflow-envvars-configmap.yaml

-apiVersion: v1
-kind: Namespace
-metadata:
-  name: bcom-airflow
-
----
-apiVersion: v1
-kind: ServiceAccount
-metadata:
-  name: bcom-airflow
-  namespace: bcom-airflow
-    
----
+#apiVersion: v1
+#kind: Namespace
+#metadata:
+#  name: bcom-airflow
+#
+#---
+#apiVersion: v1
+#kind: ServiceAccount
+#metadata:
+#  name: bcom-airflow
+#  namespace: bcom-airflow
+#
+#---
 
 apiVersion: v1
 kind: ConfigMap
@@ -36,6 +36,10 @@ data:
               value: LocalExecutor
           image: dumy-image
           imagePullPolicy: IfNotPresent
+          resources:
+            limits:
+              cpu: "1000m"
+              memory: "2Gi"
           name: base
           volumeMounts:
             - name: dags-host-volume
@@ -77,19 +81,20 @@ data:
   AIRFLOW__CORE__DEFAULT_TIMEZONE: America/Lima
   AIRFLOW__KUBERNETES__KUBE_CLIENT_REQUEST_ARGS: '{"_request_timeout": [60,60]}'
   AIRFLOW__KUBERNETES__WORKER_CONTAINER_REPOSITORY: cristianfernando/airflow_custom
-  AIRFLOW__KUBERNETES__WORKER_CONTAINER_TAG: "0.0.5"
+  AIRFLOW__KUBERNETES__WORKER_CONTAINER_TAG: "0.0.6"
   AIRFLOW__KUBERNETES__LOGS_VOLUME_CLAIM: airflow-logs-pvc
   AIRFLOW__KUBERNETES__ENV_FROM_CONFIGMAP_REF: airflow-envvars-configmap
   AIRFLOW__KUBERNETES_EXECUTOR__POD_TEMPLATE_FILE: /opt/airflow/templates/pod_template.yaml
-  AIRFLOW__CORE__EXECUTOR: KubernetesExecutor
+  AIRFLOW__CORE__EXECUTOR: LocalExecutor
   AIRFLOW__DATABASE__SQL_ALCHEMY_CONN: postgresql+psycopg2://airflow:airflow@postgres/airflow
   AIRFLOW__CORE__DAGS_ARE_PAUSED_AT_CREATION: 'true'
-  AIRFLOW__CORE__LOAD_EXAMPLES: 'true'
+  AIRFLOW__CORE__LOAD_EXAMPLES: 'false'
   _AIRFLOW_DB_UPGRADE: 'true'
   _AIRFLOW_WWW_USER_CREATE: 'true'
   _AIRFLOW_WWW_USER_USERNAME: admin
   _AIRFLOW_WWW_USER_PASSWORD: admin
   S3_DAGS_DIR: 's3://prueba1234568/dags'
+  GCS_DAGS_DIR: 'gs://prueba-rsync2/carpeta'
   SYNCHRONYZE_DAG_DIR: '30'
   MINIO_SERVER: 'http://192.168.49.2:9000'
   MINIO_DAGS_DIR: '/prueba-ca/dags'
\ No newline at end of file

deploy-k8/airflow-scheduler-deployment.yaml

@@ -22,6 +22,10 @@ spec:
       containers:
       - name: airflow-scheduler
         image: cristianfernando/airflow_custom:0.0.4
+        resources:
+          requests:
+            cpu: "1000m"
+            memory: "4Gi"
         args: ["scheduler"]
         envFrom:
           - configMapRef:

deploy-k8/airflow-volumes.yaml

@@ -5,13 +5,13 @@ metadata:
   namespace: bcom-airflow
 spec:
   capacity:
-    storage: 300Mi
+    storage: 5Gi
   accessModes:
     - ReadWriteMany
   storageClassName: airflow-dags
   nfs:
-    server: 192.168.1.9
-    path: "/mnt/nfs_share"
+    server: 10.216.137.186
+    path: "/volume1/nfs_share"
 
 ---
 
@@ -22,13 +22,13 @@ metadata:
   namespace: bcom-airflow
 spec:
   capacity:
-    storage: 8000Mi
+    storage: 16Gi
   accessModes:
     - ReadWriteMany
   storageClassName: airflow-postgres
   nfs:
-    server: 192.168.1.9
-    path: "/mnt/nfs_postgres"
+    server: 10.216.137.186
+    path: "/volume1/nfs_postgres"
 
 ---
 
@@ -39,13 +39,13 @@ metadata:
   namespace: bcom-airflow
 spec:
   capacity:
-    storage: 4000Mi
+    storage: 10Gi
   accessModes:
     - ReadWriteMany
   storageClassName: airflow-logs
   nfs:
-    server: 192.168.1.9
-    path: "/mnt/nfs_logs"
+    server: 10.216.137.186
+    path: "/volume1/nfs_logs"
 
 ---
 
@@ -60,7 +60,7 @@ spec:
   storageClassName: airflow-dags
   resources:
     requests:
-      storage: 200Mi
+      storage: 5Gi
 
 ---
 
@@ -75,7 +75,7 @@ spec:
   storageClassName: airflow-postgres
   resources:
     requests:
-      storage: 7500Mi
+      storage: 16Gi
 
 ---
 
@@ -91,5 +91,5 @@ spec:
     - ReadWriteMany
   resources:
     requests:
-      storage: 3500Mi
+      storage: 10Gi
   storageClassName: airflow-logs
\ No newline at end of file

deploy-k8/airflow-webserver-deployment.yaml

@@ -22,6 +22,10 @@ spec:
       containers:
       - name: airflow-webserver
         image: apache/airflow:2.5.3
+        resources:
+          requests:
+            cpu: "500m"
+            memory: "500Mi"
         args: ["webserver"]
         envFrom:
           - configMapRef:

deploy-k8/postgres-deployment.yaml

@@ -21,8 +21,8 @@ spec:
         image: postgres:12
         resources:
           limits:
-            memory: 128Mi
-            cpu: 500m
+            memory: "2Gi"
+            cpu: "500m"
         ports:
         - containerPort: 5432        
         env:

deploy-k8/script-apply.sh

@@ -7,4 +7,4 @@ kubectl apply -f airflow-secrets.yaml
 kubectl apply -f airflow-webserver-deployment.yaml
 kubectl apply -f airflow-webserver-service.yaml
 kubectl apply -f airflow-scheduler-deployment.yaml
-kubectl apply -f sync-dags-deployment.yaml
+kubectl apply -f sync-dags-deployment-gcs.yaml

deploy-k8/script-delete.sh

@@ -5,6 +5,6 @@ kubectl delete -f airflow-secrets.yaml
 kubectl delete -f airflow-webserver-service.yaml
 kubectl delete -f airflow-webserver-deployment.yaml
 kubectl delete -f airflow-scheduler-deployment.yaml
-kubectl delete -f sync-dags-deployment.yaml
+kubectl delete -f sync-dags-deployment-gcs.yaml
 kubectl delete -f airflow-volumes.yaml
 kubectl delete -f airflow-envvars-configmap.yaml
\ No newline at end of file

deploy-k8/sync-dags-deployment-gcs.yaml

@@ -14,9 +14,12 @@ spec:
         app: airflow-sync-dags
 
     spec:
+      serviceAccountName: bcom-airflow
+      nodeSelector:
+        iam.gke.io/gke-metadata-server-enabled: "true"
       containers:
         - args:
-            - while true; gcloud rsync -d -r ${GCS_DAGS_DIR:-gs://prueba-rsync/carpeta} /dags; 
+            - while true; gsutil rsync -d -r ${GCS_DAGS_DIR:-gs://prueba-rsync2/carpeta} /dags; 
               do sleep ${SYNCHRONYZE_DAG_DIR:-30}; done;
           command:
             - /bin/bash
@@ -24,20 +27,13 @@ spec:
             - --
           name: sync-dags-gcloud
           image: gcr.io/google.com/cloudsdktool/google-cloud-cli:alpine
+          resources:
+            limits:
+              cpu: "250m"
+              memory: "1Gi"
           envFrom:
             - configMapRef:
                 name: airflow-envvars-configmap
-          env:
-            - name: AWS_ACCESS_KEY_ID
-              valueFrom:
-                secretKeyRef:
-                  key: AWS_ACCESS_KEY
-                  name: credentials
-            - name: AWS_SECRET_ACCESS_KEY
-              valueFrom:
-                secretKeyRef:
-                  key: AWS_SECRET_KEY
-                  name: credentials
           volumeMounts:
           - name: dags-host-volume
             mountPath: /dags