Lakehouse Admin
df5ec93dc3
- Trino catalogs, Grafana, Spark jobs, LDAP LDIF, NPM compose - Airflow DAG scripts, Proxmox VM inventory, network docs - Ansible playbook, Gitea CI validate workflow - Backup and health-check scripts, cron documentation - Homepage DOCS tab with links to all documentation - Extended collect-fleet-config.sh and populate-repo.py
58 lines
1.4 KiB
Python
58 lines
1.4 KiB
Python
from kafka import KafkaConsumer
|
|
from kafka import TopicPartition
|
|
import boto3
|
|
import json
|
|
from datetime import datetime
|
|
|
|
# Configuratie
|
|
KAFKA_BROKER = "10.0.21.36:9092"
|
|
KAFKA_TOPIC = "test-lakehouse"
|
|
S3_ENDPOINT = "http://10.0.20.111:9020"
|
|
S3_BUCKET = "data"
|
|
S3_ACCESS_KEY = "REDACTED"
|
|
S3_SECRET_KEY = "REDACTED"
|
|
|
|
print("Starting S3 consumer...")
|
|
print(f"Kafka: {KAFKA_BROKER}, Topic: {KAFKA_TOPIC}")
|
|
print(f"S3: {S3_ENDPOINT}, Bucket: {S3_BUCKET}")
|
|
|
|
# S3 client
|
|
s3 = boto3.client(
|
|
's3',
|
|
endpoint_url=S3_ENDPOINT,
|
|
aws_access_key_id=S3_ACCESS_KEY,
|
|
aws_secret_access_key=S3_SECRET_KEY,
|
|
use_ssl=False,
|
|
verify=False
|
|
)
|
|
|
|
# Kafka consumer met vaste partition
|
|
consumer = KafkaConsumer(bootstrap_servers=[KAFKA_BROKER])
|
|
tp = TopicPartition(KAFKA_TOPIC, 0)
|
|
consumer.assign([tp])
|
|
consumer.seek_to_beginning()
|
|
|
|
print("Reading all messages from beginning...")
|
|
print("-" * 50)
|
|
|
|
msg_count = 0
|
|
for msg in consumer:
|
|
msg_count += 1
|
|
timestamp = datetime.now().strftime('%Y%m%d_%H%M%S_%f')
|
|
key = f"kafka/{KAFKA_TOPIC}/message_{timestamp}.json"
|
|
|
|
try:
|
|
data = json.loads(msg.value.decode('utf-8'))
|
|
except:
|
|
data = {"raw": msg.value.decode('utf-8')}
|
|
|
|
s3.put_object(
|
|
Bucket=S3_BUCKET,
|
|
Key=key,
|
|
Body=json.dumps(data, indent=2)
|
|
)
|
|
print(f"[{msg_count}] Written: s3://{S3_BUCKET}/{key}")
|
|
print(f" Data: {str(data)[:80]}...")
|
|
|
|
print(f"Done! Total {msg_count} messages written to S3")
|