Logstash: 데이터 수집과 처리

Welcome to Logstash! In the ELK Stack, Logstash is your powerful, open-source data processing pipeline.

It's designed to ingest data from various sources, transform it, and then send it to various destinations, most commonly Elasticsearch.

Logstash operates on a simple, yet powerful, pipeline concept:

• Inputs: Where data enters Logstash.
• Filters: Where data is processed and transformed.
• Outputs: Where processed data is sent.

Think of it as an assembly line for your logs!

Inputs are the first stage of the pipeline. They define how Logstash collects data. There are many input plugins available for different sources.

Common input types include:

• file: For reading log files from disk.
• beats: For receiving data from Elastic Beats (like Filebeat).
• tcp/udp: For network-based log collection.

Here's a simple Logstash configuration snippet that sets up a file input plugin to read logs from a specific path.

This tells Logstash to monitor /var/log/my_app.log for new lines.

Filters are the heart of Logstash's processing power. They take raw, unstructured data and transform it into structured, machine-readable events.

This is crucial for making your logs searchable and analyzable in Elasticsearch.

The grok filter is incredibly powerful for parsing unstructured log data. It uses regular expressions combined with predefined patterns to extract specific fields.

For example, you can extract an IP address, a timestamp, or an HTTP method from a plain text log line.

Consider a log line like: 192.168.1.1 GET /api/data 200 123ms

This grok filter extracts fields like clientip, method, request, response_code, and response_time.

While grok is essential, many other filters enhance your data:

• date: Parses timestamps from log lines into a standard format.
• mutate: Allows you to add, remove, rename, or modify fields.
• json: Parses JSON strings within your log messages into structured fields.
• kv: Extracts key-value pairs from a string.

Outputs are the final stage of the Logstash pipeline. They define where your processed events are sent.

The most common output, especially in the ELK Stack, is elasticsearch. Other useful outputs include stdout (for debugging), file, or even message queues like kafka.

Let's test your understanding of Logstash's core components and how data flows through them.

You've learned that Logstash is a vital component of the ELK Stack, acting as a flexible data processing engine.

• It uses Inputs to gather data.
• Filters (like grok) to parse and enrich that data.
• And Outputs to send the processed, structured events to destinations like Elasticsearch.

This transforms raw logs into valuable, searchable information!