# Elasticsearch / S3 Hybrid Vector Configuration Example
# ------------------------------------------------------------------------------
# This demonstrates a hybrid pipeline, writing data to both Elasticsearch and
# AWS S3. This is advantageous because each storage helps to offset its
# counterpart's weaknesses. You can provision Elasticsearch for performance
# and delegate durability to S3.

data_dir: "/var/lib/vector"

# Ingest data by tailing one or more files
# Docs: https://vector.dev/docs/reference/sources/file
sources:
  apache_logs:
    type: "file"
    include: ["/var/log/*.log"]
    ignore_older_secs: 86400 # 1 day

# Optionally parse, structure and transform data here.
# Docs: https://vector.dev/docs/reference/transforms

# Send structured data to Elasticsearch for searching of recent data
sinks:
  es_cluster:
    inputs: ["apache_logs"]
    type: "elasticsearch"
    endpoint: "79.12.221.222:9200"
    doc_type: "_doc"

  # Send structured data to S3, a durable long-term storage
  s3_archives:
    inputs: ["apache_logs"] # don't sample
    type: "aws_s3"
    region: "us-east-1"
    bucket: "my_log_archives"
    framing:
      method: "newline_delimited"
    encoding:
      codec: "json"
    compression: "gzip"
    batch:
      max_size: 10000000 # 10mb uncompressed