Skip to main content

Elasticsearch Plugin

lynx-elasticsearch integrates the official Elasticsearch client into Lynx runtime startup. Its role is to own client initialization, connection policy, metrics and health settings, and index-prefix handling, while leaving your query DSL and index mapping design in application code.

Runtime facts​

ItemValue
Go modulegithub.com/go-lynx/lynx-elasticsearch
Config prefixlynx.elasticsearch
Runtime plugin nameelasticsearch.client
Main gettersGetElasticsearch(), GetElasticsearchPlugin(), GetIndexName(name)

What the implementation actually provides​

  • initializes the Elasticsearch client from lynx.elasticsearch
  • supports multiple node addresses, auth fields, retries, timeouts, health checks, and metrics
  • exposes the raw client through GetElasticsearch()
  • exposes plugin-level helpers such as GetConnectionStats() and prefixed index name generation

Configuration​

lynx:
  elasticsearch:
    addresses:
      - "http://localhost:9200"
      - "http://localhost:9201"
    username: "elastic"
    password: "changeme"
    connect_timeout: "30s"
    max_retries: 3
    enable_metrics: true
    enable_health_check: true
    health_check_interval: "30s"
    index_prefix: "myapp"

Usage​

import elasticsearch "github.com/go-lynx/lynx-elasticsearch"

client := elasticsearch.GetElasticsearch()
plugin := elasticsearch.GetElasticsearchPlugin()
indexName := elasticsearch.GetIndexName("documents")
stats := plugin.GetConnectionStats()

_ = client
_ = indexName
_ = stats

Practical guidance​

  • use index_prefix when multiple apps or environments share one cluster
  • keep schema, aliases, and query composition in your own service layer, not in plugin config
  • if you need runtime introspection, prefer GetConnectionStats() over inventing ad hoc health probes