Se rendre au contenu

Technique

Lynx Storage - Object Indexer Pipeline

Send object-storage write events to a configurable webhook (Tika, Elasticsearch, embeddings)

lynx_storage_indexer · v19.0.1.0.0 · Premium

What this solves

Lynx Storage - Object Indexer Pipeline

Forwards every object-storage write to a configurable webhook - Apache Tika for full-text extraction, Elasticsearch for indexing, embeddings services for AI search - so files become searchable content without batch jobs.

Object storage is great at storing files but blind to their content. Most teams want full-text search or AI embeddings on top but stand up a separate indexing pipeline that lags behind reality. This module emits an HTTP POST to a configurable webhook on every write, with object metadata and a short-lived presigned URL so the receiver can fetch the bytes if needed.

Key Features

  • Webhook on write - every storage write enqueues a job that POSTs to the configured URL.

  • Receiver-agnostic - works with Apache Tika, Elasticsearch ingest, OpenAI embeddings, voyage, sentence-transformers, or anything that speaks HTTP and JSON.

  • Presigned URL payload - the receiver fetches bytes via short-lived presigned URL, not through Odoo.

  • Retry on failure - 15-minute retry cron with exponential backoff.

  • Standalone - works on manual storage.file uploads even without lynx_storage_attachment.

  • Per-bucket config - different webhooks for different storage buckets where needed.

Integrates With

  • lynx_storage - the object storage backend.

  • External indexers (Tika, Elasticsearch, embeddings services).

Try Lynx Storage - Object Indexer Pipeline on your team.

Free trial, no credit card. Talk to sales when you're ready.