MongoDB
CertifiedTasks that query and manipulate MongoDB collections.
Provide connection URI/credentials and database/collection names to run find, aggregate, insert, update, delete operations, returning JSON results for downstream processing.
MongoDB
Tasks that query and manipulate MongoDB collections.
Provide connection URI/credentials and database/collection names to run find, aggregate, insert, update, delete operations, returning JSON results for downstream processing.
tasks
How to use the MongoDB plugin
Connect to any MongoDB deployment — local, Atlas, or replica set — to query, write, and trigger flows from collection changes.
Authentication
Set the uri property to a MongoDB connection string (e.g., mongodb://user: password@host: 27017/dbname). For Atlas, use the mongodb+srv:// scheme. Store the full URI in a secret to avoid exposing credentials. Alternatively, set host, port, username, and password individually if you prefer not to construct the URI manually.
Common properties
Set database and collection on each task, or apply them globally with plugin defaults if all tasks in a flow target the same collection.
Tasks
Find retrieves documents matching a filter and returns them as Kestra internal storage for downstream tasks. Aggregate runs a pipeline and is the right choice for grouped, transformed, or multi-stage queries. InsertOne adds a single document; Bulk handles batched write operations (inserts, updates, deletes) in a single request — use Bulk when writing multiple documents to avoid per-document round trips. Update modifies matching documents and Delete removes them. Load writes records from a Kestra internal storage file into a collection — use it after a download or transform step to ingest structured data.
Trigger polls a collection on a schedule and starts one execution per batch of new documents matching a filter. Use it to drive downstream processing whenever new data arrives in MongoDB.
