Service: PostgreSQL DocumentDB - EyevinnOSC/community GitHub Wiki

PostgreSQL DocumentDB

PostgreSQL DocumentDB (serviceId ferretdb-documentdb) is a MongoDB-compatible document database built by FerretDB on top of PostgreSQL. In Eyevinn Open Source Cloud (OSC) it is a long-running service instance with up to 10 GiB of storage per instance. License: MIT. Source: FerretDB/documentdb.

What it can do

  • Let applications that speak the MongoDB wire protocol store their data in PostgreSQL without running MongoDB.
  • Provide document collections with BSON documents, queries, updates, deletes, indexes (including TTL) and aggregation pipelines through the DocumentDB API.
  • Fit into workflows that already use PostgreSQL while adding document-oriented capabilities.

Creating an instance

There are no dependencies. Store the PostgreSQL password as a service secret with create-service-secret (serviceId: "ferretdb-documentdb", for example name documentdbpwd), then call create-service-instance with serviceId: "ferretdb-documentdb":

{
  "name": "mydocdb",
  "PostgresPassword": "{{secrets.documentdbpwd}}"
}

Verify with wait-for-service-ready or describe-service-instance, which also shows the address and port to connect to. With the OSC CLI:

osc create ferretdb-documentdb mydocdb -o PostgresPassword="{{secrets.documentdbpwd}}"

Options

Option Required Type Meaning
name Yes string Name of the instance. Alphanumeric/word characters only (^\w+$), at most 20 characters.
PostgresPassword Yes string, sensitive Password for the underlying PostgreSQL. Pass as {{secrets.NAME}}; a reference must be the whole field value.
InitdbSql No string SQL that runs when the database is first initialized.

Using the instance

Use the address and port from describe-service-instance to build a MongoDB connection string. See the FerretDB documentation for the authentication settings that apply to the image.

const { MongoClient } = require('mongodb');
const client = new MongoClient(process.env.DOCUMENTDB_URL);
await client.connect();
const items = client.db('mydb').collection('items');
await items.insertOne({ hello: 'world' });

Limitations

  • Storage is limited to 10 GiB per instance.

Resources