Docs Menu

Docs HomeDevelop ApplicationsMongoDB Manual

Use Automatic Queryable Encryption with GCP

On this page

  • Overview
  • Before You Get Started
  • Set Up the KMS
  • Register a GCP Service Account
  • Create a GCP Customer Master Key
  • Create the Application
  • Create a Unique Index on your Key Vault collection
  • Create your Data Encryption Keys and Encrypted Collection
  • Configure your MongoClient for Encrypted Reads and Writes
  • Insert a Document with Encrypted Fields
  • Retrieve Your Document with Encrypted Fields
  • Learn More

This guide shows you how to build a Queryable Encryption enabled application using Google Cloud Key Management Service.

After you complete the steps in this guide, you should have:

  • A CMK hosted on Google Cloud Key Management Service.

  • A client application that inserts documents with encrypted fields using your CMK.

To complete and run the code in this guide, you need to set up your development environment as shown in the Installation Requirements page.

Tip

See: Full Application

To see the complete code for the application you make in this guide, select the tab corresponding to your programming language and follow the provided link:

1
1
2

To create a service account on Google Cloud, follow the Creating a service account guide in Google's official documentation.

3

To add a service account key on Google Cloud, follow the Managing service account keys guide in Google's official documentation.

2
1

Create a key ring and a symmetric key by following the Create a key guide from Google's official documentation.

This key is your Customer Master Key (CMK).

Record the following details of your CMK for use in a future step of this tutorial.

Field
Required
Description
key_name
Yes
Identifier for the CMK.
key_ring
Yes
Identifier for the group of keys your key belongs to.
key_version
No
The version of the named key.
location
Yes
Region specified for your key.
endpoint
No
The host and optional port of the Google Cloud KMS. The default value is cloudkms.googleapis.com.

Select the tab that corresponds to the MongoDB driver you are using in your application to see relevant code samples.

1

Create a unique index on the keyAltNames field in your encryption.__keyVault namespace.

Select the tab corresponding to your preferred MongoDB driver:

2
1

Select the tab corresponding to your preferred MongoDB driver:

Tip

You saved a file containing your service account key credentials in the Create a GCP Service Account step of this guide.

If you downloaded your credentials in JSON format, you can use the following command to extract the value of your private key, substituting <credentials-filename> with the name of your credentials file. The following command requires that you install OpenSSL:

cat <credentials-filename> | jq -r .private_key | openssl pkcs8 -topk8 -nocrypt -inform PEM -outform DER | base64

If you downloaded your credentials in PKCS12 format, you need to specify your GCP service account import password and to add a PEM pass phrase to access the key when accessing it using the following command, substituting <credentials-filename> with the name of your credentials file:

openssl pkcs12 -info -in <credentials-filename>
2

Update the following code to specify your Customer Master Key:

Tip

You recorded your Customer Master Key details in the in the Create a Customer Master Key step of this guide.

3

Construct a client with your MongoDB connection string and Key Vault collection namespace, and create the Data Encryption Keys:

Note

Key Vault Collection Namespace Permissions

The Key Vault collection is in the encryption.__keyVault namespace. Ensure that the database user your application uses to connect to MongoDB has ReadWrite permissions on this namespace.

4

Use a Queryable Encryption enabled MongoClient instance to specify what fields you must encrypt and create your encrypted collection:

The output from the code in this section should resemble the following:

Created encrypted collection!

Tip

Learn More

To view a diagram showing how your client application creates your Data Encryption Key when using an Google Cloud Key Management Service, see Architecture.

To learn more about the options for creating a Data Encryption Key encrypted with a Customer Master Key hosted in Azure Key Vault, see kmsProviders Object and dataKeyOpts Object.

Tip

See: Complete Code

3

Tip

Follow the remaining steps in this tutorial in a separate file from the one created in the previous steps.

1

Specify encryption.__keyVault as the Key Vault collection namespace.

2
3
4

Tip

Learn More

To learn more about the Automatic Encryption Shared Library, see the Automatic Encryption Shared Library page.

5

Instantiate a MongoDB client object with the following automatic encryption settings that use the variables declared in the previous steps:

4

Use your Queryable Encryption enabled MongoClient instance to insert a document with encrypted fields into the medicalRecords.patients namespace using the following code snippet:

When you insert a document, your Queryable Encryption enabled client encrypts the fields of your document such that it resembles the following:

{
"_id": { "$oid": "<_id value>" },
"firstName": "Jon",
"lastName": "Doe",
"patientId": {
"$binary": {
"base64": "<ciphertext>",
"subType": "06"
}
},
"address": "157 Electric Ave.",
"patientRecord": {
"ssn": {
"$binary": {
"base64": "<ciphertext>",
"subType": "06"
}
},
"billing": {
"$binary": {
"base64": "<ciphertext>",
"subType": "06"
}
}
},
"medications": {
"$binary": {
"base64": "<ciphertext>",
"subType": "06"
}
},
"__safeContent__": [
{
"$binary": {
"base64": "<ciphertext>",
"subType": "00"
}
},
{
"$binary": {
"base64": "<ciphertext>",
"subType": "00"
}
}
]
}

Warning

Do not Modify the __safeContent__ Field

The __safeContent__ field is essential to Queryable Encryption. Do not modify the contents of this field.

Tip

See: Complete Code

5

Retrieve the document with encrypted fields you inserted in the Insert a Document with Encrypted Fields step of this guide.

To show the functionality of Queryable Encryption, the following code snippet queries for your document with a client configured for automatic Queryable Encryption as well as a client that is not configured for automatic Queryable Encryption.

The output of the preceding code snippet should look like this:

Finding a document with regular (non-encrypted) client.
{
_id: new ObjectId("628eabeb37590e84ea742665"),
firstName: 'Jon',
lastName: 'Doe',
patientId: new Binary(Buffer.from("0798810acc0f4f46c9a76883cee80fca12102e9ddcbcdae46a821fa108a8155a850f2d0919475b6531ada68973d436a199b537a05a98a708c36d2bfec4979d59cbe66878865ce19e392d3e4789d309bdacc336e32efcc851806ae0a41b355288c10d01e39147e1c40d919c41913a0c9d2d3fad0d0d1d2873c4fc82c6c22f27b517df5f3131b331b96ed16a7c5cf89e09082a2d898c2dcd73da91d08760ba74a70077b2d0fdbbe1eea75655a19fcc397812325ad40b102cbd16b8d36b22e11e3f93404f24a8ff68cfdec3c22b0e787cb30078a5227b2a", "hex"), 6),
address: '157 Electric Ave.',
patientRecord: {
ssn: new Binary(Buffer.from("07e8b69630c32f4a00a542af768f8abcf50223edd812ff20b0ecb046ee1a9f5a0eef8d85d99cd26076411129942752516ee605c55aadce73f3d44d81ea6ddbbb8134b108a9deb40d8cab9cb4f08ef210ab0c9d2ea4347f9d235b861baf29751e60abcf059eb5c120305bd5ac05a4e07ac8ccfa6d37283f4cdbfeb7a8accb65b71857d486b5cf55e354d6a95e287d9e2dd65f3f9d9c4c9d0bdb1f26c4bd549d7be77db81796be293e08b2223bac67b212423c4e06568578b5bd7a3c33cedc1b291bcda0b27e005144d344563711a489f24b8e9b65bbb721d3a0e9d9b227a0cec0cbad", "hex"), 6),
billing: new Binary(Buffer.from("06808ae69d4caa49cf90bb688f386f097f03f870a7b8fcebb1980c9ee5488b1f0f68558fc2163adcd92d00ea5f349f56ed34e7b391f54c48ed2760b4bde73022fc818dc7486a4e046b92ce9c82e00333c7779d9d6bb476713a20632b593b7de54812662cfc4d174d05451d3f4195514e12edba", "hex"), 6)
},
medications: new Binary(Buffer.from("06665ec15d38254dc4aa16da856789d33404f27bfea53e0d2fa4deaff166989ab33f469644d89c29112d33b41dbe54ec2d89c43f3de52cdc5d454e8694046216f533614fa7b42b7c5406d6518f7ed8f9e3ce52fda6c8b2146d0f8cc51e21a3467183697e1735a9f60c18e173c1916101", "hex"), 6),
__safeContent__: [
new Binary(Buffer.from("3044b134ad0f7c8a90dab1e05bb8b296a8ede540796bd7403ab47693cdba1b26", "hex"), 0),
new Binary(Buffer.from("a22ddf9a5657cdd56bef72febbba44371899e6486962a1c07d682082c4e65712", "hex"), 0)
]
}
Finding a document with encrypted client, searching on an encrypted field
{
_id: new ObjectId("628eaca1dcf9b63e2f43162d"),
firstName: 'Jon',
lastName: 'Doe',
patientId: 12345678,
address: '157 Electric Ave.',
patientRecord: {
ssn: '987-65-4320',
billing: { type: 'Visa', number: '4111111111111111' }
},
medications: [ 'Atorvastatin', 'Levothyroxine' ],
__safeContent__: [
new Binary(Buffer.from("fbdc6cfe3b4659693650bfc60baced27dcb42b793efe09da0ded54d60a9d5a1f", "hex"), 0),
new Binary(Buffer.from("0f92ff92bf904a858ef6fd5b1e508187f523e791f51d8b64596461b38ebb1791", "hex"), 0)
]
}

Tip

See: Complete Code

To learn more about the topics mentioned in this guide, see the following links:

  • Learn more about Queryable Encryption components on the Reference page.

  • Learn how Customer Master Keys and Data Encryption Keys work on the Keys and Key Vaults page.

  • See how KMS Providers manage your Queryable Encryption keys on the KMS Providers page.

←  Use Automatic Queryable Encryption with AzureUse Automatic Queryable Encryption with KMIP →