Update Sharded Cluster to Keyfile Authentication

Create a keyfile.

With keyfile authentication, each mongod or mongos instances in the sharded cluster uses the contents of the keyfile as the shared password for authenticating other members in the deployment. Only mongod or mongos instances with the correct keyfile can join the sharded cluster.

A key's length must be between 6 and 1024 characters and may only contain characters in the base64 set. All members of the sharded cluster must share at least one common key.

You can generate a keyfile using any method you choose. For example, the following operation uses openssl to generate a complex pseudo-random 1024 character string to use as a shared password. It then uses chmod to change file permissions to provide read permissions for the file owner only:

openssl rand -base64 756 > <path-to-keyfile>
chmod 400 <path-to-keyfile>

See Keyfiles for additional details and requirements for using keyfiles.

Copy the keyfile to each component in the sharded cluster.

Every server hosting a mongod or mongos component of the sharded cluster must contain a copy of the keyfile.

Copy the keyfile to each server hosting the sharded cluster members. Ensure that the user running the mongod or mongos instances is the owner of the file and can access the keyfile.

Avoid storing the keyfile on storage mediums that can be easily disconnected from the hardware hosting the mongod or mongos instances, such as a USB drive or a network attached storage device.

Disable the Balancer.

Connect mongosh to a mongos.

sh.stopBalancer()

The balancer may not stop immediately if a migration is in progress. The sh.stopBalancer() method blocks the shell until the balancer stops.

Starting in MongoDB 6.1, automatic chunk splitting is not performed. This is because of balancing policy improvements. Auto-splitting commands still exist, but do not perform an operation. For details, see Balancing Policy Changes.

In MongoDB versions earlier than 6.1, sh.stopBalancer() also disables auto-splitting for the sharded cluster.

Use sh.getBalancerState() to verify that the balancer has stopped.

sh.getBalancerState()

See Manage Sharded Cluster Balancer for tutorials on configuring sharded cluster balancer behavior.

Shut down all mongos instances for the sharded cluster.

Connect mongosh to each mongos and shut them down.

Use the db.shutdownServer() method on the admin database to safely shut down the mongos:

db.getSiblingDB("admin").shutdownServer()

Repeat until all mongos instances in the cluster are offline.

Once this step is complete, all mongos instances in the cluster should be offline.

Shut down config server mongod instances.

Connect mongosh to each mongod in the config server deployment and shut them down.

For replica set config server deployments, shut down the primary member last.

Use the db.shutdownServer() method on the admin database to safely shut down the mongod:

db.getSiblingDB("admin").shutdownServer()

Repeat until all config servers are offline.

Shut down shard replica set mongod instances.

For each shard replica set, connect mongosh to each mongod member in the replica set and shut them down. Shut down the primary member last.

Use the db.shutdownServer() method on the admin database to safely shut down the mongod:

db.getSiblingDB("admin").shutdownServer()

Repeat this step for each shard replica set until all mongod instances in all shard replica sets are offline.

Once this step is complete, the entire sharded cluster should be offline.

Enforce Access Control on the Config Servers.

Start each mongod in the config server replica set. Include the keyFile setting. The keyFile setting enforces both Internal/Membership Authentication and Role-Based Access Control.

You can specify the mongod settings either via a configuration file or the command line.

Configuration File

If using a configuration file, for a config server replica set, set security.keyFile to the keyfile's path, sharding.clusterRole to configsvr, and replication.replSetName to the name of the config server replica set.

security:
  keyFile: <path-to-keyfile>
sharding:
  clusterRole: configsvr
replication:
  replSetName: <setname>
storage:
   dbpath: <path>

Include additional options as required for your configuration. For instance, if you wish remote clients to connect to your deployment or your deployment members are run on different hosts, specify the net.bindIp setting. For more information, see Localhost Binding Compatibility Changes.

Start the mongod specifying the --config option and the path to the configuration file.

mongod --config <path-to-config>

Command Line

If using the command line parameters, for a config server replica set, start the mongod with the -keyFile, --configsvr, and --replSet parameters.

mongod --keyFile <path-to-keyfile> --configsvr --replSet <setname> --dbpath <path>

Include additional options as required for your configuration. For instance, if you wish remote clients to connect to your deployment or your deployment members are run on different hosts, specify the --bind_ip. For more information, see Localhost Binding Compatibility Changes.

For more information on command line options, see the mongod reference page.

Make sure to use the original replica set name when restarting each member. You cannot change the name of a replica set.

Enforce Access Control for each Shard in the Sharded Cluster.

Running a mongod with the keyFile parameter enforces both Internal/Membership Authentication and Role-Based Access Control.

Start each mongod in the replica set using either a configuration file or the command line.

Configuration File

If using a configuration file, set the security.keyFile option to the keyfile's path and the replication.replSetName option to the original name of the replica set.

security:
  keyFile: <path-to-keyfile>
replication:
  replSetName: <setname>
storage:
   dbPath: <path>

Include additional options as required for your configuration. For instance, if you wish remote clients to connect to your deployment or your deployment members are run on different hosts, specify the net.bindIp setting. For more information, see Localhost Binding Compatibility Changes.

Start the mongod specifying the --config option and the path to the configuration file.

mongod --config <path-to-config-file>

Command Line

If using the command line parameters, start the mongod and specify the --keyFile and --replSet parameters.

mongod --keyfile <path-to-keyfile> --replSet <setname> --dbpath <path>

Include additional options as required for your configuration. For instance, if you wish remote clients to connect to your deployment or your deployment members are run on different hosts, specify the --bind_ip. For more information, see Localhost Binding Compatibility Changes.

For more information on startup parameters, see the mongod reference page.

Make sure to use the original replica set name when restarting each member. You cannot change the name of a replica set.

Repeat this step until all shards in the cluster are online.

Create a Shard-Local User Administrator (Optional).

For each shard replica set in the cluster, connect mongosh to the primary member over the localhost interface. You must run mongosh on the same machine as the target mongod to use the localhost interface.

Create a user with the userAdminAnyDatabase role on the admin database. This user can create additional users for the shard replica set as necessary. Creating this user also closes the Localhost Exception.

The following example creates the shard-local user fred on the admin database.

admin = db.getSiblingDB("admin")
admin.createUser(
  {
    user: "fred",
    pwd: passwordPrompt(),  // or cleartext password
    roles: [ { role: "userAdminAnyDatabase", db: "admin" } ]
  }
)

Enforce Access Control for the mongos servers.

Running a mongod with the keyFile parameter enforces both Internal/Membership Authentication and Role-Based Access Control.

Start each mongos in the replica set using either a configuration file or the command line.

Configuration File

If using a configuration file, set the security.keyFile to the keyfile`s path and the sharding.configDB to the replica set name and at least one member of the replica set in <replSetName>/<host:port> format.

security:
  keyFile: <path-to-keyfile>
sharding:
  configDB: <configReplSetName>/cfg1.example.net:27019,cfg2.example.net:27019,...

Include additional options as required for your configuration. For instance, if you wish remote clients to connect to your deployment or your deployment members are run on different hosts, specify the net.bindIp setting. For more information, see Localhost Binding Compatibility Changes.

Start the mongos specifying the --config option and the path to the configuration file.

mongos --config <path-to-config-file>

Command Line

If using command line parameters start the mongos and specify the --keyFile and --configdb parameters.

mongos --keyFile <path-to-keyfile> --configdb <configReplSetName>/cfg1.example.net:27019,cfg2.example.net:27019,...

Include additional options as required for your configuration. For instance, if you wish remote clients to connect to your deployment or your deployment members are run on different hosts, specify the --bind_ip. For more information, see Localhost Binding Compatibility Changes.

At this point, the entire sharded cluster is back online and can communicate internally using the keyfile specified. However, external programs like mongosh need to use a correctly provisioned user in order to read or write to the cluster.

Connect to the mongos instance over the localhost interface.

Connect mongosh to one of the mongos instances over the localhost interface. You must run mongosh on the same physical machine as the mongos instance.

The localhost interface is only available since no users have been created for the deployment. The localhost interface closes after the creation of the first user.

Create the user administrator.

Add a user using the db.createUser() method. The user should have at minimum the userAdminAnyDatabase role on the admin database.

The following example creates the user fred on the admin database:

admin = db.getSiblingDB("admin")
admin.createUser(
  {
    user: "fred",
    pwd:  passwordPrompt(),     // or cleartext password
    roles: [ { role: "userAdminAnyDatabase", db: "admin" } ]
  }
)

See Database User Roles for a full list of built-in roles and related to database administration operations.

Authenticate as the user administrator.

Use db.auth() to authenticate as the user administrator to create additional users:

db.getSiblingDB("admin").auth("fred", passwordPrompt()) // or cleartext password

Enter the password when prompted.

Alternatively, connect a new mongosh session to the target replica set member using the -u <username>, -p <password>, and the --authenticationDatabase "admin" parameters. You must use the Localhost Exception to connect to the mongos.

mongosh -u "fred" -p  --authenticationDatabase "admin"

If you do not specify the password to the -p command-line option, mongosh prompts for the password.

Create Administrative User for Cluster Management

The cluster administrator user has the clusterAdmin role for the sharded cluster and not the shard-local cluster administrator.

The following example creates the user ravi on the admin database.

db.getSiblingDB("admin").createUser(
  {
    "user" : "ravi",
    "pwd" : passwordPrompt(),     // or cleartext password
    roles: [ { "role" : "clusterAdmin", "db" : "admin" } ]
  }
)

See Cluster Administration Roles for a full list of built-in roles related to replica set and sharded cluster operations.

Authenticate as cluster admin.

To perform sharding operations, authenticate as a clusterAdmin user with either the db.auth() method or a new mongosh session with the username, password, and authenticationDatabase parameters.

Start the balancer.

Start the balancer.

sh.startBalancer()

Starting in MongoDB 6.1, automatic chunk splitting is not performed. This is because of balancing policy improvements. Auto-splitting commands still exist, but do not perform an operation. For details, see Balancing Policy Changes.

In MongoDB versions earlier than 6.1, sh.startBalancer() also enables auto-splitting for the sharded cluster.

Use the sh.getBalancerState() to verify the balancer has started.

See Manage Sharded Cluster Balancer for tutorials on the sharded cluster balancer.

Create additional users (Optional).

Create users to allow clients to connect and access the sharded cluster. See Database User Roles for available built-in roles, such as read and readWrite. You may also want additional administrative users. For more information on users, see Users.

To create additional users, you must authenticate as a user with userAdminAnyDatabase or userAdmin roles.