Navigation
This version of the documentation is archived and no longer supported.

mongod

Synopsis

mongod is the primary daemon process for the MongoDB system. It handles data requests, manages data access, and performs background management operations.

This document provides a complete overview of all command line options for mongod. These command line options are primarily useful for testing: In common operation, use the configuration file options to control the behavior of your database.

Note

Starting in version 4.0, MongoDB disables support for TLS 1.0 encryption on systems where TLS 1.1+ is available. For more details, see Disable TLS 1.0.

Options

Core Options

--help, -h

Returns information on the options and use of mongod.

--version

Returns the mongod release number.

--config <filename>, -f <filename>

Specifies a configuration file for runtime configuration options. The configuration file is the preferred method for runtime configuration of mongod. The options are equivalent to the command-line configuration options. See Configuration File Options for more information.

Ensure the configuration file uses ASCII encoding. The mongod instance does not support configuration files with non-ASCII encoding, including UTF-8.

--verbose, -v

Increases the amount of internal reporting returned on standard output or in log files. Increase the verbosity with the -v form by including the option multiple times, (e.g. -vvvvv.)

--quiet

Runs mongod in a quiet mode that attempts to limit the amount of output.

This option suppresses:

  • output from database commands
  • replication activity
  • connection accepted events
  • connection closed events
--port <port>

Default:

The TCP port on which the MongoDB instance listens for client connections.

--bind_ip <hostnames|ipaddresses|Unix domain socket paths>

Default: localhost

Note

Starting in MongoDB 3.6, mongod bind to localhost by default. See Default Bind to Localhost.

The hostnames and/or IP addresses and/or full Unix domain socket paths on which mongod should listen for client connections. You may attach mongod to any interface. To bind to multiple addresses, enter a list of comma-separated values.

Example

localhost,/tmp/mongod.sock

You can specify both IPv4 and IPv6 addresses, or hostnames that resolve to an IPv4 or IPv6 address.

Example

localhost, 2001:0DB8:e132:ba26:0d5c:2774:e7f9:d513

Note

If specifying an IPv6 address or a hostname that resolves to an IPv6 address to --bind_ip, you must start mongod with --ipv6 to enable IPv6 support. Specifying an IPv6 address to --bind_ip does not enable IPv6 support.

If specifying a link-local IPv6 address (fe80::/10), you must append the zone index to that address (i.e. fe80::<address>%<adapter-name>).

Example

localhost,fe80::a00:27ff:fee0:1fcf%enp0s3

Tip

When possible, use a logical DNS hostname instead of an ip address, particularly when configuring replica set members or sharded cluster members. The use of logical DNS hostnames avoids configuration changes due to ip address changes.

Warning

Before binding to a non-localhost (e.g. publicly accessible) IP address, ensure you have secured your cluster from unauthorized access. For a complete list of security recommendations, see Security Checklist. At minimum, consider enabling authentication and hardening network infrastructure.

For more information about IP Binding, refer to the IP Binding documentation.

To bind to all IPv4 addresses, enter 0.0.0.0.

To bind to all IPv4 and IPv6 addresses, enter ::,0.0.0.0 or alternatively, use the net.bindIpAll setting.

Note

--bind_ip and --bind_ip_all are mutually exclusive. Specifying both options causes mongod to throw an error and terminate.

--bind_ip_all

New in version 3.6.

If specified, the mongod instance binds to all IPv4 addresses (i.e. 0.0.0.0). If mongod starts with --ipv6, --bind_ip_all also binds to all IPv6 addresses (i.e. ::).

mongod only supports IPv6 if started with --ipv6. Specifying --bind_ip_all alone does not enable IPv6 support.

Warning

Before binding to a non-localhost (e.g. publicly accessible) IP address, ensure you have secured your cluster from unauthorized access. For a complete list of security recommendations, see Security Checklist. At minimum, consider enabling authentication and hardening network infrastructure.

For more information about IP Binding, refer to the IP Binding documentation.

Alternatively, you can set the --bind_ip option to ::,0.0.0.0 to bind to all IP addresses.

Note

--bind_ip and --bind_ip_all are mutually exclusive. That is, you can specify one or the other, but not both.

--clusterIpSourceWhitelist <string>

New in version 3.6.

A list of IP addresses/CIDR (Classless Inter-Domain Routing) ranges against which the mongod validates authentication requests from other members of the replica set and, if part of a sharded cluster, the mongos instances. The mongod verifies that the originating IP is either explicitly in the list or belongs to a CIDR range in the list. If the IP address is not present, the server does not authenticate the mongod or mongos.

--clusterIpSourceWhitelist has no effect on a mongod started without authentication.

--clusterIpSourceWhitelist accepts multiple comma-separated IPv4/6 addresses or Classless Inter-Domain Routing (CIDR) ranges:

mongod --clusterIpSourceWhitelist 192.0.2.0/24,127.0.0.1,::1

Important

Ensure --clusterIpSourceWhitelist includes the IP address or CIDR ranges that include the IP address of each replica set member or mongos in the deployment to ensure healthy communication between cluster components.

--ipv6

Enables IPv6 support. mongod disables IPv6 support by default.

Setting --ipv6 does not direct the mongod to listen on any local IPv6 addresses or interfaces. To configure the mongod to listen on an IPv6 interface, you must either:

  • Configure --bind_ip with one or more IPv6 addresses or hostnames that resolve to IPv6 addresses, or
  • Set --bind_ip_all to true.
--listenBacklog <number>

Default: Target system SOMAXCONN constant

New in version 3.6.

The maximum number of connections that can exist in the listen queue.

Warning

Consult your local system’s documentation to understand the limitations and configuration requirements before using this parameter.

Important

To prevent undefined behavior, specify a value for this parameter between 1 and the local system SOMAXCONN constant.

The default value for the listenBacklog parameter is set at compile time to the target system SOMAXCONN constant. SOMAXCONN is the maximum valid value that is documented for the backlog parameter to the listen system call.

Some systems may interpret SOMAXCONN symbolically, and others numerically. The actual listen backlog applied in practice may differ from any numeric interpretation of the SOMAXCONN constant or argument to --listenBacklog, and may also be constrained by system settings like net.core.somaxconn on Linux.

Passing a value for the listenBacklog parameter that exceeds the SOMAXCONN constant for the local system is, by the letter of the standards, undefined behavior. Higher values may be silently integer truncated, may be ignored, may cause unexpected resource consumption, or have other adverse consequences.

On systems with workloads that exhibit connection spikes, for which it is empirically known that the local system can honor higher values for the backlog parameter than the SOMAXCONN constant, setting the listenBacklog parameter to a higher value may reduce operation latency as observed by the client by reducing the number of connections which are forced into a backoff state.

--maxConns <number>

The maximum number of simultaneous connections that mongod will accept. This setting has no effect if it is higher than your operating system’s configured maximum connection tracking threshold.

Do not assign too low of a value to this option, or you will encounter errors during normal application operation.

Note

Changed in version 2.6: MongoDB removed the upward limit on the maxIncomingConnections setting.

--logpath <path>

Sends all diagnostic logging information to a log file instead of to standard output or to the host’s syslog system. MongoDB creates the log file at the path you specify.

By default, MongoDB will move any existing log file rather than overwrite it. To instead append to the log file, set the --logappend option.

--syslog

Sends all logging output to the host’s syslog system rather than to standard output or to a log file. , as with --logpath.

The --syslog option is not supported on Windows.

Warning

The syslog daemon generates timestamps when it logs a message, not when MongoDB issues the message. This can lead to misleading timestamps for log entries, especially when the system is under heavy load. We recommend using the --logpath option for production systems to ensure accurate timestamps.

--syslogFacility <string>

Default: user

Specifies the facility level used when logging messages to syslog. The value you specify must be supported by your operating system’s implementation of syslog. To use this option, you must enable the --syslog option.

--logappend

Appends new entries to the end of the existing log file when the mongod instance restarts. Without this option, mongod will back up the existing log and create a new file.

--logRotate <string>

Default: rename

New in version 3.0.0.

Determines the behavior for the logRotate command. Specify either rename or reopen:

  • rename renames the log file.

  • reopen closes and reopens the log file following the typical Linux/Unix log rotate behavior. Use reopen when using the Linux/Unix logrotate utility to avoid log loss.

    If you specify reopen, you must also use --logappend.

If auditing is enabled, the logRotate command also rotates the audit log according to the above parameters. For example, if --logRotate is set to rename, the audit log will also be renamed.

--timeStampFormat <string>

Default: iso8601-local

The time format for timestamps in log messages. Specify one of the following values:

Value Description
ctime Displays timestamps as Wed Dec 31 18:17:54.811.
iso8601-utc Displays timestamps in Coordinated Universal Time (UTC) in the ISO-8601 format. For example, for New York at the start of the Epoch: 1970-01-01T00:00:00.000Z
iso8601-local Displays timestamps in local time in the ISO-8601 format. For example, for New York at the start of the Epoch: 1969-12-31T19:00:00.000-0500
--traceExceptions

For internal diagnostic use only.

--pidfilepath <path>

Specifies a file location to store the process ID (PID) of the mongod process. The user running the mongod or mongos process must be able to write to this path. If the --pidfilepath option is not specified, the process does not create a PID file. This option is generally only useful in combination with the --fork option.

Linux

On Linux, PID file management is generally the responsibility of your distro’s init system: usually a service file in the /etc/init.d directory, or a systemd unit file registered with systemctl. Only use the --pidfilepath option if you are not using one of these init systems. For more information, please see the respective Installation Guide for your operating system.

macOS

On macOS, PID file management is generally handled by brew. Only use the --pidfilepath option if you are not using brew on your macOS system. For more information, please see the respective Installation Guide for your operating system.

--keyFile <file>

Specifies the path to a key file that stores the shared secret that MongoDB instances use to authenticate to each other in a sharded cluster or replica set. --keyFile implies --auth. See Internal Authentication for more information.

--setParameter <options>

Specifies one of the MongoDB parameters described in MongoDB Server Parameters. You can specify multiple setParameter fields.

--nounixsocket

Disables listening on the UNIX domain socket. --nounixsocket applies only to Unix-based systems.

The mongod process always listens on the UNIX socket unless one of the following is true:

New in version 2.6: mongod installed from official .deb and .rpm packages have the bind_ip configuration set to 127.0.0.1 by default.

--unixSocketPrefix <path>

Default: /tmp

The path for the UNIX socket. --unixSocketPrefix applies only to Unix-based systems.

If this option has no value, the mongod process creates a socket with /tmp as a prefix. MongoDB creates and listens on a UNIX socket unless one of the following is true:

--filePermissions <path>

Default: 0700

Sets the permission for the UNIX domain socket file.

--filePermissions applies only to Unix-based systems.

--fork

Enables a daemon mode that runs the mongod process in the background. By default mongod does not run as a daemon: typically you will run mongod as a daemon, either by using --fork or by using a controlling process that handles the daemonization process (e.g. as with upstart and systemd).

Using the --fork option requires that you configure log output for the mongod with one of the following:

The --fork option is not supported on Windows.

--auth

Enables authorization to control user’s access to database resources and operations. When authorization is enabled, MongoDB requires all clients to authenticate themselves first in order to determine the access for the client.

Configure users via the mongo shell. If no users exist, the localhost interface will continue to have access to the database until you create the first user.

See Security for more information.

--noauth

Disables authentication. Currently the default. Exists for future compatibility and clarity.

--transitionToAuth

New in version 3.4: Allows the mongod to accept and create authenticated and non-authenticated connections to and from other mongod and mongos instances in the deployment. Used for performing rolling transition of replica sets or sharded clusters from a no-auth configuration to internal authentication. Requires specifying a internal authentication mechanism such as --keyFile.

For example, if using keyfiles for internal authentication, the mongod creates an authenticated connection with any mongod or mongos in the deployment using a matching keyfile. If the security mechanisms do not match, the mongod utilizes a non-authenticated connection instead.

A mongod running with --transitionToAuth does not enforce user access controls. Users may connect to your deployment without any access control checks and perform read, write, and administrative operations.

Note

A mongod running with internal authentication and without --transitionToAuth requires clients to connect using user access controls. Update clients to connect to the mongod using the appropriate user prior to restarting mongod without --transitionToAuth.

--cpu

Forces the mongod process to report the percentage of CPU time in write lock, every four seconds.

--sysinfo

Returns diagnostic system information and then exits. The information provides the page size, the number of physical pages, and the number of available physical pages.

--noscripting

Disables the scripting engine.

--notablescan

Forbids operations that require a collection scan. See notablescan for additional information.

--shutdown

The --shutdown option cleanly and safely terminates the mongod process. When invoking mongod with this option you must set the --dbpath option either directly or by way of the configuration file and the --config option.

The --shutdown option is available only on Linux systems.

For additional ways to shut down, see also Stop mongod Processes.

--redactClientLogData

New in version 3.4: Available in MongoDB Enterprise only.

A mongod running with --redactClientLogData redacts any message accompanying a given log event before logging. This prevents the mongod from writing potentially sensitive data stored on the database to the diagnostic log. Metadata such as error or operation codes, line numbers, and source file names are still visible in the logs.

Use --redactClientLogData in conjunction with Encryption at Rest and TLS/SSL (Transport Encryption) to assist compliance with regulatory requirements.

For example, a MongoDB deployment might store Personally Identifiable Information (PII) in one or more collections. The mongod logs events such as those related to CRUD operations, sharding metadata, etc. It is possible that the mongod may expose PII as a part of these logging operations. A mongod running with --redactClientLogData removes any message accompanying these events before being output to the log, effectively removing the PII.

Diagnostics on a mongod running with --redactClientLogData may be more difficult due to the lack of data related to a log event. See the process logging manual page for an example of the effect of --redactClientLogData on log output.

On a running mongod, use setParameter with the redactClientLogData parameter to configure this setting.

--networkMessageCompressors <string>

Default: snappy

New in version 3.4.

Specifies the default compressor(s) to use for communication between this mongod instance and:

  • other members of the deployment if the instance is part of a replica set or a sharded cluster
  • a mongo shell
  • drivers that support the OP_COMPRESSED message format.

MongoDB supports the following compressors:

  • snappy
  • zlib (Available in MongoDB 3.6 or greater)

In versions 3.6 and 4.0, mongod and mongos enable network compression by default with snappy as the compressor.

To disable network compression, set the value to disabled.

Important

Messages are compressed when both parties enable network compression. Otherwise, messages between the parties are uncompressed.

If you specify multiple compressors, then the order in which you list the compressors matter as well as the communication initiator. For example, if a mongo shell specifies the following network compressors zlib,snappy and the mongod specifies snappy,zlib, messages between mongo shell and mongod uses zlib.

If the parties do not share at least one common compressor, messages between the parties are uncompressed. For example, if a mongo shell specifies the network compressor zlib and mongod specifies snappy, messages between mongo shell and mongod are not compressed.

--timeZoneInfo <path>

The full path from which to load the time zone database. If this option is not provided, then MongoDB will use its built-in time zone database.

The configuration file included with Linux and macOS packages sets the time zone database path to /usr/share/zoneinfo by default.

The built-in time zone database is a copy of the Olson/IANA time zone database. It is updated along with MongoDB releases, but the time zone database release cycle differs from the MongoDB release cycle. The most recent release of the time zone database is available on our download site.

wget https://downloads.mongodb.org/olson_tz_db/timezonedb-latest.zip
unzip timezonedb-latest.zip
mongod --timeZoneInfo timezonedb-2017b/

Warning

MongoDB uses the third party timelib library to provide accurate conversions between timezones. Due to a recent update, timelib could create inaccurate time zone conversions in older versions of MongoDB.

To explicitly link to the time zone database in versions of MongoDB prior to 4.0.25, download the time zone database. and use the timeZoneInfo parameter.

--serviceExecutor <string>

Default: synchronous

New in version 3.6.

Determines the threading and execution model mongod uses to execute client requests. The --serviceExecutor option accepts one of the following values:

Value Description
synchronous The mongod uses synchronous networking and manages its networking thread pool on a per connection basis. Previous versions of MongoDB managed threads in this way.
adaptive The mongod uses the new experimental asynchronous networking mode with an adaptive thread pool which manages threads on a per request basis. This mode should have more consistent performance and use less resources when there are more inactive connections than database requests.

Free Monitoring

New in version 4.0.

--enableFreeMonitoring <runtime|on|off>

New in version 4.0: Available for MongoDB Community Edition.

Enables or disables free MongoDB Cloud monitoring. --enableFreeMonitoring accepts the following values:

runtime

Default. You can enable or disable free monitoring during runtime.

To enable or disable free monitoring during runtime, see db.enableFreeMonitoring() and db.disableFreeMonitoring().

To enable or disable free monitoring during runtime when running with access control, users must have required privileges. See db.enableFreeMonitoring() and db.disableFreeMonitoring() for details.

on Enables free monitoring at startup; i.e. registers for free monitoring. When enabled at startup, you cannot disable free monitoring during runtime.
off Disables free monitoring at startup, regardless of whether you have previously registered for free monitoring. When disabled at startup, you cannot enable free monitoring during runtime.

Once enabled, the free monitoring state remains enabled until explicitly disabled. That is, you do not need to re-enable each time you start the server.

For the corresponding configuration file setting, see cloud.monitoring.free.state.

--freeMonitoringTag <string>

New in version 4.0: Available for MongoDB Community Edition.

Optional tag to describe environment context. The tag can be sent as part of the free MongoDB Cloud monitoring registration at start up.

For the corresponding configuration file setting, see cloud.monitoring.free.tags.

LDAP Authentication or Authorization Options

--ldapServers <host1>:<port>,<host2>:<port>,...,<hostN>:<port>

New in version 3.4: Available in MongoDB Enterprise only.

The LDAP server against which the mongod authenticates users or determines what actions a user is authorized to perform on a given database. If the LDAP server specified has any replicated instances, you may specify the host and port of each replicated server in a comma-delimited list.

If your LDAP infrastructure partitions the LDAP directory over multiple LDAP servers, specify one LDAP server or any of its replicated instances to --ldapServers. MongoDB supports following LDAP referrals as defined in RFC 4511 4.1.10. Do not use --ldapServers for listing every LDAP server in your infrastructure.

This setting can be configured on a running mongod using setParameter.

If unset, mongod cannot use LDAP authentication or authorization.

--ldapValidateLDAPServerConfig <boolean>

Available in MongoDB Enterprise

A flag that determines if the mongod instance checks the availability of the LDAP server(s) as part of its startup:

  • If true, the mongod instance performs the availability check and only continues to start up if the LDAP server is available.
  • If false, the mongod instance skips the availability check; i.e. the instance starts up even if the LDAP server is unavailable.
--ldapQueryUser <string>

New in version 3.4: Available in MongoDB Enterprise only.

The identity with which mongod binds as, when connecting to or performing queries on an LDAP server.

Only required if any of the following are true:

You must use --ldapQueryUser with --ldapQueryPassword.

If unset, mongod will not attempt to bind to the LDAP server.

This setting can be configured on a running mongod using setParameter.

Note

Windows MongoDB deployments can use --ldapBindWithOSDefaults instead of --ldapQueryUser and --ldapQueryPassword. You cannot specify both --ldapQueryUser and --ldapBindWithOSDefaults at the same time.

--ldapQueryPassword <string>

New in version 3.4: Available in MongoDB Enterprise only.

The password used to bind to an LDAP server when using --ldapQueryUser. You must use --ldapQueryPassword with --ldapQueryUser.

If unset, mongod will not attempt to bind to the LDAP server.

This setting can be configured on a running mongod using setParameter.

Note

Windows MongoDB deployments can use --ldapBindWithOSDefaults instead of --ldapQueryPassword and --ldapQueryPassword. You cannot specify both --ldapQueryPassword and --ldapBindWithOSDefaults at the same time.

--ldapBindWithOSDefaults <bool>

Default: false

New in version 3.4: Available in MongoDB Enterprise for the Windows platform only.

Allows mongod to authenticate, or bind, using your Windows login credentials when connecting to the LDAP server.

Only required if:

Use --ldapBindWithOSDefaults to replace --ldapQueryUser and --ldapQueryPassword.

--ldapBindMethod <string>

Default: simple

New in version 3.4: Available in MongoDB Enterprise only.

The method mongod uses to authenticate to an LDAP server. Use with --ldapQueryUser and --ldapQueryPassword to connect to the LDAP server.

--ldapBindMethod supports the following values:

  • simple - mongod uses simple authentication.
  • sasl - mongod uses SASL protocol for authentication

If you specify sasl, you can configure the available SASL mechanisms using --ldapBindSaslMechanisms. mongod defaults to using DIGEST-MD5 mechanism.

--ldapBindSaslMechanisms <string>

Default: DIGEST-MD5

New in version 3.4: Available in MongoDB Enterprise only.

A comma-separated list of SASL mechanisms mongod can use when authenticating to the LDAP server. The mongod and the LDAP server must agree on at least one mechanism. The mongod dynamically loads any SASL mechanism libraries installed on the host machine at runtime.

Install and configure the appropriate libraries for the selected SASL mechanism(s) on both the mongod host and the remote LDAP server host. Your operating system may include certain SASL libraries by default. Defer to the documentation associated with each SASL mechanism for guidance on installation and configuration.

If using the GSSAPI SASL mechanism for use with Kerberos Authentication, verify the following for the mongod host machine:

Linux
  • The KRB5_CLIENT_KTNAME environment variable resolves to the name of the client Linux Keytab Files for the host machine. For more on Kerberos environment variables, please defer to the Kerberos documentation.
  • The client keytab includes a User Principal for the mongod to use when connecting to the LDAP server and execute LDAP queries.
Windows
If connecting to an Active Directory server, the Windows Kerberos configuration automatically generates a Ticket-Granting-Ticket when the user logs onto the system. Set --ldapBindWithOSDefaults to true to allow mongod to use the generated credentials when connecting to the Active Directory server and execute queries.

Set --ldapBindMethod to sasl to use this option.

Note

For a complete list of SASL mechanisms see the IANA listing. Defer to the documentation for your LDAP or Active Directory service for identifying the SASL mechanisms compatible with the service.

MongoDB is not a source of SASL mechanism libraries, nor is the MongoDB documentation a definitive source for installing or configuring any given SASL mechanism. For documentation and support, defer to the SASL mechanism library vendor or owner.

For more information on SASL, defer to the following resources:

--ldapTransportSecurity <string>

Default: tls

New in version 3.4: Available in MongoDB Enterprise only.

By default, mongod creates a TLS/SSL secured connection to the LDAP server.

For Linux deployments, you must configure the appropriate TLS Options in /etc/openldap/ldap.conf file. Your operating system’s package manager creates this file as part of the MongoDB Enterprise installation, via the libldap dependency. See the documentation for TLS Options in the ldap.conf OpenLDAP documentation for more complete instructions.

For Windows deployment, you must add the LDAP server CA certificates to the Windows certificate management tool. The exact name and functionality of the tool may vary depending on operating system version. Please see the documentation for your version of Windows for more information on certificate management.

Set --ldapTransportSecurity to none to disable TLS/SSL between mongod and the LDAP server.

Warning

Setting --ldapTransportSecurity to none transmits plaintext information and possibly credentials between mongod and the LDAP server.

--ldapTimeoutMS <long>

Default: 10000

New in version 3.4: Available in MongoDB Enterprise only.

The amount of time in milliseconds mongod should wait for an LDAP server to respond to a request.

Increasing the value of --ldapTimeoutMS may prevent connection failure between the MongoDB server and the LDAP server, if the source of the failure is a connection timeout. Decreasing the value of --ldapTimeoutMS reduces the time MongoDB waits for a response from the LDAP server.

This setting can be configured on a running mongod using setParameter.

--ldapUserToDNMapping <string>

New in version 3.4: Available in MongoDB Enterprise only.

Maps the username provided to mongod for authentication to a LDAP Distinguished Name (DN). You may need to use --ldapUserToDNMapping to transform a username into an LDAP DN in the following scenarios:

  • Performing LDAP authentication with simple LDAP binding, where users authenticate to MongoDB with usernames that are not full LDAP DNs.
  • Using an LDAP authorization query template that requires a DN.
  • Transforming the usernames of clients authenticating to Mongo DB using different authentication mechanisms (e.g. x.509, kerberos) to a full LDAP DN for authorization.

--ldapUserToDNMapping expects a quote-enclosed JSON-string representing an ordered array of documents. Each document contains a regular expression match and either a substitution or ldapQuery template used for transforming the incoming username.

Each document in the array has the following form:

{
  match: "<regex>"
  substitution: "<LDAP DN>" | ldapQuery: "<LDAP Query>"
}
Field Description Example
match An ECMAScript-formatted regular expression (regex) to match against a provided username. Each parenthesis-enclosed section represents a regex capture group used by substitution or ldapQuery. "(.+)ENGINEERING" "(.+)DBA"
substitution

An LDAP distinguished name (DN) formatting template that converts the authentication name matched by the match regex into a LDAP DN. Each curly bracket-enclosed numeric value is replaced by the corresponding regex capture group extracted from the authentication username via the match regex.

The result of the substitution must be an RFC4514 escaped string.

"cn={0},ou=engineering, dc=example,dc=com"
ldapQuery A LDAP query formatting template that inserts the authentication name matched by the match regex into an LDAP query URI encoded respecting RFC4515 and RFC4516. Each curly bracket-enclosed numeric value is replaced by the corresponding regex capture group extracted from the authentication username via the match expression. mongod executes the query against the LDAP server to retrieve the LDAP DN for the authenticated user. mongod requires exactly one returned result for the transformation to be successful, or mongod skips this transformation. "ou=engineering,dc=example, dc=com??one?(user={0})"

Note

An explanation of RFC4514, RFC4515, RFC4516, or LDAP queries is out of scope for the MongoDB Documentation. Please review the RFC directly or use your preferred LDAP resource.

For each document in the array, you must use either substitution or ldapQuery. You cannot specify both in the same document.

When performing authentication or authorization, mongod steps through each document in the array in the given order, checking the authentication username against the match filter. If a match is found, mongod applies the transformation and uses the output for authenticating the user. mongod does not check the remaining documents in the array.

If the given document does not match the provided authentication name, or the transformation described by the document fails, mongod continues through the list of documents to find additional matches. If no matches are found in any document, mongod returns an error.

Example

The following shows two transformation documents. The first document matches against any string ending in @ENGINEERING, placing anything preceeding the suffix into a regex capture group. The second document matches against any string ending in @DBA, placing anything preceeding the suffix into a regex capture group.

Important

You must pass the array to --ldapUserToDNMapping as a string.

"[
   {
      match: "(.+)@ENGINEERING.EXAMPLE.COM",
      substitution: "cn={0},ou=engineering,dc=example,dc=com"
   },
   {
      match: "(.+)@DBA.EXAMPLE.COM",
      ldapQuery: "ou=dba,dc=example,dc=com??one?(user={0})"

   }

]"

A user with username alice@ENGINEERING.EXAMPLE.COM matches the first document. The regex capture group {0} corresponds to the string alice. The resulting output is the DN "cn=alice,ou=engineering,dc=example,dc=com".

A user with username bob@DBA.EXAMPLE.COM matches the second document. The regex capture group {0} corresponds to the string bob. The resulting output is the LDAP query "ou=dba,dc=example,dc=com??one?(user=bob)". mongod executes this query against the LDAP server, returning the result "cn=bob,ou=dba,dc=example,dc=com".

If --ldapUserToDNMapping is unset, mongod applies no transformations to the username when attempting to authenticate or authorize a user against the LDAP server.

This setting can be configured on a running mongod using the setParameter database command.

--ldapAuthzQueryTemplate <string>

New in version 3.4: Available in MongoDB Enterprise only.

A relative LDAP query URL formatted conforming to RFC4515 and RFC4516 that mongod executes to obtain the LDAP groups to which the authenticated user belongs to. The query is relative to the host or hosts specified in --ldapServers.

Use the {USER} placeholder in the URL to substitute the authenticated username, or the transformed username if a username mapping is specified.

When constructing the query URL, ensure that the order of LDAP parameters respects RFC4516:

[ dn  [ ? [attributes] [ ? [scope] [ ? [filter] [ ? [Extensions] ] ] ] ] ]

If your query includes an attribute, mongod assumes that the query retrieves a the DNs which this entity is member of.

If your query does not include an attribute, mongod assumes the query retrieves all entities which the user is member of.

For each LDAP DN returned by the query, mongod assigns the authorized user a corresponding role on the admin database. If a role on the on the admin database exactly matches the DN, mongod grants the user the roles and privileges assigned to that role. See the db.createRole() method for more information on creating roles.

Example

This LDAP query returns any groups listed in the LDAP user object’s memberOf attribute.

"{USER}?memberOf?base"

Your LDAP configuration may not include the memberOf attribute as part of the user schema, may possess a different attribute for reporting group membership, or may not track group membership through attributes. Configure your query with respect to your own unique LDAP configuration.

If unset, mongod cannot authorize users using LDAP.

This setting can be configured on a running mongod using the setParameter database command.

Note

An explanation of RFC4515, RFC4516 or LDAP queries is out of scope for the MongoDB Documentation. Please review the RFC directly or use your preferred LDAP resource.

Storage Options

--storageEngine string

Default: wiredTiger

Changed in version 4.0: MongoDB deprecates the MMAPv1 storage engine.

Specifies the storage engine for the mongod database. Available values include:

Value Description
wiredTiger To specify the WiredTiger Storage Engine.
inMemory

To specify the In-Memory Storage Engine.

New in version 3.2: Available in MongoDB Enterprise only.

mmapv1 (Deprecated in MongoDB 4.0) To specify the MMAPv1 Storage Engine.

If you attempt to start a mongod with a --dbpath that contains data files produced by a storage engine other than the one specified by --storageEngine, mongod will refuse to start.

--dbpath <path>

Default: /data/db on Linux and macOS, \data\db on Windows

The directory where the mongod instance stores its data.

If using the default configuration file included with a package manager installation of MongoDB, the corresponding storage.dbPath setting uses a different default.

Changed in version 3.0: The files in --dbpath must correspond to the storage engine specified in --storageEngine. If the data files do not correspond to --storageEngine, mongod will refuse to start.

--directoryperdb

Uses a separate directory to store data for each database. The directories are under the --dbpath directory, and each subdirectory name corresponds to the database name.

Changed in version 3.0: To change the --directoryperdb option for existing deployments, you must restart the mongod instances with the new --directoryperdb value and a new data directory (--dbpath <new path>), and then repopulate the data.

  • For standalone instances, you can use mongodump on the existing instance, stop the instance, restart with the new --directoryperdb value and a new data directory, and use mongorestore to populate the new data directory.
  • For replica sets, you can update in a rolling manner by stopping a secondary member, restart with the new --directoryperdb value and a new data directory, and use initial sync to populate the new data directory. To update all members, start with the secondary members first. Then step down the primary, and update the stepped-down member.

Not available for mongod instances that use the in-memory storage engine.

--noIndexBuildRetry

Changed in version 4.0: --noIndexBuildRetry cannot be used in conjunction with --replSet; i.e., you cannot use --noIndexBuildRetry for a mongod instance that is part of a replica set.

Stops the mongod standalone instance from rebuilding incomplete indexes on the next start up. This applies in cases where the mongod restarts after it has shut down or stopped in the middle of an index build. In such cases, the mongod always removes any incomplete indexes, and then also, by default, attempts to rebuild them. To stop the mongod from rebuilding incomplete indexes on start up, include this option on the command-line.

The --noIndexBuildRetry only applies to standalones.

Not available for mongod instances that use the in-memory storage engine.

--noprealloc

Only available for the MMAPv1 storage engine.

Deprecated since version 2.6: By default, MongoDB does not preallocate data files. The option exists for compatibility.

Disables the preallocation of data files.

--nssize <value>

Default: 16

Only available for the MMAPv1 storage engine.

Specifies the default size for namespace files, which are files that end in .ns. Each collection and index counts as a namespace.

Use this setting to control size for newly created namespace files. This option has no impact on existing files. The maximum size for a namespace file is 2047 megabytes. The default value of 16 megabytes provides for approximately 24,000 namespaces.

--quota

Only available for the MMAPv1 storage engine.

Enables a maximum limit for the number data files each database can have. When running with the --quota option, MongoDB has a maximum of 8 data files per database. Adjust the quota with --quotaFiles.

--quotaFiles <number>

Default: 8

Only available for the MMAPv1 storage engine.

Modifies the limit on the number of data files per database. --quotaFiles option requires that you set --quota.

--smallfiles

Only available for the MMAPv1 storage engine.

Sets MongoDB to use a smaller default file size. The --smallfiles option reduces the initial size for data files and limits the maximum size to 512 megabytes. --smallfiles also reduces the size of each journal file from 1 gigabyte to 128 megabytes. Use --smallfiles if you have a large number of databases that each holds a small quantity of data.

The --smallfiles option can lead the mongod instance to create a large number of files, which can affect performance for larger databases.

--syncdelay <value>

Default: 60

Controls how much time can pass before MongoDB flushes data to the data files via an fsync operation.

Do not set this value on production systems. In almost every situation, you should use the default setting.

Warning

If you set --syncdelay to 0, MongoDB will not sync the memory mapped files to disk.

The mongod process writes data very quickly to the journal and lazily to the data files. --syncdelay has no effect on the journal files or journaling, but if --syncdelay is set to 0 the journal will eventually consume all available disk space. If you set --syncdelay to 0 for testing purposes, you should also set --nojournal to true.

The serverStatus command reports the background flush thread’s status via the backgroundFlushing field.

Not available for mongod instances that use the in-memory storage engine.

--upgrade

Upgrades the on-disk data format of the files specified by the --dbpath to the latest version, if needed.

This option only affects the operation of the mongod if the data files are in an old format.

In most cases you should not set this value, so you can exercise the most control over your upgrade process. See the MongoDB release notes for more information about the upgrade process.

--repair

Changed in version 4.0.3.

Runs a repair routine on all databases for a mongod instance. The operation attempts to salvage corrupt data as well as rebuilds all the indexes. The operation discards any corrupt data that cannot be salvaged.

Tip

If you are running with journaling enabled, there is almost never any need to run repair since the server can use the journal files to restore the data files to a clean state automatically. However, you may need to run repair in cases where you need to recover from a disk-level data corruption.

Warning

  • Before using --repair, make a backup copy of the dbpath directory.
  • Avoid running --repair against a replica set:
    • To repair a replica set member, if you have an intact copy of your data available (e.g. a recent backup or an intact member of the replica set), restore from that intact copy instead(see Resync a Member of a Replica Set).
    • If you do choose to run mongod --repair against a replica set member and the operation modifies the data or the metadata, you must still perform a full resync in order for the member to rejoin the replica set.
  • Only use mongod --repair (and the related operations repairDatabase command and db.repairDatabase() method) if you have no other options. The operations remove and do not save any corrupt data during the repair process.
  • If repair fails to complete for any reason, you must restart the instance using the --repair option.

For MMAPv1 storage engine, see also --repairpath.

--repairpath <path>

Default: A _tmp_repairDatabase_<num> directory under the dbPath.

Only available for the MMAPv1 storage engine.

Specifies a working directory that MongoDB will use during the --repair operation. When --repair completes, the --repairpath directory is empty, and dbPath contains the repaired files.

The --repairpath must be within the dbPath. You can specify a symlink to --repairpath to use a path on a different file system.

--journal

Enables the durability journal to ensure data files remain valid and recoverable. This option applies only when you specify the --dbpath option. mongod enables journaling by default.

Not available for mongod instances that use the in-memory storage engine.

If any voting member of a replica set uses the in-memory storage engine, you must set writeConcernMajorityJournalDefault to false.

Starting in version 4.0.13 (and 3.6.14), if a replica set member uses the in-memory storage engine (voting or non-voting) but the replica set has writeConcernMajorityJournalDefault set to true, the replica set member logs a startup warning.

--nojournal

Disables journaling. mongod enables journaling by default.

Not available for mongod instances that use the in-memory storage engine.

Starting in MongoDB 4.0, you cannot specify --nojournal option or storage.journal.enabled: false for replica set members that use the WiredTiger storage engine.

--journalOptions <arguments>

Only available for the MMAPv1 storage engine.

Provides functionality for testing. Not for general use, and will affect data file integrity in the case of abnormal system shutdown.

Not available for mongod instances that use the in-memory storage engine.

--journalCommitInterval <value>

Default: 100 or 30

Changed in version 3.2.

The maximum amount of time in milliseconds that the mongod process allows between journal operations. Values can range from 1 to 500 milliseconds. Lower values increase the durability of the journal, at the expense of disk performance. The default journal commit interval is 100 milliseconds.

On MMAPv1, if the journal is on a different block device (e.g. physical volume, RAID device, or LVM volume) than the data files, the default journal commit interval is 30 milliseconds. Additionally, on MMAPv1, when a write operation with j:true is pending, mongod will reduce commitIntervalMs to a third of the set value.

On WiredTiger, the default journal commit interval is 100 milliseconds. Additionally, a write that includes or implies j:true will cause an immediate sync of the journal. For details or additional conditions that affect the frequency of the sync for WiredTiger, see Journaling Process.

Not available for mongod instances that use the in-memory storage engine.

WiredTiger Options

--wiredTigerCacheSizeGB <float>

Defines the maximum size of the internal cache that WiredTiger will use for all data. The memory consumed by an index build (see maxIndexBuildMemoryUsageMegabytes) is separate from the WiredTiger cache memory. Starting in MongoDB 3.4, the values can range from 0.25 GB to 10000 GB and can be a float.

Starting in MongoDB 3.4, the default WiredTiger internal cache size is the larger of either:

  • 50% of (RAM - 1 GB), or
  • 256 MB.

For example, on a system with a total of 4GB of RAM the WiredTiger cache will use 1.5GB of RAM (0.5 * (4 GB - 1 GB) = 1.5 GB). Conversely, a system with a total of 1.25 GB of RAM will allocate 256 MB to the WiredTiger cache because that is more than half of the total RAM minus one gigabyte (0.5 * (1.25 GB - 1 GB) = 128 MB < 256 MB).

Note

In some instances, such as when running in a container, the database can have memory constraints that are lower than the total system memory. In such instances, this memory limit, rather than the total system memory, is used as the maximum RAM available.

To see the memory limit, see hostInfo.system.memLimitMB.

Avoid increasing the WiredTiger internal cache size above its default value.

With WiredTiger, MongoDB utilizes both the WiredTiger internal cache and the filesystem cache.

Via the filesystem cache, MongoDB automatically uses all free memory that is not used by the WiredTiger cache or by other processes.

Note

The --wiredTigerCacheSizeGB limits the size of the WiredTiger internal cache. The operating system will use the available free memory for filesystem cache, which allows the compressed MongoDB data files to stay in memory. In addition, the operating system will use any free RAM to buffer file system blocks and file system cache.

To accommodate the additional consumers of RAM, you may have to decrease WiredTiger internal cache size.

The default WiredTiger internal cache size value assumes that there is a single mongod instance per machine. If a single machine contains multiple MongoDB instances, then you should decrease the setting to accommodate the other mongod instances.

If you run mongod in a container (e.g. lxc, cgroups, Docker, etc.) that does not have access to all of the RAM available in a system, you must set --wiredTigerCacheSizeGB to a value less than the amount of RAM available in the container. The exact amount depends on the other processes running in the container. See memLimitMB.

--wiredTigerMaxCacheOverflowFileSizeGB <float>

Specifies the maximum size (in GB) for the “lookaside (or cache overflow) table” file WiredTigerLAS.wt.

The setting can accept the following values:

Value Description
0 The default value. If set to 0, the file size is unbounded.
number >= 0.1 The maximum size (in GB). If the WiredTigerLAS.wt file exceeds this size, mongod exits with a fatal assertion. You can clear the WiredTigerLAS.wt file and restart mongod.

To change the maximum size during runtime, use the wiredTigerMaxCacheOverflowSizeGB parameter.

Available in 4.0-series starting in 4.0.12

--wiredTigerJournalCompressor <compressor>

Default: snappy

New in version 3.0.0.

Specifies the type of compression to use to compress WiredTiger journal data.

Available compressors are:

For details on changing the journal compressor, see Change WiredTiger Journal Compressor.

--wiredTigerDirectoryForIndexes

New in version 3.0.0.

When you start mongod with --wiredTigerDirectoryForIndexes, mongod stores indexes and collections in separate subdirectories under the data (i.e. --dbpath) directory. Specifically, mongod stores the indexes in a subdirectory named index and the collection data in a subdirectory named collection.

By using a symbolic link, you can specify a different location for the indexes. Specifically, when mongod instance is not running, move the index subdirectory to the destination and create a symbolic link named index under the data directory to the new destination.

--wiredTigerCollectionBlockCompressor <compressor>

Default: snappy

New in version 3.0.0.

Specifies the default type of compression to use to compress collection data. You can override this on a per-collection basis when creating collections.

Available compressors are:

--wiredTigerCollectionBlockCompressor affects all collections created. If you change the value of --wiredTigerCollectionBlockCompressor on an existing MongoDB deployment, all new collections will use the specified compressor. Existing collections will continue to use the compressor specified when they were created, or the default compressor at that time.

--wiredTigerIndexPrefixCompression <boolean>

Default: true

New in version 3.0.0.

Enables or disables prefix compression for index data.

Specify true for --wiredTigerIndexPrefixCompression to enable prefix compression for index data, or false to disable prefix compression for index data.

The --wiredTigerIndexPrefixCompression setting affects all indexes created. If you change the value of --wiredTigerIndexPrefixCompression on an existing MongoDB deployment, all new indexes will use prefix compression. Existing indexes are not affected.

Replication Options

--replSet <setname>

Configures replication. Specify a replica set name as an argument to this set. All hosts in the replica set must have the same set name.

Starting in MongoDB 4.0,

If your application connects to more than one replica set, each set must have a distinct name. Some drivers group replica set connections by replica set name.

--oplogSize <value>

Specifies a maximum size in megabytes for the replication operation log (i.e., the oplog).

Note

Starting in MongoDB 4.0, the oplog can grow past its configured size limit to avoid deleting the majority commit point.

By default, the mongod process creates an oplog based on the maximum amount of space available. For 64-bit systems, the oplog is typically 5% of available disk space.

Once the mongod has created the oplog for the first time, changing the --oplogSize option will not affect the size of the oplog.

To change the oplog size of a running replica set member, use the replSetResizeOplog administrative command. replSetResizeOplog enables you to resize the oplog dynamically without restarting the mongod process.

See Oplog Size for more information.

--replIndexPrefetch

Default: all

Storage Engine Specific Feature

--replIndexPrefetch is only available with the mmapv1 storage engine.

Determines which indexes secondary members of a replica set load into memory before applying operations from the oplog. By default secondaries load all indexes related to an operation into memory before applying operations from the oplog.

Set this option to one of the following:

Value Description
none Secondaries do not load indexes into memory.
all Secondaries load all indexes related to an operation.
_id_only Secondaries load no additional indexes into memory beyond the already existing _id index.
--enableMajorityReadConcern

Default: true

Starting in MongoDB 3.6, MongoDB enables support for "majority" read concern by default.

For MongoDB 4.0.3+ and 3.6.1+, you can disable read concern "majority" to prevent the storage cache pressure from immobilizing a deployment with a three-member primary-secondary-arbiter (PSA) architecture. For more information about disabling read concern "majority", see Disable Read Concern Majority.

To disable, set --enableMajorityReadConcern to false.

Important

In general, avoid disabling "majority" read concern unless necessary. However, if you have a three-member replica set with a primary-secondary-arbiter (PSA) architecture or a sharded cluster with a three-member PSA shards, disable to prevent the storage cache pressure from immobilizing the deployment.

Disabling "majority" read concern disables support for Change Streams.

Disabling "majority" does not affect multi-document transactions; i.e. you can specify read concern “majority” for multi-document transactions even if read concern “majority” is disabled.

--enableMajorityReadConcern has no effect for MongoDB versions: 4.0.0, 4.0.1, 4.0.2, 3.6.0.

Sharded Cluster Options

--configsvr

Required if starting a config server.

Declares that this mongod instance serves as the config server of a sharded cluster. When running with this option, clients (i.e. other cluster components) cannot write data to any database other than config and admin. The default port for a mongod with this option is 27019 and the default --dbpath directory is /data/configdb, unless specified.

Important

Starting in 3.4, you must deploy config servers as a replica set. The use of the deprecated mirrored mongod instances as config servers (SCCC) is no longer supported.

The replica set config servers (CSRS) must run the WiredTiger storage engine.

The --configsvr option creates a local oplog.

Do not use the --configsvr option with --shardsvr. Config servers cannot be a shard server.

Do not use the --configsvr with the skipShardingConfigurationChecks parameter. That is, if you are temporarily starting the mongod as a standalone for maintenance operations, include the parameter skipShardingConfigurationChecks and exclude --configsvr. Once maintenance has completed, remove the skipShardingConfigurationChecks parameter and restart with --configsvr.

--configsvrMode <string>

Available in MongoDB 3.2 version only

If set to sccc, indicates that the config servers are deployed as three mirrored mongod instances, even if one or more config servers is also a member of a replica set. configsvrMode only accepts the value sccc.

If unset, config servers running as replica sets expect to use the “config server replica set” protocol for writing to config servers, rather than the “mirrored mongod” write protocol.

--shardsvr

Required if starting a shard server.

Configures this mongod instance as a shard in a sharded cluster. The default port for these instances is 27018.

Important

Starting in MongoDB 3.6, you must deploy shards as replica sets. See the --replSet option to deploy mongod as part of a replica set.

Do not use the --shardsvr with the skipShardingConfigurationChecks parameter. That is, if you are temporarily starting the mongod as a standalone for maintenance operations, include the parameter skipShardingConfigurationChecks and exclude --shardsvr. Once maintenance has completed, remove the skipShardingConfigurationChecks parameter and restart with --shardsvr.

--moveParanoia

If specified, during chunk migration, a shard saves, to the moveChunk directory of the --dbpath, all documents migrated from that shard.

MongoDB does not automatically delete the data saved in the moveChunk directory.

--noMoveParanoia

Changed in version 3.2: Starting in 3.2, MongoDB uses --noMoveParanoia as the default.

During chunk migration, a shard does not save documents migrated from the shard.

TLS/SSL Options

See

Configure mongod and mongos for TLS/SSL for full documentation of MongoDB’s support.

--sslOnNormalPorts

Deprecated since version 2.6: Use --sslMode requireSSL instead.

Enables TLS/SSL for mongod.

With --sslOnNormalPorts, a mongod requires TLS/SSL encryption for all connections on the default MongoDB port, or the port specified by --port. By default, --sslOnNormalPorts is disabled.

For more information about TLS/SSL and MongoDB, see Configure mongod and mongos for TLS/SSL and TLS/SSL Configuration for Clients .

--sslMode <mode>

New in version 2.6.

Enables TLS/SSL or mixed TLS/SSL used for all network connections. The argument to the --sslMode option can be one of the following:

Value Description
disabled The server does not use TLS/SSL.
allowSSL Connections between servers do not use TLS/SSL. For incoming connections, the server accepts both TLS/SSL and non-TLS/non-SSL.
preferSSL Connections between servers use TLS/SSL. For incoming connections, the server accepts both TLS/SSL and non-TLS/non-SSL.
requireSSL The server uses and accepts only TLS/SSL encrypted connections.

Starting in version 3.4, if --sslCAFile or ssl.CAFile is not specified and you are not using x.509 authentication, the system-wide CA certificate store will be used when connecting to an TLS/SSL-enabled server.

If using x.509 authentication, --sslCAFile or ssl.CAFile must be specified unless using --sslCertificateSelector.

For more information about TLS/SSL and MongoDB, see Configure mongod and mongos for TLS/SSL and TLS/SSL Configuration for Clients .

--sslPEMKeyFile <filename>

Specifies the .pem file that contains both the TLS/SSL certificate and key.

Changed in version 4.0: Starting with MongoDB 4.0 on macOS or Windows, you can use the --sslCertificateSelector option to specify a certificate from the operating system’s secure certificate store instead of a PEM key file. --sslPEMKeyFile and --sslCertificateSelector options are mutually exclusive. You can only specify one.

  • On Linux/BSD, you must specify --sslPEMKeyFile when TLS/SSL is enabled.

  • On Windows or macOS, you must specify either --sslPEMKeyFile or --sslCertificateSelector when TLS/SSL is enabled.

    Important

    For Windows only, MongoDB 4.0 and later do not support encrypted PEM files. The mongod fails to start if it encounters an encrypted PEM file. To securely store and access a certificate for use with TLS/SSL on Windows, use --sslCertificateSelector.

For more information about TLS/SSL and MongoDB, see Configure mongod and mongos for TLS/SSL and TLS/SSL Configuration for Clients .

--sslPEMKeyPassword <value>

Specifies the password to de-crypt the certificate-key file (i.e. --sslPEMKeyFile). Use the --sslPEMKeyPassword option only if the certificate-key file is encrypted. In all cases, the mongod will redact the password from all logging and reporting output.

Starting in MongoDB 4.0:

  • On Linux/BSD, if the private key in the PEM file is encrypted and you do not specify the --sslPEMKeyPassword option, MongoDB will prompt for a passphrase. See TLS/SSL Certificate Passphrase.
  • On macOS, if the private key in the PEM file is encrypted, you must explicitly specify the --sslPEMKeyPassword option. Alternatively, you can use a certificate from the secure system store (see --sslCertificateSelector) instead of a PEM key file or use an unencrypted PEM file.
  • On Windows, MongoDB does not support encrypted certificates. The mongod fails if it encounters an encrypted PEM file. Use --sslCertificateSelector instead.

For more information about TLS/SSL and MongoDB, see Configure mongod and mongos for TLS/SSL and TLS/SSL Configuration for Clients .

--clusterAuthMode <option>

Default: keyFile

New in version 2.6.

The authentication mode used for cluster authentication. If you use internal x.509 authentication, specify so here. This option can have one of the following values:

Value Description
keyFile Use a keyfile for authentication. Accept only keyfiles.
sendKeyFile For rolling upgrade purposes. Send a keyfile for authentication but can accept both keyfiles and x.509 certificates.
sendX509 For rolling upgrade purposes. Send the x.509 certificate for authentication but can accept both keyfiles and x.509 certificates.
x509 Recommended. Send the x.509 certificate for authentication and accept only x.509 certificates.

Starting in version 3.4, if --sslCAFile or ssl.CAFile is not specified and you are not using x.509 authentication, the system-wide CA certificate store will be used when connecting to an TLS/SSL-enabled server.

If using x.509 authentication, --sslCAFile or ssl.CAFile must be specified unless using --sslCertificateSelector.

For more information about TLS/SSL and MongoDB, see Configure mongod and mongos for TLS/SSL and TLS/SSL Configuration for Clients .

--sslClusterFile <filename>

Specifies the .pem file that contains the x.509 certificate-key file for membership authentication for the cluster or replica set.

Changed in version 4.0: Starting with MongoDB 4.0 on macOS or Windows, you can use the --sslClusterCertificateSelector option to specify a certificate from the operating system’s secure certificate store instead of a PEM key file. --sslClusterFile and --sslClusterCertificateSelector options are mutually exclusive. You can only specify one.

If --sslClusterFile does not specify the .pem file for internal cluster authentication or the alternative --sslClusterCertificateSelector, the cluster uses the .pem file specified in the --sslPEMKeyFile option or the certificate returned by the --sslCertificateSelector.

If using x.509 authentication, --sslCAFile or ssl.CAFile must be specified unless using --sslCertificateSelector.

For more information about TLS/SSL and MongoDB, see Configure mongod and mongos for TLS/SSL and TLS/SSL Configuration for Clients .

Important

For Windows only, MongoDB 4.0 and later do not support encrypted PEM files. The mongod fails to start if it encounters an encrypted PEM file. To securely store and access a certificate for use with membership authentication on Windows, use --sslClusterCertificateSelector.

--sslCertificateSelector <parameter>=<value>

New in version 4.0: Available on Windows and macOS as an alternative to --sslPEMKeyFile.

Specifies a certificate property to select a matching certificate from the operating system’s secure certificate store to use for TLS/SSL.

--sslPEMKeyFile and --sslCertificateSelector options are mutually exclusive. You can only specify one.

--sslCertificateSelector accepts an argument of the format <property>=<value> where the property can be one of the following:

Property Value type Description
subject ASCII string Subject name or common name on certificate
thumbprint hex string

A sequence of bytes, expressed as hexadecimal, used to identify a public key by its SHA-1 digest.

The thumbprint is sometimes referred to as a fingerprint.

When using the system SSL certificate store, OCSP (Online Certificate Status Protocol) is used to validate the revocation status of certificates.

The mongod searches the operating system’s secure certificate store for the CA certificates required to validate the full certificate chain of the specified TLS/SSL certificate. Specifically, the secure certificate store must contain the root CA and any intermediate CA certificates required to build the full certificate chain to the TLS/SSL certificate. Do not use --sslCAFile or --sslClusterCAFile to specify the root and intermediate CA certificate

For example, if the TLS/SSL certificate was signed with a single root CA certificate, the secure certificate store must contain that root CA certificate. If the TLS/SSL certificate was signed with an intermediate CA certificate, the secure certificate store must contain the intermedia CA certificate and the root CA certificate.

--sslClusterCertificateSelector <parameter>=<value>

New in version 4.0: Available on Windows and macOS as an alternative to --sslClusterFile.

Specifies a certificate property to select a matching certificate from the operating system’s secure certificate store to use for internal x.509 membership authentication.

--sslClusterFile and --sslClusterCertificateSelector options are mutually exclusive. You can only specify one.

--sslClusterCertificateSelector accepts an argument of the format <property>=<value> where the property can be one of the following:

Property Value type Description
subject ASCII string Subject name or common name on certificate
thumbprint hex string

A sequence of bytes, expressed as hexadecimal, used to identify a public key by its SHA-1 digest.

The thumbprint is sometimes referred to as a fingerprint.

The mongod searches the operating system’s secure certificate store for the CA certificates required to validate the full certificate chain of the specified cluster certificate. Specifically, the secure certificate store must contain the root CA and any intermediate CA certificates required to build the full certificate chain to the cluster certificate. Do not use --sslCAFile or --sslClusterCAFile to specify the root and intermediate CA certificate.

For example, if the cluster certificate was signed with a single root CA certificate, the secure certificate store must contain that root CA certificate. If the cluster certificate was signed with an intermediate CA certificate, the secure certificate store must contain the intermedia CA certificate and the root CA certificate.

--sslClusterPassword <value>

New in version 2.6.

Specifies the password to de-crypt the x.509 certificate-key file specified with --sslClusterFile. Use the --sslClusterPassword option only if the certificate-key file is encrypted. In all cases, the mongod will redact the password from all logging and reporting output.

Starting in MongoDB 4.0:

For more information about TLS/SSL and MongoDB, see Configure mongod and mongos for TLS/SSL and TLS/SSL Configuration for Clients .

--sslCAFile <filename>

Specifies the .pem file that contains the root certificate chain from the Certificate Authority. Specify the file name of the .pem file using relative or absolute paths.

Windows/macOS Only
If using --sslCertificateSelector and/or --sslClusterCertificateSelector, do not use --sslCAFile to specify the root and intermediate CA certificates. Store all CA certificates required to validate the full trust chain of the --sslCertificateSelector and/or --sslClusterCertificateSelector certificates in the secure certificate store.

For more information about TLS/SSL and MongoDB, see Configure mongod and mongos for TLS/SSL and TLS/SSL Configuration for Clients .

--sslClusterCAFile <filename>

New in version 4.0.3: Specifies the .pem file that contains the root certificate chain from the Certificate Authority used to validate the certificate presented by a client establishing a connection. Specify the file name of the .pem file using relative or absolute paths. --sslClusterCAFile requires that --sslCAFile is set.

If --sslClusterCAFile does not specify the .pem file for validating the certificate from a client establishing a connection, the cluster uses the .pem file specified in the --sslCAFile option.

--sslClusterCAFile lets you use separate Certificate Authorities to verify the client to server and server to client portions of the TLS/SSL handshake.

Windows/macOS Only
If using --sslCertificateSelector and/or --sslClusterCertificateSelector, do not use --sslClusterCAFile to specify the root and intermediate CA certificates. Store all CA certificates required to validate the full trust chain of the --sslCertificateSelector and/or --sslClusterCertificateSelector certificates in the secure certificate store.

For more information about TLS/SSL and MongoDB, see Configure mongod and mongos for TLS/SSL and TLS/SSL Configuration for Clients .

--sslCRLFile <filename>

Specifies the .pem file that contains the Certificate Revocation List. Specify the file name of the .pem file using relative or absolute paths.

Note

Starting in MongoDB 4.0, you cannot specify --sslCRLFile on macOS. Use --sslCertificateSelector instead.

For more information about TLS/SSL and MongoDB, see Configure mongod and mongos for TLS/SSL and TLS/SSL Configuration for Clients .

--sslAllowInvalidCertificates

Bypasses the validation checks for TLS/SSL certificates on other servers in the cluster and allows the use of invalid certificates to connect.

Note

Starting in MongoDB 4.0, if you specify --sslAllowInvalidCertificates or ssl.allowInvalidCertificates: true when using x.509 authentication, an invalid certificate is only sufficient to establish a TLS/SSL connection but is insufficient for authentication.

When using the --sslAllowInvalidCertificates setting, MongoDB logs a warning regarding the use of the invalid certificate.

For more information about TLS/SSL and MongoDB, see Configure mongod and mongos for TLS/SSL and TLS/SSL Configuration for Clients .

--sslAllowInvalidHostnames

New in version 3.0.

Disables the validation of the hostnames in TLS/SSL certificates, when connecting to other members of the replica set or sharded cluster for inter-process authentication. This allows mongod to connect to other members if the hostnames in their certificates do not match their configured hostname.

For more information about TLS/SSL and MongoDB, see Configure mongod and mongos for TLS/SSL and TLS/SSL Configuration for Clients .

--sslAllowConnectionsWithoutCertificates

For clients that do not present certificates, mongod bypasses TLS/SSL certificate validation when establishing the connection.

For clients that present a certificate, however, mongod performs certificate validation using the root certificate chain specified by --sslCAFile and reject clients with invalid certificates.

Use the --sslAllowConnectionsWithoutCertificates option if you have a mixed deployment that includes clients that do not or cannot present certificates to the mongod.

For more information about TLS/SSL and MongoDB, see Configure mongod and mongos for TLS/SSL and TLS/SSL Configuration for Clients .

--sslDisabledProtocols <protocol(s)>

New in version 3.0.7.

Prevents a MongoDB server running with TLS/SSL from accepting incoming connections that use a specific protocol or protocols. To specify multiple protocols, use a comma separated list of protocols.

--sslDisabledProtocols recognizes the following protocols: TLS1_0, TLS1_1, TLS1_2, and starting in version 4.0.4 (and 3.6.9 and 3.4.24), TLS1_3.

  • On macOS, you cannot disable TLS1_1 and leave both TLS1_0 and TLS1_2 enabled. You must disable at least one of the other two, for example, TLS1_0,TLS1_1.
  • To list multiple protocols, specify as a comma separated list of protocols. For example TLS1_0,TLS1_1.
  • Specifying an unrecognized protocol will prevent the server from starting.
  • The specified disabled protocols overrides any default disabled protocols.

Starting in version 4.0, MongoDB disables the use of TLS 1.0 if TLS 1.1+ is available on the system. To enable the disabled TLS 1.0, specify none to --sslDisabledProtocols. See Disable TLS 1.0.

Members of replica sets and sharded clusters must speak at least one protocol in common.

--sslFIPSMode

Directs the mongod to use the FIPS mode of the TLS/SSL library. Your system must have a FIPS compliant library to use the --sslFIPSMode option.

Note

FIPS-compatible TLS/SSL is available only in MongoDB Enterprise. See Configure MongoDB for FIPS for more information.

Profiler Options

--profile <level>

Default: 0

Configures the database profiler level. The following profiler levels are available:

Level Description
0 The profiler is off and does not collect any data. This is the default profiler level.
1 The profiler collects data for operations that take longer than the value of slowms.
2 The profiler collects data for all operations.

Important

Profiling can impact performance and shares settings with the system log. Carefully consider any performance and security implications before configuring and enabling the profiler on a production deployment.

See Profiler Overhead for more information on potential performance degradation.

--slowms <integer>

Default: 100

The slow operation time threshold, in milliseconds. Operations that run for longer than this threshold are considered slow.

When logLevel is set to 0, MongoDB records slow operations to the diagnostic log at a rate determined by slowOpSampleRate. Starting in MongoDB 4.0.6, the secondaries of replica sets log all oplog entry messages that take longer than the slow operation threshold to apply regardless of the sample rate.

At higher logLevel settings, all operations appear in the diagnostic log regardless of their latency with the following exception: the logging of slow oplog entry messages by the secondaries. The secondaries log only the slow oplog entries; increasing the logLevel does not log all oplog entries.

For mongod instances, --slowms affects the diagnostic log and, if enabled, the profiler.

--slowOpSampleRate <double>

Default: 1.0

The fraction of slow operations that should be profiled or logged. --slowOpSampleRate accepts values between 0 and 1, inclusive.

--slowOpSampleRate does not affect the slow oplog entry logging by the secondary members of a replica set. Secondary members log all oplog entries that take longer than the slow operation threshold regardless of the --slowOpSampleRate.

For mongod instances, --slowOpSampleRate affects the diagnostic log and, if enabled, the profiler.

Audit Options

--auditDestination

Enables auditing and specifies where mongod sends all audit events.

--auditDestination can have one of the following values:

Value Description
syslog

Output the audit events to syslog in JSON format. Not available on Windows. Audit messages have a syslog severity level of info and a facility level of user.

The syslog message limit can result in the truncation of audit messages. The auditing system will neither detect the truncation nor error upon its occurrence.

console Output the audit events to stdout in JSON format.
file Output the audit events to the file specified in --auditPath in the format specified in --auditFormat.

Note

Available only in MongoDB Enterprise and MongoDB Atlas.

--auditFormat

New in version 2.6.

Specifies the format of the output file for auditing if --auditDestination is file. The --auditFormat option can have one of the following values:

Value Description
JSON Output the audit events in JSON format to the file specified in --auditPath.
BSON Output the audit events in BSON binary format to the file specified in --auditPath.

Printing audit events to a file in JSON format degrades server performance more than printing to a file in BSON format.

Note

Available only in MongoDB Enterprise and MongoDB Atlas.

--auditPath

New in version 2.6.

Specifies the output file for auditing if --auditDestination has value of file. The --auditPath option can take either a full path name or a relative path name.

Note

Available only in MongoDB Enterprise and MongoDB Atlas.

--auditFilter

New in version 2.6.

Specifies the filter to limit the types of operations the audit system records. The option takes a string representation of a query document of the form:

{ <field1>: <expression1>, ... }

The <field> can be any field in the audit message, including fields returned in the param document. The <expression> is a query condition expression.

To specify an audit filter, enclose the filter document in single quotes to pass the document as a string.

To specify the audit filter in a configuration file, you must use the YAML format of the configuration file.

Note

Available only in MongoDB Enterprise and MongoDB Atlas.

SNMP Options

Note

MongoDB Enterprise on macOS does not include support for SNMP due to SERVER-29352.

--snmp-disabled

Disables SNMP access to mongod. The option is incompatible with --snmp-subagent and --snmp-master.

New in version 4.0.6.

--snmp-subagent

Runs SNMP as a subagent. The option is incompatible with --snmp-disabled.

--snmp-master

Runs SNMP as a master. The option is incompatible with --snmp-disabled.

inMemory Options

--inMemorySizeGB <float>

Default: 50% of physical RAM less 1 GB

Changed in version 3.4: Values can range from 256MB to 10TB and can be a float.

Maximum amount of memory to allocate for in-memory storage engine data, including indexes, oplog if the mongod is part of replica set, replica set or sharded cluster metadata, etc.

By default, the in-memory storage engine uses 50% of physical RAM minus 1 GB.

Enterprise Feature

Available in MongoDB Enterprise only.

Encryption Key Management Options

--enableEncryption

Default: false

New in version 3.2.

Enables encryption for the WiredTiger storage engine. This option must be enabled in order to pass in encryption keys and configurations.

Enterprise Feature

Available in MongoDB Enterprise only.

--encryptionCipherMode <string>

Default: AES256-CBC

New in version 3.2.

The cipher mode to use for encryption at rest:

Mode Description
AES256-CBC 256-bit Advanced Encryption Standard in Cipher Block Chaining Mode
AES256-GCM

256-bit Advanced Encryption Standard in Galois/Counter Mode

Changed in version 4.0: MongoDB Enterprise on Windows no longer supports AES256-GCM. This cipher is now available only on Linux.

Enterprise Feature

Available in MongoDB Enterprise only.

--encryptionKeyFile <string>

New in version 3.2.

The path to the local keyfile when managing keys via process other than KMIP. Only set when managing keys via process other than KMIP. If data is already encrypted using KMIP, MongoDB will throw an error.

Requires --enableEncryption.

Enterprise Feature

Available in MongoDB Enterprise only.

--kmipKeyIdentifier <string>

New in version 3.2.

Unique KMIP identifier for an existing key within the KMIP server. Include to use the key associated with the identifier as the system key. You can only use the setting the first time you enable encryption for the mongod instance. Requires --enableEncryption.

If unspecified, MongoDB will request that the KMIP server create a new key to utilize as the system key.

If the KMIP server cannot locate a key with the specified identifier or the data is already encrypted with a key, MongoDB will throw an error

Enterprise Feature

Available in MongoDB Enterprise only.

--kmipRotateMasterKey <boolean>

Default: false

New in version 3.2.

If true, rotate the master key and re-encrypt the internal keystore.

Enterprise Feature

Available in MongoDB Enterprise only.

--kmipServerName <string>

New in version 3.2.

Hostname or IP address of the KMIP server to connect to. Requires --enableEncryption.

Starting in MongoDB 4.0.14, you can specify multiple KMIP servers as a comma-separated list, e.g. server1.example.com,server2.example.com. On startup, the mongod will attempt to establish a connection to each server in the order listed, and will select the first server to which it can successfully establish a connection. KMIP server selection occurs only at startup.

When connecting to a KMIP server, the mongod verifies that the specified --kmipServerName matches the Subject Alternative Name SAN (or, if SAN is not present, the Common Name CN) in the certificate presented by the KMIP server. If SAN is present, mongod does not match against the CN. If the hostname does not match the SAN (or CN), the mongod will fail to connect.

Enterprise Feature

Available in MongoDB Enterprise only.

--kmipPort <number>

Default: 5696

New in version 3.2.

Port number to use to communicate with the KMIP server. Requires --kmipServerName. Requires --enableEncryption.

If specifying multiple KMIP servers with --kmipServerName, the mongod will use the port specified with --kmipPort for all provided KMIP servers.

Enterprise Feature

Available in MongoDB Enterprise only.

--kmipClientCertificateSelector <string>

New in version 4.0: Available on Windows and macOS as an alternative to --kmipClientCertificateFile.

--kmipClientCertificateFile and --kmipClientCertificateSelector options are mutually exclusive. You can only specify one.

Specifies a certificate property in order to select a matching certificate from the operating system’s certificate store to authenticate MongoDB to the KMIP server.

--kmipClientCertificateSelector accepts an argument of the format <property>=<value> where the property can be one of the following:

Property Value type Description
subject ASCII string Subject name or common name on certificate
thumbprint hex string

A sequence of bytes, expressed as hexadecimal, used to identify a public key by its SHA-1 digest.

The thumbprint is sometimes referred to as a fingerprint.

Enterprise Feature

Available in MongoDB Enterprise only.

--kmipClientCertificateFile <string>

New in version 3.2.

String containing the path to the client certificate used for authenticating MongoDB to the KMIP server. Requires that a kmipServerName be provided.

Note

Starting in 4.0, on macOS or Windows, you can use a certificate from the operating system’s secure store instead of a PEM key file. See --kmipClientCertificateSelector.

Enterprise Feature

Available in MongoDB Enterprise only.

--kmipClientCertificatePassword <string>

New in version 3.2.

The password (if one exists) for the client certificate passed into kmipClientCertificateFile. Is used for authenticating MongoDB to the KMIP server. Requires that a kmipClientCertificateFile be provided.

Enterprise Feature

Available in MongoDB Enterprise only.

--kmipServerCAFile <string>

New in version 3.2.

Path to CA File. Used for validating secure client connection to KMIP server.

Note

Starting in 4.0, on macOS or Windows, you can use a certificate from the operating system’s secure store instead of a PEM key file. See --kmipClientCertificateSelector. When using the secure store, you do not need to, but can, also specify the --kmipServerCAFile.