.TH mongodb-parameters 5 .SH MONGODB SERVER PARAMETERS .SH SYNOPSIS MongoDB provides a number of configuration options that you can set using: .RS .IP \(bu 2 the \fBsetParameter\f1\f1 command: .IP .EX db.adminCommand( { setParameter: 1, : } ) .EE .IP \(bu 2 the \fBsetParameter\f1\f1 configuration setting: .IP .EX setParameter: : ... .EE .IP \(bu 2 the \fB\-\-setParameter\f1 command\-line option for \fBmongod\f1\f1 and \fBmongos\f1\f1: .IP .EX mongod \-\-setParameter = mongos \-\-setParameter = .EE .RE .PP For additional configuration options, see \fBConfiguration File Options\f1, \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .SH PARAMETERS .SS AUTHENTICATION PARAMETERS .PP \fBauthenticationMechanisms\f1 .RS .PP Remove support for the deprecated \fBMONGODB\-CR\f1 authentication mechanism. .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Specifies the list of authentication mechanisms the server accepts. Set this to one or more of the following values. If you specify multiple values, use a comma\-separated list and no spaces. For descriptions of the authentication mechanisms, see \fBAuthentication\f1\&. .RS .IP \(bu 2 .RS .IP \(bu 4 Value .IP \(bu 4 Description .RE .IP \(bu 2 .RS .IP \(bu 4 \fBSCRAM\-SHA\-1\f1 .IP \(bu 4 RFC 5802 (https://tools.ietf.org/html/rfc5802) standard Salted Challenge Response Authentication Mechanism using the SHA\-1 hash function. .RE .IP \(bu 2 .RS .IP \(bu 4 \fBSCRAM\-SHA\-256\f1 .IP \(bu 4 RFC 7677 (https://tools.ietf.org/html/rfc7677) standard Salted Challenge Response Authentication Mechanism using the SHA\-256 hash function. .IP Requires featureCompatibilityVersion set to \fB4.0\f1\&. .RE .IP \(bu 2 .RS .IP \(bu 4 \fBMONGODB\-X509\f1 .IP \(bu 4 MongoDB TLS/SSL certificate authentication. .RE .IP \(bu 2 .RS .IP \(bu 4 \fBGSSAPI\f1 (Kerberos) .IP \(bu 4 External authentication using Kerberos. This mechanism is available only in MongoDB Enterprise (http://www.mongodb.com/products/mongodb\-enterprise\-advanced?tck=docs_server)\&. .RE .IP \(bu 2 .RS .IP \(bu 4 \fBPLAIN\f1 (LDAP SASL) .IP \(bu 4 External authentication using LDAP. You can also use \fBPLAIN\f1 for authenticating in\-database users. \fBPLAIN\f1 transmits passwords in plain text. This mechanism is available only in MongoDB Enterprise (http://www.mongodb.com/products/mongodb\-enterprise\-advanced?tck=docs_server)\&. .RE .RE .PP You can only set \fBauthenticationMechanisms\f1\f1 during start\-up. .PP For example, to specify both \fBPLAIN\f1 and \fBSCRAM\-SHA\-256\f1 as the authentication mechanisms, use the following command: .PP .EX mongod \-\-setParameter authenticationMechanisms=PLAIN,SCRAM\-SHA\-256 \-\-auth .EE .RE .PP \fBclusterAuthMode\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Set the \fBclusterAuthMode\f1\f1 to either \fBsendX509\f1 or \fBx509\f1\&. Useful during \fBrolling upgrade to use x509 for membership authentication\f1 to minimize downtime. .PP For more information about TLS/SSL and MongoDB, see \fBConfigure mongod\f1 and mongos\f1 for TLS/SSL\f1 and \fBTLS/SSL Configuration for Clients\f1 . .PP .EX db.adminCommand( { setParameter: 1, clusterAuthMode: "sendX509" } ) .EE .RE .PP \fBenableLocalhostAuthBypass\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Specify \fB0\f1 or \fBfalse\f1 to disable localhost authentication bypass. Enabled by default. .PP \fBenableLocalhostAuthBypass\f1\f1 is not available using \fBsetParameter\f1\f1 database command. Use the \fBsetParameter\f1\f1 option in the configuration file or the \fB\-\-setParameter\f1\f1 option on the command line. .PP See \fBLocalhost Exception\f1 for more information. .RE .PP \fBKeysRotationIntervalSec\f1 .RS .PP \fIDefault\f1: 7776000 seconds (90 days) .PP Specifies the number of seconds for which an HMAC signing key (https://en.wikipedia.org/wiki/Hash\-based_message_authentication_code) is valid before rotating to the next one. This parameter is intended primarily to facilitate authentication testing. .PP You can only set \fBKeysRotationIntervalSec\f1\f1 during start\-up, and cannot change this setting with the \fBsetParameter\f1\f1 database command. .RE .PP \fBldapForceMultiThreadMode\f1 .RS .PP \fIDefault\f1: false .PP Enables the performance of concurrent LDAP operations. .PP Only if you are certain that your instance of \fBlibldap\f1 is safe to use in this mode, enable this flag. You may experience crashes of the MongoDB process if the \fBlibldap\f1 version you are using is not thread safe. .PP You must use \fBldapForceMultiThreadMode\f1\f1 to use LDAP connection pool. To enable LDAP connection pool, set \fBldapForceMultiThreadMode\f1\f1 and \fBldapUseConnectionPool\f1\f1 to \fBtrue\f1\&. .PP If you have any concerns regarding your MongoDB version, OS version or libldap version, please contact MongoDB Support. .RE .PP \fBldapUserCacheInvalidationInterval\f1 .RS .PP Starting in MongoDB 5.2, the update interval for cached user information retrieved from an LDAP server depends on \fBldapShouldRefreshUserCacheEntries\f1\f1: .RS .IP \(bu 2 If true, use \fBldapUserCacheRefreshInterval\f1\f1\&. .IP \(bu 2 If false, use \fBldapUserCacheInvalidationInterval\f1\f1\&. .RE .PP For use with MongoDB deployments using \fBLDAP Authorization\f1\&. Available for \fBmongod\f1\f1 instances only. .PP The interval (in seconds) that the \fBmongod\f1\f1 instance waits between external user cache flushes. After MongoDB flushes the external user cache, MongoDB reacquires authorization data from the LDAP server the next time an LDAP\-authorized user issues an operation. .PP Increasing the value specified increases the amount of time MongoDB and the LDAP server can be out of sync, but reduces the load on the LDAP server. Conversely, decreasing the value specified decreases the time MongoDB and the LDAP server can be out of sync while increasing the load on the LDAP server. .PP Defaults to 30 seconds. .RE .PP \fBldapUserCacheRefreshInterval\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP \fIType\f1: integer .PP \fIDefault\f1: 30 seconds .PP Starting in MongoDB 5.2, the update interval for cached user information retrieved from an LDAP server depends on \fBldapShouldRefreshUserCacheEntries\f1\f1: .RS .IP \(bu 2 If true, use \fBldapUserCacheRefreshInterval\f1\f1\&. .IP \(bu 2 If false, use \fBldapUserCacheInvalidationInterval\f1\f1\&. .RE .PP For MongoDB deployments using \fBLDAP Authorization\f1\&. .PP The interval in seconds that \fBmongod\f1\f1 waits before refreshing the cached user information from the LDAP server. .PP The maximum interval is 86,400 seconds (24 hours). .PP For example, the following sets \fBldapUserCacheRefreshInterval\f1\f1 to \fB4000\f1 seconds: .PP .EX mongod \-\-setParameter ldapUserCacheRefreshInterval=4000 .EE .PP Or, if using the \fBsetParameter\f1\f1 command within \fBmongosh\f1\f1: .PP .EX db.adminCommand( { setParameter: 1, ldapUserCacheRefreshInterval: 4000 } ) .EE .RE .PP \fBldapUserCacheStalenessInterval\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP \fIType\f1: integer .PP \fIDefault\f1: 90 seconds .PP For MongoDB deployments using \fBLDAP Authorization\f1\&. .PP The interval in seconds that \fBmongod\f1\f1 retains the cached LDAP user information after the last cache refresh. .PP If more than \fBldapUserCacheStalenessInterval\f1\f1 seconds elapse without a successful refresh of the user information from the LDAP server, then \fBmongod\f1\f1: .RS .IP \(bu 2 Invalidates the cached LDAP user information. .IP \(bu 2 Is unavailable for LDAP users. LDAP users are unable to authenticate until \fBmongod\f1\f1 contacts the LDAP server. .RE .PP The maximum interval is 86,400 seconds (24 hours). .PP For example, the following sets \fBldapUserCacheStalenessInterval\f1\f1 to \fB4000\f1 seconds: .PP .EX mongod \-\-setParameter ldapUserCacheStalenessInterval=4000 .EE .PP Or, if using the \fBsetParameter\f1\f1 command within \fBmongosh\f1\f1: .PP .EX db.adminCommand( { setParameter: 1, ldapUserCacheStalenessInterval: 4000 } ) .EE .RE .PP \fBldapUseConnectionPool\f1 .RS .PP Specifies whether MongoDB should use connection pooling when connecting to the LDAP server for authentication/authorization. .PP \fBStarting in version 4.2\f1, MongoDB uses the following default values: .RS .IP \(bu 2 true on Windows. .IP \(bu 2 true on Linux where MongoDB Enterprise binaries are linked against \fBlibldap_r\f1\&. .IP \(bu 2 false on Linux where MongoDB Enterprise binaries are linked against \fBlibldap\f1\&. .RE .PP \fBIn earlier versions (versions 4.0.9+)\f1, the default value is \fBfalse\f1\&. .PP You can only set \fBldapUseConnectionPool\f1\f1 during start\-up, and cannot change this setting with the \fBsetParameter\f1\f1 database command. .RE .PP \fBldapConnectionPoolUseLatencyForHostPriority\f1 .RS .PP \fINew in version 4.2.1 and 4.0.13\f1 .PP \fIDefault\f1: true .PP A boolean that determines whether the LDAP connection pool (see \fBldapUseConnectionPool\f1\f1) should use latency of the LDAP servers to determine the connection order (from lowest latency to highest). .PP You can only set \fBldapConnectionPoolUseLatencyForHostPriority\f1\f1 during start\-up, and cannot change this setting during runtime with the \fBsetParameter\f1\f1 database command. .RE .PP \fBldapConnectionPoolMinimumConnectionsPerHost\f1 .RS .PP \fINew in version 4.2.1 and 4.0.13\f1 .PP \fIDefault\f1: 1 .PP The minimum number of connections to keep open to each LDAP server. .PP You can only set \fBldapConnectionPoolMinimumConnectionsPerHost\f1\f1 during start\-up, and cannot change this setting during runtime with the \fBsetParameter\f1\f1 database command. .RE .PP \fBldapConnectionPoolMaximumConnectionsPerHost\f1 .RS .PP \fINew in version 4.2.1 and 4.0.13\f1 .PP \fIChanged starting in MongoDB versions 4.4.15, 5.0.9, and 6.0.0\f1 Changed default value to \fB2147483647\f1\&. In previous versions, the default is unset. .PP \fIDefault\f1: 2147483647 .PP The maximum number of connections to keep open to each LDAP server. .PP You can only set \fBldapConnectionPoolMaximumConnectionsPerHost\f1\f1 during start\-up, and cannot change this setting during runtime with the \fBsetParameter\f1\f1 database command. .RE .PP \fBldapConnectionPoolMaximumConnectionsInProgressPerHost\f1 .RS .PP \fINew in version 4.2.1 and 4.0.13\f1 .PP \fIChanged starting in MongoDB versions 4.4.15, 5.0.9, and 6.0.0\f1 Changed default value to \fB2\f1\&. In previous versions, the default is unset. .PP \fIDefault\f1: 2 .PP The maximum number of in\-progress connect operations to each LDAP server. .PP You can only set \fBldapConnectionPoolMaximumConnectionsInProgressPerHost\f1\f1 during start\-up, and cannot change this setting with the \fBsetParameter\f1\f1 database command. .RE .PP \fBldapConnectionPoolHostRefreshIntervalMillis\f1 .RS .PP \fINew in version 4.2.1 and 4.0.13\f1 .PP \fIDefault\f1: 60000 .PP The number of milliseconds in\-between health checks of the pooled LDAP connections. .PP You can only set \fBldapConnectionPoolHostRefreshIntervalMillis\f1\f1 during start\-up, and cannot change this setting with the \fBsetParameter\f1\f1 database command. .RE .PP \fBldapConnectionPoolIdleHostTimeoutSecs\f1 .RS .PP \fINew in version 4.2.1 and 4.0.13\f1 .PP \fIDefault\f1: 300 .PP The maximum number of seconds that the pooled connections to an LDAP server can remain idle before being closed. .PP You can only set \fBldapConnectionPoolIdleHostTimeoutSecs\f1\f1 during start\-up, and cannot change this setting with the \fBsetParameter\f1\f1 database command. .RE .PP \fBldapShouldRefreshUserCacheEntries\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP \fIType\f1: boolean .PP \fIDefault\f1: true .PP For MongoDB deployments using \fBLDAP Authorization\f1\&. .PP Starting in MongoDB 5.2, the update interval for cached user information retrieved from an LDAP server depends on \fBldapShouldRefreshUserCacheEntries\f1\f1: .RS .IP \(bu 2 If true, use \fBldapUserCacheRefreshInterval\f1\f1\&. .IP \(bu 2 If false, use \fBldapUserCacheInvalidationInterval\f1\f1\&. .RE .PP You can only set \fBldapShouldRefreshUserCacheEntries\f1\f1 during startup in the \fBconfiguration file\f1\f1 or with the \fB\-\-setParameter\f1 option on the command line. For example, the following disables \fBldapShouldRefreshUserCacheEntries\f1\f1: .PP .EX mongod \-\-setParameter ldapShouldRefreshUserCacheEntries=false .EE .RE .PP \fBmaxValidateMemoryUsageMB\f1 .RS .PP \fIDefault\f1: 200 .PP The maximum memory usage limit in megabytes for the \fBvalidate\f1\f1 command. If the limit is exceeded, \fBvalidate\f1\f1 returns as many results as possible and warns that not all corruption might be reported because of the limit. .PP You can set \fBmaxValidateMemoryUsageMB\f1\f1 during startup, and can change this setting using the \fBsetParameter\f1\f1 database command. .RE .PP \fBocspEnabled\f1 .RS .PP Available on Linux and macOS. .PP \fIDefault\f1: true .PP The flag that enables or disables OCSP. .PP You can only set \fBocspEnabled\f1\f1 during startup in the \fBconfiguration file\f1\f1 or with the \fB\-\-setParameter\f1 option on the command line. For example, the following disables OCSP: .PP .EX mongod \-\-setParameter ocspEnabled=false ... .EE .RS .IP \(bu 2 \fBocspValidationRefreshPeriodSecs\f1\f1 .IP \(bu 2 \fBtlsOCSPStaplingTimeoutSecs\f1\f1 .IP \(bu 2 \fBtlsOCSPVerifyTimeoutSecs\f1\f1 .RE .RE .PP \fBocspValidationRefreshPeriodSecs\f1 .RS .PP Available on Linux. .PP The number of seconds to wait before refreshing the stapled OCSP status response. Specify a number greater than or equal to 1. .PP You can only set \fBocspValidationRefreshPeriodSecs\f1\f1 during startup in the \fBconfiguration file\f1\f1 or with the \fB\-\-setParameter\f1 option on the command line. For example, the following sets the parameter to 3600 seconds: .PP .EX mongod \-\-setParameter ocspValidationRefreshPeriodSecs=3600 ... .EE .PP Starting in MongoDB 5.0, the \fBrotateCertificates\f1\f1 command and \fBdb.rotateCertificates()\f1\f1 method will also refresh any stapled OCSP responses. .RS .IP \(bu 2 \fBocspEnabled\f1\f1 .IP \(bu 2 \fBtlsOCSPStaplingTimeoutSecs\f1\f1 .IP \(bu 2 \fBtlsOCSPVerifyTimeoutSecs\f1\f1 .RE .RE .PP \fBopensslCipherConfig\f1 .RS .PP \fIAvailable on Linux only\f1 .PP With the use of native TLS/SSL libraries, the parameter \fBopensslCipherConfig\f1\f1 is supported for Linux/BSD and no longer supported in Windows and macOS. See \fBMongoDB 4.0 TLS/SSL\f1\&. .PP Specify the cipher string for OpenSSL when using TLS/SSL encryption. For a list of cipher strings, see https://www.openssl.org/docs/man1.1.1/man1/ciphers.html (https://www.openssl.org/docs/man1.1.1/man1/ciphers.html)\&. Multiple cipher strings can be provided as a colon\-separated list. .PP This parameter is only for use with TLS 1.2 or earlier. To specify cipher suites for use with TLS 1.3, use the \fBopensslCipherSuiteConfig\f1\f1 parameter. .PP You can only set \fBopensslCipherConfig\f1\f1 during start\-up, and cannot change this setting using the \fBsetParameter\f1\f1 database command. .PP For version 4.2 and greater, the use of \fBTLS\f1 options is preferred over \fBSSL\f1 options. The TLS options have the same functionality as the \fBSSL\f1 options. For example, the following configures a \fBmongod\f1\f1 with a \fBopensslCipherConfig\f1\f1 cipher string of \fB\(aqHIGH:!EXPORT:!aNULL@STRENGTH\(aq\f1 in MongoDB 4.2: .PP .EX mongod \-\-setParameter opensslCipherConfig=\(aqHIGH:!EXPORT:!aNULL@STRENGTH\(aq \-\-tlsMode requireTLS \-\-tlsCertificateKeyFile Certs/server.pem .EE .PP For versions 4.0 and earlier: .PP .EX mongod \-\-setParameter opensslCipherConfig=\(aqHIGH:!EXPORT:!aNULL@STRENGTH\(aq \-\-sslMode requireSSL \-\-sslPEMKeyFile Certs/server.pem .EE .RE .PP \fBopensslCipherSuiteConfig\f1 .RS .PP \fIAvailable on Linux only\f1 .PP Specify the list of supported cipher suites OpenSSL should permit when using TLS 1.3 encryption. .PP For a list of cipher suites for use with TLS 1.3, see https://www.openssl.org/docs/man1.1.1/man3/SSL_CTX_set_cipher_list.html (https://www.openssl.org/docs/man1.1.1/man3/SSL_CTX_set_cipher_list.html)\&. Multiple cipher suites can be provided as a colon\-separated list. .PP This parameter is only for use with TLS 1.3. To specify cipher strings for use with TLS 1.2 or earlier, use the \fBopensslCipherConfig\f1\f1 parameter. .PP You can only set \fBopensslCipherSuiteConfig\f1\f1 during start\-up, and cannot change this setting using the \fBsetParameter\f1\f1 database command. For example, the following configures a \fBmongod\f1\f1 with a \fBopensslCipherSuiteConfig\f1\f1 cipher suite of \fB\(aqTLS_AES_256_GCM_SHA384\(aq\f1 for use with TLS 1.3: .PP .EX mongod \-\-setParameter opensslCipherSuiteConfig=\(aqTLS_AES_256_GCM_SHA384\(aq \-\-tlsMode requireTLS \-\-tlsCertificateKeyFile Certs/server.pem .EE .RE .PP \fBopensslDiffieHellmanParameters\f1 .RS .PP \fIAvailable on Linux only\f1 .PP Specify the path to the PEM file that contains the OpenSSL Diffie\-Hellman parameters when using TLS 1.2 or previous. Specifying the OpenSSL Diffie\-Hellman parameters enables support for \fBEphemeral Diffie\-Hellman (DHE)\f1 cipher suites during TLS/SSL encryption. .PP This parameter is not supported for use with TLS 1.3. .PP Ephemeral Diffie\-Hellman (DHE) cipher suites (and Ephemeral Elliptic Curve Diffie\-Hellman (ECDHE) cipher suites) provide \fBForward Secrecy\f1\&. \fBForward Secrecy\f1 cipher suites create an ephemeral session key that is protected by the server\(aqs private key but never transmitted. This ensures that even if a server\(aqs private key is compromised, you cannot decrypt past sessions with the compromised key. .PP Starting in MongoDB 4.2, if \fBopensslDiffieHellmanParameters\f1\f1 is unset but \fBECDHE\f1 is enabled, MongoDB enables DHE using the \fBffdhe3072\f1 Diffie\-Hellman parameter, as defined in RFC\-7919#appendix\-A.2 (https://tools.ietf.org/html/7919#appendix\-A.2)\&. The \fBffdhe3072\f1 is a strong parameter (specifically, size is greater than 1024). Strong parameters are not supported with Java 6 and 7 unless extended support has been purchased from Oracle. .PP You can only set \fBopensslDiffieHellmanParameters\f1\f1 during startup, and cannot change this setting using the \fBsetParameter\f1\f1 database command. .PP If for performance reasons, you need to disable support for DHE cipher suites, use the \fBopensslCipherConfig\f1\f1 parameter: .PP .EX mongod \-\-setParameter opensslCipherConfig=\(aqHIGH:!EXPORT:!aNULL:!DHE:!kDHE@STRENGTH\(aq ... .EE .RE .PP \fBsaslauthdPath\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Specify the path to the Unix Domain Socket of the \fBsaslauthd\f1 instance to use for proxy authentication. .RE .PP \fBsaslHostName\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP \fBsaslHostName\f1\f1 overrides MongoDB\(aqs default hostname detection for the purpose of configuring SASL and Kerberos authentication. .PP \fBsaslHostName\f1\f1 does not affect the hostname of the \fBmongod\f1\f1 or \fBmongos\f1\f1 instance for any purpose beyond the configuration of SASL and Kerberos. .PP You can only set \fBsaslHostName\f1\f1 during start\-up, and cannot change this setting using the \fBsetParameter\f1\f1 database command. .PP \fBsaslHostName\f1\f1 supports Kerberos authentication and is only included in MongoDB Enterprise. For more information, see the following: .RS .IP \(bu 2 Linux: \fBConfigure MongoDB with Kerberos Authentication on Linux\f1 .IP \(bu 2 Windows: \fBConfigure MongoDB with Kerberos Authentication on Windows\f1 .RE .RE .PP \fBsaslServiceName\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Allows users to override the default \fBKerberos\f1 service name component of the \fBKerberos\f1 principal name, on a per\-instance basis. If unspecified, the default value is \fBmongodb\f1\&. .PP MongoDB only permits setting \fBsaslServiceName\f1\f1 at startup. The \fBsetParameter\f1\f1 command can not change this setting. .PP \fBsaslServiceName\f1\f1 is only available in MongoDB Enterprise. .PP Ensure that your driver supports alternate service names. .RE .PP \fBscramIterationCount\f1 .RS .PP \fIDefault\f1: \fB10000\f1 .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Changes the number of hashing iterations used for all new \fBSCRAM\-SHA\-1\f1 passwords. More iterations increase the amount of time required for clients to authenticate to MongoDB, but makes passwords less susceptible to brute\-force attempts. The default value is ideal for most common use cases and requirements. .PP If you modify this value, it does not change the iteration count for existing passwords. The \fBscramIterationCount\f1\f1 value must be \fB5000\f1 or greater. .PP For example, the following sets the \fBscramIterationCount\f1\f1 to \fB12000\f1\&. .PP .EX mongod \-\-setParameter scramIterationCount=12000 .EE .PP Or, if using the \fBsetParameter\f1\f1 command within \fBmongosh\f1\f1: .PP .EX db.adminCommand( { setParameter: 1, scramIterationCount: 12000 } ) .EE .RS .IP \(bu 2 \fBdb.changeUserPassword()\f1\f1 .IP \(bu 2 \fBdb.createUser()\f1\f1 .IP \(bu 2 \fBdb.updateUser()\f1\f1 .RE .RE .PP \fBscramSHA256IterationCount\f1 .RS .PP \fIDefault\f1: \fB15000\f1 .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Changes the number of hashing iterations used for all new \fBSCRAM\-SHA\-256\f1 passwords. More iterations increase the amount of time required for clients to authenticate to MongoDB, but makes passwords less susceptible to brute\-force attempts. The default value is ideal for most common use cases and requirements. .PP If you modify this value, it does not change iteration count for existing passwords. The \fBscramSHA256IterationCount\f1\f1 value must be \fB5000\f1 or greater. .PP For example, the following sets the \fBscramSHA256IterationCount\f1\f1 to \fB20000\f1\&. .PP .EX mongod \-\-setParameter scramSHA256IterationCount=20000 .EE .PP Or, if using the \fBsetParameter\f1\f1 command within \fBmongosh\f1\f1: .PP .EX db.adminCommand( { setParameter: 1, scramSHA256IterationCount: 20000 } ) .EE .RS .IP \(bu 2 \fBdb.changeUserPassword()\f1\f1 .IP \(bu 2 \fBdb.createUser()\f1\f1 .IP \(bu 2 \fBdb.updateUser()\f1\f1 .RE .RE .PP \fBsslMode\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Set the \fBnet.ssl.mode\f1\f1 to either \fBpreferSSL\f1 or \fBrequireSSL\f1\&. Useful during \fBrolling upgrade to TLS/SSL\f1 to minimize downtime. .PP For more information about TLS/SSL and MongoDB, see \fBConfigure mongod\f1 and mongos\f1 for TLS/SSL\f1 and \fBTLS/SSL Configuration for Clients\f1 . .PP .EX db.adminCommand( { setParameter: 1, sslMode: "preferSSL" } ) .EE .PP \fBtlsMode\f1\f1 .RE .PP \fBtlsMode\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Set to either: .RS .IP \(bu 2 \fBpreferTLS\f1 .IP \(bu 2 \fBrequireTLS\f1 .RE .PP The \fBtlsMode\f1\f1 parameter is useful during \fBrolling upgrade to TLS/SSL\f1 to minimize downtime. .PP .EX db.adminCommand( { setParameter: 1, tlsMode: "preferTLS" } ) .EE .PP For more information about TLS/SSL and MongoDB, see \fBConfigure mongod\f1 and mongos\f1 for TLS/SSL\f1 and \fBTLS/SSL Configuration for Clients\f1 . .PP \fBsslMode\f1\f1 .RE .PP \fBtlsOCSPStaplingTimeoutSecs\f1 .RS .PP Available for Linux. .PP The maximum number of seconds the \fBmongod\f1\f1/\fBmongos\f1\f1 instance should wait to receive the OCSP status response for its certificates. .PP Specify an integer greater than or equal to (\fB>=\f1) 1. If unset, \fBtlsOCSPStaplingTimeoutSecs\f1\f1 uses the \fBtlsOCSPVerifyTimeoutSecs\f1\f1 value. .PP You can only set \fBtlsOCSPStaplingTimeoutSecs\f1\f1 during startup in the \fBconfiguration file\f1\f1 or with the \fB\-\-setParameter\f1 option on the command line. For example, the following sets the \fBtlsOCSPStaplingTimeoutSecs\f1\f1 to 20 seconds: .PP .EX mongod \-\-setParameter tlsOCSPStaplingTimeoutSecs=20 ... .EE .RS .IP \(bu 2 \fBocspEnabled\f1\f1 .IP \(bu 2 \fBocspValidationRefreshPeriodSecs\f1\f1 .IP \(bu 2 \fBtlsOCSPVerifyTimeoutSecs\f1\f1 .RE .RE .PP \fBtlsOCSPVerifyTimeoutSecs\f1 .RS .PP Available for Linux and Windows. .PP \fIDefault\f1: 5 .PP The maximum number of seconds that the \fBmongod\f1\f1/\fBmongos\f1\f1 should wait for the OCSP response when verifying client certificates. .PP Specify an integer greater than or equal to (\fB>=\f1) 1. .PP You can only set \fBtlsOCSPVerifyTimeoutSecs\f1\f1 during startup in the \fBconfiguration file\f1\f1 or with the \fB\-\-setParameter\f1 option on the command line. For example, the following sets the \fBtlsOCSPVerifyTimeoutSecs\f1\f1 to 20 seconds: .PP .EX mongod \-\-setParameter tlsOCSPVerifyTimeoutSecs=20 ... .EE .RS .IP \(bu 2 \fBocspEnabled\f1\f1 .IP \(bu 2 \fBocspValidationRefreshPeriodSecs\f1\f1 .IP \(bu 2 \fBtlsOCSPStaplingTimeoutSecs\f1\f1 .RE .RE .PP \fBtlsWithholdClientCertificate\f1 .RS .PP \fIDefault\f1: false .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP A TLS certificate is set for a \fBmongod\f1\f1 or \fBmongos\f1\f1 either by the \fB\-\-tlsClusterFile\f1\f1 option or by the \fB\-\-tlsCertificateKeyFile\f1\f1 option when \fB\-\-tlsClusterFile\f1\f1 is not set. If the TLS certificate is set, by default, the instance sends the certificate when initiating intra\-cluster communications with other \fBmongod\f1\f1 or \fBmongos\f1\f1 instances in the deployment. Set \fBtlsWithholdClientCertificate\f1 to \fB1\f1 or \fBtrue\f1 to direct the instance to withhold sending its TLS certificate during these communications. Use this option with \fB\-\-tlsAllowConnectionsWithoutCertificates\f1\f1 (to allow inbound connections without certificates) on all members of the deployment. \fBtlsWithholdClientCertificate\f1 is mutually exclusive with \fB\-\-clusterAuthMode x509\f1\f1\&. .RE .PP \fBtlsX509ClusterAuthDNOverride\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP An alternative Distinguished Name (DN) that the instance can also use to identify members of the deployment. .PP For a MongoDB deployment that uses x.509 certificates for \fBclusterAuthMode\f1\f1, deployment members identify each other using x.509 certificates ( \fBnet.tls.clusterFile\f1\f1, if specified, and \fBnet.tls.certificateKeyFile\f1\f1) during intra\-cluster communications. For members of the same deployment, the \fBDN\f1 from their certificates must have the same Organization attributes (\fBO\f1\(aqs), the Organizational Unit attributes (\fBOU\f1\(aqs), and the Domain Components (\fBDC\f1\(aqs). .PP If \fBtlsX509ClusterAuthDNOverride\f1\f1 is set for a member, the member can also use the override value when comparing the \fBDN\f1 components (\fBO\f1\(aqs, \fBOU\f1\(aqs, and \fBDC\f1\(aqs) of the presented certificates. That is the member checks the presented certificates against its \fBnet.tls.clusterFile\f1\f1/\fBnet.tls.certificateKeyFile\f1\f1\&. If the DN does not match, the member checks the presented certificate against the \fBtlsX509ClusterAuthDNOverride\f1\f1 value. .PP If set, you must set this parameter on all members of the deployment. .PP You can use this parameter for a rolling update of certificates to new certificates that contain a new \fBDN\f1 value. See \fBRolling Update of x.509 Cluster Certificates that Contain New DN\f1\&. .PP For more information about membership certificate requirements, see \fBMember Certificate Requirements\f1 for details. .RE .PP \fBtlsX509ExpirationWarningThresholdDays\f1 .RS .PP \fIDefault\f1 : 30 .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Starting in MongoDB 4.4, \fBmongod\f1\f1/\fBmongos\f1\f1 logs a warning on connection if the presented x.509 certificate expires within \fB30\f1 days of the \fBmongod/mongos\f1 system clock. Use the \fBtlsX509ExpirationWarningThresholdDays\f1\f1 parameter to control the certificate expiration warning threshold: .RS .IP \(bu 2 Increase the parameter value to trigger warnings farther ahead of the certificate expiration date. .IP \(bu 2 Decrease the parameter value to trigger warnings closer to the certificate expiration date. .IP \(bu 2 Set the parameter to \fB0\f1 to disable the warning. .RE .PP This parameter has a minimum value of \fB0\f1\&. .PP You can only set \fBtlsX509ExpirationWarningThresholdDays\f1\f1 during \fBmongod/mongos\f1 startup using either: .RS .IP \(bu 2 The \fBsetParameter\f1\f1 configuration setting, \fIor\f1 .IP \(bu 2 The \fBmongod \-\-setParameter\f1\f1 / \fBmongos \-\-setParameter\f1\f1 command line option. .RE .PP See \fBx.509 Certificates Nearing Expiry Trigger Warnings\f1 for more information on x.509 expiration warnings in MongoDB 4.4. .PP For more information on x.509 certificate validity, see RFC 5280 4.1.2.5 (https://tools.ietf.org/html/rfc5280#section\-4.1.2.5)\&. .RE .PP \fBsslWithholdClientCertificate\f1 .RS .PP \fIDefault\f1: false .PP Use \fBtlsWithholdClientCertificate\f1\f1 instead. .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP A TLS certificate is set for a \fBmongod\f1\f1 or \fBmongos\f1\f1 either by the \fB\-\-tlsClusterFile\f1\f1 option or by the \fB\-\-tlsCertificateKeyFile\f1\f1 option when \fB\-\-tlsClusterFile\f1\f1 is not set. If the TLS certificate is set, by default, the instance sends the certificate when initiating intra\-cluster communications with other \fBmongod\f1\f1 or \fBmongos\f1\f1 instances in the deployment. Set \fBsslWithholdClientCertificate\f1 to \fB1\f1 or \fBtrue\f1 to direct the instance to withhold sending its TLS certificate during these communications. Use this option with \fB\-\-tlsAllowConnectionsWithoutCertificates\f1\f1 (to allow inbound connections without certificates) on all members of the deployment. \fBsslWithholdClientCertificate\f1 is mutually exclusive with \fB\-\-clusterAuthMode x509\f1\f1\&. .RE .PP \fBuserCacheInvalidationIntervalSecs\f1 .RS .PP \fIDefault\f1: 30 .PP Available for \fBmongos\f1\f1 only. .PP On a \fBmongos\f1\f1 instance, specifies the interval (in seconds) at which the \fBmongos\f1\f1 instance checks to determine whether the in\-memory cache of \fBuser objects\f1 has stale data, and if so, clears the cache. If there are no changes to user objects, \fBmongos\f1\f1 will not clear the cache. .PP This parameter has a minimum value of \fB1\f1 second and a maximum value of \fB86400\f1 seconds (24 hours). .RE .PP \fBauthFailedDelayMs\f1 .RS .PP \fIDefault\f1: 0 .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Available in MongoDB Enterprise only. .PP The number of milliseconds to wait before informing clients that their authentication attempt has failed. This parameter may be in the range \fB0\f1 to \fB5000\f1, inclusive. .PP Setting this parameter makes brute\-force login attacks on a database more time\-consuming. However, clients waiting for a response from the MongoDB server still consume server resources, and this may adversely impact benign login attempts if the server is denying access to many other clients simultaneously. .RE .PP \fBallowRolesFromX509Certificates\f1 .RS .PP \fIDefault\f1: true .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP \fIAvailable starting in MongoDB 4.0.11 (and 3.6.14 and 3.4.22)\f1 .PP A boolean flag that allows or disallows the retrieval of authorization roles from client x.509 certificates. .PP You can only set \fBallowRolesFromX509Certificates\f1\f1 during startup in the config file or on the command line. .RE .SS GENERAL PARAMETERS .PP \fBallowDiskUseByDefault\f1 .RS .PP \fIDefault\f1: True .PP Available for \fBmongod\f1\f1 only. .PP Starting in MongoDB 6.0, pipeline stages that require more than 100 megabytes of memory to execute write temporary files to disk by default. In earlier verisons of MongoDB, you must pass \fB{ allowDiskUse: true }\f1 to individual \fBfind\f1 and \fBaggregate\f1 commands to enable this behavior. .PP Individual \fBfind\f1 and \fBaggregate\f1 commands may override the \fBallowDiskUseByDefault\f1\f1 parameter by either: .RS .IP \(bu 2 Using \fB{ allowDiskUse: true }\f1 to allow writing temporary files out to disk when \fBallowDiskUseByDefault\f1 is set to \fBfalse\f1 .IP \(bu 2 Using \fB{ allowDiskUse: false }\f1 to prohibit writing temporary files out to disk when \fBallowDiskUseByDefault\f1 is set to \fBtrue\f1 .RE .PP .EX mongod \-\-setParameter allowDiskUseByDefault=false .EE .PP \fBallowDiskUseByDefault\f1\f1 only works on \fBmongod\f1\f1 not \fBmongos\f1\f1\&. \fBmongos\f1\f1 never writes temporary files to disk. Use the \fBsetParameter\f1\f1 command in a \fBmongosh\f1\f1 session that is connected to a running \fBmongod\f1\f1 to change the value of the parameter while the server is running: .PP .EX db.adminCommand( { setParameter: 1, allowDiskUseByDefault: false } ) .EE .RE .PP \fBconnPoolMaxShardedConnsPerHost\f1 .RS .PP \fIDefault\f1: 200 .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Sets the maximum size of the legacy connection pools for communication to the shards. The size of a pool does not prevent the creation of additional connections, but \fIdoes\f1 prevent the connection pools from retaining connections above this limit. .PP The parameter is separate from the connections in TaskExecutor pools. See \fBShardingTaskExecutorPoolMaxSize\f1\f1\&. .PP Increase the \fBconnPoolMaxShardedConnsPerHost\f1\f1 value \fBonly\f1 if the number of connections in a connection pool has a high level of churn or if the total number of created connections increase. .PP You can only set \fBconnPoolMaxShardedConnsPerHost\f1\f1 during startup in the config file or on the command line. For example: .PP .EX mongos \-\-setParameter connPoolMaxShardedConnsPerHost=250 .EE .RE .PP \fBconnPoolMaxShardedInUseConnsPerHost\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Sets the maximum number of in\-use connections at any given time for the legacy sharded cluster connection pools. .PP By default, the parameter is unset. .PP You can only set \fBconnPoolMaxShardedConnsPerHost\f1\f1 during startup in the config file or on the command line. For example: .PP .EX mongos \-\-setParameter connPoolMaxShardedInUseConnsPerHost=100 .EE .PP \fBconnPoolMaxShardedConnsPerHost\f1\f1 .RE .PP \fBshardedConnPoolIdleTimeoutMinutes\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Sets the time limit that a connection in the legacy sharded cluster connection pool can remain idle before being closed. .PP By default, the parameter is unset. .PP You can only set \fBshardedConnPoolIdleTimeoutMinutes\f1\f1 during startup in the config file or on the command line. For example: .PP .EX mongos \-\-setParameter shardedConnPoolIdleTimeoutMinutes=10 .EE .PP \fBconnPoolMaxShardedConnsPerHost\f1\f1 .RE .PP \fBconnPoolMaxConnsPerHost\f1 .RS .PP \fIDefault\f1: 200 .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Sets the maximum size of the legacy connection pools for outgoing connections to other \fBmongod\f1\f1 instances in the global connection pool. The size of a pool does not prevent the creation of additional connections, but \fIdoes\f1 prevent a connection pool from retaining connections in excess of the value of \fBconnPoolMaxConnsPerHost\f1\f1\&. .PP The parameter is separate from the connections in TaskExecutor pools. See \fBShardingTaskExecutorPoolMaxSize\f1\f1\&. .PP \fBOnly\f1 adjust this setting if your driver does \fInot\f1 pool connections and you\(aqre using authentication in the context of a sharded cluster. .PP You can only set \fBconnPoolMaxConnsPerHost\f1\f1 during startup in the config file or on the command line. For example: .PP .EX mongod \-\-setParameter connPoolMaxConnsPerHost=250 .EE .RE .PP \fBconnPoolMaxInUseConnsPerHost\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Sets the maximum number of in\-use connections at any given time for for outgoing connections to other \fBmongod\f1\f1 instances in the legacy global connection pool. .PP By default, the parameter is unset. .PP You can only set \fBconnPoolMaxInUseConnsPerHost\f1\f1 during startup in the config file or on the command line. For example: .PP .EX mongod \-\-setParameter connPoolMaxInUseConnsPerHost=100 .EE .PP \fBconnPoolMaxConnsPerHost\f1\f1 .RE .PP \fBglobalConnPoolIdleTimeoutMinutes\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Sets the time limit that connection in the legacy global connection pool can remain idle before being closed. .PP By default, the parameter is unset. .PP You can only set \fBglobalConnPoolIdleTimeoutMinutes\f1\f1 during startup in the config file or on the command line. For example: .PP .EX mongos \-\-setParameter globalConnPoolIdleTimeoutMinutes=10 .EE .PP \fBconnPoolMaxShardedConnsPerHost\f1\f1 .RE .PP \fBcursorTimeoutMillis\f1 .RS .PP \fIDefault\f1: 600000 (10 minutes) .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Sets the expiration threshold in milliseconds for idle cursors before MongoDB removes them; specifically, MongoDB removes cursors that have been idle for the specified \fBcursorTimeoutMillis\f1\f1\&. .PP For example, the following sets the \fBcursorTimeoutMillis\f1\f1 to \fB300000\f1 milliseconds (5 minutes). .PP .EX mongod \-\-setParameter cursorTimeoutMillis=300000 .EE .PP Or, if using the \fBsetParameter\f1\f1 command within \fBmongosh\f1\f1: .PP .EX db.adminCommand( { setParameter: 1, cursorTimeoutMillis: 300000 } ) .EE .PP Setting \fBcursorTimeoutMillis\f1\f1 to less than or equal to \fB0\f1 results in all cursors being immediately eligible for timeout. Generally, the timeout value should be greater than the average amount of time for a query to return results. Use tools like the \fBcursor.explain()\f1\f1 cursor modifier to analyze the average query time and select an appropriate timeout period. .RE .PP \fBfailIndexKeyTooLong\f1 .RS .PP \fIRemoved in 4.4\f1 .RS .IP \(bu 2 \fBMongoDB 4.4\f1 \fIremoves\f1 the deprecated \fBfailIndexKeyTooLong\f1\f1 parameter. Attempting to use this parameter with MongoDB 4.4 will result in an error. .IP \(bu 2 \fBMongoDB 4.2\f1 \fIdeprecates\f1 the \fBfailIndexKeyTooLong\f1\f1 parameter and \fIremoves\f1 the \fBIndex Key Length Limit\f1 for \fBfeatureCompatibilityVersion\f1 (fCV) set to \fB"4.2"\f1 or greater. .RE .PP For MongoDB 2.6 through MongoDB versions with \fBfeatureCompatibilityVersion\f1 (fCV) set to \fB"4.0"\f1 or earlier, \fBIndex Key Length Limit\f1 applies. If you attempt to insert or update a document whose index field exceeds the \fBIndex Key Length Limit\f1, the operation will fail and return an error to the client. .PP To avoid this issue, consider using \fBhashed indexes\f1 or indexing a computed value. If you have an existing data set and want to disable this behavior so you can upgrade and then gradually resolve these indexing issues, you can use \fBfailIndexKeyTooLong\f1\f1 to disable this behavior. .PP Setting \fBfailIndexKeyTooLong\f1\f1 to \fBfalse\f1 is a temporary workaround, not a permanent solution to the problem of oversized index keys. With \fBfailIndexKeyTooLong\f1\f1 set to \fBfalse\f1, queries can return incomplete results if they use indexes that skip over documents whose indexed fields exceed the \fBIndex Key Length Limit\f1\&. .PP \fBfailIndexKeyTooLong\f1\f1 defaults to \fBtrue\f1\&. .PP Issue the following command to disable the index key length validation: .PP .EX db.adminCommand( { setParameter: 1, failIndexKeyTooLong: false } ) .EE .PP You can also set \fBfailIndexKeyTooLong\f1\f1 at startup with the following option: .PP .EX mongod \-\-setParameter failIndexKeyTooLong=false .EE .RE .PP \fBmaxNumActiveUserIndexBuilds\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP \fIType\f1: integer .PP \fIDefault\f1: 3 .PP MongoDB limits the number of concurrent user index builds to control WiredTiger cache pressure. The primary node can run up to \fBmaxNumActiveUserIndexBuilds\f1 concurrent user index builds. This is a global limit that applies across collections. .PP System indexes are not limited to \fBmaxNumActiveUserIndexBuilds\f1, however a running system index build will count against the limit for user index builds. .PP After reaching \fBmaxNumActiveUserIndexBuilds\f1, MongodDB blocks additional user index builds until the number of concurrent index builds drops below the \fBmaxNumActiveUserIndexBuilds\f1 limit. If an index build is blocked, the server logs a message, "Too many index builds running simultaneously, waiting until the number of active index builds is below the threshold". .PP The following command sets a limit of 4 concurrent index builds: .PP .EX db.adminCommand( { setParameter: 1, maxNumActiveUserIndexBuilds: 4 } ) .EE .PP See also: \fBmaxIndexBuildMemoryUsageMegabytes\f1\f1 .RE .PP \fBnotablescan\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP Specify whether \fBall\f1 queries must use indexes. If \fB1\f1, MongoDB will not execute queries that require a collection scan and will return an error. .PP Consider the following example which sets \fBnotablescan\f1\f1 to \fB1\f1 or true: .PP .EX db.adminCommand( { setParameter: 1, notablescan: 1 } ) .EE .PP Setting \fBnotablescan\f1\f1 to \fB1\f1 can be useful for testing application queries, for example, to identify queries that scan an entire collection and cannot use an index. .PP To detect unindexed queries without \fBnotablescan\f1, consider reading the \fBEvaluate Performance of Current Operations\f1 and \fBOptimize Query Performance\f1 sections and using the \fBlogLevel\f1\f1 parameter, \fBmongostat\f1\f1 and \fBprofiling\f1\&. .PP Don\(aqt run production \fBmongod\f1\f1 instances with \fBnotablescan\f1\f1 because preventing collection scans can potentially affect queries in all databases, including administrative queries. .RE .PP \fBttlMonitorEnabled\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP \fIDefault\f1: \fBtrue\f1 .PP To support \fBTTL Indexes\f1, \fBmongod\f1\f1 instances have a background thread that is responsible for deleting documents from collections with TTL indexes. .PP To disable this worker thread for a \fBmongod\f1\f1, set \fBttlMonitorEnabled\f1\f1 to \fBfalse\f1, as in the following operations: .PP .EX db.adminCommand( { setParameter: 1, ttlMonitorEnabled: false } ) .EE .PP Alternately, you may disable the thread at startup time by starting the \fBmongod\f1\f1 instance with the following option: .PP .EX mongod \-\-setParameter ttlMonitorEnabled=false .EE .PP Do not run production \fBmongod\f1\f1 instances with \fBttlMonitorEnabled\f1 disabled, except under guidance from MongoDB support. Preventing TTL document removal can negatively impact MongoDB internal system operations that depend on \fBTTL Indexes\f1\&. .RE .PP \fBtcpFastOpenServer\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP \fIDefault\f1: \fBtrue\f1 .PP Enables support for accepting inbound TCP Fast Open (TFO) connections to the \fBmongod/mongos\f1 from a client. TFO requires both the client and \fBmongod/mongos\f1 host machine support and enable TFO: .PP \fBWindows\f1 .RS .PP The following Windows operating systems support TFO: .RS .IP \(bu 2 Microsoft Windows Server 2016 and later. .IP \(bu 2 Microsoft Windows 10 Update 1607 and later. .RE .RE .PP \fBmacOS\f1 .RS .PP macOS 10.11 (El Capitan) and later support TFO. .RE .PP \fBLinux\f1 .RS .PP Linux operating systems running Linux Kernel 3.7 or later can support inbound TFO. .PP Set the value of \fB/proc/sys/net/ipv4/tcp_fastopen\f1 to enable inbound TFO connections: .RS .IP \(bu 2 Set to \fB2\f1 to enable only inbound TFO connections. .IP \(bu 2 Set to \fB3\f1 to enable inbound and outbound TFO connections. .RE .RE .PP This parameter has no effect if the host operating system does not support \fIor\f1 is not configured to support TFO connections. .PP You can only set this parameter on startup, using either the \fBsetParameter\f1\f1 configuration file setting or the \fB\-\-setParameter\f1\f1 command line option. .PP See \fBSupport for TCP Fast Open\f1 for more information on MongoDB TFO support. .PP RFC7413 (https://tools.ietf.org/html/rfc7413)\&. .RE .PP \fBtcpFastOpenClient\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP \fIDefault\f1: \fBtrue\f1 .PP \fILinux Operating System Only\f1 .PP Enables support for outbound TCP Fast Open (TFO) connections from the \fBmongod/mongos\f1 to a client. TFO requires both the client and the \fBmongod/mongos\f1 host machine support and enable TFO. .PP Linux operating systems running Linux Kernel 4.11 or later can support outbound TFO. .PP Set the value of \fB/proc/sys/net/ipv4/tcp_fastopen\f1 to enable outbound TFO connections: .RS .IP \(bu 2 \fB1\f1 to enable only outbound TFO connections. .IP \(bu 2 \fB3\f1 to enable inbound and outbound TFO connections. .RE .PP This parameter has no effect if the host operating system does not support \fIor\f1 is not configured to support TFO connections. .PP You can only set this parameter on startup, using either the \fBsetParameter\f1\f1 configuration file setting or the \fB\-\-setParameter\f1\f1 command line option. .PP See \fBSupport for TCP Fast Open\f1 for more information on MongoDB TFO support. .PP RFC7413 (https://tools.ietf.org/html/rfc7413)\&. .RE .PP \fBtcpFastOpenQueueSize\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP \fIDefault\f1: \fB1024\f1 .PP As part of establishing a TCP Fast Open (TFO) connection, the client submits a valid TFO cookie to the \fBmongod/mongos\f1 \fIbefore\f1 completion of the standard TCP 3\-way handshake. The \fBmongod/mongos\f1 keeps a queue of all such pending TFO connections. .PP The \fBtcpFastOpenQueueSize\f1 parameter sets the size of the queue of pending TFO connections. While the queue is full, the \fBmongod/mongos\f1 falls back to the normal three\-way handshake for incoming client requests and ignores the presence of TFO cookies. Once the queue size falls back below the limit, the \fBmongod/mongos\f1 begins accepting new TFO cookies. .RS .IP \(bu 2 Increasing the default queue size may improve the effect of TFO on network performance. However, large queue sizes also increase the risk of server resource exhaustion due to excessive incoming TFO requests. .IP \(bu 2 Decreasing the default queue size may reduce the risk of resource server resource exhaustion due to excessive incoming TFO requests. However, small queue sizes may also reduce the effect of TFO on network performance. .IP The minimum queue size is \fB0\f1\&. A queue of \fB0\f1 effectively disables TFO. .RE .PP This parameter has no effect on host operating systems that do not support or are not configured for TFO connections. See \fBSupport for TCP Fast Open\f1 for more information on MongoDB TFO support. .RS .IP \(bu 2 RFC7413 TCP Fast Open Section 5: Security Considerations (https://tools.ietf.org/html/rfc7413#section\-5) .IP \(bu 2 RFC7413 TCP Fast Open Section 6: TFO Applicability (https://tools.ietf.org/html/rfc7413#section\-6) .RE .RE .PP \fBdisableJavaScriptJIT\f1 .RS .PP The JavaScript engine\(aqs JIT compiler is now disabled by default. .PP Available for \fBmongod\f1\f1 only. .PP The MongoDB JavaScript engine uses SpiderMonkey, which implements Just\-in\-Time (JIT) compilation for improved performance when running scripts. .PP To enable the JIT, set \fBdisableJavaScriptJIT\f1\f1 to \fBfalse\f1, as in the following example: .PP .EX db.adminCommand( { setParameter: 1, disableJavaScriptJIT: false } ) .EE .PP \fB$where\f1\f1 will reuse existing JavaScript interpreter contexts, so changes to \fBdisableJavaScriptJIT\f1\f1 may not take effect immediately for these operations. .PP Alternately, you may enable the JIT at startup time by starting the \fBmongod\f1\f1 instance with the following option: .PP .EX mongod \-\-setParameter disableJavaScriptJIT=false .EE .RE .PP \fBindexMaxNumGeneratedKeysPerDocument\f1 .RS .PP \fIDefault\f1: 100000 .PP Limits the maximum number of keys generated for a document to prevent out of memory errors. It is possible to raise the limit, but if an operation requires more keys than the \fBindexMaxNumGeneratedKeysPerDocument\f1\f1 parameter specifies, the operation will fail. .RE .PP \fBmaxIndexBuildMemoryUsageMegabytes\f1 .RS .PP \fIDefault\f1: .RS .IP \(bu 2 200 (For versions 4.2.3 and later) .IP \(bu 2 500 (For versions 4.2.2 and earlier) .RE .PP Limits the amount of memory that simultaneous index builds on one collection may consume for the duration of the builds. The specified amount of memory is shared between all indexes built using a single \fBcreateIndexes\f1\f1 command or its shell helper \fBdb.collection.createIndexes()\f1\f1\&. .PP The memory consumed by an index build is separate from the WiredTiger cache memory (see \fBcacheSizeGB\f1\f1). .PP Index builds may be initiated either by a user command such as \fBCreate Index\f1 or by an administrative process such as an \fBinitial sync\f1\&. Both are subject to the limit set by \fBmaxIndexBuildMemoryUsageMegabytes\f1\f1\&. .PP An \fBinitial sync operation\f1 populates only one collection at a time and has no risk of exceeding the memory limit. However, it is possible for a user to start index builds on multiple collections in multiple databases simultaneously and potentially consume an amount of memory greater than the limit set in \fBmaxIndexBuildMemoryUsageMegabytes\f1\f1\&. .PP To minimize the impact of building an index on replica sets and sharded clusters with replica set shards, use a rolling index build procedure as described on \fBRolling Index Builds on Replica Sets\f1\&. .RS .IP \(bu 2 For \fBfeature compatibility version (fcv)\f1 \fB"4.2"\f1, the index build memory limit applies to all index builds. .IP \(bu 2 For \fBfeature compatibility version (fcv)\f1 \fB"4.0"\f1, the index build memory limit only applies to foreground index builds. .RE .RE .PP \fBreportOpWriteConcernCountersInServerStatus\f1 .RS .PP \fIDefault\f1: false .PP A boolean flag that determines whether the \fBdb.serverStatus()\f1\f1 method and \fBserverStatus\f1\f1 command return \fBopWriteConcernCounters\f1\f1 information. .PP You can only set \fBreportOpWriteConcernCountersInServerStatus\f1\f1 during startup in the config file or on the command line. For example: .PP .EX mongod \-\-setParameter reportOpWriteConcernCountersInServerStatus=true .EE .RE .PP \fBwatchdogPeriodSeconds\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP \fIType\f1: integer .PP \fIDefault\f1: \-1 (disabled) .RS .IP \(bu 2 Starting in MongoDB 4.2, the \fBStorage Node Watchdog\f1 is available in both the Community and MongoDB Enterprise editions. .IP \(bu 2 In earlier versions (3.2.16+, 3.4.7+, 3.6.0+, 4.0.0+), the \fBStorage Node Watchdog\f1 is only available in MongoDB Enterprise edition. .RE .PP Determines how frequent the \fBStorage Node Watchdog\f1 checks the status of the monitored filesystems: .RS .IP \(bu 2 The \fB\-\-dbpath\f1\f1 directory .IP \(bu 2 The \fBjournal\f1 directory inside the \fB\-\-dbpath\f1\f1 directory if \fBjournaling\f1\f1 is enabled .IP \(bu 2 The directory of \fB\-\-logpath\f1\f1 file .IP \(bu 2 The directory of \fB\-\-auditPath\f1\f1 file .RE .PP Valid values for \fBwatchdogPeriodSeconds\f1\f1 are: .RS .IP \(bu 2 \fB\-1\f1 (the default), to disable/pause \fBStorage Node Watchdog\f1, or .IP \(bu 2 An integer greater than or equal to 60. .RE .RS .IP \(bu 2 If a filesystem on a monitored directory becomes unresponsive, it can take a maximum of nearly \fItwice\f1 the value of \fBwatchdogPeriodSeconds\f1\f1 to terminate the \fBmongod\f1\f1\&. .IP \(bu 2 If any of its monitored directory is a symlink to other volumes, the Storage Node Watchdog does not monitor the symlink target. For example, if the \fBmongod\f1\f1 uses \fBstorage.directoryPerDB: true\f1\f1 (or \fB\-\-directoryperdb\f1\f1) and symlinks a database directory to another volume, the Storage Node Watchdog does not follow the symlink to monitor the target. .RE .PP To enable \fBStorage Node Watchdog\f1, \fBwatchdogPeriodSeconds\f1\f1 must be set during startup. .PP .EX mongod \-\-setParameter watchdogPeriodSeconds=60 .EE .PP You can only enable the \fBStorage Node Watchdog\f1 at startup. However, once enabled, you can pause the \fBStorage Node Watchdog\f1 or change the \fBwatchdogPeriodSeconds\f1\f1 during runtime. .PP Once enabled, .RS .IP \(bu 2 To pause the \fBStorage Node Watchdog\f1 during runtime, set \fBwatchdogPeriodSeconds\f1\f1 to \-1. .IP .EX db.adminCommand( { setParameter: 1, watchdogPeriodSeconds: \-1 } ) .EE .IP \(bu 2 To resume or change the period during runtime, set \fBwatchdogPeriodSeconds\f1\f1 to a number greater than or equal to 60. .IP .EX db.adminCommand( { setParameter: 1, watchdogPeriodSeconds: 120 } ) .EE .RE .PP It is an error to set \fBwatchdogPeriodSeconds\f1\f1 at runtime if the \fBStorage Node Watchdog\f1 was not enabled at startup time. .RE .PP \fBtcmallocReleaseRate\f1 .RS .PP Default: 1.0 .PP Specifies the tcmalloc release rate (TCMALLOC_RELEASE_RATE (https://gperftools.github.io/gperftools/tcmalloc.html#runtime)). Per https://gperftools.github.io/gperftools/tcmalloc.html#runtime (https://gperftools.github.io/gperftools/tcmalloc.html#runtime) TCMALLOC_RELEASE_RATE is described as the "Rate at which we release unused memory to the system, via madvise(MADV_DONTNEED), on systems that support it. Zero means we never release memory back to the system. Increase this flag to return memory faster; decrease it to return memory slower. Reasonable rates are in the range [0,10]." .PP To modify the release rate during runtime, you can use the \fBsetParameter\f1\f1 command; for example: .PP .EX db.adminCommand( { setParameter: 1, tcmallocReleaseRate: 5.0 } ) .EE .PP You can also set \fBtcmallocReleaseRate\f1\f1 at startup time; for example: .PP .EX mongod \-\-setParameter "tcmallocReleaseRate=5.0" .EE .RE .PP \fBfassertOnLockTimeoutForStepUpDown\f1 .RS .PP Default: 15 seconds .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Allows a server that receives a request to step up or step down, to terminate if it is unable to comply (for example due to faulty server disks) within the timeout. This enables a cluster to successfully elect a new primary node and thus continue to be available. .PP \fBfassertOnLockTimeoutForStepUpDown\f1 defaults to 15 seconds. To disable nodes from fasserting, set \fBfassertOnLockTimeoutForStepUpDown=0\f1\&. .PP The following example disables nodes from fasserting: .PP .EX mongod \-\-setParameter fassertOnLockTimeoutForStepUpDown=0 .EE .RE .SS LOGGING PARAMETERS .PP \fBlogLevel\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Specify an integer between \fB0\f1 and \fB5\f1 signifying the verbosity of the \fBlogging\f1, where \fB5\f1 is the most verbose. .PP The default \fBlogLevel\f1\f1 is \fB0\f1 (Informational). .PP The following example sets the \fBlogLevel\f1\f1 to \fB2\f1: .PP .EX db.adminCommand( { setParameter: 1, logLevel: 2 } ) .EE .RS .IP \(bu 2 \fBlogComponentVerbosity\f1\f1 .IP \(bu 2 \fBsystemLog.verbosity\f1\f1 .RE .RE .PP \fBlogComponentVerbosity\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Sets the verbosity levels of various \fBcomponents\f1 for \fBlog messages\f1\&. The verbosity level determines the amount of \fBInformational and Debug\f1 messages MongoDB outputs. .PP The verbosity level can range from \fB0\f1 to \fB5\f1: .RS .IP \(bu 2 \fB0\f1 is the MongoDB\(aqs default log verbosity level, to include \fBInformational\f1 messages. .IP \(bu 2 \fB1\f1 to \fB5\f1 increases the verbosity level to include \fBDebug\f1 messages. .RE .PP For a component, you can also specify \fB\-1\f1 to inherit the parent\(aqs verbosity level. .PP To specify the verbosity level, use a document similar to the following: .PP .EX { verbosity: , : { verbosity: }, : { verbosity: , : { verbosity: } }, ... } .EE .PP For the components, you can specify just the \fB: \f1 in the document, unless you are setting both the parent verbosity level and that of the child component(s) as well: .PP .EX { verbosity: , : , : { verbosity: , : } ... } .EE .PP The top\-level \fBverbosity\f1 field corresponds to \fBsystemLog.verbosity\f1\f1 which sets the default level for all components. The default value of \fBsystemLog.verbosity\f1\f1 is \fB0\f1\&. .PP The components correspond to the following settings: .RS .IP \(bu 2 \fBaccessControl\f1\f1 .IP \(bu 2 \fBcommand\f1\f1 .IP \(bu 2 \fBcontrol\f1\f1 .IP \(bu 2 \fBftdc\f1\f1 .IP \(bu 2 \fBgeo\f1\f1 .IP \(bu 2 \fBindex\f1\f1 .IP \(bu 2 \fBnetwork\f1\f1 .IP \(bu 2 \fBquery\f1\f1 .IP \(bu 2 \fBreplication\f1\f1 .IP \(bu 2 \fBreplication.election\f1\f1 .IP \(bu 2 \fBreplication.heartbeats\f1\f1 .IP \(bu 2 \fBreplication.initialSync\f1\f1 .IP \(bu 2 \fBreplication.rollback\f1\f1 .IP \(bu 2 \fBrecovery\f1\f1 .IP \(bu 2 \fBsharding\f1\f1 .IP \(bu 2 \fBstorage\f1\f1 .IP \(bu 2 \fBstorage.journal\f1\f1 .IP \(bu 2 \fBtransaction\f1\f1 .IP \(bu 2 \fBwrite\f1\f1 .RE .PP Unless explicitly set, the component has the verbosity level of its parent. For example, \fBstorage\f1 is the parent of \fBstorage.journal\f1\&. That is, if you specify a \fBstorage\f1\f1 verbosity level, this level also applies to: .RS .IP \(bu 2 \fBstorage.journal\f1\f1 components \fIunless\f1 you specify the verbosity level for \fBstorage.journal\f1\f1\&. .IP \(bu 2 \fBstorage.recovery\f1\f1 components \fIunless\f1 you specify the verbosity level for \fBstorage.recovery\f1\f1\&. .RE .PP For example, the following sets the \fBdefault verbosity level\f1\f1 to \fB1\f1, the \fBquery\f1\f1 to \fB2\f1, the \fBstorage\f1\f1 to \fB2\f1, and the \fBstorage.journal\f1\f1 to \fB1\f1\&. .PP .EX db.adminCommand( { setParameter: 1, logComponentVerbosity: { verbosity: 1, query: { verbosity: 2 }, storage: { verbosity: 2, journal: { verbosity: 1 } } } } ) .EE .PP You can also set parameter \fBlogComponentVerbosity\f1\f1 at startup time, passing the verbosity level document as a string. .PP .EX mongod \-\-setParameter "logComponentVerbosity={command: 3}" .EE .PP \fBmongosh\f1\f1 also provides the \fBdb.setLogLevel()\f1\f1 to set the log level for a single component. For various ways to set the log verbosity level, see \fBConfigure Log Verbosity Levels\f1\&. .RE .PP \fBmaxLogSizeKB\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP \fIType\f1: non\-negative integer .PP \fIDefault\f1: 10 .PP Specifies the maximum size, in kilobytes, for an individual attribute field in a log entry; attributes exceeding this limit are truncated. .PP Truncated attribute fields print field content up to the \fBmaxLogSizeKB\f1\f1 limit and excise field content past that limit, retaining valid JSON formatting. Log entries that contain truncated attributes append a \fBtruncated\f1 object to the end of the log entry. .PP See \fBlog message truncation\f1 for more information. .PP A value of \fB0\f1 disables truncation entirely. Negative values for this parameter are not valid. .PP Using a large value, or disabling truncation with a value of \fB0\f1, may adversely affect system performance and negatively impact database operations. .PP The following example sets the maximum log line size to \fB20\f1 kilobytes: .PP .EX mongod \-\-setParameter maxLogSizeKB=20 .EE .RE .PP \fBquiet\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Sets quiet logging mode. If \fB1\f1, \fBmongod\f1\f1 will go into a quiet logging mode which will not log the following events/activities: .RS .IP \(bu 2 connection events; .IP \(bu 2 the \fBdrop\f1\f1 command, the \fBdropIndexes\f1\f1 command, the \fBdiagLogging\f1\f1 command, the \fBvalidate\f1\f1 command; and .IP \(bu 2 replication synchronization activities. .RE .PP Consider the following example which sets the \fBquiet\f1 parameter to \fB1\f1: .PP .EX db.adminCommand( { setParameter: 1, quiet: 1 } ) .EE .PP \fBsystemLog.quiet\f1\f1 .RE .PP \fBredactClientLogData\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP \fIType\f1: boolean .PP Available in MongoDB Enterprise only. .PP Configure the \fBmongod\f1\f1 or \fBmongos\f1\f1 to redact any message accompanying a given log event before logging. This prevents the program 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. .PP Use \fBredactClientLogData\f1\f1 in conjunction with \fBEncryption at Rest\f1 and \fBTLS/SSL (Transport Encryption)\f1 to assist compliance with regulatory requirements. .PP To enable log redaction on a running \fBmongod\f1 or \fBmongos\f1, use the following command: .PP .EX db.adminCommand( { setParameter: 1, redactClientLogData : true } ) .EE .PP \fBsecurity.redactClientLogData\f1\f1 .RE .PP \fBtraceExceptions\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Configures \fBmongod\f1\f1 to log full source code stack traces for every database and socket C++ exception, for use with debugging. If \fBtrue\f1, \fBmongod\f1\f1 will log full stack traces. .PP Consider the following example which sets the \fBtraceExceptions\f1 to \fBtrue\f1: .PP .EX db.adminCommand( { setParameter: 1, traceExceptions: true } ) .EE .PP \fBsystemLog.traceAllExceptions\f1\f1 .RE .PP \fBsuppressNoTLSPeerCertificateWarning\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP \fIType\f1: boolean .PP \fIDefault\f1: false .PP By default, a \fBmongod\f1\f1 or \fBmongos\f1\f1 with \fBTLS/SSL enabled\f1 and \fBnet.ssl.allowConnectionsWithoutCertificates\f1\f1 : \fBtrue\f1 lets clients connect without providing a certificate for validation while logging an warning. Set \fBsuppressNoTLSPeerCertificateWarning\f1 to \fB1\f1 or \fBtrue\f1 to suppress those warnings. .PP The following operation sets \fBsuppressNoTLSPeerCertificateWarning\f1 to \fBtrue\f1: .PP .EX db.adminCommand( { setParameter: 1, suppressNoTLSPeerCertificateWarning: true} ) .EE .RE .SS DIAGNOSTIC PARAMETERS .PP To facilitate analysis of the MongoDB server behavior by MongoDB engineers, MongoDB logs server statistics to diagnostic files at periodic intervals. .PP For \fBmongod\f1\f1, the diagnostic data files are stored in the \fBdiagnostic.data\f1 directory under the \fBmongod\f1\f1 instance\(aqs \fB\-\-dbpath\f1 or \fBstorage.dbPath\f1\f1\&. .PP For \fBmongos\f1\f1, the diagnostic data files, by default, are stored in a directory under the \fBmongos\f1\f1 instance\(aqs \fB\-\-logpath\f1 or \fBsystemLog.path\f1\f1 directory. The diagnostic data directory is computed by truncating the logpath\(aqs file extension(s) and concatenating \fBdiagnostic.data\f1 to the remaining name. .PP For example, if \fBmongos\f1\f1 has \fB\-\-logpath /var/log/mongodb/mongos.log.201708015\f1, then the diagnostic data directory is \fB/var/log/mongodb/mongos.diagnostic.data/\f1 directory. To specify a different diagnostic data directory for \fBmongos\f1\f1, set the \fBdiagnosticDataCollectionDirectoryPath\f1\f1 parameter. .PP The following parameters support diagnostic data capture (FTDC): .PP The default values for the diagnostic data capture interval and the maximum sizes are chosen to provide useful data to MongoDB engineers with minimal impact on performance and storage size. Typically, these values will only need modifications as requested by MongoDB engineers for specific diagnostic purposes. .PP \fBdiagnosticDataCollectionEnabled\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP \fIType\f1: boolean .PP \fIDefault\f1: true .PP Determines whether to enable the collecting and logging of data for diagnostic purposes. Diagnostic logging is enabled by default. .PP For example, the following disables the diagnostic collection: .PP .EX mongod \-\-setParameter diagnosticDataCollectionEnabled=false .EE .RE .PP \fBdiagnosticDataCollectionDirectoryPath\f1 .RS .PP \fIType\f1: String .PP Available for \fBmongos\f1\f1 only. .PP Specify the directory for the diagnostic directory for \fBmongos\f1\f1\&. If the directory does not exist, \fBmongos\f1\f1 creates the directory. .PP If unspecified, the diagnostic data directory is computed by truncating the \fBmongos\f1\f1 instance\(aqs \fB\-\-logpath\f1 or \fBsystemLog.path\f1\f1 file extension(s) and concatenating \fBdiagnostic.data\f1\&. .PP For example, if \fBmongos\f1\f1 has \fB\-\-logpath /var/log/mongodb/mongos.log.201708015\f1, then the diagnostic data directory is \fB/var/log/mongodb/mongos.diagnostic.data/\f1\&. .PP If the \fBmongos\f1\f1 cannot create the specified directory, the diagnostic data capture is disabled for that instance. This is often caused by: .RS .IP \(bu 2 a file with the same name already exists in the path, or .IP \(bu 2 the process does not have permissions to create the directory. .RE .RE .PP \fBdiagnosticDataCollectionDirectorySizeMB\f1 .RS .PP Increased default size to 200 megabytes. .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP \fIType\f1: integer .PP \fIDefault\f1: 200 .PP Specifies the maximum size, in megabytes, of the \fBdiagnostic.data\f1 directory. If directory size exceeds this number, the oldest \fBdiagnostic files in the directory\f1 are automatically deleted based on the timestamp in the file name. .PP For example, the following sets the maximum size of the directory to \fB250\f1 megabytes: .PP .EX mongod \-\-setParameter diagnosticDataCollectionDirectorySizeMB=250 .EE .PP The minimum value for \fBdiagnosticDataCollectionDirectorySizeMB\f1\f1 is \fB10\f1 megabytes. \fBdiagnosticDataCollectionDirectorySizeMB\f1\f1 must be greater than maximum diagnostic file size \fBdiagnosticDataCollectionFileSizeMB\f1\f1\&. .RE .PP \fBdiagnosticDataCollectionFileSizeMB\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP \fIType\f1: integer .PP \fIDefault\f1: 10 .PP Specifies the maximum size, in megabytes, of each \fBdiagnostic file\f1\&. If the file exceeds the maximum file size, MongoDB creates a new file. .PP For example, the following sets the maximum size of each diagnostic file to \fB20\f1 megabytes: .PP .EX mongod \-\-setParameter diagnosticDataCollectionFileSizeMB=20 .EE .PP The minimum value for \fBdiagnosticDataCollectionFileSizeMB\f1\f1 is \fB1\f1 megabyte. .RE .PP \fBdiagnosticDataCollectionPeriodMillis\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP \fIType\f1: integer .PP \fIDefault\f1: 1000 .PP Specifies the interval, in milliseconds, at which to collect diagnostic data. .PP For example, the following sets the interval to \fB5000\f1 milliseconds or 5 seconds: .PP .EX mongod \-\-setParameter diagnosticDataCollectionPeriodMillis=5000 .EE .PP The minimum value for \fBdiagnosticDataCollectionPeriodMillis\f1\f1 is \fB100\f1 milliseconds. .RE .SS REPLICATION AND CONSISTENCY .PP \fBdisableSplitHorizonIPCheck\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP \fIType\f1: boolean .PP \fIDefault\f1: false .RE .PP To configure cluster nodes for split horizon DNS (https://en.wikipedia.org/wiki/Split\-horizon_DNS), use host names instead of IP addresses. .PP Starting in MongoDB v5.0, \fBreplSetInitiate\f1\f1 and \fBreplSetReconfig\f1\f1 reject configurations that use IP addresses instead of hostnames. .PP Use \fBdisableSplitHorizonIPCheck\f1\f1 to modify nodes that cannot be updated to use host names. The parameter only applies to the configuration commands. .PP \fBmongod\f1\f1 and \fBmongos\f1\f1 do not rely on \fBdisableSplitHorizonIPCheck\f1\f1 for validation at startup. Legacy \fBmongod\f1\f1 and \fBmongos\f1\f1 instances that use IP addresses instead of host names will start after an upgrade. .PP Instances that are configured with IP addresses log a warning to use host names instead of IP addresses. .PP To allow configuration changes using IP addresses, set \fBdisableSplitHorizonIPCheck=true\f1 using the command line: .PP .EX /usr/local/bin/mongod \-\-setParameter disableSplitHorizonIPCheck=true \-f /etc/mongod.conf .EE .PP To allow configuration changes using IP addresses, set \fBdisableSplitHorizonIPCheck=true\f1 using the node\(aqs configuration file: .PP .EX setParameter: disableSplitHorizonIPCheck: true .EE .PP If you attempt to update \fBdisableSplitHorizonIPCheck\f1 at runtime, \fBdb.adminCommand()\f1\f1 returns an error: .PP .EX db.adminCommand( { setParameter: 1, "disableSplitHorizonIPCheck": true } ) MongoServerError: not allowed to change [disableSplitHorizonIPCheck] at runtime .EE .PP \fBenableOverrideClusterChainingSetting\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP \fIType\f1: boolean .PP \fIDefault\f1: false .PP If \fBenableOverrideClusterChainingSetting\f1\f1 is \fBtrue\f1, replica set \fBsecondary\f1 members can replicate data from other secondary members even if \fBsettings.chainingAllowed\f1\f1 is \fBfalse\f1\&. .PP You can only set \fBenableOverrideClusterChainingSetting\f1\f1 at startup and cannot change this setting with the \fBsetParameter\f1\f1 command. .PP For example, to set the \fBenableOverrideClusterChainingSetting\f1\f1 for a \fBmongod\f1\f1 instance to \fBtrue\f1: .PP .EX mongod \-\-setParameter enableOverrideClusterChainingSetting=true .EE .RE .PP \fBlogicalSessionRefreshMillis\f1 .RS .PP New in version 4.0.4 (and version 3.6.9). .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP \fIType\f1: integer .PP \fIDefault\f1: 300000 (5 minutes) .PP The interval (in milliseconds) at which the cache refreshes its logical session records against the main session store. .PP You can only set \fBlogicalSessionRefreshMillis\f1\f1 at startup and cannot change this setting with the \fBsetParameter\f1\f1 command. .PP For example, to set the \fBlogicalSessionRefreshMillis\f1\f1 for a \fBmongod\f1\f1 instance to 10 minutes: .PP .EX mongod \-\-setParameter logicalSessionRefreshMillis=600000 .EE .RE .PP \fBlocalLogicalSessionTimeoutMinutes\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP \fIType\f1: integer .PP \fIDefault\f1: 30 .PP This parameter is intended for testing purposes only and not for production use. .PP The time in minutes that a \fBsession\f1 remains active after its most recent use. Sessions that have not received a new read/write operation from the client or been refreshed with \fBrefreshSessions\f1\f1 within this threshold are cleared from the cache. State associated with an expired session may be cleaned up by the server at any time. .PP This parameter applies only to the instance on which it is set. To set this parameter on replica sets and sharded clusters, you must specify the same value on every member; otherwise, sessions will not function properly. .PP You can only set \fBlocalLogicalSessionTimeoutMinutes\f1\f1 at startup and cannot change this setting with the \fBsetParameter\f1\f1 command. .PP For example, to set the \fBlocalLogicalSessionTimeoutMinutes\f1\f1 for a test \fBmongod\f1\f1 instance to 20 minutes: .PP .EX mongod \-\-setParameter localLogicalSessionTimeoutMinutes=20 .EE .RE .PP \fBmaxAcceptableLogicalClockDriftSecs\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP \fIType\f1: integer .PP \fIDefault\f1: 31536000 (1 year) .PP The maximum amount by which the current cluster time can be advanced; specifically, \fBmaxAcceptableLogicalClockDriftSecs\f1\f1 is the maximum difference between the new value of the cluster time and the current cluster time. Cluster time is a logical time used for ordering of operations. .PP You cannot advance the cluster time to a new value if the new cluster time differs from the current cluster time by more than \fBmaxAcceptableLogicalClockDriftSecs\f1\f1\&. .PP You can only set \fBmaxAcceptableLogicalClockDriftSecs\f1\f1 at startup and cannot change this setting with the \fBsetParameter\f1\f1 command. .PP For example, to set the \fBmaxAcceptableLogicalClockDriftSecs\f1\f1 for a \fBmongod\f1\f1 instance to 15 minutes: .PP .EX mongod \-\-setParameter maxAcceptableLogicalClockDriftSecs=900 .EE .RE .PP \fBmaxSessions\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP \fIType\f1: integer .PP \fIDefault\f1: 1000000 .PP The maximum number of sessions that can be cached. .PP You can only set \fBmaxSessions\f1\f1 during start\-up. .PP For example, to set the \fBmaxSessions\f1\f1 for a \fBmongod\f1\f1 instance to 1000: .PP .EX mongod \-\-setParameter maxSessions=1000 .EE .RE .PP \fBoplogBatchDelayMillis\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP \fIType\f1: integer .PP \fIDefault\f1: 0 .PP The number of milliseconds to delay applying batches of oplog operations on secondary nodes. By default, \fBoplogBatchDelayMillis\f1 is \fB0\f1, meaning oplog batches are applied with no delay. When there is no delay, MongoDB may apply frequent, small oplog batches to secondaries. .PP Increasing \fBoplogBatchDelayMillis\f1 causes MongoDB to apply oplog batches less frequently on secondaries, with each batch containing larger amounts of data. This reduces IOPS (Input/Output Operations Per Second) on secondaries, but adds latency for writes with write concern \fB"majority"\f1\f1\&. .PP You can only set \fBoplogBatchDelayMillis\f1 at startup. You cannot set \fBoplogBatchDelayMillis\f1 during runtime. .PP For example, run the following command to set the \fBoplogBatchDelayMillis\f1 for a \fBmongod\f1\f1 instance to 20 milliseconds: .PP .EX mongod \-\-setParameter oplogBatchDelayMillis=20 .EE .RE .PP \fBstoreFindAndModifyImagesInSideCollection\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP \fIType\f1: boolean .PP \fIDefault\f1: true .PP Determines whether the temporary documents required for \fBretryable\f1 \fBfindAndModify\f1\f1 commands are stored in the \fIside\f1 collection (\fBconfig.image_collection\f1). .PP If \fBstoreFindAndModifyImagesInSideCollection\f1\f1 is: .RS .IP \(bu 2 \fBtrue\f1, the temporary documents are stored in the side collection. .IP \(bu 2 \fBfalse\f1, the temporary documents are stored in the \fBreplica set oplog\f1\&. .RE .PP Keep \fBstoreFindAndModifyImagesInSideCollection\f1\f1 set to \fBtrue\f1 if you: .RS .IP \(bu 2 Have a large \fBretryable\f1 \fBfindAndModify\f1\f1 workload. .IP \(bu 2 Require more temporary document space for \fBretryable\f1 \fBfindAndModify\f1\f1 commands than is available in the \fBreplica set oplog\f1\&. .RE .PP \fBSecondaries\f1 may experience increased CPU usage when \fBstoreFindAndModifyImagesInSideCollection\f1\f1 is \fBtrue\f1\&. .PP For example, to set \fBstoreFindAndModifyImagesInSideCollection\f1\f1 to \fBfalse\f1 during startup: .PP .EX mongod \-\-setParameter storeFindAndModifyImagesInSideCollection=false .EE .PP During runtime, you can also set the parameter with the \fBsetParameter\f1\f1 command: .PP .EX db.adminCommand( { setParameter: 1, storeFindAndModifyImagesInSideCollection: false } ) .EE .RE .PP \fBTransactionRecordMinimumLifetimeMinutes\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP \fIType\f1: integer .PP \fIDefault\f1: 30 .PP The minimum lifetime a transaction record exists in the \fBtransactions\f1\f1 collection before the record becomes eligible for cleanup. .PP You can only set \fBTransactionRecordMinimumLifetimeMinutes\f1\f1 at startup and cannot change this setting with the \fBsetParameter\f1\f1 command. .PP For example, to set the \fBTransactionRecordMinimumLifetimeMinutes\f1\f1 for a \fBmongod\f1\f1 instance to 20 minutes: .PP .EX mongod \-\-setParameter TransactionRecordMinimumLifetimeMinutes=20 .EE .PP \fBlocalLogicalSessionTimeoutMinutes\f1\f1 .RE .PP \fBenableFlowControl\f1 .RS .PP \fIType\f1: boolean .PP \fIDefault\f1: true .PP Enables or disables the mechanism that controls the rate at which the primary applies its writes with the goal of keeping the secondary members\(aq \fBmajority committed\f1\f1 lag under a configurable maximum value. .PP For flow control to engage, the replica set/sharded cluster must have: \fBfeatureCompatibilityVersion (FCV)\f1 of \fB4.2\f1 and read concern \fBmajority enabled\f1\f1\&. That is, enabled flow control has no effect if FCV is not \fB4.2\f1 or if read concern majority is disabled. .RE .PP \fBflowControlTargetLagSeconds\f1 .RS .PP \fIType\f1: integer .PP \fIDefault\f1: 10 .PP The target maximum \fBmajority committed\f1\f1 lag when running with flow control. When flow control is enabled, the mechanism attempts to keep the \fBmajority committed\f1\f1 lag under the specified seconds. The parameter has no effect if flow control is disabled. .PP The specified value must be greater than 0. .PP In general, the default settings should suffice; however, if modifying from the default value, decreasing, rather than increasing, the value may prove to be more useful. .RE .PP \fBflowControlWarnThresholdSeconds\f1 .RS .PP \fIType\f1: integer .PP \fIDefault\f1: 10 .PP The amount of time to wait to log a warning once the flow control mechanism detects the majority commit point has not moved. .PP The specified value must be greater than or equal to 0, with 0 to disable warnings. .RE .PP \fBinitialSyncTransientErrorRetryPeriodSeconds\f1 .RS .PP \fIType\f1: integer .PP \fIDefault\f1: 86400 .PP The amount of time in seconds a secondary performing initial sync attempts to resume the process if interrupted by a transient network error. The default value is equivalent to 24 hours. .RE .PP \fBinitialSyncSourceReadPreference\f1 .RS .PP \fIType\f1: String .PP Available for \fBmongod\f1\f1 only. .PP The preferred source for performing \fBinitial sync\f1\&. Specify one of the following read preference modes: .RS .IP \(bu 2 \fBprimary\f1\f1 .IP \(bu 2 \fBprimaryPreferred\f1\f1 (Default for voting replica set members) .IP \(bu 2 \fBsecondary\f1\f1 .IP \(bu 2 \fBsecondaryPreferred\f1\f1 .IP \(bu 2 \fBnearest\f1\f1 (Default for newly added \fIor\f1 non\-voting replica set members) .RE .PP If the replica set has disabled \fBchaining\f1\f1, the default \fBinitialSyncSourceReadPreference\f1\f1 read preference mode is \fBprimary\f1\f1\&. .PP You cannot specify a tag set or \fBmaxStalenessSeconds\f1 to \fBinitialSyncSourceReadPreference\f1\f1\&. .PP If the \fBmongod\f1\f1 cannot find a sync source based on the specified read preference, it logs an error and restarts the initial sync process. The \fBmongod\f1\f1 exits with an error if it cannot complete the initial sync process after \fB10\f1 attempts. For more information on sync source selection, see \fBInitial Sync Source Selection\f1\&. .PP \fBinitialSyncSourceReadPreference\f1\f1 takes precedence over the replica set\(aqs \fBsettings.chainingAllowed\f1\f1 setting when selecting an initial sync source. After a replica set member successfully completes initial sync, it defers to the value of \fBchainingAllowed\f1\f1 when selecting a replication sync source. .PP You can only set this parameter on startup, using either the \fBsetParameter\f1\f1 configuration file setting or the \fB\-\-setParameter\f1\f1 command line option. .RE .PP \fBinitialSyncMethod\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP \fIType\f1: String .PP \fIDefault\f1: \fBlogical\f1 .PP Available only in MongoDB Enterprise. .PP Method used for \fBinitial sync\f1\&. .PP Set to \fBlogical\f1 to use \fBlogical initial sync\f1\&. Set to \fBfileCopyBased\f1 to use \fBfile copy based initial sync\f1\&. .PP This parameter only affects the sync method for the member on which it is specified. Setting this parameter on a single replica set member does not affect the sync method of any other replica set members. .PP You can only set this parameter on startup, using either the \fBsetParameter\f1\f1 configuration file setting or the \fB\-\-setParameter\f1\f1 command line option. .RE .PP \fBmaxNumSyncSourceChangesPerHour\f1 .RS .PP \fIType\f1: integer .PP \fIDefault\f1: 3 .PP \fBSync sources\f1 are evaluated each time a sync source is updated and each time a node fetches a batch of oplog entries. If there are more than \fBmaxNumSyncSourceChangesPerHour\f1 source changes in an hour, the node temporarily stops re\-evaluating that sync source. If this parameter is set with a high value, the node may make unnecessary source changes. .PP This parameter will not prevent a node from starting to sync from another node if it doesn\(aqt have a sync source. The node will re\-evaluate if a sync source becomes invalid. Similarly, if the primary changes and chaining is disabled, the node will update to sync from the new primary. .RE .PP \fBoplogFetcherUsesExhaust\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP \fIType\f1: boolean .PP \fIDefault\f1: true .PP Enables or disables \fBstreaming replication\f1\&. Set the value to \fBtrue\f1 to enable streaming replication. .PP Set the value to \fBfalse\f1 to disable streaming replication. If disabled, secondaries fetch batches of \fBoplog\f1 entries by issuing a request to their \fIsync from\f1 source and waiting for a response. This requires a network roundtrip for each batch of \fBoplog\f1 entries. .PP You can only set this parameter on startup, using either the \fBsetParameter\f1\f1 configuration file setting or the \fB\-\-setParameter\f1\f1 command line option. .RE .PP \fBoplogInitialFindMaxSeconds\f1 .RS .PP \fIType\f1: integer .PP \fIDefault\f1: 60 .PP Available for \fBmongod\f1\f1 only. .PP Maximum time in seconds for a member of a \fBreplica set\f1 to wait for the \fBfind\f1\f1 command to finish during \fBdata synchronization\f1\&. .RE .PP \fBreplWriterThreadCount\f1 .RS .PP \fIType\f1: integer .PP \fIDefault\f1: 16 .PP Available for \fBmongod\f1\f1 only. .PP Maximum number of threads to use to apply replicated operations in parallel. Values can range from 1 to 256 inclusive. However, the maximum number of threads used is capped at twice the number of available cores. .PP You can only set \fBreplWriterThreadCount\f1\f1 at startup and cannot change this setting with the \fBsetParameter\f1\f1 command. .PP \fBreplWriterMinThreadCount\f1\f1 .RE .PP \fBreplWriterMinThreadCount\f1 .RS .PP \fIType\f1: integer .PP \fIDefault\f1: 0 .PP Available for \fBmongod\f1\f1 only. .PP Minimum number of threads to use to apply replicated operations in parallel. Values can range from 0 to 256 inclusive. You can only set \fBreplWriterMinThreadCount\f1\f1 at startup and cannot change this setting with the \fBsetParameter\f1\f1 command. .PP Parallel application of replication operations uses up to \fBreplWriterThreadCount\f1\f1 threads. If \fBreplWriterMinThreadCount\f1\f1 is configured with a value less than \fBreplWriterThreadCount\f1\f1, the thread pool will timeout idle threads until the total count of threads in the thread pool is equal to \fBreplWriterMinThreadCount\f1\f1\&. .PP \fBreplWriterMinThreadCount\f1\f1 must be configured with a value that is less than or equal to \fBreplWriterThreadCount\f1\f1\&. .RE .PP \fBrollbackTimeLimitSecs\f1 .RS .PP \fIType\f1: 64\-bit integer .PP \fIDefault\f1: 86400 (1 day) .PP Maximum age of data that can be rolled back. Negative values for this parameter are not valid. .PP Starting in MongoDB 4.2+ and 4.0.13+, if the time between the end of the to\-be\-rolledback instance\(aqs oplog and the first operation after the common point (the last point where the source node and the to\-be\-rolledback node had the same data) exceeds this value, the rollback will fail. .PP In MongoDB 4.0.0\-4.0.12, if the time between the end of the to\-be\-rolledback instance\(aqs oplog and the common point (the last point where the source node and the to\-be\-rolledback node had the same data) exceeds this value, the rollback will fail. .PP To effectively have an unlimited rollback period, set the value to \fB2147483647\f1 which is the maximum value allowed and equivalent to roughly 68 years. .RE .PP \fBwaitForSecondaryBeforeNoopWriteMS\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP \fIType\f1: integer .PP \fIDefault\f1: 10 .PP The length of time (in milliseconds) that a secondary must wait if the \fBafterClusterTime\f1 is greater than the last applied time from the oplog. After the \fBwaitForSecondaryBeforeNoopWriteMS\f1 passes, if the \fBafterClusterTime\f1 is still greater than the last applied time, the secondary makes a no\-op write to advance the last applied time. .PP The following example sets the \fBwaitForSecondaryBeforeNoopWriteMS\f1\f1 to 20 milliseconds: .PP .EX mongod \-\-setParameter waitForSecondaryBeforeNoopWriteMS=20 .EE .PP During runtime, you can also set the parameter with the \fBsetParameter\f1\f1 command: .PP .EX db.adminCommand( { setParameter: 1, waitForSecondaryBeforeNoopWriteMS: 20 } ) .EE .RE .PP \fBcreateRollbackDataFiles\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP \fIType\f1: boolean .PP \fIDefault\f1: true .PP Flag that determines whether MongoDB creates \fBrollback files\f1 that contains documents affected during a rollback. .PP By default, \fBcreateRollbackDataFiles\f1\f1 is \fBtrue\f1 and MongoDB creates the rollback files. .PP The following example sets \fBcreateRollbackDataFiles\f1\f1 to false so that the rollback files are not created: .PP .EX mongod \-\-setParameter createRollbackDataFiles=false .EE .PP During runtime, you can also set the parameter with the \fBsetParameter\f1\f1 command: .PP .EX db.adminCommand( { setParameter: 1, createRollbackDataFiles: false } ) .EE .PP For more information, see \fBCollect Rollback Data\f1\&. .RE .PP \fBenableElectionHandoff\f1 .RS .PP \fIType\f1: boolean .PP \fIDefault\f1: true .PP A flag that can reduce the downtime after the primary steps down from either the \fBrs.stepDown()\f1\f1 method or the \fBreplSetStepDown\f1\f1 command. .PP If the flag is true, when a primary steps down after \fBrs.stepDown()\f1\f1 (or the \fBreplSetStepDown\f1\f1 command without the \fBforce: true\f1), the primary nominates an eligible secondary to call an election immediately. .PP If the flag is false, after the step down, secondaries can wait up to \fBsettings.electionTimeoutMillis\f1\f1 before calling an election. .PP An eligible secondary must be caught up with the stepped down primary and have \fBpriority\f1\f1 greater than 0. If multiple secondary members meet this criteria, the stepped down primary selects the eligible secondary with the highest \fBpriority\f1\f1\&. If the more than one eligible secondary members have the same \fBpriority\f1\f1, the stepped down primary selects the secondary with the lowest \fB_id\f1\f1\&. The stepped down primary does not wait for the effects of the handoff. .PP The parameter has no impact if the primary steps down for reasons other than \fBrs.stepDown()\f1\f1 (or the \fBreplSetStepDown\f1\f1 command without the \fBforce: true\f1). .RE .PP \fBreplBatchLimitBytes\f1 .RS .PP \fIDefault\f1: 104857600 (100MB) .PP Sets the maximum oplog application batch size in bytes. .PP Values can range from 16777216 (16MB) to 104857600 (100MB) inclusive. .PP The following example sets \fBreplBatchLimitBytes\f1\f1 to 64 MB so that the rollback files are not created: .PP .EX mongod \-\-setParameter replBatchLimitBytes=67108864 .EE .PP During runtime, you can also set the parameter with the \fBsetParameter\f1\f1 command: .PP .EX db.adminCommand( { setParameter: 1, replBatchLimitBytes: 64 * 1024 * 1024 } ) .EE .RE .PP \fBmirrorReads\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP \fINew in version 4.4\f1 .PP \fIType\f1: Document .PP \fIDefault\f1: \fB{ samplingRate: 0.01, maxTimeMS: 1000 }\f1 .PP Specifies the settings for \fBmirrored reads\f1 for the \fBmongod\f1\f1 instance. The settings only take effect when the member is a primary. .PP The parameter \fBmirrorReads\f1\f1 takes a JSON document with the following fields: .RS .IP \(bu 2 .RS .IP \(bu 4 Field .IP \(bu 4 Description .RE .IP \(bu 2 .RS .IP \(bu 4 \fBsamplingRate\f1 .IP \(bu 4 The sampling rate used to mirror a subset of \fBoperations that support mirroring\f1 to a subset of electable (specifically, \fBpriority greater than 0\f1\f1) secondaries. That is, the primary mirrors reads to each electable secondary at the specified sampling rate. .IP Valid values are: .RS .IP \(bu 6 .RS .IP \(bu 8 \fB0.0\f1 .IP \(bu 8 Turns off mirroring. .RE .IP \(bu 6 .RS .IP \(bu 8 \fB1.0\f1 .IP \(bu 8 The primary mirrors all \fBoperations that supports mirroring\f1 to each electable secondary. .RE .IP \(bu 6 .RS .IP \(bu 8 Number between \fB0.0\f1 and \fB1.0\f1 (exclusive) .IP \(bu 8 The primary randomly samples each electable secondary at the specified rate to be sent mirrored reads. .RE .RE .IP For example, given a replica set with a primary and two electable secondaries and a sampling rate of \fB0.10\f1, the primary mirrors reads to each electable secondary at the sampling rate of 10 percent such that one read may be mirrored to one secondary and not to the other or to both or to neither. That is, if the primary receives \fB100\f1 operations that can be mirrored, the sampling rate of \fB0.10\f1 may result in \fB8\f1 reads being mirrored to one secondary and \fB13\f1 reads to the other or \fB10\f1 to each, etc. .IP The default value is \fB0.01\f1\&. .RE .IP \(bu 2 .RS .IP \(bu 4 \fBmaxTimeMS\f1 .IP \(bu 4 The maximum time in milliseconds for the mirrored reads. The default value is \fB1000\f1\&. .IP The \fBmaxTimeMS\f1 for the mirrored reads is separate from the \fBmaxTimeMS\f1 of the original read being mirrored. .RE .RE .PP You can set \fBmirrorReads\f1\f1 during startup in the \fBconfiguration file\f1\f1 or with the \fB\-\-setParameter\f1 option on the command line. If specifying from the configuration file or on the command line, \fBenclose the\f1 \fBmirrorReads\f1 \fBdocument in quotes\f1\&. .PP For example, the following sets the mirror reads sampling rate to \fB0.10\f1 from the command line: .PP .EX mongod \-\-setParameter mirrorReads=\(aq{ samplingRate: 0.10 }\(aq .EE .PP Or, to specify in a configuration file: .PP .EX setParameter: mirrorReads: \(aq{samplingRate: 0.10}\(aq .EE .PP Or if using the \fBsetParameter\f1\f1 command in a \fBmongosh\f1\f1 session that is connected to a running \fBmongod\f1\f1, do \fBnot\f1 enclose the document in quotes: .PP .EX db.adminCommand( { setParameter: 1, mirrorReads: { samplingRate: 0.10 } } ) .EE .RE .PP \fBallowMultipleArbiters\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP \fIType\f1: boolean .PP \fIDefault\f1: false .PP Specifies whether the replica set allows the use of multiple arbiters. .PP The use of multiple arbiters is not recommended: .RS .IP \(bu 2 Multiple arbiters prevent the reliable use of the majority write concern. MongoDB counts arbiters in calculating a membership majority, but arbiters do not store data. With the inclusion of multiple arbiters, it\(aqs possible for a majority write operation to return success before the write replicates to a majority of data bearing nodes. .IP \(bu 2 Multiple arbiters allow replica sets to accept writes even when the replica set doesn\(aqt have sufficient secondaries for data replication. .RE .PP For more information, see \fBConcerns with Multiple Arbiters\f1\&. .PP The parameter can only be set during startup: .PP .EX mongod \-\-setParameter allowMultipleArbiters=true .EE .RE .SS SHARDING PARAMETERS .PP Starting in version 4.2, MongoDB removes the parameter \fBAsyncRequestsSenderUseBaton\f1 and always enables the performance enhancement controlled by the parameter. .PP \fBchunkDefragmentationThrottlingMS\f1 .RS .PP \fIType\f1: integer .PP \fIDefault\f1: 0 .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Specifies the minimum time period (in milliseconds) between consecutive split and merge commands run by the \fBbalancer\f1 when the \fBchunks\f1 in a \fBsharded\f1 collection are defragmented. \fBchunkDefragmentationThrottlingMS\f1\f1 limits the rate of split and merge commands. .PP The following example sets \fBchunkDefragmentationThrottlingMS\f1\f1 to \fB10\f1 milliseconds: .PP .EX mongod \-\-setParameter chunkDefragmentationThrottlingMS=10 .EE .PP During runtime, you can also set the parameter with the \fBsetParameter\f1\f1 command: .PP .EX db.adminCommand( { setParameter: 1, chunkDefragmentationThrottlingMS: 10 } ) .EE .RE .PP \fBdisableResumableRangeDeleter\f1 .RS .PP \fIType\f1: boolean .PP \fIDefault\f1: false .PP Available for \fBmongod\f1\f1 only. .PP If set on a shard\(aqs primary, specifies if range deletion is paused on the shard. If set to \fBtrue\f1, cleanup of \fBchunk\(garanges containing :term:\(gaorphaned documents\f1 is paused. The shard can continue to donate chunks to other shards, but the donated documents will not be removed from this shard until you set this parameter to \fBfalse\f1\&. This shard can continue to receive chunks from other shards as long as it does not have a pending range deletion task in the \fBconfig.rangeDeletions\f1\f1 collection that overlaps with the incoming chunk\(aqs range. .PP When \fBdisableResumableRangeDeleter\f1\f1 is \fBtrue\f1, chunk migrations fail if \fBorphaned documents\f1 exist on the recipient shard\(aqs primary in the same range as the incoming chunks. .PP The parameter has no effect on the \fBmongod\f1\f1 if it is not the shard\(aqs primary. .PP If you set \fBdisableResumableRangeDeleter\f1\f1 parameter to \fBtrue\f1, ensure that you apply it consistently for all members in the shard\(aqs replica set. In the event of a failover, this setting\(aqs value on the new primary dictates the behavior of the range deleter. .PP You can only set this parameter during start\-up and cannot change this setting using the \fBsetParameter\f1\f1 database command. .PP .EX mongod \-\-setParameter disableResumableRangeDeleter=false .EE .RE .PP \fBenableShardedIndexConsistencyCheck\f1 .RS .PP \fINew in version 4.4 (and 4.2.6).\f1 .PP \fIType\f1: boolean .PP \fIDefault\f1: true .PP Available for \fBmongod\f1\f1 only. .PP If set on the config server\(aqs primary, enables or disables the index consistency check for sharded collections. The parameter has no effect on the \fBmongod\f1\f1 if it is not the config server\(aqs primary. .PP The following example sets \fBenableShardedIndexConsistencyCheck\f1\f1 to \fBfalse\f1 for a config server primary: .PP .EX mongod \-\-setParameter enableShardedIndexConsistencyCheck=false .EE .PP During runtime, you can also set the parameter with the \fBsetParameter\f1\f1 command: .PP .EX db.adminCommand( { setParameter: 1, enableShardedIndexConsistencyCheck: false } ) .EE .RS .IP \(bu 2 \fBshardedIndexConsistencyCheckIntervalMS\f1\f1 parameter .IP \(bu 2 \fBshardedIndexConsistency\f1\f1 metrics returned by the \fBserverStatus\f1\f1 command. .RE .RE .PP \fBshardedIndexConsistencyCheckIntervalMS\f1 .RS .PP \fINew in version 4.4 (and 4.2.6).\f1 .PP \fIType\f1: integer .PP \fIDefault\f1: 600000 .PP Available for \fBmongod\f1\f1 only. .PP If set on the config server\(aqs primary, the interval, in milliseconds, at which the config server\(aqs primary checks the index consistency of sharded collections. The parameter has no effect on the \fBmongod\f1\f1 if it is not the config server\(aqs primary. .PP You can only set the parameter during startup, and cannot change this setting using the \fBsetParameter\f1\f1 database command. .PP For example, the following sets the interval at 300000 milliseconds (5 minutes) at startup: .PP .EX mongod \-\-setParameter shardedIndexConsistencyCheckIntervalMS=300000 .EE .RS .IP \(bu 2 \fBenableShardedIndexConsistencyCheck\f1\f1 parameter .IP \(bu 2 \fBshardedIndexConsistency\f1\f1 metrics returned by the \fBserverStatus\f1\f1 commandq .RE .RE .PP \fBenableFinerGrainedCatalogCacheRefresh\f1 .RS .PP \fIType\f1: boolean .PP \fIDefault\f1: true .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP This parameter allows the catalog cache to be refreshed only if the shard needs to be refreshed. If disabled, any stale chunk will cause the entire chunk distribution for a collection to be considered stale and force all \fBrouters\f1 who contact the shard to refresh their shard catalog cache. .PP You can only set this parameter during start\-up and cannot change this setting using the \fBsetParameter\f1\f1 database command. .PP .EX mongod \-\-setParameter enableFinerGrainedCatalogCacheRefresh=true mongos \-\-setParameter enableFinerGrainedCatalogCacheRefresh=true .EE .RS .IP \(bu 2 \fBSharding\f1 .IP \(bu 2 \fBshardingStatistics.catalogCache\f1\f1 .RE .RE .PP \fBmaxTimeMSForHedgedReads\f1 .RS .PP \fIType\f1: integer .PP \fIDefault\f1: 150 .PP Available for \fBmongos\f1\f1 only. .PP Specifies the maximum time limit (in milliseconds) for the \fBhedged read\f1\&. That is, the additional read sent to hedge the read operation uses the \fBmaxTimeMS\f1 value of \fBmaxTimeMSForHedgedReads\f1\f1 while the read operation that is being hedged uses the \fBmaxTimeMS\f1 value specified for the operation. .PP For example, to set the limit to 200 milliseconds, you can issue the following during startup: .PP .EX mongos \-\-setParameter maxTimeMSForHedgedReads=200 .EE .PP Or if using the \fBsetParameter\f1\f1 command in a \fBmongosh\f1\f1 session that is connected to a running \fBmongos\f1\f1: .PP .EX db.adminCommand( { setParameter: 1, maxTimeMSForHedgedReads: 200 } ) .EE .RS .IP \(bu 2 \fBreadHedgingMode\f1\f1 .IP \(bu 2 \fBHedged Reads\f1 .RE .RE .PP \fBmaxCatchUpPercentageBeforeBlockingWrites\f1 .RS .PP \fIType\f1: integer .PP \fIDefault\f1: 10 .PP Available for \fBmongod\f1\f1 only. .PP For \fBmoveChunk\f1\f1 operations, specifies the maximum percentage of untrasferred data allowed by the migration protocol (expressed in percentage of the total chunk size) to transition from the \fBcatchup\f1 phase to the \fBcommit\f1 phase. .PP Setting a higher catchup percentage can decrease the amount of time it takes for the migration to complete at the cost of increased latency during concurrent \fBupsert\f1\f1 and \fBdelete\f1\f1 operations. .PP For example, to set the maximum percentage to 20, you can issue the following during startup: .PP .EX mongod \-\-setParameter maxCatchUpPercentageBeforeBlockingWrites=20 .EE .PP You cannot change \fBmaxCatchUpPercentageBeforeBlockingWrites\f1\f1 during runtime. .PP Live Migration Protocol (https://github.com/mongodb/mongo/blob/master/src/mongo/db/s/README.md#the\-live\-migration\-protocol) .RE .PP \fBmetadataRefreshInTransactionMaxWaitBehindCritSecMS\f1 .RS .PP \fIType\f1: integer .PP \fIDefault\f1: 500 .PP Available for \fBmongod\f1\f1 only. .PP Limits the time a shard waits for a critical section within a transaction. .PP When a query accesses a shard, a \fBchunk migration\f1 or \fBDDL operation\f1 may already hold the critical section for the collection. If the query finds the critical section is taken, the shard waits until the critical section has been released. When the shard returns control to \fBmongos\f1\f1, \fBmongos\f1\f1 retries the query. However, if a multi\-shard transaction interacts with an operation that takes the critical section on multiple shards, the interaction can result in a distributed deadlock. .PP \fBmetadataRefreshInTransactionMaxWaitBehindCritSecMS\f1\f1 limits the maximum time a shard waits within a transaction for the critical section to be released. .PP To reduce the maximum wait time for the critical section within a transaction, lower the the value of \fBmetadataRefreshInTransactionMaxWaitBehindCritSecMS\f1\f1\&. .PP If \fBmetadataRefreshInTransactionMaxWaitBehindCritSecMS\f1\f1 is too low, \fBmongos\f1\f1 could use all of its retry attempts and return an error. .PP You can set \fBmetadataRefreshInTransactionMaxWaitBehindCritSecMS\f1\f1 at startup and during runtime. .PP For example, to set \fBmetadataRefreshInTransactionMaxWaitBehindCritSecMS\f1\f1 to 400 milliseconds: .PP .EX db.adminCommand( { setParameter: 1, metadataRefreshInTransactionMaxWaitBehindCritSecMS: 400 } ) .EE .RE .PP \fBreadHedgingMode\f1 .RS .PP \fIType\f1: string .PP \fIDefault\f1: on .PP Available for \fBmongos\f1\f1 only. .PP Specifies whether \fBmongos\f1\f1 supports hedged reads for those read operations whose \fBread preference\f1 have enabled the hedged read option. .PP Available values are: .RS .IP \(bu 2 .RS .IP \(bu 4 Value .IP \(bu 4 Description .RE .IP \(bu 2 .RS .IP \(bu 4 \fBon\f1 .IP \(bu 4 The \fBmongos\f1\f1 instance supports hedged reads for read operations whose \fBread preference\f1 have enabled the hedged read option. .RE .IP \(bu 2 .RS .IP \(bu 4 \fBoff\f1 .IP \(bu 4 The \fBmongos\f1\f1 instance does not support hedged reads. That is, hedged reads are unavailable, even for read operations whose read preference have enabled the hedged read option. .RE .RE .PP For example, to turn off hedged read support for a \fBmongos\f1\f1 instance, you can issue the following during startup: .PP .EX mongos \-\-setParameter readHedgingMode=off .EE .PP Or if using the \fBsetParameter\f1\f1 command in a \fBmongosh\f1\f1 session that is connected to a running \fBmongos\f1\f1: .PP .EX db.adminCommand( { setParameter: 1, readHedgingMode: "off" } ) .EE .RS .IP \(bu 2 \fBHedged Reads\f1 .IP \(bu 2 \fBmaxTimeMSForHedgedReads\f1\f1 .RE .RE .PP \fBshutdownTimeoutMillisForSignaledShutdown\f1 .RS .PP \fIType\f1: integer .PP \fIDefault\f1: 15000 .PP Available for \fBmongod\f1\f1 only. .PP Specifies the time (in milliseconds) to wait for any ongoing database operations to complete before initiating a shutdown of \fBmongod\f1\f1 in response to a \fBSIGTERM\f1 signal. .PP For example, to set the time to 250 milliseconds, you can issue the following during startup: .PP .EX mongod \-\-setParameter shutdownTimeoutMillisForSignaledShutdown=250 .EE .PP Or if using the \fBsetParameter\f1\f1 command in a \fBmongosh\f1\f1 session that is connected to a running \fBmongod\f1\f1: .PP .EX db.adminCommand( { setParameter: 1, shutdownTimeoutMillisForSignaledShutdown: 250 } ) .EE .RE .PP \fBmongosShutdownTimeoutMillisForSignaledShutdown\f1 .RS .PP \fIType\f1: integer .PP \fIDefault\f1: 15000 .PP Available for \fBmongos\f1\f1 only. .PP Specifies the time (in milliseconds) to wait for any ongoing database operations to complete before initiating a shutdown of \fBmongos\f1\f1 in response to a \fBSIGTERM\f1 signal. .PP For example, to set the time to 250 milliseconds, you can issue the following during startup: .PP .EX mongos \-\-setParameter mongosShutdownTimeoutMillisForSignaledShutdown=250 .EE .PP Or if using the \fBsetParameter\f1\f1 command in a \fBmongosh\f1\f1 session that is connected to a running \fBmongos\f1\f1: .PP .EX db.adminCommand( { setParameter: 1, mongosShutdownTimeoutMillisForSignaledShutdown: 250 } ) .EE .RE .PP \fBreplMonitorMaxFailedChecks\f1 .RS .PP \fIAvailable in MongoDB 3.2 only\f1 .PP Type: integer .PP Default: 30 .PP The number of times the \fBmongod\f1\f1 or \fBmongos\f1\f1 instance tries to reach the replica sets in the sharded cluster (for example, shard replica sets, config server replica set) to monitor the replica set status and topology. .PP When the number of consecutive unsuccessful attempts exceeds this parameter value, the \fBmongod\f1\f1 or \fBmongos\f1\f1 instance denotes the monitored replica set as unavailable. If the monitored replica set is the config server replica set: .RS .IP \(bu 2 For MongoDB 3.2.0\-3.2.9, the monitoring \fBmongod\f1\f1 or \fBmongos\f1\f1 instance will become unusable and needs to be restarted. See the v3.2 troubleshooting guide (https://www.mongodb.com/docs/v3.2/tutorial/troubleshoot\-sharded\-clusters/#a\-config\-server\-replica\-set\-member\-become\-unavailable) for more details. .IP \(bu 2 For MongoDB 3.2.10 and later 3.2\-series, see also \fBtimeOutMonitoringReplicaSets\f1\f1\&. .RE .RE .PP \fBtimeOutMonitoringReplicaSets\f1 .RS .PP \fIAvailable in MongoDB 3.2.10 and later 3.2\-series only\f1 .PP Type: integer .PP Default: false .PP The flag that determines whether the \fBmongod\f1\f1 or \fBmongos\f1\f1 instance should stop its attempt to reach the monitored replica set after unsuccessfully trying \fBreplMonitorMaxFailedChecks\f1\f1 number of times. .PP If the monitored replica set is the config server replica set and \fBtimeOutMonitoringReplicaSets\f1\f1 is set to \fBtrue\f1, you must restart \fBmongod\f1\f1 or \fBmongos\f1\f1 if the \fBmongod\f1\f1 or \fBmongos\f1\f1 instance cannot reach any of the config servers for the specified number of times. See the v3.2 troubleshooting guide (https://www.mongodb.com/docs/v3.2/tutorial/troubleshoot\-sharded\-clusters/#a\-config\-server\-replica\-set\-member\-become\-unavailable) for more details. .RE .PP \fBShardingTaskExecutorPoolHostTimeoutMS\f1 .RS .PP Type: integer .PP Default: 300000 (5 minutes) .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Maximum time that \fBmongos\f1\f1 goes without communication to a host before \fBmongos\f1\f1 drops all connections to the host. .PP If set, \fBShardingTaskExecutorPoolHostTimeoutMS\f1\f1 should be greater than the sum of \fBShardingTaskExecutorPoolRefreshRequirementMS\f1\f1 and \fBShardingTaskExecutorPoolRefreshTimeoutMS\f1\f1\&. Otherwise, \fBmongos\f1\f1 adjusts the value of \fBShardingTaskExecutorPoolHostTimeoutMS\f1\f1 to be greater than the sum. .PP The following example sets \fBShardingTaskExecutorPoolHostTimeoutMS\f1\f1 to \fB120000\f1 during startup: .PP .EX mongos \-\-setParameter ShardingTaskExecutorPoolHostTimeoutMS=120000 .EE .PP During runtime, you can also set the parameter with the \fBsetParameter\f1\f1 command: .PP .EX db.adminCommand( { setParameter: 1, ShardingTaskExecutorPoolHostTimeoutMS: 120000 } ) .EE .RE .PP \fBShardingTaskExecutorPoolMaxConnecting\f1 .RS .PP Type: integer .PP Default: 2 .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Maximum number of simultaneous initiating connections (including pending connections in setup/refresh state) each TaskExecutor connection pool can have to a \fBmongod\f1\f1 instance. You can set this parameter to control the rate at which \fBmongos\f1\f1 adds connections to a \fBmongod\f1\f1 instance. .PP If set, \fBShardingTaskExecutorPoolMaxConnecting\f1\f1 should be less than or equal to \fBShardingTaskExecutorPoolMaxSize\f1\f1\&. If it is greater, \fBmongos\f1\f1 ignores the \fBShardingTaskExecutorPoolMaxConnecting\f1\f1 value. .PP The following example sets \fBShardingTaskExecutorPoolMaxConnecting\f1\f1 to \fB20\f1 during startup: .PP .EX mongos \-\-setParameter ShardingTaskExecutorPoolMaxConnecting=20 .EE .PP During runtime, you can also set the parameter with the \fBsetParameter\f1\f1 command: .PP .EX db.adminCommand( { setParameter: 1, ShardingTaskExecutorPoolMaxConnecting: 20 } ) .EE .RE .PP \fBShardingTaskExecutorPoolMaxSize\f1 .RS .PP Type: integer .PP Default: 2 64 \- 1 .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Maximum number of outbound connections each TaskExecutor connection pool can open to any given \fBmongod\f1\f1 instance. The maximum possible connections to any given host across all TaskExecutor pools is: .PP .EX ShardingTaskExecutorPoolMaxSize * taskExecutorPoolSize .EE .PP The following example sets \fBShardingTaskExecutorPoolMaxSize\f1\f1 to \fB20\f1 during startup: .PP .EX mongos \-\-setParameter ShardingTaskExecutorPoolMaxSize=20 .EE .PP During runtime, you can also set the parameter with the \fBsetParameter\f1\f1 command: .PP .EX db.adminCommand( { setParameter: 1, ShardingTaskExecutorPoolMaxSize: 20 } ) .EE .PP \fBmongos\f1\f1 can have up to \fBn\f1 TaskExecutor connection pools, where \fBn\f1 is the number of cores. See \fBtaskExecutorPoolSize\f1\f1\&. .PP \fBShardingTaskExecutorPoolMinSize\f1\f1 .RE .PP \fBShardingTaskExecutorPoolMaxSizeForConfigServers\f1 .RS .PP Type: integer .PP Default: \-1 .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Optional override for \fBShardingTaskExecutorPoolMaxSize\f1\f1 to set the maximum number of outbound connections each TaskExecutor connection pool can open to a \fBconfiguration server\f1\&. .PP When set to: .RS .IP \(bu 2 \fB\-1\f1, \fBShardingTaskExecutorPoolMaxSize\f1\f1 is used. This is the default. .IP \(bu 2 an integer value greater than \fB\-1\f1, overrides the maximum number of outbound connections each TaskExecutor connection pool can open to a configuration server. .RE .PP Parameter only applies to sharded deployments. .PP The following example sets \fBShardingTaskExecutorPoolMaxSize\f1\f1 to \fB2\f1 during startup, which sets the maximum number of outbound connections each TaskExecutor connection pool can open to a configuration server to \fB2\f1: .PP .EX mongos \-\-setParameter ShardingTaskExecutorPoolMaxSizeForConfigServers=2 .EE .PP During runtime, you can also set the parameter with the \fBsetParameter\f1\f1 command: .PP .EX db.adminCommand( { setParameter: 1, ShardingTaskExecutorPoolMaxSizeForConfigServers: 2 } ) .EE .RE .PP \fBShardingTaskExecutorPoolMinSize\f1 .RS .PP Type: integer .PP Default: 1 .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Minimum number of outbound connections each TaskExecutor connection pool can open to any given \fBmongod\f1\f1 instance. .PP \fBShardingTaskExecutorPoolMinSize\f1 connections are created the first time a connection to a new host is requested from the pool. While the pool is idle, the pool maintains this number of connections until \fBShardingTaskExecutorPoolHostTimeoutMS\f1\f1 milliseconds pass without any application using that pool. .PP For a \fBmongos\f1\f1 using the \fBwarmMinConnectionsInShardingTaskExecutorPoolOnStartup\f1\f1 parameter, the \fBShardingTaskExecutorPoolMinSize\f1 parameter also controls how many connections to each shard host are established on startup of the \fBmongos\f1\f1 instance before it begins accepting incoming client connections. .PP In MongoDB 4.4, the \fBwarmMinConnectionsInShardingTaskExecutorPoolOnStartup\f1\f1 parameter is enabled by default for the \fBmongos\f1\f1\&. .PP The following example sets \fBShardingTaskExecutorPoolMinSize\f1\f1 to \fB2\f1 during startup: .PP .EX mongos \-\-setParameter ShardingTaskExecutorPoolMinSize=2 .EE .PP During runtime, you can also set the parameter with the \fBsetParameter\f1\f1 command: .PP .EX db.adminCommand( { setParameter: 1, ShardingTaskExecutorPoolMinSize: 2 } ) .EE .PP \fBmongos\f1\f1 can have up to \fBn\f1 TaskExecutor connection pools, where \fBn\f1 is the number of cores. See \fBtaskExecutorPoolSize\f1\f1\&. .RS .IP \(bu 2 \fBShardingTaskExecutorPoolMaxSize\f1\f1 .IP \(bu 2 \fBwarmMinConnectionsInShardingTaskExecutorPoolOnStartup\f1\f1 .RE .RE .PP \fBShardingTaskExecutorPoolMinSizeForConfigServers\f1 .RS .PP Type: integer .PP Default: \-1 .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Optional override for \fBShardingTaskExecutorPoolMinSize\f1\f1 to set the minimum number of outbound connections each TaskExecutor connection pool can open to a \fBconfiguration server\f1\&. .PP When set to: .RS .IP \(bu 2 \fB\-1\f1, \fBShardingTaskExecutorPoolMinSize\f1\f1 is used. This is the default. .IP \(bu 2 an integer value greater than \fB\-1\f1, overrides the minimum number of outbound connections each TaskExecutor connection pool can open to a configuration server. .RE .PP Parameter only applies to sharded deployments. .PP The following example sets \fBShardingTaskExecutorPoolMinSize\f1\f1 to \fB2\f1 during startup, which sets the minimum number of outbound connections each TaskExecutor connection pool can open to a configuration server to \fB2\f1: .PP .EX mongos \-\-setParameter ShardingTaskExecutorPoolMinSizeForConfigServers=2 .EE .PP During runtime, you can also set the parameter with the \fBsetParameter\f1\f1 command: .PP .EX db.adminCommand( { setParameter: 1, ShardingTaskExecutorPoolMinSizeForConfigServers: 2 } ) .EE .RE .PP \fBShardingTaskExecutorPoolRefreshRequirementMS\f1 .RS .PP Type: integer .PP Default: 60000 (1 minute) .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Maximum time the \fBmongos\f1\f1 waits before attempting to heartbeat a resting connection in the pool. An idle connection may be discarded during the refresh if the pool is above its \fBminimum size\f1\&. .PP If set, \fBShardingTaskExecutorPoolRefreshRequirementMS\f1\f1 should be greater than \fBShardingTaskExecutorPoolRefreshTimeoutMS\f1\f1\&. Otherwise, \fBmongos\f1\f1 adjusts the value of \fBShardingTaskExecutorPoolRefreshTimeoutMS\f1\f1 to be less than \fBShardingTaskExecutorPoolRefreshRequirementMS\f1\f1\&. .PP The following example sets \fBShardingTaskExecutorPoolRefreshRequirementMS\f1\f1 to \fB90000\f1 during startup: .PP .EX mongos \-\-setParameter ShardingTaskExecutorPoolRefreshRequirementMS=90000 .EE .PP During runtime, you can also set the parameter with the \fBsetParameter\f1\f1 command: .PP .EX db.adminCommand( { setParameter: 1, ShardingTaskExecutorPoolRefreshRequirementMS: 90000 } ) .EE .RE .PP \fBShardingTaskExecutorPoolRefreshTimeoutMS\f1 .RS .PP Type: integer .PP Default: 20000 (20 seconds) .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Maximum time the \fBmongos\f1\f1 waits for a heartbeat before timing out the heartbeat. .PP If set, \fBShardingTaskExecutorPoolRefreshTimeoutMS\f1\f1 should be less than \fBShardingTaskExecutorPoolRefreshRequirementMS\f1\f1\&. Otherwise, \fBmongos\f1\f1 adjusts the value of \fBShardingTaskExecutorPoolRefreshTimeoutMS\f1\f1 to be less than \fBShardingTaskExecutorPoolRefreshRequirementMS\f1\f1\&. .PP The following example sets \fBShardingTaskExecutorPoolRefreshTimeoutMS\f1\f1 to \fB30000\f1 during startup: .PP .EX mongos \-\-setParameter ShardingTaskExecutorPoolRefreshTimeoutMS=30000 .EE .PP During runtime, you can also set the parameter with the \fBsetParameter\f1\f1 command: .PP .EX db.adminCommand( { setParameter: 1, ShardingTaskExecutorPoolRefreshTimeoutMS: 30000 } ) .EE .RE .PP \fBShardingTaskExecutorPoolReplicaSetMatching\f1 .RS .PP Type: string .PP Default: "automatic" .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP On a \fBmongos\f1\f1 instance, this parameter sets the policy that determines the minimum size limit of its connection pools to nodes within replica sets. .PP On a \fBmongod\f1\f1 instance, this parameter sets the policy that determines the minimum size limit of its connection pools to nodes within \fIother\f1 replica sets. .PP Note that this parameter only manages connections for operations that are directly related to user requests and CRUD operations. .PP Available values are: .RS .IP \(bu 2 .RS .IP \(bu 4 Matching Policy .IP \(bu 4 Description .RE .IP \(bu 2 .RS .IP \(bu 4 \fB"automatic"\f1 (Default) .IP \(bu 4 Starting in 5.0 (and 4.4.5 and 4.2.13), \fB"automatic"\f1 is the new default value. .IP When set for a \fBmongos\f1\f1, the instance follows the behavior specified for the \fB"matchPrimaryNode"\f1 option. .IP When set for a \fBmongod\f1\f1, the instance follows the behavior specified for the \fB"disabled"\f1 option. .IP If the \fBShardingTaskExecutorPoolReplicaSetMatching\f1\f1 is set to \fB"automatic"\f1, the \fBreplicaSetMatchingStrategy\f1\f1 still describes the actual policy being used, not \fB"automatic"\f1\&. To find the value of the \fBShardingTaskExecutorPoolReplicaSetMatching\f1\f1, use \fBgetParameter\f1\f1 which returns the value of the server parameter. .RE .IP \(bu 2 .RS .IP \(bu 4 \fB"matchPrimaryNode"\f1 .IP \(bu 4 When set for a \fBmongos\f1\f1, the minimum size limit of the instance\(aqs connection pool to each secondary of a replica set in the sharded cluster (specifically, shard replica set and config servers) is equal to the size of its connection pool to that replica set\(aqs primary. .IP When set for a \fBmongod\f1\f1, the minimum size limit of the instance\(aqs connection pool to each secondary of another replica set in the sharded cluster (specifically, shard replica set and config servers) is equal to the size of its connection pool to that replica set\(aqs primary. .IP If multiple shard servers in your topology can experience a rapid influx of cross\-shard operations, do not set this option on your \fBmongod\f1\f1 instances. .IP In case of a primary stepdown, \fBmatchPrimaryNode\f1 ensures that any secondary that becomes the primary can handle the current level of primary reads and writes. .RE .IP \(bu 2 .RS .IP \(bu 4 \fB"matchBusiestNode"\f1 .IP \(bu 4 When set for a \fBmongos\f1\f1, the instance\(aqs minimum size limit of the connection pool to each member of a replica set in the sharded cluster (specifically, shard replica set and config servers) is equal to the largest among the active connection counts to the primary and each secondary member of that replica set. .IP When set for a \fBmongod\f1\f1, the instance\(aqs minimum size limit of the connection pool to each member of another replica set in the sharded cluster (specifically, shard replica set and config servers) is equal to the largest among the active connection counts to the primary and each secondary member of that replica set. .IP With \fB"matchBusiestNode"\f1, \fBmongos\f1\f1 maintains enough connections to each secondary to handle the current level of primary and secondary reads and writes. The number of connections to maintain in the pool decreases as the number of active connections decreases. .RE .IP \(bu 2 .RS .IP \(bu 4 \fB"disabled"\f1 .IP \(bu 4 When set for a \fBmongos\f1\f1, the instance\(aqs minimum number of connections in the instance\(aqs connection pool to each node of a replica set in the sharded clusterv (specifically, shard replica set and config servers) is equal to the \fBShardingTaskExecutorPoolMinSize\f1\f1\&. .IP When set for a \fBmongod\f1\f1, the instance\(aqs minimum number of connections in the instance\(aqs connection pool to each node of another replica set in the sharded cluster (specifically, shard replica set and config servers) is equal to the \fBShardingTaskExecutorPoolMinSize\f1\f1\&. .RE .RE .PP The following example sets the \fBShardingTaskExecutorPoolReplicaSetMatching\f1\f1 to \fB"automatic"\f1 during startup: .PP .EX mongod \-\-setParameter ShardingTaskExecutorPoolReplicaSetMatching="automatic" .EE .PP During runtime, you can also set the parameter with the \fBsetParameter\f1\f1 command: .PP .EX db.adminCommand( { setParameter: 1, ShardingTaskExecutorPoolReplicaSetMatching: "automatic" } ) .EE .RE .PP \fBtaskExecutorPoolSize\f1 .RS .PP Type: integer .PP Default: 1 .PP Available for \fBmongos\f1\f1 only. .PP The number of Task Executor connection pools to use for a given \fBmongos\f1\f1\&. .PP If the parameter value is \fB0\f1 or less, the number of Task Executor connection pools is the number of cores with the following exceptions: .RS .IP \(bu 2 If the number of cores is less than 4, the number of Task Executor connection pools is 4. .IP \(bu 2 If the number of cores is greater than 64, the number of Task Executor connection pools is 64. .RE .PP Starting in MongoDB 4.0, the default value of \fBtaskExecutorPoolSize\f1\f1 is \fB1\f1: .RS .IP \(bu 2 In MongoDB 4.0 deployment, you can set \fBtaskExecutorPoolSize\f1\f1 to \fB0\f1 and, on Linux, set AsyncRequestsSenderUseBaton (https://www.mongodb.com/docs/v4.0/reference/parameters/#param.AsyncRequestsSenderUseBaton) to \fBfalse\f1 for the previous behavior. .IP \(bu 2 In MongoDB 4.2+ deployment, MongoDB removes the \fBAsyncRequestsSenderUseBaton\f1 parameter and always enables the performance enhancement controlled by the parameter. .RE .PP You can only set this parameter during start\-up and cannot change this setting using the \fBsetParameter\f1\f1 database command. .PP .EX mongos \-\-setParameter taskExecutorPoolSize=6 .EE .RS .IP \(bu 2 \fBShardingTaskExecutorPoolMaxSize\f1\f1 .IP \(bu 2 \fBShardingTaskExecutorPoolMinSize\f1\f1 .RE .RE .PP \fBloadRoutingTableOnStartup\f1 .RS .PP Type: boolean .PP \fIDefault\f1: true .PP Available for \fBmongos\f1\f1 only. .PP Configures a \fBmongos\f1\f1 instance to preload the routing table for a sharded cluster on startup. With this setting enabled, the \fBmongos\f1\f1 caches the cluster\-wide routing table for each sharded collection as part of its startup procedure, before it begins accepting client connections. .PP Without this setting enabled, the \fBmongos\f1\f1 only loads a routing table as needed for incoming client connections, and only loads the specific routing table for the namespace of a given request. .PP A \fBmongos\f1\f1 instance with the \fBloadRoutingTableOnStartup\f1\f1 parameter enabled may experience longer startup times, but will result in faster servicing of initial client connections once started. .PP \fBloadRoutingTableOnStartup\f1\f1 is enabled by default. .PP You can only set this parameter on startup, using either the \fBsetParameter\f1\f1 configuration file setting or the \fB\-\-setParameter\f1\f1 command line option. .RE .PP \fBwarmMinConnectionsInShardingTaskExecutorPoolOnStartup\f1 .RS .PP Type: boolean .PP \fIDefault\f1: true .PP Available for \fBmongos\f1\f1 only. .PP Configures a \fBmongos\f1\f1 instance to prewarm its connection pool on startup. With this parameter enabled, the \fBmongos\f1\f1 attempts to establish \fBShardingTaskExecutorPoolMinSize\f1\f1 network connections to each shard server as part of its startup procedure, before it begins accepting client connections. .PP A timeout for this behavior can be configured with the \fBwarmMinConnectionsInShardingTaskExecutorPoolOnStartupWaitMS\f1\f1 parameter. If this timeout is reached, the \fBmongos\f1\f1 will begin accepting client connections regardless of the size of its connection pool. .PP A \fBmongos\f1\f1 instance with this parameter enabled may experience longer startup times, but will result in faster servicing of initial client connections once started. .PP \fBwarmMinConnectionsInShardingTaskExecutorPoolOnStartup\f1\f1 is enabled by default. .PP You can only set this parameter on startup, using either the \fBsetParameter\f1\f1 configuration file setting or the \fB\-\-setParameter\f1\f1 command line option. .RS .IP \(bu 2 \fBwarmMinConnectionsInShardingTaskExecutorPoolOnStartupWaitMS\f1\f1 .IP \(bu 2 \fBShardingTaskExecutorPoolMinSize\f1\f1 .RE .RE .PP \fBwarmMinConnectionsInShardingTaskExecutorPoolOnStartupWaitMS\f1 .RS .PP Type: integer .PP \fIDefault\f1: 2000 (2 seconds) .PP Available for \fBmongos\f1\f1 only. .PP Sets the timeout threshold in milliseconds for a \fBmongos\f1\f1 to wait for \fBShardingTaskExecutorPoolMinSize\f1\f1 connections to be established per shard host when using the \fBwarmMinConnectionsInShardingTaskExecutorPoolOnStartup\f1\f1 parameter. If this timeout is reached, the \fBmongos\f1\f1 will begin accepting client connections regardless of the size of its connection pool. .PP You can only set this parameter on startup, using either the \fBsetParameter\f1\f1 configuration file setting or the \fB\-\-setParameter\f1\f1 command line option. .RS .IP \(bu 2 \fBwarmMinConnectionsInShardingTaskExecutorPoolOnStartup\f1\f1 .IP \(bu 2 \fBShardingTaskExecutorPoolMinSize\f1\f1 .RE .RE .PP \fBmigrateCloneInsertionBatchDelayMS\f1 .RS .PP The parameter is also available starting in 3.4.18 and 3.6.10 .PP Available for \fBmongod\f1\f1 only. .PP Type: Non\-negative integer .PP Default: 0 .PP Time in milliseconds to wait between batches of insertions during cloning step of the migration process. This wait is in addition to the \fBsecondaryThrottle\f1\&. .PP The default value of \fB0\f1 indicates no additional wait. .PP The following sets the \fBmigrateCloneInsertionBatchDelayMS\f1\f1 to 200 milliseconds: .PP .EX mongod \-\-setParameter migrateCloneInsertionBatchDelayMS=200 .EE .PP The parameter may also be set using the \fBsetParameter\f1\f1 command: .PP .EX db.adminCommand( { setParameter: 1, migrateCloneInsertionBatchDelayMS: 200 } ) .EE .RE .PP \fBmigrateCloneInsertionBatchSize\f1 .RS .PP The parameter is also available starting in 3.4.18 and 3.6.10 .PP Available for \fBmongod\f1\f1 only. .PP Type: Non\-negative integer .PP Default: 0 .PP The maximum number of documents to insert in a single batch during the cloning step of the migration process. .PP The default value of \fB0\f1 indicates no maximum number of documents per batch. However, in practice, this results in batches that contain up to 16 MB of documents. .PP The following sets the \fBmigrateCloneInsertionBatchSize\f1\f1 to 100 documents: .PP .EX mongod \-\-setParameter migrateCloneInsertionBatchSize=100 .EE .PP The parameter may also be set using the \fBsetParameter\f1\f1 command: .PP .EX db.adminCommand( { setParameter: 1, migrateCloneInsertionBatchSize: 100 } ) .EE .RE .PP \fBorphanCleanupDelaySecs\f1 .RS .PP Default: 900 (15 minutes) .PP Available for \fBmongod\f1\f1 only. .PP Minimum delay before a migrated \fBchunk\f1 is deleted from the source shard. .PP Before deleting the chunk during chunk migration, MongoDB waits for \fBorphanCleanupDelaySecs\f1\f1 or for in\-progress queries involving the chunk to complete on the shard primary, whichever is longer. .PP However, because the shard primary has no knowledge of in\-progress queries run on the shard secondaries, queries that use the chunk but are run on secondaries may see documents disappear if these queries take longer than the time to complete the shard primary queries and the \fBorphanCleanupDelaySecs\f1\f1\&. .PP This behavior only affects in\-progress queries that start before the chunk migration. Queries that start after the chunk migration starts will not use the migrating chunk. .PP If a shard has storage constraints, consider reducing this value temporarily. If running queries that exceed 15 minutes on shard secondaries, consider increasing this value. .PP The following sets the \fBorphanCleanupDelaySecs\f1\f1 to 20 minutes: .PP .EX mongod \-\-setParameter orphanCleanupDelaySecs=1200 .EE .PP This may also be set using the \fBsetParameter\f1\f1 command: .PP .EX db.adminCommand( { setParameter: 1, orphanCleanupDelaySecs: 1200 } ) .EE .RE .PP \fBrangeDeleterBatchDelayMS\f1 .RS .PP The parameter is also available starting in 3.4.17 and 3.6.7. .PP Available for \fBmongod\f1\f1 only. .PP Type: Non\-negative integer .PP Default: 20 .PP The amount of time in milliseconds to wait before the next batch of deletion during the cleanup stage of \fBchunk migration\f1 (or the \fBcleanupOrphaned\f1\f1 command). .PP In MongoDB 3.4, consider whether \fB_secondaryThrottle\f1 is set before modifying the \fBrangeDeleterBatchDelayMS\f1\f1\&. In MongoDB 3.4, the \fB_secondaryThrottle replication delay\f1 occurs after each document deletion instead of after the batch deletion. .PP In MongoDB 3.6+, the \fB_secondaryThrottle replication delay\f1 occurs after each batch deletion. .PP The following sets the \fBrangeDeleterBatchDelayMS\f1\f1 to 200 milliseconds: .PP .EX mongod \-\-setParameter rangeDeleterBatchDelayMS=200 .EE .PP The parameter may also be set using the \fBsetParameter\f1\f1 command: .PP .EX db.adminCommand( { setParameter: 1, rangeDeleterBatchDelayMS: 200 } ) .EE .RE .PP \fBrangeDeleterBatchSize\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP Type: Non\-negative integer .PP Default: 2147483647 starting in MongoDB 5.1.2, 5.0.6, and 4.4.12 (128 in earlier MongoDB versions) .PP The maximum number of documents in each batch to delete during the cleanup stage of \fBchunk migration\f1 (or the \fBcleanupOrphaned\f1\f1 command). .PP A value of \fB0\f1 indicates that the system chooses the default value. .PP The following example sets \fBrangeDeleterBatchSize\f1\f1 to 32 documents: .PP .EX mongod \-\-setParameter rangeDeleterBatchSize=32 .EE .PP The parameter may also be set using the \fBsetParameter\f1\f1 command: .PP .EX db.adminCommand( { setParameter: 1, rangeDeleterBatchSize: 32 } ) .EE .RE .PP \fBskipShardingConfigurationChecks\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP Type: boolean .PP Default: false .PP When \fBtrue\f1, allows for starting a shard member or config server member as a standalone for maintenance operations. This parameter is mutually exclusive with the \fB\-\-configsvr\f1\f1 or \fB\-\-shardsvr\f1\f1 options. .PP You can only set this parameter during start\-up and cannot change this setting using the \fBsetParameter\f1\f1 database command. .PP .EX mongod \-\-setParameter skipShardingConfigurationChecks=true .EE .PP Once maintenance has completed, remove the \fBskipShardingConfigurationChecks\f1\f1 parameter when restarting the \fBmongod\f1\f1\&. .PP The parameter is also available for MongoDB versions: .RS .IP \(bu 2 MongoDB 3.2.19+ .IP \(bu 2 MongoDB 3.4.11+ .RE .RE .PP \fBfindChunksOnConfigTimeoutMS\f1 .RS .PP Available for both \fBmongod\f1\f1 and \fBmongos\f1\f1\&. .PP Type: Non\-negative integer .PP Default: 900000 .PP The timeout in milliseconds for find operations on \fBchunks\f1\f1\&. .PP If there is a large number of chunks in the cluster and chunk loading fails with the error \fBExceededTimeLimit\f1, increase the parameter value: .PP .EX mongod \-\-setParameter findChunksOnConfigTimeoutMS=1000000 .EE .RE .SS HEALTH MANAGER PARAMETERS .PP \fBactiveFaultDurationSecs\f1 .RS .PP \fIType\f1: Document .PP Available for \fBmongos\f1\f1 only. .PP The amount of time to wait from a \fB\f1 failure until the \fBmongos\f1 is removed from the cluster, in seconds. .PP When a failure is detected and a Health Manager is configured as \fBcritical\f1, the server waits for the specified interval before removing the \fBmongos\f1 from the cluster. .PP For example, to set the duration from failure to crash to five minutes, issue the following at startup: .PP .EX mongos \-\-setParameter activeFaultDurationSecs=300 .EE .PP Or if using the \fBsetParameter\f1\f1 command in a \fBmongosh\f1\f1 session that is connected to a running \fBmongos\f1\f1: .PP .EX db.adminCommand( { setParameter: 1, activeFaultDurationSecs: 300 } ) .EE .PP Parameters set with \fBsetParameter\f1\f1 do not persist across restarts. See the \fBsetParameter page\f1 for details. .PP To make this setting persistent, set \fBactiveFaultDurationSecs\f1 in your \fBmongos config file\f1 using the \fBsetParameter\f1\f1 option as in the following example: .PP .EX setParameter: activeFaultDurationSecs: 300 .EE .RE .PP \fBhealthMonitoringIntensities\f1 .RS .PP \fIType\f1: Array of documents .PP Available for \fBmongos\f1\f1 only. .PP Use this parameter to set intensity levels for Health Managers\&. .PP \fBhealthMonitoringIntensities\f1 accepts an array of documents, \fBvalues\f1\&. Each document in \fBvalues\f1 takes two fields: .RS .IP \(bu 2 \fBtype\f1, the Health Manager facet .IP \(bu 2 \fBintensity\f1, the intensity level .RE .SS HEALTH MANAGERS .RS .IP \(bu 2 .RS .IP \(bu 4 Facet .IP \(bu 4 What the Health Observer Checks .RE .IP \(bu 2 .RS .IP \(bu 4 \fBconfigServer\f1 .IP \(bu 4 Cluster health issues related to connectivity to the config server. .RE .IP \(bu 2 .RS .IP \(bu 4 \fBdns\f1 .IP \(bu 4 Cluster health issues related to DNS availability and functionality. .RE .IP \(bu 2 .RS .IP \(bu 4 \fBldap\f1 .IP \(bu 4 Cluster health issues related to LDAP availability and functionality. .RE .RE .SS INTENSITY LEVELS .RS .IP \(bu 2 .RS .IP \(bu 4 Intensity Level .IP \(bu 4 Description .RE .IP \(bu 2 .RS .IP \(bu 4 \fBcritical\f1 .IP \(bu 4 The Health Manager on this facet is enabled and has the ability to move the failing \fBmongos\f1 out of the cluster if an error occurs. The Health Manager waits the amount of time specified by \fBactiveFaultDurationSecs\f1\f1 before stopping and moving the \fBmongos\f1 out of the cluster automatically. .RE .IP \(bu 2 .RS .IP \(bu 4 \fBnon\-critical\f1 .IP \(bu 4 The Health Manager on this facet is enabled and logs errors, but the \fBmongos\f1 remains in the cluster if errors are encountered. .RE .IP \(bu 2 .RS .IP \(bu 4 \fBoff\f1 .IP \(bu 4 The Health Manager on this facet is disabled. The \fBmongos\f1 does not perform any health checks on this facet. This is the default intensity level. .RE .RE .PP For example, to set the \fBdns\f1 Health Manager facet to the \fBcritical\f1 intensity level, issue the following at startup: .PP .EX mongos \-\-setParameter \(aqhealthMonitoringIntensities={ values:[ { type:"dns", intensity: "critical"} ] }\(aq .EE .PP Or if using the \fBsetParameter\f1\f1 command in a \fBmongosh\f1\f1 session that is connected to a running \fBmongos\f1\f1: .PP .EX db.adminCommand( { setParameter: 1, healthMonitoringIntensities: { values: [ { type: "dns", intensity: "critical" } ] } } ) } ) .EE .PP Parameters set with \fBsetParameter\f1\f1 do not persist across restarts. See the \fBsetParameter page\f1 for details. .PP To make this setting persistent, set \fBhealthMonitoringIntensities\f1 in your \fBmongos config file\f1 using the \fBsetParameter\f1\f1 option as in the following example: .PP .EX setParameter: healthMonitoringIntensities: "{ values:[ { type:\"dns\", intensity: \"critical\"} ] }" .EE .RE .PP \fBhealthMonitoringIntervals\f1 .RS .PP \fIType\f1: Array of documents .PP Available for \fBmongos\f1\f1 only. .PP How often this Health Manager will run, in milliseconds. .PP \fBhealthMonitoringIntervals\f1 accepts an array of documents, \fBvalues\f1\&. Each document in \fBvalues\f1 takes two fields: .RS .IP \(bu 2 \fBtype\f1, the Health Manager facet .IP \(bu 2 \fBinterval\f1, the time interval it runs at, in milliseconds .RE .SS HEALTH MANAGERS .RS .IP \(bu 2 .RS .IP \(bu 4 Facet .IP \(bu 4 What the Health Observer Checks .RE .IP \(bu 2 .RS .IP \(bu 4 \fBconfigServer\f1 .IP \(bu 4 Cluster health issues related to connectivity to the config server. .RE .IP \(bu 2 .RS .IP \(bu 4 \fBdns\f1 .IP \(bu 4 Cluster health issues related to DNS availability and functionality. .RE .IP \(bu 2 .RS .IP \(bu 4 \fBldap\f1 .IP \(bu 4 Cluster health issues related to LDAP availability and functionality. .RE .RE .PP For example, to set the \fBldap\f1 Health Manager facet to the run health checks every 30 seconds, issue the following at startup: .PP .EX mongos \-\-setParameter \(aqhealthMonitoringIntervals={ values:[ { type:"ldap", interval: "30000"} ] }\(aq .EE .PP Or if using the \fBsetParameter\f1\f1 command in a \fBmongosh\f1\f1 session that is connected to a running \fBmongos\f1\f1: .PP .EX db.adminCommand( { setParameter: 1, healthMonitoringIntervals: { values: [ { type: "ldap", interval: "30000" } ] } } ) } ) .EE .PP Parameters set with \fBsetParameter\f1\f1 do not persist across restarts. See the \fBsetParameter page\f1 for details. .PP To make this setting persistent, set \fBhealthMonitoringIntervals\f1 in your \fBmongos config file\f1 using the \fBsetParameter\f1\f1 option as in the following example: .PP .EX setParameter: healthMonitoringIntervals: "{ values: [{type: \"ldap\", interval: 200}] }" .EE .RE .PP \fBprogressMonitor\f1 .RS .PP \fIType\f1: Document .PP Available for \fBmongos\f1\f1 only. .PP \fBProgress Monitor\f1 runs tests to ensure that Health Manager checks do not become stuck or unresponsive. Progress Monitor runs these tests in intervals specified by \fBinterval\f1\&. If a health check begins but does not complete within the timeout given by \fBdeadline\f1, Progress Monitor stops the \fBmongos\f1 and removes it from the cluster. .SS PROGRESSMONITOR FIELDS .RS .IP \(bu 2 .RS .IP \(bu 4 Field .IP \(bu 4 Description .IP \(bu 4 Units .RE .IP \(bu 2 .RS .IP \(bu 4 \fBinterval\f1 .IP \(bu 4 How often to ensure Health Managers are not stuck or unresponsive. .IP \(bu 4 Milliseconds .RE .IP \(bu 2 .RS .IP \(bu 4 \fBdeadline\f1 .IP \(bu 4 Timeout before automatically failing the \fBmongos\f1 if a Health Manager check is not making progress. .IP \(bu 4 Seconds .RE .RE .PP To set the \fBinterval\f1 to 1000 milliseconds and the \fBdeadline\f1 to 300 seconds, issue the following at startup: .PP .EX mongos \-\-setParameter \(aqprogressMonitor={"interval": 1000, "deadline": 300}\(aq .EE .PP Or if using the \fBsetParameter\f1\f1 command in a \fBmongosh\f1\f1 session that is connected to a running \fBmongos\f1\f1: .PP .EX db.adminCommand( { setParameter: 1, progressMonitor: { interval: 1000, deadline: 300 } ) } ) .EE .PP Parameters set with \fBsetParameter\f1\f1 do not persist across restarts. See the \fBsetParameter page\f1 for details. .PP To make this setting persistent, set \fBprogressMonitor\f1 in your \fBmongos config file\f1 using the \fBsetParameter\f1\f1 option as in the following example: .PP .EX setParameter: progressMonitor: "{ interval: 1000, deadline: 300 }" .EE .RE .SS STORAGE PARAMETERS .PP \fBhonorSystemUmask\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP \fIDefault\f1: \fBfalse\f1 .PP If \fBhonorSystemUmask\f1\f1 is set to \fBtrue\f1, new files created by MongoDB have permissions in accordance with the user\(aqs \fBumask\f1 settings. You cannot set \fBprocessUmask\f1\f1 if \fBhonorSystemUmask\f1\f1 is set to \fBtrue\f1\&. .PP If \fBhonorSystemUmask\f1\f1 is set to \fBfalse\f1, new files created by MongoDB have permissions set to \fB600\f1, which gives read and write permissions only to the owner. New directories have permissions set to \fB700\f1\&. You can use \fBprocessUmask\f1\f1 to override the default permissions for groups and other users on all new files created by MongoDB. .PP You can only set this parameter during start\-up and cannot change this setting using the \fBsetParameter\f1\f1 database command. .PP .EX mongod \-\-setParameter honorSystemUmask=true .EE .PP \fBhonorSystemUmask\f1\f1 is not available on Windows systems. .RE .PP \fBjournalCommitInterval\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP Specify an integer between \fB1\f1 and \fB500\f1 signifying the number of milliseconds (ms) between journal commits. .PP Consider the following example which sets the \fBjournalCommitInterval\f1\f1 to \fB200\f1 ms: .PP .EX db.adminCommand( { setParameter: 1, journalCommitInterval: 200 } ) .EE .PP \fBstorage.journal.commitIntervalMs\f1\f1 .RE .PP \fBminSnapshotHistoryWindowInSeconds\f1 .RS .PP \fIDefault\f1: \fB300\f1 .PP Available for \fBmongod\f1\f1 only. .PP The minimum time window in seconds for which the storage engine keeps the snapshot history. If you query data using read concern \fB"snapshot"\f1\f1 and specify an \fBatClusterTime\f1 value older than the specified \fBminSnapshotHistoryWindowInSeconds\f1, \fBmongod\f1\f1 returns a \fBSnapshotTooOld\f1 error. .PP Specify an integer greater than or equal to (\fB>=\f1) 0. .PP Consider the following example which sets the \fBminSnapshotHistoryWindowInSeconds\f1 to \fB600\f1 seconds: .PP .EX db.adminCommand( { setParameter: 1, minSnapshotHistoryWindowInSeconds: 600 } ) .EE .PP Increasing the value of \fBminSnapshotHistoryWindowInSeconds\f1\f1 increases disk usage. For more information, see \fBSnapshot History Retention\f1\&. .PP To modify this value for a MongoDB Atlas (https://www.mongodb.com/docs/atlas/) cluster, you must contact Atlas Support (https://www.mongodb.com/docs/atlas/support/)\&. .RE .PP \fBprocessUmask\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP Overrides the default permissions used for groups and other users when \fBhonorSystemUmask\f1\f1 is set to \fBfalse\f1\&. By default, when \fBhonorSystemUmask\f1\f1 is set to \fBfalse\f1, new files created by MongoDB have permissions set to \fB600\f1\&. Use the \fBprocessUmask\f1 parameter to override this default with a custom \fBumask\f1 value. The file owner inherits permissions from the system \fBumask\f1\&. .PP You cannot set this parameter if \fBhonorSystemUmask\f1\f1 is set to \fBtrue\f1\&. You can only set this parameter during start\-up and cannot change this setting using the \fBsetParameter\f1\f1 database command. .PP Consider the following example, which sets the permissions for groups and other users to read/write only and retains the system \fBumask\f1 settings for the owner: .PP .EX mongod \-\-setParameter processUmask=011 .EE .PP \fBprocessUmask\f1\f1 is not available on Windows systems. .RE .PP \fBsyncdelay\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP Specify the interval in seconds between \fBfsync\f1 operations where \fBmongod\f1\f1 flushes its working memory to disk. By default, \fBmongod\f1\f1 flushes memory to disk every 60 seconds. In almost every situation you should not set this value and use the default setting. .PP Consider the following example which sets the \fBsyncdelay\f1 to \fB60\f1 seconds: .PP .EX db.adminCommand( { setParameter: 1, syncdelay: 60 } ) .EE .RS .IP \(bu 2 \fBjournalCommitInterval\f1\f1 .IP \(bu 2 \fBstorage.syncPeriodSecs\f1\f1 .RE .RE .SS WIREDTIGER PARAMETERS .PP \fBwiredTigerMaxCacheOverflowSizeGB\f1 .RS .PP MongoDB deprecates the \fBwiredTigerMaxCacheOverflowSizeGB\f1 parameter. The parameter has no effect starting in MongoDB 4.4. .PP \fIDefault\f1: 0 (No specified maximum) .PP Available for \fBmongod\f1\f1 only. .PP Specify the maximum size (in GB) for the "lookaside (or cache overflow) table" file WiredTigerLAS.wt for MongoDB 4.2.1\-4.2.x and 4.0.12\-4.0.x. The file no longer exists starting in version 4.4. .PP The parameter can accept the following values: .RS .IP \(bu 2 .RS .IP \(bu 4 Value .IP \(bu 4 Description .RE .IP \(bu 2 .RS .IP \(bu 4 \fB0\f1 .IP \(bu 4 The default value. If set to \fB0\f1, the file size is unbounded. .RE .IP \(bu 2 .RS .IP \(bu 4 number >= 0.1 .IP \(bu 4 The maximum size (in GB). If the WiredTigerLAS.wt file exceeds this size, \fBmongod\f1\f1 exits with a fatal assertion. You can clear the WiredTigerLAS.wt file and restart \fBmongod\f1\f1\&. .RE .RE .PP You can only set this parameter during runtime using the \fBsetParameter\f1\f1 database command: .PP .EX db.adminCommand( { setParameter: 1, wiredTigerMaxCacheOverflowSizeGB: 100 } ) .EE .PP To set the maximum size during start up, use the \fBstorage.wiredTiger.engineConfig.maxCacheOverflowFileSizeGB\f1\f1 instead. .PP \fIAvailable starting in MongoDB 4.2.1 (and 4.0.12)\f1 .RE .PP \fBwiredTigerConcurrentReadTransactions\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP Available for the WiredTiger storage engine only. .PP Specify the maximum number of concurrent read transactions allowed into the WiredTiger storage engine. .PP .EX db.adminCommand( { setParameter: 1, wiredTigerConcurrentReadTransactions: } ) .EE .PP \fBwiredTiger.concurrentTransactions\f1\f1 .RE .PP \fBwiredTigerConcurrentWriteTransactions\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP Available for the WiredTiger storage engine only. .PP Specify the maximum number of concurrent write transactions allowed into the WiredTiger storage engine. .PP .EX db.adminCommand( { setParameter: 1, wiredTigerConcurrentWriteTransactions: } ) .EE .PP \fBwiredTiger.concurrentTransactions\f1\f1 .RE .PP \fBwiredTigerEngineRuntimeConfig\f1 .RS .PP Available for \fBmongod\f1\f1 only. .PP Specify \fBwiredTiger\f1 storage engine configuration options for a running \fBmongod\f1\f1 instance. You can \fIonly\f1 set this parameter using the \fBsetParameter\f1\f1 command and \fInot\f1 using the command line or configuration file option. .PP Avoid modifying the \fBwiredTigerEngineRuntimeConfig\f1\f1 unless under the direction from MongoDB engineers as this setting has major implication across both WiredTiger and MongoDB. .PP Consider the following operation prototype: .PP .EX db.adminCommand({ "setParameter": 1, "wiredTigerEngineRuntimeConfig": "