Represents an InnoDB ReplicaSet. More...

Methods
Undefined	addInstance (String instance, Dictionary options)
	Adds an instance to the replicaset. More...

Undefined	disconnect ()
	Disconnects all internal sessions used by the replicaset object. More...

String	getName ()
	Returns the name of the replicaset. More...

Dictionary	listRouters (Dictionary options)
	Lists the Router instances. More...

Undefined	rejoinInstance (String instance, Dictionary options)
	Rejoins an instance to the replicaset. More...

Undefined	removeInstance (String instance, Dictionary options)
	Removes an Instance from the replicaset. More...

Dictionary	status (Dictionary options)
	Describe the status of the replicaset. More...

Undefined	setPrimaryInstance (String instance, Dictionary options)
	Performs a safe PRIMARY switchover, promoting the given instance. More...

Undefined	forcePrimaryInstance (String instance, Dictionary options)
	Performs a failover in a replicaset with an unavailable PRIMARY. More...

Boolean	removeRouterMetadata (RouterDef routerDef)
	Removes metadata for a router instance. More...

Undefined	setupAdminAccount (String user, Dictionary options)
	Create or upgrade an InnoDB ReplicaSet admin account. More...

Undefined	setupRouterAccount (String user, Dictionary options)
	Create or upgrade a MySQL account to use with MySQL Router. More...

Undefined	setOption (String option, String value)
	Changes the value of an option for the whole ReplicaSet. More...

Undefined	setInstanceOption (String instance, String option, String value)

shcore::Value	options ()
	Lists the ReplicaSet configuration options. More...

Properties
String	name
	Returns the name of the replicaset.

Detailed Description

Represents an InnoDB ReplicaSet.

The ReplicaSet object is used to manage MySQL server topologies that use asynchronous replication. It can be created using the dba.createReplicaSet() or dba.getReplicaSet() functions.

Member Function Documentation

◆ addInstance()

Undefined addInstance	(	String	instance,
		Dictionary	options
	)

Adds an instance to the replicaset.

Parameters

instance	host:port of the target instance to be added.
options	Optional dictionary with options for the operation.

Returns: Nothing.

This function adds the given MySQL instance as a read-only SECONDARY replica of the current PRIMARY of the replicaset. Replication will be configured between the added instance and the PRIMARY. Replication will use an automatically created MySQL account with a random password.

Once the instance is added to the replicaset, the function will wait for it to apply all pending transactions. The transaction sync may timeout if the transaction backlog is large. The timeout option may be used to increase that time.

The Shell connects to the target instance using the same username and password used to obtain the replicaset handle object. All instances of the replicaset are expected to have this same admin account with the same grants and passwords. A custom admin account with the required grants can be created while an instance is configured with dba.configureReplicaSetInstance(), but the root user may be used too.

The PRIMARY of the replicaset must be reachable and available during this operation.

Pre-Requisites

The following pre-requisites are expected for instances added to a replicaset. They will be automatically checked by addInstance(), which will stop if any issues are found.

binary log and replication related options must have been validated and/or configured by dba.configureReplicaSetInstance()

If the selected recovery method is incremental:

transaction set in the instance being added must not contain transactions that don't exist in the PRIMARY
transaction set in the instance being added must not be missing transactions that have been purged from the binary log of the PRIMARY

If clone is available, the pre-requisites listed above can be overcome by using clone as the recovery method.

Options

The options dictionary may contain the following values:

dryRun: if true, performs checks and logs changes that would be made, but does not execute them
label: an identifier for the instance being added, used in the output of status()
recoveryMethod: Preferred method of state recovery. May be auto, clone or incremental. Default is auto.
waitRecovery: Integer value to indicate the recovery process verbosity level.
recoveryProgress: Integer value to indicate the recovery process verbosity level.
cloneDonor: host:port of an existing replicaSet member to clone from. IPv6 addresses are not supported for this option.
interactive: boolean value used to disable/enable the wizards in the command execution, i.e. prompts and confirmations will be provided or not according to the value set. The default value is equal to MySQL Shell wizard mode. Deprecated.
timeout: timeout in seconds for transaction sync operations; 0 disables timeout and force the Shell to wait until the transaction sync finishes. Defaults to 0.
certSubject: instance's certificate subject to use when 'memberAuthType' contains "CERT_SUBJECT".

The recoveryMethod option supports the following values:

incremental: waits until the new instance has applied missing transactions from the PRIMARY
clone: uses MySQL clone to provision the instance, which completely replaces the state of the target instance with a full snapshot of another ReplicaSet member. Requires MySQL 8.0.17 or newer.
auto: compares the transaction set of the instance with that of the PRIMARY to determine if incremental recovery is safe to be automatically chosen as the most appropriate recovery method. A prompt will be shown if not possible to safely determine a safe way forward. If interaction is disabled, the operation will be canceled instead.

If recoveryMethod is not specified 'auto' will be used by default.

The waitRecovery option supports the following values:

0: not supported.
1: do not show any progress information.
2: show detailed static progress information.
3: show detailed dynamic progress information using progress bars.

The recoveryProgress option supports the following values:

0: do not show any progress information.
1: show detailed static progress information.
2: show detailed dynamic progress information using progress bars.

By default, if the standard output on which the Shell is running refers to a terminal, the waitRecovery option has the value of 3. Otherwise, it has the value of 2.

The cloneDonor option is used to override the automatic selection of a donor to be used when clone is selected as the recovery method. By default, a SECONDARY member will be chosen as donor. If no SECONDARY members are available the PRIMARY will be selected. The option accepts values in the format: 'host:port'. IPv6 addresses are not supported.

Attention: The waitRecovery option will be removed in a future release. Please use the recoveryProgress option instead.; The interactive option will be removed in a future release.

◆ disconnect()

void disconnect ( )

Disconnects all internal sessions used by the replicaset object.

Returns: Nothing.

Disconnects the internal MySQL sessions used by the replicaset to query for metadata and replication information.

◆ getName()

String getName ( )

Returns the name of the replicaset.

Returns: name of the replicaset.

◆ listRouters()

String listRouters ( Dictionary options )

Lists the Router instances.

Parameters

options Optional dictionary with options for the operation.

Returns: A JSON object listing the Router instances associated to the ReplicaSet.

This function lists and provides information about all Router instances registered for the ReplicaSet.

Whenever a Metadata Schema upgrade is necessary, the recommended process is to upgrade MySQL Router instances to the latest version before upgrading the Metadata itself, in order to minimize service disruption.

The options dictionary may contain the following attributes:

onlyUpgradeRequired: boolean, enables filtering so only router instances that support older version of the Metadata Schema and require upgrade are included.

◆ rejoinInstance()

Undefined rejoinInstance	(	String	instance,
		Dictionary	options
	)

Rejoins an instance to the replicaset.

Parameters

instance	host:port of the target instance to be rejoined.
options	Optional dictionary with options for the operation.

Returns: Nothing.

This function rejoins the given MySQL instance to the replicaset, making it a read-only SECONDARY replica of the current PRIMARY. Only instances previously added to the replicaset can be rejoined, otherwise the <ReplicaSet>.addInstance() function must be used to add a new instance.

The PRIMARY of the replicaset must be reachable and available during this operation.

Pre-Requisites

The following pre-requisites are expected for instances rejoined to a replicaset. They will be automatically checked by rejoinInstance(), which will stop if any issues are found.

If the selected recovery method is incremental:

transaction set in the instance being rejoined must not contain transactions that don't exist in the PRIMARY
transaction set in the instance being rejoined must not be missing transactions that have been purged from the binary log of the PRIMARY
executed transactions set (GTID_EXECUTED) in the instance being rejoined must not be empty.

If clone is available, the pre-requisites listed above can be overcome by using clone as the recovery method.

Options

The options dictionary may contain the following values:

dryRun: if true, performs checks and logs changes that would be made, but does not execute them
recoveryMethod: Preferred method of state recovery. May be auto, clone or incremental. Default is auto.
waitRecovery: Integer value to indicate the recovery process verbosity level.
recoveryProgress: Integer value to indicate the recovery process verbosity level.
cloneDonor: host:port of an existing replicaSet member to clone from. IPv6 addresses are not supported for this option.
interactive: boolean value used to disable/enable the wizards in the command execution, i.e. prompts and confirmations will be provided or not according to the value set. The default value is equal to MySQL Shell wizard mode. Deprecated.
timeout: timeout in seconds for transaction sync operations; 0 disables timeout and force the Shell to wait until the transaction sync finishes. Defaults to 0.

The recoveryMethod option supports the following values:

incremental: waits until the rejoining instance has applied missing transactions from the PRIMARY
clone: uses MySQL clone to provision the instance, which completely replaces the state of the target instance with a full snapshot of another ReplicaSet member. Requires MySQL 8.0.17 or newer.
auto: compares the transaction set of the instance with that of the PRIMARY to determine if incremental recovery is safe to be automatically chosen as the most appropriate recovery method. A prompt will be shown if not possible to safely determine a safe way forward. If interaction is disabled, the operation will be canceled instead.

If recoveryMethod is not specified 'auto' will be used by default.

The waitRecovery option supports the following values:

0: not supported.
1: do not show any progress information.
2: show detailed static progress information.
3: show detailed dynamic progress information using progress bars.

The recoveryProgress option supports the following values:

0: do not show any progress information.
1: show detailed static progress information.
2: show detailed dynamic progress information using progress bars.

By default, if the standard output on which the Shell is running refers to a terminal, the waitRecovery option has the value of 3. Otherwise, it has the value of 2.

The cloneDonor option is used to override the automatic selection of a donor to be used when clone is selected as the recovery method. By default, a SECONDARY member will be chosen as donor. If no SECONDARY members are available the PRIMARY will be selected. The option accepts values in the format: 'host:port'. IPv6 addresses are not supported.

Attention: The waitRecovery option will be removed in a future release. Please use the recoveryProgress option instead.; The interactive option will be removed in a future release.

◆ removeInstance()

Undefined removeInstance	(	String	instance,
		Dictionary	options
	)

Removes an Instance from the replicaset.

Parameters

instance	host:port of the target instance to be removed
options	Optional dictionary with options for the operation.

Returns: Nothing.

This function removes an Instance from the replicaset and stops it from replicating. The instance must not be the PRIMARY and it is expected to be reachable.

The PRIMARY of the replicaset must be reachable and available during this operation.

Options

The instance definition is the connection data for the instance.

The options dictionary may contain the following attributes:

force: boolean, indicating if the instance must be removed (even if only from metadata) in case it cannot be reached. By default, set to false.
timeout: maximum number of seconds to wait for the instance to sync up with the PRIMARY. 0 means no timeout and <0 will skip sync.

The force option (set to true) is required to remove instances that are unreachable. Removed instances are normally synchronized with the rest of the replicaset, so that their own copy of the metadata will be consistent and the instance will be suitable for being added back to the replicaset. When the force option is set during removal, that step will be skipped.

◆ status()

String status ( Dictionary options )

Describe the status of the replicaset.

Parameters

options Optional dictionary with options.

Returns: A JSON object describing the status of the members of the replicaset.

This function will connect to each member of the replicaset and query their state, producing a status report of the replicaset as a whole, as well as of their individual members.

Options

The following options may be given to control the amount of information gathered and returned.

extended: Can be 0, 1 or 2. Default is 0.

Option 'extended' may have the following values:

0 - regular level of details. Only basic information about the status of the instance and replication is included, in addition to non-default or unexpected replication settings and status.
1 - includes Metadata Version, server UUID and the raw information used to derive the status of the instance, size of the applier queue, value of system variables that protect against unexpected writes etc.
2 - includes important replication related configuration settings, such as SSL, worker threads, replication delay and heartbeat delay.

◆ setPrimaryInstance()

Undefined setPrimaryInstance	(	String	instance,
		Dictionary	options
	)

Performs a safe PRIMARY switchover, promoting the given instance.

Parameters

instance	host:port of the target instance to be promoted.
options	Dictionary with additional options.

Returns: Nothing

This command will perform a safe switchover of the PRIMARY of a replicaset. The current PRIMARY will be demoted to a SECONDARY and made read-only, while the promoted instance will be made a read-write master. All other SECONDARY instances will be updated to replicate from the new PRIMARY.

During the switchover, the promoted instance will be synchronized with the old PRIMARY, ensuring that all transactions present in the PRIMARY are applied before the topology change is committed. If that synchronization step takes too long or is not possible in any of the SECONDARY instances, the switch will be aborted. These problematic SECONDARY instances must be either repaired or removed from the replicaset for the fail over to be possible.

For a safe switchover to be possible, all replicaset instances must be reachable from the shell and have consistent transaction sets. If the PRIMARY is not available, a forced failover must be performed instead, using forcePrimaryInstance().

Options

The following options may be given:

dryRun: if true, will perform checks and log operations that would be performed, but will not execute them. The operations that would be performed can be viewed by enabling verbose output in the shell.
timeout: integer value with the maximum number of seconds to wait until the instance being promoted catches up to the current PRIMARY (default value is retrieved from the 'dba.gtidWaitTimeout' shell option).

◆ forcePrimaryInstance()

Undefined forcePrimaryInstance	(	String	instance,
		Dictionary	options
	)

Performs a failover in a replicaset with an unavailable PRIMARY.

Parameters

instance	host:port of the target instance to be promoted. If blank, a suitable instance will be selected automatically.
options	Dictionary with additional options.

Returns: Nothing

This command will perform a forced failover of the PRIMARY of a replicaset in disaster scenarios where the current PRIMARY is unavailable and cannot be restored. If given, the target instance will be promoted to a PRIMARY, while other reachable SECONDARY instances will be switched to the new PRIMARY. The target instance must have the most up-to-date GTID_EXECUTED set among reachable instances, otherwise the operation will fail. If a target instance is not given (or is null), the most up-to-date instance will be automatically selected and promoted.

After a forced failover, the old PRIMARY will be considered invalid by the new PRIMARY and can no longer be part of the replicaset. If the instance is still usable, it must be removed from the replicaset and re-added as a new instance. If there were any SECONDARY instances that could not be switched to the new PRIMARY during the failover, they will also be considered invalid.

Attention: a forced failover is a potentially destructive action and should only be used as a last resort measure.

Data loss is possible after a failover, because the old PRIMARY may have had transactions that were not yet replicated to the SECONDARY being promoted. Moreover, if the instance that was presumed to have failed is still able to process updates, for example because the network where it is located is still functioning but unreachable from the shell, it will continue diverging from the promoted clusters. Recovering or re-conciliating diverged transaction sets requires manual intervention and may some times not be possible, even if the failed MySQL servers can be recovered. In many cases, the fastest and simplest way to recover from a disaster that required a forced failover is by discarding such diverged transactions and re-provisioning a new instance from the newly promoted PRIMARY.

Options

The following options may be given:

dryRun: if true, will perform checks and log operations that would be performed, but will not execute them. The operations that would be performed can be viewed by enabling verbose output in the shell.
timeout: integer value with the maximum number of seconds to wait until the instance being promoted catches up to the current PRIMARY.
invalidateErrorInstances: if false, aborts the failover if any instance other than the old master is unreachable or has errors. If true, such instances will not be failed over and be invalidated.

◆ removeRouterMetadata()

String removeRouterMetadata ( RouterDef routerDef )

Removes metadata for a router instance.

Parameters

routerDef identifier of the router instance to be removed (e.g. 192.168.45.70::system)

Returns: Nothing

MySQL Router automatically registers itself within the InnoDB cluster metadata when bootstrapped. However, that metadata may be left behind when instances are uninstalled or moved over to a different host. This function may be used to clean up such instances that no longer exist.

The <ReplicaSet>.listRouters() function may be used to list registered router instances, including their identifier.

◆ setupAdminAccount()

Undefined setupAdminAccount	(	String	user,
		Dictionary	options
	)

Create or upgrade an InnoDB ReplicaSet admin account.

Parameters

user	Name of the InnoDB ReplicaSet administrator account.
options	Dictionary with options for the operation.

Returns: Nothing.

This function creates/upgrades a MySQL user account with the necessary privileges to administer an InnoDB ReplicaSet.

This function also allows a user to upgrade an existing admin account with the necessary privileges before a dba.upgradeMetadata() call.

The mandatory argument user is the name of the MySQL account we want to create or upgrade to be used as Administrator account. The accepted format is username[@host] where the host part is optional and if not provided defaults to ''.

The options dictionary may contain the following attributes:

password: The password for the InnoDB ReplicaSet administrator account.
interactive: boolean value used to disable/enable the wizards in the command execution, i.e. prompts and confirmations will be provided or not according to the value set. The default value is equal to MySQL Shell wizard mode. Deprecated.

If the user account does not exist, either the password, requireCertIssuer or requireCertSubject are mandatory.

If the user account exists, the update option must be enabled.

Attention: The interactive option will be removed in a future release.

◆ setupRouterAccount()

Undefined setupRouterAccount	(	String	user,
		Dictionary	options
	)

Create or upgrade a MySQL account to use with MySQL Router.

Parameters

user	Name of the account to create/upgrade for MySQL Router.
options	Dictionary with options for the operation.

Returns: Nothing.

This function creates/upgrades a MySQL user account with the necessary privileges to be used by MySQL Router.

This function also allows a user to upgrade existing MySQL router accounts with the necessary privileges after a dba.upgradeMetadata() call.

The mandatory argument user is the name of the MySQL account we want to create or upgrade to be used by MySQL Router. The accepted format is username[@host] where the host part is optional and if not provided defaults to ''.

The options dictionary may contain the following attributes:

password: The password for the MySQL Router account.
interactive: boolean value used to disable/enable the wizards in the command execution, i.e. prompts and confirmations will be provided or not according to the value set. The default value is equal to MySQL Shell wizard mode. Deprecated.

If the user account does not exist, either the password, requireCertIssuer or requireCertSubject are mandatory.

If the user account exists, the update option must be enabled.

Attention: The interactive option will be removed in a future release.

◆ setOption()

Undefined setOption	(	String	option,
		String	value
	)

Changes the value of an option for the whole ReplicaSet.

Parameters

option	The option to be changed.
value	The value that the option shall get.

Returns: Nothing.

This function changes an option for the ReplicaSet.

The accepted options are:

replicationAllowedHost string value to use as the host name part of internal replication accounts. Existing accounts will be re-created with the new value.
tag:<option>: built-in and user-defined tags to be associated to the Cluster.

Tags

Tags make it possible to associate custom key/value pairs to a ReplicaSet, storing them in its metadata. Custom tag names can be any string starting with letters and followed by letters, numbers and _. Tag values may be any JSON value. If the value is null, the tag is deleted.

◆ setInstanceOption()

Undefined setInstanceOption	(	String	instance,
		String	option,
		String	value
	)

◆ options()

shcore::Value options ( )

Lists the ReplicaSet configuration options.

Returns: A JSON object describing the configuration options of the ReplicaSet.

This function lists the configuration options for the ReplicaSet and its instances.

Methods

Properties

Detailed Description

Member Function Documentation

◆ addInstance()

◆ disconnect()

◆ getName()

◆ listRouters()

◆ rejoinInstance()

◆ removeInstance()

◆ status()

◆ setPrimaryInstance()

◆ forcePrimaryInstance()

◆ removeRouterMetadata()

◆ setupAdminAccount()

◆ setupRouterAccount()

◆ setOption()

◆ setInstanceOption()

◆ options()