Persistence configuration

Transformer’s built-in Persistence Service provides a persistent key-value store, shareable between nodes in a Transformer cluster. The configuration items here are used to define how persistence databases and files are accessed and controlled.

The Persistence Service is disabled by default. See Activating the Persistence Service 6.2.

The Persistence Service 6.2 persists data to a relational database server or to the local filesystem, but only a relational database server is supported as a persistence store in production environments. File-backed persistence (persisting data to the local filesystem) is provided as an option of convenience for development environments, but it is unsupported in production environments.

File-backed persistence is unsupported in production environments, and, from version 7 of the Persistence Service, file-backed persistence cannot be used in Transformer clusters.

All the nodes of a Transformer cluster can persist data and share it. If data is persisted to files, each Transformer in the cluster maintains its own persistence file, but the Transformer also sends a copy of its persisted data to every other Transformer so that they can write it to their own persistence files as well. If data is persisted to a database, there’s just one copy of the database that’s shared by all the Transformers in the cluster.

If you’re persisting the data to a database (rather than a file), besides defining the relevant configuration items that are documented here, you’ll also need to set up the database, giving it a name, defining the table and columns needed to store the persisted data, and defining the usernames(s) and access permissions that will allow Transformer to write and read the data.

Add your persistence configuration to the file <Framework-root>/global_config/overrides/PersistenceService/Transformer/etc/persistence.conf, where <Framework-root> is your Deployment Framework’s top-level directory.

In the following documentation, the term "this Transformer" means the Transformer for which you are defining the configuration, and "this node" means the cluster node that "this Transformer" represents.

add-database-params

add-database-params specifies parameters for the database or file used to persist data.

Syntax:

add-database-params
   init-string    [string]
   driver-name    [string]
   username       [string]
   password       [string]
   table-name     [string]
   columns        [array of strings]
   extra-params   [array of strings]
end-add-database-params
Option Type Default Description

columns

array of strings

[none]

A space-separated list of the names of the database columns where the persisted data is stored.

This option is not needed when persisting to a file.

Each piece of data is stored as a key-value pair. The names of the database columns should be listed in this order: <key column> <value column>. For example, to persist keys to a column called 'pers_key' and values to a column called 'pers_val', use the configuration below:

columns pers_key pers_val
If you’re implementing your own Transformer module that persists data to the persistence database, make sure the keys you store have names that are unique to your module and don’t clash with the keys used by other modules, such as the Watchlist Service.

driver-name

string

[none]

The name of the JDBC driver; for example, com.mysql.jdbc.Driver

This option’s not needed when persisting to a file.

extra-params

array of strings

[implementation dependent]

A list of additional parameters that are used by this Transformer’s implementation of the Java Persistor interface.

This option’s not needed when persisting to a file.

The type, number, usage and default values of the parameters in the list are dependent on implementation of Persistor that’s used; see the API documentation for the Persistor interface in the Transformer Persistence API for Java.

For example, the basic JDBC implementation of Persistor (JdbcPersistorImpl) that’s provided with, and used by, Transformer’s Persistence Config blade provides for two extra parameters: the Java debug level, and the number of pooled connections (the latter has a default value).

Also see persistence-classid.

init-string

string

[none]

Initialisation string for the persistence database or file.

  • If data is persisted to a file, this is the directory name and path of the file. In the Persistence Config blade the default file path and name is ${ccd}/persistence, where ${ccd} is the Transformer’s current configuration directory.

  • If data is persisted to a database, this is the JDBC URL of the database; for example:

    jdbc:mysql://localhost/mydatabase

password

string

[none]

The database access password associated with username. This option’s not needed when persisting to a file.

table-name

string

[none]

The name of the database table where the persisted data is stored. This option’s not needed when persisting to a file.

username

string

[none]

The username that allows access to the database. This option’s not needed when persisting to a file.

disable-cluster-replication

Persistence Service 6.2 only. The Persistence Service 7 does not replicate persisted values between nodes in a cluster.

disable-cluster-replication when TRUE,prevents this Transformer from sending to the other Transformers in the cluster data that it has already persisted.

When persistence is file-based, don’t specify disable-cluster-replication. The default value of FALSE then applies, and, although each Transformer in the cluster maintains its own persistence file, this Transformer sends a copy of its persisted data to every other Transformer so that they can also write it to their own persistence files.

Conversely, when a (shared) persistence database is used, you should set disable-cluster-replication to TRUE on every node in the cluster to prevent multiple copies of the same data being persisted to the (shared) database. You should set dont-persist-cluster-messages to TRUE as well.

Syntax: disable-cluster-replication <boolean>

Type: boolean

Default value: FALSE

dont-persist-cluster-messages

Persistence Service 6.2 only. The Persistence Service 7 does not replicate persisted values between nodes in a cluster.

dont-persist-cluster-messages when TRUE, ensures that messages received from another Transformer in the cluster are not persisted on this Transformer.

When persistence is file-based, don’t specify dont-persist-cluster-messages. The default value of FALSE then applies, so that this Transformer saves to its own persistence file all the persisted data that other Transformers in the cluster send to it.

Conversely, when a (shared) persistence database is used, you should set dont-persist-cluster-messages to TRUE on every node in the cluster, to prevent the same data being persisted to the database multiple times (the data’s then persisted solely by the Transformer where it originated). You should set disable-cluster-replication to TRUE as well.

Syntax: dont-persist-cluster-messages <boolean>

Type: boolean

Default value: FALSE (messages from another cluster node are always persisted on this node)

enable-file-database

File-backed persistence is unsupported in production environments, and, from version 7 of the Persistence Service, file-backed persistence cannot be used in Transformer clusters.

enable-file-database when TRUE, enables file-based persistence.

Syntax: enable-file-database <boolean>

Type: boolean

Default value: FALSE (data is persisted to a database)

logfile

logfile specifies the name of the file in which events about persisting data are recorded.

The filename can contain the parameters %a and %h At run time, %a is replaced by the Transformer’s application name (see the DataSource configuration item application-name), and %h is replaced by the host name of the machine on which the Transformer is running.

Syntax: logfile <log-file-name>

Type: string

Default value: persistence.log

log-level

log-level specifies the severity of the errors and events that Transformer’s persistence module writes to its log file (see logfile).

Syntax: log-level <log-level-name>

Type: string

Default value: INFO

Values accepted:

log-level-name DESCRIPTION

DEBUG

Reports all errors and events.

INFO

Reports events and information regarding normal operation of the persistence module, and all errors included in the WARN, NOTIFY, ERROR and CRIT log levels.

WARN

Reports minor errors and all errors included in the NOTIFY, ERROR and CRIT log levels.

NOTIFY

Reports errors regarding data corruptions and all errors included in the ERROR and CRIT log levels.

ERROR

Reports serious errors and all errors included in the CRIT log level.

CRIT

Reports critical errors that prevent the persistence module from running.

persistence-classid

persistence-classid specifies the Java class ID of the implementation of the Java Persistor interface that provides this Transformer’s database persistence capability. There’s a basic JDBC implementation of Persistor that’s provided with, and used by, Transformer’s Persistence Config blade; it’s called JdbcPersistorImpl. (For more about Persistor and JdbcPersistorImpl, see the Transformer Persistence API for Java documentation.)

The class ID is conventionally the package name of the Java class (with . replaced by /), plus the class name itself (see the example below).

Syntax: persistence-classid <java-class-id>

Type: string

Default value: (none; but persistence-classid is only required if enable-file-database is FALSE)

Example:

persistence-classid com/caplin/transformer/persistence/JdbcPersistorImpl

If you’ve implemented your own Persistance Adapter, you must:

  • Specify your custom Persistance Adapter class in an add-javaclass configuration item.

  • Specify the same class id for the Persistance Adapter in both the class-id option of the add-javaclass item and the persistence-classid item.

See also: