PureMessage is best understood as a set of components that combine to form a comprehensive system for filtering spam and viruses. While most of these components are essential to the operation and administration of PureMessage, some, such as the the End User Web Interface, Groups Web Interface and PureMessage API, are optional. PureMessage consists of the following main components:
A web-based graphical user interface for managing and configuring PureMessage. The Manager is installed as part of each PureMessage "role" during installation and runs as the HTTPD (Manager) service. The Manager contains tabs that correspond with key areas of PureMessage functionality (for example, Policy and Reports). PureMessage features are also accessible from the command line. By default, these commands are run as the 'pmx6' user.
PureMessage operates as a series of services. Background services such as HTTPD (Manager) and Milter (Policy) are activated when PureMessage starts. PureMessage also uses a set of scheduled jobs to perform a variety of administrative tasks at specified times. These jobs are controlled collectively by the "Scheduler", which itself is a Background Service. The status of all enabled services is displayed on the Local Services tab of the PureMessage Manager.
A powerful configuration and management tool for message filtering, the PureMessage policy engine uses a set of rules to process messages that pass through the PureMessage mail filter. The Sieve-based policy engine runs as the Milter (Policy) service. It can be managed and configured from either the PureMessage Manager or the command line. Policy lists and maps can also be configured so that conditions and actions are applied to the email addresses or hostnames contained in a particular list or map.
The PureMessage quarantine holds messages that have been quarantined according to the rules applied by the PureMessage policy engine. Quarantined messages are managed using either the pmx-qman command-line program or via the Quarantine tab in the PureMessage Manager. Depending on PureMessage installation options, end users can manage their own quarantined messages using the End User Web Interface (EUWI).
PureMessage generates pre-defined reports that provide graphical or tabular data on key performance statistics. To use these reporting features, the PostgreSQL database must be enabled. In multi-server configurations, data is collected from various network locations and is stored in a centralized database.
End User Web Interface
A web-based interface that allows end users to manage their own quarantined messages. The amount of control granted to users depends on the settings specified by the PureMessage administrator. The EUWI allows users to perform such tasks as approving blocked messages, deleting messages, and creating lists of approved and blocked senders. The EUWI runs as a PureMessage service and, by default, uses port 28443.
Groups Web Interface
A web-based interface that allows a global administrator to delegate administrative responsibilities to "group" administrators based on groups/domains and/or roles. Delegated tasks can include quarantine management, reporting, list management and the configuration of certain policy settings. Even if you do not plan to deploy PureMessage under the groups model, you can use the quarantine and reporting features as an alternative to equivalent features in the PureMessage Manager.
The PerlMx module specifies the PureMessage interface for writing sendmail filter modules in Perl. In addition, it includes the "glue" code required to register a new filter and run the filter as a standalone process.