The Data Server is an option when it is not desirable, or possible, to directly connect the Display Viewer Application, Display Viewer Applet or Display Server to data sources. Select the Data Server tab in the Application Options dialog to configure your Data Server(s).

The Data Server uses EII and XML technologies to gather, federate and distribute information from disparate data sources based on information currently in demand. It also caches the data so that multiple demands are delivered to any number of clients - without need of subsequent data queries. These important factors greatly enhance processing speed. Because the Data Server can exist behind firewalls, it also greatly simplifies and strengthens the secured delivery of information to clients beyond the firewall.

The Data Server can be used to serve data to the Display Viewer Application, Display Viewer Applet and Display Server. See the Deployment section for more information.

The following scenarios, or a combination of them, would indicate that one or multiple instances of the Data Server would be useful.

Data Access
In some cases the data source resides in a sub-network where it is not possible or desirable to connect directly. For example, you may have a data source that communicates via a TCP/IP port that is not exposed across a security firewall. In this case, the Data Server would exist behind the firewall and connect to the chosen display technology via HTTP or an available exposed port. In most cases, at least one Data Server must be configured if the Display Viewer applet is used. Applets are restricted to making connections back to the webserver host and will throw an exception if connection attempts are made to any other host.

Data Centralization
While a direct TCP/IP connection to the data sources may be possible, it may not be desirable for each data source and each client. For example, with a Display Viewer Application deployment you might not want a direct TCP connection from each client to every data source. In this case, the Data Server could act as a proxy by directly connecting to all data sources and providing data to each Java application client.

Data Reduction/Aggregation
In some cases, there may be a lot of raw data that does not need to be sent except upon demand. Using the Data Server in this situation can reduce network traffic by performing data calculations from within a sub-network and providing either only the aggregated data or raw data on demand to the chosen display technology. For example the Data Server may publish an average metric over all servers in a subnet, but provide raw data on drill down to a selected server.

Scalability
The Data Server has several data processing features including: an in memory data cache, a persistent alert rules engine, and the ability to perform data calculations. This processing load can be scaled out by dividing the labor across multiple Data Servers that will connect to the chosen display technology.

High Availability
To ensure client access to data in the event of a server failure, a pair of redundant Data Servers can be deployed on two separate hosts. In such a deployment each server has the same configuration and access to the same data sources. Typically, one server is considered the primary and the other is the backup. Each client is configured with two Data Server connections, one to the primary and another to the backup.

At startup, the client connects to the primary server if available, otherwise it connects to the backup server. If neither is available, the client periodically retries connecting to both. Client connections are specified on the Data Server tab of the Option dialog in the Builder, as or via the -dataserver command line options when launching the Display Builder, Display Viewer Application or Display Server.

A running Data Server is either "active" or "standby". An active server loads all global, cache, and alert definitions and activates them. In a standby server, none of these actions are performed, thereby avoiding the overhead of maintaining alert and cache data sources until the server is active.

NOTE: All servers specified should use the same initialization files (*OPTIONS.ini) and should have access to the same external data sources, so that displays will appear the same regardless of which server is connected.

A primary/backup redundant server pair can be configured in the following three modes:

1. Active / Active: In this mode, both servers are always active. This configuration is the simplest and provides the quickest response to clients in the event of a server failure. However, this mode increases data processing as both servers are always storing cache data, as well as processing and generating duplicate alerts.

2. Active / Standby with manual reset: In this mode, the backup server is started with the -standby:warm command line option. The backup does not become active until a client connects to it (typically after the primary fails). However, after the backup server becomes active it remains active and clients stay connected to it, even if the primary server becomes active. This mode reduces data processing in normal conditions but after a primary server fails and recovers both servers might become active, possibly with clients connected to both servers. To restore the system to its initial active / standby state, the backup server must be manually stopped and restarted (with the -standby:warm command line option).

3. Active / Standby with auto-reset. In this mode both servers belong to a server group, where one server is designated as the preferred primary server and the other as the preferred backup server. Each server monitors the status of the other. The backup runs in standby mode until it detects the failure of the primary and then becomes active. If the primary server then becomes active the backup server automatically returns to standby mode. Clients then automatically disconnect from the backup and (re)connect to the primary. This mode ensures that only one server in the group is active at any time. This means that only one cache and alert data source are utilized, thereby avoiding duplicate cache storage and duplicate alert generation.

NOTE: The cache and alert persistence features can also be used to ensure that the backup server gets the correct state of those data sources when it becomes active.

See the Server Group section (below) for details about configuring this high availability mode using the Builder GUI. This mode can also be configured on the command line. For details, see the -group_member argument.

Running the Data Server

You can run the Data Server as an application, a daemon process, or in the background as a Windows Service. NOTE: The first time you run the Data Server you must run it as an application to configure the initial settings.

Running the Data Server as a daemon process will allow you to run without a display. To do so, run the -daemon command line parameter from a Windows Command Prompt or UNIX terminal window.

Several command line options are supported for the Data Server. Java options specified in RTV_JAVAOPTS will be used by the run_dataserver scripts. NOTE: These options will not be used when the Data Server is started from the Windows Start menu.

Data source specific options are read in from initialization files created in the Display Builder. For information on creating initialization files or command line options for your data source, refer to Application Options or Command Line Options under the Data Sources section of this documentation.

NOTE: To stop the Data Server, you can use the kill_dataserver Command Line Option.

Run the Data Server: Windows or UNIX

Windows			UNIX
From the Windows Start Menu: Select Start-->Programs-->RTView Data Server NOTE: When you start the Data Server from the Windows Start menu, it runs from the demos directory. The Data Server will try to read initialization (.ini) files and resolve XML sources containing relative paths from that directory. To start the Data Server in another directory, you must run RTView from a command window. NOTE: It is possible to specify a directory for your initialization files.	From a Command Prompt window: Select Start-->Programs-->RTView--> GMS Command Prompt In the command window type: run_dataserver		In an initialized terminal window type: run_dataserver

To begin serving data, click Start Serving Data. Click Stop Serving Data to end this process. If you configure the Data Server to output to an XML file, when you start serving data information will be added to the output XML file each time new data is received for each attachment in all of the specified data configuration files. If you configure the Data Server to output data via socket, when you start serving data the Data Server will receive data requests from RTView clients (Display Builder, Display Viewer Application, Display Viewer Applet) indicating what data is needed for currently running displays.
 

Server Modes
This information must correspond with information entered in the Application Options - Data Server dialog. NOTE: By default, the Data Server starts in File Mode. If you change the Server Mode, then you must save the configuration and restart the Data Server.

File
Data redirected from configuration files will be output to an XML file. NOTE: The XML data source is never redirected through the Data Server when you output data to a file. When you start serving data, information will be added to the output XML file each time new data is received for each data attachment in all of the specified data configuration files.

Output File Name - Name of file output by the Data Server containing data from specified configuration files. Default output file name is rtvdata.xml.

Data Configuration Files
Specify which data configuration (.rtv) file(s) will be used to generate the output XML file. Data configuration files are display (.rtv) files created in the Display Builder that contain attachments to data you want to redirect through the Data Server.

Data Configuration Files	Add - Add a data configuration file to the list. If a data configuration file is added while the Data Server is serving data, it will start serving data for attachments in that file as soon as new information becomes available.
	Remove - Remove the selected data configuration (.rtv) file from the list. If the Data Server is serving data when a file is removed, it will stop serving data for that file immediately.
Substitutions	To add or edit a substitution on a specified data configuration (.rtv) file, double-click in the corresponding field of the Substitutions column.

Socket
Data requested by RTView clients will be output via socket. When you start serving data, the Data Server will receive data requests from RTView clients indicating what data is needed for currently running displays. NOTE: By default the RTView client will connect directly to the XML data source, while all other data sources are redirected through the Data Server. For information on redirecting the XML data source, see Application Options - Data Server.
 

Port - Specify the port over which the Data Server will communicate with RTView. You do not need to specify a port if you are using the default: 3278.

Use Secure Sockets - If selected, a secure socket layer (SSL) will be used. RTView clients may connect directly via socket or through the intermediary Data Servlet using HTTP or HTTPS requests to receive data. NOTE: Using the servlet requires a separate installation process, see the Data Servlet section for details.

Use Client Credentials for Database Login - Pass RTView login information into all data sources that have the Use Client Credentials option enabled. NOTE: Use Client Credentials for Database Login only works in Socket mode. Some data sources do not support this feature. For information on Application Options for your data source, refer to the Data Sources section of this documentation.

Send Changed Data Only - If selected, data will be sent only when it has changed. If you want to continuously plot data that has not changed in the trend graph, deselect the Send Changed Data Only checkbox to send all the data regardless of whether or not it has changed. NOTE: Sending all data can lead to performance issues given the amount of information that will be sent over the socket.

Preload Data for Configuration Files
You may specify configuration files for data that you want the Data Server to preload so it will be immediately available to any clients that request it.

Data Configuration Files
Specify which data configuration (.rtv) file(s) will be preloaded by the Data Server. This data will be immediately available to any clients that request it later, and will be updated during Data Server updates. Data configuration files are display (.rtv) files created in the Display Builder that contain attachments to data you want to redirect through the Data Server. Enabled if the Preload Data for Configuration Files flag is selected.
 

	Add - Add a data configuration file to the list.
	Remove - Remove the selected data configuration (.rtv) file from the list.

Console Options

Show Data in Console - If selected, the Data Server Console will print out a line for each piece of data that is being served.

General Options

Start/Stop Serving Data - Start or stop serving data.

Save Configuration - Save settings to an initialization file (DATASERVER.ini), which will be used next time you run the Data Server. NOTE: Unless you specify a directory for your initialization files, you must run the Data Server from the same directory in which the initialization  (.ini) file was saved.

About - Click on to read about RTView.

Exit - Exit the Data Server, stop serving data and close the Data Server window.

Console
The Console tab records errors and information. If you select the Show Data in Console check box on the Configuration tab, the console will also display a line for each piece of data that is being served. Click the Clear button to purge all data from the console. NOTE: By default, timestamps are displayed in the Console window. Refer to Setup>RTV_JAVAOPTS for details on how to disable timestamps.

Clients
The Clients tab appears when you start the Data Server in Socket mode. NOTE: By default, the Data Server starts in File Mode. If you change the Server Mode, then you must save the configuration and restart the Data Server.

Connections
This table displays one row for each client that is connected via socket to the Data Server. When clients are connected through the Data Servlet, one connection is shown for the servlet and separate connections are displayed for each HTTP/HTTPS client. If an HTTP/HTTPS client exits under abnormal circumstances, it may not be removed from the table for several minutes.

NOTE: Use the Get Data Server Connection Status function to access information about the status of connections from the Display Builder and Display Viewer Application to the default Data Server and any Named Data Servers. For each client connection, the table contains the following columns.
 

Client ID - A unique number assigned by the Data Server to this client connection. The first client connection is assigned an ID of 1, the second is assigned an ID of 2, etc. The ID for an HTTP/HTTPS client will contain extra digits, starting with the ID of the Data Servlet's connection. For example, if the servlet is assigned a client ID of 1, then the first HTTP/HTTPS client will be assigned an ID of 10001, the next will be 10002, etc. ID numbers will not be reused during a Data Server session.

Address - IP address of the client. Clients running on the same host as the Data Server will display the IP address 127.0.0.1.

Host - Name of the host on which the client is running. Clients running on the same host as the Data Server will display the host name localhost. If a client's host cannot be determined, the IP address will be shown instead.

Duration - Time elapsed in this client's session with the Data Server.

Last Data Sent - Number of characters of XML data contained in this client's most recent update.

Total Data Sent - Number of characters of XML data contained in all updates during this client's session with the Data Server.

Number of Clients
This trend graph (above) displays the number of clients (direct socket, servlet, HTTP/HTTPS) connected to the Data Server over the past hour.

Server Group
Use the Server Group tab to configure the Active / Standby with auto-reset high availability mode. This mode can also be configured on the command line. For details, see the -group_member argument. 

NOTE: For best performance, configure clients to connect to the primary server first (before attempting to connect to the backup server). For example, if the default primary Data Server is A:5555 and the default backup server is B:5555, set each client's connection string to the server pair as "A:5555,B:5555". Or, if connections are made via the rtvdata servlet, set each client's connection string to "http://A/rtvdata,http://B/rtvdata".

Group Options for this Server
Configure high availability settings for the Data Server.

Include in Group
Select to include the server in the server group.

Priority
Enter a value of 1 or more, with 1 being the lowest priority. A larger value assigns a higher priority. The running server in the pair with the highest priority becomes the primary server. The priority defaults to a value of 1 (the lowest priority) if Include in Group is selected and the priority is not specified.

Timeout
Enter the amount of time (in seconds) that the backup server waits for a connection or response from the primary server, after which it assumes the primary server has failed. The default is 5 seconds. (Note that in most cases, if the primary server terminates or the host on which it is running is shut down, the backup server detects this immediately).

NOTE: You can use the group_initial_wait option to lengthen the wait time for the backup server to connect to its pair at startup. This prevents a failure when the primary and backup server are both started and the backup server attempts to connect to the primary server before it is connection ready.

Connection to other Group Member
Specify the members of the fault tolerant pair.

Member
Enter the hostname and port number connection of the other Data Server in this redundant pair using the following format:

hostname:port

If a hostname is specified without a port number, port 3278 is assumed. Do not specify a URL for the rtvdata servlet. NOTE: Selecting Include in Group mitigates the need to specify the -standby:warm option. The server remains in standby mode until its determined which server is the primary.